Home
  By Author [ A  B  C  D  E  F  G  H  I  J  K  L  M  N  O  P  Q  R  S  T  U  V  W  X  Y  Z |  Other Symbols ]
  By Title [ A  B  C  D  E  F  G  H  I  J  K  L  M  N  O  P  Q  R  S  T  U  V  W  X  Y  Z |  Other Symbols ]
  By Language
all Classics books content using ISYS

Download this book: [ ASCII | HTML | PDF ]

Look for this book on Amazon


We have new books nearly every day.
If you would like a news letter once a week or once a month
fill out this form and we will give you a summary of the books for that week or month by email.

Title: Explorations in Australia - The Journals of John McDouall Stuart
Author: Stuart, John McDouall, 1815-1866
Language: English
As this book started as an ASCII text book there are no pictures available.


*** Start of this LibraryBlog Digital Book "Explorations in Australia - The Journals of John McDouall Stuart" ***


EXPLORATIONS IN AUSTRALIA.


THE JOURNALS

OF

JOHN McDOUALL STUART

DURING THE YEARS

1858, 1859, 1860, 1861, & 1862,

WHEN HE FIXED THE CENTRE OF THE CONTINENT AND
SUCCESSFULLY CROSSED IT FROM SEA TO SEA.


EDITED FROM MR. STUART'S MANUSCRIPT
BY WILLIAM HARDMAN, M.A., F.R.G.S., &c.

With Maps, a Photographic Portrait of Mr. Stuart, and twelve Engravings
drawn on wood by George French Angas, from Sketches taken during
the different expeditions.

(SANS CHANGER.
S.O. AND CO.)

SECOND EDITION.


1865.


ADVERTISEMENT

TO THE

SECOND EDITION.

Since the first edition of this work was published Mr. Stuart has arrived
in England, and at a recent meeting of the Geographical Society he
announced that, taking advantage of his privilege as a discoverer, he had
christened the rich tract of country which he has opened up to the South
Australians Alexandra Land.

December 1st, 1864.


PREFACE BY THE EDITOR.

The explorations of Mr. John McDouall Stuart may truly be said, without
disparaging his brother explorers, to be amongst the most important in
the history of Australian discovery. In 1844 he gained his first
experiences under the guidance of that distinguished explorer, Captain
Sturt, whose expedition he accompanied in the capacity of draughtsman.
Leaving Lake Torrens on the left, Captain Sturt and his party passed up
the Murray and the Darling, until finding that the latter would carry him
too far from the northern course, which was the one he had marked out for
himself, he turned up a small tributary known to the natives as the
Williorara. The water of this stream failing him, he pushed on over a
barren tract, until he suddenly came upon a fruitful and well-watered
spot, which he named the Rocky Glen. In this picturesque glen they were
detained for six months, during which time no rain fell. The heat of the
sun was so intense that every screw in their boxes was drawn, and all
horn handles and combs split into fine laminae. The lead dropped from
their pencils, their finger-nails became as brittle as glass, and their
hair, and the wool on their sheep, ceased to grow. Scurvy attacked them
all, and Mr. Poole, the second in command, died. In order to avoid the
scorching rays of the sun, they had excavated an underground chamber, to
which they retired during the heat of the day.

When the long-expected rain fell, they pushed on for fifty miles to
another suitable halting-place, which was called Park Depot. From this
depot Captain Sturt made two attempts to reach the Centre of the
continent. He started, accompanied by four of his party, advancing over a
country which resembled an ocean whose mighty billows, fifty or sixty
feet high, had become suddenly hardened into long parallel ridges of
solid sand. The abrupt termination of this was succeeded at two hundred
miles by what is now so well known as Sturt's Stony Desert, to which
frequent allusion is made by Mr. Stuart in his journals. After thirty
miles more, this stony desert ceased with equal abruptness, and was
followed by a vast plain of dried mud, which Captain Sturt describes as
"a boundless ploughed field, on which floods had settled and subsided."
After advancing two hundred miles beyond the Stony Desert, and to within
one hundred and fifty miles of the Centre of the continent, they were
compelled to return to Park Depot, where they arrived in a most exhausted
condition.

A short rest at the Depot was followed by another expedition, Captain
Sturt being on this occasion accompanied by Mr. Stuart and two men. The
seventh day of their journey brought them to the banks of a fine creek,
now so well known as Cooper Creek in connection with the fate of those
unfortunate explorers, Burke and Wills. At two hundred miles from Cooper
Creek Captain Sturt and his party were again met by the Stony Desert, but
slightly varied in its aspect. Before abandoning his attempt to proceed,
the leader of the expedition laid the matter before his companions, and
he writes as follows: "I should be doing an injustice to Mr. Stuart and
my men, if I did not here mention that I told them the position we were
placed in, and the chance on which our safety would depend if we went on.
They might well have been excused if they expressed an opinion contrary
to such a course; but the only reply they made me was to assure me that
they were ready and willing to follow me to the last."

With much reluctance, however, Captain Sturt determined to return to
Cooper Creek without delay. They travelled night and day without
interruption, and on the morning of their arrival at the creek, one of
those terrible hot north winds, so much dreaded by the colonists, began
to blow with unusual violence. Lucky was it for them that it had not
overtaken them in the Desert, for they could scarcely have survived it.
The heat was awful; a thermometer, graduated to 127 degrees, burst,
though sheltered in the fork of a large tree, and their skin was
blistered by a torrent of fine sand, which was driven along by the fury
of the hurricane. They still had fearful difficulties to encounter, but
after an absence of nineteen months they returned safely to Adelaide.

The discouraging account of the interior which was brought by Captain
Sturt did not prevent other explorers from making further attempts; but
the terrible fate of Kennedy and his party on York Peninsula, and the
utter disappearance of Leichardt's expedition, both in the same year
(1848), had a very decided influence in checking the progress of
Australian exploration. Seven years later, in 1855, Mr. Gregory landed on
the north-west coast for the purpose of exploring the Victoria River, and
after penetrating as far south as latitude 20 degrees 16 minutes,
longitude 131 degrees 44 minutes, he was compelled to proceed to the head
of the Gulf of Carpentaria, and thence to Sydney along the route taken by
Dr. Leichardt in 1844. Shortly after his return Mr. Gregory was
despatched by the Government of New South Wales in 1857, to find, if
possible, some trace of the lost expedition of the lamented Leichardt;
his efforts, however, did nothing to clear up the mystery that enshrouds
the fate of that celebrated explorer.* (* It is possible that Mr.
McKinlay has been hasty in the opinion he formed from the graves and
remains of white men shown to him by Keri Keri, and the story related of
their massacre. May they not belong to Leichardt's party?)

The colonists of South Australia have always been distinguished for
promoting by private aid and public grant the cause of exploration. They
usually kept somebody in the field, whose discoveries were intended to
throw light on the caprices of Lake Torrens, at one time a vast inland
sea, at another a dry desert of stones and baked mud. Hack, Warburton,
Freeling, Babbage, and other well-known names, are associated with this
particular district, and, in 1858, Stuart started to the north-west of
the same country, accompanied by one white man (Forster) and a native. In
this, the first expedition which he had the honour to command, he was
aided solely by his friend Mr. William Finke, but in his later journeys
Mr. James Chambers also bore a share of the expense.* (* It is greatly to
be regretted that both these gentlemen are since dead. Mr. Chambers did
not survive to witness the success of his friend's later expeditions, and
the news of Mr. Finke's death reached us while these sheets were going
through the press.) This journey was commenced in May, 1858, from Mount
Eyre in the north to Denial and Streaky Bays on the west coast of the
Port Lincoln country. On this journey Mr. Stuart accomplished one of the
most arduous feats in all his travels, having, with one man only (the
black having basely deserted them), pushed through a long tract of dense
scrub and sand with unusual rapidity, thus saving his own life and that
of his companion. During this part of the journey they were without food
or water, and his companion was thoroughly dispirited and despairing of
success. This expedition occupied him till September, 1858, and was
undertaken with the object of examining the country for runs. On his
return the South Australian Government presented him with a large grant
of land in the district which he had explored.

Mr. Stuart now turned his attention to crossing the interior, and, with
the assistance of his friends Messrs. Chambers and Finke, he was enabled
to make two preparatory expeditions in the vicinity of Lake Torrens--from
April 2nd to July 3rd, 1859, and from November 4th, 1859, to January
21st, 1860. The fourth expedition started from Chambers Creek (discovered
by Mr. Stuart in 1858, and since treated as his head-quarters for
exploring purposes), on March 2nd, 1860, and consisted of Mr. Stuart and
two men, with thirteen horses. Proceeding steadily northwards, until the
country which his previous explorations had rendered familiar was left
far behind, on April 23rd the great explorer calmly records in his
Journal the following important announcement: "To-day I find from my
observations of the sun that I am now camped in the CENTRE OF AUSTRALIA."
One of the greatest problems of Australian discovery was solved! The
Centre of the continent was reached, and, instead of being an
inhospitable desert or an inland sea, it was a splendid grass country
through which ran numerous watercourses.

Leaving the Centre, a north-westerly course was followed, but, after
various repulses, a north-easterly course eventually carried the party as
far as latitude 18 degrees 47 minutes south, longitude 134 degrees, when
they were driven back by the hostility of the natives. As has already
been stated, Mr. Gregory in 1855, starting from the north-west coast, had
penetrated to the south as low as latitude 20 degrees 16 minutes,
longitude 127 degrees 35 minutes. Mr. Stuart had now reached a position
about half-way between Gregory's lowest southward point and the head of
the Gulf of Carpentaria. Without actually reaching the country explored
by Gregory, he had overlapped his brother explorer's position by one
degree and a half, or more than one hundred miles, and was about two
hundred and fifty miles in actual distance from the nearest part of the
shores of the Gulf. It is important to remark that the attack of the
savages which forced Mr. Stuart to return occurred on June 26th, 1860, so
that he had virtually crossed the continent two months before Messrs.
Burke and Wills had left Melbourne.* (* They did not leave Cooper Creek
until December 14th, rather more than a fortnight before Mr. Stuart
started on his fifth expedition.)

On New Year's day 1861, Mr. Stuart again left Adelaide, aided this time
by a grant from the Colonial Government of 2500 pounds, in addition to
the assistance of his well-tried friends Messrs. Chambers and Finke. He
made his former position with ease, and advanced about one hundred miles
beyond it, to latitude 17 degrees, longitude 133 degrees; but an
impenetrable scrub barred all further progress, and failing provisions,
etc., compelled him, after such prolonged and strenuous efforts that his
horses on one occasion were one hundred and six hours without water, most
reluctantly to return. The expedition arrived safely in the settled
districts in September, and the determined explorer, after a delay of
less than a month, was again despatched by the South Australian
Government along what had now become to him a familiar road. This time
success crowned his efforts; a passage was found northwards through the
opposing scrub, and leaving the Gulf of Carpentaria far to the right, the
Indian Ocean itself was reached. Other explorers had merely seen the rise
and fall of the tide in rivers, boggy ground and swamps intervening and
cutting off all chance of ever seeing the sea. But Stuart actually stood
on its shore and washed his hands in its waters! What a pleasure it must
have been to the leader when, knowing well from his reckoning that the
sea must be close at hand, but keeping it a secret from all except Thring
and Auld, he witnessed the joyful surprise of the rest of the party!

The expedition reached Adelaide safely, although for a long time the
leader's life was despaired of, the constant hardships of so many
journeys with scarcely any intermission having brought on a terrible
attack of scurvy. The South Australian Government in 1859 liberally
rewarded Mr. Stuart and his party for their successful enterprise.* (*
Mr. Stuart's qualities as a practised Bushman are unrivalled, and he has
always succeeded in bringing his party back without loss of life.) On the
10th of March a resolution was passed to the effect that a sum of 3500
pounds should be paid as a reward to John McDouall Stuart, Esquire, and
the members of his party, in the following proportions: Mr. Stuart 2000
pounds; Mr. Keckwick 500 pounds; Messrs. Thring and Auld 200 pounds each;
and Messrs. King, Billiatt, Frew, Nash, McGorrerey, and Waterhouse, 100
pounds each. Perhaps this is the most fitting place to express Mr.
Stuart's appreciation of the honour done him by the Royal Geographical
Society of London, in awarding him their gold medal and presenting him
with a gold watch. He wishes particularly to express his hearty thanks to
Sir Roderick Murchison, and the other distinguished members of the
society, for the lively interest they have evinced in his welfare.

Mr. Stuart's experiences have led him to form a very decided opinion as
to the cause of the well-known hot winds of Australia, so long the
subject of scientific speculation. North and north-west of Flinders Range
are large plains covered with stones, extending as far as latitude 25
degrees. To the north of that, although the sun was intensely hot, there
were no hot winds; in fact from that parallel of latitude to the Indian
Ocean, either going or returning, they were not met with. "On reaching
latitude 27 degrees on my return," writes Mr. Stuart, "I found the hot
winds prevailing again as on my outward journey. I saw no sandy desert to
which these hot winds have been attributed, but, on lifting some of the
stones that were lying on the surface,* I found them so hot that I was
obliged to drop them immediately. (* On the surface, as I suppose, of the
large plains North of Flinders Range. ED.) It is my opinion that when a
north wind blows across those stone-covered plains, it collects the heat
from them, and the air, becoming rarified, is driven on southwards with
increased vehemence. To the north of latitude 25 degrees, although
exposure to the sun in the middle of the day was very oppressive, yet the
moment we got under the shade of a tree we felt quite alive again; there
was none of that languid feeling which is experienced in the south during
a hot wind, as for example that which blew on the morning after reaching
the Hamilton,* in latitude 26 degrees 40 minutes. (* Journal 1861 to
1862.) That was one of the hottest winds I ever experienced. I had the
horses brought up at 7 o'clock, intending to proceed, but seeing there
was a very hot wind coming on, I had them turned out again. It was well I
did so, for before 10 o'clock all the horses were in small groups under
the trees, and the men lying under the shade of blankets unable to do
anything, so overpowering was the heat." Unfortunately, Mr. Stuart had no
thermometer.

Mr. Stuart is anxious to direct attention to the establishment of a
Telegraph line along his route. On this subject he writes as follows:--

"On my arrival in Adelaide from my last journey I found a great deal of
anxiety felt as to whether a line could be carried across to the mouth of
the Adelaide river. There would be a few difficulties in the way, but
none which could not be overcome and made to repay the cost of such an
undertaking. The first would be in crossing from Mr. Glen's station to
Chambers Creek, in finding timber sufficiently long for poles, supposing
that no more favourable line than I travelled over could be adopted, but
I have good reason for supposing that there is plenty of suitable timber
in the range and creek, not more than ten miles off my track: the
distance between the two places is one hundred miles. From Chambers Creek
through the spring country to the Gap in Hanson Range the cartage would
be a little farther, in consequence of the timber being scarce in some
places. There are many creeks in which it would be found, but I had not
time to examine them in detail. Another difficulty would be in crossing
the McDonnell Range, which is rough and ragged, but there is a great
quantity of timber in the Hugh; the distance to this in a straight line
is not more than seven miles; from thence to the Roper River there are a
few places where the cartage might be from ten to twenty miles, that is
in crossing the plains where only stunted gum-trees grow, but tall timber
can be obtained from the rising ground around them. From latitude 16
degrees 30 minutes south to the north coast, there would be no difficulty
whatever, as there is an abundance of timber everywhere. I am promised
information, through the kindness of Mr. Todd, of the Telegraph
department, as to the average cost of establishing the lines through the
outer districts of this colony, and it is my intention to make a
calculation of the cost of a line on my route, by which the comparative
merits and expense will be tested, and I am of opinion I shall be able to
show most favourable results. I should have been glad for this
information to have accompanied my works, but I find I cannot postpone
them longer for that purpose, as parties have already taken advantage of
the delay occasioned by my illness at the time of, and since, my arrival
home to collect what scraps of information they could obtain, with the
intention of publishing them as my travels. I leave the reward of such
conduct to a discriminating public; I shall not fail to carry out my
intention with regard to a Telegraph line; and should I have no
opportunity of submitting it to the public, I shall take care to advance
the matter in such channels as may be most likely to lead to a successful
issue. I beg reference to my map accompanying this work, which will at
once show the favourable geographical situation of the Adelaide River for
a settlement, and the short and safe route it opens up for communication
and trading with India: indeed when I look upon the present system of
shipping to that important empire, I cannot over-estimate the advantages
that such an extended intercourse would create."

Mr. Stuart is also very anxious for the formation of a new colony on the
scene of his discoveries on the River Adelaide, and would fain have been
one of the first pioneers of such an enterprise, but his health has been
so much shattered by his last journey that he can only now hope to see
younger men follow in the path which he had made his own. He writes as
follows:--

"Judging from the experience I have had in travelling through the
Continent of Australia for the last twenty-two years, and also from the
description that other explorers have given of the different portions
they have examined in their journeys, I have no hesitation in saying,
that the country that I have discovered on and around the banks of the
Adelaide River is more favourable than any other part of the continent
for the formation of a new colony. The soil is generally of the richest
nature ever formed for the benefit of mankind: black and alluvial, and
capable of producing anything that could be desired, and watered by one
of the finest rivers in Australia. This river was found by Lieutenant
Helpman to be about four to seven fathoms deep at the mouth, and at one
hundred and twenty miles up (the furthest point he reached) it was found
to be about seven fathoms deep and nearly one hundred yards broad, with a
clear passage all the way up. I struck it about this point, and followed
it down, encamping fifteen miles from its mouth, and found the water
perfectly fresh, and the river broader and apparently very deep; the
country around most excellent, abundantly supplied with fresh water,
running in many flowing streams into the Adelaide River, the grass in
many places growing six feet high, and the herbage very close--a thing
seldom seen in a new country. The timber is chiefly composed of
stringy-bark, gum, myall, casurina, pine, and many other descriptions of
large timber, all of which will be most useful to new colonists. There is
also a plentiful supply of stone in the low rises suitable for building
purposes, and any quantity of bamboo can be obtained from the river from
two to fifty feet long. I measured one fifteen inches in circumference,
and saw many larger. The river abounds in fish and waterfowl of all
descriptions. On my arrival from the coast I kept more to the eastward of
my north course, with the intention of seeing further into the country. I
crossed the sources of the running streams before alluded to, and had
great difficulty in getting more to the west. They take their rise from
large bodies of springs coming from extensive grassy plains, which proves
there must be a very considerable underground drainage, as there are no
hills of sufficient elevation to cause the supply of water in these
streams. I feel confident that, if a new settlement is formed in this
splendid country, in a few years it will become one of the brightest gems
in the British Crown. To South Australia and some of the more remote
Australian colonies the benefits to be derived from the formation of such
a colony would be equally advantageous, creating an outlet for their
surplus beef and mutton, which would be eagerly consumed by the races in
the Indian Islands, and payment made by the shipment of their useful
ponies, and the other valuable products of those islands; indeed I see
one of the finest openings I am aware of for trading between these
islands and a colony formed where proposed."

Mr. Stuart was accompanied on his last journey by Mr. Waterhouse, a
clever naturalist, whose report to the Commissioner of Crown Lands of
South Australia, although too long for insertion here, is full of most
interesting information. Unfortunately, the interests of geographical
science were apparently lost sight of in the hurry to effect the grand
object of the expedition, namely, to cross from sea to sea. Thermometers
were forgotten; two mounted maps of the country from Chambers Creek to
Newcastle Water, in a tin case, never came to hand, and the expedition
was provided with no means of estimating even the approximate height of
the elevated land or of the mountains in the interior. As Mr. Waterhouse
remarks: "The thermometers were much needed, as it would have been very
desirable to have kept a register of the temperature, and to have tested
occasionally the degree of heat at which water boiled on the high table
lands. The loss of the maps prevented my marking down at the time on the
maps the physical features of the country, and the distribution of its
fauna and flora."

Mr. Waterhouse divides the country into three divisions. The first, which
extends from Goolong Springs to a little north of the Gap in Hanson
Range, latitude 27 degrees 18 minutes 23 seconds, may be called the
spring and saltbush country. The second division commences north of the
Gap in Hanson Range, and extends to the southern side of Newcastle Water,
latitude 17 degrees 36 minutes 29 seconds. It is marked by great scarcity
of water--in fact, there are few places where water can be relied on as
permanent--and also by the presence of the porcupine grass (Triodia
pungens of Gregory, and Spinifex of Stuart), which is the prevailing
flora. The third division commences from the north end of Newcastle
Water, latitude 17 degrees 16 minutes 20 seconds, and extends to Van
Diemen Gulf, latitude 12 degrees 12 minutes 30 seconds; it comprises a
large part of Sturt Plains, with soil formed of a fine lacustrine
deposit, the valleys of the Roper filled with a luxuriant tropical
vegetation, and thence to the Adelaide River and the sea-coast.

On visiting Hergott Springs, Mr. Waterhouse learnt that Mr. Burtt, whose
station* is only a few miles distant, in opening these springs discovered
some fossil bones, casts of which were forwarded to Professor Owen, who
pronounced them to be the remains of a gigantic extinct marsupial, named
Diprotodon Australis. (* Hergott Springs were only discovered and named
by Stuart three years before, yet we now find a station close by them.
The explorer is not far ahead of his fellow-colonists, as is well
remarked by the Edinburgh Review for July, 1862: "Australian occupation
has kept close on the heels of Australian discovery.") Bones of this
animal have also been found in a newer tertiary formation in New South
Wales. Mr. Waterhouse considers that a great tertiary drift extends over
this part of the country, obscuring and concealing at no great depth
below the surface many springs, which may hereafter be discovered as the
country becomes better known.

The Louden Spa is a hot spring arising out of a small hillock, and
proceeds from the fissures of volcanic rock. This water is medicinal, but
not disagreeable to the taste: the damper made with it was very light,
and tasted like soda-bread.

In his remarks on the second division Mr. Waterhouse states much that is
valuable. He estimates the height of Mount Hay at two thousand feet,
regarding it as the highest point of the McDonnell Range, which is the
natural centre of this part of the continent. Mr. Waterhouse only saw
Chambers Pillar from a distance, but he had an opportunity of examining a
smaller hill of the same character, and found it to be composed of a soft
loose argillaceous rock, at the top of which was a thin stratum of a hard
siliceous rock, much broken up. "The isolated hills appear to have been
at some remote period connected, but from the soft and loose nature of
the lower rock meeting with the action of water, had arisen a succession
of landslips. These have been washed away and others have followed in
their turn; the upper rock, from being undermined, has fallen down and
broken up, supplying the peculiar siliceous stones so widely distributed
on parts of the surface of the country."

The vegetation of this district is poor; the myall is scarce, but the
mulga (Acacia aneura) generally plentiful. Both these shrubs are species
of acacia, the myall being of much larger growth and longer lived than
the mulga. Nutritious grass is seldom found except in the immediate
vicinity of the creeks, and the scrubs are very extensive.

Mr. Waterhouse collected a great number of specimens of natural history,
but, from want of the convenience for carrying them, many of the more
delicate objects were broken.

In the Appendix will be found some remarks by Mr. John Gould, F.R.S.,
etc., on the birds collected by Mr. Waterhouse during Mr. Stuart's
expedition, including a description of a new and beautiful parrakeet.
There are also descriptions of new species of Freshwater Shells from the
same expedition, by Mr. Arthur Adams, F.L.S., and Mr. G. French Angas, to
the skill of which latter gentleman this work is indebted for its
admirable illustrations.

Dr. Muller, the Government Botanist, Director of the Botanic Garden at
Melbourne, in his report to both Houses of the Legislature of Victoria,
April 15th, 1863, says, "A series of all the plants collected during Mr.
J.M. Stuart's last expedition was presented by the Hon. H. Strangways,
Commissioner of Crown Lands for South Australia, and those of the former
expeditions of that highly distinguished explorer, by the late J.
Chambers, Esquire, of North Adelaide." Of this collection, Dr. Muller has
furnished a systematic enumeration, which will be found in the Appendix.
This enumeration must not, however, be accepted as final, for Dr. Muller
has forwarded all the specimens to England for the inspection of Mr.
Bentham, the learned President of the Linnaean Society of London, who is
now elaborating his great and exhaustive work on the Flora of Australia,
the second volume of which will shortly be before the public.

WILLIAM HARDMAN.



CONTENTS.

JOURNAL OF MR. STUART'S EXPEDITION TO THE NORTH-WEST. MAY TO SEPTEMBER,
1858.

JOURNAL OF MR. STUART'S SECOND EXPEDITION (IN THE VICINITY OF LAKE
TORRENS). APRIL TO JULY, 1859.

JOURNAL OF MR. STUART'S THIRD EXPEDITION (IN THE VICINITY OF LAKE
TORRENS). NOVEMBER, 1859, TO JANUARY, 1860.

JOURNAL OF MR. STUART'S FOURTH EXPEDITION (FIXING THE CENTRE OF THE
CONTINENT). FROM MARCH TO SEPTEMBER, 1860.

JOURNAL OF MR. STUART'S FIFTH EXPEDITION. FROM NOVEMBER, 1860, TO
SEPTEMBER, 1861.

JOURNAL OF MR. STUART'S SUCCESSFUL EXPEDITION ACROSS THE CONTINENT OF
AUSTRALIA. FROM DECEMBER, 1861, TO DECEMBER, 1862.


APPENDIX.


(LIST OF ILLUSTRATIONS.

PORTRAIT OF JOHN MACDOUALL STUART. Adelaide, April 1863. Professor Hall.
Photograph.

SKETCH MAP OF AUSTRALIA, SHOWING THE POSITION OF MR. STUART'S ROUTE.

NORTH-WEST OF STUART'S CREEK.

NORTH-WEST POINT OF LAKE TORRENS.

THE HERMIT HILL AND FINNISS SPRINGS.

ELIZABETH SPRINGS.

SOUTH SHORE OF LAKE EYRE.

CHAMBERS PILLAR.

CENTRAL MOUNT STUART.

ATTACK CREEK.

BRINKLEY BLUFF.

CHAMBERS CREEK.

TABLE LAND AND VALLEY OF THE ADELAIDE.

PLANTING THE FLAG ON THE SHORES OF THE INDIAN OCEAN.)



EXPLORATIONS IN AUSTRALIA.



JOURNAL OF MR. STUART'S EXPEDITION TO THE NORTH-WEST. MAY TO SEPTEMBER,
1858.

On the 14th of May, 1858, Mr. Stuart started from Oratunga (the head
station of Mr. John Chambers), accompanied by Mr. Barker, with six
horses, and all that was requisite (with one important exception, as will
be seen hereafter), for an excursion to the north-west of Swinden's
Country. They arrived at Aroona the same evening. On the following day
(the 15th) they made Morleeanna Creek, and reached Ootaina on the 16th,
about 7 p.m. Here they remained for a couple of days, as sufficient rain
had not fallen to enable them to proceed. On the afternoon of the 19th
they arrived at Mr. Sleep's, who informed them that Mr. M. Campbell had
returned from the West, being hard pushed for water; very little rain
having fallen to the west. The next day (20th) Mr. Stuart arrived at Mr.
Louden's, but, in consequence of some difficulties about the horses, he
returned to Ootaina. Various preparations, combined with want of rain,
compelled him to delay his start until the 10th of June. Here the journal
commences:--

Thursday, 10th June, 1858. Started from Ootaina at 1 p.m. for Beda.
Camped on the plain, about thirteen miles from Mount Eyre.

Friday, 11th June, West Plain. Made Mudleealpa at 11 a.m. The horses
would not drink the water. Proceeded for about five miles towards Beda.
The plains are fearfully dry; they have the appearance as if no rain had
fallen here for a long time, and I am very much afraid there will be no
water at Beda. If such should be the case, the horses will suffer too
much in the beginning of their journey to be without a drink to-night. I
think it will be best to return to Mudleealpa, leave our saddles,
rations, etc. there, and drive the horses back to water. I sent Mr.
Forster back with them, telling him if he can find no water between this
and Mr. Sleep's, to take them there, remain for the night, give them a
drink in the morning, and return; we shall then be able to make a fresh
start to-morrow. Bearings: Mount Arden, 154 degrees 30 minutes; Mount
Eyre, 77 degrees 30 minutes; Beda Hill, 272 degrees; Mount Elder, 64
degrees 50 minutes; Dutchman's Stern, 162 degrees 15 minutes.

Saturday, 12th June, Mudleealpa. In examining the creek a little higher
up, we found another well. By cleaning it out, the water is drinkable.
The horses did not arrive until it was too late to start, and having
water here now, that they can drink, we camped here another night.

Sunday, 13th June, Mudleealpa. Started for Beda. Some of the horses would
not drink the water, and others drank very little: they will be glad to
drink far worse than this before they come back, or I am much mistaken.
Arrived at Beda at sundown. I was right in my opinion; no fresh water to
be found; nothing but salt, salter than the sea. I can see nothing of Mr.
Babbage's* encampment; he must be higher up the creek. All the country we
have come over to-day is very dry. (* It will probably be recollected
that Mr. Babbage was sent out by the Government to make a north-west
course through the continent, but, when at the Elizabeth, he made an
unaccountable detour, and found himself at Port Augusta, his original
starting-point. On my return from this journey he called on me at Mount
Arden, when I furnished him with such information as he required, and he
again started, and made Chambers' Creek, which I had previously found and
named after my old friend, Mr. James Chambers, but which he called
Stuart's Creek in acknowledgment of my information, etc. J. McD. Stuart.)

Monday, 14th June, Beda. This morning we have searched all round, but can
find no fresh water, although there are numerous places that would retain
water if any quantity had fallen. Mr. Forster, whom I had sent up the
creek to Mr. Babbage's, to inquire if there was any water at Pernatta,
has returned with the information that Mr. B. was up there with all his
horses, and that there was still a little water, but not much. Started at
11.30 a.m. for that place; camped in the sand hills one hour after dark.
Here we found some pig-faces* which the horses eat freely. (* These
pig-faces belong to the Mesembryaceae, of which the common ice-plant of
our gardens is an example.) There is a great deal of moisture in them,
and they are a first-rate thing for thirsty horses; besides, they have a
powerful diuretic effect. I was unable to fix Beda Hill, all my time
being taken up in looking for water, but I hope to get its position at
Pernatta. The country was very heavy--sand hills.

Tuesday, 15th June, Sand Hills. Started at break of day for Pernatta.
About 10 a.m. met Mr. Babbage's two men returning with some of the horses
for rations. They informed me that the water was nearly all gone, but
that there was plenty in the Elizabeth, nineteen miles from Pernatta. I
intended to keep on the track, but our black insisted that Pernatta lay
through a gap, and not round the bluff. I allowed him to have his own
way. Our route was through a very stony saddle. When there we saw a gum
creek, and made for it; when we arrived at the creek he told us that was
Pernatta. We looked for water, and found a little hole, which, to our
great disappointment, contained salt water. Could see nothing of Mr.
Babbage's camp. I then asked our black where there was another water; he
said, "Down the creek," which we followed. He took us to five or six
water holes, with native names, every one dry. The last one he called
Yolticourie. It being now within an hour of sundown, I would follow him
no longer, but unsaddled, and told Mr. Forster to take the black and the
horses, and to steer for the bluff; if he found no water between, to
intersect Mr. Babbage's tracks, and follow them up and get water. I
remained with our provisions. The black fellow evidently does not know
the country. I am sorry that I have taken him with me. I think I shall
send him back; he is of little use in assisting to get the horses in the
morning.

Wednesday, 16th June, Yolticourie. The horses have returned; they found
no water last night; they were obliged to camp for the night, it being so
dark, but they found Mr. Babbage's camp very early. The horses drank all
the water. I was wrong in blaming the black fellow; he took us to the
RIGHT Pernatta. It is another water that Mr. B. is encamped at. He moves
to-day for the Elizabeth, which I also will do. He found the remains of
poor Coulthard yesterday. We must have passed quite close to them in our
search for water. He has sent for me to come and assist at the burial. It
being so late in the day (12 o'clock), and the horses requiring more
water, and he having four men besides himself, I do not see that I can be
of any use, and it might cause me to lose another day, and the horses to
be another night without water, which would be an injury to them, they
not having had sufficient this morning. Mr. B. also sent to say that he
would accompany me to the Elizabeth. I have delayed an hour for him, and
he has not yet made his appearance; it being now 1 o'clock, and having to
travel seventeen miles, I can wait no longer. Started for Bottle Hill;
arrived on the south side of the hill an hour and a half before sundown,
found some water and plenty of grass; encamped for the night. Distance
to-day, seventeen miles. The former part of the journey was over very
stony country; the latter part very heavy sand hills.

Thursday, 17th June, Bottle Hill. Got on the top of Bottle Hill to take
bearings, but was disappointed; could see no hill except one, which was
either Mount Deception or Mount North-west; the bearing was 51 degrees 30
minutes. There is a small cone of stones on the top, and a flat stone on
the top of it, with the names of Louden and Burtt. From here I saw the
gum trees in the Elizabeth; course to them 325 degrees 30 minutes, seven
miles to the creek. The country from the hill here is of the very worst
description--nothing but sand and salt bush.

Friday, 18th June, The Elizabeth. We must rest our horses to-day, they
have not yet recovered from their long thirst. I am quite disappointed
with this creek and the surrounding country. The water is not permanent,
it is only rain water; since we arrived yesterday it has shrunk a great
deal. There are small plains on each side from a quarter to half a mile
broad with salt bush; the hills are very stony with a little salt bush,
and destitute of timber, except the few gum-trees in the creek and the
mulga bushes in the sand hills.

Saturday, 19th June, The Elizabeth. The sky was quite overcast with cloud
during the night, and a few drops of rain fell, but of no consequence.
Started at 9.30 a.m., on a bearing of 308 degrees for six miles; changed
the bearing to 355 degrees for one mile and a half; next bearing 328
degrees for four miles, to the north side of a dry swamp; next bearing 4
degrees for ten miles and a half; next bearing 350 degrees for four miles
to a sand hill. Camped. Distance to-day, twenty-five miles, over a very
bad country, with large fragments of a hard flinty stone covering the
surface. Salt bush with small sand hills. No water.

Sunday, 20th June, Sand Hill. Started at 9 a.m., on a course of 25
degrees for sixteen miles. At 1 p.m., came upon a creek, in which I
thought there might be water; examined it and found two water holes, with
plenty of grass upon their banks. The water is not permanent. Our course
to-day has been across stony plains (covered on the surface with
fragments resembling hard white quartz), with sand hills about two miles
broad dividing them. The black did not know of this water; I am very
doubtful of his knowing anything of the country. The stony plains are
surrounded by high heavy sand hills, especially to the west and
north-west; I dare not attempt to get through them without rain. They are
much higher than the country that I am travelling through. It seems as if
there had been no rain for twelve months, every thing is so dried and
parched up. On further examination of the creek we have found a large
hole of clear water, with rushes growing round it; I almost think it is
permanent, and intend to run the risk of falling back upon it should I be
forced to retreat and wait for rain. The creek seems to drain the large
stony plains that we crossed; the water is three and a half feet deep,
ten yards wide, by forty yards long.

Monday, 21st June, Water Creek. Started at 9.30 a.m. on a course of 25
degrees. At a mile passed a small table-topped hill to the west of our
line; at three miles and a half crossed the creek; at four miles passed
another table-topped hill connected with the low range to the east, and
passed the first ironstone hill; at seven miles changed to 55 degrees; at
eight miles halted at a large permanent water hole (Andamoka). I can with
safety say that this is permanent; it is a splendid water hole, nearly as
large as the one at the mouth of the gorge in the John. The low range to
the east of our course, and running nearly parallel with it, is composed
of conglomerate, quartz, and a little ironstone. Part of to-day's journey
was over low undulating sandy and very well grassed country. There seems
to have been a little rain here lately; the grass is springing
beautifully. At eleven miles we came upon a salt lagoon (Wealaroo) two
miles long by one broad. From the north end of it, on a bearing of 55
degrees, one mile and a half will strike Andamoka. I think we have now
left the western sand hills behind us; and now that we have permanent
water to fall back on, I shall strike into the north-west to-morrow. The
distance travelled to-day was fifteen miles. The country around this
water consists of bold stony rises and sand, with salt bush and grass; no
timber except mulga and a few myall bushes in the creek. On an
examination of the creek, we have found salt water above and below this
hole. In one place above there are cakes of salt one inch and a half
thick, a convincing proof that this is supplied by springs.

Tuesday, 22nd June, Andamoka. Started on a bearing of 342 degrees. At
seven miles and a half, crossed a low stony range running east-north-east
and west-south-west, which turned out to be table land, with sand hills
crossing our line, bearing to a high range east of us 93 degrees 30
minutes. About eight miles in the same direction there is the appearance
of a long salt lake. At nine miles and a half, on a sand hill, I obtained
the following bearings: Mount North-west, 60 degrees 30 minutes; Mount
Deception, 95 degrees. At eleven miles and a half passed a large reedy
swamp on our left, dry. At seventeen miles sand hills ceased. At eighteen
miles and a half the sand hills again commenced, and we changed our
course to north for three miles. Camped for the night at a creek of
permanent water, very good. The last four miles of to-day's journey have
been over very stony rises with salt bush and a little grass.

Wednesday, 23rd June, Permanent Water Creek. The horses had strayed so
far that we did not get a start until 10 a.m. Bearing to-day, 318
degrees. At two miles crossed a tea-tree creek, in which there is water,
coming from the stony rises, and running to the north of east. At six
miles the sand hills again commence. To this place we have come over a
stony plain, covered on the surface with fragments of limestone, quartz,
and ironstone, with salt bush and grass. In a watery season it must be
well covered with grass; the old grass is lying between the salt bushes.
We have a view of part of the lake (Torrens) bearing north-east about
fifteen or twenty miles from us; to the west again the stony rises,
apparently more open. At ten miles, in the sand hills, we have again a
view of Flinders range. The bearings are: Mount North-west, 78 degrees 35
minutes; Mount Deception, 107 degrees. At fourteen and a half miles we
found a clay-pan of water, with beautiful green feed for the horses. As
we don't know when we shall find more water, and as Forster has a damper
to bake, I decide to camp for the rest of the day. Our route has lain
over heavy sand hills for the last eight miles.

Thursday, 24th June, Sand Hills. At 8.30 we left on a course of 340
degrees, commencing with about two miles of rather heavy sand hills. At
eight miles these sand hills diminished, and the valleys between them
became much wider--both sand hills and valleys being well covered with
grass and salt bush, with courses of lime and ironstone cropping out and
running east and west. At twelve miles changed our course to 79 degrees,
to examine a gum creek (Yarraout), which we ran down for water, but did
not obtain it before four miles, when we found a small hole of rain
water. This creek seems to be a hunting-ground of the natives, as we saw
a great many summer worleys on its banks. They had evidently been here
to-day, for, a little above where we first struck the creek, we saw some
smoke, but on following it up, we found they had gone; most likely they
had seen us and run away. The latter part of our journey to-day was over
a stony plain, bounded on the west by the stony table land with the sand
hills on the top. All this country seems to have been under water, and is
most likely the bed of Lake Torrens, or Captain Sturt's inland sea. In
travelling over the plains, one is reminded of going over a rough,
gravelly beach; the stones are all rounded and smooth. Distance to-day,
thirty miles.

Friday, 25th June, Yarraout Gum Creek. Started at 9.40 from the point
where we first struck the creek last night, bearing 20 degrees for two
miles, thence 61 degrees for one mile to a high sand hill, thence 39
degrees for one mile to a stony rise. My doubt of the black fellow's
knowledge of the country is now confirmed; he seems to be quite lost, and
knows nothing of the country, except what he has heard other blacks
relate; he is quite bewildered and points all round when I ask him the
direction of Wingillpin. I have determined to push into the westward,
keeping a little north of west. Bearing 292 degrees for five miles, sand
hills; thence 327 degrees to a table-hill nine miles. Camped without
water. Our route to-day has been through sand hills, with a few miles of
stones and dry reedy swamp, all well grassed, but no water. We came
across some natives, who kept a long distance off. I sent our black up to
them, to ask in which direction Wingillpin lay. They pointed to the
course I was then steering, and said, "Five sleeps." They would not come
near to us. About three-quarters of an hour afterwards I came suddenly
upon another native, who was hunting in the sand hills. My attention
being engaged in keeping the bearing, I did not observe him until he
moved, but I pulled up at once, lest he should run away, and called to
him. What he imagined I was I do not know; but when he turned round and
saw me, I never beheld a finer picture of astonishment and fear. He was a
fine muscular fellow, about six feet in height, and stood as if riveted
to the spot, with his mouth wide open, and his eyes staring. I sent our
black forward to speak with him, but omitted to tell him to dismount. The
terrified native remained motionless, allowing our black to ride within a
few yards of him, when, in an instant, he threw down his waddies, and
jumped up into a mulga bush as high as he could, one foot being about
three feet from the ground, and the other about two feet higher, and kept
waving us off with his hand as we advanced. I expected every moment to
see the bush break with his weight. When close under the bush, I told our
black to inquire if he were a Wingillpin native. He was so frightened he
could not utter a word, and trembled from head to foot. We then asked him
where Wingillpin was. He mustered courage to let go one hand, and
emphatically snapping his fingers in a north-west direction, again waved
us off. I take this emphatic snapping of his fingers to mean a long
distance. Probably this Wingillpin may be Cooper's Creek. We then left
him, and proceeded on our way through the sand hills. About an hour
before sunset, we came in full sight of a number of tent and table-topped
hills to the north-west, the stony table land being to the south of us,
and the dip of the country still towards Lake Torrens. I shall keep a
little more to the west to-morrow if possible, to get the fall of the
country the other way. The horses' shoes have been worn quite thin by the
stones, and will not last above a day or two. Nay, some of the poor
animals are already shoeless. It is most unfortunate that we did not
bring another set with us. Distance to-day, twenty-four miles.

Saturday, 26th June, Edge of Plain. Started at 9.30 a.m., on a bearing of
314 degrees 30 minutes, over an undulating plain, with low sand hills and
wide valleys, with plenty of grass and salt bush. After ten miles the
sand hills ceased, and at thirteen miles we reached the point of the
stony table land. Here we saw, to the north-north-west, what was
apparently a large gum creek, running north-east and south-west. Changing
our bearing to 285 degrees, after seven miles of very bad stony plain,
thinly covered with salt bush and grass, we came upon the creek, and
found long reaches of permanent water, divided here and there by only a
few yards of rocks, and bordered by reeds and rushes. The water hole, by
which we camped, is from forty to fifty feet wide, and half a mile in
length; the water is excellent, and I could see small fish in it about
two inches long. About ten miles down the creek the country seems to be
more open, and the gum-trees much larger, and in a distant bend of the
creek I can perceive a large body of water. The first of the seven or
eight tent-like hills that were to the east of our route to-day presents
a somewhat remarkable appearance. Of a conical form, it comes to a point
like a Chinaman's hat, and is encircled near the top by a black ring,
while some rocks resembling a white tower crown the summit. Distance
to-day, twenty miles.

Sunday, 27th June, Large Water Creek. Cloudy morning, with prospect of
rain. A swan visited the water hole last night, and to-day we have seen
both the mountain duck and the large black duck. Having a shoe to fix
upon Jersey, and my courses to map down, we did not get a start until 10
o'clock, and we were obliged to stop early in consequence of the grey
mare getting so lame that we were unable to proceed. We had an old shoe
or two, and Mr. Forster managed to get one on the mare. We started to-day
on a bearing of 270 degrees for eight miles to a low flat-topped hill,
when we changed to 220 degrees for five miles to a gum creek with rain
water. About five miles to the north of our line there are flat-topped
ranges, running north-east. The main creek runs on the south side of this
course, and nearly parallel to it. Further to the south, at a distance of
about ten miles, is still the stony table land with the sand hills. The
country is fearfully stony, but improves a little in grass as we get
west. It seems to be well watered. Distance to-day, about twelve miles.

Monday, 28th June, Gum Creek. There has been a little rain during the
night, and it is still coming down. As I am so far north, I regret that I
am unable to go a little further, fearing the lameness of the horses from
the stony nature of the country. I intend to follow the creek up, if it
comes from the west, or a little to the north of west, to see if I cannot
make the fall of the country to the south-west, and get on a better road
for the horses. We started on a bearing of 305 degrees, but after a mile
and a half, finding the creek wind too much to the north, we changed our
course to 287 degrees for five miles to a small flat-topped hill. Changed
our bearing again to 281 degrees for twenty-two miles to a tent hill, on
the south side of which we camped. This part of the country is very stony
and bad, with salt bush and very little grass. It has evidently been the
course of a large water at some time, and reminded me of the stony desert
of Captain Sturt. Bleak, barren, and desolate, it grows no timber, so
that we scarcely can find sufficient wood to boil our quart pot. The
rain, which poured down upon us all day, so softened the ground that the
horses could tread the stones into it, and we got along much better than
we expected. Distance to-day, twenty-eight miles and a half.

Tuesday, 29th June, South Side of Tent Hill. Started at 8.30 a.m. on a
bearing of 305 degrees. At eight miles crossed a gum creek, with
polyganum, running to the north. At twelve miles crossed another,
trending in the same direction. These creeks are wide and formed into
numerous channels. I expected to have done thirty miles to-day, but am
disappointed, for we were obliged to halt early, after having gone only
eighteen miles, as my horse was quite lame. How much do we feel the want
of another set of horse-shoes! We have, however, still got an old shoe
left, which is put on this afternoon. It had continued raining all last
night, but not heavily, and cleared off in the morning shortly after we
started. Our travelling to-day has been still very stony, over stony
rises; the stony table land that has been all along on our left is now
trending more to the south-west. The country is more open: in looking at
it from one of the rises it has the appearance of an immense plain,
studded with isolated flat-topped hills. The last eight miles is better
grassed and has more salt bush. Camped on a small creek in the stony
rises. Distance to-day, eighteen miles.

Wednesday, 30th June, Stony Rises. We had a little rain in the former
part of the night, and a very heavy dew in the morning. Started at 9.30
a.m., bearing 305 degrees; at five miles crossed the upper part of a gum
creek, and at twelve miles ascended a high flat-topped hill, commanding a
view of an immense stony plain, but it is so hazy that we can see nothing
beyond ten miles. From this hill we changed our course to 309 degrees to
a saddle in the next range. At four miles halted at a gum creek, with
plenty of green feed. Made a very short journey to-day in consequence of
the horses being quite lame. In addition to their want of shoes, a stiff,
tenacious brown clay adhered to the hoof, and picked up the small round
stones, which pressed on the frog of the foot. These pebbles were as
firmly packed as if they had been put in with cement, so that we had hard
work to keep the hoofs clear. Distance travelled, sixteen miles. Weather
showery.

Thursday, 1st July, Gum Greek. The horses have had such poor food for the
last week that I shall rest them to-day. About half a mile below us there
is a large water hole a quarter of a mile long, with a number of black
ducks upon it, but they are very shy. It rained very heavily and without
intermission all last night and to-day. This creek is visited by a great
many natives. We saw them making away as we approached.

Friday, 2nd July, Same Place. The creek came down last night: it is now a
sheet of water two hundred yards broad. Started at 8.45 a.m. over a stony
plain on a bearing of 309 degrees, to the saddle in the range. I ascended
one of the highest hills in this range, but the day was too dull to see
far. I could, however, distinguish what appeared to be a wooded country*
in the distance, from south-west to north-east. (* This "wooded country"
afterwards turned out to be sand hills, with scrub.) Observing that the
country a little more to the north was less stony, I changed our course
to a bearing of 344 degrees, over a plain thinly covered with gravelly
stones, consisting of quartz, ironstone, and a dark reddish-brown stone,
with a good deal of gypsum cropping out. The soil is of a light-brown
colour, with plenty of dry grass upon it, and very little salt bush. In
the spring time it must look beautiful. The country was so boggy from the
heavy rains, that for the sake of my horse I was obliged to stop early.
Camped at a gum creek coming from the south-west, and running a little to
the east of north. Distance to-day, eighteen miles.

Sunday, 4th July, Same Place. Not the slightest appearance of a change.
It rained in torrents all night and all day, though at sundown it seemed
to be breaking a little. The creek came down in the forenoon, overflowed
its banks, and left us on an island before we knew what we were about. We
were obliged to seek a higher place. Not content with depriving us of our
first worley, it has now forced us to retreat to a bare hill, without any
protection from the weather. The rain has come from the north-east.

Monday, 5th July, Same Place. The rain lasted the greater part of the
night, but became light before morning. Started at 12.30 on a bearing of
312 degrees for eleven miles to some sand hills. A fearfully hard day's
work for the poor horses over a stony plain, sinking up to their knees in
mud, until at eight miles we crossed a reedy swamp two miles in breadth,
and how many in length I know not, for it seemed all one sheet of water:
it took our horses up to their bellies.

Tuesday, 6th July, Sand Hills. All our rations and everything we have got
being perfectly saturated with wet, I have made up my mind to stop and
put them to rights; if we neglect them it will soon be all over with us.
This was a beautiful day, not a cloud to be seen. There are a great many
natives' tracks in these sand hills, and plenty of grass.

Wednesday, 7th July, Sand Hills. Heavy dew last night. Started on a
bearing of 312 degrees at 9 a.m. At eleven miles the sand hills cease,
and stony plain commences. The sand hills were well grassed: also the
stony plain. Dip of the country still north-east. We crossed two
watercourses--one at this side of the plain, and the other two miles
back, broad and shallow. I could see gum-trees on the latter about two
miles to the north-east as if it formed itself into a deeper channel.
Travelling very heavy. Distance to-day, twenty-five miles.

Thursday, 8th July, Sand Hills. A very heavy dew again last night.
Started at 9 a.m. At one mile we came on yesterday's course; could see
nothing; changed the bearing to 272 degrees. At seven miles crossed a
creek running north and a little west, the water being up to our
saddle-flaps. At twelve miles the sand hills ceased, and we came upon an
elevated plain, of a light-brown soil, with fragments of stone on the
surface. At twenty-five miles, in the middle of this plain, we camped,
without wood, and in sight of a large range in the far distance to the
west. Distance to-day, twenty-five miles.

Friday, 9th July, Large Plain. Left our camp at 8.50 a.m. on the same
bearing as yesterday, 272 degrees. At one mile and half came upon a creek
of water, seemingly permanent. Judging from the immense quantity of dry
grass that is strewn over the plain, this must be a beautiful country in
spring. The dip of the country is to the north and west. Our horses are
all very lame for want of shoes, and the boggy state of the soil to-day
has tried them severely. If the country does not become less stony, I
shall be compelled to leave some of them behind. We camped on a gum creek
about three miles to the west of the range. My only hope now of cutting
Cooper's Creek is on the other side of the range. The plain we crossed
to-day resembles those of the Cooper, also the grasses; if it is not
there, it must run to the north-west, and form the Glenelg of Captain
Grey. Distance to-day, twenty-one miles.

Saturday, 10th July, Gum Creek, West End of Large Stony Plain. Rested the
horses to-day. This evening we were surprised to hear a dog barking* at
the grey mare; its colour was black and tan. (* It is commonly supposed
that the native dingo or wild dog does not bark. This is an error. The
dog in this instance being black and tan, was probably a hybrid. (See
below.))

Sunday, 11th July, Same Place. This morning the sun rose at 62 degrees.
Bearing to-day, 272 degrees, so as to round the point of range, which
seems to have a little mallee in the gullies on this side, and some trees
on the west side. Started at 8.30 a.m., and at four miles ascended the
highest point of the range. The view to the north-east is over an immense
stony plain with broken hills in the distance. To the north is also the
plain, with table-hills in the far distance. To the north-west is the
termination of the range running north-east and south-west, distant about
ten miles; about half-way between is a gum creek running to north-east.
To the west is the same range, and a number of conical hills between.
Changed our bearing to 220 degrees in order to break through the range.
This range is very stony, composed of a hard milky-white flint stone, and
white and yellow chalky substance, with a gradual descent on the other
side to the south, which is the finest salt-bush country that I have
seen, with a great quantity of grass upon it. The grey mare has been very
bad; her belly was very much swollen, but this morning she seemed better.
Towards afternoon, however, she fagged very much, which caused me to stop
so soon. I am almost afraid that I shall lose her. I shall see how she is
in the morning, and, if she is no better, I will endeavour to get her on
to some permanent water or creek running to the south. I think we have
now made the dip of the country to the south, but the mirage is so
powerful that little bushes appear like great gum-trees, which makes it
very difficult to judge what is before us; it is almost as bad as
travelling in the dark. I never saw it so bright nor so continuous as it
is now; one would think that the whole country was under water. Camped
without water. No timber as yet on this side of the range, except a few
bushes in the creek. A good deal of rain has fallen here lately, and the
vegetation is looking fresh.

Monday, 12th July, Large Salt-Bush and Grass Plain. The mare seems a
little better this morning, and I shall be able to make a short journey.
There was a very heavy white frost during the night, and it was bitterly
cold. Not a hill to be seen either to the south-west or west--nothing but
plain. Left our camp at 8.30 a.m. on a bearing of 220 degrees; at two
miles and a half changed to 112 degrees for three miles to a small creek
running south with plenty of feed and water. We found our horses very
much done up this morning; they could scarcely travel over the stones,
which caused me to alter my course to the eastward, where I found the
travelling generally better. All the horses are now so lame that I shall
require to rest them before I can proceed. They will not walk above two
miles an hour among the stones. The stony plain seems to continue a long
way to the south-west, but the country being undulating and the mirage so
strong, I cannot say precisely. I intend to see where this creek will
lead me to, for I cannot face the stones again. Our distance to-day, five
miles and a half.

Tuesday, 13th July, Mulga Creek. Went to the highest point on the stony
range east of us, but could only see a very short distance. There are a
number of creeks on the eastern side running into this one. The range is
low and very stony, composed of flints and pebbles of all colours. No
timber.

Wednesday, 14th July, Same Place. During the night it became very cloudy,
and I was afraid we were going to have more rain, but it has ended in a
light shower, and cleared off this morning. I shall follow down the creek
and see what it leads to. The grey mare still seems very bad, and I must
make short journeys until she gets a little better. Started at 8.30 a.m.,
bearing 180 degrees for eight miles to Large Mulga Creek, thence 192
degrees for four miles. The country to-day is good on both sides of the
creek, a good salt-bush country with plenty of grass, but rather stony.
The gum trees are becoming a little larger on the creek, which at present
is formed into a great many channels. The timber consists of mulga and
dwarf gum, with saplings. There is plenty of water in the creek at
present, from the late rain, but I see nothing to indicate its becoming
permanent. Distance to-day, twelve miles.

Thursday, 15th July, Mulga and Gum Creek. Left the camp at 9 a.m. on a
bearing of 190 degrees for two miles, thence 230 degrees for one mile and
a half, thence 250 degrees for four miles and a half, thence 286 degrees
for two miles, thence 290 degrees for one mile, thence 270 degrees for
five miles, thence 320 degrees for one mile, to camp at some mallee. The
country on both sides of the creek is good, but subject to be flooded;
the width of the plain is about fifteen miles, bounded on the south side
by bare stony rises, and on the north by scrubby rises. The creek spreads
itself all over the plain, which seems to be very extensive. It has been
excessively cold to-day: wind from the west. Distance to-day, seventeen
miles.

Friday, 16th July, Large Plain, Mulga and Gum Creek. Left the camp at 9
a.m., on a bearing of 270 degrees for nine miles. The first six miles was
a continuation of the creek and plain; it then turned to the north-west
and the sand hills commenced. At nine miles we had a good view of the
surrounding country, from the east to the north-west. To the west we
could see the range that we crossed on the 11th instant trending away to
the north-west as far as the eye could reach, apparently a sandy and
scrubby country with small patches of open ground intervening. There also
appeared to be a gum creek, about five miles west of this point. Seeing
there was no hope for anything to the west for a long distance, I changed
my course to the south on a bearing of 190 degrees to cross the stony
rise, keeping on the sand hills for the benefit of the horses' feet. At
five miles found that the sandy country swept round the stony rise, the
country still having the appearance of scrub and sand hills all round. I
altered my course to south-east to 132 degrees for fourteen miles; on
this course we have ridden over a scrubby plain of a light sandy soil,
most beautifully grassed but dry, the young feed not having sprung. We
have not seen a drop of water on the surface; the ground evidently
absorbs all that falls; the scrub is principally the mulga and hakea
bushes and acacia, with a few other small bushes, but very little salt
bush. Camped to-night without water. The grey mare appears to be getting
round again; it seems to have been an affection of the chest, and has now
fallen down into the left knee, which has become very much swollen, but
it seems to have relieved her chest; she now feeds as well as ever.
Distance to-day, twenty-eight miles.

Saturday, 17th July, Scrub and Sandy Plain without Water. Started at 8.10
a.m. on the same course, 132 degrees. At two miles and a half, rain
water; at seven miles crossed a stunted gum creek running towards the
south-west; at twenty-five miles came upon a little rain water. Camped.
The plain still continues with very low rises at intervals; the scrub is
much thicker and the greater part of it dead, which makes it very
difficult to travel through. The grass is not so plentiful, and it is
more sandy. The creek that we crossed at seven miles was running; it had
salt tea-tree on its banks, and seems likely to have some permanent water
either above or below. I did not examine it, because, the surrounding
country being so sandy and scrubby, it will be of little use. Distance
to-day, twenty-five miles.

Sunday, 18th July, Dense Scrubby Plain. Rain Water. Left at 9.15 a.m. on
the same bearing, 132 degrees. We saw some native worleys, and the tracks
of a number of natives having passed this place a day or two ago, going
to the south-west. Distance to-day, twenty miles. Had to halt early in
consequence of grey mare being done up and unable to proceed. The first
part of the day's journey the scrub became more open and splendidly
grassed, the latter part was fearfully thick, it is composed of mulga,
dead and alive, and a few hakea and other bushes, with salt bush and
plenty of grass of two or three different sorts. We have a view of rising
ground a little to the north of our line, about from fifteen to twenty
miles distant. To-morrow I shall alter my course to strike the highest
point; it is a range, and seems to be wooded. I suppose it is the same
range that we crossed on the 11th instant. It is very cloudy, and seems
as if it will rain. Distance to-day, twenty miles.

Monday, 19th July, Dense Scrubby Plain. Started at 9.15 a.m. on a bearing
of 120 degrees to the highest point of the range. A slight shower fell
early this morning; it still looks very cloudy. We could only accomplish
ten miles to-day in consequence of the grey mare being unable to proceed
farther; if I can get her on to permanent water I shall leave her; she
only keeps me back, and endangers the other horses. I shall be very sorry
to do so, for she is a great favourite. We are now camped at a place
where there are five or six small watercourses; if we can find water I
shall give her until to-morrow to rest. The country that we have come
over to-day is most splendidly grassed, of a red light sandy soil, but
good; the mulga bushes in some places grow thick, and a great many are
very tall. Forster caught an opossum--the first that we have seen; we
intend making a dinner from him to-day. This is the first game we have
been able to secure, except two small ducks we had at the beginning of
our journey. We have found water a little way down the valley, which I
think will become a large creek further to the south-west. We are again
in the country of the kangaroos. Distance to-day, ten miles.

Tuesday, 20th July, Grassy Valley. We had another shower this morning. I
must try and make the hills to-day if I can. Started at 10.10 a.m. on the
same bearing as yesterday, 120 degrees, and at four miles ascended the
peak on the range. I see around me a scrubby country, with open patches,
and here and there in the far distance what appear to be belts of mulga.
Four miles beyond this hill we halted at some rain water. We have seen
three or four kangaroos to-day; they were the red sort with white
breasts. Distance travelled, eight miles.

Wednesday 21st July, Grass and Salt-Bush Plains. Left the camp at 9 a.m.
on a bearing of 97 degrees. Camped at some rain water in a clay-pan. At
twelve miles there is low rising ground running north-west and
south-east, which divides the two plains; there are no creeks, but the
dip of the country is to the south-west. This is as fine a salt-bush and
grass country as I have seen. It is a pity there is no permanent water.
Distance to-day, twenty miles.

Thursday, 22nd July, Open, Good Country. Started at 9 a.m. on the same
course as yesterday, 97 degrees. At ten miles crossed a small watercourse
running to the south-south-west; at sixteen miles came through the saddle
of a low range running north-west and south-east composed of limestone;
it forms one of the boundaries of a large plain, which seems well adapted
for pastoral purposes; it is well grassed, with salt bush, although we
could find no permanent water. I think I can see a gum creek to the east
of us, but the mirage is so powerful that I am not quite certain.
Distance to-day, twenty miles.

Friday, 23rd July, Large East Plain. Started at 9.10 a.m. on a bearing of
82 degrees, and at four miles ascended an isolated hill, but can see
nothing of the gum creek. Changed our course to 122 degrees, and at four
miles crossed a mulga creek running to the east. Camped on the south-east
side of a flat-topped hill, which, although the highest I have yet seen,
enabled me to see nothing but the range to the north-east, and a high
conical hill about ten miles south-west, connected with the ranges. The
country is without timber except a few mulga bushes at intervals.
Distance to-day, twenty-one miles.

Saturday, 24th July, South-east Side of Flat-topped Hill. Left at 8.10
a.m. on the same course, 122 degrees, over an undulating stony plain,
with narrow sand hills at intervals, and a number of lagoons containing
rain water, where we camped. I intend to move to-morrow to another large
lagoon that we have seen from a small rise, and rest the horses there;
they have had a very severe day of it, and feel the want of shoes very
much. The stones are mostly white quartz and ironstone, small and
water-washed. I conclude they have come from the hills that are to the
south-west. Distance to-day, twenty-four miles.

Sunday, 25th July, A Lagoon of Rain Water. Finding that we have sand
hills to cross, and being anxious to meet with the gum creek that the
blacks have talked about, I have determined to proceed to-day, but if I
do not find it on this course I shall turn to the south. Started at eight
a.m. on a bearing of 122 degrees. At five miles, one mile to the south is
a large reedy swamp. At fourteen miles changed the bearing to 135 degrees
to the head of a swamp, two miles and a half, found it dry, a large
clay-pan about three miles in circumference. I am obliged to halt, the
horses are very tired and want rest; and there being plenty of beautiful
green feed about, I have halted without water. Our journey has been
through a very thick mulga scrub and sand hills, very heavy travelling.
The trees in the scrub are of a different description to any that I have
seen; they grow high and very crooked, without branches until near the
top, and with a rough, ragged bark; seven or eight seem to spring from
one root. The wood is very tough and heavy, and burns a long time, giving
out a glowing heat. The leaves resemble the mulga, but are of a darker
colour and smaller size. The native name is Moratchee. Shot a wallaby,
and had him for dinner. They are very wild, no getting within shot of
them, which is unfortunate, as our provisions are getting rather short.
From the number of native tracks about, this would seem to be their
season for hunting in the sand hills, which accounts for everything being
so wild. We saw five turkeys yesterday, but could not get within shot of
them. All the water seems to drain into the reedy swamp and clay-pans. I
shall go no further to the east on this course, for I can see no
inducement. I shall go south to-morrow, and see what that produces; if I
cross no large creek within forty-five miles in that direction, I shall
then direct my course for the north-west of Fowler's Bay to see what is
there. Distance to-day, sixteen miles.

Monday, 27th July, Sand Hills and Dense Scrub. Left our camp at 9.20 a.m.
on a southerly course, 182 degrees. At thirteen miles we camped at some
rain water to give the horses a little rest. We have come through a very
thick scrub of mulga, with broken sand hills and a few low rises of lime
and ironstone. We have seen two or three pines for the first time, and a
few black oaks. No appearance of a change of country. From a high sand
ridge I could see a long way to the north-east, seemingly all a dense
scrub. The grey mare is unwell again. Distance to-day, thirteen miles.

Tuesday, 27th July, Sandy Undulations. Started at 9 a.m. on the same
bearing as yesterday, 182 degrees. At twenty-one miles changed our course
to 235 degrees to some gum-trees. The first part of our journey the scrub
became lower and more open, with limestone and sand rises at intervals,
and with a good deal of grass in places. The last ten miles the mulga
scrub was so dense that it was with difficulty we managed to get through.
We have seen no water on this day's route, except that in the lagoon we
are now camped at, and which is as salt as the sea. There is another
large lagoon about a mile to the westward of us, which I will examine
to-morrow to see if it gives rise to any creek. Distance to-day,
twenty-two miles.

Wednesday, 28th July, Sand Hills. Started at 9 a.m. on a bearing of 283
degrees for two miles to examine the other lagoon, which is about three
miles long, water salt. Changed our course to 182 degrees for ten miles
to a large lake crossing our course. Changed our bearing to 240 degrees,
and at four miles changed to 270 degrees, crossing some horse-tracks
going towards the large lake. This seems to be a country of salt lagoons,
for we passed three, and have seen a great many more. The large one that
crossed our south course is evidently the head of Lake Gairdner. I could
see it winding away in that direction. We have now got upon a plain
slightly undulating with thick scrub and the unceasing mulga, intermixed
with a few black oaks; no signs of water, no creeks. I intend to proceed
north of west to intersect any creek or country that may come from the
good country that we found on our south-east course, and the land of
kangaroos; there is no hope of anything here. Camped without water.
Distance to-day, twenty miles.

Thursday, 29th July, Mulga Plain, West of Lake Gairdner. Our course
to-day is 310 degrees. Left our camp at 8.30, and accomplished twenty
miles of the same scrubby plain, slightly undulating. Plenty of grass,
but no water. Same description of country as on the 18th instant.

Friday, 30th July, Mulga Plain. Started at 7.35 on same course, 310
degrees. The scrub is so dense that I cannot see above one hundred yards
ahead, and sometimes not that. During the night some swans and two ducks
flew over, apparently from Lake Gairdner, and going in our direction. At
ten miles, having met with some rain water, we halted, for the horses had
been three nights without it. I have given them the rest of the day to
drink their fill. This seems to be a continuation of the stony plain we
crossed on our south-eastern line. The country appears open to the south,
but no sign of any permanent water. Forster bakes the last of our flour
this afternoon--the last of our provisions. Distance to-day, ten miles.

Saturday, 31st July, South Stony Plain. Left at 8.30 on the same bearing,
310 degrees. At ten miles we ascended a low range running north and
south. We did not see a drop of water all day. Our course was over a
gradually rising plain, well grassed at intervals, with plenty of salt
bush, and with stone on the surface, composed of quartz, ironstone, and
the hard white flinty stone so frequently met with. The scrub has nearly
ceased. The dip of the country is south. During the night we again heard
a dog barking at one of the horses, and during the day we saw two
kangaroos. At ten miles we crossed a valley, through which water has been
flowing to the south-south-west. Camped without water. Distance to-day,
fifteen miles.

Sunday, 1st August, Stony Plain Valley. Left at 8.45 on the same bearing,
310 degrees. My reason for keeping this bearing is that there seems to
have been very little rain to the south of us, and I am unwilling to get
too far away from where it has fallen, in case I have to put to my former
line for it. If I should meet with it to-day I shall turn south-west or
west. This country is very dry, and absorbs all that falls. It is of a
bright red soil, mixed with sand and, in some places, lime. At ten miles
I am obliged to stop, in consequence of the grey mare being quite done
up; the stones play the mischief with her. I have great doubts of her
living through the journey. Distance to-day, ten miles.

Monday, 2nd August, Salt Bush--a Stony Plain. We had a little rain during
the night. Started at 9 on a bearing of 315 degrees. At three miles
changed our course to 230 degrees. The last three miles of this day's
journey were through rather a thick scrub, but well grassed, with few
stones. The former part was through a very well-grassed country, with a
little salt bush and low scrub. Saw a number of kangaroos, but they were
too wild to get near them. Distance to-day, twenty miles.

Tuesday, 3rd August, Good Country. It has rained during the whole night,
and is likely to do so to-day. Started at 9, on the same course as
yesterday, 230 degrees. The first portion of our journey was over six
miles of splendid alluvial country, covered with grass--partly spear
grass--with a little salt bush intermixed with it, also a few mulga
bushes at intervals; no other timber. It is a most beautiful open piece
of country, and looks much better than the Adelaide plains did at the
commencement of the colony. Four miles further it was not so good; the
soil became a little lighter, with more salt bush, and a little scrub.
The last eleven miles the soil is good, with grass and salt bush in
abundance, but much thicker with mulga and other low scrubs. It seems to
be a continuation of the same scrub that we passed over on the 19th
ultimo, and I observe that the ants build their habitations in the same
style as they did there. They are about one foot in diameter at the base,
and formed in the shape of a cone, and are supported by the dead root of
a mulga. Others, however, stand from eighteen inches to three feet in
height, built of clay, and on the surface. The kangaroo and emu inhabit
the country. We have also found a number of places where the natives have
been encamped. They seem to be numerous, judging from the number of
places where they have had their fires; but we have not seen any of them.
We have had it raining nearly all day, and it still looks bad. Our black
fellow left us during the night; he seemed to be very much frightened of
the other natives. He knows nothing of the country, and if he follows our
tracks back, I don't envy him his walk. He was of very little use to us,
and I wish I had sent him off before, but I thought he might be useful in
conversing with the other natives when we should meet them. He was of no
other use than for tracking and assisting in getting the horses in the
morning, for I have given them every advantage--they have been seldom
hobbled. There are three small valleys on our line in which water seems
to have run at some former period. We have crossed no course of rocks of
any description since our northern line; from which I am of opinion that
the drainage is underneath, so that there ought to be numerous springs
near the sea-coast. Camped without water. Distance to-day, twenty miles.

Wednesday, 4th August, Scrubby Good Country. Started at 8 on the same
bearing as yesterday, 230 degrees. At thirteen miles ascended a low red
granite range in which there is water. Changed our bearing to 209 degrees
to a hill on the opposite range; when I returned I found the grey mare so
done up that she is unable to proceed. I should not like to leave her,
but I cannot delay longer with her. For about half a mile under the range
where we are now camped is beautiful feed up to the horses' knees. Six
cockatoos passed over to another range. We have also found a small
running stream where I shall leave the mare to-morrow; I will make an
attempt to regain her as I return.

Thursday, 5th August, Granite Range. Started at 8 on the same bearing for
the hill on the opposite range. At six miles another low granite range
with water, where we left the mare. At twelve miles went to the highest
point of the range composed of hard flinty quartz and ironstone. We had a
good view of the surrounding country, which was generally low and
undulating, with salt lakes crossing at about ten miles. This region
appears to be dotted with the lagoons from nearly the foot of the range.
Changed our bearing to 268 degrees for nine miles. Camped under a range
of low hills with good feed for the horses. On our west course we crossed
a plain of red light soil, with abundance of grass and a little salt bush
with a very thick scrub close to the range, but as we advanced it became
more open, and the scrub lower. Shot a wallaby and had him for supper.
Distance to-day, twenty-five miles.

Friday, 6th August, Under the Low Range. Left at 8.30 a.m. on a bearing
of 239 degrees to avoid the stones on the hills. At five miles and a half
got some rain water; at nine miles changed our bearing to 255 degrees; at
fifteen miles camped among the sand hills. Shot another wallaby. The
timber about here is very large, consisting of black oaks, mallee, mulga,
the native peach, the nut, and numerous low scrubs. The grass is good in
some places. The mountain that I am steering for is further off than I
anticipated; we got sight of it a short time before we halted; it seems
to be very high, and I expect something good will be the result of our
visit to it to-morrow. The hills that we were camped under last night are
composed of quartz, and are connected with the range that we were on
running to the south-west. Distance to-day, twenty six miles.

Saturday, 7th August, Sand Hills going to the High Mount. Left at 8.30
a.m. on the same bearing, 255 degrees, for eighteen miles to the foot of
the mountain. At fifteen miles camped under the highest point, which is
composed of quartz rock. The journey to-day has been through horrid dense
scrub and heavy sand hills, to the foot of the hill, which I have named
Mount Finke. It is as high as Mount Arden; I have not light to get on the
top of it to-night. Very little rain has fallen here, and we have been
without water for the last two nights: the country is of such a light
sandy soil that it will not retain it. I almost give up hopes of a good
country; this is very disheartening after all that I have done to find
it. If I see nothing from the top of the mount to-morrow, I must turn
down to Fowler's Bay for water for the horses. As I could not remain
quiet, I got on one of the lower spurs of Mount Finke to see what was
before me. The prospect is gloomy in the extreme! I could see a long
distance, but nothing met the eye save A DENSE SCRUB AS BLACK AND DISMAL
AS MIDNIGHT. On my return I found that Forster had succeeded in finding
water by digging in the creek. Distance to-day, twenty miles.

Sunday, 8th August, Mount Finke. At dawn of day I ascended the mountain,
but was unable to see much more than I did last night, in consequence of
there being a mist all round. No high rising ground is to be seen in any
direction. A FEARFUL COUNTRY. Left the mount at 9.30 a.m. on a bearing of
270 degrees. At eighteen miles halted to give the horses some food, as
they were obliged to be tied up all last night, there not being any feed
for them, and the scrub very dense. The horse Blower seems to be very
unwell; he has lain down twice this morning, and an hour's rest will do
him good. After leaving the mount we have a thick mallee and mulga scrub
to go through with spinifex. At ten miles changed our bearing to 190
degrees; at eight miles camped. The whole of our journey to day has been
through a dreadful desert of sand hills and spinifex. In the last eight
miles we have not seen a mouthful for the horses to eat and not a drop of
water; it is even WORSE than Captain Sturt's desert, where there was a
little salt bush; but here there is not a vestige. Distance to-day,
twenty-five miles.

Monday, 9th August, Desert. Started at 8.30 on the same bearing, 190
degrees. At five miles there is a change in the country; the spinifex has
suddenly ceased and low scrub taken its place; the sand ridges are spread
and the valley wider. At seven miles discovered some rock water in the
middle of a valley with plenty of salt bush and green grass, first rate
for the horses, which have had nothing to eat for two nights. I shall
give them the rest of the day to recover. They were beginning to be very
much done up, and it was with difficulty we could get them to face the
spinifex. Shot a pigeon and had him for supper. We have seen where a
horse has been a long time ago. Distance to-day, seven miles.

Tuesday, 10th August, Rock Water. Started at 8.30 on a bearing of 180
degrees. Camped at eighteen miles without water, and a very little food
for the horses, only a little salt bush. The appearance of a change from
the dreary desert lasted only for about one mile from where we camped
last night; it then became even worse than before--the sand hills higher,
steeper and closer together, the spinifex thicker and higher; we got the
horses through it with difficulty. It rained all last night and all day.
There is some rising ground to the west. Distance to-day, eighteen miles.

Wednesday, 11th August, Dense Scrub. Left our camp at 8 on the same
bearing, 180 degrees. At 9 obliged to halt for the remainder of the day,
the horses being too tired to proceed further; the fearful sand hills are
very trying for them. To-day's few miles have been through the same
DREARY, DREADFUL, DISMAL DESERT of heavy sand hills and spinifex with
mallee very dense, scarcely a mouthful for the horses to eat. When will
it have an end? We again saw the rising ground a little to the north of
west of us; I should have gone and examined it, but our small remaining
quantity of provisions being nearly exhausted, I could not venture; my
object now being to make Fowler's Bay for water for our horses, and
thence to Streaky Bay, to endeavour to get some provisions there to carry
us home. We have now travelled considerably upwards of a thousand miles,
and in that journey my horses have had only four clear days to
themselves; they have done most excellently well. No water.

Thursday, 12th August. Dense Scrub. Left at 8.25 on a bearing of 165
degrees. Camped at ten miles; the horses done up. The same dreary desert.
No water.

Friday, 13th August. Dense Scrub. The horses look very bad this morning.
I hope we shall be able to make the sea-coast to-day. Started at 8.30 on
the same bearing, 165 degrees, but was unable to get more than ten miles
out of the horses; Bonney is nearly done up, and there is no water for
the poor animals. I hope I shall not be obliged to leave the poor old
horse behind, but I very much fear that I shall have to do so if nothing
turns up to-morrow. The country is still the same. This is dreadful work!

Saturday, 14th August, Dense Scrub. Started at 8.15 on the same bearing,
165 degrees. At ten miles came upon some green feed for the horses, and
gave them the benefit of it for the rest of the day. Bonney still very
bad. For the last two miles we have had no sand hills, but very dense
mallee and tea-tree, with a light sandy soil with a little limestone,
also salt bush and pig-face in abundance. No water.

Sunday, 15th August, Dense Mallee Scrub. Started at 8.45 on same bearing,
165 degrees. At two miles and a half changed our course to 225 degrees,
having found some fresh horse-tracks; at seven miles camped for the
remainder of the day to recruit the horses, having come upon some new
green grass. Distance actually travelled, fifteen miles.

Monday, 16th August, Dense Mallee Scrub. Started at 9 on a course of 205
degrees. Twelve miles to Miller's Water. I intended to have given the
horses two days' rest here, but there is not sufficient water; there are
only three holes in the limestone rock, and the thirsty animals have
nearly drunk it all: there will not be enough for them in the morning.
The country that we have come through yesterday and to-day resembles the
scrub between Franklin Harbour and Port Lincoln--mallee with grassy
plains occasionally--only the mallee is larger, and the plains are met
with at shorter intervals, more numerous and of larger extent. The soil
is good but light, being produced by decomposed limestone, of which the
low range to the north-west is composed. I am unable to go to Fowler's
Bay as I intended; our provisions are exhausted, and the horses unable to
do the journey. I must now shape my course for Streaky Bay to get
something to eat.

Tuesday, 17th August, Miller's Water. Watered our horses from a
waterproof with a quart pot. Started at 9.15, our course 160 degrees, six
miles to Bectimah Gaip. For the first three miles the grassy plains are
very good, and seem to run a considerable distance between belts of large
mallee, in some places wider than in others, and seem to be connected by
small gaips; I think water could be easily obtained by digging. The last
three miles to the coast is very dense small mallee. Actual distance,
twelve miles. I intend to give the horses a rest to-morrow. I regret
exceedingly that I was unable to make Fowler's Bay. It is with difficulty
that I have been able to save Bonney; he is still very weak and unable to
do a day's journey; we can scarcely get him to do the short journeys we
have been doing lately. For upwards of a month we have been existing upon
two pounds and a half of flour cake daily, without animal food. Since we
commenced the journey, all the animal food we have been able to obtain
has been four wallabies, one opossum, one small duck, one pigeon, and
latterly a few kangaroo mice, which were very welcome; we were anxious to
find more, but we soon got out of their country.

These kangaroo mice are elegant little animals, about four inches in
length, and resemble the kangaroo in shape, with a long tail terminating
with a sort of brush. Their habitations are of a conical form, built with
twigs and rotten wood, about six feet in diameter at the base, and rising
to a height of three or four feet. When the natives discover one of these
nests they surround it, treading firmly round the base in order to secure
any outlet; they then remove the top of the cone, and, as the mice
endeavour to escape, they kill them with the waddies which they use with
such unfailing skill. When the nest is found by only a few natives, they
set fire to the top of the cone, and thus secure the little animals with
ease. For the last month we have been reduced to one meal a-day, and that
a very small one, which has exhausted us both very much and made us
almost incapable of exertion. We have now only TWO meals left to take us
to Streaky Bay, which is distant from this place ONE HUNDRED MILES. We
have been forced to boil the tops of the pigface, to satisfy the wants of
nature. Being short of water, we boiled them in their own juice. To a
hungry man they were very palatable, and, had they been boiled in fresh
water, would have made a good vegetable. Yesterday we obtained a few
sow-thistles, which we boiled, and found to be very good.

Wednesday, 18th August, Bectimah Gaip. Rested the horses and obtained a
few shell-fish from the beach: there are very few, which was a
disappointment to us.

Thursday, 19th August, Bectimah Gaip. Started at 8 a.m. for Streaky Bay.
I managed to get thirty miles to-day, which is a great help. I only hope
that Mr. Gibson is at Streaky Bay, so that we may be able to get
something to eat; we must endure three days' more starving before we
shall be able to reach there.

Friday, 20th August, Smoky Bay. Started at 7.15. Mallee scrub in some
places very dense, in others open, with good grassy plains at intervals,
in which I think water could be had by digging; very few birds about, and
those small. At twenty-five miles we got some rock water. Distance
to-day, thirty-five miles.

Saturday, 21st August, Small Grassy Plains. Started at 7.30 on a
south-easterly course. Got a little water in the limestone rock for our
horses. Camped on the shore at Streaky Bay at sundown. The last sixteen
miles were through very dense scrub; the former part through scrub with
good grassy plains at intervals. Distance, thirty-eight miles.

Sunday, 22nd August, On the Shore at Streaky Bay. Started at 11 a.m. to
make Mr. Gibson's station. The horses did not arrive until 10.30, as they
had gone back on their tracks of yesterday. During the time Forster was
after them, I managed to shoot a crow, and cooked him in the ashes. We
had him for breakfast--the first food we have had for the last three
days; it was very agreeable to taste and stomach, for we were beginning
to feel the cravings of nature rather severely. I hope Mr. Gibson will be
at the Depot; it will be a fine trouble if he is not, and we have to
travel two hundred and forty miles on the chance of shooting something.
Twenty-four miles to Mr. Gibson's station, where we were received and
treated with great kindness, for which we were very thankful. We enjoyed
a good supper, which, after three days' fasting, as may readily be
imagined, was quite a treat.

Monday, 23rd August, Mr. Gibson's Station. Both Forster and myself felt
very unwell, especially Forster, who is very bad; the sudden change from
a state of starvation to plenty of good and wholesome food has been the
cause. I am suffering chiefly from weakness and a very severe pain
between the shoulder-blades, which I have felt for some weeks back. It is
a dreadful pain, and nearly incapacitated me from sitting in the saddle
all day yesterday; I thought I should not have been able to reach here, I
was so very bad with it. I have been obliged to send down to the next
station, about thirty miles distant, to try and get some horseshoes. I
must rest here a few days to recover.

Tuesday, 24th August, Mr. Gibson's Station. Forster appears to be a
little better this morning, but very weak; I also feel a little better
this morning from yesterday's rest.

Wednesday, 25th August, Mr. Gibson's Station. I have succeeded in getting
some shoes for the horses from Mr. Miller, to whom I am deeply indebted
for his kindness in allowing me to have them.

Thursday, 26th August, Mr. Gibson's Station. Shoeing the horses and
preparing for a start at the beginning of next week.

Friday, 27th August, Mr. Gibson's Station. At the same thing. Improving
in health and strength.

Saturday, 28th August, Mr. Gibson's Station. I have been very unwell all
night.

Sunday, 29th August, Mr. Gibson's Station. Still very ill; unable to do
anything.

Monday, 30th August, Mr. Gibson's Station. The same.

Tuesday, 31st August, Mr. Gibson's Station. I had a dreadful night of it;
seized with cramp in the stomach, and thought I should never see morning;
no medicine to relieve me. I intended to have started to-day, but am
quite unable to do so.

Wednesday, 1st September, Mr. Gibson's Station. Can stay no longer; made
a start to-day, and got as far as one of Mr. Gibson's out-stations,
twenty-five miles. Quite done up.

Thursday, 2nd September, One of Mr. Gibson's Out-Stations. Raining this
morning; unable to proceed. Very unwell.

Friday, 3rd September, Same Place. Feel better this morning. Started at
8.30 for Parla. I am unable to make any attempt to recover the grey mare.
Made Parla at 1 p.m.; camped at ten miles beyond. Distance to-day,
twenty-five miles.

Saturday, 4th September, Ten Miles beyond Parla. Started at 8.15 on an
east bearing twenty-three miles to Rock Water. Camped. Very poor country.
The granite range that Mr. Hack has laid down on his chart, I cannot
find. I have come east from Parla, and ought to have crossed about the
middle of it.

Sunday, 5th September, Rock Water. I shall shape my course for the
Freeling range, and see what that is made of. Started at 7.30 on a
bearing of 84 degrees twenty-two miles. Rock water with plenty of grass.
Gave the horses the rest of the day.

Monday, 6th September, South of Mount Sturt. Started at 8.15 on a bearing
of 84 degrees for twenty-five miles. Changed the bearing to 60 degrees
for three miles to a fine plain covered with grass. Halted. No water.
There are some high hills to the east-north-east, to which I have now
changed my course, and which I conclude to be the Freeling range. Our
journey to-day has been through very scrubby and sandy country,
especially the last fifteen miles. At six miles south there is a high
table-topped hill, which I think is granite. I intended going down to it,
but the country, so far as I could see, was apparently not good, and,
having crossed the tracks of some horses going towards it, and being very
unwell myself, I thought it would be useless my going. Distance to-day,
thirty-eight miles. No water.

Tuesday, 7th September, Freeling Range. Started for the range at 8 on a
bearing of 60 degrees. At eleven miles ascended the south-west hill of
Freeling range, Mount Sturt bearing 266 degrees. Changed the bearing to
96 degrees to a stony hill of granite. Found a little water, and halted
for the remainder of the day. Distance, fifteen miles.

Wednesday, 8th September, Freeling Range. Started at 7.30 for Separation
Camp, bearing 72 degrees. Halted at thirty-three miles. The first
twenty-five miles were mallee scrub with patches of grass; the last eight
miles were over elevated table land, salt bush, and a little grass with a
few patches of scrub, the soil being red, with a few fragments of quartz
and ironstone on the surface. No water.

Thursday, 9th September, Salt-Bush Country. Started at 9.15 on the same
bearing, 72 degrees, fourteen miles; changed to 160 degrees (1.30 p.m.)
two miles and a half; thence 80 degrees three miles to a small creek,
where we can obtain water by digging in the sand. Camped. Distance
to-day, twenty miles. Did not see Separation Camp; it is wrongly placed
on the map.

Friday, 10th September, Small Creek. Started at 9 on a bearing of 110
degrees for Cooroona; at seventeen miles made Cooroona. Camped fifteen
miles beyond.

Saturday, 11th September. Arrived at Mr. Thompson's station, Mount Arden.

I cannot conclude this narrative of my first journey, without
acknowledging that it was with the advice and assistance of my friend Mr.
Finke SOLELY, that I undertook this exploration of the country. I
therefore look upon him as the original pioneer (if I may be allowed so
to express myself) of all my subsequent expeditions, in which our friend
Mr. Chambers afterwards joined.


JOURNAL OF MR. STUART'S SECOND EXPEDITION (IN THE VICINITY OF LAKE
TORRENS). APRIL TO JULY, 1859.

Saturday, 2nd April, 1859. Started from Mr. Glen's for St. A'Becket's
Pool, where we camped. This water hole is a large one, and likely to last
a long time. The country around is good--a large salt bush and grassy
plain, with upwards of 300 cattle feeding upon it. Found the native
cucumber growing.

Sunday, 3rd April. Shortly after sunrise started from St. A'Becket's
Pool, over low sand hills with large valleys between, well grassed, as
described by Mr. Parry. Camped about two miles to the north-east of it,
in a polyganum and grassy valley.

Monday, 4th April. The saddles injuring our horses' backs, we must stop
and repair them. Herrgott and I rode to Shamrock Pool. There is still
water there. It may last about a month, but it is not permanent.

Tuesday, 5th April. The horses could not be found before noon. One of
them has lost a shoe, which will require to be put on. It is too late to
start to-day for St. Francis' Ponds, the distance being thirty-two miles,
and no water between. I deem it advisable to remain until to-morrow.

Wednesday, 6th April. Started on a bearing of 330 degrees, and at six
miles came upon a gum creek, with abundance of water, which I believe is
permanent. For fifty yards on each side of the creek there is a great
quantity of polyganum and other water-bushes. On the water there are a
great many ducks, cranes, and water-hens. The water hole is upwards of
three-quarters of a mile long; at the broadest place it is fifty yards in
breadth. There are two trees marked "J.G. and W. Latitude, 30 degrees 4
minutes 1 second." At one mile struck Mr. Parry's tracks; had a view of
the country on the bearing that I intended to steer; saw that it would
lead me into a very rough country, therefore followed his tracks to where
he had camped. Camped south of Mount Delusion, without water. I do not
doubt that there is water further down the creek to the eastward.

Thursday, 7th April. Went to the top of Mount Delusion and took bearings.
Had some difficulty in finding St. Francis' Ponds. Towards sunset we
found them, and, to our great disappointment, quite dry; all the water
had disappeared, except a little in one of the creeks, which was salter
than the sea, and of no use to us. There seems to have been no rain here
this season; I have searched the country all round, but can see no sign
of water. I must return to-morrow morning to the creek that I passed
yesterday. The horses have now been two nights without water; they appear
to feel it very much.

Friday, 8th April. Started back on a straight line, 6.40, for the gum
creek, and arrived at 1.40 p.m., the horses being so much done up that I
must give them two days' rest. I expect they will endure it better next
time; they now know what it is to be without. In our course we crossed
the middle of Mr. Parry's dry lake. It can be crossed at any time, for
there are large courses of slate running through it in a north and south
direction, level with the bed of the lake. The country around St.
Francis' Ponds is as Mr. Parry describes it, with the exception of the
water, which is gone. There is a great deal of Cooper's Creek grass
growing in places. It is my intention to start with one man (as soon as
the horses recover), and endeavour to find water nearer Mount North-west
range. If I can find water east or west of St. Francis I shall then be
able to make the Finniss Spring.

Saturday, 9th April. Resting the horses.

Sunday, 10th April. I intended to have gone to the north to-day to search
for water, but I am so unwell from the effects of the water of this creek
that I am unable to do so. I have been very ill all yesterday and all
night, but I hope I shall be right to-morrow.

Monday, 11th April. I am unable to go and search for water, being too
weak and not able to ride. I have sent Herrgott and Muller to find St.
Stephen's Ponds, and see if there is water; they are to return by the
foot of the range and endeavour to find water there also. I have been
very ill indeed during the night; I have had no sleep for the last two
nights, and I am so weak that I am scarcely able to move.

Tuesday, 12th April. Feel a little better this morning, but still very
unwell.

Wednesday, 13th April. I feel a good deal better. I hope by to-morrow I
shall be all right again. Herrgott did not return until noon to-day. He
reports that there is no water in St. Stephen's Ponds, which I expected;
but he also states that he has found a batch of springs three miles on
this side of the ponds, with abundance of water. They are twelve in
number. I shall go to-morrow with the party to them. I am very glad he
has found them. There will now be no difficulty in taking stock to
Chambers Creek. From this camp to the springs will be the longest journey
to be encountered in a season like this, in which so little rain has
fallen. After rain has fallen there will be no difficulty at all. The
native cucumber grows about here.

Thursday, 14th April. Started at 8.10. The country travelled over was
fine salt-bush country, but there was no water on our course, although we
disturbed numerous pigeons and other birds. There are three table-topped
hills to the east of the end of our north line; I think they are those
within a short distance of which Major Warburton mentions that he found
water. It would take me too much to the east of my course to examine them
at present. I should have gone that way if Herrgott had not found those
twelve springs, which we hope to make early to-morrow morning, and then
proceed to the Finniss Springs. Camped on the east side of Decoy Hill,
without water.

Friday, 15th April, East Side of Decoy Hill. At daybreak despatched
Campbell for the horses. At 7.30 he returned with only five, and said
that he found them on the track, going back for the water from which we
have come, and that the others had left the tracks and gone west towards
the hills. I immediately despatched Muller on horseback to track and
bring them back, and I sent the others by Herrgott to get water at the
springs. Sundown: no appearance of the horses. They must have gone back.
If they have, it will be the middle of the night before Muller can be
here. It is vexing to be delayed thus with the brutes.

Saturday, 16th April, Same Place. Muller and the horses have not yet
come. I must go to the top of Decoy Hill to take some bearings. At 9.30
returned to the camp, and found Muller had just returned, but no horses;
he had followed upon their tracks until they crossed a stony hill, where
he lost them, and, on purpose to find them again, he tied the mare to a
bush; she broke loose, and would not allow him to catch her until she got
to the water. It was then sundown; he remained there during the greater
part of the night to see if the others would come in: they did not, and
he therefore came up to inform me of what had occurred. He was without
fire, blankets, or anything to eat. I did not pity him; he ought to have
been more careful. I had several times warned him not to leave the mare
insecurely tied, or she would be off. I gave him a fresh horse, and sent
him and Campbell off to follow them up to wherever they go, and not to
come back without them. It is most dreadfully annoying to be kept back in
this manner, all through the carelessness of one man: he must have been
quite close to them when the mare got away. They were short hobbled, and
I had looked at them at half-past two in the morning, to see if they were
all right, and found them feeding quietly, so that they could not have
gone far. Sundown: no appearance of the horses. I feel much better
to-day.

Sunday, 17th April, Same Place. Still neither horses nor men. At 1.30
they arrived; my men had gone over to the range, and had searched every
creek, but without success. When found, the runaway animals were standing
on a rise looking very miserable and at a loss what to do; they had
skirted the hill as far down as Mount Delusion. The men took them to the
last water, remained there through the night, and left for this place
this morning. I will give them an hour's rest, and go to the springs
to-night. Arrived at the springs at sundown; they are about nine miles
from Decoy Hill.

Monday, 18th April, Same Place. Resting horses. I went to the top of
Mount Attraction, accompanied by Herrgott, to see what appearance the
country had to the north of west. I observed a high red table-topped hill
bearing 276 degrees from this point, for which I started in search of
water. I had a good view of the country all round; it seems very low to
the westward with low ranges and valleys between; plenty of salt bush and
grass. There is copper with the ironstone on the top of Mount Attraction;
native copper is adhering to the sides of the large pieces of ironstone.
No water. Changed our course to north one mile and a half, thence to
north-east five miles, thence to the springs, but could neither find
water nor Major Warburton's tracks. To-day's journey forty-five miles.
Arrived at the springs after dark.

Tuesday, 19th April, Springs. To the south of our tracks yesterday there
was the appearance of a gum creek, and I think it advisable to send
Herrgott to-day to examine it for water. It would be a great advantage
for stock going to the new country. Seen from a little distance these
springs, at which we are camped, resemble a salt lagoon covered with
salt, which however is not the case; it is the white quartz which gives
them that appearance. There are seven small hillocks from which flow the
springs; their height above the plain is about eight feet, and they are
surrounded with a cake of saltpetre, but the water is very good indeed,
and there is an unlimited supply. Herrgott has taken a sketch of them. He
has returned from examining the gum creek, but can find no water. I must
push on to-morrow for Finniss Springs, and trust to find water on the
way.

Wednesday, 20th April, Same Place. Started at 7.30 on a bearing of 275
degrees over a stony, undulating country with plenty of grass and salt
bush, but no water. At twenty miles we saw a smoke raised by the blacks
to the south of our line, under the range. Camped at 5.15 under a low
range about thirty feet high and very perpendicular, running nearly
north-east and south-west. Distance to-day, thirty-three miles.

Thursday, 21st April. Started at daybreak this morning. Same course. Cut
Major Warburton's tracks at two miles, and changed to his course, 252
degrees. At one mile, saw Finniss Springs a mile and a half to the south
of us; went down to them and camped. There is an immense quantity of
water flowing from them. I shall raise a large cone of stones upon the
hill, which is very prominent and can be seen from a long distance.

Friday, 22nd April, Finniss Springs. Went to the top of Hermit Hill,
whence I obtained a very extensive view of Lake Torrens from north-west
to north-east. Mount Hermit is surrounded by low hills, and in the far
distance there seems to be rising ground. To the south are broken hills,
the termination of the Mount North-west range. I shall examine that part
of the country to-morrow. Between this and the lake (Eyre) to the north
the country is very rough--broken cliffs, with sand; the good country
does not extend more than three miles. The springs are very numerous all
round this mount, and seem to drain into the lake; they give out an
immense quantity of water, and there are many streams of water running
from them. The ground is covered round about the springs with a cake of
soda and saltpetre. I intended to have moved on to Gregory Creek this
afternoon, but took the precaution to send my stockman to see in what
state the water was. He reports the water in the creek to be quite salt,
and many of the small fish dead; he also found some very perfect fossil
shells, the mussel and oyster; they have now become a solid limestone;
they were found in a large circular piece of limestone.

Saturday, 23rd April, Finniss Springs. Started at 8 a.m. with Herrgott to
examine the country south of this. Between this and the range the land is
good in places. It is a little rotten and stony, but the range is a
beautiful grass country to the very top. In the creeks the grass and
other plants are growing luxuriantly, but we could find no water. I was
unable to prosecute the search as far as I wished, in consequence of my
horse having lost a shoe and becoming quite lame, which forced me to
return to the camp, where we arrived at 9 p.m. The view from a high
conical hill of white granite with black spots at the north-west point of
the range, is very extensive, except to the south, which is limited. We
saw smoke in one of the creeks to the east; but as I was anxious to
examine the creek to the south-west, which we saw from the top of the
conical hill, I did not go to where the smoke was rising, thinking that
the blacks might only be hunting. I therefore crossed the hills to the
creek over a good feeding country, timbered with box and gum-trees. We
expected to find water in it, from the great number of birds of all
descriptions that were flying about; we followed it down, but were
unsuccessful, although the birds continued all the way. There must be
water about the hills in some place. At sundown, my horse becoming very
lame, I was forced reluctantly to return. The flow of the waters is
northward into North Lake Torrens. On Monday I shall start again to the
south-west, and leave the examination of the range to the south-east
until my return.

Sunday, 24th April, Finniss Springs. Latitude, 29 degrees 33 minutes 30
seconds. Rested.

Monday, 25th April, Finniss Springs. As it seemed likely to rain, in
which case the country would be very soft, I started at 9.30 on a bearing
of 242 degrees for Chambers Creek. After three miles of gravelly soil and
scanty feed we came to the banks of the two creeks passed by Major
Warburton, splendidly grassed, but the water very salt. They flow into
Lake Torrens. After leaving these creeks we had four miles of sand hills,
very rich with feed, thence over some stony ground to the creek, all
good; my course brought me about three-quarters of a mile to the south of
the creek, which I expected. Distance from the springs to this water
hole, two miles; this is a very long water hole, with plenty of water in
it, and the feed good. We saw some fresh tracks of natives to-day, but
did not meet with any of them.

Tuesday, 26th April, Chambers Creek. I intend to remain here to-day to
fix this place and examine the country about it. Latitude, 29 degrees 39
minutes 9 seconds. I sent Campbell (my stockman) in one direction, and
Muller (the botanist) in another; they report quantities of water, also a
great deal of salt water, with plenty of salt for the use of stations,
with abundance of feed. The stockman saw numerous fresh tracks, but did
not see any natives. The fires were still burning. Muller saw an old man,
a woman, and a child. They were very much frightened, and when he
approached, they called out "Pompoy!" and moved their hands for him not
to come any nearer. As they seemed quite unwilling to hold any
conversation, he left them.

Wednesday, 27th April, Chambers Creek. Started at sunrise this morning,
accompanied by my botanist. After travelling thirty miles in a fruitless
search for water, we camped upon a large stony plain with plenty of
vegetation. The horses were very much tired by reason of the heavy sand.
We could see no sign of Lake Torrens. Latitude, 29 degrees 53 minutes 58
seconds.

Thursday, 28th April, Large Stony Plain. Saddled by break of day. Changed
my course to see if the water is still at Yarra Wirta. In order to avoid
the heavy sand hills, which will not do for the horses if there is no
water, I steered for the creek, struck it a little to the north of where
I crossed it on my former expedition, and followed it down. Passed my
former encampment, and found no water there, but on following it down to
where I considered it permanent, I found water still there. I shall give
the horses the afternoon to recruit, and start early in the morning.
Distance to-day, twenty-three miles.

Friday, 29th April, Chambers Creek. Started at sunrise for about a mile
to that part of the north shore of the lake opposite to where the Yarra
Wirta empties itself into it. The country close to the lake is very stony
and scanty of feed; there is some water in it, but it is very salt; a few
salt creeks run into it, but no great body of water. I ascended a hill
for which I had been steering, and obtained an observation of the sun and
bearings. Latitude, 30 degrees 8 minutes 11 seconds. There is no
appearance of any lake between this point and Mount Deception; it appears
to be a stony plain with some ridges of sand hills. This hill, which I
have named Mount Polly, for distinction, is the easternmost of the
flat-topped hills on the north side of the lake, and is a spur from the
Stuart range. It is very stony, and there is grass nearly to the top; it
is very level, and extends for six miles in a north-westerly direction. I
saw that there was little prospect of my obtaining water to-night; and
knowing that the natives had been seen within a few miles of the camp, I
felt anxious about the safety of my party. I determined to proceed
towards the camp on a north-westerly course. Arrived at the creek at
11.30 p.m. and found all right; the natives had paid them a visit, as I
anticipated, but my people could get no information from them. They were
six in number; one was very forward, wishing to examine everything. I had
left orders that, if they came, they were not to be allowed to come near
the camp, but were to be met a little distance from it. They remained for
some time, and then stole off one by one without being perceived, and
were out of sight in a moment. The one that remained to the last in his
flight did not forget to carry along with him a piece of blanket that had
been a saddle-cloth, and which happened to be lying outside the camp.

Saturday, 30th April, Chambers Creek. Sent Muller and my stockman to
build a cone of stones upon the highest of the three table-topped hills,
for the base line of the survey. They are three remarkable hills close
together; two only can be seen coming from the south and from the
north-east. Latitude, 29 degrees 40 minutes 27 seconds. From the hill the
men saw a number of native fires smoking to the westward on the creek,
but have not seen any natives.

Sunday, 1st May, Chambers Creek. This morning we had a heavy dew. Went to
the top of the three table-tops, and had a fine view of Mount Hamilton
and the lagoon where the springs are, and the other hills; they are the
same hills that I saw on my north-west course, when on my last journey.

Monday, 2nd May, Chambers Creek. Sent Muller and Campbell to build a cone
of stones on Mount Strangways, which I have fixed as a south point of my
base line. The mean of all the observations that I have got to-day makes
the latitude to be 29 degrees 39 minutes 15 seconds.

Tuesday, 3rd May, Chambers Creek. Spent the day examining the
neighbourhood for water, and in taking numerous bearings.

Wednesday, 4th May, Chambers Creek. I intend to move to-day to the large
water holes westward, where I first struck the creek. The horses having
strayed a long way off this morning, made it 11 o'clock before we got a
start. About four miles from last night's camp the chain of large water
holes commences, and continues beyond to-night's camp. They are indeed
most splendid water holes--not holes, but very long ponds; they are
nearly one continuous sheet of water, and the scenery is beautiful. I am
sorry I did not name it a river in my former journal. I must bring my
survey up to this night's camp to-morrow. It is very cloudy to-night,
with a strong wind from the south-west, from which quarter the clouds are
coming. The country is a little stony, but well grassed.

Thursday, 5th May, Chambers Creek. Moved the camp to a better situation.
Ascended a hill, got some bearings to fix it, and built a cone of stones
upon it. I have had the creek, which joins this, run up for three miles
to the sources to-day. There is no more permanent water. There are an
immense number of small fish in the ponds, and on the banks there is a
shrub growing that tastes and smells like cinnamon; we happened to stir
up the sugar in a pannikin of tea with a small twig of the bush, and it
left quite the flavour of it in the tea. I have had Herrgott to take
sketches of some of the ponds, also of the fish and other remarkable
things. It has been rather cloudy to-day, and I could not depend upon my
observations. There are numerous tracks of natives about, but we have not
seen any of them; we have also found some new plants in the creek.

Friday, 6th May, Chambers Creek. Moved further up the creek on the south
side to the last water that we knew of. It is a hole of rain water, very
large, and will last a long time, being well sheltered by gum-trees and
other shrubs.

Saturday, 7th May, Chambers Creek. Sent Muller to see if there is any
more water to the west, and went myself to the top of a small hill, and
built a cone of stones to connect this point with the last point. Muller
returned after dark, and reported that there was no more permanent water.
I shall start to the north to-morrow.

Sunday, 8th May, Chambers Creek. Started to the north over the range,
which is rather difficult to get the horses up and down. On the top it is
very stony, with salt bush and scanty grass. Crossed the Margaret and a
salt creek, in which there is water, some of which is salt and some
brackish, but not unfit for the use of cattle. There is abundance of feed
all round. We arrived at Hamilton Springs a little before sundown.
Distance, twenty-one miles.

Monday, 9th May, Mount Hamilton. Some of the horses require to be shod
to-day. I shall also require to build a cone of stones upon Mount
Hamilton (the one built by Major Warburton having fallen down), and get
an observation of the same. Latitude, 29 degrees 27 minutes 37 seconds.
The springs are certainly very remarkable, and Major Warburton gives a
very good description of them.

Tuesday, 10th May, Mount Hamilton. Started for the Beresford Springs.
Arrived at Mount Hugh at 11 o'clock, seven miles distant from Mount
Hamilton, and, as I anticipated, found a number of splendid springs,
giving out a fine stream of water, not the least brackish. The hill from
which this stream issues is one hundred feet above the level of the
plain, the water coming from the very top. My horse got bogged on the
top, and I had some difficulty in getting him out, but I did so at last
without injuring him. Started from the mount at 12.30, and, after three
miles and a half, arrived at Beresford Springs. The Beresford Springs are
nothing in comparison to the others; there are only two that are running,
but they are very good. The country travelled over to-day has been very
well grassed, with salt bush; take it altogether I have not seen better
runs in the colony, and in the driest summer the furthest distance from
water will not be above five miles at the most, but the feed is so
abundant that they would not require to go so far. On that account they
will feed double and treble the number of stock that the runs down the
country do. At two miles on this side of the Hugh Springs discovered
another batch of springs with plenty of water running from them; there
are about eight or nine of them very good; those springs have not been
visited by Major Warburton. We examined all round, but could find no
tracks. I have named them the Elizabeth Springs. There is enough water
running to drive a flour-mill in two or three places. They are really
remarkable springs--such a height above the level of the plain; I saw
them from a hill on Chambers Creek (the Twins). From whence do they
derive their supply of water, to cause them to rise to such a height? It
must be from some high ranges to the north-west, or a large body of fresh
water lying on elevated ground. This is another strange feature of the
mysterious interior of Australia. I shall remain here until after 12
to-morrow, to get an observation of the sun to fix this hill. I shall
return to Mount Hamilton, and proceed to examine the country west of
North Lake Torrens, for one of the east runs, which will complete my
survey of them, and I shall despatch thence a messenger to Oratunga.

Wednesday, 11th May, Elizabeth Springs. Latitude, 29 degrees 17 minutes
43 seconds. I omitted to mention yesterday that, two miles before we
reached Beresford Hill, we crossed Pasley Ponds and saw one of the
Major's camps. The water is brackish, but not bad. The white deposit
round these springs, and also round the Elizabeth, is soda. In returning,
I examined the Coward Springs; the water is good, and running. There is a
plentiful supply. It was dark when I arrived at Mount Hamilton. Saw four
natives to-day, but they gave us a wide berth; they do not like to come
near us.

Thursday, 12th May, Mount Hamilton. Some of the horses require shoeing,
and I wish to get another observation of the sun. I shall remain here
to-day, and examine the country to the north-east. About seven miles in
that direction is the salt creek of Major Warburton. The country is of a
light sandy soil covered with grass.

Friday, May 13th, Mount Hamilton. Started to the eastward, to complete
the survey of the runs, and see if there are any more springs. To the
south of east, about four miles, we discovered four springs not seen by
the Major; there is a plentiful supply of water, and would be more if
they were opened. One is choked up with reeds, but the other two are
running. Saw some natives; they seemed frightened at first, but were
induced to come close up: they were very much amused at our equipments.
Two had seen or heard of whites before; they knew the name of horse, but
no more; they call water courie, and some of their words very much
resemble those of the natives in Port Lincoln. We could make nothing of
them--they repeat every word of the question we ask them. They followed
us over to the Margaret, and took us to some fresh-water springs in the
creek, the water of which is very good. There is a quantity of reeds
growing round them, also tea-tree. From this we followed the creek to the
north, thence north-east towards the lake, but the water being too
brackish, I returned to the springs, the natives walking with us all the
time; they seemed very inoffensive. In following down the creek, another
native joined us from the creek, carrying a net in which were some small
fish; the net was a hoop one, well made.

Saturday, May 14th, The Margaret Creek. The morning very cloudy; every
appearance of rain. Saddled and proceeded in search of Emerald Spring, on
a north course. At seven miles made Mr. Babbage's old camp on a sand
hill. Camped a little way from it. I did not know the position of the
spring, but Herrgott informed me that it was three miles to the west. It
commenced raining before we started, has rained all the way up, and is
still doing so; it is a very light rain, but the wind is very strong and
cold from the south-west. Intended to have brought up my plan, but the
rain and wind prevent me.

Sunday, 15th May, Mr. Babbage's Old Camp. It cleared off during the
night, but the clouds have come up again this morning and look very
threatening. Sent Herrgott to find the spring. The wind is still from the
same quarter, and too strong for me to do anything to the plan, which is
a great annoyance. I will finish the survey of the runs from this place,
and send Campbell back to Oratunga with the plan. Herrgott did not return
until after sundown: he could not find the spring.

Monday, 16th May, Same Place. Sent Muller to the west; he returned at 10
o'clock, having found the spring about two miles and a half distant from
the camp; it is not hot, but a little warmer than milk-warm. There is a
good stream running from it, and the water is excellent; to me it has a
mineral taste, very good. There were some small fish lying dead on the
bank, near the mouth; they seemed to have been left there by the retiring
of the flood--they were quite dried up. I intended to have taken some
with me, but they were too dry--nothing but skin and bone. The creek
empties itself into the lake, about a mile north from where Chambers
Creek goes into it.

Tuesday, 17th May, Same Place. Again very cloudy, with a little rain.
Busy finishing the survey. Could not obtain an observation of the sun.
Wind still very strong.

Wednesday, 18th May, Same Place. Weather clearing up. Engaged with
survey.

Thursday, 19th May, Same Place. Finishing tracings, etc.

Friday, 20th May, Same Place. At sunrise started Campbell for Oratunga
with tracings, letter, etc., with orders to proceed to Finniss Springs,
thence to Herrgott Springs, thence to St. A'Becket's Pool, thence to
Mount Glenns, thence to Mount Stuart, and thence to Oratunga, taking six
days to perform the journey. Preparing my other plans for a start
to-morrow for the north-west, to see what the Davenport range is.
Latitude, 29 degrees 23 minutes 20 seconds.

Saturday, 21st May, Same Place. Started at 8 o'clock on a bearing of 310
degrees for the Davenport range. At twenty-two miles changed our course
to examine a large lagoon to the south-west of us, bearing 238 degrees.
At two miles reached the lagoon, which we examined for springs, but found
none. I suppose it receives Major Warburton's salt creek. It is caked
with a crust of salt, and is dry; it is seven miles long by three broad,
running north-west and south-west. On the south-west side it is bounded
by steep cliffs, and high sand hills on the top. Changed to 310 degrees,
our original course. Came upon some rain water at four miles, and camped
for the night. Distance to-day, twenty-eight miles.

Sunday, 22nd May, Rain Water. Sent Herrgott to examine the south-west
side of the lagoon which we passed last night, with orders to overtake me
by 11.30, so that I may get an observation of the sun at noon. The horses
having strayed some distance during the night, our start was delayed
until 9.15. Started on the same bearing as yesterday, 310 degrees.
Stopped at 11.20 for Herrgott to come with the instruments, but he did
not come up until 1.15, so that I lost my observation. I had told him, if
there was no appearance of springs not to go far, but to return
immediately; instead of which he went round the lagoon. Camped on a stony
rise, with a little wood. Distance to-day, twenty-one miles.

Monday, 23rd May, Stony Rise. Started towards the Davenport range. The
sand hills again commenced with beautiful feed upon them--low, with broad
valleys; they continued for five miles, when the stony plain again
commenced. The highest part of the range seems to be at the north-eastern
point, which has the appearance of a detached hill. At three miles and a
quarter from the last of the sand hills we saw the Douglas, and changed
our bearing to 328 degrees 30 minutes. At one mile and a quarter struck
the creek, but found no water in it. There were a number of gums, but not
very large, also plenty of myalls there. The bed of the creek is bad, and
will not retain water. We followed it down for three miles to see if
there was water; but no sign of it, the creek still continuing broad and
sandy. I was obliged to return to where I struck it, because it was
nearly sundown, and I had found a little rain water about a mile to the
south, which would do for the horses in the morning.

Tuesday, 24th May, The Douglas. Herrgott's horse in want of shoes. Could
not get a start until late. Found a little more rain water in a clay-pan.
If I can find no water near the range, I shall have to fall back upon
Strangway Springs. I am anxious to see what is on the other side of the
range, or I would run this creek down. There are numerous tracks of
natives about the creek; we have also seen three fires three or four days
old. Latitude, 28 degrees 45 minutes 4 seconds. Started at 12.30 on a
bearing of 313 degrees for the highest point of the range east, over
stony table land. The creek runs in the same direction for four miles, it
then turns to the westward, and is lost sight of among some hills. At ten
miles struck a stony box-tree creek; its bed was sand and gravel, but no
water. At 11.30 descended from the table land, and camped at a gum creek
at sundown; the bed the same as the last, and no water. There were
numerous native foot-tracks here also. I am sorry I could not reach the
range to-night, but we had some very bad ground to travel over, and no
water.

Wednesday, 25th May, Dry Gum Creek. Examined the creek for water, but
found none. Started on the same course as yesterday, 313 degrees, for the
north-east highest point, which I suppose to be the Mount Margaret of
Major Warburton. Native tracks seen in the creek. There may be water some
distance down the creek, but here it is too sandy to retain it. At four
miles struck another gum creek in turning round the south side of the
range; it was of the same description as the others, too sandy to hold
water. Proceeded towards the highest point of the range, and obtained an
observation of the sun within a mile and a half of the mount. Left the
horses in charge of Muller and ascended the mount, which was very
difficult; it took us an hour to go up, and three-quarters of an hour to
come down. The hill is composed of a greenish slate, lying horizontally
at the base, and courses of quartz and granite, with ironstone; but I can
see nothing of Major Warburton's quartz cliffs; they must be more to the
south-west. The range has a very peculiar appearance from a short
distance off; it seems to be an immense number of rugged conical hills
all thrown together. From the top, the view to the north-west was hidden
by a higher point of the range. To the north-north-west there is another
range, about twenty miles distant, apparently higher than this, running
south-west and north-east. To the north is another far-distant range; to
the east, broken hill and stony plain, with a number of clay-pans. A
number of creeks run to the eastward from this range; they become gum
creeks further down, but in and close to the range they have myall
bushes, and other shrubs. No water to be obtained in this range. Changed
my course to the north-east to examine a white clay-pan that I thought
might contain some fresh water. At three miles came upon it, and was very
much disappointed to find it salt. This being the second day that the
horses have been without water, I must give up the search for springs and
return to one mile south of the Douglas, where we had found a little rain
water. It being nearly sundown, I made for the last large gum creek,
striking it lower down, also cutting the other creeks between, hoping to
find water in some, but there was none. Made the large gum creek at 10
o'clock. Camped for the night. Horses very much done up, in consequence
of the ground that we have been travelling over being so rotten and
stony. The country is not good, nor the range; but at three miles to the
east it becomes less stony and better grassed. No water.

Thursday, 26th May, Large Gum Creek. Started at daylight for beyond the
Douglas. At 3 o'clock arrived at water. Horses so much done up that I
shall require to give them two days' rest, if the water will hold so
long, and then I must return to the Strangway Springs, as we know that to
be permanent water. There are some heavy clouds coming up from the
south-west, which I hope will bring rain.

Friday, 27th May, The Douglas. Rain all gone after a slight shower, which
did not assist me much. Very sorry for it.

Saturday, 28th May, The Douglas. Horses looking better this morning, so I
will give them this day also. I have sent Muller down the creek to the
eastward, to see if there is any water in it. I should have gone again
to-day to the Davenport range, to see if I could find the quartz reefs by
striking it more to the south-west, but it would be too much for the
horses, which are my mainstay, and this water will not last longer than
to-day; it is going very fast. I do wish to goodness it would rain, for I
do hate going back. Muller returned at sundown. He has been about twelve
miles down the creek, but can find no water. It still continued sandy. He
shot three new parrots.

Sunday, 29th May, The Douglas. Not being satisfied with my hurried
examination of the range, I shall make another attempt to-day, and
endeavour to find water. If we do not succeed we must fall back upon the
springs. Started on a course of west-north-west. Crossed the Douglas
three times. It turned to the south-west, but I continued my course, over
low hills and valleys, with plenty of feed, with quartz, ironstone, and
granite. At fifteen miles changed a little more to the north towards a
rise. The country becomes very broken and rough, but still plenty of
grass. At twenty miles crossed the upper part of the gum creek that I
camped on on the 25th instant. The banks are nearly perpendicular cliffs
of slate. Followed it up for two miles, but no water. I continued my
course for the rising ground. At six miles I found that I was getting
upon high table land; so, as the sun was nearly down, I returned to the
creek, where there is some green feed for the horses, as they will be
without water to-night. It was after sundown before I reached the creek
and camped. I have named this creek Davenport Creek, after the Honourable
Mr. Davenport, M.L.C.

Monday, 30th May, Davenport Creek. Started at sunrise determined to
follow down the creek, for I think there must be water somewhere before
it enters the plain. The flow is to the east. At five miles came upon a
beautiful spring in the bed of the creek, for which I am truly thankful.
I have named this The Spring of Hope. It is a little brackish, not from
salt, but soda, and runs a good stream of water. I have lived upon far
worse water than this: to me it is of the utmost importance, and keeps my
retreat open. I can go from here to Adelaide at any time of the year, and
in any sort of season. Camped for the rest of the day. Latitude, 28
degrees 33 minutes 34 seconds.

Tuesday, 31st May, The Spring of Hope. Shoeing horses, and repairing
various things.

Wednesday, 1st June, The Spring of Hope. Not being satisfied with my
hurried view of the salt clay-pan that I visited on the 25th ultimo, I
have sent Muller to-day to examine it for springs, before I proceed to
the north-west. On a further examination of this water, I find a very
large portion of magnesia in it, and also salt, but very little. Muller
has returned, having been down the creek, and, as I expected, has found a
small spring of very good water on the banks of the salt creek. I expect
there will be others. I shall move down there to-morrow and examine it. I
expect we have fallen upon the line of springs again, which I hope will
continue towards the north. No rain seems to have fallen here for a long
time.

Thursday, 2nd June, The Spring of Hope. Started at 9 o'clock for the
springs, and arrived there in the afternoon. Travelled over a stony but
very good feeding country, which became better as we approached the
springs. There is a creek with a large water hole, and around the small
hills are numerous springs. On the banks of the creek and round the
springs an immense quantity of rushes, bulrushes, and other water-plants
are growing. The quantity of land they cover is very great, amounting to
several square miles. Some of the springs are choked up, others are
running, though not so active as those further to the south. Round about
them there is a thin crust of saltpetre, magnesia, and salt. The water of
these springs is very good, but that of the creek is a little brackish,
but will do very well for cattle. Some of the holes in the creek are
rather salt. There is enough of good water for the largest station in the
colony. Round the small hill, where I am now camped, there are twelve
springs, and the water is first-rate. I have named them Hawker Springs,
after G.C. Hawker, Esquire, M.L.A.* (* Now the Honourable G.C. Hawker,
Speaker of the House of Assembly at Adelaide.) The hills are composed of
slate, mica, quartz (resembling those of the gold country), and
ironstone. Latitude, 28 degrees 24 minutes 17 seconds. One of the horses
seems to be very unwell to-day; he has endeavoured to lie down two or
three times during the journey, but I hope he will be better by the
morning.

Friday, 3rd June, Hawker Springs. I find that the horse is too unwell to
proceed. I shall give him another day, for fear I should lose him
altogether. I sent Muller to see if there are any springs round the hill
about six miles to the east. He states that the creek flows past that
hill, and on towards other hills of the same kind. The springs continue
to within half a mile of the hill, where he found two large springs
running over, covered with long reeds. I do not doubt but that they still
continue on towards the lake, (wherever that may be), which I intend to
examine on my return.

Saturday, 4th June, Hawker Springs. This morning the horse does not look
much better, but still I must push on. Started at 8 towards the highest
point of the next range. At one mile struck a gum creek coming from the
Davenport range, and running to the north of east; the bed sandy and
grassy. At four miles another gum creek of the same description, with the
gum-trees stunted. At eight miles and a half struck three creeks joining
at about a quarter of a mile to the east; the centre one is gum, and the
other two myall. At twelve miles changed my course to 29 degrees to
examine three dark-coloured hills, where I think there will be springs.
At a mile and a quarter came upon a small batch of springs round the
north side of the hills in a broad grassy valley, with plenty of good
water. Changed my course again to 318 degrees towards the highest point
of the range. At one mile a myall and gum creek; at three miles another
gum creek; at seven miles a very large and broad gum creek, spread out
into numerous channels. I have not the least doubt but there is water
above and below, judging from the number of tracks of natives and emus
that have been up and down the creek. As this is the largest creek that I
have passed, and is likely to become as good as Chambers Creek, which it
very much resembles, I have called it The Blyth, after the Honourable
Arthur Blyth. I have named the range to the east The Hanson Range, after
the Honourable R.D. Hanson. At nine miles and a half attained the highest
point of the range, and built a cone of stones thereon, and have named it
Mount Younghusband, after the Honourable William Younghusband. From it I
had a good view of the surrounding country, which seems to be plentifully
supplied with springs. To the north-west is another isolated range like
this; I should think it is about seven hundred feet high. I have named it
Mount Kingston, after the Honourable G.S. Kingston, Speaker of the House
of Assembly. To the north the broken ranges continue, and in the distance
there is a long flat-topped range, broken in some places. It seems to be
closing upon my course on the last bearing. I cannot judge of the
distance, the mirage being so great. Descended from the mount, and
proceeded on a bearing of 336 degrees towards a spring that I saw from
the top. As we were rounding the mount to the east, we found eight
springs before we halted, in a distance of three miles; some were
running, and others were choked up, but soft and boggy. At dark arrived
at another batch of springs--not those that I intended going to--they are
on the banks of a small creek, close to and coming from the range; they
are not so active as the others, and taste a little brackish; they are
coated with soda, saltpetre, and salt. The horse seems to be very ill; he
has again attempted to lie down two or three times. I cannot imagine what
is the matter with him.

Sunday, 5th June, Mount Younghusband. I must remain where I am to-day;
the horse is so bad that he cannot proceed; he neither eats nor drinks. I
have sent Muller to the west side of the mount to see the extent of the
springs; they are on the banks of a creek which has brackish water in it,
large and deep, and a great quantity of rushes. The water comes from the
limestone banks which are covered with soda. He rode round the mount: it
is all the same, and the feed is splendid right to the top of the mount.
It is a wonderful country, scarcely to be believed. I have had one of the
springs opened to-day, and the water to-night tastes excellent; it could
not be better. Native tracks about; I am surprised we see none of them;
we are passing old fires constantly. Latitude, 28 degrees 1 minute 32
seconds.

Monday, 6th June, Mount Younghusband. The horses being some distance off,
and my horse requiring a shoe, I was unable to make a start until 10
o'clock, on a bearing of 307 degrees 45 minutes, passing Mount Kingston
on the south-west side. At three-quarters of a mile came upon the springs
that I intended to have camped at on Saturday night: they are flowing in
a stream strong enough to supply any number of cattle. I named them The
Barrow Springs, after J.U. Barrow, Esquire, M.L.A. At four miles and a
half struck a large broad valley, in which are the largest springs I have
yet seen. The flow of water from them is immense, coming in numerous
streams, and the country around is beautiful. I have named these The
Freeling Springs, after the Honourable Major Freeling, M.L.C. After
leaving the springs I ascended a rough stony hill, to have a view of
them, but I could not see them all, their extent is so great. They extend
to under the Kingston range, and how much further I do not know. From
this point I changed my course to 322 degrees. I can just see the top of
a distant range, for which I will go on that bearing. At one mile and a
half crossed a broad gum salt creek, coming from the west, with a
quantity of salt water in it. I have named this Peake Creek, after C.J.
Peake, Esquire, M.L.A. After crossing this, we travelled over low rises
with quartz, ironstone, and slate; the quartz predominating. Herrgott and
Muller, who have both been long in the Victoria gold diggings, say that
they have not seen any place that resembles those diggings so much as
this does. The country seems as if it were covered with snow, from the
quantity of quartz. At eleven miles passed a brackish water creek and
salt lagoon; searched for springs but could find none, although reeds and
rushes abound, but no water on the surface. I thence proceeded
three-quarters of a mile, and struck a gum creek with a number of
channels and very long water holes, but the water is brackish; it might
do for cattle. This I have named The Neale, after J.B. Neale, Esquire,
M.L.A. I think by following it down, there will be a large quantity of
water, and good, and that it will become a very important creek. No
person could wish for a better country for feed than that we have passed
over to-day; it resembles the country about Chambers Creek.

Tuesday, 7th June, The Neale. At 8 o'clock started on a bearing of 180
degrees for the northernmost of the isolated hills, to see if there are
springs round it. At four miles ascended it, but could see no springs.
This I have named Mount Harvey, after J. Harvey, Esquire, M.L.A. from
Mount Kingston it bears 47 degrees 45 minutes. Thence I started for the
other mount, which I have named Mount Dutton, after the Honourable F.T.
Dutton; four miles and a half to the top. The Hanson range is closing
upon my course, and I think to-morrow's journey will cut it. On the north
side are a few springs, some of them a little brackish, and some very
good. We cleared out one, and found it very good. Here I camped for the
night. From south-west to north-west it seems to be an immense plain,
stony on the surface, with salt bush and grass. Mount Dutton is well
grassed to the top; it is composed of the same rock as the others.

Wednesday, 8th June, Mount Dutton. at 9.15 started on a course of 310
degrees. At three-quarters of a mile passed another batch of springs,
some of them brackish, and some very good indeed. Leaving them we passed
over a good feeding country, crossing several gum and myall creeks, one
with polyganum, all coming from Hanson range and flowing into the Neale.
At nine miles crossed the top of Hanson range. From it I could see, about
fifteen miles to the west of north, a high point of this range, which I
have named Mount O'Halloran (after the Honourable Major O'Halloran), on
the west side of which there appears to be a large creek coming from the
north-west. We then proceeded on a course of 324 degrees towards Mount
O'Halloran. At four miles and a half struck a large gum creek coming from
the range and running for about four miles north-west on our course;
examined it for water, but found none. It divides itself into numerous
channels, and when full must retain a large quantity of water for a long
time. The gum-trees are large and numerous, and numbers of pigeons
frequent its banks. At a mile further came upon some rain water in a
stony flat, where we camped for the night between low sand rises covered
with grass.

Thursday, 9th June, Stony Flat. This country must be examined today for
springs. I have therefore sent Muller down the creek to search that,
whilst I must remain and get an observation of the sun. My party is far
too small to examine the country well. I cannot go myself and leave the
camp with the provisions to one man; the natives might attack him, and
destroy the lot, there seem to be a great many tracks about. Three
o'clock. Muller has returned; he has run the creek down until it joined
another very large gum creek coming from the north-west--the one that I
saw from the top of the range. The gum-trees were large; from one of them
the natives had cut a large sheet of bark, evidently for a canoe. He also
saw two large water holes, one hundred yards wide and a quarter of a mile
long, with very high and steep banks. It seems to be the same creek as
the Neale. Can it be Cooper's Creek? the country very much resembles it.
My course will strike it more to the north-west to-morrow.

Friday, 10th June, Same Place. I have been very unwell during the night
with cramp in the stomach, but hope I shall get better as I go on.
Started at 8 o'clock on a bearing of 32 degrees 4 minutes. At four miles
went to the top of Mount O'Halloran. The creek is about three miles to
the west; it breaks through the Hanson range. Changed my course to 317
degrees to get away from the stones, which are very rough close to the
hill. At six miles changed my course to 270 degrees to examine an
isolated hill for springs, but found none. The creek winds round this
hill, and spreads out into numerous channels, covering a space of two
miles; but there is no water here, nor for three miles further up the
creek. We have, however, found some rain water; and, as I feel so unwell
that I am unable to ride, I have camped here for the night, and sent
Muller to examine the creek for water. He has been unsuccessful.

Saturday, 11th June, Rain Water. I feel a little better this morning.
Started at 9.20 on a bearing of 317 degrees. Crossed the creek, which is
about a mile wide. For five miles it ran parallel to my course, and then
turned more to the west. There is a beautiful plain along the bank, about
three miles wide, and completely covered with grass. At nine miles and a
half, on a small rise, changed my course to 318 degrees 30 minutes, to a
distant hill. Travelled for nine miles and a half over another large and
well-grassed plain of the same description; thence over some low stony
hills to a myall flat, the soil beautiful, of a red colour, covered with
grass; after four miles it became sandy. Camped for the night, after
having gone thirty-one miles. The country of to-day surpasses all that I
have yet travelled over for the abundance of feed. We have passed a
number of native tracks, but only one or two are fresh. We have found no
water to-day, except some little rain water, which is nearly all mud. I
have no doubt but there is plenty towards the east.

Sunday, 12th June, Myall Flat. I feel still very unwell. We are now come
to our last set of shoes for the horses, and, having experienced the
misery of being without them in my previous journey, I am, though with
great reluctance, forced to turn back. My party is also too small to make
a proper examination of such splendid country. Started back, keeping more
to the east to examine a high hill in search of water. If I can find
water, I shall endeavour to reach the north boundary. At 11.40 arrived at
the hill. Latitude, 27 degrees 12 minutes 30 seconds. Can see no
appearance of water, although the country seems good all round. Ten
degrees to the east of north is a large dark-coloured hill, which I saw
from last night's camp, from fifteen to twenty miles distant. I should
like to go to it, but can find no water. I have named it Mount Browne,
after Mr. J.H. Browne, of Port Gawler, my companion in Captain Sturt's
expedition. I dare not risk the horses another night without water, the
grass is so very dry; had there been green grass, I would not have
hesitated a moment. Turned towards the Neale by a different course to try
and find water; was unsuccessful until within an hour of sundown, when we
struck some muddy water. As I expected, the horses were very thirsty and
devoured the lot. Reached the creek after dark.

Monday, 13th June, The Neale. Found some rain water on the bank of the
creek, and, two of the horses requiring shoes, I stopped for the day. At
noon sent Muller up the creek to see if he could find any water holes,
but he saw none. At six miles another creek coming from the south-west
joins this. I am afraid I shall not have enough shoes to carry me into
the settled districts. The creek does not seem to have been running for a
number of years. The water has, some time or other, been ten feet high.
The breadth of the plain where the channels flow is a mile and a half,
and the quantity of water must be immense. It drains a very large extent
of country. After examining the country during the next two or three
days, I shall endeavour to follow this creek down, and learn where it
empties itself.

Tuesday, 14th June, The Neale. Started at 9 o'clock. Running the creek
down. At eight miles crossed another branch of the creek coming from the
south of west. We found no water. At twelve miles changed my bearing to
south. At three miles and a half camped at the two water holes that
Muller found when I sent him to examine the creek on the 9th instant. I
can not with certainty say they are permanent, there are neither reeds
nor rushes round them; they are very large and very deep, and, when
filled with rain, must hold a large quantity of water for a long time.
There are ducks upon them. The water does not taste like rain water,
which leads me to think that it may be permanent and supplied by springs
from below.

Wednesday, 15th June, Water Holes found by Muller on the 9th. Started at
9.15 a.m. Following the creek down. As we approached Hanson range, where
it broke through, we came upon two nice water holes with ducks upon them.
They are long, wide, and deep, with clay banks, and about three feet of
water in the middle. There are no reeds nor rushes round them, and it is
doubtful whether they are permanent. At seven miles and a half the creek
winds a little more to the west. Shortly afterwards we struck (in the
gap) two very long and large water holes a quarter of a mile long, and
between forty and fifty yards wide, and very deep. These I may safely say
are permanent. After getting through the range, the creek spreads out
over a large plain in numerous courses, bearing towards the south-east.
At four miles and a half changed my course. At six miles, going more to
the east, changed again, and at eight miles camped for the night, without
water. We have found no water since leaving the last water hole, although
I do not doubt of there being some. It would have taken us too long a
time to examine it more than I have done, my party being so small. We
have passed several winter worleys of the natives, built with mud in the
shape of a large beehive, with a small hole as the entrance. Numerous
tracks all about the creek, but we see no natives. We are now approaching
the spring country again.

Thursday, 16th June, The Neale. Started at 11.15. Still following the
creek, which continues to spread widely over the plain. At five miles I
observed some white patches of ground on the south-west side of Mount
Dutton, resembling a batch of springs. I changed my course and steered
for them, crossing the Neale at two miles and three-quarters. On the
south-west side of the Neale the country is rather stony, and for about a
mile from it the feed is not very good, in consequence of its being
subject to inundation, but beyond that the feed is beautiful. At three
miles and a half made the white patches, and found them to be springs
covering a large extent of country, but not so active as those already
described. Leaving the springs at two miles, crossed the Neale at a place
where it becomes narrower and the channel much deeper, with long sheets
of salt and brackish water. I shall now leave the creek. In the time of a
flood an immense body of water must come down it. At the widest part,
where it spreads itself out in the plain, the drift stuff is from
fourteen to fifteen feet up in the trees. Camped at 4 p.m.

Friday, 17th June, The Neale. Discovered another large quantity of water
supplied by springs. This country is a wonderful place for them. There is
an immense quantity of water running now.

Saturday, 18th June, The Neale. Started early in the morning to examine
the country. Found large quantities of quartz, samples of which I brought
with me. Still well watered, but without any timber.

Sunday, 19th June, The Neale. Water in abundance, with large quantities
of quartz. The course the quartz seems to take is from the south-west to
the north-east. The plain we examined to-day is a large basin, surrounded
by the hills from Mount Younghusband and Mount Kingston, with the creek
running through the centre. To-morrow I shall have a look along the
north-east side of Mount Kingston, for I see the quartz apparently goes
through the range and breaks out again on the north-east side, which is
very white.

Monday, 20th June, Mount Kingston. Started at 8 o'clock a.m. to examine
the quartz on the east side of Mount Kingston. Crossed the creek, and at
three miles struck a quartz reef. The Freeling Springs still continue,
but seem inclined to run more to the eastward. Changed my course to a
peak in a low range which has a white appearance. At eight miles reached
the peak; the quartz ceases altogether, and the country is stony from
here. I can see the line of the Neale running eastward; it spreads out
over the plain. It was my intention to follow it until it reached the
lake, but I find the ground too stony for me to do so. Being reduced to
my last set of shoes, and some of them pretty well worn out, I am obliged
to retreat. Changed my course at seven miles across the bed of the creek,
three miles broad, with a number of brackish water holes in it, some very
salt. At this point the trees cease. I can see nothing of the lake.
Camped on a gum creek without water. The latter part of our course was
over a very barren and rotten plain, surrounded by cliffs of gypsum,
quite destitute of vegetation. It has evidently been the bed of a small
lake at some time. There is no salt about it.

Tuesday, 21st June, Dry Gum Creek. At 7.40 started on the same course as
last night, and after various changes of bearings arrived at the hill,
whither I had sent Muller, and where he found two springs. Instead of
two, they are numerous all round the hill; some are without water on the
surface, and others have plenty. It is a perfect bed of springs. A little
more east they are stronger, surrounded with green reeds and rushes.

Wednesday, 22nd June, Mount Younghusband. Started at 8.40. At three miles
and a half came to a large bed of springs with reeds and rushes, water
running and good, with numerous other small springs all round. They are a
continuation of those we camped at last night, with an abundant supply of
excellent water. At four miles crossed the salt creek coming from Hawker
Springs. At eight miles crossed three salt and soda lagoons, surrounded
by lime and gypsum mounds, in which are numerous springs up to the foot
of the hills (ten miles and a half) and all round them. I have named
these hills Parry Hills, after Samuel Parry, Esquire. It was my intention
to have gone to the east from this, but the horses' shoes will not admit
of it. To the south-east I observed three conical hills, for which I will
now steer. At seven miles crossed a gum creek, in which are large water
holes, where water had been lately, but there is now only mud. There must
be water either up or down the creek, for there are numerous native
tracks leading both ways. At ten miles crossed a large gum (stunted)
creek with abundant springs of rather brackish water. At nineteen miles
and a half camped on a broad creek, but no water. The country good.

Thursday, 23rd June, Dry Creek. Started at 8.30 on the same course for
one of the conical hills. At three miles ascended it, and found it to be
flat-topped. I can see nothing of any lake to the east. The view is
interrupted by a flat-topped range. From this I changed my course, and at
three miles and a half observed a peculiar-looking spot to the
south-west, which had the appearance of springs. Changed my course for
it, and at six miles came upon a hill of springs surrounded by a number
of smaller ones, with an ample supply of first-rate water. The hill is
covered with reeds and rushes; it is situated at the west side of a large
plain, and is bounded by stony table land on the east side, which has an
abrupt descent of about thirty feet into the plain. On the west side are
a number of broken hills, and a small range composed of gypsum and lime,
having the surface covered with fragments of quartz and ironstone, and a
number of other pebbles. On the hill where the springs are we have found
lava. There are numerous small creeks coming from the hill, and running
in every direction. They seem to be all in confusion. The plain is about
five miles wide. These I have named the Louden Springs.

Friday, 24th June, Louden Springs. I must remain here to-day, and put the
last of the shoes upon some of the horses which are getting rather lame.
I have been making them go without as long as I can.

Saturday, 25th June, Louden Springs. Started at 7.50. At 8.45 (three
miles) crossed a gum creek, and at 12 o'clock (eleven miles) crossed the
Douglas, but no water. The channel still broad and sandy.

Sunday, 26th June, The Douglas. Started at 8.25, on a bearing of 217
degrees. Crossed the lagoon, which was rather boggy in some places. It is
now more than two miles broad, with a white crust on the top, composed of
soda and salt, but mostly salt. It must be supplied by springs. At three
miles crossed a salt creek, with salt water. It empties itself into the
lagoon, and is the same that passes by the Strangway Springs. I can see
nothing of any springs at this part of the creek. Steered upon the same
course to intersect my outward tracks. Saw some natives walking along a
valley. They did not observe us. I hailed them, and an old man came up to
us. He was rather frightened, and trembled a good deal. He seemed to
wonder and be pleased at my smoking a pipe of tobacco. I gave one to him
and a piece of tobacco, but he did not know how to manage the cutting,
filling, and lighting operations. I did these for him. In the first
attempt he put the wrong end into his mouth, which he found rather hot,
and quickly took it out. I then showed him the right end. He managed a
whiff or two, but he did not fancy it. He seemed very much pleased with
the pipe, which he kept. I then made him understand that I wanted water.
He pointed the same course that I was steering. In a short time another
made his appearance in the distance. By a little persuasion from the old
fellow, he was induced to come up, and in a short time became very
talkative, and very anxious to show us the water. In a few minutes a
third made his appearance, and came up. He was the youngest--a stout,
able-bodied fellow, about twenty-four years old. The others were much
older, but were very powerful men, and all three in excellent condition.
The women did not come up, but remained in the flat. I expected they were
going to take us to some springs, and was disappointed when they showed
us some rain water in a deep hole. They were quite surprised to see our
horses drink it all. They would go no further with us, nor show us any
more, and, in a short time after, left us. We struck our outward tracks,
and steered for the Elizabeth Springs, where we arrived after dark.

Monday, 27th June, Elizabeth Springs. Gave the horses a half-day, and
made the Mount Hamilton Springs in the afternoon.

Tuesday, 28th June, Hamilton Springs. Started for Chambers Creek to my
first encampment. Arrived there in the afternoon. Distance, eighteen
miles.

Wednesday, 29th June, Chambers Creek. Resting the horses and preparing
for a trip down on the west side of Mount North-west, to see if I can
find a road and water that way.

Friday, 1st July, Chambers Creek. Started at 8 a.m. on a bearing of 120
degrees. At twenty-four miles camped on a water hole in Gregory Creek,
where it comes out of the hills. There are three remarkable peaks north
of the water, one in particular having a white face to the east, with a
course of black stones on the summit, distant about one mile. The first
five miles was over a well-grassed country, with stones on the surface,
slightly undulating, with a number of good valleys, very broad, emptying
themselves into Gregory Creek. At twenty-two miles crossed the main
channel of the creek. It is divided into a number of courses, with some
very deep holes in them. When they are filled, they must retain water for
a great length of time. There are a great many native encampments all
about the creek. The gums are dwarf.

Saturday, 2nd July, Gregory Creek. Started at 10.8. Course, 120 degrees.
At three miles, opposite a long permanent water hole, with rushes growing
round it. At seven miles, crossed the upper part of the Gregory; eight
miles and a half, top of dividing range; thirteen miles, crossed a creek
with rain water; fourteen miles, crossed another deep channel. Camped at
twenty-three miles, within twelve miles of Termination Hill. The country
for ten miles before we halted was very good.

Sunday, 3rd July. Rounded Termination Hill, and arrived at Mr. Glen's
station.


JOURNAL OF MR. STUART'S THIRD EXPEDITION (IN THE VICINITY OF LAKE
TORRENS). NOVEMBER, 1859, TO JANUARY, 1860.

Friday, 4th November, 1859. Started from Chambers Creek for the Emerald
Spring. At ten miles crossed nine fresh horse-tracks going eastward; I
supposed them to be those of His Excellency the Governor-in-Chief. I have
not as yet seen his outward track. Arrived at the spring before sundown.

Saturday, 5th November, Emerald Spring. Started at 7.30 on a course of
340 degrees. At seven miles and a half changed to 38 degrees, for three
miles to a high sand hill, from which I could see two salt lagoons, one
to the south and the other to the north; examined them, but could find no
springs. Next bearing, 18 degrees, to clear the lagoon, two miles and a
half sandy, with salt bush and grass. Changed to our first bearing, 340
degrees, for six miles, and then to 350 degrees, for five miles, when we
reached the top of a high hill, from which we could see the lake lying to
the north of us about three miles distant. Changed to 315 degrees for
three miles and a half to get a good view of the lake. This is a large
bay; from north-east to north-west there is nothing visible but the dark,
deep blue line of the horizon. To the north-north-east there is an island
very much resembling Boston Island (Port Lincoln) in shape; to the east
of it there is a point of land coming from the mainland. To the
north-north-west are, apparently, two small islands. A short distance to
the east of the horn of the bay there seems to be much white sand or salt
for two or three miles from the beach towards the blue water (on this
side of which there is a white line as if it were surf): this again
appears at the shores of the island, and also at the horn of the bay.
From the south shore to the island the distance is great; I should say
about twenty-five miles, but it is very difficult to judge correctly. At
three miles and a half camped at sundown, without water.

Sunday, 6th November, Lake Eyre. Got up before daybreak to get the first
glimpse of the lake, to see if there is any land on the horizon, and,
with a powerful telescope, can see none. It has the same appearance as I
described last night. I watched it for some time after sunrise, and it
still continued the same. After breakfast went to examine the shore:
course north, two miles and a half; found it to be caked with salt, with
ironstone and lime gravel. When flooded, at about fifty yards from the
hard beach, the water will be about three feet deep. I tried to ride to
the water, but found it too soft, so I dismounted and tried it on foot.
At about a quarter of a mile I came upon a number of small fish, all
dried and caked in salt; they seem to have been left on the receding of
the waters, or driven on shore by a heavy storm; they were scattered over
a surface of twelve yards in breadth all along the shore; very few,
especially of the larger ones, were perfect. I succeeded in obtaining
three as nearly perfect as possible; one measured eight inches by three,
one six inches by two and a half, and another five inches by two. They
resemble the bream. I should think this a sufficient proof of the depth
of the water. I then proceeded towards the water, but the ground became
soft, and the clay was so very tenacious and my feet so heavy, that it
was with difficulty I could move them, and so I was obliged to return.
The salt is about three inches thick, and underneath it is clay. I would
have tried it in some other places, but as my horses were without water
(and as I intend to visit this place again), I think it more prudent to
search for water for them, and, if I cannot find any, to return to the
camp. Started on a south course to examine the country for springs. At
six miles found we were running parallel to sand ridges, and no chance of
water. Changed to 160 degrees, crossed a number of sand ridges, but no
water, except a little rain water that we found in a hole. Proceeded to
the camp, and arrived there about sundown.

Monday, 7th November, Emerald Springs. Finding that the weevil is at work
with my dried beef, I must remain to-day and put it to rights. Prepared a
package with the fish, etc., to be left for Mr. Barker when he comes
here, to be sent to town. There are fish in this spring about three
inches long. We have also found a cold-water spring among the warm ones.

Tuesday, 8th November, Emerald Springs. Not being satisfied about one of
the lagoons I saw yesterday, I have sent Kekwick and Muller to see if
there are any springs, while I and the others proceed to the Beresford
Springs; they are to overtake me. Arrived at the springs at 3 p.m. We
could find no fresh water on our way, but plenty of salt and brackish in
the creek which we first struck at six miles from the Emerald Springs.
Sundown: the two men have not come up; they must have found something to
detain them; they had only to do about eight miles more than I had. I
expect they will arrive during the night.

Wednesday, 9th November, Beresford Springs. No signs of the two men; they
must have stopped at some water during the night. It is very tiresome to
be delayed in this way: what can they be about? At 12 noon they arrived;
they had passed my tracks and gone on to Mount Hugh instead of coming on
here. I will give their horses an hour's rest and go on to the Strangway
Springs. The Paisley Ponds are dry, but there is salt and brackish water
three miles lower down the creek. Started at 2 p.m., and at 5 p.m.
arrived at the springs, which are about ten miles from the Beresford.
They are upon a high hill about one hundred feet above the level of the
plains; there are a great number of them, and abundance of water, but
very much impregnated with salt and soda. My eyes are very bad.

Thursday, 10th November, Strangway Springs. Suffering very much from bad
eyes and the effects of the water of these springs; cannot help it, but
must go and examine the country to north-west and west. Sent Muller to
the east in search of springs, with instructions to strike my former
tracks and examine all the country between. Started at 7 a.m. with one
man, on a course of 315 degrees, and at one mile crossed a salt creek
with water; at three miles the sand hills commenced, crossing our course
at right angles. At 2 p.m. struck a large lagoon (salt) about two miles
broad and five miles long, running north-east and south-west, narrowing
at the ends; distance, fourteen miles; tried to cross it but found it too
boggy; rounded it on the south-west point, where we discovered a spring;
no surface water, but soft, and the same all round for about two acres
square, covered with grass reeds of a very dark colour and very thick,
showing the presence of water underneath. Proceeded round the lagoon to a
high hill, which seemed to have reeds upon the top of it; after a good
deal of bogging and crossing the bends of the lagoon, we arrived at the
hill, and found it to be very remarkable. Its colour is dark-green from
the reeds and rushes and water-grass which cover it. It is upwards of one
hundred feet high, the lower part red sand; but a little higher up is a
course of limestone. On the top is a black soil, sand and clay, through
and over which the water trickles, and then filters through the sand into
the lagoon. Where the water is, on the top, it is upwards of one hundred
feet long. Immense numbers of tracks of emus and wild dogs, also some
native tracks, all fresh. On the north-west side there is one solitary
gum-tree, and about half a mile in the same direction is another bed of
reeds, and a spring with water in it. All the banks round the lagoon are
of a spongy nature. I am very glad I have found this; it will be another
day's stage with water nearer to the Spring of Hope. We can now make that
in one day, if we can get an early start. By the discovery of springs on
this trip, the road can now be travelled to the furthest water that I saw
on my last trip from Adelaide, and not be a night without water for the
horses. The country to the south and south-east of the last springs
(which I have named the William Springs, after the youngest son of John
Chambers, Esquire), is sand hills and valleys, rich in grass and other
food for cattle. Thence I proceeded to hill bearing 10 degrees south of
north, distant three miles, from the top of which I could see no rising
ground to the westward, nothing but sand hills. Changed my course to
south, to a white place under some stony hills; at ten miles reached it,
and found it to be a salt creek, but no springs. The last ten miles were
through hills not so high as those I crossed on my way out, but more
broken, with plenty of feed. It is my intention to push for the Strangway
Springs tonight, so as to get an early start in the morning. Arrived at
10 p.m., found that one of the horses had not been seen all day;
something always does go wrong when I am away; I shall have to make a
search for him in the morning. My eyes very bad from the effects of the
glare of the sun on the sand hills, and the heat reflected from them, and
that everlasting torment, the flies.

Friday, 11th November, Strangway Springs. My eyes so bad I cannot see;
unable to go myself in search of the missing horse; despatched two of the
men at daybreak to circuit the spring, and cut her tracks if she has left
them. They have returned, but can see no tracks leaving the spring; she
must be concealed among the reeds; sent three men to examine them. They
found her at 1 p.m. Started at 2 p.m., and arrived at William Springs at
sundown. Distance, fourteen miles. By keeping a little more to the east,
the sand hills can nearly be avoided, and a good road over stony country,
with good feed, can be had to this spring.

Saturday, 12th November, William Springs. Very unwell, unable to move
to-day; I am almost blind and suffering greatly from the effects of the
water at Strangway Springs. As I wished to examine round this spring, I
remained here to-day; and, as I could not go myself, sent two of the men
in different directions. At sundown they returned, and reported that
there are no springs for ten miles distant from east-south-east to north.
To the east about three miles there is another lagoon resembling this
one, but not so large, and no springs; plenty of grass about a mile from
the lagoon. Saw two natives at a distance, but could not get near them.

Sunday, 13th November, William Springs. I feel a little better to-day,
but suffer very much from the eyes. I hope I shall be able to travel
to-morrow, for it is misery to remain in camp in the hot weather.
Latitude, 28 degrees 57 minutes 24 seconds. Variation, 4 degrees 47
minutes east.

Monday, 14th November, William Springs. Started on a course of 317
degrees for the Hope Springs, and arrived at 5 p.m. I kept to the west in
order to see what the country was in that direction, in the hope of
finding some more springs. At twenty-one miles crossed the Douglas,
coming from north-north-west; the country from it to the north-west and
north looked quite white with quartz, and showed signs of being
auriferous. From the Douglas to north-west the feed was not quite so
plentiful, salt bush with grass, the salt bush predominating; but as we
approached the Spring of Hope it improved, and became good as we neared
the creek. Distance, thirty miles.

Tuesday, 15th November, Spring of Hope. The spring is still good,
yielding a plentiful supply of water. Sent one of the men to the east and
south-east to examine some white patches of country that I saw on our
journey up here, while I, with one man and two days' provisions, started
south-west to a high and prominent hill in the range. At 11 a.m. arrived
at the top, from which I had a good view of the country all round. It is
a table-topped hill, standing on high table land, which is intersected
with numerous small watercourses, flowing towards the Douglas on the
south and west sides of the mount, which I have named Mount Anna. It is
compound of ironstone, quartz, granite, and a chalky substance, also an
immense quantity of conglomerate quartz and ironstone, which has the
appearance of having been run together in a smelting works. There are
also numerous courses of slate of different descriptions and colours; the
quartz, which exists in white patches, predominates, and gives the
country the appearance of numerous springs. These patches have deceived
me two or three times to-day. At twenty miles the sand hills begin again;
the country being rather poor, with a number of isolated hills, and also
some white chalky cliffs of twenty feet high and upwards. No water nor
appearance of any to the west for a considerable distance. Changed to the
north-west to look at some more white country. I am again disappointed;
it turns out to be quartz with low chalky cliffs, and a large quantity of
igneous stone. Country the same, with salt bush and a little grass in
places. I can see no inducement for me to go further, so I shall return
to the camp. Arrived after dark. My eyes are still very bad, and I suffer
dreadfully from them. To-day has been hot, and the reflection from the
white quartz and the heated stones was almost insufferable: what a relief
it was when the sun went down! Distance, forty-five miles.

Wednesday, 16th November, Spring of Hope. Still very ill, and unable to
go out myself. Sent Muller to examine the creek nearer Mount Margaret for
water; if he finds any near the mount, I shall move there, as it will be
nearer, for building the cone of stones on the top of the mount, than
Hawker Springs. Shod our horses, and built a small cone of stones on a
reef of rocks that runs along the top of a hill about half a mile
west-north-west from the spring, to which it will act as a land mark.
Muller has returned, and reports having found water in the other creek,
about five miles north-north-west from this; the water is in the centre
of the creek, in three or four holes, some of which are brackish, but one
of them is very good. A number of natives were camped about it, but took
to flight the moment they saw him; he tried to induce them to come near
him, but they would not; they appeared to be very much frightened, and
climbed up the cliffs to get out of his way. Plenty of feed between the
two waters; through the hills there is an abundance. I find the water
discovered to-day (which I have named The George Creek, after G.
Davenport, Esquire), will be of no advantage to me when building the cone
of stones; I shall therefore move to the Hawker Springs to-morrow.

Thursday, 17th November, Spring of Hope. Arrived at the Hawker Springs at
noon, and commenced the survey. Springs still good; some of them at this
point will require to be opened. We have opened one, and the water is
beautiful. Immense quantities of reeds and rushes. Built a cone of stones
on the hill at the westernmost spring.

Friday, 18th November, Hawker Springs. Building a cone of stones on the
top of Mount Margaret, and making other preparations for the survey.
To-day very hot, wind south-east; a great deal of lightning to the south.
Obtained bearings of the following points from the hill at Hawker
Springs--namely, Mount Margaret, Mount Younghusband, hill at Parry
Springs, Mount Charles, and Mount Stevenson.

Saturday, 19th November, Hawker Springs. Sent the party on to Fanny
Springs, where I intend to lay down my base-line. Went with Kekwick to
the top of Mount Margaret. This hill is composed of grey and red granite,
quartz, and ironstone; on the lower hill is a blue and brown stratum. I
then proceeded to examine the creeks running to the east; in following
one of them down we came upon another spring of water, running and very
good. The creek is bounded on both sides for about a mile by nearly
perpendicular cliffs, which appeared to get much lower and broken to the
west. It is situated about one mile north of Mount Margaret, and runs
into the Hawker Springs valley. Could see no more higher up. Followed the
creek down to the opening. Proceeded about half a mile, entered another
gorge, and rode up it about three-quarters of a mile; came upon another
spring, running also, water excellent. Numerous native camps in the
creek. Country the same as in the other creek; cliffs slate and not so
high, but more broken, with watercourses between them, through which
cattle could find their way to the tops of the hills, where there appears
to be plenty of grass; there is also an abundance at the mouth of the
gorge and on the plains. This creek also runs into the valley of the
Hawker Springs. Distance from Mount Margaret, two miles and a half, 8
degrees east of north. As it was getting towards sunset I found I must
make for the camp, which was about twelve miles off. Arrived after dark.
Springs still as good as when I first saw them. Very tired, having had a
very long day of it.

Sunday, 20th November, Fanny Springs. Got up at daybreak, and went to the
top of Mount Charles, on which I had ordered the men to build a cone of
stones after their arrival here yesterday. On my return to the camp the
men informed me that Smith had absconded during the night. He generally
made a practice of sleeping some little distance from the others, when I
did not see him lie down; I had checked him for it several times. It did
not appear that he had gone to sleep, but waited an opportunity to steal
away, taking with him the mare which he used to ride, and harness, etc.,
also some provisions. As I had started very early to walk to Mount
Charles, his absence was not observed until some time after I had left,
and being detained some hours on the top of the hill, in consequence of
the atmosphere being so thick that I could not obtain my observations, it
was 7 a.m. before I heard of his departure. That moment I sent Kekwick
for my own horse (he being the swiftest), and ordered him to saddle,
mount, pursue, overtake, and bring Smith back; but during the time he was
preparing, I had time to think the matter over, and decided upon not
following him, as it would only knock up my horse and detain me three or
four days. Smith must have started about midnight, for I was up taking
observations from 12.30 a.m. until daybreak, and neither saw nor heard
any one during that time. I could ill afford to lose the time in pursuing
him, situated as I was in the midst of my survey, and he being a lazy,
insolent, good-for-nothing man, and, worse than all, an incorrigible
liar, I could place no dependence upon him. We are better without him; he
has been a very great annoyance and trouble to me from the beginning
throughout the journey. What could have caused him to take such a step I
am at a loss to imagine; he has had no cause to complain of bad treatment
or anything of that sort; he never mentioned such a thing to the other
men, nor was he heard to complain of anything. Such conduct on an
expedition like ours deserves the severest punishment: there is no
knowing what fatal consequences may follow such a cowardly action. Had he
not stolen the mare, I should have cared little about his running away,
but I am short of riding horses and have a great deal for them to do
during the time I am surveying and examining the country. The vagabond
went off just as the heavy work was beginning, and it was principally for
that work that I engaged him. He put on a pair of new boots, leaving
those he had been wearing, evidently intending to push the mare as far as
she would go, expecting he would be pursued, and then leave her and walk
the rest. I expect, when he reaches the settled districts, he will tell
some abominable lie about the matter. If such conduct is not severely
dealt with, no confidence can be placed in any man engaged in future
expeditions.

Monday, 21st November, Fanny Springs. Kekwick and I commenced chaining
the base-line from the top of Mount Charles, bearing 131 degrees.
Distance chained, four miles thirty chains. I ordered H. Strong to come
to me with two horses, which he did about 1.30 p.m.; we had finished the
line, and were waiting for him. I had seen some country that looked very
much like springs, to the north-east, a mile or so from the line; went to
examine it, and found some splendid springs--one in particular is a very
large fountain, about twenty yards in diameter, quite circular and
apparently very deep, from which there is running a large stream of water
of the very finest description; it is one of the largest reservoirs I
have yet seen, three times the size of the one at the Hamilton Springs,
with abundance of water for any amount of cattle; the water is running a
mile below it.

Tuesday, 22nd November, Fanny Springs. Engaged chaining the base-line to
north-west. Saw some more springs a mile or two to the east; too tired to
examine them to-day. It is dreadfully hot. Returned to the camp at
sundown.

Wednesday, 23rd November, Fanny Springs. Finished the remaining part of
base-line. The line is ten miles and forty chains long, crossing the top
of Mount Charles.

Thursday, 24th November, Fanny Springs. Fixing the angles of runs. Found
another batch of springs close to north-west boundary of large run,
covering four or five acres of ground, with an immense quantity of reeds;
they are not so active as the others. The ground round about is very
soft, and the water is most excellent. After fixing the north-east
corner, I proceeded to examine the country beyond the boundaries of the
runs in search of springs. Having gone several miles north, I saw the
appearance of a lagoon north-east, for which I started, but on my arrival
found no springs round it. Still continued on the same course for a
considerable distance further to a high sand hill, from which we could
see the Neale winding through a broad valley. One part of the creek being
much greener than the other, I went to examine it, and found the green
appearance to be caused by fresh gum-trees, young saplings, rushes, and
other fresh-water plants and bushes. The creek spreads over the plain in
numerous channels, four miles wide, but the main channel has only
gum-trees, with a chain of water holes, some salt, some brackish. By
scratching on the bank where the rushes were growing we got some
beautiful water in the gravel, a few inches below the surface. There was
plenty of feed, and the wild currant, or rather grape, grew in great
abundance, and was very superior to any I had tasted before. There were
two kinds; one grew upon a dark-green bush, and had a tart and saltish
taste, the other grew upon a bush of a much lighter colour, the fruit
round and plump and much superior to the former; in taste it very much
resembled some species of dark grape, only a little more acid. From this
I went in a north-east direction to a mound I had seen on my former
journey, and found it to be hot springs with a large stream of warm water
flowing from them nearly as large as the Emerald Springs, and, as it
seemed to me, warmer. It was a very hot day, and I had been riding fast.
It was as much as I could bear to keep my hand in the spring for a few
minutes, six inches below the surface. I put in a staff about four feet
long, but could find no bottom--nothing but very soft mud; the staff came
up quite hot. It is a very remarkable hill. From the west side it would
be taken for a very high sand hill with scrub growing on it--in fact it
is so. The springs are not seen until the top is reached. From them all
the east side is covered with green reeds to the base of the hill. The
hot springs are near the top, and cold ones on one side to the south;
some at the bottom and some half-way up. There is a large lagoon to the
east, which I will examine when I move the party up to this, for I have
no time to-day. Returned towards the camp and fixed the north-west corner
of the second run; I am obliged to drive pickets into the ground to show
them. I would have built cones of stones, but could get none large enough
to do it with. Arrived at the camp very late; fourteen hours on
horseback.

Friday, 25th November, Fanny Springs. Started shortly after sunrise to
mark the other two corners of the two runs. On approaching the south-west
angle of the second run (Parry Spring run), I discovered three other
springs close to the boundary of the first run. Two of them are outside,
and one inside, or rather on the boundary. The latter is a large spring,
having seven streams of water coming from it, one large, the others
smaller. The other two have abundance of water, covered with reeds.
Proceeded and marked the other corners, but, having no stones, was
obliged to put down pickets. Returned to camp, keeping outside the south
boundary in search of springs, but found none. Crossed over table land,
salt bush and grass, with stones on the surface. Arrived at the camp a
little before sundown.

Saturday, 26th November, Fanny Springs. Started for Parry Springs. In the
evening commenced putting up a cone of stones on the northernmost hill.
The day was excessively hot. One great thing here is that the nights are
very cool, so that we are obliged to have a good fire on all night. We
have had one or two warm nights since I have been out this time. I
suppose the reason must be that a large body of water exists in the lake
not far distant from us, the wind coming from north-east. From north-west
to south-south-east the winds are generally cool. It is so cold in the
morning that the men are wearing their top-coats; the day does not get
hot until the sun is a considerable height.

Sunday, 27th November, Parry Springs. Cold wind this morning from the
east. In the afternoon the sky became overcast, the clouds coming from
the south-east.

Monday, 28th November, Parry Springs. Building a cone of stones on the
northernmost of the hills, fixing the south-east corner of run Number 2,
and moving to the hot springs. Arrived at sundown. Saw a number of holes
where the natives had been digging for water. Cleaned out one, and found
water at two feet from the surface, above the water in the creek. It is
very good. On examining this spring, I find there is a great deal more
water coming from it than from the Emerald Springs. The hot springs are
on the top of the sand hill, and the cold ones at the foot. There are
large quantities of the wild grape growing here, both red and white. They
are very good indeed, and, if cultivated, would, I think, become a very
nice fruit.

Tuesday, 29th November, Primrose springs. Surveying run. Sent Muller to
the north to a distant range, and Strong to the north-east to look for
springs. Towards evening both returned without being successful. They
passed over plenty of good feeding country, but the range is high and
stony, with very little grass, only salt bush. It is a continuation of
Hanson range, all table land.

Wednesday, 30th November, Primrose springs. Surveying, etc. North-east
corner of run Number 2 is about two miles west of the Neale. I scratched
a few inches deep from the surface in the gravel, and found very good
water. The wild grape is in abundance here, and grows as large as the
cultivated one. I have obtained some choice seeds.

Thursday, 1st December, Primrose springs. At daybreak started with
Kekwick to find the lake on an easterly course, keeping to south of east,
to avoid a soft lagoon. Travelled over a fair salt-bush and grass
country, with stones on the surface. In places the grass is abundant,
though dry. At seven miles the sand hills commenced; they are low, with
broad valleys between, covered with stone. On the sand hills there was
plenty of grass, and numerous native and emu tracks going towards the
Neale, which is to the south of us. At fourteen miles struck a gum creek
with salt water. Searched for springs, but could find none with
fresh-water. Continued on a course east over sand hills and stony plain,
and at twenty miles crossed the Neale. It is very broad, with numerous
channels. In the main one there was plenty of water, but it was very
brackish. We scratched a hole on the bank about two feet from the salt
water, and found plenty of good water at six inches from the surface, of
which our horses drank very readily. This seems to be the mode in which
the natives obtain good water in a dry season like this. The emus and
other birds also adopt the same plan. An immense quantity of water must
come down this creek at times. The drift stuff was upwards of thirteen
feet high in the gum-trees. A number of native tracks all about the
creek, quite fresh, but we could not see any one. After giving our horses
as much water as they would drink, we crossed the creek, which now runs
north, and proceeded, still on our easterly course, over stony plains for
four miles, then over sand hills, which continued to the lake, which we
struck at thirty-five miles. The atmosphere is so thick, it is impossible
to say what it is like to-night. Camped without water under a high sand
hill, so that I may have a good view of the lake in the morning. I like
not the appearance of it to-night; I am afraid we are going to lose it.

Friday, 2nd December, Lake Torrens. Got up at the first peep of day and
ascended the sand hill. I fear my conjecture of last night is too true. I
can see a small dark line of low land all round the horizon. The line of
blue water is very small. So ends Lake Torrens! Started on a course of 30
degrees west of north to where the Neale empties itself into the lake. At
seven miles struck it; found plenty of water, but very salt, with
pelicans and other water-birds upon it. Traversed the creek to the
south-west in search of water for the horses. At five miles came upon a
number of water-bushes growing on the banks of a large brackish water
hole. Scraped a hole about two feet from the bad water, and got good
water six inches from the surface for ourselves and horses. Gave them an
hour's rest and started on a west course for the camp, where we arrived
at 9.30 p.m. The country was similar to that on our outward route; feed
more abundant. At sundown we crossed the broad channel of a creek, with
moisture in the centre. Having neither time nor light to examine it
to-night, I must do so to-morrow, as I think there must be springs to
supply the moisture.

Saturday, 3rd December, Primrose Springs. Sent Kekwick to examine the
creek we crossed last night. I cannot go myself, for my eyes are so very
bad I can scarcely see anything. This is the first time I have had such a
long continuance of this complaint. I am trying every remedy I can
imagine, but each seems to have very little or no effect. At sundown
Kekwick returned, and reported having found the springs which supply the
creek, but they are salter than the sea, or the strongest brine that ever
was made. He brought in a fine sample of crystal of salt, which he got
from under the water, attached to the branch of a bush which had blown
into it. The creek is the upper part of the first gum creek crossed
yesterday, and flows into the Neale, which accounts for the water being
so salt at the mouth of it. No fresh-water springs to be seen round
about.

Sunday, 4th December, Primrose Springs. Examining the Neale for
fresh-water springs. The water holes are abundant, but all more or less
brackish; plenty of rushes on the banks, where fresh water can be had by
scratching a little below the surface. I have not the least doubt but
there will be plenty of fresh water on the surface for a long time after
the creek comes down and sweeps all the soda and salt into the lake. It
is the rapid evaporation that causes it to be so brackish, and I should
think the consumption by stock would make a great improvement in it;
there would not be so much of it exposed to the sun, and the evaporation
would be much less. After considering the matter of having seen the
northern boundary of Lake Torrens, I am inclined to think I have been in
error. What I have taken for the lake may have been a large lagoon, which
receives the waters of the Neale before going into the large lake: I must
examine it again. After my surveys are completed, I shall move my party
down the creek to where we found the good water, and from there see what
it really is. I cannot bring my mind to think it is the northern boundary
of the lake.

Monday, 5th December, Primrose Springs. Moved the party down to the South
Parry Springs. My eyes are still very bad.

Tuesday, 6th December, South Parry Springs. Shortly after daybreak
started for Louden Springs, taking different courses, in search of more
springs, but can find none. Examined the George Creek, where the small
run is to be laid off; found some good water by scratching in the creek,
where there are plenty of rushes. A little before sundown we arrived at
the springs. I did not observe before that the higher springs on the top
of the hill are warm, but not nearly so hot as the others; the lower ones
are cold. Some other party has been here; we have seen their fresh tracks
and the place where they have camped; they seem to have been wandering
about a good deal before they found these springs.

Wednesday, 7th December, Louden Springs. Went to the top of Mount
Stevenson, built a cone of stones, and obtained bearings to fix it. No
appearance of any springs to the east of this, nor of the lake.

Thursday, 8th December, Louden Springs. Surveying and building
trigonometrical station on a light-coloured hill to the south of this. My
eyes very bad; can scarcely see; can do nothing.

Friday, 9th December, Louden Springs. Nearly blind; dreadful pain; can do
nothing to-day; no sleep last night.

Saturday, 10th December, Louden Springs. All yesterday the wind was hot
and strong from west and north-west; heavy clouds from south and
south-west. In the evening the wind changed to south. This morning still
the same; heavy clouds from same direction. My eyes are a little better,
so that I shall be able to do something. The sky being overcast I shall
put up some of the corners of this run.

Sunday, 11th December, Louden Springs. Still cloudy, but no rain.

Monday, 12th December, Louden Springs. Still very cloudy; wind south;
heavy clouds to north-west; no rain. Finishing the east boundary of
Number 3 run. Can find no more springs in or about this run. At sundown
still very cloudy, but no rain.

Tuesday, 13th December, Louden Springs. Started at 7.15 a.m. to find the
lake on an east course. The horses being a long distance off, it was late
before they came up. At nine miles crossed the gum creek running north,
spread out in a broad valley into numerous courses rich in food for
cattle. At twelve miles sand hills commenced, and continued to the shores
of the lake, with broad stony plains between, and plenty of grass. At
twenty miles crossed the Douglas, running north through sand hills in a
broad valley divided into numerous courses, with dwarf gum-trees, mallee,
tea-tree, and numerous other bushes; the bed sandy, and no water. At
thirty-five miles struck the lake where the Douglas joins it. The country
travelled over to-day has been stony plain (undulating), and low sand
hills, with abundance of feed, but no water. There is some water at the
mouth of the Douglas, but it is salter than the sea. The water in the
lake seems to be a long distance off, but the mirage is so very strong
that I can form no opinion of it to-night. This seems also a bay I have
got into. There is a point of land to the south bearing 25 degrees east
of south, and the other bearing 25 degrees east of north. Searched about
for water, but could find none. Camped in the creek without any. The
country at this part is very low, and nearly on a level with the lake.
The only sand hill I shall be able to get a view from is not above thirty
feet high. At sundown I got on the top of the sand hill, but could see
nothing distinctly; must wait until morning. This creek seems to be very
little frequented by natives; can see very few tracks and no worleys.

Wednesday, 14th December, Lake Torrens. At the first dawn of day I got to
the top of the hill, and remained there some time after sunrise. To the
south-east there is the appearance of a point of land, which I suppose to
be the island which I saw when I first struck the lake. There is the
appearance of water between. A little more to the eastward I can see
nothing but horizon. To the east there is again the appearance of very
low distant land--a mere dark line when seen through a powerful
telescope. To the north of that there is nothing visible but the horizon,
with a blue and white streak between. To the north-north-east beyond the
point, a little low land is to be seen running out from the point, with
water in the far distance. Rode down to the beach to see what that was
composed of; found it to be sand, mud and gravel; firm ground next the
shore. Tried a little distance with the horses, but found it too soft to
proceed with them. I then dismounted, and tried it on foot, but could
only get about two miles; it became so soft, that I was sinking to the
ankles, and the clay was so very tenacious that it completely tired me
before I got back to the horses. The quantity of salt was not so great
here as at the first place I examined. What I thought was a point of land
bearing north-north-east turns out to be an island, which I can see from
here. The point of the bay is north from where I took the bearings.
Between the island and the point I can see nothing but horizon; too low
to see any water. Traced the creek up for seven miles in search of water
or springs, but could see none, nor any indications. Had breakfast, and
started on a course of 20 degrees north of west in search of water or
springs. Crossed the Davenport and ascended a low range, but still could
not see any indications of water; the country similar to that passed over
yesterday. Changed my bearing towards the camp, and arrived there a
little before sundown. The horses were very thirsty, and drank an awful
quantity of water, but being hot it will do them no harm. It is
remarkable that to east of the hot springs I can find no others. This is
the third time I have tried it, and been unsuccessful. I am almost afraid
that the next time I try the lake I shall not find the north boundary of
it. Where can all this water drain to? It is a mystery.

Thursday, 15th December, Louden Springs. Surveyed run Number 4, and sent
Kekwick to correct observations from Mount Stevenson.

Friday, 16th December, Louden Springs. Finished Number 4 run. To-day we
have discovered a large fresh-water hole in a creek joining the George
and coming from the south-west. The water seems to be permanent; it is
half a mile long and seems to be deep. On the banks a number of natives
have been encamped; round about their fires were large quantities of the
shells of the fresh-water mussel, the fish from which they had been
eating: I should think this a very good proof of the water being
permanent. After finishing the survey I followed the creek up for a
number of miles in search of more water, but could find none. It spread
into a number of courses over a large plain, on which there was splendid
feed.

Saturday, 17th December, Louden Springs. Started for the springs under
Mount Margaret to finish the western boundary of Number 1 run. Arrived
towards sundown. Found the creek occupied by natives, who, as soon as
they caught sight of us, bolted to the hill and got upon the top of a
high cliff, and there remained for some time, having a good view of us. I
did everything in my power to induce them to come down to us, but they
would not, and beckoned us to be off back the road we came. At night they
had fires round us, but at some distance off.

Sunday, 18th December, Mount Margaret. About 9 a.m. the natives made
their appearance on the hill, and made signs for us to be off; they were
eight in number. I found that we had camped close to a large quantity of
acacia seed that they had been preparing when we arrived, but had no time
to carry it away before we were on them. One old fellow was very
talkative. I went towards them to try and make friends with them, but
they all took to the hills. By signs I induced the old fellow to stop,
and in a short time got him to come a little nearer. When I came to the
steep bank of the creek he made signs for me to come no further. I showed
him I had no arms with me, and wished him to come up. I could understand
him so far that he wished us to go away, that they might get their seed.
I thought it as well not to aggravate them, but to show them that we came
as friends; and as I had completed all I had to do here, I moved the camp
towards the Freeling Springs, at which they seemed very glad, and made
signs for us to come back at sundown. They seemed to be a larger race
than those down below; the men are tall and muscular, the females are low
in stature and thin. I examined the Mount Margaret range in going along;
there are a number of gum creeks coming from the north side which flow
into the Neale. We searched them up and down, but could find no water.
The number of channels that join them in the range is so great that it
would take weeks to examine them minutely for water. We camped in one of
them without water, although the country promises well for it.

Monday, 19th December, Gum Creek. Started on a north-west course to
examine the country between this and the Mount Younghusband range. We
could see no springs until we reached the Blyth, in which there is water,
but a little brackish; it will do well for cattle. Rode through the
middle of the range, and came upon some horse-tracks, not very old; saw
where the party had camped, and a cairn of stones they had erected on the
top of one of the hills. Followed their tracks some distance down the
gully; they seemed to be going to the Burrow Springs; they appear,
however, to have gone back again. Left the tracks, and proceeded to the
Freeling Springs. Arrived there in the afternoon. No one has been here
since I was, as far as I can see. The country we have passed over
yesterday and to-day has been really splendid for feed. The springs
continue the same, running in a strong stream and of the finest quality.

Tuesday, 20th December, Freeling Springs. Sent Kekwick and one of the men
to examine the goldfield, and to select a place for sinking to-morrow
morning. My eyes were so bad that I was unable to go. They returned in
the afternoon, bringing with them samples from the quartz reefs, in which
there was the appearance of gold. Kekwick said he had not seen such good
quartz since he left the diggings in Victoria. There was every indication
of gold, and I determined to give the place a good trial before leaving
it.

Wednesday, 21st December, Freeling Springs. Commenced digging, but found
the rocks too near. Surface indications were very slight here, but I
found another place which seemed to promise better, so began sinking
there, and at four feet came upon some large boulders, round which was
very good-looking stuff for washing; took some of it to camp and washed
it. No gold, but good indications; a quantity of black sand and emery,
also other good signs. I shall continue the hole, and see what is in the
bottom. Thunderstorm this afternoon; south-west hot wind.

Thursday, 22nd December, Freeling Springs. Occupied in sinking, but made
little progress in consequence of the stones being so large, and the want
of proper tools, crowbar, etc. Washed some more stuff from round about
the boulders; the produce same as yesterday; no gold.

Friday, December 23rd, Freeling Springs. Found that we could do nothing
with the stones with the tools we have. Examined the country round about,
and found another place, which will be commenced to-morrow. Examined a
quartz reef which had every indication of gold. I regretted that I had
not another man, so that I might be able to examine the country for some
distance round. It is necessary to have two men at the camp, which cannot
be moved to where we are sinking, as there is no water within two miles.
It would not be safe to leave the camp with one man only, and two
digging, which is all our strength. Heavy thunderstorm from the
south-west, but very little rain. The wind blew my tent in two. At
sundown it passed over and cleared up, which I regretted to see, as I
expected heavy rains at this season, to enable me to make for the north
or north-west.

Saturday, 24th December, Freeling Springs. Sank upwards of six feet
through gravel, shingle, stones, and quartz. Wind south-west. Heavy
clouds; wind hot.

Sunday, 25th December, Freeling Springs. Wind south; heavy clouds, but no
rain; towards evening changed to south-east. Cool.

Monday, 26th December, Freeling Springs. Got to the bottom of the hole;
washed the stuff, but no gold. Commenced another hole by the side of the
quartz reef, which looks well. In the morning the wind was from the
north; at 10 a.m. it suddenly changed to south, and blew a perfect
hurricane during the whole day, with heavy clouds; but no rain has
fallen.

Tuesday, 27th December, Freeling Springs. The storm continued during the
night, until about 3 o'clock this morning, when a few drops of rain fell,
but not enough to be of any service to me. Bottomed the hole by the side
of the quartz reef: no gold, and I think we shall not be able to sink any
more; our tools are getting worn out. For the rest of the day examined
the quartz reef, in which there is every appearance of gold; I shall stop
the search for it and proceed to the north-east to-morrow, for I think
some rain has fallen in that direction, which will enable me to examine
the country and see if the lake still continues.

Wednesday, 28th December, Freeling Springs. At 7 a.m. started with
Kekwick on a north-east course. At seven miles crossed the Neale, spread
over a large grassy plain four miles broad, and ascended a low ridge of
table-topped hills, stony, with salt-bush and grassed. Crossed another
creek, at twenty miles, with myall and stunted gums running over a plain
in numerous courses. Plenty of grass but no water. After crossing it,
ascended a high peak, which I supposed to be the top of the Hanson range,
but found another long table-topped hill, higher, about three miles
distant. Ascended that, but could see nothing but more table-topped
ranges in the distance. This hill is thirty-five miles from Freeling
Springs. Searched for water, and after some time found a little water in
one of the creeks, where we camped, it being after sundown. The country
from the last creek is not so good, and very stony, so much so that it
has lamed my horse, and nearly worn his shoes through at the tips. The
horses have drunk all the water, and left none for the morning.

Thursday, 29th December, Hanson Range. Started at 6 a.m. on the same
course for another part of the range. At six miles crossed a grassy creek
of several channels, with myall and gum, but no water, running to
north-east, nearly along our line. At seventeen miles struck the same
creek again where it is joined with several others coming from the
west-north-west and north. They are spread over a large broad plain
covered with grass. Searched for water, but could not find any. Crossed
the plains and creeks to a white hill on a north course, and at three
miles reached the top; it was a low chalky cliff on the banks of the
creek. Changed our course to the first hill I had taken. At seven miles
and a half reached the top, which I found very stony. To the north can be
seen the points of three other table-topped hills; to the north-east is a
large stony plain about ten miles broad, beyond which are high sand
hills, and beyond them again, in the far distance, is the luminous
appearance of water. Not being on the highest part of the range I
proceeded two miles to the south-east to get a better view. From here we
could see the creek, winding in a south-east direction, until it reached
the lake, which seemed to be about twenty-five miles off. We could not
distinctly see it, the mirage and sand hills obscuring our view. My horse
having lost both his fore shoes and there being no prospect of water
further on, I was reluctantly obliged to return to the camp. We had seen
a little rain water on the plain, about seven miles back, at which we
decided to camp to-night. Arrived there a little before sundown. My horse
very lame, scarcely able to walk along the stones. I am disappointed that
there is not more rain water; there seems only to have been a slight
shower.

Friday, 30th December, Hanson Range. The horses having strayed some
distance, we did not get a start till half-past seven on a course of 323
degrees, to a white hill, to see whether there are any springs on the
other side; at one mile and a half reached it, but no springs. Changed
our course to a very prominent hill (which I have named Mount Arthur)
bearing 275 degrees, and after crossing two small myall creeks and a
stony plain with salt bush and grass, at ten miles we struck a large
myall and gum creek, coming from the north-west, with some very deep
channels. We went some miles up it, but could find no water, the courses
for the water being too sandy and gravelly to retain it. At twenty-four
miles from the last hill arrived at the summit of Mount Arthur. Changed
course to 195 degrees. At ten miles struck another myall and gum creek of
the same description as the others, coming from the range; no water.
Camped. My horse is nearly done up; I am almost afraid he will not be
able to reach the camp to-morrow.

Saturday, 31st December, Hanson Range. Started shortly after daybreak for
the camp. At fifteen miles struck another myall and gum creek running
into the Neale, and at twenty miles came upon the Neale, which is here
three miles broad. Here we saw some recent native tracks and places where
fires had been. Arrived at the camp at sundown; horses quite done up. I
am sorry that I have been unable to make the lake on this journey; I
could have done it, but should most likely have had to leave my horse; he
never could have done it. I should then have been obliged to walk the
distance back, with all the water dried up. Had I seen the least
indication of water on ahead, I should have gone.

Sunday, 1st January, 1860, Freeling Springs. In the afternoon it became
cloudy. Wind north. No rain.

Monday, 2nd January, Freeling Springs. Having observed a hill on Saturday
that seemed to me a spring, where the Neale comes through the range, I
sent Kekwick to examine it, my eyes being too bad. Sent Muller to examine
some more quartz reefs in which I think gold exists. Towards sundown he
returned with two good specimens, in which I am almost sure there is
gold. The reef is twelve feet wide. Shortly after, Kekwick returned and
reported springs and two large water holes, and numerous smaller ones,
with abundance of permanent water, although slightly brackish. I shall
move up and fix their position as soon as I am satisfied with the search
for gold.

Tuesday, 3rd January, Freeling Springs. Sent Kekwick and Muller to get
some more specimens of quartz. They returned with some in which there
were very good indications of gold. It was useless for us to try any
more, our tools being of no use. The reefs would require to be blasted. I
am afraid there will be no surfacing here. I have done all that lies in
my power to get at the gold; but without proper tools we can do nothing,
so I shall be obliged to give it up, and start to-morrow for the Neale,
to where I sent Kekwick yesterday.

Wednesday, 4th January, Freeling Springs. Started at 8 a.m., and arrived
in about thirteen miles. The large water hole is upwards of a mile long,
with fully forty yards of water: in width, from bank to bank, it is
seventy yards, and upwards of fifteen feet deep; there are large mussel
shells on the banks, and plenty of good feed. All round to the south
there are low sand hills covered with grass. To the east, in some places,
it is stony, with salt-bush, and many broad well-grassed valleys coming
from the Mount Kingston range. About a quarter of a mile to the west of
the large hole there is a course of springs coming from the Kingston
Hills and sand hills, and emptying themselves into the creek. The water
is delicious, and plentiful, and, if opened, these springs will yield an
ample supply for all purposes. To the west are hills, with the creek
coming through them, with water all the way up to where I crossed it in
my return last trip. To the north are stony undulating rises, with
salt-bush and grass.

Thursday, 5th January, The Neale. Examining the country round to the
north and round Mount Harvey. It is poor and stony. On the eastern and
northern sides it becomes bad at three miles from the creek. The country
in the other directions is good, and will make a first-rate run. This, in
connexion with the Mildred and McEllister Springs, will feed any number
of cattle.

Friday, January 6th, The Neale. As my rations are now drawing to a close
(for we started with provisions only for three months, and have been out
now for three months and more), I must sound a retreat to get another
supply at Chambers Creek. It was my intention to have sent two men down
for them, but I am sorry to say that I have lost confidence in all except
Kekwick. I cannot trust them to be sent far, nor dare I leave them with
our equipment and horses while Kekwick and I go for the provisions.
Situated as I am with them, I must take all the horses down; and if I can
get men to replace them at Chambers Creek, I will send them about their
business. They have been a constant source of annoyance to me from the
very beginning of my journey. The man that I had out with me on my last
journey has been the worst of the two. They seem to have made up their
minds to do as little as possible, and that in the most slovenly and lazy
manner imaginable. They appear to take no interest in the success of the
expedition. I have talked to them until I am completely wearied out;
indeed, I am surprised that I have endured it so long. Many a one would
have discharged them, and sent them back walking to Adelaide; in fact, I
had almost made up my mind to do so from here, and to run the chance of
getting others at Mr. Barker's. Although they have behaved so badly, and
so richly deserve to be punished (for they have taken advantage of me
when I could get no others to supply their places), I could not find in
my heart to do it. Kekwick is everything I could wish a man to be. He is
active, pushing, and persevering. At any time, and at any moment, he is
always ready, and takes a pleasure in doing all that lies in his power to
forward the expedition. Would that the two others were like him! I should
then have no trouble at all. Started at 7 a.m. on my return on a
south-east course, and camped at a small spring on the east side of Mount
Younghusband. Distance, twenty miles.

Saturday, 7th January, Mount Younghusband. Started at 7 a.m. for the
Milne Springs, where I shall remain for a day or two to get all the
horses fresh shod, and leave what things I do not require, intending to
get them on my return. Arrived there at 11 o'clock. Found the water much
the same as it was when I first saw it.

Sunday, 8th January, Milne Springs. Severe attack of lumbago. Sun hot;
but cool breeze from south-east.

Monday, 9th January, Milne Springs. Unable to ride, so I was obliged to
send Kekwick and one of the men to the westward. This was a great
disappointment to me, as I should like to have seen the country myself to
have connected it with my farthest north-west point on my first journey.
The other man was shoeing the horses. Sun hot. Cool breeze from
south-east. Very cold night and morning.

Tuesday, 10th January, Milne Springs. Latitude, 28 degrees 15 minutes 45
seconds. Shoeing horses. Flies a great trouble; can do nothing for them.
If they are allowed to remain a moment on the eye, it swells up
immediately, and is very painful. Kekwick and the other man returned at 9
o'clock p.m. They report having found two springs, one about nine miles
west, and the other about thirty miles, in a large spring country, which
they had not time to examine well. Although I am so unwell, I must start
to-morrow and see what it is. Judging from their description, there must
be something good; and I cannot leave without seeing it, although my
provisions are nearly done.

Wednesday, 11th January, Milne Springs. Shortly after sunrise started
with Kekwick on a west course for the larger spring country, leaving the
near one until our return. At eleven miles and a half crossed the Blyth,
coming from the south. At twenty-eight miles reached the spring country.
Changed to 150 degrees, and at two miles camped at the spring. The
springy place has the appearance of a large salt lagoon, three miles
broad and upwards of eight miles long. At the south end of it is a creek
with brackish water, and on its banks are the springs, the water from
which is very good; they are not running.

Thursday, 12th January, West Springs. There are a number of natives at
these springs. We have seen their smoke, and both old and recent tracks.
Started on a south course. At four miles and a half came upon a creek,
with reeds and brackish water, running a little to the west of north.
Traced it down for upwards of a mile and a half. Saw that it ran into the
swamp west of where we struck it. Could see no springs upon its banks.
Returned to the place where we first struck it, and proceeded a mile on a
course of 120 degrees to three large patches of very green reeds, which
turned out to be eight feet high. Could find no surface water except what
was brackish. The country was moist all round. Thence on the same bearing
for two miles. Sent Kekwick to examine some places that looked like
springs. They were in the middle of a large salt lagoon, having a crust
of limestone, under which the water was, and if broken open, in many
places where there was no sign of water, a beautiful supply could be
obtained. Changed to 245 degrees, and, at about fifteen miles, changed to
90 degrees, through sand hills. We have seen many places where water can
be obtained at a few inches below the surface. Camped at the spring. Feel
very ill; can scarcely sit on my horse.

Friday, 13th January, West Springs. Being anxious to see the nature of
the country between this and the Mount Margaret range, I started at 6.30
on a course of 110 degrees over occasional sand hills and stony places,
with splendid feed. At ten miles and a half reached a stony rise, and
changed my course to 76 degrees, for five miles, to a black hill composed
of ironstone. Changed to 105 degrees, for one mile, to examine a white
place coming down from the range, which had the appearance of springs,
but found it to be composed of white quartz. Changed again to 50 degrees
to a rough hill, which had also the appearance of springs. At two miles
crossed the bed of the Blyth, which takes its rise in the range. No water
in it, but loose sand and gravel. At seven miles reached the rough hill,
after crossing three small tributaries; was disappointed in not finding
water. Ascended the hill, from which we had a good view of the
surrounding country, but see no indications of water. I must now make for
the second spring found by my men three days ago. Course north, over
stony hills and table land, in which I crossed my former tracks going to
the Freeling Springs. Arrived at the spring at 7.30 p.m. All of us, men
and horses, very tired.

Saturday, 14th January, Springs South of Mount Younghusband. Examined the
spring, and found it to be a very good one; it is situated near the banks
of the Blyth, on the same spongy ground that I discovered last time, and
which was marked off as a run. Searched about, and found two more good
springs. There was plenty of water in the creek, but the dry season had
made it brackish. Discovered a spring in one of the creeks that runs east
from Mount Margaret. The natives had cleared it out, and the water, which
was very good, was about two feet from the surface. In the other two
creeks we also found springs which only required opening. I then made for
the camp, where I found everything all right.

Sunday, 15th January, Milne Springs. Preparing for a start to-morrow for
Chambers Creek, by way of Louden Springs; I must endeavour to find some
more springs, for I am not quite satisfied yet about that country. Very
much annoyed by the misconduct of the two men I left behind at the camp;
they have had the impertinence to open my plan-case, and have so damaged
my principal plan with their hot moist hands, that I know not what to do
with it. This is not the first time they have done it.

Monday, 16th January, Milne Springs. Started at 7.10 a.m. on a bearing of
138 degrees 30 minutes. At about twenty-two miles struck four other
springs, beyond the Messrs. Levi's boundary; from one of them there is a
strong stream of water flowing. They are almost completely hidden, and
one cannot see them until almost on the top of them. I have taken
bearings to fix them, and have named them Kekwick Springs. Five o'clock
p.m. Arrived at Louden Springs. Distance, thirty-one miles.

Tuesday, 17th January, Louden Springs. Started shortly after daybreak, on
a course of 110 degrees, over as fine a grass country as I have yet
travelled over. At sixteen miles crossed the Douglas, running through
sand hills covered with grass, but no water, nor any signs of springs.
Proceeded in the same direction for eight miles, when we were stopped by
a lagoon. Changed my course to south-south-west to a hill that had the
appearance of water, but found beyond it another large dry lagoon, on the
banks of which we saw the tracks of a single horse crossing the end of
the lagoon, and steering for Lake Torrens; they seemed to be about two
months old. Can they be the tracks of that infatuated man who left me on
the 20th of November? In all probability he has lost my downward track
and himself also. They are only about two miles to the east of mine.
Camped without water on a sand hill.

Wednesday, 18th January, Sand Hill. Started shortly after daybreak on a
south-south-east course, still in search of springs (crossing my outward
track of last journey), at a place where I thought it most likely for
them, but was unsuccessful. If I could have found one here, I should have
gone direct to the Emerald Springs, but the horses would suffer very much
if they were to be another night without water; the food is so dry, and
the weather so hot, they cannot endure more than two days and one night
without it. Changed my course to Strangway Spring. Arrived there at 2.30.
Some of the horses very much done up. Camped, and gave them the rest of
the day to recruit.

Thursday, 19th January, Strangway Springs. Started for the Beresford
Springs. At nine miles and a half arrived there; and, at eight miles
beyond, made the Hamilton Springs, where we camped for the night.

Friday, 20th January, Hamilton Springs. Started by way of the Emerald
Springs, to see if Mr. Barker's party is there, or if any person had been
there and got the parcel, and forwarded it to Mr. Chambers. Arrived at
the springs, and found that some one had got it. Mr. B.'s party had gone.
Went on to Chambers Creek, and found them there.

Saturday, 21st January, Chambers Creek. Here we found provisions awaiting
us, as we expected; but the two men still exhibit a spirit of
non-compliance, and refuse to proceed again to the north-west; they are
bent upon leaving me and returning to Adelaide although they know that
there are no men here to supply their places. They have demanded their
wages and a discharge, which, under all the circumstances of the case,
and considering how badly they have served me, I feel myself justified in
withholding. I shall therefore be compelled to send Kekwick down as far
as Mr. Chambers' station with my despatches, etc., and to procure other
assistance. This will be a great loss of time and expense, which the
wages these men have forfeited by not fulfilling their agreements will
ill repay. Here we heard of the man Smith, who, it seems, left the mare,
whether dead or alive we know not at present. He was lost for four days
without water (according to his own account), and, after various
adventures, and picking up sundry trifles from different travelling
parties, who relieved him out of compassion, reached the settled
districts in a most forlorn condition. Mr. Barker had left his station
some three weeks before we arrived.


JOURNAL OF MR. STUART'S FOURTH EXPEDITION--FIXING THE CENTRE OF THE
CONTINENT. FROM MARCH TO SEPTEMBER, 1860.

Friday, 2nd March, 1860, Chambers Creek. Left the creek for the
north-west, with thirteen horses and two men. The grey horse being too
weak to travel was left behind. Camped at Hamilton Springs.

Saturday, 3rd March, Mount Hamilton. Camped at the Beresford Springs,
where it was evident that the natives, whose camp is a little way from
this, had had a fight. There were the remains of a body of a very tall
native lying on his back. The skull was broken in three or four places,
the flesh nearly all devoured by the crows and native dogs, and both feet
and hands were gone. There were three worleys on the rising ground, with
waddies, boomerangs, spears, and a number of broken dishes scattered
round them. The natives seemed to have run away and left them, or to have
been driven away by a hostile tribe. Between two of the worleys we
observed a handful of hair, apparently torn from the skull of the dead
man, and a handful of emu feathers placed close together, the feathers to
the north-west, the hair to the south-east. They were between two pieces
of charred wood, which had been extinguished before the feathers and hair
were placed there. It seemed to be a mark of some description.

Sunday, 4th March, Beresford Springs. Night and morning cold; day very
hot. Wind south-west.

Monday, 5th March, Beresford Springs. Wind changed to the east during the
night. Morning very cold. Arrived at the Strangway Springs. Day very hot.
Wind variable.

Tuesday, 6th March, Strangway Springs. Very hot during the night. Made
William Springs and camped. The day exceedingly hot, wind south-west, in
which direction a heavy bank of clouds arose about noon; in the evening
there was a great deal of lightning, and apparently much rain falling
there, but none came down our way.

Wednesday, 7th March, William Springs. The night very hot and cloudy,
with the wind from the west, but without rain. Started for Louden Spa,*
(* The Louden Springs of the two last expeditions.) the first few miles
being over low sand rises and broad valleys of light sandy soil, with
abundance of dry grass; by keeping a little more to the north-west the
sand rises can be avoided. At seven miles we struck a swamp, but could
see no springs. On approaching the Douglas the country becomes more
stony, and continues so to the Spa, where we camped.

Thursday, 8th March, Louden Spa. Cold wind this morning from the
south-east; the clouds are gone. Camped at Hawker Springs.

Friday, 9th March, Hawker Springs. Very cold last night. Wind from the
south. During the day it changed to the south-east, and the sun was very
hot. Camped at the Milne Springs, and found the articles we had left*
there all right (* See last expedition.); the natives had opened the
place where we had put them, but had taken nothing.

Saturday, 10th March, Milne Springs. At half past 11 last night it began
to rain, and continued doing so nearly all day. Wind south-east.

Sunday, 11th March, Milne Springs. About 10 o'clock last night we were
flooded with water, although upon rising ground, and were obliged to move
our camp to the top of a small hill. It rained all night and morning, but
there are signs of a break in the clouds. During the day it has rained at
intervals. The creek is coming down very rapidly, covering all the valley
with a sheet of water.

Monday, 12th March, Milne Springs. A few heavy showers during the night,
but now there seems a chance of a fine day, which will enable us to get
our provisions dried again. The country is so boggy that I cannot proceed
to-day, but if it continues fair I shall attempt it to-morrow morning.
This rain is a great boon to me, as it will give me both feed and water
for my horses, and if it has gone to the north-west it will save me a
great deal of time looking for water.

Tuesday, 13th March, Milne Springs. Started for Freeling Springs. The
country in some places is very soft, but the travelling is better than I
expected. As we approached the Denison ranges the rain did not seem to
have been so heavy, but when we came to the Peake, we found it running
bank high, and very boggy. Impossible to cross it here, so I shall follow
it up in a west-south-west direction. Camped at Freeling Springs.

Wednesday, 14th March, Freeling Springs. Started on a course a little to
the south of west, to try and find a crossing-place. At two miles it
turned a little to the north of west, but at ten miles it turned to the
south-west, and was running very rapidly, about five miles an hour. I was
obliged to stop at this point, as I could not cross the creek, the banks
being so boggy. I have discovered another spring at eleven miles on the
same bearing as the Freeling Springs, but I cannot get to it. From here
it has the appearance of being very good; a hill covered with reeds at
the top, the creek running round the east side of it. I shall endeavour
to cross to-morrow and examine it.

Thursday, 15th March, The Peake. The creek being still impassable, I
remained here another day. Yesterday the horse that was carrying my
instruments broke away from the man who was leading him, burst the
girths, and threw the saddlebags on the ground. The instruments were very
much injured, in fact very nearly ruined; the sextant being put out of
adjustment, has taken me all day to repair, and I am not sure now whether
it is correct or not. It is a great misfortune. Wind north; clouds
north-east.

Friday, 16th March, The Peake. Saddled and started to cross the Peake
about three miles to the south-west, but had a fearful job in doing so,
the banks being so boggy, and the current so strong. The horses could
hardly keep on their feet, and most of them were up to their
saddle-flaps, and some under water altogether. One poor old fellow we
were obliged to leave in it, as he was unable to get out, and we were
unable to help him, although we tried for hours. He is of very little use
to me, for he has never recovered his trip to Moolloodoo and back. He has
had nothing to carry since we started, and seemed to be improving every
day. I wish now that I had left him at Chambers Creek along with the
grey, but as he looked in better condition, I thought he would mend on
the journey, and I intended him to bring the horses in every morning,
when we got further out. We have been from 10 a.m. to 3.30 p.m. in
getting across, including the time spent in trying to extricate Billy. I
cannot proceed further to-day, and have therefore camped on the west side
of the springs that we saw from the last encampment, which I named
Kekwick Springs. There are six springs. The largest one will require to
be opened; the reeds on it are very thick, and from ten to twelve feet in
height. We tried again to get the horse on shore, but could not manage
it; the more we try to extricate him, the worse he gets. I have left him;
I do not think he will survive the night. It is now sundown, and raining
heavily; the night looks very black and stormy. Wind from the south-west.

Saturday, 17th March, Kekwick Springs. About 8 o'clock last evening the
wind changed to the north-west, and we had some very heavy rain, which
lasted the greater part of the night. Early in the morning the wind
changed again to the south-east, with occasional showers. At sunrise it
looked very stormy. I must be off as soon as possible out of this boggy
place. The old horse is still alive, but very weak. The water has lowered
during the night. If no more rain falls to the south-west it will soon be
dry, when he may have a chance of getting out. I cannot remain longer to
assist him; it would only be putting the rest of my horses in danger. I
would have remained here to-day to have dried my provisions, but the
appearance of the weather will not allow me. They must take their chance.
Started on a north-west course for the Neale. At fifteen miles struck it,
and changed to the west to a creek coming south from the stony rises. The
banks of the Neale are very boggy. The first four miles to-day were along
the top of a sandy rise, with swampy flats on each side, with a number of
reeds growing in them, also rushes and water-grass. At four miles was a
strong rise, but before we arrived at it we had to cross one of the
swamps, in which we encountered great difficulty. After many turnings and
twistings, and being bogged up to the shoulders, we managed to get
through all safe. It was fearfully hard work. For three miles, on the top
of a stony rise, the country is poor (stones on the top of gypsum
deposit), but after that it gradually improves, and towards the creek it
becomes a good salt-bush country. Wind from the south-east; still very
cloudy.

Sunday, 18th March, Neale River. Wind south-east; heavy clouds. I
observed a bulbous plant growing in this creek resembling the Egyptian
arum; it was just springing. I will endeavour to get some of the seed, if
I can. I hoped we should have got our provisions dried to-day, but it was
so showery we could not get it done. The creek is so boggy that we cannot
cross it, and must follow it round to-morrow. A sad accident has happened
to my plans. There was a small hole in the case that contains them, which
I did not observe, and in crossing the Peake the water gained admittance
and completely saturated them; it is a great misfortune. Sundown: still
raining; wind same direction.

Monday, 19th March, Neale River. Rained during the night, and looks very
stormy this morning. Followed the Neale round to where it goes through
the gap in Hanson range; in places it was rather boggy, but good
travelling in this wet weather--firmer than I expected. We had much
difficulty in crossing some of the side creeks. Camped on the south side
of the gap. Wind south-east; cloudy, with little rain.

Tuesday, 20th March, Neale River Gap, Hanson Range. Wind south-east; a
few showers during the night. Still no chance of getting my provisions
dried. It cleared off about noon, and became a fine day. Followed the
Neale round, and camped on one of the side creeks coming from the south
of west. Ground still soft. Wind south-east. Saw some smoke in the hills
this morning, but no natives. Good country along both sides of the range
on the west side of the Neale.

Wednesday, 21st March, Neale River. Beautiful sunshine. Shall remain here
to-day, in order to dry my provisions. On examining them I find that a
quantity of our dried meat is quite spoiled, which is a great
loss--another wet day, and we should have lost the half of it.

Thursday, 22nd March, Side Creek of the Neale River. Wind south-west;
clear sky. I intended to have gone north-west from this point, but, in
attempting to cross the creek, we found it impassable. My horse got
bogged at the first start, and we had some difficulty in getting him out.
We were obliged to follow the creek westward for seven miles, where it
passes between two high hills connected with the range. We managed at
last, with great labour and difficulty, to get across without accident.
At this place four creeks join the main one, and spread over a mile in
breadth, with upwards of twenty boggy water-courses; water running. It
has taken us five hours, from the time we started, to cross it. The
principal creek comes from the south-west. I ascended the two hills to
get a view of the surrounding country, and I could see the creek coming
from a long way off in that direction. At this point the range seems
broken or detached into numerous small ranges and isolated hills. I now
changed my course to north-west, over table land of a light-brown colour,
with stones on the surface; the vegetation was springing all over it and
looking beautifully green. At six miles on this course camped on a myall
creek. The work for the horses has been so very severe to-day that I have
been induced to camp sooner than I intended. Wind south.

Friday, 23rd March, Myall Creek. Wind south. Started on the same course,
north-west. At three miles crossed another tributary--gum and myall. The
country, before we struck the creek, was good salt-bush country, with a
plentiful supply of grass. The soil was of a light-brown colour, gypsum
underneath, and stones on the surface, grass and herbs growing all round
them. After crossing the creek, which was boggy, we again ascended a low
table land of the same description. At ten miles came upon a few low sand
rises, about a mile in breadth. We then struck a creek, another
tributary, spread over a large plain, very boggy, with here and there
patches of quicksand. We had great difficulty in getting over it, but at
last succeeded without any mishap. We then entered a thick scrubby
country of mulga and other shrubs; the soil now changed to a dark red,
covered splendidly with grass. After the first mile the scrub became much
thinner; ground slightly undulating. After crossing this good country, at
twenty miles we struck a large creek running very rapidly at five miles
per hour; breadth of water one hundred feet, with gum-trees on the bank.
From bank to bank it was forty-four yards wide. This seemed to be only
one of the courses. There were other gum-trees on the opposite side, and
apparently other channels. Wind south. A few clouds from the north-west.

Saturday, 24th March, Large Gum-Tree Creek. Found it impossible to cross
the Neale here; the banks were too boggy and steep. We therefore followed
it round on a west course for three miles, and found that it came a
little more from the north. Changed to 290 degrees, after trying in vain
to cross the creek at this point. At about four or five miles
south-south-west from this point there are two high peaks of a low range.
The higher one I have named Mount Ben, and the range Head's Range; its
general bearing is north-west to opposite this point; it turns then more
to the west. I can see another spur further to the west, trending
north-west. At four miles and a half after leaving we found a ford, and
got the horses across all safe. I then changed to the north-west again,
through a scrubby country--mulga, acacia, hakea, salt bush, and numerous
others, with a plentiful supply of grass. The soil is of a red sandy
nature, very loose, and does not retain water on the surface. We had
great difficulty in getting through, many places being so very thick with
dead mulga. We have seen no water since we left the creek. Distance,
eighteen miles. I was obliged to camp without water for ourselves. As we
crossed the Neale we saw fish in it of a good size, about eight inches
long, from which I should say that the water is permanent. I shall have
to run to the west to-morrow, for there is no appearance of this scrubby
country terminating. I must have a whole day of it.

Sunday, 25th March, Mulga Scrub. I can see no termination, on this
course, to this thick scrub. I can scarcely see one hundred yards before
me. I shall therefore bear to the west, cut the Neale River, and see what
sort of country is in that direction. At ten miles made it; the water
still running, but not so rapidly. The gum-trees still existed in its
bed, and there were large pools of water on the side courses. We had the
same thick scrub to within a quarter of a mile of the creek, where we met
a line of red sand hills covered with a spinifex. The range on the
south-west side of the creek seemed to terminate here, and become low
table land, apparently covered with a thick scrub, the creek coming more
from the north. I did not like the appearance of the spinifex, an
indication of desert to the westward. Camped on the creek. Wind
north-west; heavy clouds from the same direction.

Monday, 26th March, The Neale River West. I am obliged to remain here
to-day to repair damages done to the packs and bags, which have been torn
all to pieces; it will take the whole of the day to put them in order. We
have seen very few signs of natives visiting this part of the country. I
shall go north to-morrow and try to get through this scrub. Wind south,
sky overcast with heavy clouds; looks very like rain.

Tuesday, 27th March, West Neales. Rained very heavily during the night,
and is still doing so, but less copiously. About noon it cleared up a
little. I have sent Kekwick to get a notion of the country on the other
side of the low range, while I endeavour to obtain an observation of the
sun. The range is scrubby, composed of a light-coloured and dark-red
conglomerate volcanic rock, easily broken. The view from it is not
extensive. At a mile from the creek the sand ceases, and stony ground
succeeds up to the range. Feed excellent south-west from the camp. To the
eastward rugged hills, apparently with fine open grass and forest lands.
Numerous rows of water holes visible. To the south-east, country more
open. To the south-south-east and south still the same good country. From
south to west the same; hills to the west from five to eight miles
distant. View from another hill north-west two miles and a half. The
hills on the west still continue towards the north-west, but become
lower. Country scrubby, with occasional patches of open grass land. Creek
coming in from north-north-west. From north-west to north-north-east
mulga scrub. From north-north-east to east low range in the distance,
like table land. Too cloudy to take an observation; occasional showers
during the day. Wind south-south-east; still looking very black.
Repairing my saddles; some of my horses are getting bad backs.

Wednesday, 28th March, West Neale River. Started on a north course to get
through the mulga scrub. At ten miles could see the range to the
north-east. The scrubby land now became sand hills; I could see no high
ground on ahead, the scrub becoming thicker; it seemed to be a country
similar to that I passed through on my south-east course (first journey),
and I think is a continuation of it. I therefore changed my course to the
north-east range, bearing 35 degrees. After five miles through the same
description of country, mulga scrub with plenty of grass, we arrived at
water, where three creeks join, one from the south-west, one
west-north-west, and the other from about north-west. The water was still
running in the one from the west-north-west with large long water holes;
also water holes in the other two; gum-trees in the creek. I suppose this
to be the Frew; excellent feed on the banks of the creek up to the range,
which is stony. I ascended the table range in order to have a view of the
country round. To this point the range comes from east-south-east, but
here it takes a turn to the east of north, all flat-topped and stony,
with mulga bushes on the top and sides; the rocks are of a light, flinty
nature. At about six miles north the country seems to be open and stony.
That country I shall steer for to-morrow. To the north-east is the range,
but it seems to drop into low table land; distant about fifteen miles. To
the north-west and west is the thick mulga, scrubby country. There are
numerous tracks of natives in the different creeks, quite fresh,
apparently made to-day. Wind south-east; clouds.

Thursday, 29th March, The Frew. Started on a north course. At one mile,
after crossing a stony hill with mulga, we suddenly came upon the creek
again; it winds round the hill. Here another branch joins it from the
north, the other coming from the east of north. Along the base of the
range there were very large water holes in both branches. The natives had
evidently camped here last night; their fires were still alight; they
seemed only just to have left. From the numerous fires I should think
there had been a great number of natives here. All round about in every
direction were numerous tracks. We also observed a number of winter
habitations on the banks of the creek; also a large native grave,
composed of sand, earth, wood, and stones. It was of a circular form,
about four feet and a half high, and twenty to twenty-four yards in
circumference. The mulga continued for about six miles; but at three
miles we again crossed the north branch of the creek, coming now from the
north-west. The mulga was not thick except on the top of the rises, where
splendid grass was growing all through it. We now came upon the open
stony country, with a few mulga creeks. There was a little salt bush, but
an immense quantity of green grass, growing about a foot high, which gave
to the country a beautiful appearance. It seemed to be the same all round
as far as I could see. At fourteen miles we struck the other branch,
where it joined, with splendid reaches of water, to the main one, which
now came from the west of north, and continued to where our line cut the
east branch. This seems to be the place where it takes its rise. Camped
for the night. The whole of the country that we have travelled through
to-day is the best for grass that I have ever gone through. I have
nowhere seen its equal. From the number of natives, from there being
winter and summer habitations, and from the native grave, I am led to
conclude the water there is permanent. The gum-trees are large. I saw
kangaroo-tracks.

Friday, 30th March, Small Branch of The Frew. Course north. At two miles
and a half changed to 332 degrees to a distant hill, apparently a range
of flat-topped hills. At sixteen miles crossed a large gum creek running
to the south of east; it spreads out over a flat between rough hills of
half a mile wide. The bed is very sandy; it will not retain water long.
On the surface it very much resembles the Douglas, but is broader, and
the gum-trees much larger. There were some rushes growing in its bed. I
have named it the Ross. We then ascended the low range for which I had
been steering. Four miles from the creek it is rough and stony, composed
of igneous rock, with scrub, mulga, and plenty of grass quite to the top.
To continue this course would lead me again into the mulga scrub, where I
do not want to get if I can help it. It is far worse than guiding a
vessel at sea; the compass requires to be constantly in hand. I again
changed to the north, which appears to be open in the distance. I could
see another range of flat-topped hills. After crossing over several small
spurs coming from the range, and a number of small creeks, volcanic, and
stony, we struck another large gum creek coming from the south of west,
and running to the south-east. It was a fine creek. These courses of
water spread over a grassy plain a mile wide; the water holes were long
and deep, with numerous plants growing on their banks, indicating
permanent water. The wild oats on the bank of the creek were four feet
high. The country gone over to-day, although stony, was completely
covered with grass and salt bush; it was even better than that passed
yesterday. Some of the grass resembled the drake, some the wild wheat,
and some rye--the same as discovered by Captain Sturt. There is a light
shade over the horizon from south-east to north-west, indicating the
presence of a lake in that direction. I have named it after my friend Mr.
Stevenson. There are small fish in the holes of this creek, and mussel
shells, also crabs about two inches by one inch and a half.

Saturday, 31st March, The Stevenson. I am obliged to remain here to-day;
my horses require shoeing. The country cuts up the shoes very much.

Sunday, 1st April, The Stevenson. I find to-day that my right eye, from
the long continuation of bad eyes, is now become useless to me for taking
observations. I now see two suns instead of one, which has led me into an
error of a few miles. I trust to goodness my other eye will not become
the same; as long as it remains good, I can do. Wind east; cool. Heavy
clouds.

Monday, 2nd April, The Stevenson. Started at 8 o'clock; course 355
degrees to distant hills. At six miles we struck a gum creek with water
in it, but not permanent. At ten miles we crossed another, running
between rugged hills; a little water coming from the west and running
east-south-east through a mass of hills. At twelve miles crossed a valley
a quarter of a mile broad, through which a gum creek runs, with an
immense quantity of drift timber lying on its banks. At twenty miles
arrived at the first part of the range, and at twenty-eight miles camped
on a gum creek running east and coming from the south of west. The first
three miles of to-day's journey were over good country; it then became
rather scrubby, with numerous small creeks and valleys running to the
east. Plenty of grass and salt-bush, with gravel, ironstone, and lime on
the surface. At a mile before we made the rugged creek the ironstone
became less, and a hard white stone took its place, and continued to the
range, on which it is also found. Gypsum, chalk, ironstone, quartz, and
other stones, are the chief materials of which it and the other hills are
composed. There are also a few of a hard red sandstone. The range is
broken, and running nearly east and west. The country round is slightly
undulating; numerous small creeks running to the eastward, with a deal of
grass and salt-bush. No water in this creek. Camped without. Wind east.

Tuesday, 3rd April, Gum Creek, South of Range. Ascended the hill at three
miles from last night's camp. The country very rough, stony, and scrubby
to the base. The view from it is very extensive. I have named it Mount
Beddome, after S. Beddome, Esquire, of Adelaide. To the west is another
broken range, about fifteen miles distant, of a dark-red colour, running
nearly north and south. The country between is apparently open, with
patches of scrub. A gum creek comes from the south-west and runs some
distance to the north-east; it then turns to the east. In the distant
west appears a dense scrub. On a bearing of 330 degrees there is a large
isolated table hill, for which I shall shape my course, to see if I can
get an entrance that way. To the north are a number of broken hills and
peaks with scrub between; they are of every shape and size. To the east
another flat-topped range; country between also scrubby; apparently open.
Close to the range, distant about twenty miles saw hills in the far
distance; to the east another flat-topped small range; between it and the
other the creek seems to run. The highest point of it bears 80 degrees,
and I have named it Mount Daniel, after Mr. Daniel Kekwick, of Adelaide.
From east to south-east the country is open and grassy; low ranges in the
distance. Saw some rain water, bearing 30 degrees, to which I will go,
and give the horses a drink; they had none last night. Distance, two
miles. Obtained an observation of the sun, 118 degrees 17 minutes 30
seconds. At six miles crossed the broad bed of a large gum creek; gravel;
no water. At eight miles the red sand hills commence, covered with
spinifex; and on the small flats mulga scrub, which continues to the base
of the hill. Red loose sand; no water. Distance, twenty miles from Mount
Beddome to this hill. The country good, until we get among the spinifex.

Wednesday, 4th April, Mount Humphries. At break of day ascended the
mount, which is composed of a soft white coarse sandstone. On the top is
a quantity of water-worn quartz, cemented into large masses. The view is
much the same as from Mount Beddome, broken ranges all round the horizon,
and apparently a dense scrub from south-west to west. It then becomes an
open and grassy country, with alternate patches of scrub. I can see a gum
creek about two miles distant; I can also see water in it, which the
horses have not yet discovered. I shall therefore go in that direction,
and give them a drink. To the north and eastward the country appears
good. Went to the aforesaid water, to see if there is any that I can
depend upon. On my return, wanting to correct my instrument, which met
with an accident three or four days ago, by the girths getting under the
horse's belly (he bolted and kicked it off), I sent Kekwick to examine
the creek that I saw coming from the north. He says there is plenty of
water to serve our purpose. The creek is very large, with the finest
gum-trees we have yet seen, all sizes and heights. This seems to be a
favourite place for the natives to camp, as there are eleven worleys in
one encampment. We saw here a number of new parrots, the black cockatoo,
and numerous other birds. The creek runs over a space of about two miles,
coming from the west; the bed sandy. After leaving it, on a bearing of
329 degrees, for nine miles, we passed over a plain of as fine a country
as any man would wish to see--a beautiful red soil covered with grass a
foot high; after that it becomes a little sandy. At fifteen miles we got
into some sand hills, but the feed was still most abundant. I have not
passed through such splendid country since I have been in the colony. I
only hope it may continue. The creek I have named the Finke, after
William Finke, Esquire, of Adelaide, my sincere and tried friend, and one
of the liberal supporters of the different explorations I have had the
honour to lead. Wind south-east. Cloudy.

Thursday, April 5th, Good Country. Started on the same course to some
hills, through sand hills and spinifex for ten miles. Halted for half an
hour to obtain an observation of the sun, 117 degrees 6 minutes. Within
the last mile or two we have passed a few patches of shea-oak, growing
large, having a very rough and thick bark, nearly black. They have a
dismal appearance. The spinifex now ceased, and grass began to take its
place as we approached the hills. From the top of the hill the view is
limited, except to the south-west, where, in the far distance, is a long
range. The country between seems to be scrub, red sand hills, and
spinifex. To the west the country is open, but at five miles is
intercepted by the point of the range that I am about to cross. To the
north-west and east is a mass of flat-topped hills, of every size and
shape, running always to the east. Camped on the head of a small gum
creek, among the hills, which are composed of the same description of
stones as the others. This water hole is three feet deep, and will last a
month or so. The native cucumber is growing here.

Friday, 6th April, Small Gum Creek in Range of Hills. Started on the same
course, 330 degrees, to a remarkable hill, which has the appearance at
this distance of a locomotive engine with its funnel. For three miles the
country is very good, but after that high sand hills succeeded, covered
with spinifex. At six miles we got to one of the largest gum creeks I
have yet seen. It is much the same as the one we saw on the 4th, and the
water in it is running. Great difficulty in crossing it, its bed being
quicksand. We were nearly across, when I saw a black fellow among the
bushes; I pulled up, and called to him. At first he seemed at a loss to
know where the sound came from. As soon, however, as he saw the other
horses coming up, he took to his heels, and was off like a shot, and we
saw no more of him. As far as I can judge, the creek comes from the
south-west, but the sand hills are so high, and the large black shea-oak
so thick, that I cannot distinguish the creek very well. These trees look
so much like gums in the distance; some of them are very large, as also
are the gums in the creek. Numerous tracks of blacks all about. It is the
upper part of the Finke, and at this point runs through high sand hills
(red), covered with spinifex, which it is very difficult to get the
horses through. We passed through a few patches of good grassy country.
In the sand hills the oak is getting more plentiful. We were
three-quarters of an hour in crossing the creek, and obtained an
observation of the sun, 116 degrees 26 minutes 15 seconds. We then
proceeded on the same course towards the remarkable pillar, through high,
heavy sand hills, covered with spinifex, and, at twelve miles from last
night's camp, arrived at it. It is a pillar of sandstone, standing on a
hill upwards of one hundred feet high. From the base of the pillar to its
top is about one hundred and fifty feet, quite perpendicular; and it is
twenty feet wide by ten feet deep, with two small peaks on the top. I
have named it Chambers Pillar, in honour of James Chambers, Esquire, who,
with William Finke, Esquire, has been my great supporter in all my
explorations. To the north and north-east of it are numerous remarkable
hills, which have a very striking effect in the landscape; they resemble
nothing so much as a number of old castles in ruins; they are standing in
the midst of sand hills. Proceeded, still on the same course, through the
sand rises, spinifex, and low sandstone hills, at the foot of which we
saw some rain water, where I camped. To the south-west are some high
hills, through which I think the Finke comes. I would follow it up, but
the immense quantity of sand in its bed shows that it comes from a sandy
country, which I wish to avoid if I can. Wind south-east. Heavy clouds;
very like rain.

Saturday, 7th April, Rain Water under Sandstone Hills. Started on the
same course 330 degrees, over low sand rises and spinifex, for six miles.
It then became a plain of red soil, with mulga bushes, and for seven
miles was as fine a grassed country as any one would wish to look at; it
could be cut with a scythe. Dip of the country to the east, sand hills to
the west; afterwards it became alternate sand hills and grassy plains,
mulga, mallee, and black oak. From the top of one of the sand hills, I
can see a range which our line will cut; I shall make to the foot of that
to-night, and I expect I shall find a creek with water there. Proceeded
through another long plain sloping towards the creek, and covered with
grass. At about one mile from the creek we again met with sand hills and
spinifex, which continued to it. Arrived and camped; found water. It is
very broad, with a sandy bottom, which will not retain water long;
beautiful grass on both banks. Wind east, and cool.

Sunday, 8th April, The Hugh Gum Creek. I have named this creek the Hugh,
and the range James Range. It is scrubby on this side and is not
flat-topped as all the others have been, which indicates a change of
country. On the other side the bearing is nearly east and west. Examined
the creek, but cannot find sufficient water to depend upon for any length
of time; the gum-trees are large. Numerous parrots, black cockatoos, and
other birds. Wind east; very cold during the night.

Monday, 9th April, The Hugh Gum Creek. Started for the highest point of
the James range. At four miles arrived on the top, through a very thick
scrub of mulga; the range is composed of soft red sandstone, long blocks
of it lying on the side. To the east, apparently red sand hills, beyond
which are seen the tops of other hills to the north-east. On the
north-west the view is intercepted by a high, broken range, with two very
remarkable bluffs about the centre. I shall direct my course to the east
bluff, which is apparently the higher of the two. In the intermediate
country are three lower ranges, between which are flats of green grass,
and red sand hills. To the west are grassy flats next to the creek;
beyond these are seen the tops of distant ranges and broken hills; at
about six miles the Hugh seems to turn more to the north, towards a very
rough range of red sandstone. We then descended into a grassy flat with a
few gum-trees. We have had a very great difficulty in crossing the range,
and now I am again stopped by another low range of the same description,
which is nearly perpendicular--huge masses of red sandstone on its side,
and in the valley a number of old native camps. After following the range
three miles, we at last found out a place to cross it. Although this is
not half the height of James range, we encountered far more difficulty;
the scrub was very dense, a great quantity having withered and fallen
down: we could scarcely get the horses to face it. Our course was also
intercepted by deep, perpendicular ravines, which we were obliged to
round after a great deal of trouble, having our saddlebags torn to
pieces, and our skin and clothes in the same predicament. We arrived at
the foot nearly naked, and got into open sandy rises and valleys, with
mulga and plenty of grass, among which there is some spinifex growing. At
sundown, after having gone about eight miles further, we made a large gum
creek, in which we found some water; it is very broad, with a sand and
gravel bottom. Camped, both men and horses being very tired.

Tuesday, 10th April, Gum Creek, Bend of the Hugh. I find our saddle-bags
and harness are so much torn and broken that I cannot proceed until they
are repaired. I am compelled with great reluctance to remain here to-day.
This creek is running to the west. On ascending a sand hill this morning,
I find that it is the Hugh (which seems to drain the sand hills) that we
saw to the east from the top of James range. There is another branch
between us and the high ranges. At about four miles west it seems to
break through the rough range and join the Hugh. A large number of native
encampments here, and rushes are growing in and about the creek: there is
plenty of water.

Wednesday, 11th April, Bend of the Hugh. Got the things put pretty well
to rights, and started towards the high bluff. I find that my poor little
mare, Polly, has got staked in the fetlock-joint, and is nearly dead
lame; but I must proceed. At six miles and a half we again crossed the
Hugh, and at another mile found it coming through the range, which is a
double one. The south range is red sandstone, the next is hard white
stone, and also red sandstone, with a few hills of ironstone; a
well-grassed valley lying between. The two gorges are rocky, and in some
places perpendicular, with some gum-trees growing on the sides. The
cucumber plant thrives here in great quantities, and water is abundant.
At twelve miles we got through both the gorges of the range, which I have
named the Waterhouse Range, after the Honourable the Colonial Secretary.
The country between last night's camp and the range is a red sandy soil,
with a few sand hills, on which is growing the spinifex, but the valleys
between are broad and beautifully grassed. At fifteen miles again crossed
the Hugh, coming from the east, with splendid gum-trees of every size
lining the banks. The pine was also met with here for the first time.
There is a magnificent hole of water here, long and deep, with rushes
growing round it. I think it is a spring; the water seems to come from
below a large bed of conglomerate quartz. I should say it was permanent.
Black cockatoos and other birds abound here, and there are numbers of
native tracks all about. I hoped to-day to have gained the top of the
bluff, which is still seven or eight miles off, and appears to be so very
rough that I anticipate a deal of difficulty in crossing it. I am forced
to halt at this bend of the creek, in consequence of the little mare
becoming so lame that she is unable to proceed further to-day. Our hands
are very bad from being torn by the scrub, and the flies are a perfect
torment. Indications of scurvy are beginning to show themselves upon us.
Wind west; cool night.

Thursday 12th April, The Hugh. Started for the bluff. At eight miles we
again struck the creek coming from the west, and several other gum creeks
coming from the range and joining it. We have now entered the lower hills
of the range. Again have we travelled through a splendid country for
grass, but as we approached the creek it became a little stony. At twelve
miles we found a number of springs in the range. Here I obtained an
observation of the sun. As we approached near the bluff, our route became
very difficult; we could not get up the creek for precipices, and were
obliged to turn in every direction. About two miles from where I obtained
the observation, we arrived with great difficulty at the foot of the
bluff; it has taken us all the afternoon. I expected to have gone to the
top of it to-night, but it is too late. It will take half a day, it is so
high and rough. We are camped at a good spring, where I have found a very
remarkable palm-tree, with light-green fronds ten feet long, having small
leaves a quarter of an inch in breadth, and about eight inches in length,
and a quarter of an inch apart, growing from each side, and coming to a
sharp point. They spread out like the top of the grass-tree, and the
fruit has a large kernel about the size of an egg, with a hard shell; the
inside has the taste of a cocoa-nut, but when roasted is like a potato.
Here we have also the india-rubber tree, the cork-tree, and several new
plants. This is the only real range that I have met with since leaving
the Flinders range. I have named it the McDonnell Range, after his
Excellency the Governor-in-Chief of South Australia, as a token of my
gratitude for his kindness to me on many occasions. The east bluff I have
named Brinkley Bluff, after Captain Brinkley, of Adelaide, and the west
one I have named Hanson Bluff, after the Honourable R. Hanson, of
Adelaide. The range is composed of gneiss rock and quartz.

Friday, 13th April, Brinkley Bluff, McDonnell Range. At sunrise I
ascended the bluff, which is the most difficult hill I have ever climbed;
it took me an hour and a half to reach the top. It is very high, and is
composed principally of igneous rock, with a little ironstone, much the
same as the ranges down the country. On reaching the top, I was
disappointed; the view was not so good as I expected, in consequence of
the morning being so very hazy. I have, however, been enabled to decide
what course to take. To the south-west the Waterhouse and James ranges
seem to join. At west-south-west they are hidden by one of the spurs of
the McDonnell range. To the north-west the view is intercepted by another
point of this range, on which is a high peak, which I have named Mount
Hay, after the Honourable Alexander Hay, the Commissioner of Crown Lands.
About five miles to the north are numerous small spurs, beyond which
there is an extensive wooded or scrubby plain; and beyond that, in the
far distance, is another range, broken by a high conical hill, bearing
about west-north-west, to which I will go, after getting through the
range. To the north-east is the end of another range coming from the
south. On this, which I have named Strangway Range, after the Honourable
the Attorney-General, is another high hill. Beyond is a luminous, hazy
appearance, as if it proceeded from a large body of water. A little more
to the east there are three high hills; the middle one, which I should
think is upwards of thirty miles from us, is the highest, and is bluff at
both ends; it seems to be connected with Strangway range. To the east is
a complete mass of ranges, with the same luminous appearance behind them.
I had a terrible job in getting down the bluff; one false step and I
should have been dashed to pieces in the abyss below; I was thankful when
I arrived safely at the foot. I find that I have taken the wrong creek to
get through the bluff. The Hugh still comes in that way, but more to the
westward. Started at 10 o'clock; the hills very bad to get over; wind
easterly. Camped at sundown on the creek; there is an abundance of water,
which apparently is permanent, from the number of rushes growing all
about it. The feed is splendid. There are a number of fresh native
tracks.

Saturday, 14th April, McDonnell Range. Started at 8 o'clock to follow the
creek, as it seems to be the best way of getting through the other
ranges; but, as it comes too much from the east, I must leave it, and get
through at some of the low hills further down. This we at last contrived
to do after a severe struggle. It has taken us the whole day to come
about five miles. We are now camped, north of the bluff, at a gorge, in
which there is a good spring of water; the creeks now run north from the
range.

Sunday, 15th April, The North Gorge of McDonnell Range. I ascended the
high hill on the east side of the gorge; the atmosphere being much
clearer, I got a better view of the country. To the north-west, between
the McDonnell range and the conical hill north-north-west, is a large
plain, apparently scrub; no hills on the horizon, but a light shade in
the far distance; the conical hill bears 340 degrees from this; it
appears to be high. From the foot of this, for about five miles, is an
open grassy country, with a few small patches of bushes. A number of gum
creeks come from the ranges, and seem to empty themselves in the plains.
The country in the ranges is as fine a pastoral hill-country as a man
would wish to possess; grass to the top of the hills, and abundance of
water through the whole of the ranges. I forgot to mention that the nut
we found on the south side of the range is not fit to eat; it caused both
men to vomit violently. I ate one, but it had no bad effect on me.

Monday, 16th April, The North Gorge of McDonnell Range. Started at 9
o'clock to cross the scrub for the distant high peak. For five miles the
plain was open and well grassed: afterwards it became thick, with mulga
bushes and other scrubs. At twenty miles we again encountered the
spinifex, which continued until we camped after dark. Distance, thirty
miles. Met with no creek or watercourse after leaving the McDonnell
ranges.

Tuesday, 17th April, In the Scrub. Got an early start, and continued
through the scrub and spinifex on the same course, 340 degrees. At three
miles passed a small stony hill, about two miles to the west of our
course. At eighteen miles saw to the west two prominent bluff hills, and
two or three small ones, about ten miles distant from us. At thirty-two
miles crossed a strong rise. There are three reap-hook hills about three
miles west, their steep side facing the south. At sundown reached the
hills. At two miles passed a small sandy gum creek, the only watercourse
we have seen between the two ranges. Followed the range to the north-west
till after dark, hoping to find a gum creek coming from the range, but
without success; nothing but rocky and sandy watercourses. Camped. The
poor horses again without water; I trust that I shall find some for them
in the morning; if not, I shall have to return to the McDonnell range.
Very little rain seems to have fallen here; the grass is all dried up.
The spinifex continues until within a mile of the range. The small gum
creek that we passed is running south-west into the scrub.

Wednesday, 18th April, Under the High Peak, Mount Freeling. At daybreak
sent Kekwick in search of water, while I ascended the high mount to see
if any could be seen from that place. To my great delight I beheld a
little in a creek on the other side of the range, bearing 113 degrees,
about a mile and a half. I find this is not quite the highest point of
the range; there is another hill, still higher, about fifteen miles
further to the north-north-west. About two miles off I can see a gum
creek looking very green, coming from the range in the direction in which
I have sent Kekwick, where I hope he will find water. The country from
west to north-east is a mass of hills and broken ranges; to the
south-west high broken ranges. To the north-north-east is another hill,
with a plain of scrub between. To the south-east scrub, with tops of
hills in the far distance. Brinkley Bluff bears 166 degrees and Mount Hay
186 degrees. Returned to the camp, and find to my great satisfaction that
Kekwick has discovered some water in the creek about two miles off. I am
very glad of it, for I am sure that some of my horses would not have
stood the journey back without it. I must not leave this range without
endeavouring to find a permanent water, as no rain seems to have fallen
to the north of us; everything is so dry, one would think it was the
middle of summer. The sun is also very hot, but the nights and mornings
are cool. Wind east. Old tracks and native camps about. The range is
composed of the same description of rocks as the McDonnell ranges, with
rather more quartz than mica. We here found new shrubs and flowers, also
a small brown pigeon with a crest. I have built a small cone of stones on
the peak, and named it Mount Freeling, after the Honourable Colonel
Freeling, Surveyor-General. The range I have called the Reynolds, after
the Honourable Thomas Reynolds, the Treasurer.

Thursday, 19th April, Mount Hugh. The horses separated during the night,
and were not found until after one o'clock. Moved to the east side of the
mount to where I had seen the water from the top. We found plenty of
water in the gum creek which is the head of the one we crossed on Tuesday
night, just before making the range. We were obliged to come a long way
round before we could get to it, the hills being all rough sharp rocks,
impassable for horses; abundance of grass with a little spinifex on the
hills. At this camp I have marked a tree "J. M.D. S."; the cone of stones
on the top of the mount bears 293 degrees. Ten miles distant in a branch
creek about half a mile to the north of this is more water; and a little
higher up, in a ledge of rocks, is a splendid reservoir of water, thirty
yards in diameter and about one hundred yards in circumference. We could
not get to the middle to try the depth, but where we tried it it was
twelve feet deep. A few yards higher up is another ledge of rocks, behind
which is a second reservoir, but smaller, having a drainage into the
former one. Native tracks about. Wind north. I have named this Anna's
Reservoir, after Mr. James Chambers' youngest daughter.

Friday, 20th April, East Side of Mount Hugh. Started to the south-east to
find a crossing place over the range; this was not an easy matter, from
the roughness of the hills; at last, however, we got over it. On the
other side we found a large gum creek with water in it, running to the
north-east. Camped. The range is well grassed, with gum creeks coming
from it, and a little mulga scrub. Here we have discovered a new tree,
whose dark-green leaf has the shape of two wide prongs; the seed or bean,
of which I have obtained a few, is of a red colour; the foliage is very
thick. The stem of the largest we have seen is about eighteen inches in
diameter. The wood is soft; when in the state of a bush it has thorns on
it like a rose. Here we have also obtained some seed of the vegetable we
have been using; we have found this vegetable most useful; it can be
eaten as a salad, boiled as a vegetable, or cooked as a fruit. We have
also some other seeds of new flowers. The bearing from this to the cone
of stones on Hugh Mount, 233 degrees 45 minutes.

Saturday, 21st April, Gum Creek, East Side of Mount Freeling. Started at
half-past seven across the scrub to another high hill. For seven miles
the scrub is open, and the land beautifully grassed. At twelve miles from
the camp we crossed another gum creek, coming from the range; as far as I
could see it ran to the north-east. After seven miles the scrub became
much thicker. We had great difficulty in getting through, from the
quantity of dead timber, which has torn our saddle-bags and clothes to
pieces. There are a number of gum-trees, and the new tree that was found
on Captain Sturt's expedition, 1844, but mulga predominates. At fourteen
miles we struck a large gum plain, but after a short time again entered
the scrub. At about twenty-two miles met another arm of the gum plains,
with large granite rocks nearly level with the surface. We found rain
water in the holes of these rocks. At thirty-two miles crossed the sandy
bed of a large gum creek divided into a number of channels; too dark to
see any water. Four miles further on, camped on a small gum creek with a
little rain water; the creeks are running to the north-east. The soil is
of a red sandy colour: the grass most abundant throughout the whole day's
journey. Occasionally we met with a few hundred yards of spinifex. Wind
south-east. Native tracks quite fresh in the scrub and plain; we also
passed several old worleys.

Sunday, 22nd April, Small Gum Creek, under Mount Stuart, Centre of
Australia. To-day I find from my observations of the sun, 111 degrees 00
minutes 30 seconds, that I am now camped in the centre of Australia. I
have marked a tree and planted the British flag there. There is a high
mount about two miles and a half to the north-north-east. I wish it had
been in the centre; but on it to-morrow I will raise a cone of stones,
and plant the flag there, and name it Central Mount Stuart. We have been
in search of permanent water to-day, but cannot find any. I hope from the
top of Central Mount Stuart to find something good to the north-west.
Wind south. Examined a large creek; can find no surface water, but got
some by scratching in the sand. It is a large creek divided into many
channels, but they are all filled with sand; splendid grass all round
this camp.

Monday, 23rd April, Centre. Took Kekwick and the flag, and went to the
top of the mount, but found it to be much higher and more difficult of
ascent than I anticipated. After a deal of labour, slips, and knocks, we
at last arrived on the top. It is quite as high as Mount Serle, if not
higher. The view to the north is over a large plain of gums, mulga, and
spinifex, with watercourses running through it. The large gum creek that
we crossed winds round this hill in a north-east direction; at about ten
miles it is joined by another. After joining they take a course more
north, and I lost sight of them in the far-distant plain. To the
north-north-east is the termination of the hills; to the north-east, east
and south-east are broken ranges, and to the north-north-west the ranges
on the west side of the plain terminate. To the north-west are broken
ranges; and to the west is a very high peak, between which and this place
to the south-west are a number of isolated hills. Built a large cone of
stones, in the centre of which I placed a pole with the British flag
nailed to it. Near the top of the cone I placed a small bottle, in which
there is a slip of paper, with our signatures to it, stating by whom it
was raised. We then gave three hearty cheers for the flag, the emblem of
civil and religious liberty, and may it be a sign to the natives that the
dawn of liberty, civilization, and Christianity is about to break upon
them. We can see no water from the top. Descended, but did not reach the
camp till after dark. This water still continues, which makes me think
there must certainly be more higher up. I have named the range John
Range, after my friend and well-wisher, John Chambers, Esquire, brother
to James Chambers, Esquire, one of the promoters of this expedition.

Tuesday, 24th April, Central Mount Stuart. Sent Kekwick in search of
water, and to examine a hill that has the appearance of having a cone of
stones upon it; meanwhile I made up my plan, and Ben mended the
saddlebags, which were in a sad mess from coming through the scrub.
Kekwick returned in the afternoon, having found water higher up the
creek. He has also found a new rose of a beautiful description, having
thorns on its branches, and a seed-vessel resembling a gherkin. It has a
sweet, strong perfume; the leaves are white, but as the flower is
withered, I am unable to describe it. The native orange-tree abounds
here. Mount Stuart is composed of hard red sandstone, covered with
spinifex, and a little scrub on the top. The white ant abounds in the
scrubs, and we even found some of their habitations near the top of Mount
Stuart.

Wednesday, 25th April, Central Mount Stuart. There is a remarkable hill
about two miles to the west, having another small hill at the north end
in the shape of a bottle; this I have named Mount Esther, at the request
of the maker of the flag. Started at 9 o'clock, on a course a little
north of west, to the high peak that I saw from the top of Mount Stuart,
which bears 272 degrees. My reason for going west is that I do not like
the appearance of the country to the north for finding water; it seems to
be sandy. From the peak I expect to find another stratum to take me up to
the north-north-west. Around the mount and on the west side, the country
is well grassed, and red sandy soil; no stones. To the north and south of
our line are several isolated hills, composed principally of granite. At
ten miles there is a quartz reef on the north side of the south hills. At
twelve miles struck a gum creek coming from the south and running to the
north; it has three channels. We found a little rain water in one, and
camped, to enable us to finish the mending of the saddle-bags. Wind east;
very cold morning and night. The large creek that flowed round Mount
Stuart is named the Hanson, after the Honourable R. Hanson, of Adelaide.

Thursday, 26th April, Gum Creek on West Course. Started at a quarter past
8 o'clock on the same course for the high peak. At two miles crossed some
low granite and quartz hills; and at four miles crossed a gum creek
running to the north with sand and gravel beds. No water. The country
then became difficult to get through, in consequence of the number of
dead mulga bushes. At ten miles the grass ceased, and spinifex took its
place, and continued to the banks of the next gum creek, which we crossed
at twenty-two miles; the bed sandy, and divided into a number of
channels, coming from the south-east, and running a little to the east of
north, but no water in them. Native tracks in its bed. On the west side
of the creek the grass again begins, and continues to the hills, where we
arrived at five minutes to 7. Camped without water. There seems to have
been very little rain here--the grass and everything else is quite dry.
Distance, thirty-eight miles.

Friday, 27th April, East Side of Mount Denison. Sent Kekwick to the
south-west to a remarkable hill which I hope may yield some water, with
orders to return immediately if he should find any nearer, so that we
might get some for the horses. I waited till past 12, but he did not
return, so I started, intending to go to the top of the mount. On getting
to the north-east side of the ranges, I liked the appearance of the
country for water, and seeing that the top of the mount was still some
distance off, and that it would make it too late to return, I set to work
myself to look for water. After an hour's search I was successful,
finding some rain water in a gum creek coming from the hills. The natives
must have been there quite recently, as their fires were still warm; and,
as I had left the camp and provisions with only one man, I hurried back,
had the horses saddled and packed, and brought them down to the water,
leaving a note for Kekwick to follow in a west-north-west direction to a
gum creek about three miles distant. Kekwick's search was also
successful; he found permanent water under the high peak to which I sent
him, and which I have named Mount Leichardt, in memory of that
unfortunate explorer, whose fate is still a mystery. I have seen no trace
of his having passed to the westward. Kekwick describes the water he has
found as abundant and beautifully clear, springing out of conglomerate
rock much resembling marble; its length is upwards of a quarter of a
mile, falling into natural basins in the solid rock, some six feet in
depth and of considerable capacity. The country round the base of the
range is covered with the most luxuriant grass and vegetation. Mount
Leichardt and the range are composed, at their base, of a soft
conglomerate rock in immense irregular masses, heaped one on the other;
the higher part where the spring appears is of the same conglomerate, but
broad and solid, having smooth faces, which makes the ascent very
difficult.

Saturday, 28th April, Gum Creek under Mount Denison. As soon as the
horses were caught I started for the top of the mount. I left my horse in
a small rocky gum creek which I thought would lead me to the foot of the
mount. At about a quarter of a mile from the mouth of the gorge, I came
upon some water in a rocky hole, followed it up, and, two hundred yards
further, was stopped by a perpendicular precipice with water trickling
over it into a large reservoir. I had now to take to the hills, which
were very rough, and after a deal of difficulty I arrived, as I thought,
at the top, but to my disappointment I had to go down a fearfully steep
gully. At it I went, and again I arrived, as I fancied, at the top, but
here again was another gully to cross, and a rise still higher. I have at
last arrived at the summit, after a deal of labour and many scratches.
This is certainly the highest mountain I have yet ascended; it has taken
me full three hours to get to the summit. The view is extensive, but not
encouraging. Central Mount Stuart bears 95 degrees. Mount Leichardt, 155
degrees 30 minutes. To the south, broken ranges with wooded plains before
them, and in the far distance, scarcely visible, appears to be a very
high mountain, a long, long way off. To the south-west the same
description of range. About thirty miles to the west is a high mount with
open country, and patches of woodland in the foreground. At the
north-west there appears to be an immense open plain with patches of
wood. To the north is another plain becoming more wooded to the
north-east. As this is the highest mountain that I have seen in Central
Australia, I have taken the liberty of naming it Mount Denison, after his
Excellency Sir William Denison, K.C.B., Governor-General. The next range
(bearing 334 degrees), being the last of the highest ones north, I have
named Mount Barkly, after his Excellency Sir Henry Barkly,
Governor-in-Chief of Victoria. When on the second highest point of this
mount, I saw a native smoke rise up in the creek below, a short distance
from where I had tied my horse. This naturally made me very anxious for
his safety, and when I descended I was rejoiced to find him safe. The
natives have been in the creek and on the mount: their tracks, which are
quite fresh, lead me to conclude that they have been running. The descent
was difficult, but I discovered a shorter route, and it has taken me two
hours to come down. Arrived at the camp at 4.30, and found all right. I
intended to have built a large cone of stones on the summit; but, when I
arrived there, I was too much exhausted to do so. I have, however,
erected a small one, placing a little paper below one of the stones, to
show that a white man has been there. I have also marked a tree "J. M.D.
S." on the creek where we are now camped. Mount Denison bears from here
249 degrees.

Sunday, 29th April, Gum Creek under Mount Denison. Latitude, 21 degrees
48 minutes. Variation, 3 degrees 20 minutes east. Mount Denison and the
surrounding hills are composed of a hard reddish-brown sandstone. About
one hundred yards from the summit is a course of conglomerate, composed
of stones from half an inch to four inches in diameter, having the
appearance of being rounded at a former period by water. From the foot to
the top of this course is about ten feet, and the breadth on the top is
about twelve feet. There is red sandstone on the summit, with three or
four pines growing. The mount and adjoining hill are covered with
spinifex, but the plain is grassed. The wind has now changed to the west,
and it is much hotter.

Monday, 30th April, Under Mount Denison. The wind changed again to the
south-east during the night, and is much colder. Started on a course, 315
degrees, across the plain towards Mount Barkly. The highest point of the
mount is eighteen miles distant from our camp on the creek. We had to
round the west side of it, finding no water until we came upon a little
in the gorge coming from the highest point. It was dark before we
arrived, so that we could not take the horses up to-night. Wind
south-east, blowing a hurricane, and very cold.

Tuesday, 1st May, North-west Side of Mount Barkly. On examining the
water, I find it is only a drainage from the rocks, and there is not more
than two gallons for each horse. I ascended the hill, but could see
nothing more than I had seen from Mount Denison. The base is composed of
a hard red sandstone, the top of quartz rock. I do not like the
appearance of the country before us. Started on a course of 335 degrees,
and at six miles and a half came upon a large gum creek divided into
numerous channels: searched it carefully, without finding any surface
water; but I discovered a native well about four feet deep, in the east
channel, close to a small hill of rocks. Cleared it out, and watered the
horses with a quart pot, which took us long after dark--each horse
drinking about ten gallons, and some of them more. Natives have been here
lately, and from the tracks they seem to be numerous. We also observed
the rose-coloured cockatoo. I have named this creek The Fisher, after Sir
James Hurtle Fisher; it runs a little east of north.

Wednesday, 2nd May, The Fisher. We did not start until 11 o'clock in
consequence of it taking a long time to water the horses. We steered for
some hills that I had seen from the top of the last two mounts. At
thirteen miles arrived at the hills, but found them low, and no
appearance of water. Changed my course west 35 degrees north to some
higher hills. At 6.30 camped in the scrub without water. The country from
Mount Denison to this is a light-red sandy soil, covered with spinifex,
with very little grass, and is nearly a dead level. In some places it is
scrubby, having a number of gum-trees, and the new tree of Captain Sturt
growing all over it. From a distance it has the appearance of a good
country, and is very deceiving; you constantly think you are coming upon
a gum creek. Wind south-east; very cold at night and morning.

Thursday, 3rd May, Spinifex and Gum Plains. Started on the same course,
west 35 degrees north, and at four miles reached the top of the hills,
which are low and composed of dark-red sandstone and quartz. The bearings
to Mount Denison, 146 degrees; Mount Barkly, 142 degrees; to another hill
west-north-west, 302 degrees, distant about ten miles, which I have named
Mount Turnbull, after the late Gavin Turnbull, Esquire, Surgeon in the
Indian Army. The morning is very hazy, and I cannot see distinctly;
besides, my eyes are again very bad. The appearance of the country all
round is that of having gum creeks everywhere. To the north there are
some more low hills. A short distance off, on a bearing of 328 degrees,
there appears to be a gum creek with something white as if it were water,
so I shall change my course. At 3.50 camped, some of my horses being
nearly done up from want of water, and having nothing to eat but
spinifex. I have now come eighteen miles, and the plain has the same
appearance now as when I first started--spinifex and gum-trees, with a
little scrub occasionally. We are expecting every moment to come upon a
gum creek, but hope is disappointed. I have not so much as seen a
water-course since I left the Fisher, and how far this country may
continue it is impossible to tell. I intended to have turned back sooner,
but I was expecting every moment to meet with a creek. It is very
alluring, and apt to lead the traveller into serious mistakes. I wish I
had turned back earlier, for I am almost afraid that I have allowed
myself to come too far. I am doubtful if all my horses will be able to
get back to water. In rainy weather this country will not retain the
water on the surface, and we have not so much as seen a clay-pan of the
smallest dimensions. The gum-trees on this plain have a smooth white
bark, and the leaves are some light-green and some dark. Most of the
trees seem very healthy; there are very few dead ones about. To-morrow
morning I must unwillingly retreat to water for my horses. There is no
chance of getting to the north-west in this direction, unless this plain
soon terminates. From what I could see there is little hope of its doing
so for a long distance.

Friday, 4th May, Gum and Spinifex Plains. At times this country is
visited by blacks, but it must be seldom, as since we left the Fisher we
have only seen the track of one, who seems to have come from the east,
and to have returned in that direction. The spinifex in many places has
been burnt, and the track of the native was peculiar--not broad and flat,
as they generally are, but long and narrow, with a deep hollow in the
foot, and the large toe projecting a good deal; the other in some
respects more like the print of a white man than of a native. Had I
crossed it the day before, I would have followed it. My horses are now
suffering too much from the want of water to allow me to do so. If I did,
and were not to find water to-night, I should lose the whole of the
horses and our own lives into the bargain. I must now retreat to Mount
Denison, which I do with great reluctance; it is losing so much time, and
my provisions are limited. Started back at 7.10 a.m., and at thirty miles
came upon a native well, with a little grass round it; the bottom was
moist. Unsaddled, and turned the horses out. Commenced clearing out the
well the best way we could, with a quart pot and a small tin dish, having
unfortunately lost our shovel in crossing the McDonnell ranges. We had
great difficulty in keeping the horses out while we cleared it. To our
great disappointment we found the water coming in very slowly. We can
only manage, in an hour and a half, to get about six gallons, which must
be the allowance for each horse, and it will take us till to-morrow
morning to water them all. One of us is required to be constantly with
them to keep them back, and that he can hardly do; some of them will get
away from him do all he can. Kekwick's horse was nearly done up before we
reached this place; also one of the others. Those nearest to the
cart-breed give in first.

Saturday, 5th May, Native Well. Got all the horses watered by 11 o'clock
a.m., and could only get about five gallons for each horse, although we
were employed the whole of the night, and got no sleep. Started for the
Fisher, and arrived at the native well at sundown. Were obliged to tie
the horses up, to keep them from getting into it. We could scarcely get
some of them as far as this, as they are quite done up. What was still
worse, we found the native well had fallen in since we left. It cannot be
helped: we must take things as they come. Commenced immediately to cut a
number of stakes, rushes, and grass, to keep the sand back, and by 3
o'clock in the morning we got them all watered, and very thankful we were
to do so. It has been, and is still, bitter cold throughout the night and
morning, the wind still coming from the south-east. We had a pot of tea,
although we could ill afford it, and lay down and got a little sleep,
completely tired and worn out with hard work and want of rest.

Sunday, 6th May, The Fisher. Got up at daybreak and went to the well, but
found that the rascals of horses had been there before us, and trodden in
one side of the well. They had as much water last night and morning as
they could drink, and the quantity that some of them drank was enormous.
I had no idea that a horse could hold so much, yet still they want more.
I shall remain here two days, put down more stakes, clear out the well,
and give them as much as they will drink. During this trying time I have
been very much pleased with the conduct of Kekwick and Ben; they have
exerted themselves to the utmost, and everything has been done with the
greatest alacrity and cheerfulness. Although they have only had two
hours' sleep during the last two nights, there has not been a single word
of dissatisfaction from either of them, which is highly gratifying to me.
It is, indeed, a great pleasure to have men that will do their work
without grumbling. Watered the horses as they came in. They do not now
drink a fourth part of what they did at first.

Monday, 7th May, The Fisher. Had a good night's rest, and felt recovered
from the past fatigue. Started for the creek on the east side of Mount
Denison, to the water at which we camped before, keeping to the north
side of Mount Barkly in search of water, but could find none. Arrived at
the creek after dark. Kekwick's horse is entirely done up; he had to get
off and lead him for two miles. Another of the horses is nearly as bad,
but he managed to get to the creek. We found the water greatly reduced,
but still enough for us.

Tuesday, 8th May, Creek East of Mount Denison. I must remain here two
days to allow the horses to recover. I am afraid if we have such another
journey, I shall have to leave some of them behind. I do not know what is
the cause of their giving in so soon; I have had horses that have
suffered three times as much privation, and yet have held out. The light
ones are all right; it is the heavy ones, of the cart-horse breed, that
feel it most. I had been keeping them up on purpose for an occasion like
this, and they all looked in first-rate condition, but the work of the
past week has made a great alteration in some of them. I suppose the
young grass is not yet strong enough for them. It is very vexing to be
thus disappointed and delayed. To think that they should fail me at the
very moment when I expected them to do their best, and after all the
trouble and loss of time I have incurred in giving them short journeys!
However, I cannot improve it by complaining, and must rest contented and
hope for the best. Wind south-east. Storm brewing.

Wednesday, 9th May, Creek East of Mount Denison. Resting horses and
putting our things in order. Wind blowing very strong from the
south-east; it has continued nearly in the same quarter since March.

Thursday, 10th May, Creek East of Mount Denison. I find that I must give
the horses another day; they have not yet recovered, and I expect we
shall have some more hard work for them. We have not quite finished
mending.

Friday, 11th May, Creek East of Mount Denison. Ben was taken very ill
during the night, and is still so bad that I am obliged to remain here
another day. Afternoon: Ben feels much better, so I shall start
to-morrow.

Saturday, 12th May, Creek East of Mount Denison. Ben is better, and the
horses look as if they can stand a little more hardship. Started at 8.20
on a bearing of 28 degrees east of north, to see if I can get on in that
direction. For fourteen miles our course was through mulga scrub and
spinifex, in some places very thick. At twenty-seven miles camped without
water. The country that we have passed over the last two days is
apparently destitute of water, even in rainy weather. I do not think the
ground would retain it a single day. Very little feed for the horses.

Sunday, 13th May, Scrub and Gum Flat. I do not like the appearance of the
country. As I can see no hope of obtaining water on this course, I shall
change to the east, in order to cut the large gum creek that I crossed on
the 26th ultimo, and, if I find water in it, to follow it out to wherever
it goes. At three miles cut a small gum creek: searched for water both up
and down, but could find none, nor any appearance of it. Still keeping my
east course, we then passed through a very thick mulga scrub, and at ten
miles struck a low range of hills, composed of quartz, with a conical
peak, which I ascended. The prospect from this is very extensive, but
disheartening, apparently the same sort of scrubby country that I have
endeavoured to break through to the north-west. The view to the north is
dismal; there are a few isolated hills, seemingly the termination of John
range, and of the same formation as this that I am now on. To
east-south-east there appears to be a creek, to which I shall now go. At
three miles I reached what I had supposed to be a creek, but it is a
small narrow gum flat which receives the drainage from this low range. We
found a hole where there had been water, but it was all gone. I have
named the peak Mount Rennie, after Major Rennie of the Indian army. In
this small flat we shot a new macaw, which I shall carry with me, and
preserve the skin, if we get to water to-night. The front part of the
neck and underneath the wings is of a beautiful crimson hue, the back is
of a light lead colour, the tail square, the beak smaller than a
cockatoo's, and the crest the same as a macaw's. After leaving this flat,
we passed through some scrub, and came upon another of the same
description. Here I narrowly escaped being killed. My attention being
engaged looking for water, my horse took fright at a wallaby, and rushed
into some scrub, which pulled me from the saddle, my foot and the staff
that I carry for placing my compass on catching in the stirrup-iron.
Finding that he was dragging me, he commenced kicking at a fearful rate;
he struck me on the shoulder joint, knocked my hat off, and grazed my
forehead. I soon got clear, but found the kick on my shoulder very
painful. Mounted again, and at seven miles we came upon some more low
hills with another prominent peak of a dark-red sandstone. This I have
named Mount Peake, after E.J. Peake, Esquire, of Adelaide. I now find
that the gum creek which I crossed between Central Mount Stuart and Mount
Denison runs out and forms the gum plains we have just passed. No hope of
water. I must now bear in for the centre to get it. Passed through a very
thick, nasty mulga scrub for five miles, and camped again without water
under some low stony hills. I feel the effects of my accident very much.

Monday, 14th May, Stony Hills, Mulga Scrub. Feel very stiff and ill.
Started at daylight, and passed through three belts of thick mulga scrub,
between which there were low stony hills. At three miles passed a small
gum creek, emptying itself into the scrub. At seventeen miles passed
another, doing the same; at twenty miles another, and at twenty-four
miles a third, under the hills north-west of Central Mount Stuart. This
has a very remarkable hill at the north-west, in the shape of a large
bottle with a long neck. We have had the greatest difficulty in getting
all our horses to the water; three of them are very bad; two have been
down a dozen times during the journey to-day. On approaching the range,
we passed through some large patches of kangaroo grass, growing very
thickly, and reaching to my shoulder when in the saddle.

Tuesday, 15th May, Centre. The horses look very bad to-day; I shall
therefore give them three or four days' rest. It is very vexing, but it
cannot be helped. The water here will last about ten days. I shall cause
another search for more to be made; I myself am too unwell to assist.
Yesterday I rode in the greatest pain from the effects of my fall, and it
was with great difficulty that I was able to sit in the saddle until we
reached here. Scurvy also has taken a very serious hold of me; my hands
are a complete mass of sores that will not heal, but, when I remain for
two or three days in some place where I can get them well washed, they
are much better; if not, they are worse than ever, and I am rendered
nearly helpless. My mouth and gums are now so bad that I am obliged to
eat flour and water boiled. The pains in my limbs and muscles are almost
insufferable. Kekwick is also suffering from bad hands, but, as yet, has
no other symptoms. I really hope and trust that it will not be the cause
of my having to turn back. I suffered dreadfully during the past night.
This afternoon the wind has changed to the west--the first time since
March; a few clouds are coming up in that direction.

Wednesday, 16th May, Centre. I despatched Kekwick at daybreak in search
of permanent water, with orders to devote the whole of two days to that
purpose. I must now do everything that is in my power to break this
barrier that prevents me from getting to the north. If I could only get
one hundred and twenty miles from this, I think there would be a chance
of reaching the coast. I wish the horses could endure the want of water a
day or two longer, but I fear they cannot; this last journey has tried
them to their utmost. Two of them look very wretched to-day, and will
with difficulty get over it; one I scarcely think will do so. I should
not have been afraid to have risked two more days with five of them. If
they had been all like these five, I should have tried to the north-west
a degree and back again without water. I have been suffering dreadfully
during the past three weeks from pains in the muscles, caused by the
scurvy, but the last two nights they have been most excruciating. Violent
pains darted at intervals through my whole body. My powers of endurance
were so severely tested, that, last night, I almost wished that death
would come and relieve me from my fearful torture. I am so very weak that
I must with patience abide my time, and trust in the Almighty. This
morning I feel a little easier; the medicines I brought with me are all
bad, and have no effect. The wind still from the north-west, with a few
light clouds. Towards sundown the wind has changed to the south-west;
heavy clouds coming from the north-west.

Thursday, 17th May, Centre. Wind from the south; the heavy clouds
continued until sunrise, and then cleared off. I fully expected some
rain, but was disappointed. I have again had another dreadful night of
suffering; I had, however, about two hours' sleep, which, as it was the
only sleep I have had for the last three nights, was a great boon. This
morning I observe that the muscles of my limbs are changing from
yellow-green to black; my mouth is getting worse, and it is with
difficulty that I can swallow anything. I am determined not to give in; I
shall move about as long as I am able. I only wish the horses had been
all right, and then I should not have stayed here so long. Kekwick
returned at 3 o'clock, and reported having found water in the Hanson,
about fifteen miles from Central Mount Stuart, but only a small supply.
Beyond that the creek divides into two, one running north and the other
east, but he could see no more water further down. He also saw two
natives, armed with long spears, about three hundred yards off; they did
not observe him, and he thought it most prudent not to show himself, but
to remain behind a thick bush until they were gone. In this instance I
regret his caution, for I am anxious to see or hear what is the
appearance of the Central natives. Wind variable, with heavy clouds from
north-west.

Friday, 18th May, Centre. I have again had a very bad night, and feel
unable to move to-day. Wind the same.

Saturday, 19th May, Centre. I had a few hours' sleep last night, which
has been of great benefit to me. I shall attempt to move down to the
water in the Hanson. Arrived there about 1.30 completely done up from the
motion of the horse. The water is a few inches below the surface in the
sand. East side of Mount Stuart bearing 250 degrees, about ten miles
distant. I do not think the water is permanent.

Sunday, 20th May, The Hanson. Another dreadful night for me. Wind and
clouds still coming from the north-west, but no rain.

Monday, 21st May, The Hanson. Unable to move; very ill indeed. When shall
I get relief from this dreadful state?

Tuesday, 22nd May, The Hanson. I got a little sleep last night, and feel
a great deal easier this morning, and shall try my horse back again. I
shall now steer north-east to a range of hills that I saw from the top of
Central Mount Stuart, and hope from these to obtain an entrance to the
north-west or north-east. I also hope to cut the creek that carries off
the surplus water from all the creeks which I have passed since March. It
must go somewhere, for it is difficult to believe that those numerous
bodies of water can be consumed by evaporation. Started on a bearing of
48 degrees, crossed the Hanson, running a little on our right; at six
miles crossed it again, running more to the north for two miles further.
We crossed four more of its courses, all running in the same direction.
The most easterly one is spread over a large salt-creek valley, and forms
a lagoon at the foot of some sand ridges, the highest of which is ten
miles and a half from our last camp. On the east side of it there is a
large lagoon, five miles long by one mile and a half broad, in which
water has lately been, but it is now dry. We then proceeded through a
little scrub, with splendid grass, and at twelve miles cut a small gum
creek, coming from the range. We saw a number of birds about, and there
were tracks of natives, quite fresh, in the creek. Sent Kekwick down it
to see if there were water, while I went up and examined it. This is the
large gum plain that we met with the day we made the Centre; it is
completely covered with grass. Kekwick ran the creek out. At about two
miles he observed a little water in the creek, where the natives had been
digging. He also came upon two of them, and two little children. They did
not observe him until he was within fifty yards, when they stood for a
few minutes paralysed with astonishment; then, snatching up the children,
ran off as quickly as their legs could carry them. They did not utter a
sound, although he called to them. He remarked that they had no hair on
their heads, or it was as short as if it had been burned off close. I
wish I had seen them; I should have overtaken them and seen if it were a
fact that the hair was burnt. It is reported in Adelaide that there are
natives in the interior without hair on their bodies. At fourteen miles
we again struck the creek, and found plenty of water in it. It winds all
over the plain in every direction. Camped for the night very much done
up. I could hardly sit in my saddle for this short distance. Wind
north-west.

Wednesday, 23rd May, Gum Creek, East Range, the Stirling. The wind has
changed again to the south-east. I have named this creek the Stirling,
after the Honourable Edward Stirling, M.L.C. Followed it into the range
on the same course towards a bluff, where I think I shall find an easy
crossing. At one mile from the camp the hills commenced on the south-east
side of the creek, but on the north-west side they commenced three miles
further back. There was abundance of water in the creek for thirteen
miles; at ten miles there was another large branch with water coming from
the south-east. At fourteen miles ascended the bluff and obtained the
following bearings: South side of the creek, to a high part of the range
about two miles off (which I have named Mount Gwynne, after his Honour,
Justice Gwynne), 186 degrees. North side of the creek, to another hill
about two miles and a half off (which I have named Mount Mann, in memory
of the late Commissioner of Insolvency), 249 degrees. Central Mount
Stuart bears 131 degrees to the highest point. At the north-west
termination of the next range, to which I shall now go, there are two
very large hills, the north one, which is the highest, I have named Mount
Strzelecki, after Count Strzelecki, bearing 358 degrees. I have named the
high peak on the same range Mount Morphett, after the Honourable John
Morphett, M.L.C. The view from this bluff is extensive, except to the
west-north-west, which is hidden by this range just alluded to, which I
have named Forster Range, after the Honourable Anthony Forster, M.L.C.
From the south-west it has the appearance of a long continuous range,
but, on entering it, it is much broken into irregular and rugged hills:
on this side, the north-east, it consists of table-hills, with a number
of rugged isolated ones on the north side. To the north-west there is
another scrubby and gum-tree plain; to the north-north-west are some
isolated low ranges; to the north are grassy plains and low ranges; to
the east are several spurs from this range, which is composed of a very
hard dark-red stone, mixed with small round quartz and ironstone, and in
some places a hard flinty quartz. The range and hills are covered with
spinifex, but the valleys are beautifully grassed. We descended, and at
four miles struck a creek coming from the range, and running between two
low ranges towards the north-east. At seven miles changed my course to
north-east to camp in the creek, and endeavour to get water for the
horses before encountering the scrubby plains to-morrow morning. At five
miles came upon a low range, but no creek; it must have gone further to
the eastward. It being now quite dark, we camped under the ranges. Since
I changed my course I have come through a patch of mulga and other scrubs
with plenty of grass, but no watercourses. Wind south-east; heavy clouds
from the north-west; lightning in the south and west.

Thursday, 24th May, Range of Low Hills. This morning I feel very ill from
climbing the bluff yesterday; I had no sleep during the night, the pains
being so very violent. About 9 o'clock we had a heavy shower of rain, and
a little more during the night. Very late before the horses were found,
and the atmosphere very thick, with the prospect of rain for the rest of
the day. This and my being so ill have decided me to remain here until
to-morrow, there being sufficient rain water for the horses. A few more
light showers during the afternoon and evening. Wind still the same;
heavy clouds from the north-west.

Friday, 25th May, Range of Low Hills. I feel better this morning. The
clouds have all gone during the night, and it is now quite clear. Started
for Mount Strzelecki, passing through some very thick mulga scrub, with a
few gum-trees and plenty of grass. At twenty-one miles came upon a small
gum creek, where we gave the horses water, filled our own canteens, and
proceeded to the foot of the mount and camped. At a mile from its base
the spinifex begins again. Wind south-east. Very cold.

Saturday, 26th May, Mount Strzelecki. Ascended the mount, and built a
cone of stones. To the east are hills connected with this range, which I
have named Crawford Range, after ---- Crawford, Esquire, of Adelaide. To
the east-north-east is a large wooded undulating plain, with another
range in the extreme distance. To the north-east the distant range
continues with the same plain between. At a bearing of 55 degrees is a
large lagoon, in which there appears to be a little water. To the
north-north-east the plain appears to be rather more scrubby, and with a
few sand hills. To the north the point of the distant range is lost sight
of by some high scrubby land. To the west there are a few low hills, from
fifteen to twenty-five miles distant. This range is composed of a hard
flinty quartz, partly of a blue colour, with a little ironstone. We can
find no permanent water in this range, but, from the two or three native
tracks, quite fresh, which we have passed, I think there must be some
about. Descended, and proceeded round the range to the lagoon, the range
being too rough to cross. There is not enough water to be a drink for the
horses. Camped. Very heavy clouds from the north-west. The mount is about
four miles distant. At sundown there was a beautiful rain for an hour. It
is very strange, the clouds come from the north-west, and the wind from
the south-east. The rain seems to be coming against the wind.

Sunday, 27th May, Lagoon North-east of Mount Strzelecki. We had a few
heavy showers during the night, but it seems as if the rain would now
clear off. I hope not, for there is only about two inches of water in the
lagoon. I am again suffering much pain from the exertion it cost me to
climb Mount Strzelecki, and from assisting in building the cone of
stones; but if I did not put my hands to almost everything that is
required, I should never get on. My party is too small. It is killing
work.

Monday, 28th May, Lagoon North-east of Mount Strzelecki. We could not get
a start till 9.15, the horses having strayed to a distant bank for
shelter from the wind, which was piercingly cold. I had, in the first
instance, to go three miles north-north-west, in order to clear the low
stony range that runs on to the east side of the lagoon. I then changed
to 22 degrees to the far-distant range. For the first three miles our
course was through a very thick mulga scrub, with plenty of grass, and
occasionally a little spinifex; it then changed to a slightly undulating
country of a reddish soil, with gum and cork-trees, and numerous low
sandy plains, much resembling the gum and spinifex plains to the west,
where I was twice beaten back. It certainly is a desert country. Camped
without water on a little patch of grass. Distance to-day, twenty-eight
miles. Wind south-east. Very cold all day.

Tuesday, 29th May, Scrub, Spinifex and Gum-Trees. Started at 8 on the
same course for the range, which is still distant, through the same
description of country. At seven miles we came upon a plain of long
grass, which seems to have been flooded. It is about two miles broad.
Between this and the first hill of the range we passed four more of the
same description. Distance to the first hill, fourteen miles. In another
mile we struck a small creek; searched for water, but could find none,
although birds were numerous; thence through another mulga scrub, and
after crossing a number of rough stony hills, we arrived at the top of
the range, which I have named Davenport Range, after the Honourable
Samuel Davenport, M.L.C. It is composed of hard red sandstone, with
courses of quartz. I find this is not the range for which I am bound.
Although this one is high, the other is still higher, and, I should
think, is still forty or fifty miles distant. The day is thick, and I
cannot see distinctly. Between these ranges is a large plain, more open
than those we have come over. To the north the range appears to
terminate; to the west of north, in the far distance, just visible, are
two high hills, the northernmost of which is conical. To the east and
south-east is the plain and range; to the west, continuation of the same
plain that we have come over in the last two days' journey. Although we
had some heavy showers at the lagoon, we have not passed a single
water-course, except the one we crossed a few miles before we made this
range, nor did we see a drop of surface water: it seems to be all
absorbed the moment that it falls. Descended the north-north-east side of
the range, and at a mile and a half found some rain water in a creek,
coming from the range. Camped. Wind south-east. Distance, twenty miles.

Wednesday, 30th May, The Davenport Range. I find this water will not last
more than three days. I have determined to remain here to-day, and have
sent Kekwick in search of more water. As I am now a little better, I must
get my plan brought up. It has got in arrear, in consequence of my hands
being so bad with the scurvy. My limbs are much easier, yet the riding is
still very painful; my mouth also is much better, so that I am led to
hope that the disease will soon leave me. Native tracks about here, and
when I was on the top of the range I saw smoke in the scrub a few miles
to the north-west. Sundown: I am quite surprised that Kekwick has not
returned, as my instructions to him were not to go above five or six
miles, and then to return whether he found water or not. I am very much
afraid that something has happened to him.

Thursday, 31st May, The Davenport Range. Kekwick has not returned. I
begin to feel very uneasy about him. I must be off and follow up his
tracks. Sent Ben for the horses. He was a long time in finding them, as
is generally the case when one wants a thing in a hurry. 9.30: Kekwick
has arrived before the horses; he overshot his mark last night, and got
beyond the camp. I am very glad he is all safe. He informs me that he
came upon plenty of water a few miles from here, which compensates for
the anxiety he caused me during the night. His reason for not returning
as I had directed was that he crossed a gum creek which had so promising
an appearance, that he was induced to follow it to the plains, where he
found an abundance of water. While he was riding he was taken very ill,
and was unable to come on for some time, which made it so late that he
could not see to reach the camp. He is unable to proceed to-day, which is
vexing, for I wish to get on as quickly as possible.

Friday, 1st June, The Davenport Range. The horses having strayed, we did
not get a start till late. Our course was 22 degrees, and at two miles we
struck a small gum creek coming from the range and running
west-north-west. At three miles and a half we crossed a larger one coming
from, and running in, the same direction. Then commenced again the same
sort of country that we passed through the other day. At eight miles
struck a splendid large gum creek or river, having long and deep reaches
of water with fish four or five inches in length; it is running through
the plain as far as I can see, which is only a short distance, the ground
being low and level. Its course at this place is to the west-north-west;
it is very broad, and in some places the banks are perpendicular, and are
well grassed and covered with fine gum-trees, mulga and other bushes.
From bank to bank its width is about ten chains. This is the finest creek
for water that we have passed since leaving Chambers Creek. The day being
far advanced, I shall camp here, and get to the range to-morrow. I am
very much inclined to follow this creek and see where it empties itself;
but I expect to find a large one close to the range, or on the other
side. I wish also to get on the top to see what the country on ahead is
like. The fact of fish being in this creek leads me to think that it does
not empty itself into the gum plains, like others lately passed, but that
it must flow either into the sea on the north-west coast, or into a lake.
I have named it the Bonney Creek, after Charles Bonney, Esquire, late
Commissioner of Crown Lands for South Australia.

Saturday, 2nd June, The Bonney Creek. Started at 8.20 on the same course,
22 degrees, for the range, through a country of alternate spinifex and
grass with a little mulga scrub. At seven miles we struck another large
gum creek with every appearance of water, but I had no time to look for
it, being anxious to make the range to-night, and endeavour to find water
either on this side or on the other. The creek is large, and resembles
the last. I have named it the McLaren, after John McLaren, Esquire, late
Deputy Surveyor-General of South Australia. At seventeen miles, after
passing through a well-grassed country with a little scrub, we reached
the top of the first range, which is composed of a hard white
granite-looking rock, with courses of quartz running through it. I have
three or four spurs to cross yet before I make the main range. So far as
I can see, McLaren Creek is running much in the same direction as the
Bonney. Started from the top of the range and had a very difficult job in
crossing the spurs. About sundown arrived all safe on a gum flat, between
the ranges, and attempted to get upon what appears to be the highest
range, but getting up the horses deterred us. We then sought for water
among the numerous gum creeks which cover the plain, and at dark found
some, and camped. There is a good supply of water, but I do not think it
is permanent; it will last, however, for a month or six weeks. I have
named these ranges the Murchison, after Sir Roderick Murchison, President
of the Royal Geographical Society, London. Wind varying.

Sunday, 3rd June, Murchison Ranges. I feel very unwell this morning, from
the rough ride yesterday. It was my intention to have walked to the top
of the range to-day, but I am not able to do so. The small plain between
the ranges is a bed of soft white sandstone, through which the different
creeks have cut deep courses; the stones on the surface (igneous
principally), are composed of iron, quartz, dark black and blue stone,
also a bright red one, all run together and twisted into every sort of
nick, as also with the limestone, and many other sorts which I do not
know. This plain is covered with a most hard spinifex, very difficult to
get the horses to face. In another creek, about one mile south-west from
the camp, is a large water hole which will last six months; it is ten
yards long by twenty yards wide.

Monday, 4th June, Murchison Ranges. Started on a course of 330 degrees to
round this spur of the ranges, and at four miles and a half changed to 15
degrees to the high point of the range, and at three miles arrived on the
top. I have named it Mount Figg. The view from this is extensive. The
course of this range from the south to this point is 25 degrees; it then
makes a turn to the north-north-west, in which direction the country
appears more open, with some patches of thick scrub, and high ranges in
the distance. From north-west to west it appears to be gum plain, with
open patches of grass, and a number of creeks running into it from the
range. I shall change my course to a high peak on the north-west point of
the range, which bears from this 340 degrees 30 minutes. This range is
volcanic here, and is of the same formation as I have already given.
Started from the top of the mount at 12 o'clock. Went for eight miles
along the side of the range, and met with a small gum creek running on
our course; followed it up for three miles without finding water; it then
took a more westerly course, so I left it to pursue my route. After
leaving the mount, the range is composed of red sandstone with a little
quartz. We have occasionally met with a little limestone gravel. Camped
at 6 o'clock, without water.

Tuesday, 5th June, Gum-Tree Plain. Started on the same course at 7
o'clock for the high peak, through the same sort of country as yesterday.
No watercourse. At fifteen miles ascended the peak, which I have named
Mount Samuel, after my brother. The top is a mass of nearly pure
ironstone. It attracted the compass 160 degrees. From north to west are
broken ranges and isolated hills of a volcanic character, in all sorts of
shapes. The isolated hills seem to be the termination of these ranges,
which run nearly north and south. I have named them the McDouall Ranges,
after Colonel McDouall, of the 2nd Life Guards, Logan, Wigtownshire. I
then changed my course to the north-north-east in search of water, there
being no appearance of any to the north-north-west. After travelling five
miles over small grassy, scrubby plains, between isolated hills and
gum-trees, I could not find a water-course, so I changed to the east, to
try if I could see anything from a high hill, which I ascended, and
discovered a gum creek coming from the range on the east side. Followed
it down, and, one mile and a half from the top, found a splendid hole of
water in the rock, very deep, and permanent. The creek is very rocky, and
its course here is north-east into the plain. Wind south-east. Clouds
from the north-west.

Wednesday, 6th June, Gum Creek, North-east Side of the McDouall Ranges.
There being nothing but spinifex on the ranges and creeks, the horses had
been travelling nearly all night in search of food, and had gone a long
way before they were overtaken. This morning saddled and got a start by
11 o'clock on a course of 340 degrees, crossing numerous creeks and stout
spinifex, through which we had great difficulty in driving the horses. At
five miles struck a gum creek in which we found water. The banks have
excellent feed upon them, and in abundance, so, for the sake of the
horses, I have determined to remain here to-day. This creek, which I have
named Tennant Creek, after John Tennant, Esquire, of Port Lincoln, runs
east. In searching for the horses this morning Ben found three or four
more large water holes in the adjoining creek, a little south-east from
this. Before we reached this, we crossed some marks very much resembling
old horse-tracks.

Thursday, 7th June, Tennant Creek, McDouall Ranges. Started at 7.20.
Course, 340 degrees. At three miles passed through an immense number of
huge granite rocks piled together and scattered about in every direction,
with a few small water-courses running amongst them to the eastward. We
then encountered a rather thick scrub, and occasionally crossed a few low
quartz rises coming from the McDouall ranges. At fourteen miles ascended
the highest of them, which I have named Mount Woodcock, after the
Venerable the Archdeacon of Adelaide. To the north-west and north is
another range, about ten miles distant, which seems to continue a long
way. I will change my course to 315 degrees, which will take me to the
highest point. At two miles on this course came upon a gum creek running
to the north-east, which I named Bishop Creek; followed it for one mile
and a half, and found water, which will last a month or six weeks, and an
immense number of birds. This is a camping-place of the natives, who seem
to have been here very lately. We watered the horses and proceeded
towards the range. At about two miles passed a low rugged ironstone
range, peculiar in having a large square mass of ironstone standing by
itself about the centre. I have named it Mount Sinclair, after James
Sinclair, of Port Lincoln. Passed through a thick scrub, among which we
saw a very handsome bush that was new to us, having a blue-green leaf ten
inches long by six inches broad. We looked for some seed, but could not
find any. At five miles crossed a grassy gum plain, where a creek empties
itself. The same scrub continues to the range, which we reached at twelve
miles from the water. It is not very high, but rough and steep, and we
had great difficulty in getting to the top, but after many twistings and
turnings and scramblings, we arrived there all right, and found it to be
table land. At fourteen miles camped without water. The range is composed
of ironstone, granite, quartz and red sandstone, running north of west
and south of east. I have named it Short Range, after the Right Reverend
the Lord Bishop of Adelaide.

Friday, 8th June, Short Range. Started at 8 o'clock on the same course,
315 degrees, to some very distant rising grounds. Short range seems to
run nearly parallel to our course, as also does another distant range to
the north, which I have named Sturt Range, after Captain Sturt. The table
land continued about two miles, and then there was a gradual descent to
the plains, and we entered a thick scrub with spinifex and gums. At
eighteen miles came upon a beautiful plain of grass, having large
gum-trees, and a new description of tree, the foliage of which is a
dark-green and rather round, and the bark rough and of a dark colour.
Here also was the cork-tree, and numerous other shrubs. This grassy plain
continued for thirty-one miles, until we camped, but the last part is not
so good. When I struck this plain, I was in great hopes of finding a
large creek of water, but have been disappointed; we have not crossed a
single water-course in thirty-one miles. Camped at sundown. No water.
Wind south-east.

Saturday, 9th June, Grassy Plain. There is some rising ground a few miles
further on, to which I shall go in search of a creek; I might be able to
see something from it. If I do not find water I shall have to retreat to
Bishop Creek, as the horses have now been two nights without water.
Started at 7 o'clock, same course, 315 degrees, through scrub and a light
sandy soil. At four miles got to the rise, which is a scrubby sand-hill.
From this I can see nothing, the scrub being so thick; it is of a nasty,
tough, wiry description, and has torn our hands and saddle-bags to
pieces. I got up a tree to look over the top of this scrub, which is
about twelve feet high, and I could see our course for a long distance;
it appears to be the same terrible scrub, with no sign of any creeks. It
is very vexing to get thus far, and have to turn back, when perhaps
another day's journey would bring me to a better country. I shall now try
a south course, and cut the grassy plains to the westward, in the hope of
finding water; if so, I shall be able to make two days' journey to the
north-west. Started on a south course for fourteen miles, through scrub
and small grassy plains alternately, but we could find neither creek nor
water. I now regret that I attempted the south course, which makes the
distance from the water so much greater. Wind still south-east; heavy
clouds coming from the north-west, I trust it will rain before morning.

Sunday, 10th June, Grassy Plains. Started at sunrise, and at two miles
again got into the scrub. Three of the horses we can scarcely get along;
they are very much done up. At 11 o'clock, one horse gave in altogether.
We cannot get him up; we have tried everything in our power to do
something for him. The other horses have been carrying his load, and he
has had nothing to carry for this last hour and a half; all our efforts
are in vain, and I am obliged, although with great reluctance, to leave
him to his fate. Had this occurred nearer the water, I should have put an
end to his existence and taken part of him to eat, for we are now very
short of provisions, and the other horses have quite enough to carry
without sharing his load; I wish I had left him sooner. At 12 o'clock, I
find I shall lose some more of them, if they do not get water to-night,
and it will be to-morrow before I can reach Bishop Creek. I shall now go
to Short range and try to find some. The little bay mare Polly has become
nearly mad, running about among the other horses, and kicking them as she
passes; even the men do not escape from her heels. At five miles made the
range. There are no large creeks coming from this side--nothing but small
ones which empty themselves into the plain; sand up to the foot of the
hills. Before we reach the range another of the horses is done up; he has
only been carrying about 30 pounds in consequence of his back having been
bad for the last three weeks. We lightened all the weak horses two days
since. We shall now try if he will go without anything on his back. We
are now amongst the granite ridges, and hope we shall find water on this
side. The horse has given in before we can get to the other side. We must
leave him for the sake of the others. Too much time has already been lost
in endeavouring to get them on. Reached the other side and searched the
different creeks, but cannot find any water. Crossed a spur of the range
running south, and can see a nice-looking creek with gum-trees. Our hopes
and spirits are again revived; the sight of it has even invigorated the
horses, and they are hurrying on towards it. Traversed it down, but, to
our great disappointment, find that it loses itself in a grassy plain. It
is now dark, so I must remain here for the night. The sky is quite
overcast, and I trust that Providence will send us rain before morning.
An accident has happened to the water we were carrying; it was all lost
yesterday. If it clears during the night, so that I can see the stars to
guide me, I shall move on.

Monday, 11th June, Short Range. During the night there were a few drops
of rain, which again raised our hopes, and about 4 o'clock it looked as
if we were to have a deluge, but, alas! it only rained for about two
minutes, and as much fell as would wet a pocket-handkerchief. Saddled and
started through the range, my poor little mare looking very bad this
morning; I have taken everything off her, so that she may hold out until
we get to water, and I have been obliged to leave as many things at this
camp as I could possibly do without. The mare lies down every few yards,
I am therefore compelled to leave her for the sake of the others. From
the number of birds about here, I think there must be water near; I hope
she may find it, although I am afraid she is too far gone even to try it.
At 1 o'clock, at the foot of Mount Woodcock, the horses' spirits revived
at sight of their old track. I shall now be able to get all the rest of
them safe to water, although there is one still doubtful. My own black
mare shows a few symptoms of madness, but still keeps on, and does her
work well. About an hour before sundown arrived at the water without any
more losses, for which I sincerely thank the Almighty. We have had a
terrible job to keep the horses from drinking too much water, but, as
they have now eaten a few mouthfuls of grass, I have allowed them to
drink as much as they thought proper. The natives have been here since we
left.

Tuesday, 12th June, Bishop Creek. Resting: the horses look very bad; they
remained by the water all night.

Wednesday, 13th June, Bishop Creek. The horses still look very bad this
morning; they have again stayed by the water nearly all night; they had
been one hundred and one hours without a drop, and have accomplished a
journey of one hundred and twelve miles; they will require a week to
recover; one of them is very lame from a kick the little mare gave him in
her madness. Thus ends my last attempt, at present, to make the Victoria
River; three times have I tried it, and have been forced to retreat.
About 11 o'clock I heard the voice of a native; looked round and could
see two in the scrub, about a quarter of a mile distant. I beckoned to
them to approach, but they kept making signs which I could not
understand. I then moved towards them, but the moment they saw me move,
they ran off immediately. About a quarter of an hour afterwards they
again made their appearance on the top of the quartz reef, opposite our
camp, and two others showed themselves in about the same place as the two
first did. Thinking this was the only water, I made signs to the two on
the reef to go to the water; but they still continued talking and making
signs which I could not understand; it seemed as if they wished us to go
away, which I was determined not to do. They then made a number of
furious frantic gestures, shaking their spears, and twirling them round
their heads, etc. etc., I suppose bidding us defiance. I should think the
youngest was about twenty-five years of age. He placed a very long spear
into the instrument they throw them with, and, after a few more gestures,
descended from the reef, and gradually came a little nearer. I made signs
of encouragement for him to come on, at the same time moving towards him.
At last we arrived on the banks of the creek, he on one side, and I on
the other. He had a long spear, a womera, and two instruments like the
boomerang, but more the shape of a scimitar, with a very sharp edge,
having a thick place at the end, roughly carved, for the hand. The
gestures he was making were now signs of hostility, and he came fully
prepared for war. I then broke a branch of green leaves from a bush, and
held it up towards him, inviting him to come across to me. As he did not
seem to fancy that, I crossed to where he was, and got within two yards
of him. He thought I was quite near enough, and would not have me any
nearer, for he kept moving back as I approached. I wished to get close up
to him, but he would not have it; we then stood still, and I tried to
make him understand, by signs, that all we wanted was water for two or
three days. At last he seemed to understand, nodded his head, pointed to
the water, then to our camp, and held up his five fingers. I then
endeavoured to learn from him if there was water to the north or
north-east, but I could make nothing of him. He viewed me very steadily
for a long time, began talking, and seeing that I did not understand him,
he made the sign that natives generally do of wanting something to eat,
and pointed towards me. Whether he meant to ask if I was hungry, or to
suggest that I should make a very good supper for him, I do not know, but
I bowed my head as if I understood him perfectly. We then separated, I
keeping a careful watch upon him all the time I was crossing the creek.
Before I left him the other one joined. The first was a tall, powerful,
well-made fellow, upwards of six feet; his hair was very long, and he had
a red-coloured net tied round his head, with the ends of his hair lying
on his shoulders. I observed nothing else that was peculiar about them.
They had neither skins nor anything round their bodies, but were quite
naked. They then took their departure. A short time afterwards I saw them
joined by five others. We have seen nothing more of them to-day, and I
hope they will not trouble us any more, but let me get my horses rested
in peace. Wind south, all the clouds gone; nights and mornings very cold.
Occupied during the day in shoeing horses, and repairing and making
saddle-bags.

Thursday, 14th June, Bishop Creek. On examining the water holes, I find
there are small crab fish in them, which leads me to think this water is
permanent. This morning we again hear the voices of the natives up the
creek to the west. There must be plenty more water up there, as most of
the birds go in that direction to drink, passing by this water. The
natives have not come near us to-day, but we have seen the smoke of their
fires. Shoeing horses, repairing and making saddle-bags, which were torn
all to pieces by the scrub.

Friday, 15th June, Bishop Creek. Resting horses, and getting our
equipment in order for another trial, as I think the horses will be ready
to start on Monday morning. No more of the natives but their smoke is
still visible. Wind south; day hot, night cool.

Saturday, 16th June, Bishop Creek. The horses are still drinking an
immense quantity of water; they are at it five and six times a day; they
must have suffered dreadfully. The grass here is as dry as if it were the
middle of summer, instead of winter. I hope we may soon have rain, which
would be a great blessing to me.

Sunday, 17th June, Bishop Creek. The horses still pay frequent visits to
the water. We have found more about a mile up the creek, and there seems
to be plenty further up in the hills; I cannot examine it just now, in
consequence of the natives being about. It would not do for me to leave,
as the party is so small, nor do I like sending one of them, for he might
be taken by surprise and cut off, which would ruin me altogether, being
able to do scarcely anything myself. Although I am much better, I am
still very weak; the pains in my limbs are not so constant. I attribute
the relief to eating a number of native cucumbers which are in quantities
on this creek. The horse that was kicked by the mare is still very lame.
Wind south-east.

Monday, 18th June, Bishop Creek. Started at 9.30 on a bearing of 18
degrees, through a plain of alternate grass, scrub, and spinifex, and at
five miles passed a number of isolated hills close together, composed of
large masses of ironstone, quartz, and a hard brown rock, very irregular,
and all sorts of shapes; the stones seem as if they had undergone the
action of fire. We then proceeded through some very bad spinifex,
dark-coloured, long, hard and dry; we could scarcely get the horses to
face it. We then came upon a grassy plain, and at ten miles struck a gum
creek coming from the west of north-west, and running (at this place)
east-north-east; followed it and found an abundance of water in long deep
holes, with shells of the crab fish lying on the banks. The water is
upwards of a mile in length; the creek then spreads out over a grassy
plain with scrub and gum-trees, and is joined by the other creeks coming
from the McDouall range. I thought it advisable to camp here for the rest
of the day, as a further journey would be a risk for the horse that is
lame, and I do not wish to lose any more; as it is, I am afraid he will
not be able to cross Short range, which I hope to do in a few hours.
Natives about. Splendid grass on this plain, and on the banks of the
creek, which I have named Phillips Creek, after John Phillips, Esquire,
J.P., of Kanyaka. Wind variable.

Tuesday, 19th June, Phillips Creek. Started at 8 o'clock on the same
bearing, 18 degrees. We first passed through a well-grassed plain with a
little scrub, then again through hard spinifex to the range. At one mile
crossed another gum creek with water in it, coming from Short range. At
four miles reached the top of the spur of the range; and at seven miles,
the top of the range. About two miles to the east, the range seems to
terminate in a gum plain, a spur from the McDouall range running on the
other side of the plain, and crossing our line a few miles further on.
Short range here is composed of quartz, ironstone, and red granite, with
a little limestone. Descended into the plain, and at ten miles came upon
another gum creek, spreading over a grassy plain, but could find no
water. At thirteen miles came upon some dry swamps with a number of birds
about them. At fourteen miles reached the top of the next range. From
this the appearance of the country, on this course, is evidently very
scrubby. On a bearing of 55 degrees, in the far distance, is the
termination of another range. I do not like facing the scrub again so
soon after my late loss, and with my horses not yet recovered. I shall
return to the swamps and look for water. If I find any, I shall start in
the morning for the end of the distant range. My lame horse is unable to
do more to-day; crossing the range has been very hard upon him. Returned
to the swamps and found a fine pond of water. Camped. The water is
derived from the creek that we passed in the middle of the day. I have
named these ponds after Kekwick, in token of the zeal and activity he has
displayed during the expedition.

Wednesday, 20th June, Kekwick Ponds. Saddled at sunrise, and proceeded to
the top of the low range, from which I turned back yesterday, and changed
my course to 56 degrees to the northernmost point of the distant hills,
through a plain of alternate grass and spinifex. At 3 o'clock struck the
William Creek again, with splendid grass on its banks. It ran nearly our
course for about three miles, and then turned to the east. We then
entered the same sort of scrub as that in which I lost my horses; this
continued until we reached the hills, which we did in about eighteen
miles. From this we can see a range to the south-south-east. About ten
miles off there is a large lake, with red sand hills on the east side. I
cannot see the extent of it, the hills that I am now on being so low;
they are composed of granite, and run north and south. To the north and
north-east is another lake, about the same distance, to which I shall go
on a course of 32 degrees 30 minutes. On the north side of this one there
are also sand hills with scrub. For two miles after leaving the hills we
passed through a soft, sandy, scrubby country and spinifex. It then
became harder, with grass and spinifex alternately. At four miles from
the hills we camped without water. My horses have not recovered from
their last trial, and seem to be very tired to-night, although to-day's
journey was not a long one, but it has been very hot, and the scrub thick
and difficult to get through.

Thursday, 21st June, Scrub. The horses having gone back on the track, we
did not get a start until 8.30--course, 32 degrees 30 minutes to a high
hill on the other side of the lake, passing through a thick scrub of
cork-tree and gums, with spinifex and grass. At seven miles came upon
what I thought was the lake, but it turns out to be a large plain of rich
alluvial soil covered with dry grass, which gave it the appearance of a
lake. It was three miles across to the top of the hill; no water-course
through, nor any water to be seen. The hills on the north side are
composed of ironstone and granite, and, from the distance, looked very
much like sand hills. From the top of the hill I can see the plain
extending a little to the west of north, but I cannot see far for the
mirage. To the north-north-east is another plain of the same description,
but much smaller, about a mile and a half broad, and nearly circular. To
the north-east is another very extensive one; its dimensions I cannot
see. I seem to have got into the land of grassy plains and low stony
hills. I wish my horses had had water last night or yesterday. They seem
to be very much in want of it. I must devote the rest of this day to a
search for it. I shall now direct my course for the south part of the
plain that I have just crossed; it seems to be the lowest part, and the
flight of the birds is directed that way. Searched all round, but can
find no water; so I must return to Kekwick Ponds. The day is extremely
hot, and my horses cannot stand two more nights without water. Would that
they had more endurance! It is dreadful to have to turn back almost at
the threshold of success. I cannot be far from the dip of the country to
the Gulf. Returned by another course to where I camped last night, but
still no water. I would fain try the plain to the south, but I dare not
risk the loss of more horses. Proceeded to the low range that I crossed
yesterday; examined round it, but cannot find any water. Camped. Two of
the horses very much done up. I must go back through that nasty scrub
again.

Friday, 22nd June, Under the West Low Range. Started at sunrise for the
ponds, and at 1.30 arrived; the horses being very much exhausted. I am
glad I did not remain another night without water; three of them are
completely done up, and it has been with difficulty that we have got them
here. Wind south-west.

Saturday, 23rd June, Kekwick Ponds. Resting horses. About 1 o'clock we
were visited by two natives, who presented us with four opossums and a
number of small birds and parrots. They were much frightened at first,
but after a short time became very bold, and, coming to our camp, wanted
to steal everything they could lay their fingers on. I caught one
concealing the rasp that is used in shoeing the horses under the netting
he had round his waist, and was obliged to take it from him by force. The
canteens they seemed determined to have, and it was with difficulty we
could get them from them. They wished to pry into everything, until I
lost all patience and ordered them off. In about half an hour two other
young men approached the camp. Thinking they might be in want of water,
and afraid to come to it on account of the horses, I sent Ben with a tin
dishful, which they drank. They were very young men, and too much
frightened to come any nearer. About an hour before sundown, one of the
first that had come, returned, bringing with him three others, two of
whom were young, tall, powerful, well made, and good-looking, and as fine
specimens of the native as I have yet seen. On their heads they had a
neatly-fitting hat or helmet close to the brow, and rising straight up to
a rounded peak, three or four inches above the head and gradually
becoming narrower towards the back part. The outside was net-work; the
inside was composed of feathers very tightly bound together with cord
until it was as hard as a piece of wood; it may be used as a protection
from the sun, or as armour for the battle-field. One of them had a great
many scars upon him, and seemed to be a leading man. Only two had helmets
on, the others had pieces of netting bound round their foreheads. One was
an old man, and seemed to be the father of these two fine young men. He
was very talkative, but I could make nothing of him. I have endeavoured,
by signs, to get information from him as to where the next water is, but
we cannot understand each other. After some time, and having conferred
with his two sons, he turned round, and surprised me by giving me one of
the Masonic signs. I looked at him steadily; he repeated it, and so did
his two sons. I then returned it, which seemed to please them much, the
old man patting me on the shoulder and stroking down my beard. They then
took their departure, making friendly signs until they were out of sight.
We enjoyed a good supper from the opossums, which we have not had for
many a day. The men are complaining of weakness from the want of
sufficient nourishment. I find the quantity of rations is not enough;
five pounds of flour per week is too little for many weeks together. It
may do very well for a month or so, but when it comes to the length of
time we have been out, we all feel it very much; and the dried meat that
I brought with me being very young, it has not half the strength in it
that old meat has.

Sunday, 24th June, Kekwick Ponds. Our black friends have not made their
appearance to-day.

Monday, 25th June, Kekwick Ponds. Started again on a bearing of 345
degrees to some very distant hills, to see if I can get into the face of
the country to the Gulf of Carpentaria. At two miles crossed a large gum
creek (with long beds of concrete ironstone), which I have named Hayward
Creek, after Frederick Hayward, Esquire. The banks are beautifully
grassed, and extend for four miles on the north side. At fourteen miles
struck a gum creek with large sheets of water in which were plenty of
ducks, native companions, black shags, cranes, and other birds. Camped
here for the remainder of the day. The course of the creek at this point
is to the north of east, and coming from the north of west, apparently
from the range, which is distant about ten miles. It very much resembles
Chambers Creek. The ponds (in which we found some small fish) are about
eighty yards broad, and about three quarters of a mile long, having large
masses of concrete ironstone at both ends, separating the one pond from
the other; large gum-trees being in the ponds. Wind north-west. Very hot.

Tuesday, 26th June, Large Gum Creek, with Sheets of Water. I have
resolved to follow this creek down to-day, and, if the water continues,
to follow it out. Started on a course 77 degrees, and at six miles
crossed the creek, which is running a little more to the north. There are
long sheets of water all the way down to this, the banks in some places
being steep, with the lower part formed of concrete, and the upper red
sandy soil, which gives me a bad opinion of it for water, if the concrete
ceases. Here we saw some blacks; they would not come near us, but walked
off as fast as they could. From the top of the rise we saw where they
were camped, on the banks of a large sheet of water; we passed on without
taking any more notice of them, and at nine miles, not seeing any
appearance of the creek, I changed my course to 25 degrees. At three
quarters of a mile cut it again, but without water in it; it is much
narrower and deeper, having sandy banks and bed. Changed again to 77
degrees, the creek frequently crossing our course, and at fifteen miles
saw there was no hope of obtaining water. The country is becoming more
sandy, and is thickly covered with spinifex and scrub. We crossed down to
the banks of the creek; no rising ground visible. I must keep closer to
the hills, and, as the day has been very hot, I shall return and camp at
nine miles from our last camp, if there is water; if not, I shall have to
camp a short way above where we saw the natives this morning. I do not
wish to get too near them, or to annoy them in any way. We could find no
water below where they were camped; I therefore pushed on to get above
them before dark. At half-past one o'clock, about three miles from the
creek, we saw where they had been examining our tracks, and as we
approached the creek their tracks became very numerous on ours. When we
arrived on the top of the rise, where we had previously seen their camp
and fires, we could now see nothing of them, neither smoke, fires, nor
anything else: it was then nearly dark. I concluded they had left in
consequence of having seen us pass in the morning, as natives in general
do. I was moving on to the place where we crossed the creek in the
morning, when suddenly from behind some scrub which we had just entered,
up started three tall powerful fellows fully armed, having a number of
boomerangs, waddies, and spears. Their distance from us was about two
hundred yards. It being so nearly dark, and the scrub we were then in
placing us at a disadvantage, I wished to pass without taking any notice
of them, but such was not their intention, for they continued to approach
us, calling out and making all sorts of gestures apparently of defiance.
I then faced them, making every sign of friendship I could think of. They
seemed to be in a great fury, moving their boomerangs above their head,
bawling at the top of their voices, and performing some sort of a dance.
They were now joined by more of their tribe, so that in a few minutes
their numbers had increased to upwards of thirty; every bush seemed to
produce a man. Putting the horses on towards the creek, and placing
ourselves between them and the natives, I told my men to get their guns
ready, for I could see they were determined upon mischief. They paid no
regard to all the signs of friendship I kept constantly making, but were
still gradually approaching nearer and nearer to us. I felt very
unwilling to fire upon them, and still continued making signs of peace
and friendship, but all to no purpose. Their leader, an old man, who was
in advance, made signs with his boomerang, which we took as a signal for
us to be off. They were, however, intended as tokens of defiance, for I
had no sooner turned my horse's head to comply with what I thought were
their wishes, than we received a shower of boomerangs, accompanied by a
fearful yell; they then set fire to the grass, and commenced jumping,
dancing, yelling, and throwing their arms into all sorts of postures,
like so many fiends. In addition to the thirty that already confronted
us, I could now see many others getting up from behind the bushes. Still
I felt unwilling to fire upon them, and tried again to make them
understand that we wished to do them no harm. Having now approached
within about forty yards of us, they made another charge, and threw their
boomerangs, which came whistling and whizzing past our ears, one of them
striking my horse. I then gave orders to fire, which stayed their mad
career for a little. Our pack-horses, which were on before us, took
fright when they heard the firing and fearful yelling, and made off for
the creek. Seeing some of the blacks running from bush to bush, with the
intention of cutting us off from our horses, while those in front were
still yelling, throwing their boomerangs, and coming nearer to us, we
gave them another reception, and I sent Ben after the horses to drive
them on to a more favourable place, while Kekwick and I remained to cover
our rear. We soon got in advance of those who were endeavouring to cut us
off, but they still kept following, though beyond the reach of our guns,
the fearful yelling still continuing from more numerous voices, and fires
springing up in every direction. It being now quite dark, with the
country scrubby, and our enemies bold and daring, we could be easily
surrounded and destroyed by such determined fellows as they have shown
themselves to be. Seeing there is no hope with such fearful odds (ten to
one at least) against us, and knowing all the disadvantages under which
we labour, I very unwillingly make up my mind to push on to our last
night's camp. We have done so, and now I have had a little time to
consider the matter over I do not think it prudent to remain here
to-night; I shall therefore continue on until I reach the open grassy
plain or gum creek. They are still following us up; I only wish that I
had four more men, for my party is so small that we can only fall back
and act on the defensive. If I were to stand and fight them (which I wish
I could) our horses must remain unprotected, and we, in all probability,
should be cut off from them. Our enemies seem to be aiming at that, and
to prevent our advance up the creek; by this time they have found out
their mistake, as we did not go a step out of our course for them.
Arrived at Hayward Creek at 11 o'clock at night.

Wednesday, 27th June, Hayward Creek. This morning we see signal fires all
around us. It was my intention last night to have gone this morning to
Kekwick Ponds to water the horses, then to give them the day to rest, and
proceed to-morrow back again to the large creek, and go on to the distant
hills that I was steering for on the 25th instant, but, after considering
the matter over the whole night, I have most reluctantly come to the
determination to abandon the attempt to make the Gulf of Carpentaria.
Situated as I now am, it would be most imprudent. In the first place my
party is far too small to cope with such wily, determined natives as
those we have just encountered. If they had been Europeans they could not
better have arranged and carried out their plan of attack. They had
evidently observed us passing in the morning, had examined our tracks to
see which way we had gone, and knew we could get no water down the creek,
but must retrace our steps to obtain it above them; they therefore lay in
wait for our return. Their charge was in double column, open order, and
we had to take steady aim, to make an impression. With such as these for
enemies in our rear, and, most probably, far worse in advance, it would
be destruction to all my party for me to attempt to go on. All the
information of the interior that I have already obtained would be lost.
Moreover, we have only half rations for six months, four of which are
gone, and I have been economizing as much as I possibly could in case of
our having to be out a longer time, so that my men now complain of great
weakness, and are unable to perform what they have to do. Again, only two
showers of rain have fallen since March, and I am afraid of the waters
drying up to the south, and there is no appearance of rain at present.
The days are now become very hot again, and the feed for the horses as
dry as if it were the middle of summer. The poor animals are very much
reduced in condition, so much so that I am afraid of their being longer
than one night without water. Finally, my health is so bad, that I am
hardly able to sit in the saddle. After taking all those things into
consideration, I think it would be madness and folly to attempt more. If
my own life were the only sacrifice, I would willingly risk it to
accomplish my purpose; but it seems that I am destined to be
disappointed; man proposes, but the Almighty disposes, and his will must
be obeyed. Seeing the signal fires around, and dreading lest our black
friends at Kekwick Ponds might have been playing a double part with us,
in spite of their Masonic signs, I gave them a wide berth, and steered
for Bishop Creek. Arrived there in the afternoon, and found that the
creek had not been visited by natives since we left. These natives do not
deposit their dead bodies in the ground, but place them in the trees,
and, judging from the number of these corpses which we have passed
between this and the large creek, where they made their attack upon us,
they must be very numerous. These natives have quite a different cast of
features from those in the south; they have neither the broad flat nose
and large mouth, nor the projecting eyebrows, but have more of the Malay;
they are tall, muscular, well-made men, and I think they must have seen
or encountered white men before.

Thursday, 28th June, Bishop Creek. Camped at the rocky water hole
north-east side of the McDouall range.

Friday, 29th June, Anderson Creek. Crossed the McDouall ranges and camped
on a gum creek on the north-east side of the Murchison ranges, which I
have named Gilbert Creek, after Thomas Gilbert, Esquire, late Colonial
Storekeeper.

Saturday, 30th June, Gilbert Creek. Crossed the Murchison ranges, and the
large gum creek coming from them, and running west-north-west, which I
have named Baker Creek, after the Honourable John Baker, M.L.C. I did not
examine it, but should think from its appearance that there is water in
it; besides, I can distinguish the smoke of a native encampment.
Proceeded to the creek where we camped before, but found all the water
gone, except a little moisture in the bottom of the holes. I was rather
surprised at this, for I thought it would have lasted three months at
least. Went to another creek, where there was a large hole of water in
conglomerate rock; this we found also to be very much reduced; when we
last saw it, its depth was four feet, and now it is only eighteen inches.
Camped.

Sunday, 1st July, Murchison Ranges. My horses very tired, and three of
them are nearly done up.

Monday, 2nd July, Murchison Ranges. Proceeded to the Bonney Creek to get
feed for the horses, there being very little besides spinifex under the
ranges. Smoke of native encampments on and about the creek; I must be
very careful.

Tuesday, 3rd July, The Bonney Creek. We have not seen any more of the
natives yet. I shall rest the horses to-day, there being plenty of feed,
which they very much want. Being so very few of us, I am obliged to turn
them out with the saddles on; so that, if we are attacked again, one can
put the packs on, while I and the other defend him. The water in this
hole is very much reduced, but I think it will not fail altogether, in
consequence of the small fish being in it. From the diminution of the
water in this creek since I left it, a month ago, I am inclined to think
that I shall have a very hard push to get back; my horses being so weak
from the hardships they have undergone, that they are now unable to do as
much as they did before. I fear that I shall not get any water between
this and Forster's range, a distance of upwards of eighty miles, so I
shall rest them here for a week, if the natives will be quiet; if not, I
must run the risk of losing more of them. To-day, I had made up my mind
to follow out this creek, to see if the waters continue, and if it would
take me to the north of the spinifex and gum-tree plain which I had to
turn back from on my north-west course from Mount Denison, and if rain
falls to try again for the Victoria River. I am, however, disappointed,
for, on weighing the rations, I find I am terribly short, which I did not
expect, and which cuts off all hope of my attaining that point. My
troubles and vexations seem to come upon me all at once. Had I but a
stronger party, and six months' rations, I think I should be able to
accomplish something before my return. I have done my best, and can do no
more. My eyesight is now so bad that I cannot depend upon my
observations, which will be a great loss to me; and the scurvy has
returned with greater severity. Before I start on my return, if
everything goes right, I shall run down this creek a short distance. It
may, at some future time, turn out to be the road to the Victoria River,
or one of its tributaries. Wind south and south-west.

Wednesday, 4th July, The Bonney Creek. The water in this hole has been
diminishing very rapidly since we were here; it is falling at the rate of
six inches per day, which is a poor look-out for us on our homeward
course. I have not a day to spare now, as the weather is becoming very
hot, and will dry it up much faster. I must push back as soon as my
horses are rested and able to undergo the eighty miles without water. I
must give up the examination of this creek, for every day now is of the
utmost importance, and I must not give the horses one mile more than I
can help. Oh! that rain would fall before I leave this. It would indeed
be an inestimable blessing. Wind from all points. At sundown a few clouds
have made their appearance.

Thursday, 5th July, The Bonney Creek. During the night it became very
cloudy from the west, and this morning still continues. My hopes are
again raised. If it should rain, I shall try for the Victoria River
again, even though I should be without rations for my return; I could
kill one of the horses and dry his flesh, and that would take me back.
Still very cloudy, and every sign of rain. I am making preparations for
another trial. At sundown there are still heavy black clouds coming from
the west, which have raised our hopes of success to the highest point,
and I ardently trust they will be realized. No natives have come near us,
yet they are still about.

Friday, 6th July, The Bonney Creek. A sad, sad disappointment; all our
most sanguine hopes are again gone, for, during the night, the clouds
broke up and have all vanished; it is very vexing. I shall rest the
horses till Monday, and then, ill and dispirited, commence my homeward
journey. I dare not venture into a new route, for, want of water, and the
low condition of my horses, compel me to keep my former track. Last night
about 10 o'clock, I observed the comet for the first time, above the west
horizon; it set at 7 o'clock 20 degrees north of west. At sundown it has
become overcast with heavy clouds, and my hopes are again raised; I trust
we may get it now. Midnight: still cloudy, and every appearance of rain.
Wind changeable.

Saturday, 7th July, The Bonney Creek. Alas! all the clouds are again
gone; our hopes were only raised to be dashed down with greater
disappointment. The wind has returned to its old quarter, south-east.
Natives still about, but they do not come near us. I shall now prepare
for my return on Monday morning; it is very disheartening.

Sunday, 8th July, The Bonney Creek. The weather has every appearance of
being dry for some time to come, not a cloud to be seen; the wind
south-east, and very cold night and morning. All hope of making the coast
is now gone. On weighing our rations to-day, I find that we are again
short since we halted here. The man Ben has been making it a regular
practice to steal them since he has been with me. I have caught him
several times doing so, and all the threats and warnings of the
consequences have had no effect upon him. They deter him for a day or
two, and then he is as bad as ever. I have been in the habit of reducing
our allowance to make up for the loss, which has been very hard upon
Kekwick and myself; he has helped himself to about double his allowance
during the journey.

Monday, 9th July, The Bonney Creek. Started for the Davenport range,
where we camped before; the water is all dried up. Ascended the range,
and changed my bearing to Mount Morphett, 196 degrees, in the Crawford
range, in the hope of finding water there. At four miles struck the creek
that I have before crossed nearer to the range, found water, and camped
to give my horses every chance. I have named this creek Barker Creek,
after Mr. Chambers' brother-in-law. I do not think this water is
permanent, but, from the number of birds that are passing up the creek, I
think there must be permanent water higher up. This range seems to yield
a deal of water on both sides. Native graves about.

Tuesday, 10th July, Barker Creek. Started at 6.30 on a bearing of 196
degrees towards Mount Strzelecki. At six miles crossed a gum creek,
coming from the range, and running to the west, on my former track. I
crossed it where it lost itself on the plain. The country is well
grassed, with a little spinifex occasionally, from the range to this
point. At twelve miles it became scrubby and sandy with a little grass,
spinifex predominating, which continued to where we camped. Wind,
south-east.

Wednesday, 11th July, Scrub North-north-east of Mount Strzelecki. One of
the horses having parted from the others, and gone a long distance off in
search of water, it was 9 o'clock before we could get a start. At seven
miles arrived at a lagoon north-east of Mount Strzelecki. Found a little
water and feed for the horses. Camped to give them the benefit of it.
Wind, south-south-east. Cold.

Thursday, 12th July, Lagoon North-east of Mount Strzelecki. Made an early
start, crossing the range, on a south course. Very rough and difficult.
Could see no water. To the south-east of Mount Morphett there is the
appearance of a creek, and on the south-west there are also the signs of
a watered country, which is more hilly. Proceeded on through the thick
dead mulga scrub, to the north side of Forster range, where we camped at
dark without water. The country passed over to-day is splendidly grassed,
especially as we approached the range. There is also a little spinifex,
but not much. Distance to-day, thirty-two miles.

Friday, 13th July, North Side of Forster Range. Started early, proceeding
to the gum creek coming from the north side of Forster's range, where we
found a little water, numerous fresh tracks of natives, and a great
number of birds. I have named this the Barrow Creek, after J.H. Barrow,
Esquire, M.P. Crossed the range to the Stirling Creek, which we followed
down, and found an abundant supply of water. The upper part of it is now
dry, and it is difficult to say whether it is permanent or not; but, to
judge from the number of native tracks and encampments, and the many
birds, I should think it is. The wood-duck is also on some of the pools.
At dark we can hear the natives down the creek.

Saturday, 14th July, Stirling Creek. I shall give the horses a rest
to-day and to-morrow, for I do not expect to get water before we reach
the reservoir in the Reynolds range. I am afraid it will be all gone in
the Hanson and at the Centre.

Sunday, 15th July, Stirling Creek. Resting horses, etc., etc.

Monday, 16th July, Stirling Creek. The natives were prowling about during
the night, and startled three horses, which separated from the others,
went off at full gallop, and were not recovered till noon, about four
miles off. Too late to start to-day, for which I am very sorry, as every
hour is now of the utmost value to us, in consequence of the evaporation
of the water. Not the slightest appearance of any rain yet. Wind, south.

Tuesday, 17th July, Stirling Creek. Proceeded to the Hanson. Shortly
after we started, we were followed by the natives, shouting as they came
along, but keeping at a respectful distance. They followed us through the
scrub for about two hours, but when we came to the open ground at the
lagoons they went off. I intended to have halted and spoken to them
there, thinking it would not be safe to do so in the scrub. They were
tall, powerful-looking fellows, and had their arms with them. We then
went on to the Hanson, crossing numerous fresh native tracks. On nearing
the water, we saw five blacks, who took fright and went off at full
speed. There were many more in the distance; in fact, they seemed to be
very numerous about here. The country all round was covered with their
tracks. Found water still there, but had to clear the sand away a little
to give the horses a drink. Thinking that it would not be safe to camp in
the neighbourhood of so many natives, I went on to the Central Creek, and
in going through some scrub, we again disturbed some more, but could only
see children, one a little fellow about seven years old, who was cleaning
some grass seeds in a worley, with a child who could just walk. The
moment he saw us he jumped up, and, seizing his father's spear, took the
child by the hand and walked off out of our way. It was quite pleasing to
see the bold spirit of the little fellow. On nearing Central Mount Stuart
we saw two men, who made off into the scrub. Arrived at the creek after
dark, but the water is all gone. On examining the hole where the water
was, we discovered a small native well, with a very little water, too
little to be of any service to me. To-morrow morning I must push on
through the scrub to Anna's Reservoir. My horses are still very weak, and
I do not think they will be able to do it in a day. Wind variable.

Wednesday, 18th July, Centre. Starting early, we crossed the Hanson, and
got through the scrub to the gum plains, where we camped at sundown, the
horses not being able to do the whole journey in one day. The creeks
empty themselves into the plains, but there is no water. Still, from the
number of birds that are about, I think there must be water not far away,
but I have no time to search for it. If I do not find water in the gum
creek (which is doubtful) the horses will have another long day's
journey. They are suffering much from the dryness of the feed, three of
them being infected with worms. Wind, south-east.

Thursday, 19th July, Gum Plains. Made our way through the remainder of
the scrub, and arrived in the afternoon at the gum creek, where we found
a little water, and clearing away the sand, obtained enough for our
horses. There will be enough for them to-day and to-morrow morning. I
shall therefore stop here for the rest of the day. There are some heavy
clouds coming up from the west and south-west, which I hope will give us
rain. Wind still from the south-east. The natives have been upon our old
tracks through the whole of the scrub in great numbers, and there are
many traces of them about this creek, some of which are quite fresh. The
drying up of the water round about has compelled them to collect round
this and other creeks which are permanent.

Friday, 20th July, Gum Creek North-east of Mount Freeling. Crossed the
Reynolds range to Anna's Reservoir, which is still full of water. I may
now say that this is permanent. The water we camped at is gone, but there
is still a little down the creek. We could not get enough for the horses
this morning in the creek we have left. Judging from the number of native
tracks that we have crossed this morning, there must be permanent water
on the north side of the range, which is composed of immense blocks of
granite, apparently on the top of mica slate, with occasional courses of
quartz and ironstone. To the north-east of where we camped last night,
about three miles distant, is the point of the range, on which there is a
very remarkable high peak, composed of ironstone, with a number of very
rough rounded ironstone hills. I have named this Mount Freeling. Here I
found indications of copper, the only place I have seen it in all this
journey. The natives do not seem to have frequented this reservoir much
of late, as there were no fresh tracks within two miles of it. In the
creek close by, there were some very old worleys. No rain;
clouds all gone. Wind, still south-east.

Saturday, 21st July, Anna's Reservoir, Reynolds Range. I shall remain
here till Monday morning to rest the horses, for they need it much; they
all have sore backs. A small pimple made its appearance under the saddle,
and has gradually spread into a large sore, which we cannot heal up; it
makes them very weak. The clouds have again made their appearance from
the north-west, and the wind has also changed to that quarter. I hope we
shall now get some rain, so that I can make short journeys for my horses,
to enable them to gather strength. Two long journeys on successive days
without water would reduce them again to the same state of weakness as
they were in at the Bonney Creek. For the last fourteen days we have been
getting a quantity of the native cucumber and other vegetables, which
have done me a great deal of good; the pains in my limbs and back are
much relieved, and I trust will soon go away altogether if these
vegetables hold out. We boil and eat the cucumbers with a little sugar,
and in this way they are very good, and resemble the gooseberry; we have
obtained from one plant upwards of two gallons of them, averaging from
one to two inches in length, and an inch in breadth.

Sunday, 22nd July, Anna's Reservoir. On examining the creek near the
reservoir, we have found some more large and deep water holes. I have
named this Wicksteed Creek. The clouds are again heavy, and have every
appearance of rain; they and the wind both come from the north-west.

Monday, 23rd July, Anna's Reservoir. No rain has fallen; again all the
clouds are gone. Started early for the spring in the North gorge,
McDonnell range, which we noticed on April 14th. Camped at dark in the
thick scrub and spinifex. No feed for the horses, so we had to tie them
up during the night. Wind, south-east again.

Tuesday, 24th July, Dense Scrub and Spinifex. Started through the
remainder of the scrub to the gorge, where we arrived at 7 o'clock, after
twelve hours' journey. Camped outside, and drove the horses up to the
spring. There is still the same supply of water; it is an excellent
spring, and might be of great importance to future exploration. I have
named it Hamilton Spring. Wind, variable.

Wednesday, 25th July, Hamilton Spring, McDonnell range. Resting the
horses. Yesterday afternoon we passed a great number of fresh tracks of
natives apparently going to Hamilton Peak, which leads me to think there
must be permanent water there. The peak is very high--quite as high as
Mount Arden, but there is another part of the spur higher than it, to
which I have given the name of Mount Hugh; further to the west-north-west
is a mount, still higher, which I have named Mount Hay. Wind, north-east.
It has been very hot to-day.

Thursday, 26th July, Hamilton Spring, McDonnell Range. Started across the
ranges to Brinkley Bluff, and camped on the east side. There is still
plenty of water in the Hugh, although greatly reduced. The natives have
been following our former tracks in great numbers; some of their
foot-prints are very large. There is a great quantity of marble in this
creek.

Friday, 27th July, Brinkley Bluff, McDonnell Range. Started down the
Hugh, and camped on the south side of Brinkley Bluff, finding plenty of
water all the way, in holes of various sizes, with reeds and rushes
growing round them, with plenty of feed on the banks. Wind, variable.

Saturday, 28th July, The Hugh, South Side of Brinkley Bluff, McDonnell
Range. Proceeded towards the Waterhouse range, and stopped at my former
camp of the 11th April. The spring still gives out an abundance of water;
we have also found another good spring on the south side of the creek,
which is here very broad, nearly two hundred yards wide, with a good
feeding country all round, and a small strip of salt-bush on the banks.
Splendid gum-trees in the creek. Wind, east; sun, hot.

Sunday, 29th July, The Hugh, between McDonnell and Waterhouse Ranges.
Wind variable; some clouds coming from the south-west.

Monday, 30th July, The Hugh, between McDonnell and Waterhouse Ranges.
Proceeded towards the range; at four miles crossed the creek, and half a
mile further entered the ranges. We made our former camp of April 9th on
the creek, but no water, so followed it down to the westward, and after
clearing a hole, found sufficient for our wants in the sand. Camped. Very
unwell. Wind, south-east. Not a drop of rain has fallen since we were
here before.

Tuesday, 31st July, Between the Waterhouse and James Ranges. Started on a
course of 220 degrees, following down the creek through James range,
instead of crossing it. I am afraid there will be no water at our camp on
the south side. I have a chance of getting some in the range. At two
miles met with a good water hole, under a sandstone hill. At seven miles
the creek enters the range; the bed is broad, sandy, and gravelled. At
twelve miles we found some water, and camped, as I am too unwell to
continue in the saddle any longer. Cleared a hole, and obtained water
sufficient for our purpose. Wind, south-east.

Wednesday, 1st August, In James Range, on the Hugh. Followed the creek
through the remainder of the range, and found water in four different
places. I have not the least doubt that there is plenty, but the creek is
so broad, and divided into so many courses, that it would require four
men at least to examine it well. On arriving at our camp of the 7th
April, we found all the water gone. Scratched in the sand, and found a
little moisture, but no water; after a fruitless search of an hour, I was
going back to the last water that I had seen, six miles distant, when two
emus came into the creek, and made for a large gum-tree in the middle. On
going to it, I found a fine hole of water round its roots. Camped. Wind
the same.

Thursday, 2nd August, The Hugh, South Side of James Range. Went down the
south side of the creek, through good grassy country. At fourteen miles
in a side creek we found a native well about four feet deep. We camped
here, as there is little prospect of finding any more water in the Hugh,
which is become broad and sandy. As to surface water, my men have neither
the strength nor the appliances for digging. There is plenty of water
under this sand, but having only a small tin dish, the labour is too
great. My men have now lost all their former energy and activity, and
move about as if they were a hundred years old; it is sad to see them;
our horses, too, suffer very much from their sore backs. On the south
side of the creek are some isolated hills, chiefly composed of limestone,
ironstone, quartz, and granite. This morning there was ice on the water
left in the tin dish, and also in the canteens, an eighth of an inch
thick. It was very cold.

Friday, 3rd August, James Range. I find the water in the well is nearly
all gone this morning. It would take us nine hours to water the horses
here, so slowly does it come in; I must therefore go back to our last
camp. I shall follow the creek round, for there might be a chance of
getting some nearer. Saddled, and proceeded up the creek, and at four
miles found a little under the limestone rocks coming from a small side
creek; gave the horses a drink turning back, and made for the Finke on a
course of 160 degrees. Crossing a few stony hills and small plains, at
ten miles, we ascended a broken table range, which I have named Warwick
Range; it is composed of hard grey limestone and ironstone. We then
proceeded through a well-grassed country, with mulga bushes, and at
twenty miles camped under a redstone hill, not being able to get any
further. No water.

Saturday, 4th August, Small Hill between the Hugh and the Finke. The
horses strayed a long way in the night, so that I did not get them till
after 11 o'clock this morning, and could not start until noon. Passed
over a country of much the same description as yesterday, crossing three
stony hills running nearly east and west, and at nine miles camped,
without water, in a fine grassy country, which, as the grass is green,
will be quite a treat for the horses. About six miles north of Chambers
Pillar. Wind, south-east.

Sunday, 5th August, North of Chambers Pillar. At sunrise heavy clouds
came up from the south-east, bringing with them a very thick fog, through
which I had great difficulty in steering my course; it cleared off about
10 o'clock. I expected rain, but none has fallen; it is now quite clear
again. Arrived at the Finke at 12 o'clock, and was very much surprised to
find so little water. I had no idea it would have gone away so soon. The
bed is very broad and sandy, which is the cause of the rapid
disappearance of the large quantity that I saw when I crossed before.
This is a great disappointment, as it was my intention to run it down, in
the hope that it would take me into South Australia. I shall go one day's
journey down, and see what it is; if I can find no more water I must
return to this, to rest my horses, and push for the Stevenson. I cannot
remain here, for this water will only last a short time. My provisions
will barely carry me down, and there is not the least appearance of rain.
I am afraid my retreat is cut off. Wind, south-east. Clouds.

Monday, 6th August, The Finke. Thick fog again this morning. From the
heavy clouds that have passed yesterday to the south of us, I think a
shower of rain may have fallen there; I ought not to allow the chance of
it to escape, as it is likely to be my only one until the equinox, and I
have not provisions sufficient to remain until that time, so I must push
the horses as far as they will go, and then we must walk the rest, which
is a very black prospect, considering the weak state we all are in.
Proceeded to the south-east, having camped on my former course at two
clay-pans, where I think there is a chance of water, if a shower has
fallen there. Started on our former course and arrived at the clay-pans
without seeing a drop of water; neither is there any in them. Camped; the
horses being very tired, from coming through so many sand hills.

Tuesday, 7th August, Clay-pans in Sand Hills. A light dew fell last night
and this morning, which I am very glad of; it will be a good thing for
the horses. Kekwick was unwell last night, but I cannot stop on his
account. He must endure it the best way he can. If I find water at where
I suppose the Finke joins the gum creek that runs a little north of Mount
Humphries, I will remain there a day to give him rest. He is completely
done up. I hope he will not get worse. I must push back as quickly as
possible, and get him into the settled districts. At noon we made the
Finke. Still the same white, sandy bed; but here it is about a quarter of
a mile broad, and the east bank is composed of white sandstone, with a
course of light slate on the top of it, then courses of limestone and
other rocks, and, on the top of all, red sand hills. The gum-trees are
not so large as they are further north. On first striking the creek we
could find no water, but, by following it down for a short distance, we
discovered a little, which will do for us. It is more than I expected,
and I feel most thankful for it. Kekwick still very ill. Poor fellow, he
is suffering very much. I dare not show him much pity, or I should have
the other giving in altogether. I hope and trust he will soon get better
again, and that to-morrow's rest may do him good. He has been a most
valuable man to me. I place entire confidence in him. A better one I
could not have got. I wish the other had been like him, and then neither
he nor I should have suffered so much from hunger. Wind, south-east.

Wednesday, 8th August, The Finke. Resting Kekwick and shoeing horses.
This water was going away very rapidly, so I rode down the creek for ten
miles to see if there were any more, that I may risk following it down.
After joining the West Creek it spreads itself over a broad valley,
bounded on the north by sand hills and on the south by stony hills.
Course, eastward. It is divided into numerous courses; very sandy, and
immense quantities of drift wood about it. Some very large gum-trees
piled high on the banks, and a great number of birds of every
description; but I could find no water. It is so broad, with so many
courses, that it would require half a dozen men to examine it well. If we
were to stay searching for water here, and be unsuccessful, and the
creeks on ahead were to be dried up, we should lose our horses and have
to walk, which Kekwick could not do. I do not consider it would be right
thus to risk his life. I shall therefore make for the Stevenson, where I
am almost certain to find water. Wind, east.

Thursday, 9th August, The Finke. Started early on our former tracks,
passing Mount Humphrey and Mount Beddome. Camped at our old place. I
should think from the appearance of the country that the Finke takes a
south-east course from where I left it yesterday. The hills run that way.
Wind, south-south-east.

Friday, 10th August, South of Mount Beddome. Proceeded on our former
course to the Stevenson, which we made a little before dark, and found
water, but I am quite surprised to see so little of it left. The fine
large holes are nearly dry. Wind, east.

Saturday, 11th August, The Stevenson. The horses having lost some shoes,
I am forced to remain here to-day to put others on. There is more water a
little further down the creek, at which I camped. No rain seems to have
fallen since I was here before. The sun has been very hot to-day. Wind,
east-south-east.

Sunday, 12th August, The Stevenson. I was too unwell to move yesterday,
but, feeling a little better this morning, I rode down the creek. For
three miles it takes a south-east course, then east-south-east through
table land, with rocky and precipitous hills on each side. I then went on
a south-east course for nine miles, through a splendidly-grassed country,
with numerous small creeks running into the Stevenson. During my ride I
found plenty of water, and splendid grass, up to the saddle-flaps, and
quite green. Ducks and numerous other birds abound here; the water is
quite alive with them. I regret that I have not provisions enough to
enable me to follow this creek round its different bends. It is a
splendid feeding country for cattle, and much resembles Chambers Creek.
Wind, south-east.

Monday, 13th August, The Stevenson. Started on a course of 135 degrees to
see if the Stevenson comes from the south; continued on the table land,
from where I left it yesterday for sixteen miles from last night's camp,
when we suddenly dropped into the bed of a large broad sandy gum-creek,
coming from the west, which I find to be the Ross. There are many rushes
about it; it runs in three or four courses, in all of which water can be
obtained by scratching in the sand. There are plenty of birds. It is
evidently raining to the east of this. Camped. My course takes me across
the middle of a range, which I shall endeavour to cross to-morrow. There
are two small springs, but they are brackish. Wind, south.

Tuesday, 14th August, The Ross. Started on the same course, 135 degrees,
and again ascended the stony table land. Crossing thence, we met two
small myall-creeks running north-east with birds upon them. At seven
miles crossed another, and found a fine large deep water hole with ducks
on it. We again ascended the table land, which continued to the range,
and at sixteen miles gained the top, which is table land about a mile
broad; the view is extensive to the east-north-east and north. We
descended on a course of 175 degrees to search for water in the creek
below. We crossed a number of myall-creeks, coming from the range, and
running south-east; in many the water has just dried up. At six miles on
the same course we found water and camped, the horses being tired by
their rough journey. This water hole is not permanent although when full
it is deep and large, and will last a considerable time. The Stevenson
and Ross seem to take a north-east course. On a further examination of
this creek I found a large hole of water about two hundred yards long and
thirty broad, with birds upon it, and plants that grow round permanent
water. I also found shells. This creek I have named Anderson Creek, after
James Anderson, Esquire, of Port Lincoln, and the range Bagot Range,
after the Honourable the Commissioner of Crown Lands.

Wednesday, 15th August, Anderson Creek. Started towards the south-east
point of Bagot range, which I find to be five miles distant. The country
between is undulating and stony, with plenty of grass. To the east, about
thirty miles, is a high isolated hill, bearing 100 degrees. At six miles
and a half crossed a myall and gum creek, in which, about a mile to the
east, under a red bank, is a large water hole, seemingly permanent. At
ten miles crossed the Frew, whose bed is sandy, and has many courses, the
banks being covered with rushes. The rest of the day's journey was
through mallee scrub and sand hills, in which we camped without water;
the feed, however, is abundant, yet not so thick as when I crossed
before.* (* See ante, March 28, 29, and 30.) Wind, south.

Thursday, 16th August, Mulga Scrub and Sand Hills. Started at 7 o'clock
on a course of 170 degrees, and in four hours made the Neale, and camped,
as there was still plenty of water.

Friday, 17th August, The Neale. Proceeded on a south-east course, and
camped on a side branch of the Neale, with plenty of water in large
holes. Wind, east.

Saturday, 18th August, Side Branch of the Neale. Proceeded towards the
gap in Hanson range, and camped near one of the large water holes. It is
very cloudy.

Sunday, 19th August, Gap in the Hanson Range. Still cloudy, and looks
like rain, so we must push on to-day, in case the Peake River should come
down and stop us, which would not suit the state of my provisions, as we
have lost a quantity of flour by the scrub scoring the bags, and we have
not enough to take us to Chambers Creek. At eight miles camped
west-north-west of Freeling Springs, having given the horses a drink in
crossing the Neale.

Monday, 20th August, Sand Hills West-north-west of Freeling Springs. It
still threatens for rain. Proceeded to Kekwick Springs to see if the
horse we had left in the Peake had got out. We found his bones; he does
not seem to have made a struggle since we left him, as he is in the same
position. From the number of tracks, the natives must have visited him.
Proceeded to Freeling Springs and camped. There were a number of ducks
and two swans on the large water hole. We shot one of the latter, which
was a great treat to our half-starved party. Wind variable.

Tuesday, 21st August, Freeling Springs. Still cloudy, and we had a few
drops of rain during the night; also distant thunder and lightning.
Resting horses. Wind, north-east.

Wednesday, 22nd August, Freeling Springs. Proceeded through Denison
range, and camped at the Milne Springs. Wind, north-east. Still cloudy,
but no rain.

Thursday, 23rd August, Milne Springs. Went on and camped at Louden Spa.
Wind variable.

Friday, 24th August, Louden Spa. Camped at the William Springs. Wind,
north-west.

Saturday, 25th August, William Springs. Proceeded to the Strangway and
Beresford Springs, and camped at Paisley Ponds. Wind, north-east.

Sunday, 26th August, Paisley Ponds. During the night thunder and
lightning from the north-west, with a few drops of rain. Cloudy this
morning; had a few showers on our journey to Hamilton Springs. Found Mr.
Brodie camped there three miles south-east of Mount Hamilton. He received
and treated us with the greatest kindness.

Mr. Stuart and his party remained at Hamilton Springs until 1st
September, when they proceeded to Chambers Creek, where, having reached
the settled districts, his journal ends.


JOURNAL OF MR. STUART'S FIFTH EXPEDITION. FROM NOVEMBER, 1860, TO
SEPTEMBER, 1861.

When Mr. Stuart reached Adelaide, in October, 1860, on his return from
his last expedition, bringing with him the intelligence that he had
penetrated to the northward almost as far as the eighteenth degree of
south latitude, and had only been forced to retreat by the hostility of
the natives, the South Australian Parliament voted a sum of 2500 pounds
for a larger, better-armed, and more perfectly organized party, of which
he was to be the leader. The ill-fated Victorian expedition, under Burke
and Wills, had already started from Melbourne, on the previous 20th of
August, amid all the excitement of a popular ovation, but a messenger was
instantly despatched by the Victorian Government to overtake them, in
order to give them what information the South Australian Government
allowed to be known. On the 29th of November Mr. Stuart was ready to
start once more, and left Moolooloo with seven men and thirty horses,
arriving at Mr. Glen's station on the 1st of December, and at Goolong
Springs on the 4th. He was delayed at the latter place for several days,
in consequence of the horses, and more especially the town horses, being
unmanageable and unequal to their work. The party reached Welcome Springs
on the 8th, and Finniss Springs on the 11th. The water at Finniss Springs
seemed to have an injurious influence on the town horses, but those that
had been with Mr. Stuart on his previous journeys were not so much
affected. The following evening they arrived at Chambers Creek, where
they remained until the end of the month.

During their stay at Chambers Creek they were occupied in killing and
drying bullocks, mending saddles, weighing rations, shoeing horses, and
generally preparing to start. Several of the horses, which had been
knocked up and left behind on the way, had to be brought up; others
became quite blind, one was lost, and one died. On the 31st of December
four fresh horses arrived, which had been kindly sent up by Mr. Finke the
moment he heard of the difficulty in which Mr. Stuart was placed. The
party was also further increased, both by horses and men, so that when it
left Chambers Creek, on the 1st of January, 1861, it numbered twelve men
and forty-nine horses. The following is the list of those who started:--

John McDouall Stuart, Leader of the Expedition.
William Kekwick, Second in Command.
F. Thring, Third Officer.
-- Ewart, Storekeeper.
-- Sullivan, Shoeing Smith.
-- Thompson, Saddler.
-- Lawrence.
-- Masters.
J. Woodforde.
-- Wall.
E.E. Bayliffe.
J. Thomas.

Shortly after starting, the horses that Mr. Finke sent up went off at a
gallop, taking with them one of the others; but, at about a mile, they
were headed by Ewart, Wall, and Lawrence, and brought back covered with
sweat. Not content with this gallop, in a short time afterwards they
bolted again. This last one seemed to content them, for they went very
quietly for the rest of the day; they had, however, lost a pick, which
could not be found. The party arrived at Mr. Ferguson's station, at
Hamilton Springs, that evening. Louden Spa was reached on the 8th of
January. The next day Mr. Stuart writes:

"Wednesday, 9th January, Louden Spa. I am obliged to leave two horses. I
thought that I should have been able to have got them down as far as Mr.
Levi's station. There are three others that I must leave behind; they are
now nearly useless to me, and cause more delay than I can afford. I shall
reduce my party to ten individuals, in order to lighten the horses that I
take with me. I shall take thirty weeks' provisions; the rest I shall
leave there (Mr. Levi's station). The two men who are to return are to
have a month's provisions to carry them down. They will be here two
weeks, and if the horses have not recovered by that time, they will
remain another week, when they will have one week's provisions to take
them to Chambers Creek, where they will get enough to carry them to the
mine."

Bayliffe and Thomas were the two men selected to return, and it may not
be without interest to follow them back to the settled districts. They
did not arrive at Melrose, Mount Remarkable, until the latter end of
March. Thomas was suffering severely from rheumatism, and had to be
conveyed in a cart for the last six miles of his journey from a place
where he and his companion had camped for the purpose of recruiting
themselves. They had been obliged to leave two of the horses at Mr.
Mather's station, and two more had died on the road. The men arrived with
one horse only, which they were using as a pack-horse.

But to return to the rest of the party, who reached Mr. Levi's station
the same evening (January 9th) on which they parted from the two men. On
Friday, January 11th, Mr. Stuart writes:

"I have now all put in order, and consider myself fairly started, with
thirty weeks' provisions. Day extremely hot. An eclipse of the sun took
place at noon. Although our poor little dog Toby is carried on one of the
pack-horses, he is unable to bear this great heat. I fear he will not
survive the day. Arrived at Milne Springs about 5 p.m. At sundown poor
little Toby died, regretted by us all, for he had already become a great
favourite."

On January 21st Mr. Stuart reached the Neale Creek, a little to the east
of where he struck it before, but found that the large bodies of water
had nearly all gone; by digging in the sand of the main channel, however,
they obtained sufficient for their immediate wants. Exploring parties
were despatched up and down the creek, and returned, reporting abundance
of water eight miles above and five miles below where they were. They
also brought back with them some fish, resembling the bream, which were
very palatable when cooked. An attack of dysentery prevented Mr. Stuart
from proceeding for a few days, and, during his stay, the natives, while
studiously keeping themselves out of sight, set fire to the surrounding
grass. On the 27th the expedition arrived at the Hamilton, after a heavy
journey of thirty-five miles. "I observed," says Mr. Stuart, "a peculiar
feature in one of the families of the mulga bushes; the branches seemed
to be covered with hoar frost, but on closer examination it turned out to
be a substance resembling honey in taste and thickness. It was
transparent, and presented a very pretty appearance when the sun shone
upon it, making the branches look as though they were hung with small
diamonds."

The course now taken was through Bagot range to the Stevenson, where they
arrived on February 1st. The next day they proceeded northward, and at
eight miles came upon a large water hole, which was named Lindsay Creek,
after J. Lindsay, Esquire, M.L.A. This water hole was one hundred and
fifty yards long, thirty wide, and from eight to fifteen feet deep in the
deepest parts. The native cucumber was growing upon its banks, and the
feed was abundant. Here they met with immense numbers of brown pigeons,
of the same description as those found by Captain Sturt in 1845. There
were thousands of them; in fact, they flew by in such dense masses that,
on two occasions, Woodforde killed thirteen with a single shot. The
travellers pronounced them first-rate eating. Many natives, tall,
powerful fellows, were seen, but they did not speak with them. After
trying for water in the neighbourhood of Mount Daniel, they were
compelled to return to Lindsay Creek, which they did not quit until
February 9th, when they camped on another creek, which was named the
Coglin, after P.B. Coglin, Esquire, M.L.A. From this place Mr. Stuart
started, accompanied by Thring and Woodforde, to examine the condition of
the Finke, and found its bed broad, and filled with white drift sand, but
without water. A hole ten feet deep was sunk in the sand, but just as the
increasing moisture gave them hope of finding water, the sides gave way,
and Thring had a narrow escape of being buried alive. After sinking
several other holes, but without success, they turned to another creek,
coming more from the westward, and in a short time discovered six native
wells near to what was evidently a large camping-place of the natives.
The ground for one hundred yards round was covered with worleys, and at
one spot they seemed to have had a grand corroberrie, the earth being
trodden quite hard, as if a large number had been dancing upon it in a
circle. They had left one of their spears behind, a formidable weapon
about ten feet long, with a flat round point, the other end being made
for throwing with the womera. On the 13th Mr. Stuart and his two
companions returned to the camp on the Coglin, after discovering a place
about four miles from the six native wells, where sufficient water could
be obtained by digging. On the 14th three of the men were sent in advance
to dig a hole at this place, and the following day the whole party moved
forward to join them. Here the natives annoyed them much by setting fire
to the grass in every direction.

Marchant Springs (on the Finke) were reached on February 22nd, and here
Mr. Stuart noticed a remarkable specimen of native carving. He says: "The
natives had made a drawing on the bark of two trees--two figures in the
shape of hearts, intended, I suppose, to represent shields. There was a
bar down the centre, on either side of which were marks like broad
arrows. On the outside were also a number of arrows, and other small
marks. I had a copy of them taken. This was the first attempt at
representation by the natives of Australia which I had ever seen."

Following the course of the Finke, they arrived on the 25th at some
springs which were rendered memorable by Mr. Stuart's favourite mare
Polly. She became very ill, and on the morning of the 26th slipped her
foal. Polly had been with her master on all his previous journeys, and
was much too valuable and faithful a creature to be left behind; besides,
she was second to none in enduring hardship and fatigue. They therefore
waited another night to give her time to recover, and Mr. Stuart named
the springs Polly Springs in her honour. On the 27th they again moved
northwards, still following the course of the Finke, and, after a short
journey of ten miles, camped at what were afterwards called Bennett
Springs. It is worthy of remark that while the horses were in this water
drinking, one of them kicked out a fish about eight inches long and three
broad--an excellent sign of the permanency of the water. Here several of
the horses were taken violently ill, and the next morning one of them
could not be found. Mr. Stuart writes:

"Thursday, 28th February, The Finke, Bennett Springs. Found all the
horses but one named Bennett. Sent two of the party out in search of him;
at 9 a.m. they returned, having been all round, but could see nothing of
him. I then sent out four, to go round the tracks and see if he had
strayed into the sand hills. At noon they returned unsuccessful. Sent
five men to search, but at 2 p.m. they likewise returned without having
discovered him. I then went out myself, and, in half-an-hour, found the
poor animal lying dead in a hole, very much swollen. Blood seemed to have
come from his mouth and nostrils. He must have died during the night. I
am afraid that there is some description of poisonous plant in the sand
hills, and that the horses have eaten some of it. As he lay he appeared
to have been coming from the sand hills, and making for the water. He
seemed to have fallen down three times before he died. I never saw horses
taken in the same way before--in a moment they fell down and became quite
paralysed. The cream-coloured horse, that was taken so ill last night,
must also have eaten the poison. We were upwards of two hours before we
could get him right. As soon as he got on his legs, his limbs shook so
that he immediately fell down. This he did for more than a dozen times.
As we were very much in want of hobble-straps, I sent Mr. Kekwick, with
three others, to take Bennett's skin and shoes off. We found no
indication of poison on opening him. This is a very great loss to me, for
he was one of my best packhorses--one that had been with me before, and
that I could depend upon for a hard push."

On the 2nd March, while still following the course of the Finke, they
passed two or three holes containing fish about eight inches long, and
enclosed by small brush fences, apparently for the purpose of catching
fish. They also saw a lot of shields, spears, waddies, etc., which the
natives had deposited under a bush. As to the aborigines themselves,
although it was evident there were plenty of them about, they never
allowed themselves to be seen. There was an abundance of timber which Mr.
Stuart says would be well suited for electric-telegraph poles.

Mr. Stuart's journal continues as follows:

Tuesday, 5th March, The Finke. Started at 8.5 a.m., bearing 345 degrees,
for the Hugh, with Thring and Lawrence. On arriving there found the water
nearly all gone, only a little in a well dug by the natives; cleared it
out, but it took us until 12 p.m. to water the four horses. At three
miles further, we passed round a high conspicuous table hill, having a
slanting and shelving front to the south; this I have named Mount Santo,
after Philip Santo, Esquire, M.P. The country passed over to-day has been
sand hills, with spinifex, grassy plains, with mulga and other shrubs,
and occasionally low table-topped hills, composed of sand, lime, and
ironstone, also the hard whitish flinty rock; kangaroo plentiful, but
very wild. Wind south-east. The day has been very hot; horses very tired.

Wednesday, 6th March, The Hugh. Started at 8.45 a.m. on a bearing of 209
degrees. At nine miles, finding the water gone that I had seen on my last
return, I dug down to the clay, and obtained a little, but not enough for
us. Followed the creek up into the gorge, and found it very dry. Our
former tracks are still visible in the bed of the creek. No rain seems to
have fallen here since last March. I had almost given up all hopes of
finding any water, when, at seven miles, we met with a few rushes, which
revived our sinking hopes; and, at eight miles, our eyes and ears were
delighted with the sight and sound of numerous diamond birds, a sure sign
of the proximity of water. At the mouth of a side creek coming from the
James range, on the eastern side of the Hugh, found an excellent water
hole, apparently both deep and permanent. We saw a native and his lubra
at the upper end at a brush fence in the water; they appeared to be
fishing, and did not see us until I called to them. The female was the
first who left the water; she ran to the bank, took up her child, and
made for a tree, up which she climbed, pushing her young one up before
her. She was a tall, well-made woman. The man (an old fellow), tall,
stout, and robust, although startled at our appearance, took it leisurely
in getting out of the water, ascended the bank, and had a look at us; he
then addressed us in his own language, and seemed to work himself up into
a great passion, stopping every now and then and spitting fiercely at us
like an old tiger. He also ascended the tree, and then gave us a second
edition of it. We leisurely watered our horses, and he was very much
surprised to see Thring dismount and lead the pack-horse down to the
water, so much so that he never said another word, but remained staring
at us until we departed, when he commenced again. This water being
sufficient for my purpose, I will go no further up the creek, but return
to the last night's camp. Wind, south-east.

Thursday, 7th March, The Hugh. As my horses are very tired, and the
distance between my main camp on the Finke and the water we discovered
yesterday being upwards of fifty miles, I will remain here to-day, dig
down to the clay, and try if I can obtain enough water for all the party;
for, owing to the extreme heat, and the dryness of the feed, many of our
weak horses are unable to go a night without water. By 8 p.m. we dug a
trench ten feet long, two feet and a half deep, and two feet and a half
broad; it is about twelve feet below the level of the creek. We have had
a very hard day's work. Wind, south-east. Day very hot.

Friday, 8th March, The Hugh. This morning very cold; wind, still
south-east. The trench is quite full; our four horses made very little
impression on it. I shall send up and enlarge the trench, so that we may
be enabled to water the whole lot. At 6.40 a.m. started back for the
camp. At 1.45 p.m. halted to give the horses a little rest. At 2.30 p.m.
changed to 184 degrees, and at four miles reached the table hills, but
there was no creek, only a number of clay-pans, all quite dry, with
stunted gum-trees growing round them. Changed my bearing to Mount Santo,
passing a number of clay-pans of the same description; from thence
proceeded to the camp; arrived there at sundown, and found all right.
Plenty of water; the horses make little impression on it. Wind,
south-east.

Saturday, 9th March, The Finke. I shall give Thring a rest to-day, and
will send him with two others, and a part of the horses, to-morrow to the
Hugh, to make a place large enough to water all. From about 2 a.m. until
after sunrise the morning has been very cold. Wind, south-east.

Sunday, 10th March, The Finke. At 7 a.m. despatched Thring, Thompson, and
Sullivan, with eleven pack and three riding-horses, to the Hugh to dig a
tank. Wind, still south-east; clouds east.

Monday, 11th March, The Finke. Clouds all gone; wind still south-east. I
will remain here to-day with the rest of the party, to give the others
time to have all ready for us when we arrive. One of the horses missing;
found him in the afternoon. Wind variable.

Tuesday, 12th March, The Finke. Started at 8.30 a.m. for the Hugh, course
345 degrees, following our former tracks. The day has been exceedingly
hot; wind from east and south-east, with heavy clouds in the same
direction. About 3 p.m. missed the party that was behind; they were last
seen about one mile and a half back. Thinking that the packs had gone
wrong, and that they were remaining behind to repair them, I waited an
hour, but finding they did not come up, I sent Ewart back to the place
where they were last seen to find out what was wrong; in an hour he
returned, and informed me that their tracks were going away to the
eastward. As the James range was in sight, and two of the party had been
there before, I concluded that they must have lost my tracks and were
pushing on for the water. This loss of two hours would make it late
before we arrived there, so we hurried on; but within four miles it
became so dark, from the sky being overcast with heavy clouds, and the
mulga bushes being so thick, that we were in great danger of losing some
of our pack-horses, for we could not see them more than ten yards off. I
therefore camped until daylight, having to tie the horses during the
night. Wind variable.

Wednesday, 13th March, Between the Finke and the Hugh. Started at
daybreak; and in a little more than an hour arrived at the Hugh; found
that Thring had gone up the creek to the other water, not finding enough
here for the horses he had with him. We could only get sufficient for ten
of ours. As the fire was still alight, I was led to believe that the
other party had arrived here last night, having had two hours more
sunlight than we, and that they, seeing Thring's note to me, which he had
fastened on a tree, and also the small quantity of water, had watered
their horses last night, and gone on this morning, leaving the water that
had accumulated during the night for us and our horses; we cleared out
the hole in order to obtain sufficient for our other five. At about 10
a.m. had breakfast; before we finished, the other party came in sight;
they had lost the tracks, and could not find them again. They made the
creek about one mile to the eastward. Unsaddled and gave their horses a
rest, and as much water as we could get for the weak ones; those of mine
which have had none will have to go without. By 1 p.m. obtained a drink
for seven of them. Pushed on to the other water, fifteen miles up the
creek; arrived there a little before sundown. The day, although cloudy,
has been very hot. Found Thring and his party all right. They had seen no
more of our spitting friend. Wind variable, with heavy clouds from east
and south-east, but still no rain.

Thursday, 14th March, The Hugh, James Range. As the done-up horses will
not be able to travel to-day, I have sent Thring and Wall up the creek to
look for other water. Sky still overcast. No rain. Thring and Wall
returned in the afternoon, having found water a little below the surface,
about nine miles up; a very light shower has fallen. Wind all round the
compass.

Friday, 15th March, The Hugh, James Range. A few drops of rain have
fallen during the night, but this morning it seems to be breaking up
again, which is a great disappointment. Started at 8 a.m., course 10
degrees west of north; passed through the gorge in James range, found all
the water gone that I had seen on my journey down; followed up the creek
to the native wells that Thring found yesterday. This water is situated
about one mile and a half from where the creek enters the gorge in James
range, and under a concrete bank on the north side. The natives seem to
have quitted this water on hearing us coming, for they have left behind
them a large, long, and unfinished spear, two smaller ones, and some
waddies, one of which was quite wet, as if the owner had been in the act
of clearing out one of the wells when he heard or saw us coming: he also
left a shield cut out of solid wood, which I think was, from its
lightness, cork-wood. I also observed on one of the gum-trees, marks
similar to those which I saw on the Finke, broad arrows and a wavy line
round the tree. Still cloudy, but much broken. No rain. Wind, south-east.

Saturday, 16th March, The Hugh, James Range. Rain all gone. Proceeded up
the creek, course 30 degrees, to examine the east bend before it enters
the Waterhouse range; in about six miles arrived and followed it upwards,
pushing on through the gorge to the large water I had previously seen on
the north side of the range; found it gone, but water in some native
wells in its bed. Proceeded on to the second bend of the creek from
Waterhouse range, to a water which I consider to be a spring (it is under
conglomerate rock), and am glad to see that there is still a large hole
of beautiful water, with bulrushes growing round about it. Camped. This
water I have named Owen Springs, after William Owen, Esquire, M.P. Wind
variable, from south-east to north-east. Cloudy.

Sunday, 17th March, Owen Springs, The Hugh. During the night we had a few
light showers, which will be of great advantage to us, causing the green
feed to spring up. The morning still cloudy; wind from the east, with a
few drops of rain. Wind still variable--all round the compass.

Monday, 18th March, Owen Springs, The Hugh. Very heavy clouds this
morning; and it seemed as if it was setting in for a wet day, but it
cleared off, and only a little rain fell. Wind still all round the
compass.

Tuesday, 19th March, Owen Springs, The Hugh. Saddled and started for
Brinkley Bluff, bearing 349 degrees. After entering the McDonnell range
the water is permanent. It has been here for twelve months; no rain has
fallen during that time, for my former tracks, both up and down, are as
distinct as if they had been made a month ago. At 3.30 p.m. camped at the
waterhole about a mile north-west of Brinkley Bluff; it is situated under
a rocky cliff. There are some seams of beautiful grey granite crossing
the creek, and abundance of marble of all colours, also a little iron and
limestone. We found some specimens of the palm tree, but there is neither
seed nor blossom at this season of the year. Lawrence got one of the
leaves, ate the lower end of it, and found it sweet--resembling
sugar-cane; he ate a few inches of it, and in about two hours became very
sick, and vomited a good deal during the evening. Wind variable; but
mostly south-east, with heavy thunder clouds.

Wednesday, 20th March, Brinkley Bluff, McDonnell Range. About 1 p.m. we
were delighted with the sight and feeling of heavy rain. At about 4 the
creek came down, and by sunrise it was running at the rate of five miles
an hour--a new and delightful sight to behold. At about 9 the clouds were
breaking and the rain lighter. We were all truly thankful for this great
boon. It is too wet to move to-day; the horses are bogging up to their
knees. After sundown we had a heavy thunder storm, accompanied by vivid
lightning, and heavy rain from south-east and east. Wind from
same direction.

Thursday, 21st March, Brinkley Bluff, McDonnell Range. Rain has continued
at intervals during the night; a great deal has fallen. A horse having
gone into the creek to drink during the night, one of his hobbles became
undone, and got fastened to his hind shoe. He was found this morning up
to his body in water, and unable to move. Having relieved him, it was
with difficulty he could get out. He is in a tremble all over, and can
scarcely walk. The ground is so soft, even on the hills, that we cannot
walk without sinking above the ankle. I should gain nothing by starting
to-day. It would injure the horses more than a week's travelling.

Friday, 22nd March, Brinkley Bluff, McDonnell Range. About 1 a.m. the
rain came down in torrents, and continued until nearly sunrise, from
south-east. Wind from same quarter. It is impossible to move to-day. The
creek is higher than it has been before, and running with great rapidity.
All the horses were found right this morning but the one which got into
the creek yesterday. After searching all the hills and the creeks round
about, he was found in a small gully by himself.

Saturday, 23rd March, Brinkley Bluff, McDonnell Range. Heavy shower of
rain about 4 a.m. this morning. After sunrise it all cleared away and
became fine. Started at 8.20 to cross the northern portion of the range
by following the creek up. We have had a very hard and difficult journey
of it. It is now 4 p.m., and we have arrived at Hamilton Springs. The
ground was so soft, even at the top of the ranges, that we had the
greatest difficulty in getting the horses through. We did so, however,
with the loss of a great number of shoes, and many of the horses were
very lame. Wind still south-east.

Sunday 24th March, Hamilton Springs. I am compelled to have some of the
horses shod to-day, and also to have a number of saddle-bags mended,
which were torn by the scrub yesterday. This afternoon there is a great
deal of thunder and lightning in the north and north-east.

Monday, 25th March, Hamilton Springs. Part of the horses missing this
morning in consequence of the green feed; did not get a start until 10.20
a.m.; bearing 43 degrees. The country became so boggy after seven miles
that we were unable to proceed further than eleven miles. There being no
surface water, although the ground was so soft that the horses kept
bogging up to their bodies, we were forced to retreat five miles to
obtain some for them. Wind south-east, the stormy weather apparently
breaking up. Camped at 5 p.m. Latitude, 23 degrees 28 minutes 51 seconds.

Tuesday, 26th March, Scrub North-east of Hamilton Springs. Started at 9
a.m. on a south-south-east course to round the boggy country. At about
six miles we were enabled to cross the lower part, and go in the
direction of a low range. Camped on the north-east side of it. The last
four miles were over fair travelling-country of a red soil, with mulga
and other bushes, in some places rather thick, abounding in green grass.
We also passed many bushes of the honey mulga, but the season is passed,
and it is all dried up. Wind, east. Latitude by Pollux, 23 degrees 24
minutes 51 seconds; by Jupiter, 23 degrees 24 minutes 52 seconds.

Wednesday, 27th March, Low Granite Range in Scrub. More than half of the
horses are missing this morning; at noon we have managed to get all but
ten; they are scattered all over the place; at 5 p.m. they cannot be
found, and the water is nearly all gone, and the country much dried
towards Strangway range. I have sent the horses four miles back to a
large clay-pan that we saw yesterday, to remain there to-night and in the
morning to return. Two of the party to separate from there, and to go in
search of the missing horses, which I suppose have gone back to the
Hamilton Springs; it is very vexing, some of our best are amongst them.
Wind, east.

Thursday, 28th March, Low Granite Range in Scrub. At 11 a.m. the horses
were brought back from the clay-pan. Two of the missing ones were found
about a mile after they started, making towards where they had camped
last night. I think that the other eight must be also in that direction;
we find that all the tracks have gone that way; I shall therefore move
down to-day to the south end of the swampy country, which I know they
cannot cross, and endeavour, if possible, to find them to-night. By 1
p.m. arrived at the end of the swamp; camped, and despatched Thring in
one direction and Sullivan in another to try and cut their tracks; at a
little before sunset Sullivan returned with three of the missing ones.
Five are still wanting. Wind, south-east.

Friday, 29th March, South End of Swamp in Scrub. At sunrise sent Thring
and Sullivan again to look for the missing horses; they arrived at 5 p.m.
with three of them. If we do not find the other two to-morrow, I shall
push on without them, and endeavour to pick them up on our return.

Saturday, 30th March, South End of Swamp in Scrub. Again sent Thring and
Sullivan in search of the two remaining horses; at about 11 a.m. they
returned with them. I shall now move up to our camp of 25th instant.
Camped at some rain water a little south of our former place, where there
is plenty of feed for the horses. Wind, south-east; clouds from
north-west.

Sunday, 31st March, Rain Water in Scrub. All day the sky has been
overcast with clouds from the north-west. Wind from south-east.

Monday, 1st April, Same Place. Started at 7.30 a.m.; course, 330 degrees.
At 1 p.m. we came upon a very pretty flat of beautiful grass, with water
in the middle of it; and, as the afternoon has every appearance of rain,
I have camped--to go on in the rain will only spoil our provisions. We
had scarcely got the packs off when it came on heavily, and lasted about
an hour: it then ceased until sundown, when it came on again, and
continued till 10.30 p.m.

Tuesday, 2nd April, Green Flat in Scrub. Started at 8.20 a.m. on same
course, and camped at 1.30 p.m. under a prominent rocky hill, which I
ascended and have named Mount Harris, after Peter G. Harris, Esquire, of
Adelaide. I obtained bearings of the different points all round. The last
seven miles was sandy soil, with spinifex and scrub, which was mostly
young cork-tree, and the broad-leafed mallee.

Wednesday, 3rd April, Mount Harris. We have put up a small cone of stones
on the top of this mount. Started at 8 a.m. for Anna's Reservoir. Arrived
at the creek about two miles south-south-east of it, and, finding it
running, camped amongst excellent feed. By keeping to west of my former
track I have found the country much opener; but nearly all day the
journey has been through spinifex. Wind from west.

Thursday, 4th April, The Wicksteed, Reynolds Range. Started at 7.40 a.m.
to cross the range, bearing to Mount Freeling 312 degrees. At 1.30 p.m.
crossed the range, and arrived at the creek, camping at the same place as
I did on my previous journey, and finding water and feed abundant. I have
named this creek the Woodforde, after Dr. Woodforde, of Adelaide. After
crossing the range, we found the bean-tree in blossom; it was
magnificent. I have obtained a specimen of it; also some beans, a number
of which were of a cream colour; we have roasted a few of them, and find
that they make very good coffee. Wind, south-east.

Friday, 5th April, The Woodforde, Reynolds Range. Started at 7.30 a.m.
Camped at 4.30 p.m. on the Hanson, which is now a running stream. About
five miles back we passed a freshly-built native worley. I observed a
peculiarity in it which I never noticed in any before--namely, that it
was constructed with greater care than usual. It was thatched with grass
down to the ground. Inside the worley there was a quantity of grass laid
regularly for a bed, on which some one had been lying. Round about the
front was collected a large quantity of firewood, as much as would have
done for us for a night. Latitude, 22 degrees 5 minutes 30 seconds,
bearing to Central Mount Stuart, 25 degrees. Wind, south-south-west.

Saturday, 6th April, The Hanson. Started at 8 a.m., on a course of 46
degrees 30 minutes, to the springs in the Hanson; this course led me
through about four miles of very thick mulga. After crossing the central
line we arrived on the creek and camped, below the springs, at 1.30 p.m.
Bearing to Central Mount Stuart, 251 degrees 20 minutes. Wind variable.

Sunday, 7th April, The Hanson, East-north-east of Centre. Day hot. Wind
variable, with a few clouds.

Monday, 8th April, The Hanson, East-north-east of Centre. Five of the
horses missing this morning. Started at 9.45, course 45 degrees; camped
on the Stirling at 3.50 p.m. Through all the day's journey the country
abounded in grass and water. Wind from south.

Tuesday, 9th April, The Stirling, Forster Range. Started at 7.30 a.m., to
cross Forster range on the same course. At 10.50 a.m. camped on north
side of it, on a large gum creek with water. I have named this the
Taylor, after John Taylor, Esquire, of the firm of Messrs. Elder,
Stirling, & Co., of Adelaide. This is a most beautiful place, a plain
four miles broad between two granite ranges, completely covered with
grass, and a gum creek winding through the centre. I made a short journey
to-day in consequence of having some of the horses lame, and some weak
through the effects of the green grass, and to-morrow's journey will be a
long one. Had I gone on to-day, they would in all probability be without
water, and would require to be tied up during the night. I shall now be
able to get through in one day, and keep them in good condition for the
unexplored country, which I expect to commence next Monday.

Wednesday, 10th April, The Taylor. Started at 7.25 a.m. on a course of 11
degrees 30 minutes for Mount Morphett; at 12.30 ascended the summit. On
the north side we had some difficulty in getting the horses down;
however, we managed without accident. Ran a creek down and found some
water; gave the horses a drink; still followed it until it was lost in a
grassy plain. Proceeded on to the next hills, passed through a gap, and
made for a creek on the north side, in which we found water, and camped
at 4 p.m.

Thursday, 11th April, North Side Mount Morphett, Crawford Range. Started
at 7.45 a.m. on a course of 10 degrees. The first four miles was over a
beautiful grassy plain, with mulga wood, not very thick; it then became
more sandy, and covered with gum, cork-trees, and other scrubs, which
continued within a mile of where we camped, in a small, but beautiful
grassed plain; no water. Latitude 20 degrees 38 minutes 33 seconds. Wind,
south-east.

Friday, 12th April, Grassy Plain. Started at 6.15 a.m., same course. At 1
p.m. arrived at the Bonney; it is now running--green feed abundant. As
some of the horses are still very lame, I will rest them to-morrow and
Sunday, and start into the unexplored country on Monday morning. Wind
from south-east; a few clouds from north-east.

Saturday, 13th April, The Bonney. Sent Thring down the creek to see what
its course is, and if the country gets more open; the men mending
saddle-bags, cleaning and repairing saddles, shoeing horses, etc. While I
and Woodforde were endeavouring to get a shot at some ducks on the long
water holes, a fish, which he describes as being about two feet long,
with dark spots on either side, came to the surface; he fired at it, but
was unsuccessful in killing it. A little before sundown Thring returned;
he gave a very bad account of the creek; it was a dry deep channel. Wind,
variable; cloudy.

Sunday, 14th April, The Bonney. Wind from every quarter, with clouds; a
few drops of rain fell about the middle of the day; after sundown much
lightning in the south-west.

Monday, 15th April, The Bonney. Cloudy; wind still variable. Mount
Fisher, bearing 120 degrees. Started at 7.15 a.m., bearing 290 degrees;
at 11.40 changed to 264 degrees, to some rising ground; at 12.45 p.m.,
after crossing stony hills, we crossed a gum creek on the west side, with
long reaches of water in it running north-west, which I supposed to be
the Bonney; but as there appeared to be more and larger gum-trees farther
on, I continued, to see if there were not another channel. Proceeded
three miles over low limestone rises, with small flats between, on which
was growing spinifex, and the gum-trees which I had seen--exactly the
same description of country from which I was forced to return through
want of water on my former journey from Mount Denison to north-west. I
therefore returned to the creek, which I find to be the Bonney, now much
smaller, but containing plenty of water--followed it down to
north-north-west for about one mile, and then camped. The water is in
long reaches, which I think are permanent.

Tuesday, 16th April, The Bonney. Still cloudy. Started at 8 a.m. on a
bearing of 380 degrees. At 11.15 changed to 40 degrees, with the
intention of cutting the McLaren. Camped at 3.40 p.m. Three miles from
our start the creek spreads itself over a large grassy plain, thickly
studded with gum-trees, covered with long grass, and a great number of
white ants' nests of all sizes and shapes, putting one in mind of walking
through a large cemetery. In many places it was very boggy. We followed
it for ten miles, but it still continued the same; I could not see more
than one hundred yards before me, the gum-trees, and sometimes a low
scrub, being so thick. Not seeing anything of the McLaren coming into the
plain, I changed my course to cut it and run it down, as I think that it
will form a large creek where they join. In three miles we got out of the
plain upon a red sandy soil, with spinifex, and scrubs of all kinds, in
some places very thick, and difficult to get the horses through. When we
were in the gum plain the atmosphere was so close and heavy, and the
ground so soft, that the sweat was running in streams from the horses;
and when we halted for a few minutes they were puffing and blowing as
though they had just come in from running a race. I continued the second
course for fourteen miles, but saw nothing of the McLaren; it must have
joined the plain before I left it. Thus ends the Bonney and the McLaren.
We passed over several quartz and ironstone ranges of low hills crossing
our course, and camped under a high one, without water. Wind south-east.
Cloudy.

Wednesday, 17th April,* (* The Journal of this Expedition, as published
by the Royal Geographical Society, commences here.) Quartz Hill, West
Mount Blyth. Started at 7.25 a.m. on a bearing of 70 degrees. We again
passed quartz hills running as yesterday; the spinifex still continuing,
with a little grass, until we came within a mile of the hills in the
Murchison range; finding some water, I camped, and gave the horses the
rest of the day to recruit. Last night after sundown, and during the
night, we had a few slight showers of rain, and a great deal of thunder
and lightning, mostly from south-west. About 11 to-day the clouds all
cleared away. About a mile before camping, we observed the ground covered
with numerous native tracks; also that a number of the gum-trees were
stripped of their bark all round.

Thursday, 18th April, West Mount Blyth. Started at 7.40 a.m., same
bearing, across the Murchison range, in which we found great difficulty.
On the north-east side of Mount Blyth we found a large gum creek of
permanent water, and camped. I have named this Ann Creek. I then rode to
the highest point of the range, taking Thring with me, to see if there is
any rising ground to north-west by which I may cross the gum plain. I
could see no rise, nothing but a line of dark-green wood on the horizon.
We had great difficulty in getting to the top, the rocks being so
precipitous. In coming down the eastern side we were gratified by the
sight of a beautiful waterfall, upwards of one hundred feet high, over
columns of basaltic rock, its form, two sides of a triangle, the water
coming over the angle. Wind, south-east.

Friday, 19th April, Ann Creek. Started at 7.45 a.m., on a course of 324
degrees, towards Mount Samuel. After sundown arrived at Goodiar Creek;
one of the horses done up; had to leave him a little distance back; he is
unwell. On leaving the Murchison range we crossed a number of quartz
reefs and hills running east and south-west. Wind, south-south-east.

Saturday, 20th April, Goodiar Creek. Three horses missing this morning,
in consequence of the scarcity of feed. The horse left behind last night
has been brought in; he looks very bad indeed. About 11 a.m. the other
horses were found, brought in, and saddled, and we proceeded on a
north-north-west course for Bishop Creek, but found the sick horse too
ill to proceed further than Tennant Creek, where we camped, there being
plenty of water and feed. Two natives were seen by Masters this morning
when in search of the horses--he could not get them to come near him.
Wind, south-west.

Sunday, 21st April, Tennant Creek. Wind from south-west; a few clouds
from east.

Monday, 22nd April, Tennant Creek. Started at 7.30 a.m., course 21
degrees, for Bishop Creek, and at twelve miles made it. I find that two
of the horses are so weak that they are unable to go any further without
giving in, I have therefore camped, giving them the remainder of the day
to recruit. Native fires are smoking all around us, but at some distance
off. Wind, east.

Tuesday, 23rd April, Bishop Creek. It is late before we can get a start
to-day, in consequence of one of the horses concealing himself in the
creek. He is an unkind brute, we have much trouble with him in that
respect; he is constantly hiding himself somewhere or other. Started at
9.30 a.m., on a course of 17 degrees, to cross Short range. Found plenty
of water in Phillips Creek; the grass on its banks, and on the plains
where it empties itself, is splendid, two feet and a half long, fit for
the scythe to go into, and an abundant crop of hay could be obtained. We
then crossed the range a little north of where I passed before, and found
some slight difficulty. After descending, we struck a small creek which
supplies Kekwick Ponds, and is a tributary to Hayward Creek; found plenty
of water and camped at 3 p.m. Feed abundant. Wind, south-east.

Wednesday, 24th April, Hayward Creek. Started at 7.40 a.m.; course 17
degrees. At 9.30 changed to 14 degrees 30 minutes west of north, and at
12.30 arrived at Attack Creek; camped at the same place that I did on my
former journey. Tracks of natives about, but we have seen none of them. I
kept about a mile to the west of my former track, and found the country
much more open. The banks of both creeks for two or three miles are
splendidly covered with grass, in some places over the horses' heads.
Four of the horses are ill, and looking very bad indeed. Wind,
south-west.

Thursday, 25th April, Attack Creek. Started at 7.50 a.m., on a course of
294 degrees, to the top of the range, which I have named Whittington
Range, after William S. Whittington, Esquire, of Adelaide. At six miles
reached the top. At 9.50 changed to north-west, and at 11.30 struck a
large gum creek running east, with large water holes in it. At about two
hundred yards crossed it again, running to the west, and shortly
afterwards crossed it again, running to the east. I have called it
Morphett Creek, after the Honourable John Morphett, Chief Secretary. We
then ascended another portion of the range, and continued along a spur on
our course. This range presents quite a new feature, in having gums
growing on the top and all round it; it is composed of masses of
ironstone, granite, sand, and limestone, and in some places white marble.
Thinking that the creek we had passed might break through a low part of
the range, which I could see to the north-west, at ten miles I changed to
west, and crossed to the other range, but found the dip of the country to
the south. We could find no water; traced the creek to the south-east for
two miles, found some water and camped. The range is very rough and
stony, covered with spinifex; but the creeks are beautifully grassed.
Native smoke to east. This is one of the sources of Morphett Creek, and
flows to the east; it is as large, if not larger, than Attack Creek, and,
in all probability, contains water holes quite as fine to the eastward.
Latitude, 18 degrees 50 minutes 40 seconds.

Friday, 26th April, Morphett Creek. At 8 a.m. started on a course of 300
degrees to cross the north-west part of the range. Camped upon a plain of
the same description as John Plain, that I met with on my former journey
to the north-east of Bishop Creek, a large open plain covered with grass,
and with only a few bushes on it. The journey to-day has been very rough
and stony. Not a drop of water have we passed to-day, nor is there the
appearance of any on before us. I shall be compelled to fall back
to-morrow to the water of last night. Four of the horses, I am afraid,
will not be able to get there. I must try more to the north, and
endeavour to get quit of the plains, and get amongst the creeks. There is
no hope of success on this course. Latitude, 18 degrees 38 minutes. Wind,
east.

Saturday, 27th April, Grassy Plains. Started at 7.10 a.m., course 110
degrees, to the other side of the plain. At three miles came upon a small
creek running towards the north; I followed it down to the north. At
three miles came upon a fine large creek, coming from the south-east,
with plenty of water. Returned to the party, took them down to the large
creek on north course, and at three miles camped. Two of the horses are
nearly done up. Wind, south-east. Latitude, 18 degrees 35 minutes 20
seconds.

Sunday, 28th April, Tomkinson Creek. Sent Thring down to examine and see
how the creek runs. I have named it after S. Tomkinson, Esquire, Manager
of the Bank of Australasia, at Adelaide. We have found many new plants
and flowers, also some trees, one of which grows to a considerable size,
the largest being about a foot in diameter. The fruit is about the size
and colour, and has the appearance of plums; the bark is of a grey
colour; the foliage oval, and dark-green. Another is more of a bush, and
has a very peculiar appearance; the seed vessel is about the size of an
orange, but more pointed. When ripe it opens into four divisions, which
look exactly like honeycomb inside, and in which the seeds are
contained; they are about the size of a nut, the outside being very
hard. The natives roast and eat them. The leaves resemble the mulberry,
and are of a downy light-green. We have obtained a few of the seeds of
it. The bean-tree does not seem to grow up here. Mr. Kekwick, in looking
for plants this morning, discovered one which very much resembles wheat
in straw (which is very tough), ear, and seed. It grows two feet high.
The seed is small, but very much like wheat both in shape and colour. At
about 3 p.m. Thring returned, having run the creek out into a large
grassy plain. The course of this creek is west-north-west for about nine
miles; it then turns to west, and empties itself into the plain. There
is plenty of water about, but where it empties itself it becomes quite
dry. The native companion, the emu, and the sacred ibis are on this
creek. The country is splendidly grassed. We have got to the north side
of the Whittington range. I shall have to leave my two done-up horses
here, and will get them when I return. The hills and rocks are of the
same description as the first part of the range. Wind, south. Sun hot,
but the nights and mornings are very cold.

Monday, 29th April, Tomkinson Creek. Had a late start this morning in
consequence of my having to take a lunar observation. Started at 10.30
a.m. At 2.10 p.m. reached the top of a high hill; from this we could see
a gum creek. Started at 2.30 to examine it; found water, and camped at 4.
I have named the hill Mount Primrose, after John Primrose, Esquire, of
North Adelaide. This water will last us six or eight weeks. The country
passed to-day has been mostly stony rises of the same description as the
other parts of the range. The valleys have a light sandy soil, nearly all
with spinifex and scrub. The view from the top of Mount Primrose is not
extensive, except to the west and south-west, which appears to be thick
wood or scrub. Near the top we met with the Eucalyptus Dumosa. Wind,
south-east. Latitude, 18 degrees 25 minutes.

Tuesday, 30th April, Carruthers Creek. The creek in which we are now
camped I have named Carruthers Creek, after John Carruthers, Esquire, of
North Adelaide. Started at 8.50 a.m. At 1.50 p.m. found a creek running
from the range, with a splendid hole of permanent water situated under a
cliff, where the creek leaves the range; it is very deep, with a rocky
bottom. From the top of the range the country seems to be very thick,
which I am afraid is scrub; no high hills visible. To the north of this
the range appears to cease; I wish it had continued for another sixty
miles. The country passed to-day has been stony rises coming from the
range, very rough and rocky indeed. My horses' shoes are nearly all gone;
I am obliged to let some go without--they have felt the last four rough
days very much. Spinifex, scrub, and stunted gums all the day, with
occasionally a few tufts of grass; this is very poor country indeed.
Smoke of native fires still in south-east. The hills of the same
formation as those we first came upon in entering the ranges from Attack
Creek. I have named this creek Hunter Creek, after Mr. Hunter, of Messrs.
Hunter, Stevenson, and Co., of Adelaide. Camped. The horses seem very
tired. Wind, east. Latitude, 18 degrees 17 minutes.

Wednesday, 1st May, Hunter Creek. Started at 8 a.m., course, 305 degrees.
At 8.45 crossed the Hunter going south-west; it came round again and
continued crossing our course thirteen times in nine miles, after which
it was lost in a large grassy and gum plain. At 5.15 camped. The plain in
which the creek loses itself bears south-west; the banks are beautifully
grassed, but about a mile on either side the soil is sandy, with spinifex
and scrub, which continued for nine miles; we then entered upon a scrub
and grassy plain. Here I noticed a new and very beautiful tree--in some
instances a foot in diameter--with drooping branches. Its bark was grey
and rough, and it had a small dark-green leaf, shaped like a butterfly's
wing. Not finding a creek, nor the least indication of a watercourse, and
the scrub becoming very thick, I changed to north, to see if I could find
any water; but at three miles we lost the gums, the new tree taking their
place, and becoming very thick scrub with plenty of grass, but no signs
of a watercourse. I again changed to east in the hope of cutting one in
that direction. At one mile and a half again came upon small gums; and at
three miles, seeing neither creek nor any hope of getting water, camped.
The horses very tired. Wind light from west-north-west. Latitude, 18
degrees 3 minutes 19 seconds.

Thursday, 2nd May, Large Scrubby and Grassy Plain. Started at 10 a.m. in
consequence of some of the horses having strayed a long way to the east
during the night; course, 143 degrees 30 minutes, back to Hunter Creek. I
have taken a different course to see if there is any creek that supplies
this plain with water. For about nine miles we passed over a splendidly
grassed plain, with gum-trees, the new tree, and a number of all sorts of
bushes. One part for about three miles is subject to inundation, and the
Eucalyptus Dumosa grows thickly on it. We then passed over about two
miles of spinifex and grass, and again entered the grassy plain, which
continued to Hunter Creek. During the whole day we have not seen the
shadow of a creek or watercourse. If there had been any sign of a
watercourse, or if I could have seen any rising ground near our course, I
would have gone on another day. I sent Wall to the top of the highest
tree to see if there was anything within view; he could see nothing but
the same description of plain. If my horses can travel to-morrow, I will
try a course to the north, and run down the creek, to see if there is one
that will lead me through this plain. If I could get to some rising
ground, I think I should be all right; but there is none visible except
the end of the range, which is lost sight of to the north-east. Wind
again south-east, with a few clouds. Latitude, 18 degrees 13 minutes 40
seconds.

Friday, 3rd May, Hunter Creek. Started at 8.40 a.m.; course, north. At
11.15 (nine miles), came upon a creek; bed dry and sandy; searched for
water, and, at three quarters of a mile to east, found a nice hole;
watered the horses and proceeded on the same course--starting at 12. At
3.20 p.m. changed to 20 degrees north of east; the first ten miles were
over a plain of gums covered with grass two feet long; we had then six
miles of spinifex, and a thick scrub of dwarf lancewood, as tough as
whalebone. After that we entered upon another gum plain, also splendidly
grassed, which continued for four miles, when the gums suddenly ceased,
and it became a large open plain to north, as far as I could see. Seeing
no appearance of water, I changed my course to 30 degrees north of east,
to some high gums; and, at one mile, not finding any, I camped without
it. This seems now to be a change of country; there is no telling when or
where I may get the next water on this course, so that I shall be
compelled to go towards the range to-morrow to get some, and have a long
day's journey to the new country. The wind has been from east all day.
Latitude, 17 degrees 56 minutes 40 seconds.

Saturday, 4th May, Sturt Plains. Started at 7.15 a.m., course east, to
find water. At 3.20 p.m. came upon a little creek and found a small
quantity of water, which we gave to the horses. Started again at 9 p.m.,
course south-east, following the creek to find more; at a mile and a half
found water which will do for us until Monday morning. I proceeded to the
top of the range to obtain a view of the country round, but was
disappointed in its height; from the plain it appeared higher than it
really is. This range I have named Ashburton Range, after Lord Ashburton,
President of the Royal Geographical Society. The point upon which I am at
present is about three miles east of our camp; the view from south to
north-west is over a wooded plain; from north-west to north is a large
open plain with scarcely a tree upon it. On leaving our last night's
camp, we passed over three miles of the plain, which is subject to
inundation. There are numerous nasty holes in it, into which the horses
were constantly stumbling. It is covered with splendid grass, and is as
fine a country as I have ever crossed. These plains I have named Sturt
Plains, after the venerable father of Australian exploration and my
respected commander of the expedition in 1845. Ashburton range is
composed of sandstone and ironstone, granite, and a little quartz; it is
very rough and broken. Native tracks about here. Wind, south-east. This
creek I have named Watson Creek, after Mr. Watson, formerly of Clare.

Sunday, 5th May, Watson Creek, Ashburton Range. Sent Thring to the north
along the range to see if there is permanent water; at eight miles he
returned, having found plenty. One large hole is about a mile from here;
in another creek it is apparently permanent, having a rocky bed. A flight
of pelicans over head to-day; they seem to have come from the north-west,
which course I will try to-morrow. Wind, south-east. Latitude, 17 degrees
58 minutes 40 seconds.

Monday, 6th May, Watson Creek, Ashburton Range. Started at 8.20 a.m.,
course 300 degrees, to cross Sturt Plain. At eleven miles arrived at the
hill which I saw from Ashburton range. It turned out to be the banks of
what was once a fresh-water lake; the water-wash is quite distinct. It
had small iron and limestone gravel, with sand and a great number of
shells worn by the sun and atmosphere to the thinness of paper, plainly
indicating that it is many years since the water had left them. Judging
from the water-marks, the lake must have been about twelve feet deep in
the plain. The eucalyptus is growing here. We then proceeded over another
open part of it, for about two miles, when the dwarf eucalypti again
commenced, and continued until we camped at twenty-one miles; the horses
quite worn out. This has been the hardest and most fatiguing day's work
we have had since starting from Chambers Creek; for, from the time we
left in the morning until we camped, we have had nothing but a succession
of rotten ground, with large deep holes and cracks in it, caused at a
former period by water, into which the poor horses have been constantly
falling the whole day, running the risk of breaking their legs and our
necks, the grass being so long and thick that they could not possibly see
them before they were into them. I had a very severe fall into one of
these holes; my horse came right over and rolled nearly on top of me. I
was fortunate enough to escape with little injury. Some of the shells
resemble the cockle shell, but are much longer, many of them being three
or four inches long; the others are of the shape of periwinkles, but six
times as large. Both sorts are scattered over the plain, which is
completely matted with grass. The soil is a dark rich alluvial, and
judging from the cracks and holes, some of which are of considerable
depth, they are splendid plains, but not a drop of surface water could we
see upon them, nor a single bird to indicate that there is any. It was my
intention at starting to have gone on thirty miles, but I find it quite
impossible for the horses to do more; it would be madness to take them
another day over such a country, when from the highest tree we can see no
change. If I were to go another day and be without water, I should never
be able to get one of the horses back, and in all probability should lose
the lives of the whole party. If I could see the least chance of finding
water, or a termination of the plain, I would proceed and risk
everything. I see there is no hope of my reaching the river by this
course. I believe this gum plain to be a continuation of the one I met
with beyond the Centre, and that it may continue to the banks of the
Victoria. The features of the country are nearly the same. The absence of
all birds has a bad appearance. Day very hot. Wind, south-east. Latitude,
17 degrees 49 minutes.

Tuesday, 7th May, Sturt Plains. Before sunrise this morning I sent Wall
up a tree to see if any hills or rising grounds would be visible by
refraction. To the west, with a powerful telescope he can just see the
top of rising ground. As the grass is now quite dry, the horses feel the
want of water very much; many of them are looking wretched, and I hardly
think will be able to reach it. However reluctant, I must go back for the
safety of the party. At 3 p.m. arrived at the creek which Thring found
about one mile to the north of my former camp, with the loss of only one
horse; we had to leave him a short distance behind, he would not move a
step further, although during a great part of the journey he had been
carrying little or nothing. This water will last two months at least;
feed good. It is inside the first ironstone rise in Ashburton range, in a
gum creek which empties itself into the plains. This creek I have named
Hawker Creek, after James Hawker, Esquire, of her Majesty's Customs at
Port Adelaide. The day has been very hot. Wind, south-east. Latitude, 17
degrees 58 minutes.

Wednesday, 8th May, Hawker Creek, Ashburton Range. I have sent Masters
back to bring up the horse we left behind. Sturt Plains have been at one
time the bed of a large fresh-water lake; our journey of the 6th instant
was over the middle of it, and we were not at the end of it when I was
forced to return; the same rotten ground and shells continued, although
we had got amongst the eucalypti. I shall give the horses a rest to-day,
and to-morrow will take the best of them (those that I had out on my
former journey), and endeavour to cross the plain to the rising ground
seen yesterday morning; I shall take Thring and Woodforde, with seven
horses and one week's provisions. I may be fortunate enough to find some
water, but from the appearance of the country I have little hope. I
shall, however, leave nothing untried to accomplish the object of the
expedition. In the morning the horse we left behind could not be found;
sent Masters and Sullivan in search of him; in the afternoon they
returned with him looking miserable. He had wandered away beyond the
other camp.

Thursday, 9th May, Hawker Creek, Ashburton Range. Started at 7 a.m., with
Thring and Woodforde, and seven horses, following our tracks through the
rotten ground to the first eucalypti, for about twelve miles, as it made
it lighter for the horses, the tracks being beaten to that place. Changed
our course to 282 degrees, still journeying over Sturt Plains; at
twenty-seven miles arrived at the end of the portion of them that had
been subject to inundation, but there are still too many holes to be
pleasant. I certainly never did see a more splendid country for grass; in
many places for miles it is above the horses' knees. We entered upon red
sandy soil, with spinifex and grass, from which we changed our bearing.
The country became thickly studded with eucalypti, in one or two places
rather open, but generally thick. After the twenty-seven miles we again
met with the new small-leafed tree, the broad-leafed mallee, the
eucalypti, and many other scrubs. At sundown we camped; distance,
thirty-three miles, but not a drop of water have we seen the whole day,
or the least indication of its proximity. I hope to-morrow we may be more
fortunate, and find some. Wind, south.

Friday, 10th May, Sturt Plains. This morning there are a few birds about.
Started at 8.15 a.m., same course; at 10.30 arrived on first top of
rising ground seen from the camp of 7th instant, which turns out to be
red sand hills covered with thick scrub. Changed our course to
north-west, and at 11.15 arrived at the highest point; the view is very
discouraging--nothing to be seen all round but sand hills of the same
description, their course north-north-east, and south to west. No high
hills or range to be seen through the telescope. We can see a long
distance, apparently all sand hills with scrub and stunted gums on them.
The first ridge is about two hundred feet above Sturt Plains, but further
to the west they are much lower, and become seemingly red sandy
undulating table land; but further to the west they are much lower. There
is no hope of reaching the Victoria on this course. I would have gone on
further to-day had I seen the least chance of obtaining water to-night;
but during the greater part of yesterday and to-day we have met with no
birds that frequent country where water is. Both yesterday and to-day
have been excessively hot, and the country very heavy. From this point I
can see twenty-five miles without anything like a change. To go on now
with such a prospect, and such heavy country before me, would only be
sacrificing our horses and our own lives without a hope of success--the
horses having already come forty-five miles without a drop of water, and
over as heavy a country as was ever travelled on. I have therefore, with
reluctance, made up my mind to return to the camp and try it again
further north, where I may have a chance of rounding the sand hills; the
dip of them from here seems to be south-south-west. Turned back, and at
eighteen miles camped on Sturt Plains, where there is green grass for the
horses. Wind, south.

Saturday, 11th May, Sturt Plains. At dawn of day started for the camp;
arrived at 2 p.m. It was fortunate I did not go on further, for some of
the horses were scarcely able to reach it; a few more hours and I should
have lost half of them. The day has been so hot that it has nearly
knocked them all up. Found the rest of the party all right at the camp.
We had a job to keep the horses from injuring themselves by drinking too
much water. I gave them a little three separate times, tied them up for
twenty minutes, and then gave them a good drink, and drove them off to
feed. They took a few mouthfuls of grass, and were back again almost
immediately, and continued to do so nearly all the afternoon. They drank
an immense quantity. Wind, south.

Sunday, 12th May, Hawker Creek, Ashburton Range. My old horses that were
out with me before look very well this morning, but the others, whose
first trip of privation this has been, are looking very bad indeed. They
could not have gone another night without water; it has pulled them down
terribly. Yesterday, while Masters was looking for the horses, he saw
what appeared to him to be a piece of wood stuck upon a tree, about two
feet and a half long, sharp at both ends, broad at the bottom, and shaped
like a canoe. Having pulled it down, he found it to be hollow. On the top
of it were placed a number of pieces of bark, and the whole bound firmly
round with grass cord. He undid it, and found the skull and bones of a
child within. Mr. Kekwick brought it to me this morning for my
inspection. It certainly is the finest piece of workmanship I have ever
seen executed by natives. It is about twelve inches deep and ten wide,
tapering off at the ends. Small lines are cut along both sides of it. It
has been cut out of a solid piece of wood, with some sharp instrument. It
is exactly the model of a canoe. I told him to do it up again, and
replace it as it was found. If it is here when I return, I will endeavour
to take it to Adelaide with me. Wind, variable. A few clouds about.

Monday, 13th May, Hawker Creek, Ashburton Range. Started at 8 a.m.,
course 360 degrees. At five miles crossed the large gum-tree creek, with
water, that Thring found; proceeded along the side of Sturt Plains. At
ten miles ascended the north point of Ashburton range; descended, and the
country became red sand with spinifex, gum-tree, the new tree, and other
shrubs very thick; at fifteen miles, gained the top of another stony
rise; followed three creeks down in search of water; found a little, but
not sufficient for us; followed it still further down, leading us to the
south for about six miles, but could find no more. I thought it best to
return for water to the large creek, which I have named Ferguson Creek,
after Peter Ferguson, Esquire, of Gawler Town. From the top of the range
the view is limited. To the north and north-east are stony rises, at
about nine miles distant; from north to west are Sturt Plains, in some
places wooded; to the north they are open for a very long distance; the
country in the hills is bad, but in the plains is beautiful. I am afraid,
from the view I have of the country to the north, that I shall again meet
with the same description of sand hills that I came upon on my last
western course. Wind east-south-east, blowing strong. Latitude, 17
degrees 53 minutes 20 seconds.

Tuesday, 14th May, Ferguson Creek. Started at 8.30 a.m., on a north
course, to the place I turned back from yesterday; arrived at noon;
changed course to 345 degrees. Started again at 12.20. At 1 p.m. crossed
a gum creek that has the appearance of water. At 1.40 changed course to
260 degrees, and came upon two large water holes, apparently very deep,
situated in the rocks--they are seemingly permanent. Camped. I named this
creek Lawson Creek, after Dr. Lawson, J.P., of Port Lincoln. A number of
natives have been camped about them. We found another canoe, of the same
description as the one in which the bones of the child were found--it is
broken and burned, and seems to have been used as a vessel for holding
water. Wind south-east, blowing strong. Mornings and evenings very cold.
Latitude, 17 degrees 43 minutes 30 seconds.

Wednesday, 15th May, Lawson Creek. Started at 8.10 a.m.; went a mile west
to clear the stones; changed course to 340 degrees. At 2.45 p.m. changed
again to 45 degrees. Camped at 4.15. The first twelve miles is poor
country, being on the top of stony rises, with eucalypti, grass, and
scrubs. After descending from the rises, we crossed a wooded plain,
subject to inundation; no water. The trees are very thick indeed--they
are the eucalyptus, the Eucalyptus Dumosa, the small-leaved tree, another
small-leaved tree much resembling the hawthorn, spreading out into many
branches from the root; it rises to upwards of twenty feet in height. We
have also seen three other new shrubs, but there were no seeds on them.
After crossing the plain we got upon red sandy rises, very thick with
scrub and trees of the same description. We continued on this course
until 2.45 p.m.; then, as there is an open plain in sight, with rising
ground upon it to north-east, and as this scrubby ridge seems to
continue, without the least appearance of water, I have changed to
north-east. Crossed the plain, which is alluvial soil, covered with
grass, but very dry. At 4.15 camped on north-east side, without water. I
would have gone on to the rise, but I feel so ill that I am unable to sit
any longer in the saddle. I have been suffering for the last three days
from a severe pain in the chest. Wind, east. Latitude, 17 degrees 16
minutes 20 seconds.

Thursday, 16th May, Sturt Plains. Sent Thring to see if there is a creek
or a sign of water under the rise. At 8.20 a.m. he returned, having found
no water. It is a low sandy rise, covered with a dense scrub. Started at
8.20 a.m.; course, east. At three miles I was forced to return; the scrub
is so dense that it is impossible to get through. Came back two miles;
changed to 20 degrees west of south to get out of it. At two miles gained
the plain, then changed to the east of south at 10.45. At 2 p.m. there is
no hope of a creek or water. Changed to south-west. At two miles and a
half struck our tracks and proceeded to Lawson Creek. We found the open
parts of the plain black alluvial soil so rotten and cracked, that the
horses were sinking over their knees; this continued for six miles. It is
covered with long grass and polygonum; also a few eucalypti scattered
over it. The scrub we were compelled to return from was the thickest I
have ever had to contend with. The horses would not face it. They turned
about in every direction, and we were in danger of losing them. In two or
three yards they were quite out of sight. In the short distance we
penetrated it has torn our hands, faces, clothes, and, what is of more
consequence, our saddle-bags, all to pieces. It consists of scrub of
every kind, which is as thick as a hedge. Had we gone further into it we
should have lost everything off the horses. No signs of water. From south
to west, north and north-east nothing visible but Sturt Plains, with a
few sand rises having scrub on them, which terminate the spurs of the
stony rises. They are a complete barrier between me and the Victoria. I
should think that water could be easily obtained at a moderate depth in
many places on the plains. If I had plenty of provisions I would try to
make it by that way. The only course that I can now try is to the
north-east or east, to round the dense scrub and plains. At sundown
arrived at Lawson Creek. The horses, owing to the dryness of the grass,
drank a great quantity of water; they are falling off very much. Wind,
south-east.

Friday, 17th May, Sturt Plains. I must remain here to-day to mend
saddle-bags, etc. I have sent Thring to north-east to see if the stony
rises continue in that direction. He has returned and gives a very poor
account of the country. He crossed them in about six miles, and again
came upon the plain that we were on yesterday, extending from north-east
to south. Nothing but plains. To the north is the dense scrub, thus
forming a complete stop to further progress. From here I fear it is a
hopeless case either to reach Victoria or the Gulf. The plains and forest
are as great a barrier as if there had been an inland sea or a wall built
round. I shall rest the horses till Monday, and will then try a course to
the north-west, and another to north-east. I have not the least hope of
succeeding without wells, and I have not sufficient provisions to enable
me to remain and dig them. It is a great disappointment to be so near,
and yet through want of water to be unable to attain the desired end.
Wind, south-east.

Saturday, 18th May, Lawson Creek. Resting horses, etc. Wind, south-east.

Sunday, 19th May, Lawson Creek. Wind, south-east.

Monday, 20th May, Lawson Creek. Started at 7.25 a.m., course 45 degrees,
with Thring, Woodforde, and seven horses. The first four miles was over
the stony rises; the next three, sandy table-land, with spinifex,
eucalyptus, and scrub. Crossed part of Sturt Plains, open and covered
with grass. Five miles of it were very heavy travelling-ground, very
rotten, and full of holes and cracks. At about thirty miles camped on the
plains. We have seen no birds, nor any living thing, except kites and
numerous grasshoppers, which are in myriads on the plains. From this
place to the east, and as far as south-south-west, there is no rising
ground within range of vision--nothing but an immense open grassy plain.
The absence of birds proclaims it to be destitute of water. We have not
seen a drop, not a creek, nor a watercourse during the whole day's
journey. To-morrow I shall again try to get through the scrub. On leaving
the camp this morning, I instructed Kekwick to move the party about three
miles down the creek to another water hole, the feed not being good.
Wind, east.

Tuesday, 21st May, Sturt Plains, East. Started at 7.10 a.m. Passed
through a very thick scrub seven miles in extent. We again entered on
another portion of the open plains at ten miles from our last night's
camp. Nothing to be seen on the horizon all round but plains. Changed to
300 degrees, to where I saw some pigeons fly. At two miles came across
their feeding-ground; skirted the scrub until we cut our tracks. No
appearance of water. This is again a continuation of the open portion of
Sturt Plains; they appear to be of immense extent, with occasional strips
of dense forest and scrub. We had seven miles of it this morning as thick
as ever I went through; it has scratched and torn us all to pieces. At my
furthest on the open plain. I saw that it was hopeless to proceed, for
from the west to north, and round to south-south-west, there is nothing
to be seen but immense open plains covered with grass, subject to
inundation, having an occasional low bush upon them. I think with the aid
of the telescope I must have seen at least sixty miles; there is not the
least appearance of rising ground, watercourse, or smoke of natives in
any direction. The sun is extremely hot on the plain. Having no hope of
finding water this morning, I left Woodforde with the pack and spare
horses where we camped last night, as the heat and rough journey of
yesterday have tired them a great deal; so much so, that I fear some of
them will not be able to get back to water. Returned to where I had left
him, and followed our tracks back to the open plain. After sundown camped
among some scrub. Wind, south-east.

Wednesday, 22nd May, Sturt Plains. After sunset we saw a number of
turkeys flying towards the stony rises where our main camp is; they
appear to come from the north-west. Upwards of fifty passed over in twos
and threes; and this morning we observed them going back again. Two of
the horses which had been short hobbled walked off during the night,
following our tracks. Saddled and followed, overtaking them in three
miles and a half, standing under the shade of a tree. Unhobbled and drove
them on before us. At 12 o'clock arrived at Lawson Creek. Had great
difficulty in preventing the horses from drinking too much, and, as there
are other holes down the creek, I gave them a little at a time at each.
Found that Kekwick had moved with the party. Followed them, and at three
miles and a half west-south-west arrived at their camp, and allowed the
horses to drink as much as they chose. Poor brutes! they have had very
hard work, eighty miles over the heaviest country, under a burning sun,
without a drop of water. Three of them were those I had on my former
journeys; I could depend upon them; the rest were the best I could pick
from the other lot. They have all stood the journey very well, but could
not have done another day without water. Natives seem to have been about
this water lately, but we have not seen one since leaving our spitting
friend on the Hugh. Wind, east.

Thursday, 23rd May, Lawson Creek. Started 7.45 a.m., course 315 degrees,
with Thring, Woodforde, and seven fresh horses. At fourteen miles came
across a splendid reach of water, about one hundred and fifty yards wide,
but how long I do not know, as we could not see the end of it. It is a
splendid sheet of water, and is certainly the gem of Sturt Plains. I have
decided at once on returning, and bringing the party up to it, as it must
be carefully examined, for it may be the source of the Camfield, or some
river that may lead me through. On approaching it I saw a large flock of
pelicans, which leads me to think that there may be a lake in its
vicinity. There are mussels and periwinkles in it, and, judging from the
shells on the banks, the natives must consume a large quantity. The
gum-trees round it are not very large. The first ten miles of that part
of the plain travelled over to-day is full of large deep holes and
cracks, black alluvial soil covered with grass, with young gum-trees
thicker as we approached the water. This I have named Newcastle Water,
after his Grace the Duke of Newcastle, Secretary for the Colonies. Duck,
native companion, white crane, and sacred ibis abound here. Returned to
bring the party up to-morrow. Wind, south-east.

Friday, 24th May, Lawson Creek. Started at 8 a.m. for Newcastle Water;
arrived at noon. Camped. Sent Kekwick to north-east and Thring to west to
see the length of it; I have had the depth tried. It is about six feet
deep ten yards from the bank, and in the middle seventeen feet. I should
say it was permanent. Thring found it still the same at three miles west.
Kekwick returned after following it for four miles. At two miles there is
a break in it. At four miles it is more of a creek coming from
north-east. Gum-trees much larger. Woodforde succeeded in catching four
fish about ten inches long, something resembling the whiting. I had one
cooked for tea; the skin was as tough as a piece of leather, but the
inside was really good, as fine a fish as I have ever eaten. To-morrow I
shall follow the water to the west; its bed is limestone. Wind,
south-east, with a few clouds. Latitude, 17 degrees 36 minutes 40
seconds.

Saturday, 25th May, Newcastle Water, Sturt Plains. Started at 7.50 a.m.
and followed the water nine miles round. It still continued, but became a
chain of ponds. As I could see some rising ground north-north-east about
four miles distant, I camped the party and took Thring with me to see
what the country was before us. At four miles we found that the first
part of the rise was stony, but on the top it was sandy table-land,
covered with thick scrub. The view is obstructed to the east-north-east
to north by it; but to the north-west and west there is an appearance of
rising ground, thickly wooded, about twenty miles off. Wind, west.
Latitude, 17 degrees 30 minutes 30 seconds.

Sunday, 26th May, Newcastle Water, Sturt Plains. This morning we were
visited by seven natives, tall, powerfully-made fellows. At first they
seemed inclined for mischief, making all manner of gestures and shaking
their boomerangs, waddies, etc. We made friendly signs to them, inviting
them to come nearer; they gradually approached, and Kekwick and Lawrence
got quite close to them; in a short time they appeared to be quite
friendly. I felt alarmed for the safety of J. Woodforde (who had gone
down the water in search of ducks, and in the direction from which they
had come), and endeavoured to make them friends by giving them pieces of
handkerchiefs, etc. During the time we were talking with them I heard the
distant report of his gun; at the same time Thring and Masters returned
from collecting the horses that were missing. I told them to remain until
the natives were gone, as I wished to keep them as long as possible to
give Woodforde a chance of coming up before they left us; shortly
afterwards they went off apparently quite friendly. Sent Thring and Wall
to round up the horses which were close at hand, and while they were
doing so the natives again returned, running quite close up to the camp
and setting fire to the grass. It was now evident they meant mischief. I
think they must have seen or heard Woodforde, and have lit the grass in
order to engage our attention from him. I felt very much inclined to fire
upon them, but desisted, as I feared they would revenge themselves on him
in their retreat. They did very little injury by their fire, which we
succeeded in putting out. By signs I ordered them to be off, and after
much bother they left us, setting fire to the grass as they went along. I
now ordered Thring and Wall to go with all speed to protect Woodforde. In
about twenty minutes he came into the camp. After leaving us they had
attacked him, throwing several boomerangs and waddies at him; he had only
one barrel of his gun loaded with shot; they all spread out and
surrounded him, gradually approaching from all sides. One fellow got
within five yards of him, and was in the act of aiming his boomerang at
him. Seeing it was useless to withhold any longer, while the black was in
the act of throwing he gave him the contents of his gun in his face, and
made for the camp. In a short time Thring and Wall returned at full
speed; they had passed where he was, and hearing the report of his gun,
made for the place, overtook the blacks, gave chase and made them drop
the powder-flask and ducks (which Woodforde had laid down before firing
when they attacked him); knowing them to be his, they gave up the chase
to look for him, but seeing nothing of him, and two of the natives
supporting one apparently wounded, they returned to the camp, where they
saw him all safe, relating his adventure, his shot-belt still missing. I
sent Thring and him to look for it, and to bring up the missing horses
which they had seen. Wind variable. Cloudy.

Monday, 27th May, Newcastle Water, Sturt Plains. Started at 8.10 a.m.,
course 335 degrees. At 10.20 changed to north; at 1.20 p.m. changed to 90
degrees; and at one mile found water; gave the horses some, and proceeded
north-north-east; at 3.40 changed to 90 degrees to some gums: at one mile
and a half camped. The gums turn out to be thick wood. I went
north-north-west this morning, with the expectation of meeting with
water, or rather a chain of ponds; at four miles, I could see nothing of
them; and, as we were getting into a very thick scrub of lancewood, I
changed to north; and at ten miles on that course, still seeing nothing
of them, I changed to east; at one mile came upon them, found water, and
followed them; their course now, 20 degrees; at one mile found another
pond; in a short time, lost the bed of them in a thick wooded plain.
Found a native path running nearly in my course; followed it, thinking it
would lead me to some other water, but in a few miles it became
invisible. I continued on the same course for nine miles, and found
myself on Sturt Plains, with belts of thick wood and scrub; to the north,
nothing visible but open plains; to north-east, apparently thick wood or
scrub; to north-west and west, apparently scrubby sand hills. The ponds
seem to drain this portion of the plains. Changed to east, to what seemed
to be large gum-trees, thinking there might be a creek; arriving there, I
found them to be stunted gums on the edge of the plain. There is no hope
of succeeding in this quarter. Camped without water. Wind, east.
Latitude, 17 degrees 12 minutes 30 seconds.

Tuesday, 28th May, Sturt Plains, North. Fourteen of the horses missing
this morning before sunrise. From the highest tree nothing is to be seen
from east to north and north-west but immense open grassy plains, without
a tree on them; no hope of water. I must go back to the ponds and try
again to the westward. Did not find the horses until 9.30 a.m., and
started at 10. I observed very large flocks of pigeons coming in clouds
from the plains in every direction towards the ponds. Some time
afterwards we saw them coming back and flying away into the plains as far
as the eye can reach, apparently to feed. Arrived at the water at 1.30
p.m. Wind, east-north-east.

Wednesday, 29th May, Chain of Ponds. Started at 7.20 a.m. with Thring,
Woodforde, and Wall, and nine horses, to follow a native track, which is
leading to the westward. At 9.20 made the track; its course,
west-north-west. At twenty-eight miles camped without water. The track
led us into very thick wood and scrub, and at five miles became
invisible. I still continued on the same bearing through the scrub. We
have again met with the mulga--a little different from what we have seen
before, growing very straight, from thirty to forty feet high, the bark
stringy, the leaf much larger and thicker. Amongst it is the hedge-tree.
We had seven miles of it very dense, when we again met with an open
plain. At three miles entered another dense wood and scrub, like that
passed through in the morning. To-day's journey has been over plains of
grass, through forest and scrub, without water. In the last five miles we
passed through a little spinifex, and the soil is becoming sandy. Wind,
south.

Thursday, 30th May, Sturt Plains. As I can see no hope of water, I will
leave Woodforde and Wall with the horses, take Thring with me, and
proceed ten miles, to see if there will be a change in that distance.
Went into a terrible thick wood and scrub for eleven miles and a half,
without the least sign of a change--the scrub, in fact, becoming more
dense; it is scarcely penetrable. I sent Thring up one of the tallest
trees. Nothing to be seen but a fearfully dense wood and scrub all round.
Again I am forced to retreat through want of water. The last five miles
of the eleven the soil is becoming very sandy, with spinifex and a little
grass. It is impossible to say in which way the country dips, for, in
forty-five miles travelled over, we have not seen the least sign of a
watershed, it is so level. Returned to where I left the others, followed
our tracks back, and at eleven miles camped. Horses nearly done up with
heavy travelling and the heat of the sun, which is excessive. It is very
vexing and dispiriting to be forced back with only a little more than one
hundred miles between Mr. Gregory's last camp on the Camfield and me. If
I could have found water near the end of this journey, I think I could
have forced the rest. It is very galling to be turned back after trying
so many times. Wind, east.

Friday, 31st May, Sturt Plains. Not having sufficient tethers for all the
horses, we had to short hobble two, and tie their heads to their hobbles;
and, in the morning, they were gone. I suppose they must have broken
their hobbles or fastenings; they will most likely make on to our outward
tracks. I have sent Thring and Woodforde to follow them up, while Wall
and I, with the other horses, proceed on our way to the camp. In two
hours they made the tracks before us, and I then pushed on as hard as I
could get the horses to go; being very anxious about the safety of the
party--for, on the first day that I left them, at about seven miles, we
passed fourteen or fifteen natives going in the direction of their camp;
I also observed, this morning, that they had been running our tracks both
backwards and forwards. At three o'clock we arrived, and found all safe;
they have not been visited by them, although I observed the prints of
their feet in our tracks, a short distance from the camp. It was as much
as some of our horses could do to reach the camp. The day has been
excessively hot; wind from north-north-east, with clouds. Latitude, 17
degrees 7 minutes.

Saturday, 1st June, Chain of Ponds. I must rest the horses to-day and
to-morrow, for they look very miserable; our longitude is 133 degrees 40
minutes 45 seconds. Before leaving the Ponds I shall try once more to the
westward--starting from a point three miles west of my first camp on
them. To try from this, for the Gulf of Carpentaria, I believe to be
hopeless, for the plain seems to be without end and without water. If I
could see the least sign of a hill, or hope of finding water, I would try
it; but there is none--if there is a passage it must be to the south of
this. Wind variable, with clouds.

Sunday, 2nd June, Chain of Ponds. The day has again been very hot. Wind
variable.

Monday, 3rd June, Chain of Ponds. Started back to the commencement of the
Chain of Ponds, and camped. During the day the sky has been overcast with
heavy clouds. Wind, south-east.

Tuesday, 4th June, Chain of Ponds. Last night one of the horses was
drowned in going down to drink at the water hole. He went into a boggy
place, got his hind foot fastened in his hobbles, from which he could not
extricate himself, and was drowned before we could save him. This is
another great loss, for he was a good pack-horse, and was one that I
intended taking on my next trip to the westward. At about 8 p.m. it began
to rain, and continued the whole night, coming from the east and
east-south-east. It still continues without any sign of a break. The
ground has become so soft that when walking we sink up to the ankle, and
the horses can scarcely move in it. At sundown there is no appearance of
a change. It has rained without intermission the whole of last night and
to-day. I do not know what effect this will have on my further progress,
for now it is impossible to travel. The horses in feeding are already
sinking above their knees. Wind and rain from east and east-south-east.

Wednesday, 5th June, Chain of Ponds. There is a little sign of a break in
the clouds this morning. The rain has continued the whole night. Ground
very soft; it has become about the thickness of cream. The horses can
scarcely get about to feed. Sundown: It has been showery all day; sky
overcast; clouds and rain from same direction, south-east. In the
afternoon some natives made their appearance at about six hundred yards'
distance. As the rain had damped the cartridges I caused the rifles to be
fired off in that direction; and, as the bullets struck the trees close
to them they thought it best to retreat as fast as possible, yelling as
they went.

Thursday, 6th June, Chain of Ponds. During the night it has been stormy,
with showers of rain, and is still the same this morning. Sundown: Still
stormy, with a few drops of rain. Wind, east.

Friday, 7th June, Chain of Ponds. During the night the rain ceased, and
this morning is quite bright. Ground so soft that it is impossible to
travel. Latitude, 17 degrees 35 minutes 25 seconds. Sent Thring some
miles to the west, to see in what state the country is, if fit for us to
proceed, and if he can see any water that I could move the party to, for
I do not like this place. If more rain falls it will lock us in all
together--neither do I like leaving the party with so many natives about.
At one o'clock he returned. The ground was so heavy that he had to turn
at five miles. He could see no water, but a number of native tracks going
to and coming from the west. I shall be obliged to leave the party here,
and on Monday try another trip to the west. If I find water I shall
return and take them to it. The day has been clear, but at sundown it is
again cloudy. Clouds from north-west. Wind from east.

Saturday, 8th June, Chain of Ponds. This morning it has again cleared
off, and there is every appearance of fine weather. If it hold this way I
shall be able to travel on Monday. Sundown: A few clouds. Wind,
south-east.

Sunday, 9th June, Chain of Ponds. The day has again been fine. Wind,
still south-east.

Monday, 10th June, Chain of Ponds. Started at 7.55 a.m., course 275
degrees, with Thring, Woodforde, and Wall, nine horses, and fourteen
days' provisions. The first five miles were over a grassy plain, with
stunted gum and other trees. It was very soft, the horses sinking up to
their knees. We met with a little rain water at three miles, where the
soil became sandy; continued to be more so as we advanced, with lancewood
and other scrubs growing upon it. At fourteen miles gained the top of a
sand rise, which seems to be the termination of the sand hills that I
turned back from on my west course south of this. From here the country
seems to be a dense forest and scrub; no rising ground visible. Camped at
5 p.m., distance thirty-two miles. The whole journey from the sand hills
has been through a dense forest of scrubs of all kinds--hedge-tree, gum,
mulga, lancewood, etc. We have had great difficulty in forcing the horses
through it so far; they are very tired. It is the thickest scrub I have
yet been in. Ground very soft; heavy travelling, with the exception of
the last five miles, where little rain seems to have fallen. I am afraid
this will be another hopeless journey. I fully expected to have got water
to-night from the recent rains, but there is not a drop. The country is
such that the surface cannot retain it, were it to fall in much larger
quantities. I shall try a little further on to-morrow. I had a hole dug,
to see if any rain had fallen, and found that it had penetrated two feet
below the surface, below which it is quite dry. Wind, east.

Tuesday, 11th June, Dense Forest and Scrub. Leaving Woodforde, Wall, and
the pack-horses, I took Thring with me, and proceeded on the same course
to see if I could get through the horrid forest and scrub, or meet with a
change of country, or find some water. At two miles we came upon some
grass again, which continued, and at another mile the forest became much
more open and splendidly grassed, which again revived my sinking hopes;
but alas, it only lasted about two miles, when we again entered the
forest thicker than ever. At eleven miles it became so dense that it was
nearly impenetrable. The horses would not face it; when forced, they made
a rush through, tearing everything we had on, and wounding us severely by
running against the dead timber (which was as sharp as a lancet) and
through the branches. I saw that it was hopeless to force through any
further. Not a drop of water have we seen, although the ground is quite
moist--the horses sinking above the fetlock. The soil is red and sandy;
the mulga from thirty to forty feet high and very straight; the bark has
a stringy appearance. There is a great quantity of it lying dead on the
ground, which causes travelling to become very difficult. I therefore
returned to where I left Woodforde and Wall, and came back ten miles on
yesterday's journey, and camped. This morning, about 5.30, we observed a
comet bearing 110 degrees; length of tail, 10 degrees, and 10 degrees
above the horizon. Wind, south-east.

Wednesday, 12th June, Western Dense Forest and Scrub. Proceeded to camp
and found all well. This is the third long journey by which I have tried
to make the Victoria in this latitude, but have been driven back every
time by the same description of country and the want of water. There is
not the least appearance of rising ground, or a change in the
country--nothing but the same dismal, dreary forest throughout; it may in
all probability continue to Mr. Gregory's last camp on the Camfield. My
farthest point has been within a hundred miles of it. I would have
proceeded further, but my horses are unable to do it; they look as if
they had done a month's excessive work, from their feet being so dry, the
forest so thick, and the want of water. Thus end my hopes of reaching the
Victoria in this latitude, which is a very great disappointment. I should
have dug wells if my party had been larger, and I had had the means of
conveying water to those engaged in sinking the wells. I think I could
accomplish it in that way; but by doing so, I should have to divide the
party into three, (one sinking, one carrying water, and one at the camp),
which would be too small a number where the natives appear to be so
hostile. I have not the least doubt that water could be obtained at a
moderate depth, near the end of my journeys, amongst the long thick
timber, which seems to be the lowest part of the country. I had no idea
of meeting with such an impediment as the plains and heavy scrub have
proved to be. For a telegraphic communication I should think that three
or four wells would overcome this difficulty and the want of water, and
the forest could be penetrated by cutting a line through and burning it.
In all probability there is water to be found nearer than this in the
Camfield, Mr. Gregory's last camp, somewhere about its sources, which
might be thirty miles nearer. Wind, south-east. Country drying up very
fast.

Thursday, 13th June, Chain of Ponds. To-day I shall move the camp to the
easternmost part of Newcastle Water, and now that rain has come from the
east, I shall try if I can cross Sturt Plains, and endeavour to reach the
Gulf of Carpentaria. My provisions are now getting very short. We are
reduced to four pounds of flour and one pound of dried meat per man per
week, which is beginning to show the effects of starvation upon some of
them; but I can leave nothing untried where there is the least shadow of
a chance of gaining the desired object. Started at 9.40 a.m. At three
miles and a half passed our first camp of Newcastle Water. At eight miles
and a half camped at the last water to the eastward. The ground is firmer
than I expected, travelling good. The large part of the water is reduced
two inches since 24th ultimo. The late rains seem to have no effect on
it. Wind, south-east.

Friday, 14th June, East End of Newcastle Water. Started with Thring,
Woodforde, and Wall, with one month's provisions and ten horses, at 7.45
a.m.; course, 60 degrees. At two miles crossed our former tracks, on the
top of the sandy table land, and after leaving it we again got on the
open plains, black alluvial soil, covered with grass, with deep holes and
cracks into which the horses were continually falling on their noses, and
running the risk of breaking our necks. These plains have swallowed up
every drop of rain that has fallen. The extent of the plain is seven
miles. We then entered a thick wooded country, of the same description as
the western forest, being equally thick, if not thicker, and as difficult
to penetrate. This continued for thirteen miles, when we met with another
small plain about half a mile wide, but opening out wider to north-west
and south. Not a drop of water have we seen since leaving Newcastle
Water, a distance of about thirty miles, except a little rain water about
three miles east of it. The plains are quite dry, scarcely showing that
rain has fallen. Camped. The horses have had a hard day's work and are
very tired. I wish I could have found water for them to-night. Latitude,
17 degrees 26 minutes 20 seconds. Wind, south-east.

Saturday 15th June, North-east Small Plains, Sturt Plains. Started at
7.30 a.m.; course, 60 degrees, through another ten miles of very thick
forest, the thickest we have yet seen. At eleven miles came again upon
the large open grassy plain, at the point where I turned on the 21st
ultimo. I expected to have found some rain water here, this being the
only place in all the plain I have seen that is likely to retain it. Sent
Thring and Woodforde in different directions, while I proceeded in
another, to see if we could find any, but not a drop could we see. It has
been all swallowed up by the ground, which is again dry and dusty. It
must take an immense quantity to saturate it, and leave any on the
surface; and if that were to be the case, the country would become so
soft it would be quite impassable. I am again forced to turn; it is quite
hopeless to attempt it any farther. It would be sacrificing our horses,
and, perhaps, our own lives, without the least prospect of attaining our
end. If I could see rising ground, however small, or a change in the
country to justify my risking everything, I would do so in a moment. I
only wish there was. I have tried my horses to their utmost. Even my old
horses that are inured to hardship are unable to be longer than three
days without water, owing to the heat of the sun, the dryness of the
feed, and the softness of the country. We saw a few cockatoos and
pigeons. There might be water within a short distance, but none can we
see or find; for on my course 20 degrees west of north I passed within
two miles of Newcastle Water, where the main camp is now, but could not
see it. It would require a long time to examine this country for water.
There are so many clumps of trees, and strips of scrub on the plain,
where water might be, that it would take upwards of twelve months to
examine them all. At sundown camped fifteen miles from the main camp.
Horses look very bad. It has been very heavy travelling, over rotten
ground, and tearing through thick wood and scrub, which has skinned our
legs from the knees to the ankles and caused no little pain. Wind,
variable.

Sunday, 16th June, Sturt Plains East. Proceeded to the camp, where I
found all well. No natives had been near them. This is very disheartening
work. I shall proceed to the south, and try once more to round that
horrid thick western forest; it is now my only hope; if that fail I shall
have to return. I am doubtful of the water in Ashburton range, if no rain
has fallen there; those hills are the last of the rising ground within
range of vision, which ends in about latitude 17 degrees 14 minutes. From
south-south-east round the compass to south-south-west nothing but dense
forest and Sturt Plains. Wind, south-east.

Monday, 17th June, Newcastle Water East. Returned to the Lawson and
camped. Little rain seemed to have fallen there. I kept a little to west
of my former tracks to see the nature of the large open plain. It is
completely matted with grass, having large deep holes and cracks, and is
as dry as if no rain had fallen for months. Wind, south-east.

Tuesday, 18th June, Lawson Creek. Proceeded to Hunter Creek. Tracks of
natives upon ours to Hawker Creek. Light winds, variable.

Wednesday, 19th June, Hawker Creek. Although the water holes in this
creek are full from recent rains, the water is very hard, evidently
showing it must come from a spring in the hills. Proceeded to the Hunter
along the foot of the hills, and at nine miles crossed the large gum
creek, where I watered the horses on my north course; this I have named
Powell Creek, after J.W. Powell, Esquire, of Clare. At twenty miles
crossed another gum creek, which I have named Gleeson Creek, after E.B.
Gleeson, Esquire, J.P., of Clare. Camped on the Hunter. Between this and
Hawker Creek we crossed eleven gum creeks with water in them. The country
passed over is not so good, being close to the hills: it is scrubby, and
generally covered with spinifex. Wind, south-east.

Thursday, 20th June, Hunter Creek. Three horses missing; could not be
found until too late to reach the other water to-night. Wind, calm.

Friday, 21st June, Hunter Creek. Proceeded to the water under Mount
Primrose, over stony hills, the highest of which I have named Mount
Shillinglaw, after ---- Shillinglaw, Esquire, F.R.C.S., of Melbourne, who
kindly presented me with Flinders' Charts of North Australia. The gum
creek on which we are now camped I have named Carruthers Creek, after
John Carruthers, Esquire, of Adelaide. Calm.

Saturday, 22nd June, Carruthers Creek. Proceeded to Tomkinson Creek,
where I left the two horses; I will there rest the horses a day, and have
those shod which I intend to take with me. The last two days have been
over very stony country, which has made some of the horses quite lame. I
am now running short of shoes. We can see nothing of the two horses about
our old camp. Light wind from north-east, with a few clouds. Very hot in
the middle of the day; evenings and mornings cold.

Sunday, 23rd June, Tomkinson Creek. Sent Thring and Woodforde down the
creek, and Masters up into the open plain, to see if they could find the
horses on their tracks. In the afternoon they returned unsuccessful,
except Masters, who had seen their tracks when the ground was boggy.
Recent tracks of natives were also seen. If they have not been frightened
away, they will not be far off. I have instructed Sullivan to follow
their tracks, and try to find them during my absence. Wind, north-east,
with a few clouds. The sun is very hot in the middle of the day.

Monday, 24th June, Tomkinson Creek. Started with Thring, Masters, and
Lawrence, and ten horses, with fourteen days' provisions, at 7.40 a.m.;
course, 270 degrees east. We crossed the plain and the creek several
times. At 12.20, fifteen miles, ascended a stony rise, and saw that the
creek emptied itself into an open grassy plain, about two miles north of
us. Proceeded on the same course over a gum plain covered with grass for
five miles. The country then became sandy soil, slightly undulating, with
ironstone, gravel, spinifex, gums, and occasionally a little scrub, which
continued throughout the day. Camped without water. Very little feed for
the horses, it being nearly all spinifex. Distance, twenty-eight miles.
Wind, west; a few clouds.

Tuesday, 25th June, Spinifex and Gum Plain West. Started at 7.40 a.m. on
the same course, 270 degrees. Camped at twenty-seven miles. The country
travelled through to-day is bad--red sandy light soil, covered with
spinifex, slightly undulating, and having iron gravel upon it. Scarcely a
blade of grass to be seen. Some gum-trees, and a low scrub of different
sorts. I seem to have got to the south of the dense forest, but into a
poorer country. Not a drop of water or a watercourse have we seen since
we left Tomkinson Creek. We have crossed two or three low rises of
ironstone gravel. Not having the dense forest to tear through has induced
me to go on all day in the hope of meeting with a change, but at the end
of the day there seems as little likelihood as when we first came upon
it, and it may continue to the river. I am again forced to return
disappointed. There is no hope of making the river now; it must be done
from Newcastle Water with wells. I wish that I had twelve months'
provisions and convenience for carrying water, I should then be enabled
to do it. Wind, east.

Wednesday, 26th June, Spinifex and Gum Plain. Started at 7 a.m. back
towards Tomkinson Creek. At dusk found some water on the small plain into
which the creek empties itself. Camped. Distance travelled to-day, forty
miles. One of the horses completely done up. I am fortunate in finding
this water, for another night without it and I should have lost some of
them. I am also glad we had a cool day--only two hours' heat. The horses
have travelled one hundred miles without water, and the country being
sandy, made it very heavy walking for them. Wind, east.

Thursday, 27th June, Tomkinson Creek. Started for the camp, and arrived
at noon. Sullivan had gone after the horses, and lost himself for three
days and two nights. Not making his appearance the first night, Kekwick
sent Woodforde in search of him from south-east to north. Not returning
the second night, Kekwick and Woodforde went out in another direction to
try if they could cut his tracks, but were again unsuccessful. At about 3
a.m. he came into the camp perfectly bewildered, and did not seem to
recognise anyone. From what we can learn from him he must have gone to
the south instead of the east, where the tracks of the two horses were
seen. On the first night he came close to the camp--saw the other horses
feeding, but could not find them. He can give no account of where he went
the next day and night; on the third day he cut my outward tracks to the
west, and the horse brought him to the camp. I observed his horse's
tracks upon ours this morning, about ten miles down the creek, and could
not imagine how they came there. Woodforde found the two horses he went
in search of within three miles of the camp--they had not left the creek.
The cream-coloured one had improved very much; but Reformer still looks
miserable--I think he must be ill. Wind, north, with a few clouds coming
from the same direction.

Friday, 28th June, Tomkinson Creek. Shoeing horses and preparing for
another start. I shall try once more to make the Gulf of Carpentaria from
this. There may be a chance of my being able to round Sturt Plains to the
east or north-east. Wind, varying from south-east to north.

Saturday, 29th June, Tomkinson Creek. Shoeing horses, etc. Wind,
south-east. Clouds all gone.

Sunday, 30th June, Tomkinson Creek. Wind, north-east.

Monday, 1st July, Tomkinson Creek. Started at 8.10 a.m., course 54
degrees, with Thring, Woodforde, and Masters. At 11.20 (eleven miles),
top of a high hill, which I named Mount Hawker, after the Honourable
George C. Hawker, Speaker of the House of Assembly, S.A. At 12.45, four
miles, struck a large creek; its course a little east of north, which I
have named McKinlay Creek, after John McKinlay, Esquire. The first part
of the journey was over stony undulations, gradually rising until we
reached the top of Mount Hawker, the view from which was not very
extensive on our course, being intercepted by stony spurs of the range
nearly the same height, about eight hundred feet, and very rocky and
precipitous. They are composed of sandstone, quartz, iron, limestone, and
hard white flinty rocks. The sandstone predominates. We descended with
great difficulty, crossed McKinlay Creek, and at five miles ascended
another high hill, which I have named Mount Hall, after the Honourable
George Hall, M.L.C. From this our view is most extensive, over a complete
sea of white grassy plains. At about fifteen or twenty miles south-east
are the terminations of other spurs of this range; beyond them nothing is
visible on the horizon but white grassy plains. To the east and
north-east the same. To the north apparently a strip of dense scrub and
forest, which seems to end about north-east, beyond which, in the far
distance, we can see the large grassy plain I turned back from on the
21st of May and 15th of June. No rising ground visible except the hills
of Ashburton range to north-west and south-east. Descended towards the
plains over stony rises, with gum-tree, lancewood, and other scrub and
spinifex. At five miles reached the plain. It is of the same description
as the other parts I have been over. No appearance of water. It is
hopeless to proceed further; it will only be rendering my return more
difficult, by reducing the strength of my horses, without the slightest
hope of success. All hope of gaining the Gulf without wells is now gone.
I have therefore turned back to a small plain (four miles), searched
round it, and in one of the small creeks found a little rain water, at
which I have camped. Wind, south.

Tuesday, 2nd July, Loveday Creek. This creek I have named Loveday Creek,
after R.J. Loveday, Esquire, Lithographer to the South Australian
Government. Returned towards the camp. On reaching McKinlay Creek I was
informed by Woodforde that Masters had remained behind, about six miles
back, and had not yet come up. This is against my strict orders which are
that no one shall leave the party without informing me, that I may halt
and wait for them. I have sent Thring back to one of the hills to fire
off a gun, and see if he is to be seen, as I have left my outward tracks
to avoid crossing Mount Hall--and the tracks are very difficult to be
seen over such stony country. I am afraid that he is lost. In an hour and
a half, Thring returned; he can see nothing of him. He cut our former
tracks, but can see nothing of his on them. My conjectures, I fear, are
too true. If he has missed the tracks, it is a thousand chances to one if
he is ever found again. To track a single horse is impossible. I
proceeded towards Mount Hawker, and camped on my outward tracks, at a
remarkable gorge that we had come through. Sent Thring back to the top of
Mount Hall to raise a smoke, to remain there some time, and see if he
comes up; if not, he is to proceed to our last night's camp, there to
remain all night, in case he should go there--while I and Woodforde
raised another smoke on top of Mount Hawker. A little after 2 p.m. Thring
returned with him. He found him on a hill near Mount Hall, looking for
the tracks. He was quite bewildered, and in a great state of excitement.
I am most thankful that he is found. The account that he gives is, that
his horse slipped the reins out of his hand, and that he was unable to
catch him for some time, and when he did so, he was unable to find our
tracks, or to track his own horse back, and he became quite confused. He
seems to be most thankful for his narrow escape. As it is too late to
reach the camp, I shall remain here to-night. Wind, west.

Wednesday, 3rd July, Under Mount Hawker. Proceeded to the camp on the
Tomkinson. Found all right, with the exception of one of the horses
(Reformer), which cannot be found. He is one of the two that I left here
formerly, and was looking so ill when I found him. He was last seen on
Monday night, when he looked miserable. I have sent three men in search
of him. Wind, variable.

Thursday, 4th July, The Tomkinson. Started at 8.20 a.m., course 300
degrees, with Woodforde, Thring, and Masters, ten horses, and a month's
provisions, to try once more to make the Victoria. Between my first and
last attempts, I may succeed. I am very unwilling to return without
trying all that is in my power. At three miles we left the plains, and
proceeded over stony rises for two miles. The country then became sandy,
with gum, spinifex, and lancewood scrub, not difficult to get through.
There is no grass. At twenty-five miles came to a little, and, as I am
not sure of coming upon any more soon, I camped. We have seen no water
since leaving the creek. Latitude, 18 degrees 25 minutes 40 seconds.
Wind, south-east.

Friday, 5th July, Spinifex and Gum Plains. Started at 7.50 a.m., course
360 degrees, to find water. At 9.10 (five miles), struck a creek with
water; followed it down, course 285 degrees, and at eight miles camped on
the last water. The banks in places have good feed upon them, but there
is a great deal of spinifex and scrub. The creek is getting narrower,
and, as the horses had but little to eat last night, I shall give them
the remainder of the day here, for there is no telling when they will get
another good feed. Day exceedingly hot, horses covered with sweat. This I
have named Burke Creek, after my brother explorer, Richard O'Hara Burke,
Esquire, of Melbourne. On camping I saw a remarkable bird fly up; I sent
Woodforde to try and shoot him, which he did. It was of a dark-brown
colour, and spotted like the landrail; the tail feathers were nine in
number, and twelve inches long. I have had it skinned, and will endeavour
to take it to Adelaide. Thring, Woodforde, and Masters cooked the body,
and ate it. They had scarcely finished, when, in a moment, they were
seized with violent vomiting, but in a few minutes they were all right
again. Wind, calm. Latitude, 18 degrees 19 minutes 30 seconds.

Saturday, 6th July, Burke Creek. Started at 7.45 a.m., same course, to
follow the creek (285 degrees). At three miles it was lost in a grassy
gum plain; changed to 300 degrees. On this course the plain continued for
three miles; it then became sandy soil, with spinifex, gums, and scrub.
Crossed a low sand hill at fourteen miles; descended into another low
grassy plain subject to inundation, which, I suppose, receives Hunter
Creek. It continued for two miles, at the end of which we again ascended
a sandy rise, on the top of which the country became a sandy table land,
and continued so the rest of the day's journey. Camped without water, and
with very little grass. The table land was spinifex, gums, and scrub, in
some places very difficult to get through. Distance, thirty miles. Wind,
south-east. Latitude, 18 degrees 7 minutes 5 seconds. At 7 p.m. I
observed the comet, 5 degrees above the horizon, bearing 15 degrees west
of north, the nucleus more hazy, and the tail much longer. Calm.

Sunday, 7th July, West Sandy Table Land. Started at 8 a.m. on the same
course. At 3 p.m. we got into dense scrub, and, as I could see some
distance on before, being on one of the slight undulations, I felt there
was not the slightest hope of obtaining water; there was no change, no
rising ground visible. It would be hopeless to continue such sandy
country, as it can never hold water on the surface. We dug five feet, in
one of the small plains, but came to the clay without finding water, or
even moisture. There is not a mouthful of grass for the horses to eat;
the whole of the journey, with the exception of the small grassy plains,
is spinifex, gums, and scrub. I shall have to retreat to the last plain
we passed through to get feed for the horses, which are looking very bad.
The travelling has been heavy tearing through thick scrub, which in some
places has been burned: this makes it very rough for them. I must now
give up all hope of reaching the Victoria, and am unwillingly forced to
return, my horses being nearly worn out. Wind, variable. Distance,
twenty-five miles.

Monday, 8th July, Small Grass Plain in Scrub. Started at break of day and
continued until 4.30 p.m. Meeting with a little grass, camped; some of
the horses unable to go further. Wind, south.

Tuesday, 9th July, Sandy Table Land. Started at sunrise and arrived at
Burke Creek. At 11 a.m. turned the horses out to feed for two hours, and
proceeded up the creek to where I first struck it. Camped. At a little
more than a mile down the creek from here, there is a course of concrete
ironstone running across it, which forms a large pond of water nearly a
mile in length, apparently deep and permanent. Wind, west.

Wednesday, 10th July, Burke Creek. Shortly after sunrise proceeded toward
the main camp, and arrived there at 3 p.m. Found all well. The natives
have been about. They attacked Wall while in search of the missing horse;
he and his horse narrowly escaped being hit by their boomerangs. The
missing horse cannot be found. I suppose that he has crept into some
bushes and died; for, the night before he was missed, he left the other
horses and came to the camp fire; he appeared to be very stupid, and for
some time they could not get him away; when they did so, he went off
reeling. Wind, south-west.

Thursday, 11th July, Tomkinson Creek. Shoeing horses, and repairing
saddles and bags to carry our provisions back. We have now run out of
everything for that purpose, and are obliged to make all sorts of shifts.
The two tarpaulins that I brought from Mr. Chambers's station for mending
the bags, are all used up some time ago, and nearly all the spare bags;
the sewing-twine has been used long since, and we are obliged to make
some from old bags. We are all nearly naked, the scrub has been so severe
on our clothes; one can scarcely tell the original colour of a single
garment, everything is so patched. Our boots are also gone. It is with
great reluctance that I am forced to return without a further trial. I
should like to go back, and try from Newcastle Water, but my provisions
will not allow me. I started with thirty weeks' supply at seven pounds of
flour per week, and have now been out twenty-six, and it will take me ten
weeks before I can reach the first station. The men are also failing, and
showing the effects of short rations. I only wish I had sufficient to
carry me over until the rain will fall in next March. I think I should be
able to make both the Victoria and the Gulf. I had no idea when starting
that the hills would terminate so soon in such extensive level country,
without water, or I should have tried to make the river, and see what the
country was, when I first saw the rising grounds from Mount Primrose,
which are the sand and iron undulations passed over on my southernmost
western journey. Before I went to Newcastle Water they completely
deceived me; for from the top of the mount they had the appearance of a
high range, which I was glad to see, thinking that if the range I was
then following up should cease, or if I could not find a way into the
river further north, I would be sure to get in by that distant range,
which caused me to leave the Newcastle Water country sooner than I should
otherwise have done; and now I have not provisions to take me back again.
From what I have seen of the country to the west and south of Newcastle
Water, I am of opinion that it would be no use trying again to make the
river, for I believe no water can be obtained by sinking. To the west and
north-west of Newcastle Water the country is apparently lower, and I
think that water could be obtained at a moderate depth. It is the
shortest distance between the waters; but the greatest difficulty would
be in getting through the dense forest and scrub, but that, I should
think, could be overcome. It certainly is a great disappointment to me
not to be able to get through, but I believe I have left nothing untried
that has been in my power. I have tried to make the Gulf and river, both
before rain fell, and immediately after it had fallen; but the results
were the same, UNSUCCESSFUL. Even after the rain I could not get a step
further than before it. I shall commence my homeward journey to-morrow
morning. Wind, south. The horses have had a severe trial from the long
journeys they have made, and the great hardships and privations they have
undergone. On my last journey they were one hundred and six hours without
water.

On Friday, July 12th, Mr. Stuart quitted Tomkinson Creek to return to
Adelaide, and on the following Friday reached Ann Creek on the north side
of the Murchison range. On the 30th, the party proceeded across the
Centre, and camped south-west of it, on the Hanson. The nights now became
very cold, and there was usually white frost on the grass, and ice in the
buckets every morning. On August 6th they camped under Brinkley Bluff,
and remained there until the 8th. The Hamilton was reached on the 23rd,
and here the natives again showed some hostility, contenting themselves,
however, with yelling and howling, and endeavouring to set fire to the
grass, in which they were happily unsuccessful. On Saturday, August 31st,
they arrived at Mr. Levi's station, where all of them "were overjoyed at
once more seeing the face of a white man." They were received with great
kindness and attention. After remaining there three days they proceeded
by way of Louden Spa, William Springs, Paisley Ponds, and Hamilton
Springs to Chambers Creek, where they arrived on September 7th and
remained until the 10th.

The last entry in Mr. Stuart's journal is as follows:

Sunday, 15th September, Moolooloo. I shall leave to-morrow for Port
Augusta, and proceed by steamer for Adelaide, leaving the party to be
brought into town by Mr. Kekwick.

I cannot close my Journal without expressing my warmest thanks to my
second in command, and my other companions; they have been brave, and
have vied with each other in performing their duties in such a manner as
to make me at all times feel confident that my orders were carried out to
the best of their ability, and to my entire satisfaction; and I also beg
to tender my best thanks to the promoters (Messrs. Chambers and Finke)
and the Government, for the handsome manner I was fitted out.

JOHN McDOUALL STUART,

Leader of the Expedition.


JOURNAL OF MR. STUART'S SUCCESSFUL EXPEDITION ACROSS THE CONTINENT OF
AUSTRALIA. FROM DECEMBER, 1861, TO DECEMBER, 1862.

Mr. Stuart made his public entry into Adelaide on Monday, 23rd September,
and reported himself to the authorities. Almost at the same time the
Victorian Government obtained their first traces of the survivors of the
ill-fated expedition under Burke and Wills.* (* The news of their death
reached Melbourne on November 2nd.) The South Australian Government had
such confidence in Mr. Stuart that, on his expressing his readiness to
make another attempt to cross the continent, they at once closed with his
offer, and in less than a month (on October 21st) the new expedition
started from Adelaide to proceed to Chambers Creek, and get everything in
order there for a final start. Mr. Stuart accompanied them for a few
miles to see that everything went on well, when, one of the horses
becoming restive, he advanced with the intention of cutting the rope
which was choking the animal; the horse reared and struck him on the
temple with its fore foot, knocking him down and rendering him
insensible. The brute then sprang forward and placed one of his hind feet
on Mr. Stuart's right hand, and, rearing again, dislocated two joints of
his first finger, tearing the flesh and nail from it, and injuring the
bone to such an extent that amputation of the finger was at first thought
unavoidable. By careful treatment, however, it was unnecessary to resort
to such a course, and in five weeks the leader was able to start to
overtake his party, some of whom were to remain at Moolooloo until he
joined them.

In no way discouraged either by his own unlucky accident and previous
want of success, or by the melancholy end of his brother explorers, Burke
and Wills, Mr. Stuart arrived at Moolooloo on Friday, December 20th, and
at Finniss Springs on the 29th. The names of the party were as follows:

John McDouall Stuart, Leader of the Expedition.
William Kekwick, Second Officer.
F.W. Thring, Third Officer.
W.P. Auld, Assistant.
Stephen King.
John Billiatt.
James Frew.
Heath Nash.
John McGorrerey, Shoeing Smith.
J.W. Waterhouse, Naturalist to the Expedition.

Besides these, there were at starting, Woodforde and Jeffries; but at
Finniss Springs, the latter struck one of his companions, and, on being
called to account by his leader, refused to go any further. As to the
former, when quitting Mr. Levi's station on January 21st, it was
arranged, in order to lighten the weak horses, that the great-coats of
the party should be left, but Woodforde objected to this, and said he
would not go unless he had his great-coat with him. Mr. Stuart had very
properly decided not to take any man who refused to obey orders, and he
therefore started without him. The next day Woodforde rejoined the party
near Milne Springs, but did not accompany them many days longer; for on
February 3rd, shortly after starting, he asked McGorrerey to hold his gun
while he returned to get something he had left behind at the previous
night's camp. About an hour afterwards, McGorrerey discovered a piece of
folded-up paper on the nipple of the gun, and on examination this proved
to be an insolent note, addressed to his leader, stating that he had gone
back, taking with him a horse, saddle, bridle, tether-rope, and sundry
other things not belonging to him. Mr. Stuart had been much dissatisfied
with his conduct for some days, and had made up his mind to send him
back, believing that he was doing everything in his power to discourage
the party, and bring his leader's authority into contempt.

At Marchant Springs, where they arrived on February 15th, they began to
experience annoyance from the natives. On the 17th, as Auld was
approaching the water-hole, a native who was there called to some others
who were posted in trees, and shortly afterwards a great cloud of smoke
was seen to windward, coming towards the camp. It was evidently their
intention to attack the exploring party under cover of the smoke, "but
Thring, while looking for the horses, came suddenly on three of them
concealed behind a bush, armed with spears and boomerangs; he did not
perceive them until within twelve yards of them. They immediately jumped
up, and one of them threw a boomerang at him, which fortunately missed
both him and his horse. He was obliged to use his revolver in self
defence," but with what result Mr. Stuart does not state.

The excessive heat of the weather now proved a great hindrance to the
expedition. They had already lost so many horses that a large part of
their provisions, etc. had to be abandoned on various occasions. On
February 23rd, Mr. Stuart writes:

"Before reaching this place (the Hugh) five other horses gave in, and
were unable to proceed further. I cannot understand the cause of the
horses knocking up so much; every one of them has fallen off the last
week. Whether it is the excessive heat or the brackish water of the
Finke, I am unable to say. Last night I tried some citric acid in the
water of the Finke, and it caused it to effervesce, showing that the
water contained soda." It was afterwards ascertained that the horses were
suffering from worms, which may partially account for their failing
strength.

After leaving the Hugh, on February 25th, they were again annoyed by the
natives. When about half-way through the gorge, they "set fire to the
grass and dry wood across the creek, which caused a dense smoke to blow
in our faces. I had the party prepared for an attack. After passing
through the smoke and fire, three natives made their appearance about
twenty-five yards off, on the hill side, armed with spears and shields,
and bidding us defiance by placing the spears in the womeras, and yelling
out at the highest pitch of their voices. I ordered Auld to dismount and
fire a shot a little distance on one side of them, to let them know what
distance our weapons carried. The ball struck the rock pointed out to him
to aim at, and stopped their yelling, but seemed to have no other effect.
I again ordered him to fire at the rock on which the middle one of the
three was standing; the shot was a good one, for the ball struck the
desired spot, and immediately had the effect of sending them all off at
full speed."

Again, on March 5th, while crossing the plains under Mount Hay, they came
suddenly on three natives armed with long spears and shields, who ran off
into the scrub. A short distance further, while watering the horses at
some rain water, these three natives returned, accompanied by four
others, and made signs of hostility, by yelling and shaking their spears,
and performing other threatening antics while widely separating
themselves in a half-circle. Mr. Stuart says: "I had the party prepared
to receive an attack; but when they saw us stationary they approached no
nearer. I ordered some of the party to fire close to them, to show them
we could injure them at a long distance, if they continued to annoy us,
but they did not seem at all frightened at the report of the rifles nor
the whizzing of the balls near to them, since they still remained in a
threatening attitude. With the aid of a telescope we could perceive a
number of others concealed in the belt of scrub. They all seemed fine
muscular men. There was one tall fellow in particular with a large shield
and a very long spear (upwards of twelve feet), which he seemed very
anxious to use if he could have got within distance. We crossed the
creek, and had proceeded a short distance across the plain, when they
again came running towards us, apparently determined to attack; they were
received with a discharge of rifles, which caused them to retire and keep
at a respectful distance. Having already wasted too much time with them,
I proceeded over the plain, keeping a sharp look-out; should they
threaten us again, I shall allow them to come close, and make an example
of them. It is evident their designs are hostile. Before entering the
scrub we could see no signs of them following. About sundown, arrived at
Mount Harris without further annoyance."

A week later (on March 12th) the Centre was passed; and on the 17th,
while going from Woodforde Creek through the bad country towards the
Bonney, Thring met with an awkward accident, which his leader thus
describes: "Being anxious to keep my old tracks through the scrub, as it
does not wear the saddle-bags so much as breaking through a new line, I
missed them about two miles after starting, in consequence of the
earliness and cloudiness of the morning. I sent Thring in search of them,
and he, on finding them a short distance off, fired his revolver to let
us know that the tracks were found. The young horse he was riding stood
the first report very well. I, not hearing the report, was moving on,
which caused him to fire again, whereupon his horse backed and threw him
with violence to the ground on his chest. He feels his chest is much hurt
by the fall. The horse then returned on the tracks at full gallop," and
was not recovered until shortly before sundown.

The party camped at Attack Creek on Friday, March 28th, and at Tomkinson
Creek on the 31st. On April 3rd, while crossing the Gleeson, Kekwick's
horse fell back with him in ascending the bank, and broke the stock of
his gun, but he himself escaped unhurt. On Saturday, April 5th, they
camped at the east end of Newcastle Water, and the following day, "at
about 9 o'clock a.m. Kekwick, in endeavouring to shoot some ducks, went
towards some native smoke, and was met by two natives, who ran away. In
an hour afterwards, five natives came within a hundred yards of the camp,
and seemed anxious to come up to it, but were not permitted. Two hours
afterwards we were again visited by fifteen more, to some of whom a
present was made of some looking-glasses and handkerchiefs; at the same
time they were given to understand that they must not approach nearer to
the camp, and signs were made to them to return to their own camp, which
they shortly did. In the afternoon we were again visited by nineteen of
them, who approached within a hundred yards of the camp, when they all
sat down and had a good stare at us, remaining a long time without
showing any inclination to go. At length some of them started the horses
which were feeding near the water, and made them gallop towards the camp.
This so frightened the natives that they all ran away, and we were not
troubled with them for the rest of the evening."

The next day the camp was moved to the north end of Newcastle Water,
where they remained for a week resting horses and repairing bags,
saddles, etc. The Journal then continues as follows:

Monday, 14th April, North End of Newcastle Water. Leaving Mr. Kekwick in
charge of the party, started with Thring and Frew at 7.15 a.m., on a
northerly course, in search of water; and at six miles, on the edge of
the open plains, found some rainwater, sufficient for a few days.
Proceeded across the plain on the same course; but at three miles saw
something like a watercourse, and changed my course to 20 degrees, to see
what it was. At two miles I struck a dry course running south-west.
Followed it up towards the small rise without finding any water. Three
miles further on the same course I ascended a low stony rise, from which
I could see nothing but a thick forest of tall mulga and gums. I changed
to a northerly course, and, at 4.20 p.m., camped in a forest without
water. Wind, south-east.

Tuesday, 15th April, Sturt Plains, Forest. Proceeded on a course of 250
degrees, and at five miles again struck the open plains, and changed to
180 degrees. At one mile I found a fine water hole three feet deep and
about forty feet in diameter, the edge of which was surrounded with
conglomerate ironstone rock; watered the horses, and proceeded on a
southerly course, through grassy plains with stunted gum-trees, to the
first water I found yesterday, and camped. The plains and forest are of
the same description as I have already given, only that the plains have
not quite so many holes in them, and the forest in many places is covered
with ironstone gravel. I shall try a course to the north of west
to-morrow, to see if I can find water. Wind variable.

Wednesday, 16th April, Frew's Water Hole, Sturt Plains. Started at 7.45
a.m., on a course of 302 degrees, keeping along the edge of the open
plain. I have made many twistings and turnings, but my general course is
north-west for ten miles. Seeing a small rise on the open plain, a little
to the north of west, I changed to 275 degrees; and at two miles came on
some fine ponds of water about one mile and a half long, twenty feet
broad, and three feet and a half deep. I examined them on both sides, to
see if they would do for a permanent camp for the party as it is a point
nearer; and I think I may depend on the water lasting two months without
any more rain. I shall camp here to-night, and try another day to-morrow
to the westward, and endeavour to make the Victoria, for I can see but
little chance of making the Adelaide. By my journal of the 14th,
everything is quite dry and parched up; no rain seems to have fallen
there for a long time. The last two days have been excessively hot. The
further to the west the hotter I find it. The natives seem to be
numerous, for their smoke in the scrub is to be seen in every direction.
I name these ponds after John Howell, Esquire, of Adelaide.

Thursday, 17th April, Howell Ponds. Started at 7 a.m., on a bearing 10
degrees north of west. At twelve miles crossed the open plains, and
entered a thick forest of gums and other trees and shrubs. Seeing that
there is no chance of finding water to-day, returned to the ponds. The
open plains seem to tend more to the north-west; I shall examine them
when I bring the party up to the ponds. Distance, fifteen miles. Wind,
south-east.

Friday, 18th April, Howell Ponds. Started for the camp on the Newcastle
Water. On my arrival, I found the party all right, but very anxious about
me, as I had been absent longer than I intended. No natives had been near
them during my absence at this time; smoke was seen all around. Weather
hot during the day, but cold at night and in the morning. Wind,
south-east.

Saturday, 19th April, North End of Newcastle Water. I shall remain here
till Monday, in order to take some lunar observations, as I am not quite
certain that my longitude is correct. Wind, south-east.

Sunday, 20th April, North End of Newcastle Water. Wind from the east;
blowing strongly during the day, but it dropped a little before sundown,
allowing the mosquitoes to annoy us very much.

Monday, 21st April, North End of Newcastle Water. Some of the horses
having strayed some distance made it 10 o'clock a.m. before I could get a
start. Proceeded through six miles of forest and scrub to the water that
I found on the 14th instant; from thence I changed to 301 degrees 30
minutes for nine miles, and then to 275 degrees, and at two miles camped
at the ponds I had discovered on the 16th. Native smoke all around us.
The day has been very hot, and the flies a perfect nuisance. Wind,
south-east.

Tuesday, 22nd April, Howell Ponds. Preparing for a start to-morrow to the
north-west in search of water. Wind, south-east.

Wednesday, 23rd April, Howell Ponds. Leaving Mr. Kekwick in charge of the
party, I started with Thring and Frew at 8.5 a.m., on a course of 284
degrees. At 9.55 (seven miles) changed to 320 degrees. At 11.20 (four
miles and a half) crossed the open plain, changing to 40 degrees to avoid
the scrub. At one mile and a half changed to west. At one mile changed to
north-west. At 2.20 (five miles) changed to 45 degrees. At 3 o'clock (two
miles) changed to north. At 3.25 one mile and a half changed to
north-west. At 3.45 camped without water. I have skirted the border of
the forest land in the hope of finding water, but am disappointed. I have
not seen a drop since I started. The plains are covered with beautiful
grass, two or three feet high. There are a great many different kinds of
birds about, and native smoke all round. I have searched every place
where I thought there was the least chance of finding water, but without
success. The day has been exceedingly hot. With such hot weather as this
I dare not attempt to make the Victoria. The horses could not stand a
hundred and forty miles without water. Those I have had with me to-day
seem to have suffered enough, and would not stand another two days
without. I must therefore return to the camp to-morrow. Wind, calm.

Thursday, 24th April, Sturt Plains. Returned to the camp and found all
right. The day has been excessively hot. We have seen nothing new during
the journey--the same open plains, with forest between.

Friday, 25th April, Howell Ponds. Leaving Mr. Kekwick in charge of the
party, started at 8.20 a.m., with Thring and Frew and fresh horses, on a
northerly course, in hopes of better success in that direction: course
360 degrees for twenty-two miles; grassy plains, covered in many places
with stunted gums, and a new tree with a small green leaf. After that, we
entered again a thick forest, and scrub almost impassable. At
twenty-eight miles, seeing no prospect of getting through it, I returned
two miles to a small open space, where I could tether the horses. I have
not seen a drop of water this day's journey. The forest is so very thick,
and so many twistings and turnings are required to pass through it, that,
although I travelled thirty miles, I don't believe I made more than
fifteen miles in a straight line. The day again exceedingly hot, with a
few clouds. A few birds were seen during this day's journey, but no
pigeons, which are the only sign we have now of being near water. Wind
variable.

Saturday, 26th April, Dense Forest. Returned to the camp. The horses felt
the heat and the want of water very much. In the forest the heat was
almost suffocating. I hope it will rain soon and cool the ground and
replenish the ponds, which are drying up fast. There have been a few
clouds during the day, but after sundown they all cleared away. Wind,
south-east.

Sunday, 27th April, Howell Ponds. A few clouds have again made their
appearance, but still no rain. There has not fallen a drop of rain since
I left the Woodforde, which was on the 9th of March. Wind, south-east.
Latitude, 17 degrees 5 minutes 16 seconds.

Monday, 28th April, Howell Ponds. Leaving Mr. Kekwick in charge of the
party, started with Thring and King, on a course of 338 degrees, to try
and find an opening in the dense forest and scrub, as well as water. At
ten miles we crossed the open plain, with stunted gum-trees and long
grass. At this point we met with a small ironstone rise, about twenty
feet in height. On ascending I was again disappointed in finding before
me a dense forest and scrub. Proceeding in our course, it became thicker
than any which I had ever encountered before, and was almost impassable.
Still continued, and for a short distance in some places it became more
open. A little before sundown I camped on the edge of a stunted gum-tree
plain. There are a few slate-coloured cockatoos and other birds, which
lead me to hope that, in the morning, I may drop across some water. Wind
variable, with a few clouds during the day.

Tuesday, 29th April, Sturt Plains. Started on an easterly course,
following the flight of the birds; but at five miles crossed the open gum
plain, and again encountered the thick forest. Examined every place I
could see or think of where water was likely to be found, but was again
disappointed--not a drop was to be seen. Changed my course, so as to keep
on the plain; at four miles again crossed, and again met the dense
forest, but still no water. Changed to south-east, and at ten miles found
ourselves on a large stunted-gum plain. Changed to a little east of
south, and arrived at the camp without seeing a drop of water. Wind
variable, with heavy clouds from the east.

Wednesday, 30th April, Howell Ponds. I feel so unwell to-day that I am
unable to go out, besides I shall require my compass case and other
things mended; they got torn to pieces in the last journey by the forest
and the scrub. Yesterday's clouds are all gone, and have left us no rain.
Another hot day. Wind, east.

Thursday, 1st May, Howell Ponds. Leaving Mr. Kekwick in charge of the
party, started with King and Thring to the water hole that I discovered
on the 15th ultimo; arrived in the afternoon and camped. This water hole
I have named Frew's Water Hole, in token of my approbation of his care
of, and attention to, the horses. This waterhole is about twenty feet
below the plain, surrounded by a conglomerate ironstone rock. Since my
last visit it is only reduced two inches, and is still a large body of
clear water from the drainage of the adjacent country; it will last much
longer than I anticipated. I shall use my best endeavours to-morrow to
find an opening in the thick scrub from north to north-west. The course
of the forest seems to run a little west of north, and I am afraid the
open plains are surrounded by it; however, I shall try to get through it
if I possibly can. Wind, south-east. Day excessively hot.

Friday, 2nd May, Frew's Water Hole. Started at half-past seven o'clock
a.m. Course, 335 degrees. At ten miles, a dense forest and scrub. Changed
to 10 degrees east of north. At half a mile struck a water-shed, and
followed it north for two miles. Found a little rainwater in it, and at
two miles further arrived at its source. At three miles further on the
same course changed to 30 degrees east of north. At three miles and a
half again changed to 320 degrees, and at about a mile and a half struck
some fine ponds of water. At two miles further, arrived at what seemed to
be the last water, a small shallow pond. Examined around the plain to try
and find others, but without success. A little before sundown, returned
to the last water and camped. The first part of the day's journey was
over a stunted-gum plain, covered with grass. At ten miles we again met
with thick forest and scrub. I then changed my course to get out of it,
and struck the small water-shed running to the east of south. Following
it generally for two miles on a northerly course, we met with a little
rain water. Continued the same course through a thick forest and scrub
for three miles and a half to get through it if possible. At this point
it becomes denser than ever. Sent Thring to climb to the top of a tree,
from which he saw apparently a change in one of the low scrubby rises,
which appeared to be not so thickly covered with scrub as the others. I
directed my course to it, 30 degrees east of north, to examine it. I
observe that there is some sandstone in the low scrubby rises, which
leads me to hope that I may not be far from a change of country. On this
last course we travelled three miles, through a dense thicket of
hedge-tree, when I observed some large gum-trees bearing 320 degrees, and
decided to examine them before leaving the rise. As I approached nearer
to it I again sent Thring to climb a tree to see if there was any change.
He could see nothing but the same description of forest and scrub. The
change that he saw from the other tree was the shade of the sun on the
lower mulga bushes, which caused him to suppose that it was more open
country. Not seeing any opening in that direction, I changed to the
gum-trees. At a mile and a half was delighted at the sight of a chain of
fine water holes; their course north-west to south-east, the flow
apparently to south-east. I followed one pond, which was about half a
mile long and appeared to be deep. A number of smaller ones succeeded.
They then ceased, and I crossed a small plain, which shows signs of being
at times covered with water. Observing some green and white barked
gum-trees on the west side of it, I went to them, and found a small
watercourse with small pools of water, which flowed into the plain coming
from the north-west. Following it a little further, we met with some more
water. A short distance above this it ceased in the dense forest which
seems to surround these ponds. I shall endeavour to force my way through
it to-morrow to the west of north. Wind, south-east, with a few clouds in
the same direction. These ponds I name King's Ponds, in token of my
approbation of his care of, and attention to, the horses, and his
readiness and care in executing all my orders. Wind, south-east, with a
few clouds in the same direction.

Saturday, 3rd May, King's Chain of Ponds. Started at twenty minutes past
seven a.m., on a course of 350 degrees. At twenty-four miles changed to
45 degrees; at three miles and a half changed to north; at two miles and
a half camped. At two miles from our last night's camp found an easy
passage through the forest; the rest of the twenty-four miles was over a
well-grassed country, well wooded with gum and some new trees that I had
found last year, and occasionally a little scrub, in some places thick
for a short distance. On my first course, before changing, I was crossing
low ironstone undulations, which caused me to think I was running along
the side of one of the scrubby rises. I therefore changed to 45 degrees
east of north to make the plain--if there is any--the scrub being so
thick that I cannot see more than fifty yards before me. At three miles
and a half I found that I was travelling over the same description of
small rises and getting into much thicker scrub. I again changed to
north, to see if that would lead me into a plain. At two miles and a half
it was still the same, and apparently a thick forest and scrub before us.
I camped a little before sundown at a small open place to tether the
horses. I have not seen a drop of water during the whole journey, nor any
place likely to retain it, with the exception of a small flat about six
miles from the last camp. The day very hot. Wind, south-east, with a few
clouds.

Sunday, 4th May, Dense Forest. Returned to King's Ponds. This country
seems but little frequented by the natives, as we have seen no recent
tracks of them. There are a number of cockatoos and other birds about. We
have seen no other game, except one wallaby and one kangaroo. There are
plenty of old emu tracks about the ponds. Wind, variable. Cloudy.

Monday, 5th May, King's Ponds. Returned to Frew's Water Hole and camped.
Before sundown the sky became overcast with clouds. Wind variable.

Tuesday, 6th May, Frew's Water Hole. Towards morning we had a few drops
of rain. Returned to the camp and found all well. Yesterday they were
visited by a few natives who seemed to be very friendly; they called
water ninloo: they were armed with spears, about ten feet long, having a
flat sharp flint point about six inches long, with a bamboo attached to
the other end. They pointed to the west as the place where they got the
bamboo and water also, but they seemed to know nothing of the country
north of this; they were tall, well-made, elderly men. After talking for
some time they went away very quietly. To-day they have set fire to the
grass round about us, and the wind being strong from the south-east it
travelled with great rapidity. In coming into the camp, about three miles
back, I and the two that were with me narrowly escaped being surrounded
by it; it was as much as our horses could do to get past it, as it came
rolling and roaring along in one immense sheet of flame and smoke,
destroying everything before it.

Wednesday, 7th May, Howell Ponds. Resting. The natives have not again
visited us, but their smoke is seen all around. I shall start to-morrow
on a course west of north, to try and make the Victoria by that route. I
shall take some of the waterbags with me to see how they answer. Wind,
south-east. Clouds all gone.

Thursday, 8th May, Howell Ponds. Leaving Mr. Kekwick in charge of the
party, started with Thring and McGorrerey, also with King and Nash, who
are to bring back the horses which carry the waterbags, whilst I with
Thring and McGorrerey proceed on a west course. Started at half-past
eight a.m., keeping the former tracks made on my previous journey to the
westward, to where we met with the thick forest. About a mile beyond,
struck a native track, followed it, running nearly north-west, until
nearly three o'clock p.m., when we came upon a small water hole or
opening in the middle of a small plain, which seems to have been dug by
the natives, and is now full of rain water. This is apparently the water
that the natives pointed to, for their tracks are coming into it from
every direction. This opening I have named Nash Spring, in token of my
approbation. I am very much disappointed with the water-bags; in coming
this distance of twenty-one miles they have leaked out nearly half. Wind,
south east.

Friday, 9th May, Nash Spring. Sent King and Nash with the horses that
carried the water-bags back to the depot, while I and the other two, at
twenty minutes to eight o'clock a.m. proceeded on a bearing of 290
degrees, following one of the native tracks running in that direction. At
about a mile they became invisible; for that distance I observed that a
line of trees was marked down each side of the track by cutting a small
piece of bark from off the gum-trees with a tomahawk. This I had never
seen natives do before; the marks are very old. At eighteen miles and a
half struck another track (the trees cut in the same way) crossing our
course; followed it, bearing 10 degrees east of north, and at about two
miles came on a native well with moisture in it. Followed the valley on
the same course, but seeing no more appearance of water, I again changed
to my original course, and, at a quarter to four o'clock, finding that I
was again entering the dense forest and scrub, I camped at a good place
for feed for the horses, but no water. The whole of the day's journey has
been through a wooded country, in some places very thick, but in most
open; it is composed of gums, hedge-trees, and some new trees--the gums
predominating; there were also a few patches of lancewood scrub. For the
first eighteen miles the soil was light and sandy, with spinifex and a
little grass mixed. At the end of eighteen miles I again got into the
grass country, with occasionally a little spinifex. Wind, south-east.
Cold during the night and morning.

Saturday, 10th May, The Forest. Started at five minutes to seven o'clock
a.m. (same course, 290 degrees). Almost immediately encountered a dense
forest of tall mulga, with an immense quantity of dead wood lying on the
ground. It was with the greatest difficulty that the horses could be made
to move through it. At a mile it became a little more open, which
continued for six miles. At seven miles I thought, from the appearance of
the country, that it was dipping towards the north-north-west; I
therefore changed my course to north-west, and in less than a mile again
entered a dense forest of tall mulga, thicker than I had yet been into.
Continued pushing, tearing, and winding into it for three miles. The
further I went the denser it became. I saw that it was hopeless to
continue any further. We were travelling full speed, and making little
more than a mile an hour throughout the ten miles gone over to-day. The
country is a red light soil and covered with abundance of grass, but
completely dried up. No rain seems to have fallen here for a length of
time. We have not seen a bird, nor heard the chirrup of any to disturb
the gloomy silence of the dark and dismal forest--thus plainly indicating
the absence of water in and about this country. I therefore retraced my
steps towards Nash Springs; passed our last night's camp, and continued
on till sundown, one of the horses being completely knocked up. Camped
without water. Wind, south-east.

Sunday, 11th May, The Forest. This morning the horse that was so bad last
night was found dead, which puts us in a very awkward position--without a
pack-horse. We had to leave behind the pack-saddle, bags, and all other
things we could not carry with us on our riding-horses. Proceeded to Nash
Spring, which we reached after two o'clock p.m., with another of the
horses completely knocked up. It was with difficulty that he reached it.
I suppose the days being so extremely hot, and the feed so dry that there
is little nourishment in it, is the cause of this, as they were horses
that had been out with me on my last year's journey, and had suffered
from want of water a longer time than on this occasion. I am nearly in a
fix with a long journey before me, the horses unable to do more than two
nights without water, and the water-bags losing half their contents in
one day's journey. To make the Victoria through the country I have just
passed into would be impossible. I must now endeavour to find a country
to the northward and make the Roper. I am very vexed about the water-bags
turning out so badly, as I was placing great dependence on them for
carrying me through. I must try and push through the best way possible.
Wind, south-east.

Monday, 12th May, Nash Spring, West Forest. Proceeded very slowly with
the knocked-up horse to the Depot; he appears to be very ill, and is
looking very bad this morning. Arrived there and found all right; they
had been visited by the natives twice during my absence. They appeared to
be very friendly, and were hugging Frew and King, for whom they seemed to
have taken a great fancy; they were old, young, and children. Some pieces
of white tape were given to them, which pleased them much. They still
pointed to the west, as the place where the large water is, and made
signs with a scoop to show that they have to dig for it in going through;
which I am now almost sure is the case from what I saw of the country in
my last journey in that direction. In upwards of fifty miles we did not
see the least signs of a watercourse--nor could I discover any dip in the
country; it has the same appearance all round; one cannot see more than
half a mile before one, and in many places only a few yards. I have been
deceived once or twice by what appeared to be a dip in the country, but
it turned out to be only lower trees and scrub than what we were
travelling through. With a small party I might make the Victoria from
here, but there is every chance of losing the horses in doing so; and I
should be in a sad predicament to be there without horses, and without
the possibility of receiving supplies from the party at the Depot; I
should have to perish there. Therefore, I consider it would be folly and
madness to attempt it, and might be the cause of sacrificing the lives of
both parties. Had the feed been green, or had it any substance in it, I
would have tried, but every blade of grass is parched and dried up as in
the middle of summer, and the horses have not the strength nor endurance
to undergo much privation, of which I have had a proof in the journey I
have just taken. After resting a day or two to recover the horses, and
get ourselves a little refreshed, I shall move the party up to King's
Ponds, and try to push through wherever I can find an opening. Day very
hot. Wind, south-east. A few clouds came up from that quarter after
sundown.

Tuesday, 13th May, Depot, Howell Ponds. Resting ourselves and horses. Day
again hot, with a few clouds round the horizon. The natives had again set
fire to the country all around, which increases the heat. I wish it would
come on to rain, and put out their fires, and fill the ponds, which are
shrinking a great deal more than I expected. Wind, south-east. Clouds.

Wednesday, 14th May, Depot, Howell Ponds. As I don't feel well enough
to-day, I shall remain here, and start to-morrow morning. This morning,
while Thring was collecting the horses, he came on a place where the
natives had been encamped a day or two before, and there saw the remains
of the bones of one of them that had apparently been burnt; this is
another new feature in their customs. Wind, south-east.

Thursday, 15th May, Depot, Howell Ponds. Started with the party across
the plain to Frew's Water Hole, course 15 degrees east of north; found
the plain burnt for ten miles. The fire has been so great that it has
burned every blade of grass, and scorched all the trees to their very
tops. I was very fortunate the other day in having escaped it; nothing
could have lived in such a fire, and had we been caught in it we must
have perished. Wind, south-east. Clouds all gone, Latitude, 16 degrees 54
minutes 7 seconds.

Friday, 16th May, Frew's Water Hole. Started at fourteen minutes past
eight o'clock a.m., course 345 degrees, for King's Chain of Ponds.
Arrived at about half-past three o'clock p.m. In coming through, one of
the horses separated from the rest and bolted off into the dense forest,
tearing everything down before him. We got him in again, but with a
broken saddle, and the top off one of the bags, which we afterwards
recovered. Arrived at the ponds without any further accident. Wind,
north-east. Very hot, and a few clouds. Latitude, 16 degrees 38 minutes
53 seconds.

Saturday, 17th May, King's Chain of Ponds. Sent King and Thring to follow
round the flat to see where the ponds go to. About noon they returned,
and reported that the water loses itself in a flat, which is surrounded
by thick forest and scrub. This certainly is a very pretty place, and a
great pity it is not more extensive. It reminds me much of the park land
found by Captain Sturt in 1845, where he had his second depot, named Fort
Grey. Wind, south-east, with a few clouds.

Sunday, 18th May, King's Chain of Ponds. In the afternoon the sky became
cloudy, and at sundown was quite overcast; the day exceedingly hot, and
the wind nearly calm. The clouds came from the north-west, and the little
wind there is from the south-east.

Monday, 19th May, King's Chain of Ponds. As the sky is overcast with
clouds, so that I cannot see the sun, and as it is nearly impossible to
keep a straight course in such thick country without it, I shall remain
here to-day, and if it should break up I shall endeavour to take a lunar
observation. At 9 o'clock a.m. it cleared up, which enabled me to take
one. The remainder of the day very hot. Wind variable, with clouds from
every direction; towards sundown it again settled in the south-east, and
all the clouds disappeared without any rain falling.

Tuesday, 20th May, King's Chain of Ponds. Leaving Mr. Kekwick in charge
of the party, I started with Thring, King, and Auld, at half-past nine
a.m., on a northern course; at one o'clock p.m. changed to 65 degrees, to
what appeared to be a bare hill. At a little more than a mile struck a
small watercourse running towards the north; followed it, and at about
two miles and a half came on some ponds of water, but not so large as
those at our depot; at present they are not more than three feet and a
half deep. Examined around the wooded plain to see if there was any
larger body of water, but could see none. This plain is covered with
small gums, having a dark bluish-green leaf with a grey-coloured bark;
there are also a few white ones around the ponds of water, which abound
with grass. Before reaching the plain we crossed what seemed to be
elevated sandy table land, extending about nine miles, covered with
spinifex and dark-coloured gum-trees; we also passed two or three narrow
belts of tall mulga and hedge-trees which grow on the stony rises, about
twenty feet high. These ponds I name Auld's Ponds, in token of my
approbation of his conduct. Wind, south-east. Latitude, 16 degrees 28
minutes 16 seconds.

Wednesday, 21st May, Auld's Chain of Ponds. Started at twenty minutes
past eight o'clock a.m., course north. The morning was so thick, with a
heavy fog, that I did not get a start till late. At three miles I found
another chain of ponds, but not so large; these I name McGorrerey Ponds.
Proceeded on the same course and passed through some thick belts of
hedge-tree and scrub; the country then opened and became splendidly
grassed, with gums and other trees. We also saw, for the first time, a
new gum-tree, having a large broad dark-green leaf, and the bark of a
nankeen colour, which gave a very pretty effect to the country. At
seventeen miles, not finding any water, and having passed five deep holes
surrounded with ironstone conglomerate rock similar to Frew's Water Hole,
but without any water in them, and to all appearance the dip of the
country being to the north-east, I have changed my course to that
direction, again travelling over a splendidly grassed country for ten
miles, occasionally meeting with low stony rises of ironstone and gravel,
at the foot of which were some more deep holes without water. In the last
three miles we had to get through a few patches of scrub; the grass is
all very dry. No rain seems to have fallen here for a long time. At
sundown camped without water. Day very hot. Wind variable, with a few
clouds. Latitude, 16 degrees 8 minutes 39 seconds.

Thursday, 22nd May, Fine Grass Country. Returned to McGorrerey Ponds. Day
very hot, and the horses much distressed for want of water; they have the
appearance of being half-starved for a month, and have taken an immense
quantity of water, having gone to it about four or five times in an hour.
As I am not satisfied that these ponds cease here, I shall try again
to-morrow a little more to the east. Wind, south-east.

Friday, 23rd May, McGorrerey Ponds. Gave the horses a little time to feed
after daylight in consequence of their having been tethered during the
night; the country is so thickly wooded that I dare not trust them in
hobbles the whole night, as, if they were lost sight of there would be
great difficulty in finding them here. There is still the appearance of a
small creek, which I shall follow until it runs out or trends too much to
the east. Started at half-past eight o'clock a.m., course 20 degrees east
of north, following the small creek about two miles; it seems to be
getting larger, with occasionally a little water in it. We have also
seen, on both sides of us, ponds with water surrounded by gum-trees;
these ponds, when full, must retain water for a long time. We have also
seen a new tree growing on the banks of the creek, with a large straight
barrel, dark smooth bark, with bunches of bright yellow flowers and
palmated leaves. At a mile and a half further the creek is improving
wonderfully. We have now passed some fine holes of water, which will last
at least three months; at five miles the water is becoming more plentiful
and the creek broader and deeper, but twisting and turning about very
much, sometimes running east and then turning to the west and all other
points of the compass. Having seen what I consider to be permanent water,
I shall now run a straight course, 20 degrees east of north, and strike
it occasionally to see if the water continues. I have named these Daly
Waters, in honour of his Excellency the Governor-in-Chief. Within a
hundred yards the banks are thickly wooded with tall mulga and lancewood
scrub; but to the east is open gum forest, splendidly grassed. Proceeded,
occasionally touching the creek, and always found fine reaches of water,
which continued a considerable way. At thirteen miles they become smaller
and wider apart; at fifteen miles the creek seems to be trending more to
the eastward, its bed is now conglomerate ironstone, and, as this appears
to be about the last water, I shall give the horses a drink and follow it
as far as it goes. In a short distance it has become quite dry, with a
deep broad course upwards of twenty yards wide. At seventeen miles it
separated into two channels, and at a quarter of a mile the two channels
emptied themselves into a large boggy swamp, with no surface water. I
examined the swamp, but could see no outlet. The country round about is
thickly timbered with gum and other trees. Returned to the last water and
camped. I shall return to the Depot and bring the party up here. Wind,
south-east; a few clouds at sunset.

Saturday, 24th May, Chain of Ponds, Large Creek. Followed my tracks back
to Auld's Chain of Ponds, and had difficulty in doing so, the ground
being so hard that the hoofs of the horses scarcely left any impression
on it. This would be a fearful country for any one to be lost in, as
there is nothing to guide them, and one cannot see more than three
hundred yards around, the gum-trees are so thick, and the small belts of
lancewood make it very deceptive. Should any one be so unfortunate as to
be lost, it would be quite impossible to find them again; it would be
imprudent to search for them, for by so doing the searchers would run the
risk of being lost also. Arrived at Auld's Ponds and camped. Wind,
south-east. A few clouds.

Sun day, 25th May, Auld's Chain of Ponds. Proceeded to the Depot, where I
arrived in the afternoon and found all well. No natives have been near
them, although some of their smoke has been seen at a short distance from
the Depot. Yesterday we hoisted the Union Jack in honour of her Most
Gracious Majesty's birthday, that being the only thing we had to
commemorate this happy event, with our best wishes for her long and happy
reign. Wind, south-east.

Monday, 26th May, Chain of Ponds. Removed the party on to Auld's Chain of
Ponds.

Tuesday, 27th May, Auld's Chain of Ponds. Proceeded with the party to the
fourth chain of ponds and creek. This water has every appearance of being
permanent, and I hope I may fall in with such another in the next degree
of latitude. It may be from this that the Wickham receives a supply of
water when this overflows. Wind, south-west. Latitude, 16 degrees 14
minutes 31 seconds.

Wednesday, 28th May, Daly Waters, Fourth Chain of Ponds and Creeks. Sent
Thring and King to round the swamp into which this creek flows, to see if
there is any outlet to the eastward of this within two miles. There are
other ponds and a creek, which also empties itself into a swamp a little
to the eastward of the one into which this one empties itself. In the
afternoon they returned, having found a small watercourse forming the
north-west side of the swamp; followed it, running nearly 10 degrees east
of north. In about one mile and a half they came upon a large swamp
covered with water, but shallow. They then proceeded seven miles on a
north-east course; then meeting with some white-barked gum-trees,
appearing to run to the north-west, followed them for three miles,
crossing a gum and grass plain. Observing some native smoke to
north-east, they returned. Wind, south-east.

Thursday, 29th May, Daly Waters. Leaving Mr. Kekwick in charge of the
camp, at half-past seven o'clock proceeded with Thring, Auld, and Frew
down the creek to examine the swamp found yesterday. It is about 30
degrees east of north, about three miles from the Depot at Daly Waters.
The water does not appear to be deep, but covers a large area; there were
a few pelicans and other water-birds on it. From this we proceeded, on a
course 20 degrees east of north, to search the flat where Thring and King
saw the smoke yesterday. At eighteen miles from Daly Waters, having
crossed the gum plain without meeting with any water, and being on
apparently higher ground than the plain, I changed my course to 90
degrees east of north. At two miles and a half again crossed the plain,
and got upon low rising ground of ironstone and gravel, but still no
water. Changed to former bearing of 20 degrees east of north, and at
seven miles came upon a dry swamp, covered with long blue grass and deep
holes, but still no water could we find. Proceeded another mile, and
finding I was getting on rising ground, and the horses having done a long
and heavy day's journey, camped without water. After leaving the swamp
with the water (which was very boggy all round it), the country became
similar to that of Sturt Plains surrounding Newcastle Water, being so
full of deep holes that we were in danger of getting our necks broken,
and also the horses' legs. The soil is good, and completely covered with
grass and stunted gum-trees. In rainy weather it seems to be covered with
water. There is no watercourse, or any appearance of which way the water
flows. A number of various kinds of birds were about. Wind variable, but
mostly from south-west. Latitude, 15 degrees 56 minutes 11 seconds.

Friday, 30th May, North-north-east of Blue-Grass Swamp. Wishing to see a
little more of the country further on and to find where the birds get
their water, I proceeded with Thring, leaving the other two behind with
the horses, three miles and a half on the same course, following their
flight. In half a mile came again upon the stunted gum plain, splendidly
grassed to above the horses' knees. Can find no water, although the birds
are still round about us. The same description of country continues from
the swamp with the water to beyond this, consisting of small undulations
of gravel and ironstone. Retraced my steps to where I had left the other
two, and proceeded towards the Depot at nine miles. The country was in a
blaze of fire to the east of us. I am very thankful there was scarcely a
breath of wind, which enabled us to pass within a quarter of a mile of
it: had there been a strong wind we should have been in great danger, the
grass being so long and thick. Returned to the Depot after six p.m.,
being all very tired with the shaking we have had the last two days by
the horses falling into the holes nearly every step, and they also are
nearly exhausted; twelve hours in the saddle over such a country is no
easy task. It was my intention to have come back more to the east, but
having seen the smoke I saw we should be in the middle of the fire, and
so changed my intention. Wind, south-west. Very hot.

Saturday, 31st May, Daly Waters. As there are no appearances of rain, the
weather very hot, and I have a good deal of work in plans, etc. to bring
up, I shall remain here until Monday. I feel this heavy work much more
than I did the journey of last year; so much of it is beginning to tell
upon me. I feel my capability of endurance beginning to give way. There
are a number of small fish in this water, from three to five inches long,
something resembling a perch; the party are catching them with hooks;
they are a great relish to us, who have lived so long upon dry meat. Any
change is very agreeable. Wind variable.

Sunday, 1st June, Daly Waters. The day has been as hot as if it were in
the middle of summer. Surely we must get a change soon. Wind variable,
with a few light clouds. Mr. Waterhouse has shot two new parrots.

Monday, 2nd June, Daly Waters. Leaving the party in charge of Mr.
Kekwick, I started at twenty minutes past seven (course north), with
Thring, Auld, and Frew. Camped at 4.20. The whole day's journey has been
through a splendid grass country, and open forest of gum-trees and other
shrubs, some of them new to us. Here again we have also met with the
bean-tree, the blossoms of a bright crimson, and at this season they seem
to shed their leaves. The country passed over consisted mostly of
undulations of ironstone and gravel, with a brown-coloured rock
occasionally, between which were broad valleys of a light-coloured soil,
all cracked and having many deep holes, which, being hidden with the long
grass, caused the horses to tumble into them, and made it very fatiguing
both to them and us. I have been constantly in the hope all day of coming
upon some water, but have been disappointed. After rain this country can
be passed over with the greatest facility, for we have passed holes that
will hold water for a long time. The dip of this country is now to the
eastward. To-day I think I have been running along where the dip
commences from the table land. It was my intention to have tried a
journey to the north-west; but, from what I have seen of the country
to-day, and on my other journeys to the north, as well as Mr. Gregory's
description of it on the other side, I am led to believe that it would be
hopeless to expect to find water there. To try it will only be losing
time, and reducing the strength of my horses. I must now try on a
north-east course towards the Gulf of Carpentaria. I do not wish to go
east if I can help it; but I must go where the water leads me. During the
day's journey we passed through three narrow belts of hedge-tree scrub,
which was very thick. There does not seem to be so much of that as we get
to the north, neither is there so much of the tall mulga. We have not
seen a drop of water since we left the camp. Camped without it. Wind,
south. Day very hot. Latitude, 15 degrees 50 minutes 20 seconds.

Tuesday, 3rd June, Gum Forest. Fine country. I sent Thring on three miles
to see if there was any change, there being a number of birds about that
frequent the place where water is. I think there may still be a chance of
falling in with some. He has returned and can see none. Country the same
as that travelled over yesterday. Returned to the Depot. Arrived a little
before sundown, and found all well. Wind light; south. Day again very
hot.

Wednesday, 4th June, Daly Waters. Preparing for a start to-morrow to the
north-east. I shall take the water-bags; they may retain as much as will
suffice for a drink night and morning for four horses. I shall proceed to
the blue-grass swamp that I found in my last north-north-east course,
trace that down as far as it goes, and, should there be no water, shall
strike for the sources of the Wickham River. Wind, south-south-east.

Thursday, 5th June, Daly Waters. Started at a quarter to eight with
Thring and Auld, taking all the water-bags full, also King and Billiatt
to take back the horses that carry the water. I have chosen King for this
purpose, as being the next best bushman to Thring, and one in whom I can
place the greatest dependence to execute any charge I may give him with
care and faithfulness. At four o'clock arrived at the blue-grass swamp.
Changed my course to 70 degrees east of north, following down the middle
of it, which contains a great number of large deep holes in which water
has been, but are now quite dry. Followed it until it spread itself over
the plain, causing a great number of deep cracks and holes completely
covered with grass, gums, and other trees, too thick to get an easy
passage through. At sundown camped on the plain without water. A few
hours before sundown the sky had a very peculiar appearance to the
eastward, as if a black fog were rising, or smoke from an immense fire at
a long distance off, but it was too extensive for that. At sundown it
assumed a more distinct aspect in the shape of black clouds coming from
that direction. Wind, south-east.

Friday, 6th June, Plain East of Blue Swamp. Sent King and Billiatt back
with the horses, while I proceeded with the other two on a course 70
degrees east of north. At a mile and a half came suddenly upon a scrubby
ironstone rise about twenty feet high. After passing over a rotten plain,
full of holes and covered with grass and stunted gum-trees, proceeded to
the top, from which we had a good view of the surrounding country--to all
appearance one of the blackest and most dismal views a man ever beheld;
even the splendid grass country I had been coming through has the same
appearance. The cause of it is the trees being so thick, and some of them
of a very dark colour, that nothing but their tops can be seen, which
gives it the appearance of being a dense scrub. To the west there is an
appearance of a scrubby rise--the one on which I have been on my other
journeys to the north. No hills visible; all appears to be a level
country. Proceeded down the gradual slope, crossing two other lower
ironstone undulations, meeting occasionally with small rotten plains with
holes, and covered with grass. At five miles the ground became firmer; at
seven miles met with what seemed to be a water-shed. After a long search
found that the flow of the water was to the west of north; traced it a
short distance to the south-east and found a small shallow pool of water
and gave our horses a drink; and wishing to take advantage of anything
that may take me to the north-west, I turned and traced it down; passed
three ponds with some water in them, and at three miles came upon a fine
large one two and a half feet deep; followed it still on, but was
disappointed on finding it terminate in a dry swamp, all cracked and full
of holes; circled round it to see if the creek took up again, but could
see no appearance of any. As this last pond will do for the party, I will
return and bring them up, for there is a slight appearance of rain, and I
wish to get them on as far as possible before the winter rain comes on.
Returned to our last night's camp, where we arrived at sundown. Wind,
south-east, with few clouds.

Saturday, 7th June, Plain East of Blue Swamp. Returned to the Depot;
found all well. Clouds all gone, but the wind blowing strong from the
south-east.

Sunday, 8th June, Daly Waters. Strong winds still from south-east, and
sometimes from the south. Day very hot.

Monday, 9th June, Daly Waters. Last night, a little after sundown, Mr.
Waterhouse was seized with a violent pain in the stomach, which was
followed by a severe sickness, and continued throughout the night; this
morning he is a little better. I think it was caused by eating some
boiled gum which had been obtained from the nut-tree Mr. Kekwick
discovered last year. When boiled it very much resembles tapioca, and has
much the same taste. I also ate some of it yesterday, which occasioned a
severe pain in the stomach, but soon went off. Some of the others also
felt a little affected by it, but none so bad as Mr. Waterhouse; on
others it had no effect whatever, and they still continue to eat it. Mr.
Waterhouse looks so ill that I think it desirable not to move the party
to-day, and trust by to-morrow he will be quite well. Light wind from the
south-east, with a few clouds.

Tuesday, 10th June, Daly Waters. As Mr. Waterhouse is better, I shall
move the party to-day. Started at half-past eight a.m., following my
former tracks. At half-past four p.m. camped at the blue-grass swamp;
twenty-six miles without water. The horses will require to be watched
during the night. Wind, south-east. Day very hot. Latitude, 15 degrees 56
minutes 31 seconds.

Wednesday, 11th June, Blue-Grass Swamp. Started at seven o'clock; course,
70 degrees east of north. At three miles crossed the ironstone rise, and
at eleven miles changed to north, to cut the chain of ponds, which I have
named Purdie Ponds, in honour of Dr. Purdie, of Edinburgh, M.D. At one
mile and three quarters, on the last course, camped on the largest pond.
The country that we have gone over, although there are a number of holes
and cracks in it, is really of the best description, covered with grass
up to the horses' bodies. We have passed several new trees and shrubs.
The bean-tree is becoming more numerous here. At this season and in this
latitude it sheds its leaves; the flower is in full bloom without them.
The course of the ironstone rise seems to be north and south. Wind,
south-east. Weather a little cooler, but clouds all gone. Latitude, 15
degrees 52 minutes 58 seconds.

Thursday, 12th June, Purdie Ponds. Preparing for another start to-morrow
with the water-bags. It takes two men nearly half a day to fill them. The
orifices for filling them are a great deal too small; they ought to be at
least two inches in diameter. The American cloth with which they are
lined is useless in making them watertight, and is a great annoyance in
emptying them, for the water gets between it and the leather. It takes a
long time to draw through again, and does not answer the purpose it was
intended for. A piece of calico would have done far better. It is very
vexing to bring things so far, and, when required, to find them nearly
useless. Wind, south-east. Cloudy. Nights cold, but the day hot.

Friday, 13th June, Purdie Ponds. Leaving Mr. Kekwick in charge of the
party, started at fifteen minutes past eight with Thring and Auld, also
with King and Billiatt, who were to bring back the horses carrying the
water-bags. Proceeded on a north course, and at seven miles crossed what
seemed to be a water-shed, seemingly running to the west of north. Halted
the party, and sent Thring a short distance to see if the flow was in
that direction. In a quarter of an hour he returned and informed me that
it was, but only very slightly so. Changed to north-north-west to follow
it. It gradually assumed the appearance of a small creek. At two miles
came upon three small pools of water. I now resolve to follow it down and
see where it goes to. I should think there must be more water further on.
Its course is west of north. Continued to follow it down, winding and
twisting about very much to almost every point of the compass. At seven
miles from the pools found a little more water, but not a drop between.
Allowed the horses to drink what there was, and proceeded down it. I sent
Thring to follow it on one side, while I and the rest of the party kept
on the other. By this we were enabled to cut off the bends and see all
the creeks, so that no water could escape us. Twice it became very small,
and I was afraid we were going to lose it altogether, but it commenced
again and became a fine creek. Not a drop of water. At a quarter to five
camped without it. Stony rises are now commencing, which are covered with
gum and other trees, also a low scrub. They are very rough and running
nearly west and south. The one on the west is a continuation of the one I
crossed in coming to Purdie Ponds. The general flow of the creek is
north. Some of the new trees are growing very large on its banks. The
cabbage-tree is growing here also. This is the first time I have met with
it, sometimes growing to the height of fifteen feet. All along the banks
of the creek, and apparently for some distance back, is covered with an
abundance of grass, but all dried up. In some places both horse and rider
were completely hidden by it. Wind, south-east--few clouds. Latitude, 15
degrees 30 minutes 27 seconds.

Saturday, 14th June, River Strangways. Named after the Honourable H.B.
Templar Strangways, Commissioner of Crown Lands, South Australia, and
who, since his taking office, has done all in his power to promote
exploration of the interior. Sent King and Billiatt back with the horses
to the camp at Purdie Ponds, whilst I proceed with the further
examination of the creek. I find it now running to the east of north, and
the stony rises are closing upon it at two miles and a half. They begin
to assume the shape of hills, which causes the travelling to be rather
rough. At three miles and a half the hills run close to the creek, and
are precipitous; the bed is very rough and stony--so much so that I could
not take the horses down it. Ascended a hill near the creek to see what
it and the country ahead was like; the hills being so rough that I could
not get the horses close enough to see if there was any water, dismounted
and scrambled to the top of the precipices; was delighted to see below me
a large hole of water. Sent the horses across a gully to another hill
still higher, while I descended into the creek; found the bed very rough,
having large masses of sandstone and ironstone, which rendered it
impassable for the horses. Found the water to be deep and beautifully
clear; proceeded down a little further, and saw another large one. The
hills close to the creek are very precipitous, and we shall have
difficulty in getting the horses down to water; the hills, where they
come close to the creek, are covered with spinifex. I shall therefore
require to camp the party at the mouth of the gorge, where there is
plenty of feed. The hill I had sent the horses to was so rocky they were
unable to cross it, and there being higher hills still on ahead, I have
left the horses with Auld, and, taking Thring with me, have walked to the
top of it to see what course the creek was taking, but they are all so
much of the same height and appearance that I can scarcely tell in which
direction it runs. There is an appearance of a large creek coming in from
the westward, and higher hills towards the north. I shall return and send
the party on to this permanent water, and try to find an easy road over
the ranges for them. I would have gone on to-day, but my horses are
without shoes, and some of them are already lame, and the shoes I brought
with me are nearly all exhausted; we have not been using any since
shortly after leaving South Australia. Returned to our last night's camp,
where we had left the canvas tank with some water that the horses did not
drink in the morning; gave them what remained, and proceeded up the
creeks to the last water we saw yesterday, where we arrived at sundown
and camped. Wind, south.

Sunday, 15th June, River Strangways. Returned to the Depot at Purdie
Ponds; found all well. Wind, south-east. Cool.

Monday, 16th June, Purdie Ponds. It was late before the horses could be
found. Proceeded to the first pool of water in the River Strangways,
distance about ten miles, and camped. Wind, south-east.

Tuesday, 17th June, River Strangways. Proceeded down the creek to the
gorge and camped; day very hot. We had some difficulty in finding a way
down for the horses to drink, it being so very rough and stony, but at
last succeeded. On the west side there is a layer of rocks on the top of
the hard sandstone, black and rugged, resembling lava; spinifex close to
the creek. Wind, south-east.

Wednesday, 18th June, Gorge, River Strangways. I shall require to have
some of the horses shod for further exploration, and shall therefore
remain here to-day to get that done. I sent Thring and King a little way
down the creek to see what the country is, and if there is any more
water. They went about nine miles, but could see no more. In some places
the country is sandy, and in others stony and grassy. Mr. Kekwick has
discovered four new trees that we have not seen before, and several new
shrubs. Some of the party succeeded in catching a few fine large fish,
some of them weighing two pounds and a half. Some were of the perch
family, and others resembled rock cod, with three remarkable black spots
on each side of their bodies. There are also some small ones resembling
the gold fish, and other small ones with black stripes on their sides,
resembling pilot fish. Wind, south-east. Latitude, 15 degrees 30 minutes
3 seconds.

Thursday, 19th June, Gorge, River Strangways. Leaving Mr. Kekwick in
charge of the party, started with Thring, Auld, and King, to look for
water. No rain seems to have fallen here for a long time back; the grass
is quite dry and withered. At 8.15 proceeded down the river, and, to
avoid the hills, I went about a mile to the west, and found a very
passable road; for about two miles we had sandy soil and spinifex mixed
with grass, also a few stony rises of lime and sandstone. The country
after that again became excellently grassed, the soil light and a little
sandy. No water in the bed, which appears to have a very rapid fall; its
general course is about north-north-east. At twelve miles, seeing a stony
hill of considerable elevation, I left the bed, and went towards it. At
the base of it was a deep creek; I was pleased to see a fine supply of
water in it. I immediately sent Thring back to guide the party up here
to-morrow, whilst I with the two others proceeded with the examination of
the river further down. After following it for about ten miles through a
beautifully grassed country, passing occasionally sandstone rises, with
apparently scrub on their tops, camped at the base of one of them.

Friday, 20th June, First Camp North of Gorge. Returned to the other
water, and at noon met the party and brought them on to this water. We
have passed a few stringy-bark trees. In the bed of the river there is
growing some very large and tall timber, having a dark-coloured bark, the
leaf jointed the same as the shea-oak, but has not the acid taste: the
horses eat it. There are also some very fine melaleuca-trees, which here
seem to displace the gums in the river. We have also passed some more new
trees and shrubs. Frew, in looking about the banks, found a large creeper
with a yellow blossom, and having a large bean pod growing on it. I shall
endeavour to get some of the seed as we go on to-morrow. I shall now move
on with the whole party, and I trust to find water in the river as long
as I follow it; its banks are getting much deeper and broader, and likely
to retain water; it is dreadfully slow work to keep searching for water.
Before this I could not do otherwise, in consequence of the season being
so very dry. Since the commencement of the journey the only rain that we
have had to have any effect upon the creeks was at Mr. Levi's station,
Mount Margaret. Since then we have had only two or three showers, which
have had no effect upon the creeks. Light winds, south-east. Latitude, 15
degrees 15 minutes 23 seconds.

Saturday, 21st June, First Camp North of Gorge. It was late before we
could get a start, in consequence of our not being able to find two of
the horses which separated from the rest during the night. Started,
following the river down; it frequently separating into two or three
channels, and again joining. Numerous small watercourses are coming in on
both sides, from east and west; it winds about a great deal--its general
course to-day for nine miles has been nearly north-north-east. We passed
a number of large lagoons, nearly dry, close to the stony hills: when
full they must retain water for a long time. There is very little water
in the main channel. At nine miles I found a large and excellent pool of
water in one of the side creeks; it will last some time. It being now
afternoon, and there being a nice open plain for the horses, I have
camped. The river is now running through stony hills, which are very
rough, composed of hard sandstone mixed with veins of quartz, some of
which are very hard, much resembling marble with crystalline grains in
it. We are now passing a number of stringy bark along with gum and other
trees, Mr. Kekwick still finding new shrubs. After we had camped, taking
Thring with me, I ascended a hill a little way from the camp, but was
disappointed in not having an extensive view. To the north, which is now
apparently the course of the river, there seems to be an opening in the
range of stony hills. The dip of the country seems to be that way. At 33
degrees east of north from the camp, about eight miles distant, there is
a high wooded tent-hill on the range; this I have named Mount Muller,
after my friend the well-known botanist of Victoria. All round about are
rough stony hills with grassy valleys between, having spinifex growing on
their sides and tops. The valley through which the main channel flows is
good soil, and covered with grass from two to four feet high. Towards the
north-west the hills appear to be very rugged. Wind south-east, with a
few clouds. Latitude, 15 degrees 10 minutes 40 seconds.

Sunday, 22nd June, Rock Camp, River Strangways. A few heavy clouds about.
We are now in the country discovered by Mr. Gregory. There is a great
deal of very good timber in the valley, which is getting larger and
improving as we advance. It is still very thick--so much so, that the
hills cannot be seen until quite close to them. Wind variable. Latitude,
15 degrees 10 minutes 30 seconds.

Monday, 23rd June, Rock Camp, River Strangways. This morning the sky is
overcast with light clouds coming from the south-east. Started at eight
o'clock, still following the river, which winds about very much; its
general course 10 degrees east of north. At nine miles the channel became
much smaller, and shortly afterwards separated into numerous small ones,
and was apparently lost to me. I continued a north course, and at twelve
miles struck a creek coming from the south-east; at two miles from this
creek found another large one coming from the south-west, with shea-oak
in it, which makes me suppose it is the River Strangways, and that it
formed again and joined this one. At the junction were numerous recent
fires of the natives; there must have been a great many of them, for
their fires covered the ground, also shells of the mussel which they had
been eating. Searched for water, and found a little, but not sufficient
for my horses, and too difficult to approach; the course of the river is
still to the north. One mile and a half from the junction found enough
water that will do for me at night. As there seems to be so little water,
and this day being exceedingly hot and oppressive, I have camped. The
country travelled over to-day has been of the same description,
completely covered with long grass; the soil rich, and a great many of
the cabbage-tree growing about it. Wind variable. Latitude, 14 degrees 58
minutes 55 seconds.

Tuesday, 24th June, Mussel Camp, River Strangways. With the sun there
came up a very thick and heavy fog which continued for about two hours;
it then cleared off and the day became exceedingly hot. The river, after
rounding the hills (where we were camped), ran nearly east for three
miles, meeting there a stony hill which again throws it into a northerly
course. I ascended the hill, but could see nothing distinctly, the fog
being so thick. Descended and pursued the bed, which separated frequently
into many channels, and at ten miles it spread into a large area, and its
courses became small with no water in them. The grass above our heads was
so high and thick that the rear-party lost me and could not find the
rocks; by cooeing I brought them to me again. Before I had heard them I
had sent Thring back to pick up their tracks and bring them to the clear
ground I was on with the rest of the party, but they arrived before he
made up to them. The scrub is also very thick close to the river. Mr.
Kekwick found cane growing in the bed, and also brought in a specimen of
a new water-lily--a most beautiful thing it is; it is now in Mr.
Waterhouse's collection. At twelve miles, finding some water, the horses
being tired in crossing so many small creeks, and working through the
scrub and long grass, I camped at the open ground. The country gone over
to-day is again splendidly grassed in many places, especially near the
river; it has very lately been burned by the natives. There are a great
number of them running along the banks; the country now seems to be
thickly inhabited. Towards the east and the north-east the country is in
a blaze; there is so much grass the fire must be dreadful. I hope it will
not come near us. The day has been most oppressively hot, with scarcely a
breath of wind. Latitude, 14 degrees 51 minutes 51 seconds.

Wednesday, 25th June, River Strangways. Two of the horses having
separated from the others, and crossing the river, quite hidden in the
long grass, it was late before they were found. Started at nine o'clock;
course about 70 degrees east of north, following the channel. I expect,
in two or three miles, to meet with the Roper. At three miles struck a
large sheet of deep clear water, on which were a number of natives, with
their lubras and children; they set up a fearful yelling and squalling,
and ran off as fast as they could. Rounded the large sheet of water and
proceeded along it. At a mile, three men were seen following; halted the
party, and went up to them. One was a very old man, one middle-aged, the
third a young, stout, well-made fellow; they seemed to be friendly. Tried
to make them understand by signs that I wished to get across the river;
they made signs, by pointing down the river, by placing both hands
together, having the fingers closed, which led me to think I could get
across further down. They made signs for us to be off, and that they were
going back again. I complied with their request, and after bidding each
other a friendly good-bye, we followed down the banks of the river, which
I now find is the Roper. At seven miles tried to cross it, but found it
to be impossible; it is now divided into a number of channels, very deep
and full of running water. Proceeded further, and tried it at several
places, but with the same result. At twelve miles, camped close to a
steep rocky hill on the north side of the river. Searched all round for a
crossing, but was unable to find one. To the eastward the country is all
on fire. The banks of the river are thickly lined with cabbage-trees,
also the cane, bamboo, and other shrubs. Two small turtle-shells were
picked up by the party at the native camp. The country is still of the
same fine description. We are now north of Mr. Gregory's tracks.
Latitude, 14 degrees 5 minutes. Wind variable.

Thursday, 26th June, Roper River. As I cannot find a crossing, I shall
have to return to my last camp and try to cross there. Arrived and
camped. Day again oppressively hot. Almost immediately on leaving our
camp this morning I observed native tracks on ours close to it. They must
have followed us up last night, although we saw nothing of them. They are
not to be trusted: they will pretend the greatest friendship one moment
and spear you the next. They have been following us to-day, but keeping
on the other side of the river and setting fire to the grass as they go
along. I wish it would rain and cause the grass to become green, so as to
stop them burning, as well as to give me some fresh food for the horses,
for they now begin to show the want of it very much; it is so dried up
that there is little nourishment in it. Some of them are beginning to
look very poor and are much troubled with worms. My journeys have been
very short last week, in consequence of my being so weak from the effects
of scurvy and a severe attack of dysentery, for I have scarcely been able
to endure the motion of horseback for four hours at a time; but having
lately obtained some native cucumbers, I find they are doing me a deal of
good, and hope by next week to be all right again. Wind, south. Latitude,
14 degrees 51 minutes 51 seconds.

Friday, 27th June, West Roper River. Started on a course of 320 degrees,
crossing the river, and at three miles and a half again struck the Roper,
running. Followed it up, coming nearly from the west, but winding about
very much, and having many branches, which makes it very difficult for me
to get the turns correctly. It is a splendid river. We have passed many
brooks and deep reaches of water some miles in length, and the country
could not be better: it is really magnificent. At 2.30 I was informed
that we were short of a horse. Sent Messrs. Kekwick and Thring back to
see where he was left. We have had to cross so many boggy, nasty places,
with deep water and thick scrub, that he must have been missed at one of
these. The general course of the river to-day has been 280 degrees.
Distance, fifteen miles. Messrs. Kekwick and Thring are returned. They
found the horse bogged in a side creek. It was so thick with cabbage-tree
that they passed in searching for him two or three times. They had great
difficulty in getting him out, but at last succeeded, and arrived at the
camp before dark. A short time before that, another horse got into a very
deep and rapid channel of the river, the top of the banks projecting so
much that he could not get out, and the gum-trees having fallen across
both above and below him, he was completely fixed. We endeavoured to get
him out, but it got so dark that we could not see him, and the rope
breaking that we were pulling him out by, he got his head under water,
and was drowned in a moment. We then found that the cause of the rope
breaking was that he had got one of his hind feet entangled in a sunken
tree. It being now so dark we can do no more to-night, and have left him
in the water until daylight. Wind, south-east. Latitude, 14 degrees 47
minutes 26 seconds.

Saturday, 28th June, Roper River. As I shall be short of meat, I remain
here to-day to cut up the horse and dry him. The water of this river is
most excellent; the soil is also of the first description; and the grass,
although dry, most abundant, from two to five feet high. This is
certainly the finest country I have seen in Australia. We passed three
rocky hills yesterday, not high, but having grass up to their tops, round
which the river winds at their base, forming large and long reaches of
water. On the grassy plains it forms into different channels, and is
thickly timbered with shea-oak, gum, cabbage-trees, and other trees and
shrubs. Wind variable.

Sunday, 29th June, Roper River. We are all enjoying a delightful change
of fresh meat from dry. It is a great treat, and the horse eats
remarkably well, although not quite so good as a bullock. At sundown the
meat is not all quite dry, but I think we shall be able to preserve the
greater part of it. The natives are still burning the grass round about
us, but they have not made their appearance either yesterday or to-day.
Wind variable.

Monday, 30th June, Roper River. Started at 8.10, course west, following
up the river, which winds about very much from north-west to south, and
at last to south-east. When coming close to where the grass was on fire,
finding a good ford, I crossed the party to the north-east side. At
fifteen miles came upon a large reedy swamp through which the river
seemed to flow, and again at twenty miles came upon the river running
into the swamp, and coming from the north-north-west. Although travelling
twenty miles we have not made more than ten miles in a straight line; the
general course is west. The country is of the same excellent description.
We have passed the stony rises on the north side of the river, which are
covered with grass to their tops. After crossing the river I ascended
another of the same kind. To the south are a few hills scattered over the
grassy plains, with lines of dark-green trees between them, showing that
they are creeks flowing into the river whose junctions we have been
crossing to-day; the same to the south-west, and at west 20" south the
distance appears level, with a single peak just visible. To the
north-west seemingly stony hills; to the north the same; to the east I
could see nothing, for the smoke conceals from me the country; it is all
on fire. The river is still running very rapidly, and as this is a
different branch from those previously discovered, I have named it the
River Chambers, after my late lamented friend, James Chambers, Esquire,
whose zeal in the cause of Australian exploration is already well known.
A short time before sundown a number of natives were seen approaching the
camp. We were immediately prepared for them. I sent Mr. Kekwick forward
to see what their intentions were--friendly or hostile. I immediately
followed. On reaching them they appeared quite friendly. There were three
men, four lubras, and a number of children. One, an old man, presented a
very singular appearance--his legs being about four feet long, and his
entire height seven feet, and so remarkably thin that he appeared to be a
perfect shadow. Mr. Kekwick having a fish-hook stuck in his hat, which
immediately caught the tall old fellow's eye, he made signs of its use,
and that he would like to possess it. I told Mr. Kekwick to give it to
him, which seemed to please him much. After examining it he handed it
over to a young man, seemingly his son, who was a fat, stout fellow, and
who was laughing nearly all the time. The other was a middle-aged man of
the ordinary height. The women were small, and very ugly. Wind,
south-east. Latitude, 14 degrees 47 minutes 24 seconds.

Tuesday, 1st July, Reedy Swamp, River Chambers. Before sunrise the
natives again made their appearance, sixteen in number, with small
spears. Sent Mr. Kekwick to see what they wanted. On his coming up to
them they put two fingers in their mouths, signifying that they wanted
more fish-hooks, but we had no more to spare. They remained looking at us
until the horses were packed and started. After Thring and Frew had
brought in the horses, they rode up to where they were. They (the
natives) did not fancy being too near the horses, but having dismounted,
it gave them confidence, and they returned again. Thring opened the lips
of one of the horses, and showed them his teeth, the appearance of which
did not suit their taste. Some of them thought the further off they were
from such weapons the better, and ran off the moment they saw them.
Others remained, but kept at a respectful distance. Thring pulled a
handful of grass, and it amused them much to see the horses eating it.
After starting they followed us for some miles, when Mr. Waterhouse,
observing a new pigeon, shot it. They, not liking the report of the gun,
went off, and we saw no more of them. Started at 8.20, following the
river on a course 30 degrees east of north. After a mile it gradually
came round to the south-east, and was a running stream in that direction.
As that course would take me too much out of my road, I changed my
bearing to north-west, to an opening between the hills. After passing a
number of fine ponds, many of them with water in them, came upon a large
creek, having long reaches of water in it, but not running. It winds
about a great deal. Its general course to-day has been west-north-west.
The reedy swamp must be a mass of springs, which causes the Roper to run
with such velocity. A little after one o'clock camped. The journey to-day
has been rough, having so many small creeks to cross, and the day being
excessively hot, the horses seem fagged. They have been covered with
sweat since shortly after starting until now, and as some of the drowned
horse is not quite dry, I have halted earlier than I intended. The
country gone over to-day is of the same kind, beautiful soil, covered
with grass. We occasionally met with stony hills coming down to the
creek, also well grassed and timbered to their tops. Wind west, with
heavy clouds from the south-east. Latitude, 14 degrees 41 minutes 39
seconds.

Wednesday, 2nd July, West-north-west of Reedy Swamp, River Chambers.
Started 7.40, following the river up until ten o'clock. We kept nearly a
north-west course: it then went off to the south-west; as that would take
me too much out of my course, I kept the north-west course, crossing the
saddle of broken hills, amongst which we have now got; and at twelve
again met the river, now coming from the north through the hills,
following it still, having plenty of water. At a very large water hole
surprised some natives, who ran off at full speed when the rear of the
party was passing their camp. One stout fellow came running up, armed
with spears, and loaded with fish and bags filled with something to eat.
Mr. Kekwick rode towards him. The native held up a green bough as a flag
of truce, and patting his heart with his right hand, said something which
could not be understood, and pointed in the direction we were going. We
then bade him good-bye, and proceeded on our journey. At one o'clock the
river suddenly turned to the east, coming from very rough hills of
sandstone and other rocks. At one mile and a half on that course it was
coming from the south of east, which will not do for me. Changed to the
north, and got into some terrible rough stony hills with grassy valleys
between, but not a drop of water. It being now after two o'clock, too
late to encounter crossing the table land, I again changed my course to
south-east for the Chambers, and at 5.3 camped at a large water hole at
the foot of a stony rise lined with cabbage (palm) trees. The country
although rough is well grassed to the top of the hills, with an abundance
of permanent water in the river. I am sorry it is coming from the
south-east, and have been in hopes it would carry me through this degree
of latitude. To follow it further is only losing time; I shall therefore
take to the hills to-morrow. Frew, on coming along, picked up a small
turtle alive. Light wind from the south-east; heavy clouds from the
south-west. Latitude, 14 degrees 32 minutes 30 seconds.

Thursday, 3rd July, River Chambers. Started at 8.10 o'clock, north-west
course. At one mile and a half again struck the river coming from the
west-north-west; left it and followed its north-west course: and at
another mile again came upon it with plenty of water. Saw four natives,
who ran off the moment they saw us. Followed the river, the hill coming
quite close to it, very steep and rocky, composed of a hard sandstone,
and occasionally a little ironstone. At nine miles again left the river,
finding it was coming too much from the eastward; crossed the saddle of
the two spurs again; came upon a creek, which I think is the river; ran
it up to the west for about a mile, but no appearance of water; left it,
and ascended a very rough rugged hill. In the creek we have just left
there is a deal of limestone. Crossed three more small spurs and small
creeks, but not a drop of water. It being now afternoon, and wishing to
see from what direction the river is coming, I changed to north-east, but
found that I was still among the rough hills; I then went east for a
short distance, and made the river, now quite dry, and having a sandy
bed. Followed it up, but saw there was no hope of water; turned, and
traced it down to try and find water. After following it for three miles,
came upon a fine permanent hole of water, a short distance from where we
left in the former part of the day. If it would only rain and put some
water in the deep dry holes that are in the other creeks crossed to-day,
I should then be enabled to steer a straight line for the Adelaide. It is
very tedious and tiresome having to look for water every day. We have now
reached to the top of one of the tributaries of the Chambers. This is
apparently the last water. It seems to take its rise in a grassy plain to
the east of this. The valley through which the creek flows is well
grassed, but the sides and the tops of the hills are spinifex mixed with
grass. All the small valleys are well grassed. Wind, south-east.
Latitude, 14 degrees 26 minutes 50 seconds.

Friday, 4th July, Last Water Hole in the Chambers. Started at 8.10,
course north-west, following up the river to its sources. At four miles
ascended a rise, which was very rough, composed of sandstone, ironstone,
and limestone, with ironstone gravel on the top. Descended on the other
side, and at about five miles came upon a nice running stream, but very
rough and stony round about it. After crossing several stony rises, in
which we had some difficulty in getting our horses over, arrived at a
nice broad valley with a creek running through it, course north-west. At
a mile it received a large tributary from the east of north, and the bed
seems sandy; melaleuca and gum-trees in it; also the bean-tree. The
valley is covered with grass from two to four feet high. There is a ridge
of rough sandy stone hills, with occasional ironstone on each side, from
the direction it was at first taken. I thought I was fortunate in meeting
with one of the sources of the Alligator or Adelaide River. After
following it for five miles, sometimes going west and south, it went
through a stony gorge, and seemed to run to the south, which is a great
disappointment. I ascended one of the hills to view the country, but
could see very little, it being so thickly wooded. To the north is the
appearance of a range running to the east and west that I must endeavour
to cross to-morrow if I do not find another creek running to the
north-west. There is one benefit I shall derive from following down this
creek a day; it will enable me to round the very rough sandstone range
that runs on the north side of the creek. It is so rough that I could not
take the horses over it. Camped at the gorge of this creek, which I
suppose, from the course it is now taking, to be another tributary of the
Chambers. The gorge is impassable for horses. It has a very picturesque
appearance; immense masses of rock--some thousands of tons in
weight--which had fallen from the top of the cliff into the bed of the
creek. Mr. Kekwick found a number of new plants, among them a fine
climbing fern. Light winds, east. Plenty of permanent water in the creek.
Latitude, 14 degrees 25 minutes 8 seconds.

Saturday, 5th July, Gorge on another West Branch of the River Chambers.
Started 8.15; course, 5 degrees west of north. After travelling two miles
over stony rises we ascended a low table land with coarse grass and a
little spinifex; at six miles came up to a high stony tent-hill, which I
ascended and named Mount Shillinglaw. All round are stony hills and
grassy valleys--dip of the country seemingly to the south. There is
apparently a continuous range in the distance to the north-west, the
Chambers range. Changed my course to 325 degrees, and at four miles
struck another large branch coming from the north-east, and running
apparently south--plenty of water in it. This I named the Waterhouse, in
honour of Mr. H.W. Waterhouse, naturalist to the expedition. Some of the
horses are become so lame on account of the stones they will not be able
to travel another day. I have camped early to have them shod, for on
Monday I intend taking a north-west course to strike the source of the
Adelaide. The country on the last course is again of the very best
description and well grassed. The hills are stony, but abound with grass;
they are composed of sandstone, ironstone, and occasionally a little
limestone; the trees are the same as those on the Roper. Wind,
south-east. Latitude, 14 degrees 18 minutes 30 seconds.

Sunday, 6th July, The Waterhouse River. Day again very hot. There is
another branch a short distance off, which seems to come from the
north-west; I shall follow it to-morrow if it continues the same course.
I think these creeks we are now crossing must be the sources of the
Adelaide flowing towards the dry river seen by Mr. Gregory running
towards the north-west. Wind light; sky cloudy.

Monday, 7th July, Waterhouse River. Started at 7.55; course, north-west.
At four miles the creek was coming from the west, north-east, and east; I
therefore left it, crossed two low stony rises, and again struck another
creek coming from the north-east, with plenty of water; followed it for a
short distance to the west, found it so boggy and the body of water so
large that I could not get the party round the stony hills. Returned
about half a mile, and crossed the stony rise, and again struck it. At
eight miles came upon a number of springs coming from the stony rises.
Ascended one of the rises, which are not high, and found myself on a
sandy table land, which continued for six miles, having coarse grass and
spinifex growing on it. Towards the last two miles it again became well
grassed. The timber is stringy-bark, some splendid trees; amongst them
gums and a number of pines, also very fine. The cabbage palm still
growing in the creeks in great numbers, some of them very tall, with
several branches on the top. The first eight miles was again over a
splendid country, and the last three of the same description. A stony
hill being in my course, I proceeded to the top of it, from which I had a
good view of the country before me. This hill I named after Lieutenant
Helpman. At 10 degrees south of west are two remarkable isolated table
hills, Mount Levi and Mount Watts, beyond which is the Chambers range to
the north-west; my view in other directions is obstructed by other hills,
but to the west about one mile and a half is seemingly a creek, to which
I shall go, and if there is water I shall camp. Proceeded and found it a
fine creek with plenty of water; followed it about one mile to the
north-west, when it became dry. There it seems to come from the south.
There are a great number of cabbage palms on its banks. I hope it will
soon come round to the north-west and continue on that course. Light
winds, variable. Latitude, 14 degrees 9 minutes 31 seconds.

Tuesday, 8th July, Water Creek in Stony Rises. Started at 7.40 a.m.,
course north-west; followed the creek a little way, but found it was
running too much to the west of my course; left it and proceeded to the
north-west, crossing some stony rises, now composed of granite and
ironstone, with occasionally some hard sandstone. Crossing three small
creeks running to the west, at six miles came upon a large one with broad
and long sheets of permanent water coming from the north-north-east, and
apparently running to the south-west. This I have named the Fanny, in
honour of Miss Fanny Chambers, eldest daughter of John Chambers, Esquire.
In a small tree on this creek the skull of a very young alligator was
found by Mr. Auld. The trees in this creek are melaleuca and gum, with
some others. Proceeded across the creek, still going north-west; ascended
two stony rises, and got upon low table land with spinifex and grass,
passing two stony hills, one on each side of my course. At eighteen miles
struck the head of a small creek flowing nearly on my course; followed it
down in search of water, now through a basaltic country. At two miles
came upon another large creek, having a running stream to the south of
west, and coming from the north of east. Timber, melaleuca, palm, and
gum, with some of other descriptions. This I have named the Katherine, in
honour of the second daughter of James Chambers, Esquire. The country
gone over to-day, although there is a mile or two of light sandy soil, is
good for pasturage purposes; in the valley it is of the finest
description. Light winds, variable. Latitude, 13 degrees 58 minutes 30
seconds.

Wednesday, 9th July, The Katherine. Started at five minutes to eight
o'clock, crossing the Katherine, and proceeded on a north-west course
over a basaltic country, splendidly grassed. At five miles I ascended a
high hill, which I named Mount Stow, but was disappointed in the view.
West-north-west course, over a great number of rises thickly timbered
with gum. At 20 degrees north of west is a high bluff point of the range;
the country on that bearing does not seem to be so rough. No more visible
but the range to the west and the hill between. Descended, and changed my
course to the bluff point. At one mile and a half crossed a creek with
water in it, coming from the north-east, and running to the south-west.
At three miles further arrived at the bluff. The basaltic country has now
suddenly changed to slate, limestone, sandstone, and a hard white stone.
Crossed three stony rises, and got upon a white sandy rise, with large
stringy-bark trees growing upon it; and there seemingly being a creek at
the foot of it, from the number of green gums and palm-trees, I went down
to it, and found it to be springy ground, now quite dry, although the
grass was quite green. Proceeded on the westerly course, expecting to
meet with a creek; found none, but large springs coming from sandy rises.
Having found water at thirteen miles, and being so very unwell that I
cannot proceed, I have been compelled to camp. There is an immense
quantity of water coming from these springs; the soil round them is of
the best deep black alluvial. About a mile to the west is a strong stream
running to the south-west from them. I have called them Kekwick Springs,
in honour of my chief officer. Wind light and variable. Latitude, 13
degrees 54 minutes 12 seconds.

Thursday, 10th July, Kekwick's Large Group of Springs. Started at eight
o'clock; crossed the springs without getting any of the horses bogged.
Proceeded on a north-west course, but at a mile and a half again came
upon springs and running water; the ground too boggy to cross it. Changed
to north; at three miles and a half on the course changed to north-west.
Ascended some very rough stony hills, and got on the top of sandy table
land thick with splendid stringy-bark, pines, and other trees and shrubs,
amongst which, for the first time, we have seen the fan palm, some of
them growing upwards of fifteen feet high; the bark on the stem is marked
similar to a pineapple's; the leaf very much resembles a lady's fan set
on a long handle, and, a short time after it is cut, closes in the same
manner. At half-past one crossed the table land--breadth thteen miles.
The view was beautiful. Standing on the edge of a precipice, we could see
underneath, lower down, a deep creek thickly wooded running on our
course; then the picturesque precipitous gorge in the table land; then
the gorge in the distance; to the north-west were ranges of hills. The
grass on the table land is coarse, mixed with a little spinifex; about
half of it had been burnt by the natives some time ago. We had to search
for a place to descend, and had great difficulty in doing so, but at last
accomplished it without accident. The valley near the creek, which is a
running stream, is very thickly wooded with tall stringy-bark, gums, and
other kinds of palm-trees, which are very beautiful, the stem growing
upwards of fifty feet high, the leaves from eight to ten feet in length,
with a number of long smaller ones growing from each side, resembling an
immense feather; a great number of these shooting out from the top of the
high stems, and falling gracefully over, has a very pretty, light, and
elegant appearance. Followed the creek for about two miles down this
gorge, and camped on an open piece of ground. The top course of the table
land is a layer of magnetic ironstone, which attracted my compass upwards
of 20 degrees; underneath is a layer of red sandstone, and below that is
an immense mass of white sandstone, which is very soft, and crumbling
away with the action of the atmosphere. In the valley is growing an
immense crop of grass, upwards of four feet high; the cabbage palm is
still in the creek. We have seen a number of new shrubs and flowers. The
course of the table land is north-north-west and south-south-east. The
cliffs, from the camp in the valley, seem to be from two hundred and
fifty to three hundred feet high. Beyond all doubt we are now on the
Adelaide river. Light winds, variable. Latitude, 13 degrees 44 minutes 14
seconds.

Friday, 11th July, Adelaide River, North-west Side, Table Land. The
horses being close at hand, I got an early start at 7.20, course
north-west. In a mile I got greatly bothered by the boggy ground, and
numbers of springs coming from the table land, which I am obliged to
round. At two miles got clear of them, and proceeded over a great number
of stony rises, very steep; they are composed of conglomerate quartz,
underneath which is a course of slates, the direction of which is
north-west, and lying very nearly perpendicular, and also some courses of
ironstone, and a sharp rectangular hard grey flint stone. My horses being
nearly all without shoes, it has lamed a great many of them, and, having
struck the river again at fifteen miles, I camped. They have had a very
hard day's journey. The country is nearly all burnt throughout, but those
portions which have escaped the fire are well grassed. I should think
this is a likely place to find gold in, from the quantity of quartz, its
colour, and having so lately passed a large basaltic and granite country;
the conglomerate quartz being bedded in iron, and the slate
perpendicular, are good signs. The stony rises are covered with
stringy-bark, gum, and other trees, but not so tall and thick as on the
table land and close to it, except in the creek, where it is very large;
the melaleuca is also large. Since leaving the table land we have nearly
lost the beautiful palms; there are still a few at this camp, but they
are not growing so high; the cabbage palm is still in the creek and
valleys. Light winds from south-east. Country burning all round.
Latitude, 13 degrees 38 minutes 24 seconds. This branch I have named the
Mary, in honour of Miss Mary Chambers.

Saturday, 12th July, The Mary, Adelaide River. Started at 7.30; course,
north-west. At one mile and a half came upon a running stream coming from
the north-east; had great difficulty in getting the horses across, the
banks being so boggy. One got fixed in it and was nearly drowned; in an
hour succeeded in getting them all safe across. At six miles I ascended a
high, tall, and stony hill; the view is not good, except to the westward.
In that direction there is seemingly a high range in the far distance,
appearing to run north and south; the highest point of the end of the
range is west, to which the river seems to tend. My horse being so lame
for the want of shoeing, I shall strike in for the river and follow it
for another two miles, as it seems to run so much to the westward. I have
resolved to use some of the horseshoes I have been saving to take me back
over the stony country of South Australia. To enable McGorrerey to get
them all shod on the front feet before Monday, I have camped. There is
still a slaty range on each side of the river, with quartz hills close
down to it; the timber the same as yesterday. The country has recently
all been burned; but, judging from the small patches that have escaped,
has been well grassed up to the pass of the hills. The valley and banks
of the creeks are of beautiful alluvial soil. One new feature seen to-day
is the growing of large clumps of bamboo on the banks of the river, from
fifty to sixty feet in height and about six inches in diameter at the
butt. I am now on one of the tributaries of the Adelaide River. There
must have been a dreadful fire here a few days ago; it has destroyed
everything before it, except the green trees, to the edge of the water.
Slight winds, variable. Latitude, 13 degrees 35 minutes 58 seconds.

Sunday, 13th July, The Mary, Adelaide River. Shoeing horses. Wind blowing
strong; variable from all points of the compass.

Monday, 14th July, The Mary, Adelaide River. One of the horses cannot be
found this morning, and he has been for some time very ill and weak, and
no appearance of getting better. It was my intention to have left him. We
have been all round the tracks forward and backward over the
feeding-ground and can see nothing of him. I am afraid he has gone off to
some place and died; I shall therefore waste no more time in looking for
him. If he is alive I may have a chance of recovering him on my return.
Late start, in consequence of so long looking for him. As I have now got
all the horses shod on the front feet, I shall proceed on a north-west
course through the stony rises, which are still quartz and slate,
splendidly grassed, with gums and other trees and bushes not too thick to
get through with ease. Crossed six small creeks, one with holes with
water in them; the third one, a large creek, which I crossed at nine
miles, I have named William Creek, after the second son of John Chambers,
Esquire, of Adelaide; all running at right angles to my course.
Immediately after crossing this last creek the country changed to
granite; the rises are composed of immense blocks of it, with
occasionally some quartz. The country has been all burned. The valleys
between the granite rises are broad and of first-rate soil, many of them
are quite green, caused by springs oozing from the granite rock. We have
passed a number of trees resembling the iron-bark, also some like new
ones, and many shrubs which Mr. Kekwick has found. Wind, south-east.
Latitude, 13 degrees 29 minutes 25 seconds.

Tuesday, 15th July, Billiatt Springs. I have named these springs in token
of my approbation of Billiatt's thoughtful, generous, and unselfish
conduct throughout the expedition. I started at 7.40 this morning, course
north-west. Crossed granite and quartz rises, with broad valleys between,
both splendidly grassed. At three miles crossed a small creek with water;
at another mile the same creek again; one also to my line on the
south-west side, and immediately went off to the south-west. At six miles
the river came close to the line, and immediately went off to the west.
Continued on my course through granite and quartz country, splendidly
grassed, and timbered with stringy-bark and gums, pines, palms,
nut-trees, and a wattle bush, which in some places was rather thick, but
not at all difficult to get through. At ten miles again struck the river;
it is now apparently running to the north. Changed to that course, but it
soon left me. At three miles and a half on the north course struck
another creek running from the range north-east; it has an abundance of
water, and is rather boggy. King's horse fell with him in it, but did no
further injury than giving him a wetting. A few of the other horses
stumbled and rolled about in it for a short time, but we got them all
across without accident. Changed to west of north; at half a mile reached
a saddle between two hills, and ascended the one to the west, the river
now running between ranges to the west; they seemed a good deal broken,
with some high points to the north-west. There is a higher one, seemingly
running north and south, with apparently a plain between about four miles
broad, on which are four or five lines of dark trees; this leads me to
suppose that the river is divided. The plain being very thickly timbered,
I could not see distinctly which was the main channel. Descended, and
proceeded on a north-west course. At one mile and a half struck the
river, again running north; changed to that, and at two miles and a half
camped. The country is now all burnt. I am obliged to stop where I can
get feed for the horses. One of the channels comes close to the bank,
east side, about six yards wide and two feet deep; bed sandy. The main
channel must be in the middle of the plain. The hill I ascended to-day
has been under the influence of fire; it is composed of quartz, and a
hard dark-coloured stone; the quartz runs in veins throughout it, in
places crystalline, and formed into spiral and many-sided figures; in
places there is a crust of iron, as if it had been run between the
stones, that is also crystalline. Wind, south-east. Latitude, 13 degrees
17 minutes 22 seconds.

Wednesday, 16th July, The Mary, Adelaide River. Started at 7.40, course
north. The river runs off again to the north-west, and I have passed over
an undulating country, all burnt, but the soil of the richest
description. The rises are comprised of quartz and a hard white stone,
with occasionally a little ironstone. At three miles crossed a creek with
water holes. At five miles crossed another. At seven miles came close to
a high hill--ascended it; at the foot it is composed of a hard slaty
stone covered with a cake of iron; about the middle is quartz, and on the
top conglomerated quartz. The view from south-west to north-west is
extensive, but this not being the highest hill, the rest is hidden. To
the west is a high hill, bluff at both ends, seemingly the last hill of
the range; its course apparently north-west and south-east. At this bluff
hill the range seems to cease, or drops into lower hills. A branch of the
river lies between it and me, but there are still a number of stony hills
before I can reach it. To the north-west and north there are high and
stony hills. The river now seems to run to the west, on a bearing of 30
degrees north of west. From twenty to twenty-five miles distant is
another range, at the foot of which there is a blue stripe, apparently
water, which I suppose to be the main stream of the Adelaide. Descended,
as the country is too rough and stony to continue either to the north or
north-west. I changed to 3 degrees north of west, crossed some stony
hills and broad valleys with splendid alluvial soil, the hills grassed to
the top. On that course struck the branch of the river. Still very thick
with the same kind of timber already mentioned. Most of the bamboos are
dead. I suppose the fire has been the cause of it. I again find it
running to the north; I turn to that course. At three miles struck a
large creek coming from the east with large sheets of water; had to run
it up half a mile before I could get across it. Crossed it all right, and
passed through a beautiful valley of green grass. After that, found that
I was again on the stony rise, where every blade of grass had been burned
off, and not knowing how far this may continue, I have turned off again
for the creek, to give the horses the benefit of the valley. The timber
is the same as yesterday in some places; the stringy-bark is much larger.
The banks of the river, when we first came upon it to-day, were high and
stony. The range to the east seems to cease about here. We are now
crossing low undulations. I have seen a number of kangaroos to-day; they
do not seem to be as large as those in the south. The valleys are
composed of conglomerated ironstone underneath the soil. A large number
of new birds seen to-day, some of them with splendid plumage. Wind,
south-east. Latitude, 13 degrees 7 minutes 21 seconds.

Thursday, 17th July, Tide Creek, Adelaide River. Started at eight
o'clock, course north-west; passed over some stony hills, small creeks,
and valleys well grassed. At three miles again met with the branch of the
river, with bamboos and trees of the same description as before, a
running stream, but not so rapid. At five miles, observing an open plain
among the trees, and the river trending more to the westward, I changed
my course to it, 15 degrees west of north; found it to be open plain, of
rich alluvial soil in places; at times it seemed to be subject to
inundation, I suppose the drainage from the range to the eastward, which
is distant about four miles. I am pleased it has been burnt, but where it
has not the grass is most abundant; where the water seems to remain it is
rather coarse. The plains are studded with lines of green gum-trees, and
the cabbage palms are numerous, which give them a very pretty park-like
appearance. They continued for ten miles, when we made a small stony
hill; we met with a large creek, with large holes of water in it, and
supposing I had got upon the plain that ran to the sea-coast, and seeing
those I had passed over so dry, camped; and having sent Thring to a rise
to see where the river is, he returned, but can see nothing of it, but
reports high hills to the north-west. I am glad of this, for it is not my
intention to follow the river round if I can get water in other places,
for it has already been well described south of this by Lieutenant
Helpman when he came up in a boat, and I wish to see what the country is
away from its banks. Wind south-east, with a few clouds from the north.
For the last week the weather has been excellent, not too hot during the
day, and cool and refreshing at night. The mosquitoes are very annoying,
and the flies during the day are a perfect torment. This creek I have
called Priscilla Creek. Latitude, 12 degrees 56 minutes 54 seconds.

Friday, 18th July, Priscilla Creek. Started at 8.15, course north-west.
Passed over grassy plains and stony rise; when, at three miles, seeing
the termination of a range in a bluff point, changed my course to 310
degrees. Proceeded, still crossing stony hills, consisting of ironstone,
slate, and a hard white rock, which is broken into rectangular fragments;
also over broad valleys, which are covered with grass that when green
must have stood very high, but is now so dry that it breaks off before
the horses. My horse being first, collects so much on his front legs that
I have been obliged to stop, pull him back, and allow it to fall, so that
he may step over it, go on, get another load, and do the same. At six
miles and a half, after crossing a plain, crossed a deep bamboo creek;
this I have named Ellen Creek. Proceeded over two other stony rises and
valleys of the same description, and came upon extensive plains, well
grassed, and of beautiful alluvial soil; crossing them towards the bluff
point at fifteen miles, came upon the Adelaide between me and the bluff,
which is about a mile further on; the river is about eighty yards wide,
and so still that I could not see which way the current was. I suppose
its being high tide was the cause of this. The banks are thickly lined
with bamboo, very tall and stout, very steep, and twelve feet down to the
water's edge; the water appeared to be of great depth, and entirely free
from snags or fallen timber. The range on the opposite side of the river,
for which I was directing my course, being the highest I have seen in
this new country, I have named it after His Excellency the
Governor-in-Chief of South Australia, Daly Range, and its highest peak to
the north Mount Daly. Before reaching the river, at thirteen miles, we
passed a high conspicuous tent hill, at right angle, north-east to our
line; this I have named Mount Goyder, after the Surveyor-General of South
Australia. Followed the river on a north course for about a mile, when I
was stopped by a deep side creek of thick bamboo, with water; turned to
the east, rounded the bamboo, but found myself in a boggy marsh, which I
could not cross. This marsh is covered with fine grass, in black alluvial
soil, in which is growing a new kind of lily, with a large broad
heart-shaped leaf a foot or more across; the blossoms are six inches
high, resemble a tulip in shape, and are of a deep brilliant rose colour;
the seeds are contained in a vessel resembling the rose of a
watering-pot, with the end of each egg-shaped seed showing from the
holes, and the colour of this is a bright yellow. The marsh is studded
with a great number of melaleuca-trees, tall and straight. As I could not
cross, I had to round it, which took me a little more than an hour; when
I got upon some low undulating rises, not far from Mount Goyder, composed
of conglomerate ironstone and ironstone gravel, which seem to produce the
springs which supply the marsh. Camped on the side of the marsh, to give
the horses the benefit of the green grass, for some of them are still
troubled with worms, and are very poor and miserable, and I have no
medicine to give them, and there is not a blade of grass on the banks of
the river--all has been burnt within the last four days. Native smoke in
every direction. Wind south-east, with a few clouds. Latitude, 12 degrees
49 minutes 30 seconds.

Saturday, 19th July, Lily Marsh, Adelaide River. Started at 9.10, course
20 degrees east of north. At three miles crossed some stony rises and
broad alluvial grassy valleys; at four miles met the river, had to go
half a mile to the south-east to round it. Again changed to my first
course; at seven miles and a half crossed a creek with water. The country
to this is good, with occasionally a little ironstone and gravel, timber
of stringy-bark, and a little low gum scrub. Having crossed this creek,
we ascended a sandy table land with an open forest of stringy bark (good
timber), palms, gums, other trees and bushes; it has been lately burnt,
but the roots of the grass abound. This continued for about three miles.
There is a small stony range of hills to the west, which at the end of
the three miles dropped into a grassy plain of a beautiful black alluvial
soil, covered with lines and groves of the cabbage palm trees, which give
it a very picturesque appearance; its dip is towards the river; in two
miles crossed it, and again ascended low table land of the very same
description as the other. At fourteen miles struck another creek with
water, and camped. The country gone over to-day, though not all of the
very best description, has plains in it of the very finest kind--even the
sandy table-land bears an abundant crop of grass. The trees are so thick
that I can get no view of the surrounding country; the tall beautiful
palm grows in this creek. Native smoke about, but we have not seen any
natives. There are large masses of volcanic rock on the sides of this
creek. At about a mile to the eastward is a large body of springs that
supply water to this creek, which I have named Anna Creek. Camped at ten
minutes to three o'clock. Wind variable. Latitude, 12 degrees 39 minutes
7 seconds.

Sunday, 20th July, Anna Creek. The mosquitoes at this camp have been most
annoying; scarcely one of us has been able to close his eyes in sleep
during the whole night: I never found them so bad anywhere--night and day
they are at us. The grass in, and on the banks of, this creek is six feet
high; to the westward there are long reaches of water, and the creek very
thickly timbered with melaleuca, gum, stringy-bark, and palms. Wind,
south-east.

Monday, 21st July, Anna Creek and Springs. Again passed a miserable night
with the mosquitoes. Started at eight o'clock; course, north-north-west.
At three miles came upon another extensive fresh-water marsh, too boggy
to cross. There is rising ground to the north-west and north; the river
seems to run between. I can see clumps of bamboos and trees, by which I
suppose it runs at about a mile to the north-north-west. The ground for
the last three miles is of a sandy nature, and light-brown colour, with
ironstone gravel on the surface, volcanic rock occasionally cropping out.
The borders of the marsh are of the richest description of black alluvial
soil, and when the grass has sprung after it has been burnt, it has the
appearance of a rich and very thick crop of green wheat. I am now
compelled to alter my course to 30 degrees south of east, to get across a
water creek coming into the marsh, running deep, broad and boggy, and so
thick with trees, bushes, and strong vines interwoven throughout it, that
it would take a day to cut a passage through. At three miles we crossed
the stream, and proceeded again on the north-north-west course, but at a
mile and a half were stopped by another creek of the same description.
Changed to east, and at half a mile was able to cross it also, and again
went on my original bearing. Continued on it for three miles, when we
were again stopped by another running stream, but this one I was able to
cross without going far out of my course. Proceeded on the
north-north-west course, passing over elevated ground of the same
description as the first three miles. At seventeen miles came upon a
thick clump of trees, with beautiful palms growing amongst them; examined
it and found it to have been a spring, but now dry. Proceeded on another
mile, and was again stopped by what seemed to be a continuation of the
large marsh; we now appeared to have got right into the middle of it. It
was to be seen to the south-west, north-east, and south-east of us.
Camped on a point of rising ground running into it. The timber on the
rises between the creeks is stringy-bark, small gums, and in places a
nasty scrub, very sharp, which tore a number of our saddle-bags: it is a
very good thing the patches of it are not broad. The grass, where it has
not been burned, is very thick and high--up to my shoulder when on
horseback. About a mile from here, to the west, I can see what appears to
be the water of the river, running through clumps of trees and bamboos,
beyond which, in the distance, are courses of low rising ground, in
places broken also with clumps of trees; the course of the river seems to
be north-north-west. On the east side of the marsh is also rising ground;
the marsh in that direction seems to run five or six miles before it
meets the rising ground, and appears after that to come round to the
north. Nights cool. Latitude, 12 degrees 28 minutes 19 seconds. Wind,
south-east.

Tuesday, 22nd July, Fresh-water Marsh. As the marsh seems to run so much
to the east, and not knowing how much further I shall have to go to get
across the numerous creeks that appear to come into it, I shall remain
here to-day and endeavour to find a road through it to the river, and
follow up the banks if I can. I have a deal of work to do to the plan,
and our bags require mending. After collecting the horses Thring tried to
cross the marsh to the river, and succeeded in reaching its banks,
finding firm ground all the way; the breadth of the river here being
about a hundred yards, very deep, and running with some velocity, the
water quite fresh. He having returned with this information, I sent him,
King, and Frew, mounted on the strongest horses, to follow the banks of
the river till noon, to see if there is any obstruction to prevent my
travelling by its banks. In two hours they returned with the sad tidings
that the banks were broken down by watercourses, deep, broad, and boggy;
this is a great disappointment, for it will take me a day or two longer
than I expected in reaching the sea-coast, in consequence of having to go
a long way round to clear the marsh and creeks. The edge of the marsh was
still of the same rich character, and covered with luxuriant grass. The
rise we are camped on is also the same, with ironstone gravel on the
surface; this seems to have been a favourite camping-place for a large
number of natives. There is a great quantity of fish bones, mussel, and
turtle shells, at a little distance from the camp, close to where there
was some water. There are three poles fixed in the ground, forming an
equilateral triangle, on the top of which was a framework of the same
figure, over which were placed bars of wood: its height from the ground
eight feet. This has apparently been used by them for smoke-drying a dead
blackfellow. We have seen no natives since leaving the Roper, although
their smoke is still round about us. On and about the marsh are large
flocks of geese, ibis, and numerous other aquatic birds; they are so wild
that they will not allow us to come within shot of them. Mr. Kekwick has
been successful in shooting a goose; it has a peculiar-shaped head,
having a large horny lump on the top resembling a topknot, and only a
very small web at the root of his toes. The river opposite this, about a
yard from the bank, is nine feet deep. Wind variable. Night cool.

Wednesday, 23rd July, Fresh-water Marsh. Started at 7.40, course 22
degrees east of south, one mile, to round the marsh; thence one mile
south-east; thence east for six miles, when we struck a large creek, deep
and long reaches; thence three quarters of a mile south before we could
cross it. This I have named Thring Creek, in token of my approbation of
his conduct throughout the journey; thence east, one mile and a half;
thence north for nine miles, when I again struck the large marsh. Thring
Creek has been running nearly parallel with the north course until it
empties itself into the marsh. The country gone over to-day, after
leaving the side of the marsh, as well as the banks of the creek, and
also some small plains, is of the same rich description of soil covered
with grass; the other parts are slightly elevated, the soil light with a
little sand on the surface of a brown colour; timber, mixture of
stringy-bark and gums, with many others; also, a low thick scrub, which
has lately been burnt in many places, the few patches that have escaped
abounding in grass. I have come twelve miles to the eastward to try to
round the marsh, but have not been able to do so; the plains that were
seen from the river by those who came up it in boats is the marsh; it is
covered with luxuriant grass, which gives it the appearance of extensive
grassy plains. I have camped at where the Thring spreads itself over a
portion of the marsh. There is rising ground to the north-west, on the
opposite side, which I suppose to be a continuation of the elevated
ground I passed before crossing the creek, and the same that I saw
bearing north from the last camp. I suppose it runs in towards the river.
Wind, south. Latitude, 13 degrees 22 minutes 30 seconds.

Thursday, 24th July, Thring Creek, Entering the Marsh. Started at 7.40,
course north. I have taken this course in order to make the sea-coast,
which I suppose to be distant about eight miles and a half, as soon as
possible; by this I hope to avoid the marsh. I shall travel along the
beach to the north of the Adelaide. I did not inform any of the party,
except Thring and Auld, that I was so near to the sea, as I wished to
give them a surprise on reaching it. Proceeded through a light soil,
slightly elevated, with a little ironstone on the surface--the volcanic
rock cropping out occasionally; also some flats of black alluvial soil.
The timber much smaller and more like scrub, showing that we are nearing
the sea. At eight miles and a half came upon a broad valley of black
alluvial soil, covered with long grass; from this I can hear the wash of
the sea. On the other side of the valley, which is rather more than a
quarter of a mile wide, is growing a line of thick heavy bushes, very
dense, showing that to be the boundary of the beach. Crossed the valley,
and entered the scrub, which was a complete network of vines. Stopped the
horses to clear a way, whilst I advanced a few yards on to the beach, and
was gratified and delighted to behold the water of the Indian Ocean in
Van Diemen Gulf, before the party with the horses knew anything of its
proximity. Thring, who rode in advance of me, called out "The Sea!" which
so took them all by surprise, and they were so astonished, that he had to
repeat the call before they fully understood what was meant. Then they
immediately gave three long and hearty cheers. The beach is covered with
a soft blue mud. It being ebb tide, I could see some distance; found it
would be impossible for me to take the horses along it; I therefore kept
them where I had halted them, and allowed half the party to come on to
the beach and gratify themselves by a sight of the sea, while the other
half remained to watch the horses until their return. I dipped my feet,
and washed my face and hands in the sea, as I promised the late Governor
Sir Richard McDonnell I would do if I reached it. The mud has nearly
covered all the shells; we got a few, however. I could see no sea-weed.
There is a point of land some distance off, bearing 70 degrees. After all
the party had had some time on the beach, at which they were much pleased
and gratified, they collected a few shells; I returned to the valley,
where I had my initials (J.M.D.S.) cut on a large tree, as I did not
intend to put up my flag until I arrived at the mouth of the Adelaide.
Proceeded, on a course of 302 degrees, along the valley; at one mile and
a half, coming upon a small creek, with running water, and the valley
being covered with beautiful green grass, I have camped to give the
horses the benefit of it. Thus have I, through the instrumentality of
Divine Providence, been led to accomplish the great object of the
expedition, and take the whole party safely as witnesses to the fact, and
through one of the finest countries man could wish to behold--good to the
coast, and with a stream of running water within half a mile of the sea.
From Newcastle Water to the sea-beach, the main body of the horses have
been only one night without water, and then got it within the next day.
If this country is settled, it will be one of the finest Colonies under
the Crown, suitable for the growth of any and everything--what a splendid
country for producing cotton! Judging from the number of the pathways
from the water to the beach, across the valley, the natives must be very
numerous; we have not seen any, although we have passed many of their
recent tracks and encampments. The cabbage and fan palm-trees have been
very plentiful during to-day's journey down to this valley. This creek I
named Charles Creek, after the eldest son of John Chambers, Esquire: it
is one by which some large bodies of springs discharge their surplus
water into Van Diemen Gulf; its banks are of soft mud, and boggy. Wind,
south. Latitude, 12 degrees 13 minutes 30 seconds.

Friday, 25th July, Charles Creek, Van Diemen Gulf. I have sent Thring to
the south-west to see if he can get round the marsh. If it is firm ground
I shall endeavour to make the mouth of the river by that way. After a
long search he has returned and informs me that it is impracticable,
being too boggy for the horses. As the great object of the expedition is
now attained, and the mouth of the river already well known, I do not
think it advisable to waste the strength of my horses in forcing them
through, neither do I see what object I should gain by doing so; they
have still a very long and fatiguing journey in recrossing the continent
to Adelaide, and my health is so bad that I am unable to bear a long
day's ride. I shall, therefore, cross this creek and see if I can get
along by the sea-beach or close to it. Started and had great difficulty
in getting the horses over, although we cut a large quantity of grass,
putting it on the banks and on logs of wood which were put into it. We
had a number bogged, and I was nearly losing one of my best horses, and
was obliged to have him pulled out with ropes; after the loss of some
time we succeeded in getting them all over safely. Proceeded on a
west-north-west course over a firm ground of black alluvial soil. At two
miles came upon an open part of the beach, went on to it, and again found
the mud quite impassable for horses; in the last mile we have had some
rather soft ground. Stopped the party, as this travelling is too much for
the horses, and, taking Thring with me, rode two miles to see if the
ground was any firmer in places; found it very soft where the salt water
had covered it, in others not so bad. Judging from the number of shells
banked up in different places, the sea must occasionally come over this.
I saw at once that this would not do for the weak state in which my
horses were, and I therefore returned to where I had left the party,
resolving to recross the continent to the City of Adelaide. I now had an
open place cleared, and selecting one of the tallest trees, stripped it
of its lower branches, and on its highest branch fixed my flag, the Union
Jack, with my name sewn in the centre of it. When this was completed, the
party gave three cheers, and Mr. Kekwick then addressed me,
congratulating me on having completed this great and important
undertaking, to which I replied. Mr. Waterhouse also spoke a few words on
the same subject, and concluded with three cheers for the Queen and three
for the Prince of Wales. At one foot south from the foot of the tree is
buried, about eight inches below the ground, an air-tight tin case, in
which is a paper with the following notice:

"South Australian Great Northern Exploring Expedition.

"The exploring party, under the command of John McDouall Stuart, arrived
at this spot on the 25th day of July, 1862, having crossed the entire
Continent of Australia from the Southern to the Indian Ocean, passing
through the centre. They left the City of Adelaide on the 26th day of
October, 1861, and the most northern station of the Colony on 21st day of
January, 1862. To commemorate this happy event, they have raised this
flag bearing his name. All well. God save the Queen!"

[Here follow the signatures of myself and party.]

As this bay has not been named, I have taken this opportunity of naming
it Chambers Bay, in honour of Miss Chambers, who kindly presented me with
the flag which I have planted this day, and I hope this may be the first
sign of the dawn of approaching civilization. Exactly this day nine
months the party left North Adelaide. Before leaving, between the hours
of eleven and twelve o'clock, they had lunch at Mr. Chambers' house; John
Bentham Neals, Esquire, being present, proposed success to me, and wished
I might plant the flag on the north-west coast. At the same hour of the
day, nine months after, the flag was raised on the shores of Chambers
Bay, Van Diemen Gulf. On the bark of the tree on which the flag is placed
is cut--DIG ONE FOOT--S. We then bade farewell to the Indian Ocean, and
returned to Charles Creek, where we had again great difficulty in getting
the horses across, but it was at last accomplished without accident. We
have passed numerous and recent tracks of natives to-day; they are still
burning the country at some distance from the coast. Wind, south-east.
Latitude, 12 degrees 14 minutes 50 seconds.

...

RETURN.

Saturday, 26th July, Charles Creek, Chambers Bay, Van Diemen Gulf. This
day I commence my return, and feel perfectly satisfied in my own mind
that I have done everything in my power to obtain as extensive a
knowledge of the country as the strength of my party will allow me. I
could have made the mouth of the river, but perhaps at the expense of
losing many of the horses, thus increasing the difficulties of the return
journey. Many of them are so poor and weak, from the effects of the
worms, that they have not been able for some time to carry anything like
a load, and I have been compelled to make the (symbol crescent over C)
horses stand the brunt of the work of the expedition. As yet not one of
them has failed; they have all done their work in excellent style. The
sea has been reached, which was the great object of the expedition, and a
practicable route found through a splendid country from Newcastle Water
to it, abounding, for a great part of the way, in running streams well
stocked with fish--and this has been accomplished at a season of the year
during which we have not had one drop of rain. Started, following my
tracks back. Passed my former camp on the Thring; went on and crossed it.
Proceeded on my east course to the west, about one mile and a half, to
some small green marshy plains of black alluvial soil, with a spring in
the centre, covered with fine green grass. Camped. Wind, south. Latitude,
12 degrees 30 minutes 21 seconds.

Sunday, 27th July, Small Grassy Plains. Day rather warm; mosquitoes
terrible; no sleep last night; never found them so bad before; not a
breath of wind to drive them away.

Monday, 28th July, Small Grassy Plains. Started at 7.40, course 25
degrees west of south, for my camp of the eighteenth instant. At ten
miles struck my tracks, thus avoiding the boggy creeks that flow into the
large marsh. On this course passed five small black alluvial plains,
covered with grass, three of them having springs with water on the
surface. They lie between slightly elevated country of light-brown soil,
having stringy-bark and gums, with occasionally a thin scrub abounding in
grass. On the plains there is occasionally a little of the volcanic rock
cropping out. Followed my former tracks to the camp on the Lily Marsh,
and remained for the night. We all passed a miserable night with the
mosquitoes. My hands, wrists, and neck, were all blistered over with
their bites, and were most painful.

Tuesday, 29th July, Lily Marsh. At half-past seven o'clock proceeded on
the track. Passed my camp of 17th instant, and arriving at the one of the
16th at four o'clock p.m., camped. Wind, south.

Wednesday, 30th July, Side Creek, Adelaide River. All were delighted with
a comfortable night's rest--no mosquitoes. Proceeded to Billiatt Springs
and camped. One of the horses, Jerry, has been ill for the last three
weeks, and although he has not had anything to carry, it has been as much
as we could do to get him into the camp. This afternoon he gave in
altogether, and Mr. Kekwick was quite unable to get him a step further,
and was compelled to leave him about three miles back, where there is
some water and plenty of feed. Wind, south-east.

Thursday, 31st July, Billiatt Springs. Proceeded and passed our camps of
13th and 12th instant. Crossed the Mary branch of the Adelaide: went
along the south side, expecting to avoid the boggy creek crossed on the
12th instant. When nearly opposite to it, camped. Found this part of the
branch deep, broad, and boggy; but I think we will be able to cross in
the morning by cutting down a number of cabbage palms, which are growing
very thick here. Light winds from south-east.

Friday, 1st August, South Side of the Mary. Recrossed the Mary, which is
very boggy on the banks. We were enabled to cross it safely by cutting a
large quantity of long grass, laying it on the sides of the banks, with a
few logs and pickets driven into the bed to prevent the current from
carrying away the grass. In this we succeeded very well. After crossing I
found we had still to encounter the other running and boggy creek of the
12th ultimo; but, by repeating the same operation, we were successful.
Passed our camp of the 11th ultimo, and proceeded on towards the table
land. On approaching it, where the springs come from underneath, found it
very boggy; had some difficulty in getting the horses through it. Got
them all through with the exception of Frew's horse, which stuck hard and
fast in it, and we were obliged to pull him out, which was soon
accomplished, and we got him safe on terra firma. Continued along the
foot of the table land, and halted at our camp of the 10th ultimo. At
about seven p.m. last night I heard something plunging in the river; sent
down to see what it was; found two of the horses bogged, and unable to
extricate themselves. Got ropes, and all the party to pull them out.
After an hour's hard work succeeded. On coming near the table land the
country is all on fire, causing a dense black smoke and heated
atmosphere. Wind, south-east.

Saturday, 2nd August, North-west Side of Table Land. Proceeded up the
creek to the gorge--where we came down from the top of the table land;
ascended it, which they all did well except one horse, which refused to
go up, and caused me to lose more than an hour with him; we had to take
all the things off him and carry them to the top on our backs. We had to
zigzag him backwards and forwards, and got him to the top after a deal of
trouble. Crossing on the top we met with a large fire about two miles
broad. The wind not being strong, nor the grass very long, we got through
it well, but my weak eyes suffered much from the smoke coming from the
burning logs, trees, and grass. The atmosphere very hot and almost
overpowering before we got through it. One of the horses knocked up, but
we were able to get him on to the running creek connected with Kekwick's
large group of springs, where I am obliged to camp and try to recover
him. This is the first one of the (symbol crescent over C) horses that
has failed; but he has not had fair play, through the negligence of the
man who had him. He has for some time been carrying a load of one hundred
and forty pounds without my knowledge, far more than he was able to
carry. He has been a good horse, and has done a deal of work. There are a
number of native tracks both up and down our tracks. One of the natives
seems to have a very large foot. Wind, south.

Sunday, 3rd August, Kekwick's Large Springs. Last evening, just as the
sun was dipping, five natives made their appearance, armed with spears,
and came marching boldly up to within eighty yards of the camp, where
they were met by Mr. Kekwick and others of the party who had advanced to
meet them. They were all young men, small, and very thin. Seeing so many
approaching them they soon went off. They were all smeared over with
burnt grass, charcoal, or some other substance of that description. This
morning, shortly after sunrise, the same five again made their
appearance. I went up to them to see what they wanted. Saw that they had
painted their bodies with white stripes ready for war. As it is my
intention to pass peaceably through the different tribes, I endeavoured
to make friends with them by showing them we intended them no harm if
they will leave us alone. One of them had a curious fish spear, which he
seemed inclined to part with, and I sent Mr. Kekwick to get some
fish-hooks to exchange with him, which he readily did; we then left them.
They continuing a longer time than I wished, and gradually approaching
nearer to our camp, thinking perhaps they really did not wish to part
with the spear, I sent Mr. Kekwick back with it to them to see if that
was what they wanted, and to take the fish-hooks from them. But when they
saw what was intended, they gave back the spear and retained the hooks.
They offered another with a stone head upon the same terms, which was
accepted. Mr. Kekwick had a deal of trouble before he could get them to
move off, when they were joined by another, and then went off by twos. In
a short time they set fire to the grass all round us to try to burn us
out. Two of them came again close to the camp under pretence of looking
for game before the fire, at the same time setting fire to the grass
closer to us. But Mr. Kekwick and one of the others, seeing their
intention, ran up to them, who, on their approach, ran off, setting fire
to the grass as they went along, which gave us a deal of trouble in
putting out, as we wished to save as much feed for the horses as will do
for them till to-morrow morning; we have managed that, if they do not
come and set fire to it again. If they do I shall be compelled to use
preventive means with them, for I can stand it no longer; they must be
taught a lesson that we possess a little more power than they anticipate.
I would have moved on, but some of my horses are so ill that they are
unable to travel. If the natives we have seen to-day are a sample of
those that inhabit this country, they are certainly the smallest and most
miserable race of men that I have ever seen. In height about five feet,
their arms and legs remarkably thin, they do not seem to want the
inclination of doing mischief if they could get an opportunity, but they
find we are rather too watchful to give them a chance. From their manner
I have no doubt there were many more concealed, who intended attacking us
under cover of the smoke--indeed if they see us unprepared they may yet
do it before evening. At sundown they have not again made their
appearance. Wind, south.

Monday, 4th August, Kekwick's Large springs. Proceeded to the Katherine
and camped. The horse that knocked up on Saturday gave in again two miles
before we arrived here, although the distance is only thirteen miles, and
he had a rest all Sunday. I shall be compelled to leave him here; he only
destroys other horses dragging him along, and as the season is so far
advanced, I am doubtful of the water in some of the ponds, and therefore
cannot stop with him. I have been so very unwell to-day, with symptoms of
fever, that I could scarcely reach this place; but I hope I shall be
better by to-morrow. Nights and mornings are now very cold, but the sun
is very hot during the middle and afterpart of the day. Wind, south-east.

Tuesday, 5th August, The Katherine. Leaving the knocked-up horse behind,
proceeded to the Fanny, and camped. It was as much as I could do to sit
in the saddle this distance. Wind, south.

Wednesday, 6th August, The Fanny. Proceeded to the Waterhouse and camped.
The natives have been along our track, and burned the grass to within
three miles of our camp. On arriving here I was much disappointed on
finding all the water gone, but, following back the north-west branch, I
found enough for our use to-night and to-morrow morning. The country is
all on fire to the south-east. Wind, variable. The journey has been
rather rough and stony, and my weak horses feel it very much. I am afraid
I shall be compelled to leave some more of them behind. I cannot now stay
for them to recover, after seeing the rapidity with which this water has
dried up. A long delay will cause my retreat to be cut off in the pond
country. Wind, south-east. There is still permanent water up the
north-west branch of this creek.

Thursday, 7th August, The Waterhouse. Started at half-past seven, and at
two minutes past ten o'clock I arrived at the running stream (the
Chambers) of the 4th ultimo and camped. Weak horses looking very bad.
Country on fire round about us. A number of natives have been following
on our former tracks. Wind, south.

Friday, 8th August, Running Stream, The Chambers. Crossed the hard
sandstone range, and got upon the branch of the Chambers that I followed
up, passing our camp of 3rd ultimo, with plenty of permanent water.
Followed it down to our camp of the 2nd ultimo and remained there. Had to
leave one of the done-up horses about two miles behind. Another horse
gave in, and it was as much as Mr. Kekwick could do to get him thus far.
The natives have burned all the grass throughout this day's journey. A
little has escaped at this camp, and I am now compelled to give my horses
a rest until Monday morning. I thought they would have been able to carry
me across the Chambers before I gave them a rest, but, if I proceed
further, I shall lose more of them. The weather is beginning to be again
very hot in the middle of the day. Wind, south-east.

Saturday, 9th August, River Chambers. Resting horses. Day hot. Wind
variable.

Sunday, 10th August, River Chambers. Resting horses. I have sent Thring
to bring up the one that was left behind on Friday; in a short time he
brought him up, looking a most deplorable picture; the other one that
gave in the same day is quite as bad. I shall have to leave them behind;
it is only destroying other horses to force them along. I must also
reduce the weight the others are carrying, to enable them to get along. I
have had all the saddle-bags overhauled, and shall leave everything we
can possibly do without--even boots and clothes belonging to the party
have not been spared; all were quite willing to sacrifice anything they
had, with the exception of one who had a pair of new boots he had never
put on. I told him to put them on, and leave the old ones, but he
immediately told me that he had got a bad foot; I very soon cured him of
that by telling him if that was the case he might leave the new ones. I
have managed to leave about three hundredweight; many of the things I can
ill spare, but I hope by doing this to be able in a short time to push on
a little quicker. Light winds, variable.

Monday, 11th August, River Chambers. Two of the horses having strayed
this morning, it was a quarter past nine before I could get a start. I
had to proceed very slowly, in consequence of five of the horses being so
ill that they were unable to walk quickly. Proceeded on my former tracks,
cutting off the bends of the river. In some places it is very stony. Late
in the afternoon managed to get all the horses to the first camp on this
river. Light winds, south-east.

Tuesday, 12th August, River Chambers. Horses missing again this morning.
Started at half-past eight. Proceeded to the south-east end of the reedy
swamp, and at half-past three o'clock camped. An hour before halting, we
surprised a number of native women and children who were preparing roots
and other things for their repast. The moment they saw us they seized on
their children, placed them on their shoulders, and ran off screaming at
a great rate, leaving all their things behind them, amongst which we saw
a piece of iron used as a tomahawk; it had a large round eye into which
they had fixed a handle; the edge was about the usual tomahawk breadth;
when hot it had been hammered together. It had apparently been a hinge of
some large door or other large article; the natives had ground it down,
and seemed to know the use of it. Left their articles undisturbed, and
proceeded to the river Roper. My horses are still looking very bad. The
cause must be the dry state of the grass; it is so parched up that when
rubbed between the hands it becomes a fine powder, and they must derive
very little nourishment from it. I can hear natives talking and screaming
on the other side of the river, which at this place is a strong running
stream about thirty yards wide and apparently deep. Wind, south-east,
blowing strong.

Wednesday, 13th August, Roper River, Reedy Swamp. One of the horses
missing again this morning; he is one that generally goes off and hides
himself if he can find a place to do so. Searched all round, but could
find nothing of him or his tracks. Thinking that he might be hidden
amongst the thick bushes over the river, sent Frew to look through them
on foot, and Mr. Kekwick to an open place up the river to see if he had
got into it. Mr. Kekwick returned in a short time and reported that he
saw him lying drowned in the middle of it. I am sorry for this: he was a
good horse, in fair condition, was with me last year, and has always done
his work well, although he has caused a deal of trouble and loss of time
by so frequently concealing himself. I shall feel his loss very much, as
so many of the other horses are so poor that they are able to carry but
little of a load, and I am obliged to let four go without carrying
anything; indeed it is as much as they can do to walk the day's journey,
although the journeys are short. I shall be compelled to make them still
shorter to try and get them round again. As we were saddling, one native
man and two women made their appearance and came close to the camp. Mr.
Kekwick and I went up to them; the man was middle-aged, stout and tall,
the women were also tall, one especially. Their features were not so
coarse as those we had seen before--a very great difference between this
fellow and those I saw on the source of the Adelaide River. The man made
signs that he would like to get a fishhook by bending his forefinger and
placing it in his mouth, imitating the method of catching fish. I gave
him one with which he was much pleased: I also gave a cotton handkerchief
to each of the women; one of them no sooner got it than she held out the
other hand and called out "more, more, more;" with that request I did not
feel inclined to comply. They remained until we started. Proceeding about
three quarters of a mile down the river to where I had crossed it before,
I got all the horses over without difficulty. There is now no difference
in the strength, depth, nor velocity of the stream since we were here; it
is exactly in the same state as when we previously crossed it. After
crossing it to the other side, I had to cross another deep although dry
creek coming from the east; proceeded on a south-east course to avoid the
deep boggy creek that comes into the river, but at two miles I was
stopped by an immense number of springs, very boggy, and emitting a large
quantity of water; they seem to come from the east, as far as I could
see, in a wooded valley between two hills. I had to round them until I
got upon the south-east course again. At seven miles came upon a large
creek or chain of ponds, having long broad deep reaches of water;
followed this, running nearly my course for seven miles in a straight
line. Camped. My horses cannot do more. The country that I have travelled
over to-day is of the very finest description, rich black alluvial soil,
completely matted with grass, the water most excellent and abundant. The
timber, gum and melaleuca, a few of the trees resembling the shea-oak
also; a few of the fan palms growing among the springs, very tall,
upwards of forty feet; the cabbage palm, and a number of other bushes.
The general course to-day has been about east-south-east. Wind variable.

Thursday, 14th August, Springs and Chains of Ponds South of the Roper.
Started at half-past seven, intending to follow a south-east course to
make the Mussel Camp on the 23rd of June; but, meeting with another large
creek with continuous water, deep, broad, and boggy, also a number of
springs and water creeks, so boggy that I could not cross them, had to
twist and turn about very frequently, and sometimes to go quite back
again, before I could clear them--which brought me often close to the
river again. About eleven o'clock, as I was approaching the east end of a
low rocky range of hills, where I expected to get rid of all the boggy
ground, I was again stopped by a broad, deep, and boggy sheet of water. A
few minutes before coming to it, I was seized with a violent pain under
the right shoulder-blade, which deprived me of breath and power of
utterance: it darted through my body like lightning, causing the most
excruciating pain that I have ever felt during my life. I had to halt the
party, and was lifted from the saddle completely powerless. After
dismounting, the pain became so violent, and the torture so excessive,
that I thought my career in the world was coming quickly to a close. I
was completely paralysed, and a cold perspiration was pouring in streams
over my face and body. Recollecting I had got a mixture of laudanum and
other strong aromatic tinctures, had it sought for and took a strong
dose. After suffering an hour the extremes of torture, I began to feel
the good effects of the medicine, and obtained a little relief from the
pain ceasing for a few seconds; but still very bad. In a short time
afterwards I was able to bear being lifted into the saddle; again my
sufferings commenced, for every false step the horse made sent the pain
through my body like a knife, and almost brought me to the ground. Being
determined to reach the Mussel Camp to-night, and get quit of the Roper
River, which has been so unfortunate to me in drowning two of my best
horses, I kept my saddle until I reached it--which was not till near five
o'clock. Such a day of torture I never experienced before. On reaching
our tracks, about four miles from the Mussel Camp, another of the horses
knocked up, and we could not get him a step further. I expected to have
lost him long before this; he is one of those that failed on my last
journey, and was sent back from Mount Margaret. Light winds from east.

Friday, 15th August, Mussel Camp. I have passed a miserable night, and
feel but little better this morning, and as the horses require rest, I
shall remain here to-day. Shortly after sunrise, three natives came close
to the camp; Mr. Kekwick went up to them. Two were of the number of those
who visited us the first time at the large reedy swamp. They were very
quiet, and seemed very friendly; they had come to have a look at us, and
satisfy their curiosity. I feel a little easier to-night. Light wind,
variable.

Saturday, 16th August, Mussel Camp. Started at nine o'clock. Another of
my horses very ill; I think that many of them must have eaten some
poisonous plant on the Roper and its tributaries; I never saw horses fall
away so rapidly before. The worst are those that have been in good
condition throughout the journey, and the work they have been doing since
I commenced my return journey any horses ought to have done with ease. I
have never travelled more than eight hours a day, and frequently not more
than six hours. In a day or two they fall away to perfect skeletons, are
quite stupid, and hardly able to walk. I am glad that I am now quit of
the Roper, and hope that I shall have no more of them taken ill. If I can
only get the weak ones beyond Newcastle Water, where I expect to get some
new grass for them (from the June and July rains), they would soon
recover. My old horses are all looking well, although they have had to
carry the heaviest loads throughout the journey. I should have been in a
sad way without them--they are my mainstay. Arrived at the Rock Camp,
River Strangways, at two o'clock without having to leave any more. I feel
a little better to-day, but the motion of the horse has been very severe
throughout the journey. The water at this camp is drying up very rapidly:
it is reduced three feet in depth since we left, and I am very much
afraid it will be all gone in Purdie Ponds--if such is the case, I shall
lose all the weak horses. Wind in strong puffs, variable.

Sunday, 17th August, Rock Camp. Resting horses. Winds light and variable.

Monday, 18th August, Rock Camp. Three of the best horses are missing this
morning--they are the three leading horses--while feeding; and I have
never known them to be away from the others before. The three
horse-keepers have returned at half-past ten, and can see nothing of
them; the ground is so hard that their tracks leave but little
impression, so that they might have passed them unseen. Mounted Thring
and King on fresh horses to round the feeding-tracks again, and at
half-past twelve they returned with them. They happened to come upon
their tracks on a small piece of sandy ground on the opposite side of the
creek; they traced them to a large permanent water lagoon, deep and
broad, with water-lilies growing round it, and a number of ducks upon it;
it is about three quarters of a mile west-south-west from this camp. Not
seeing them there they followed their tracks for another mile, and there
found them, at which I was very glad, for they are three of my very best
horses, on which I am placing my dependence for carrying me back. I felt
very uneasy at their being away, thinking that the natives might have cut
them off during night. Saddled and proceeded to my first camp, north of
the Rocky Gorge, but was disappointed to find all the water gone, which I
did not expect. Proceeded a mile further, and found as much as will do
for a drink for the horses to-night and to-morrow morning. Camped. Light
winds, variable.

Tuesday, 19th August, First Camp North of Rocky Gorge. Started at eight
o'clock, proceeding to the Rocky Gorge, and camped. This water has shrunk
considerably since we left it, and I have now little hopes of there being
any water in Purdie Ponds. If there is not I shall require to push
through to Daly Waters. Light winds, south-east.

Wednesday, 20th August, Rocky Gorge, River Strangways. If there is no
water in Purdie Ponds, I have six horses that will not be able to go
through to Daly Waters; they must be two nights without it, and that they
will not be able to stand. I have therefore determined to send Thring and
King to Purdie Ponds to-morrow, to see if there is any water, and also to
examine another place that I observed in coming through, where I think
there may be water. If they find none at either of these places, I shall
be compelled to leave the six weak horses at the camp, where there is and
will be plenty of food and water for them. To attempt taking them
through, and be compelled to leave them behind where there will be no
chance of their getting a drop of water, would, I consider, be a great
cruelty; here they are safe, and there is a chance of their being picked
up by the next party. If Thring succeeds in getting water, I shall still
endeavour to take them on. I am yet suffering very much from scurvy; my
teeth and gums are so bad that it causes me excessive pain to eat
anything, and what I do eat I am unable to masticate properly, which
causes me to feel very ill indeed. Light winds, south-east.

Thursday, 21st August, Rocky Gorge, River Strangways. At 7.30 despatched
Thring and King to see if there is any water in the Ponds. Resting
horses, repairing saddle-bags, etc. Day hot, night and morning cool;
wind, south-east. My sight has been very much impaired during the last
month; after sundown, I am in total darkness. Even though the moon is
full, and shining bright and clear to the others, to me it is darkness; I
can see her dimly, but she gives me no more light than if she had been
painted on a piece of canvas. I am now quite incapable of taking
observations at night, and I am most thankful this did not happen before
I was enabled to reach the ocean, as the most of my observations are
taken at night. After the equinox the sun is too high to be measured by
the sextant in the artificial horizon.

Friday, 22nd August, Rocky Gorge, River Strangways. Day exceedingly hot.
Wind still from south-east, sometimes blowing in strong puffs. A little
after two o'clock Thring and King returned with the good news that there
is still water in Purdie Ponds; there is as much as will do for us until
Monday morning. I am very glad of it, for it will enable me to get the
weak horses through to Newcastle Water. After that I hope they will soon
recover, for I expect that rain has fallen to the southward of that, and
trust I shall get some fresh feed for them, which they require very much.
I still feel very unwell to-day.

Saturday, 23rd August, Rocky Gorge, River Strangways. Started at
half-past seven, and at four o'clock arrived at the Ponds. The day has
been extremely hot, but about noon some heavy clouds came up from the
east and south-east, which made it a little cooler, and enabled me to get
all the weak horses through; one of them showed symptoms of giving in
before we reached the Ponds, but we got him in all right. I shall remain
here until Monday morning, when I shall have again another long journey
without water (thirty-five miles) to Daly Waters. At sundown the clouds
all cleared away, without giving us any rain. Wind, south-east. This
day's journey has completely knocked me up. At one time I thought I
should never have been able to reach this water. I had no idea I was in
such a weak state, and am very doubtful of my being able to stand the
journey back to Adelaide; whatever may occur I must submit to the will of
Divine Providence.

Sunday, 24th August, Purdie Ponds. Day hot. Wind light, from south-east.
About noon a few clouds came up, but they all disappeared about sundown.
Very little improvement in me to-day.

Monday, 25th August, Purdie Ponds. Started at seven o'clock on my former
tracks towards Daly Waters. At seven miles south of the Blue-grass Swamp
saw a heavy fog to the east, in the same place that I saw the black fog
in coming up; it must be caused by a large body of water in that
direction. The natives have been running our tracks, and have burnt the
grass on both sides of it for some distance. There seem to be very few of
them about this part of the country. At half-past four passed the large
swamp that receives the surplus water of Daly Waters, with water still in
it, but very much reduced. At a quarter past five o'clock arrived at Daly
Waters; found them also very much reduced, but still an abundant supply.
Got all the weak horses through, which is more than I expected. This long
journey has again completely exhausted me, and I feel very ill. Wind,
south-east, with a few clouds.

Tuesday, 26th August, Daly Waters. I feel a little better this morning,
but still very weak and languid. I shall give the horses and myself a
rest to-day, for I am quite unable to ride. Wind, south-east, with a few
clouds from the same direction.

Wednesday, 27th August, Daly Waters. Last evening, about half-past seven,
Thring observed a comet bearing about 20 degrees west of north, and about
15 degrees above the horizon; the tail is short and the nucleus large. I
regret that I am unable to see it. I cannot now see a single star,
everything at night is total darkness. I should like to take some
observations of it, but I am quite debarred from doing so. Started at
half-past seven and proceeded along the Daly Waters, in which we saw an
abundant supply. On reaching McGorrerey Ponds, and finding plenty of
water, camped. I feel a good deal better to-day, but the motion of
travelling on horseback is still very severe. Although Daly Waters is
much reduced, there is still enough to last six months longer, even
should no rain fall. These ponds will also hold out about three months
longer. Wind, strong from south-east, with a few clouds.

Thursday, 28th August, McGorrerey Ponds. Proceeded to King's Ponds and
camped. Find that the natives have been running our tracks, and have
burnt large patches of grass; at this camp they have burnt it round. The
water here is nearly all dried up; a few days later and I should not have
got a drop. There is enough to last me to-night and to-morrow morning.
Strong wind from south-east. The natives have cut on one side of my
initials, on a gum-tree by the water where we camp, a figure resembling
(a stylised flying bird).

Friday, 29th August, King's Ponds. Started at quarter past seven;
proceeded to Frew's Pond, but was disappointed to find it quite dry. Dug
down two feet, but could find no water. Proceeded on a straight course
for Newcastle Water. Crossed Sturt Plains, and after dark camped on them.
I would have gone to Howell Ponds, but finding the others so nearly dry,
I was doubtful of them. A little before sundown, after I had passed them
some distance, I observed flocks of pigeons flying towards them, showing
that there is water still there. It is too late for me to go there now,
Newcastle Water being the nearest. Wind, south-east. I feel a little
better than I did on the former long journey.

Saturday, 30th August, Sturt Plains. At dawn of day started, being still
some eight miles from Newcastle Water. The horses look very wretched this
morning, especially the weak ones. About half-past eight arrived there,
and found an abundant supply of water, though much reduced. No rain seems
to have fallen since we left this, upwards of four months ago. A short
time before we arrived a number of natives were observed following at a
distance behind the rear of the party. They followed us on to our old
camp, when I sent Mr. Kekwick up to them to keep them amused until I had
the horses unpacked and taken down to water. By giving them a
handkerchief he obtained a stone tomahawk from them. They are a fine race
of men, tall, stout, and muscular, but not very handsome in features.
They were very quiet. By making signs they were made to understand they
were not to come nearer to our camp than about one hundred and fifty
yards. They remained until noon staring at us and our horses. Some who
could not see us very well got into the gum-trees, and had a long look at
us. They were seventeen in number; four of them were boys, one of them
much lighter than the others, nearly a light yellow. At noon they all
went off, after remaining for four hours. Once more have I returned, if I
may so call it, into old country again, after an absence of four months
and ten days, exploring a new and splendid country from this to the
Indian ocean without receiving a single drop of rain, or without any
hostilities from the natives. I have returned from the coast to this in
one month and three days. The horses have been one night without water,
but got it early next morning, between eight and nine o'clock, and they
would not have been without it if I could have seen to have guided the
party after sundown. After the rays of the sun have left the earth, all
is total darkness to me, even if there is a moon; I was therefore
compelled to camp until daylight. Had my horses been in anything like a
fair condition to have done a day's journey, and my health permitting, I
could have accomplished the journey from the coast to this in three
weeks. Before sundown we were again visited by our black friends; this
time two old men accompanied them, whom Mr. Kekwick recognised as among
those who visited the Depot at Howell Ponds during my absence. They all
came up this time painted in red and white, and after remaining a short
time went quietly to their camp. Wind, south-east.

Sunday, 31st August, North Newcastle Water. The natives again visited us
this morning, and after remaining some time went off quietly. Wind,
south-east. Few clouds at sundown.

Monday, 1st September, North Newcastle Water. Whilst saddling the horses
this morning the natives again came up, and were anxious to know if they
might be permitted to visit the camp after we were gone; that of course I
had no objection to. They have been very quiet and peaceable during our
stay; but I suppose they observed that both night and day we were always
prepared to resist any aggression on their part. Started at seven
o'clock, and proceeded by the base of the Ashburton range to my former
camp on the East Newcastle Water. Distance, twenty-five miles; course
nearly south-east. Arrived at four o'clock and found the water much
reduced, but still in great abundance. Not a drop of rain has fallen
since we left. There are, apparently, two tribes of natives on this
water, one inhabiting the north and the other the south; for, on those of
the north visiting us, we could not recognise any of those we saw on the
southern water. One of the natives was a very amusing little fellow,
rather less than five feet high, having a very peculiar and comical
countenance and antics that would have eclipsed Liston in his best days,
and as supple in the movements of his joints as any clown on the stage.
He imitated every movement we made, and burlesqued them to a very high
degree, causing great laughter to his companions and us. He seems to be
the buffoon of the tribe. The other natives delighted in making sport of
him, by ridiculing the shortness of his stature and laughing at him
behind his back. Wind, south-east.

Tuesday, 2nd September, East Newcastle Water. Proceeded to Lawson Creek,
but found no water in the lower part. Went up into the gorge, and there
found as much as will do; it also is nearly gone, but there are still a
few feet of it. I had no idea that such a body of water could have
evaporated so quickly, which now makes me very doubtful of the waters to
the southward. Wind, south-east.

Wednesday, 3rd September, Lawson Creek. As I now do not expect to get
water before I reach the Hunter or the Burke, a distance of upwards of
forty miles, I shall give the horses one day's rest to enable them to do
the journey. I expect to lose some of the weak ones; to delay longer is
only making the risk the greater. This must be an uncommonly dry season;
not a single drop of rain has fallen in this part of the country since we
left it. Last year we had three days' rain about the middle of June, and
I was in hopes there would be the same this year, but am very much
disappointed. I shall lighten the horses as much as I can possibly do, by
leaving the water-bags, which are nearly useless, blankets, rugs, and
cloths, as well as any other articles that can be done without.
Provisions I MUST carry. I sincerely hope the forthcoming equinox will
give me some rain and enable me to return. I feel a little better, but
very weak and feeble from the severe attack of scurvy. My mouth and gums
are so sore that to eat any food gives me the greatest pain. I cannot
chew it, and am obliged to swallow it as it is, which makes me very ill.
I am the only one of the party that is at present troubled with it. Wind,
east.

Thursday, 4th September, Lawson Creek. Started at 6.40, and proceeded to
the Hawker, but found no water there; thence to Watson Creek, none there;
thence to Powell, Gleeson, and a number of other creeks that had water in
them last year, but there is not a drop. Continued on to the creek that I
camped at coming up. Arrived at 6.45 p.m.; found that water also gone,
although it was a large deep hole when we were here before. Camped. Weak
horses nearly done up. About 8.30 p.m. sent Thring up the creek to see if
he could find any water. In three hours he returned: he had followed it
up into the rough rocky hills until he could get no further, without
seeing a drop. Wind, east. A few clouds at sundown coming from the south
and south-east.

Friday, 5th September, Branch Creek of Hunter. Had to watch the horses
during the night to prevent them straying in search of water. Started at
5.40 a.m. for the Hunter; in an hour and three quarters found some water
in its bed. Camped, and will give the horses the benefit of it to-day.
Wind variable.

Saturday, 6th September, The Hunter. Proceeded to the Burke, and found an
abundant supply of water in the large iron conglomerate water hole that I
discovered last year; it is reduced about four feet, but is still deep,
and will last yet a long time without rain. I should say it was permanent
without a doubt. Camped. From here I shall require to send on in advance
to see if there is water in the Tomkinson; if not, I shall require to
rest the horses here for three or four days to enable them to do the
journey to Attack Creek without it. If there is none in the Tomkinson, I
do not expect to find any in the Morphett. Native smoke about. They have
burnt a great portion of the grass about here. The day has been
oppressively hot and close. Wind from the east-south-east, with heavy
clouds from the south-east to the south-west at sundown.

Sunday, 7th September, The Burke. After sunrise the clouds all gone. At
6.30 despatched Thring and King to the Tomkinson to see if there is any
water. The day again oppressively hot, with clouds from south and
south-east. Wind variable.

Monday, 8th September, The Burke. The clouds continued to come up du-ring
the night, but after sunrise they cleared off; still no rain. Between one
and two p.m. Thring and King returned with the disheartening tidings that
there was no water in or about the Tomkinson. I shall give the horses two
more days' rest, and push through to Attack Creek, where I am almost sure
of there being water. The wind variable, sometimes north, east, and west.
The clouds are broken up, and are nearly all gone, without leaving rain.

Tuesday, 9th September, The Burke. Resting horses, mending saddle-bags,
etc. Wind, north and variable, with a few clouds from the west and
south-west.

Wednesday, 10th September, The Burke. Thring on his return last Monday
saw some water about four miles higher up this creek, nearly on our
course for the Tomkinson; to that I shall go to-day, and make a start for
Attack Creek to-morrow morning. Every mile now gained is of the utmost
importance to me. Started early, to get there in the cool of the morning.
In an hour and a half arrived at the water and camped. It is situated at
the foot of some ironstone conglomerate rock, and will last a week or two
longer. It has a number of small fish in it. The soil on its banks is
light and a little sandy, with spinifex and grass mixed through it. Wind,
north and north-west; the clouds have all disappeared. This morning I
again feel very ill. I am very doubtful of my being able to reach the
settled districts. Should anything happen to me, I keep everything ready
for the worst. My plan is finished, and my journal brought up every
night, so that no doubt whatever can be thrown upon what I have done. All
the difficult country is now passed, and what remains is well known to
those who have been out with me before; so that there is no danger of the
party not finding their way back, should I be taken away. The only
difficulty they will have to encounter is the scarcity of water, caused
by the extreme dryness of the season.

Thursday, 11th September, The Upper Burke. Started at 6.40; crossed the
Tomkinson and small grassy plains; ascended the north spur of the
Whittington range. After sundown, it becoming quite dark to me, so that I
could not see the horse's head before me, I was compelled to halt on the
top of the range, four miles from my former camp on the Morphett. Day
excessively hot; myself and horses have felt it very much. Wind variable,
from the north and north-east.

Friday, 12th September, Top of Whittington Range. At break of day
started over the range to my former camp, but found all the water gone.
Proceeded down the Morphett, and at four miles found a little in the
sandy bottom of what had once been a large hole. There is as much as
will do for me until to-morrow by digging. All the clouds gone; not the
slightest appearance of rain. The country on fire all round us. Wind,
north-west and variable. Day exceedingly hot.

Saturday, 13th September, The Morphett. Started at 7.20, crossed the
other spur of Whittington range, and at 11.20 arrived at Attack Creek.
There is still an abundant supply of water, although much reduced--much
lower than I have ever seen it. In about an hour and a half after
camping, some native women came to the lower end of the hole where
Billiatt was getting some water. The moment they saw him they went off at
full speed. In a short time afterwards one man made his appearance and
came marching up towards us. Sent Mr. Kekwick to meet him. As he
approached him the black became stationary, and moving back a little,
beckoned to some others to come up. Mr. Kekwick observed five or six
others down at the lower end of the water hole, one of whom came up. I
then sent Frew to Mr. Kekwick. They approached very cautiously, but as
soon as they caught sight of Mr. Kekwick's gun, he could not get near
them. On laying it down he got a little nearer; they shrank back when he
attempted to touch them. Taking out a small strip of white calico which
he had in his pocket, he tore it into two and held it out to them. They
wished to possess it, but did not fancy coming too close to him for it.
He made a sign that he wished to tie it round their wrists; they
gradually approached nearer, holding out their arms at full length, and
so frightened were they to come close, that he had to reach out his full
length to tie them on; after which they gained a little more confidence,
pointed towards the gun, imitated the report with their mouth, and held
up three fingers, signifying that they recollected my first visit and
number, which they do not seem to have forgotten, and seem to dread the
appearance of a gun. The first one that came up had a very long spear,
with a flat, sharp, and barbed point. They were two elderly stout men,
one very much diseased and lame. They remained a long time looking at us.
None of the others came up. In a little more than three hours they went
off and we saw no more of them during the evening. Wind, south-west, with
heavy clouds from the same direction and from the south.

Sunday, 14th September, Attack Creek. During the night the sky frequently
became overcast with heavy clouds, which seemed to indicate rain, but
none fell. About eight o'clock the wind changed to north-east, bringing
up very heavy clouds, which led me to expect rain, but I was much
disappointed, for at half-past twelve they all broke up and went off.
This morning, at sunrise, I despatched Thring and Nash to see if there is
water in Hayward, Phillip, Bishop, Tennant, or Goodiar Creeks. If there
is none I shall require to rest the horses for three days, and then push
on for the Bonney. It is a very long distance, and only the very best of
them will be able to do it. I feel a little better this morning, but
still very weak. The pains are increasing in my limbs, and my mouth is so
bad I can eat nothing but a little boiled flour. How I am to get over
such long pushes I do not know. I must trust entirely to Divine
Providence. The natives have not visited us this morning. A little before
four o'clock p.m. Nash returned. Thring had sent him back to report that
there was water, by digging in the sand, at Hayward Creek, while he goes
on to see if there is any other creek. Wind variable, with heavy clouds
at sundown.

Monday, 15th September, Attack Creek. Started at 8.40. On crossing the
creek, one of the weak horses, which had eaten some poison about the
Roper, and which has been getting weaker every day, in attempting to get
up the bank, which was not steep, fell and rolled back into the creek.
There he had to be some time before he was able to get up. I saw that it
was useless taking him any further, therefore left him where he will get
plenty of feed and water. Proceeded to the Hayward, where I met Thring.
There is some soft mud in Phillip Creek, but none in Bishop Creek.
Camped, and cleared out a place for the horses to drink at. A number of
natives have been camped on the opposite side of the creek, where they
have left their spears, dishes, etc. Thring had arrived here some time
before. About twenty of them coming closer to him than was safe, he
mounted his horse and chased them to the hills, where they are now seated
watching us. Some of them are approaching nearer. Mr. Kekwick could not
get them to come near him until one of the old men who visited us at
Attack Creek arrived and came up to him, which gave the others
confidence. A number of them then came forward--tall, stout, well-made
fellows, armed with long heavy spears, having bamboo at one end. One of
them had also part of a large sea-shell, but it is so broken and ground
down for a scoop that I cannot say of what description it is. The bamboo
and the sea-shell show that this tribe has communicated with the
sea-coast. They remained until sundown, and then did not seem inclined to
go away, but prepared sleeping-places for the night--a proof that this is
the only water near. There are upwards of thirty men, besides women and
children. Wind, south-east. Clouds all gone.

Tuesday, 16th September, Hayward Creek. The natives showed themselves
again at daybreak, but kept on the opposite bank of the creek, having a
long look at us, and calling out something at the top of their voices
which we could not understand. Watered our horses, saddled, and moved on
amidst a succession of yells and screeches from old and young. Proceeded
across Short ranges, and Phillip, and Bishop Creeks. Looked into every
place I could think of, but could not find a drop. Moved on to Tennant
Creek. Found that dry. Tried digging in the sand, without effect. Pushed
on to the large rocky water hole in Goodiar Creek, where I made almost
sure that I should find some. On arriving, was sadly disappointed to find
that dry also. Proceeded across the McDouall range, and camped on a
grassy plain between it and Mount Samuel. The natives followed us nearly
to Tennant Creek, raising a line of smoke all the way. They kept about a
mile to the east of us, on some rising ground that runs nearly parallel
with my tracks. We have had to lighten a heavy cart-horse named Charley.
When any hardship is to be undergone, he is always the first to show
symptoms of giving in. He had only thirty pounds to carry to-day, and he
looks ten times worse than those that are carrying one hundred and
twenty. I shall require to let him go without anything to-morrow. We
shall have to watch the horses during the night to prevent them from
straying in search of water. Wind, south-east.

Wednesday, 17th September, McDouall Range. Started at daybreak for the
Murchison range. About eleven o'clock the cart-horse gave in, and would
not move a step further. I am obliged to leave him; he has been carrying
nothing all the morning. Two others that have been very weak from eating
some poisonous plant will, I fear, give in before the end of the day. A
little after four o'clock I found I must leave them. At dark arrived at
the Baker, which I found dry. Camped. This is another night the horses
will be without water, and will require to be watched. A quantity of
native smoke about. There must be permanent water about this range
somewhere, but I have no time to look for it now. Tomorrow I must push on
for the Bonney. If that fails me I shall be in a sad predicament, but I
trust that the Almighty will still continue to show me the same great
kindness that he has done throughout my different journeys. There is very
little improvement in my health. I feel very much being in the saddle so
long. Twelve hours is almost too much for my weak state, but I must
endure it. Wind, south-east.

Thursday, 18th September, Murchison Range. Proceeded at daydawn to the
Gilbert. Found it dry. Went on towards the Bonney; crossed the
McLaren--no water. At two o'clock arrived at the Bonney, and am most
thankful to Divine Providence that there is still a good supply of water
that will last some time longer. My horses look very bad indeed. I
expected to have lost more of them. They have got over this first
difficulty very well. Towards the end of the journey my old horse took
the lead. Day hot. Wind, south-east.

Friday, 19th September, The Bonney. From this camp Mount Fisher bears 119
degrees 30 minutes. I must remain here some time to get my horses round
again. A large number of them are looking very ill this morning. Being so
long without water and the dry state in which the grass is, has reduced
them more than three months' hard work would have done. If the grass had
any nourishment in it, two or three days would have done for them. Not a
drop of rain seems to have fallen here for the last twelve months;
everything is dry and parched up. This appears to be the driest part of
the year. I am very doubtful of the water in the Stirling, the next place
that I was depending upon. From the very reduced state in which this is,
I have very little hope of there being any there. The day has been again
oppressively hot. I trust we shall soon have rain. Wind variable. Native
smoke about.

Saturday, 20th September, The Bonney. Resting horses. I feel very ill
again; being so long in the saddle is very severe upon me. Day again very
hot. Wind from the west, with a few clouds, which I trust will bring up
rain.

Sunday, 21st September, The Bonney. Resting horses. Day very hot. Wind,
west; clouds broken up.

Monday, 22nd September, The Bonney. This morning sent Thring up the creek
to see if there is any larger water than this that can be depended on for
some time to come. Very hot. Clouds all gone. Wind variable.

Tuesday, 23rd September, The Bonney. Recruiting horses, etc. About eleven
o'clock Thring returned. He has been about twenty miles up the creek to
where it became much narrower and was joined by a number of small ones
coming from very rough and stony hills. Its general course is about
east-south-east. At four miles from this he found a pool of water four
feet deep, two hundred yards long, and thirty feet broad. There is a
considerable quantity of water all the way up, but shallow, and none of
the extent of the former one found. Should I be forced to retreat, that
will be a safe place to fall back on until rain falls. Day again
oppressively hot. Wind, east.

Wednesday, 24th September, The Bonney. Shortly after sunrise despatched
Thring to see if there is any water in Thring Ponds, or any between them
and this. I would have gone myself, but was quite unable to do so, being
very little better. One of my good horses has met with an accident in
feeding along the bank of the creek in places where it is very
precipitous. A portion must have given way and thrown him into the creek,
injuring him very much in the chest and other parts of the body. I am
afraid he will not be able to travel with me, which will be a great loss,
having so many weak ones already. Wind, south-east, with a few clouds.

Thursday, 25th September, The Bonney. Clouds all gone, no rain. Resting
horses, etc. Day hot, morning and evening cool, with strong wind from
east and south-east. I have been obliged to reduce the rations to five
pounds of flour and one pound of dried meat per week for each man, which
will leave me provisions at that rate until the end of January, in case I
should be locked in with the dry state of the season. The flies at this
place are a perfect torment. A little after three o'clock p.m. Thring
returned. There was no water in the Barker, none in the Sutherland, and
when he got to the ponds, found them quite dry also; he then returned two
miles to where there was some good feed for the horse, and camped for the
night without water, intending to return to this in the morning. In
saddling he observed some crested pigeons fly past him to the south of
east; he thought it would be as well to follow them some distance in that
direction, as they might be going to water, as about that time in the
morning is generally the time they fly towards it. After going a few
miles he surprised fourteen natives at breakfast. As soon as they saw him
they ran off at full speed. Observing some small wooden troughs with
water in them, he collected it together and gave it to his horse.
Examined the small creek for more, but could find none, and knowing the
natives would not carry it very far, and that there must be some no great
way off, went on a little further and found a fine pool of water with
ducks on it, but shallow. He then returned. This will bring the Stirling
within visiting distance. I shall remove the party down to the pool
to-morrow. Strong wind, still from the south-east.

Friday, 26th September, The Bonney. In consequence of the horses
separating during the night, I did not get a start before nine o'clock;
followed my former tracks across Younghusband's range; thence on a
bearing 25 degrees east of south; arrived at the pool of water at 5.15
p.m. Before reaching the water we crossed four red sand hills, with
spinifex, running north-east and south-west, having broad valleys
between, in which are growing melaleucas, gum-trees, and grass. After
rain they retain water, but now are quite dry. This one that we are now
camped at is much larger, having the same description of timber, with
polyganum growing round about it; the water is shallow, and will not last
long. There are a number of ducks, geese, and other water-fowl on it, but
too shy to be approached. A quantity of native smoke about. I am very ill
to-day; I am scarcely able to endure the motion of the horse thus far.
The horse that injured himself so much knocked up about two miles from
this water, but we were able to get him to it before sundown. I shall
have to kill him and eat what is good of him; it is useless to attempt
taking him on a long journey without water--he would never be able to do
it; and, as we are now upon half rations of meat, I shall kill and eat
him, so that he will not be lost altogether. Wind variable. Day
exceedingly hot.

Saturday, 27th September, Pool of Water. Before attempting to see if
there is water in the Stirling, I have sent Thring on course 20 degrees
west of south, to see if there is any creek or water between two stony
ranges of hills that lie east of Mount Morphett. At sundown he has not
returned. Wind, west. Day very hot. After sundown we shot the black horse
that was not able to travel; shall cut him up and dry him to-morrow;
there are some parts very much injured by bruises he got in his tumble.
He also showed evidence of having drunk too much water at the Bonney.
Being so exhausted and knocked up on my arrival there, I was unable to go
and see they did not drink too much, and had to leave it to others. In
all my journeys (and my horses have been much longer time without water
than this), this is the first horse that has injured himself in that way.

Sunday, 28th September, Pool of Water. About eight o'clock, Thring
returned, being out all night without food or blankets; he had found a
large gum creek in the place I had sent him to, with water in it, by
sinking in its sandy bed. I shall move the party to it to-morrow morning.
Wind variable, mostly from the north and north-east. Day very hot.
Latitude, 20 degrees 47 minutes 59 seconds.

Monday, 29th September, Pool of Water. Started at seven o'clock, course
20 degrees west of south. For the first five miles we passed over a fine
country, soil red, and in places a little sandy, with gums, grass, a
little scrub, and in places a little spinifex. After this it became
covered with spinifex until within five miles of the creek, where the
mulga commenced, with plenty of grass, which continued to its banks,
where we arrived after twenty-six miles, and had to dig six feet in the
sand before we could get sufficient water for the horses; by ten o'clock
p.m. however we got them all watered. I am inclined to believe this is a
continuation of the Taylor and other creeks coming from Forster range
more to the eastward. After my arrival here, I sent Thring up the creek
to see if he could find any surface water. After dark he returned and
informed me that he had followed it into the Crawford range, and that it
came through the range; if such is the case, there is no doubt of it
being the Taylor with the creeks from Forster range. There is no surface
water, but apparently plenty by digging in the bed of the creek, judging
from the number of native wells that he saw with water in them. At one of
the wells he saw several natives, who ran off on his approach. Latitude,
21 degrees 9 minutes 30 seconds. Wind variable. Day oppressively hot.

Tuesday, 30th September, The Taylor. As soon as I could get the horses, I
despatched Thring to the Stirling to see if there is water. I have sent
King on with him, with a pack-horse carrying two bags of water for the
horse that carries him to the Stirling. They are to follow this creek up,
and, if it is the Taylor, they are to stop to-night at our last camp on
it. Next morning King is to return to me, whilst Thring goes on to
examine the Stirling. Still all hands engaged in sinking for water for
the horses. Wind from the south-east, with heavy clouds from the
north-west and south-west, showing every indication of rain, which I
sincerely hope will fall before morning.

Wednesday, 1st October, The Taylor. About nine o'clock last night there
were a few drops of rain, and almost immediately afterwards the clouds
broke up and went off to the south-east, to our very great
disappointment. This morning there are still a few light ones about, but
very high, and no more appearance of rain. Wind still strong and blowing
from the same quarter. We have now got enough water for the horses, and
can water them all in about two hours. No natives have shown themselves
since we have been here, although their smoke was quite close to us
yesterday. In the afternoon Thring and King returned, having found a fine
pool of water about fifteen miles up the creek, four feet deep, which
will serve us for a short time. Sundown: still blowing strong from the
south-east; clouds all gone.

Thursday, 2nd October, The Taylor. Started at five minutes to eight,
course 3 degrees west of south; at five miles got through the gap in the
range, then changed to 20 degrees west of south, and after ten miles on
that course reached the water hole. The journey to-day has been over
first-rate travelling-ground, avoiding crossing the range at Mount
Morphett. The country in many places along the creek has large grassy
plains with mulga, gum-trees, and scrub, not too thick to get easily
through. Native smoke under the hills to the east. Strong cool wind
blowing all day from the south-east. A little before sundown three
natives came within three hundred yards of the camp, setting fire to the
grass as they came along. We could not get them to come any nearer.
Latitude 21 degrees 22 minutes 12 seconds.

Friday, 3rd October, Surface Water, The Taylor. Shortly after sunrise
despatched Thring and King in search of water higher up the creek. I feel
so weak and ill that I am now scarcely able to move about the camp. This
morning Frew, in searching for some of the horses, came upon the three
natives we saw last night; the moment they saw him off they went at full
speed, and he saw no more of them. They must have been sneaking about and
watching our camp during the night. Wind still blowing strong south-east.

Saturday, 4th October, Surface Water, The Taylor. It still continues to
blow very strong from the same quarter. A little before two p.m. King
returned. They had followed up this creek for a considerable distance
beyond where the Taylor joined it, and as it came more from the
south-east than I had expected, and approached near to Forster range,
Thring changed his course to the Stirling, according to my instructions.
A little before sundown they arrived at my former camp on the Stirling;
found the water hole quite dry; dug down, but could find no moisture.
They had not seen a drop of water during the whole day. In the morning
King returned to me, giving Thring's horse the water that he had carried
with him to enable him to search the Stirling down and round about the
adjoining country. Still blowing strong from the same direction. No
clouds visible.

Sunday, 5th October, Surface Water, The Taylor. Still blowing strong and
cool from the same quarter. About half-past one o'clock Thring returned;
he could find no surface water, neither any to be had by digging. He then
crossed over to the foot of the Hanson, where he saw some native smoke;
on his arriving at it he surprised a native busily engaged in sinking for
water, about six feet deep, in the bed of the creek, who, as soon as he
saw him, jumped out of the well and ran off as fast as he could. He then
tried to see what quantity of water was in the bottom of the well, but
having nothing but a quart pot to clear it out with, he was unable to
form a correct opinion, but from all appearances he thinks there will be
sufficient for our use for some time, only it will require an immense
deal of labour and time to remove the great body of sand to enable the
horses to get down to it. To-morrow I shall send Thring with McGorrerey
and Nash, with four horses and sufficient provisions for a fortnight. On
their arrival at the native well on the Hanson they will be able easily
to get water enough for their four horses that night. McGorrerey and Nash
will then clear out the well and see what quantity there is in it, while
Thring will proceed up the Hanson to see if there is water in the springs
that I discovered on my first journey through the centre. If they are dry
he will proceed with the examination of the Hanson to above where we
crossed it; he will then return to the diggers; by that time they will be
able to judge if there is sufficient water for the whole party. If there
is sufficient he will leave them to dig, and come on to me; if not, and
there is no more water higher up, he will bring them on with him, and I
shall require to try a course more to the south-east. In the afternoon
the three natives again made their appearance, bawling out as they came
near, but retreated as Mr. Kekwick went towards them to see what they
wanted. Wind still south-east.

Monday, 6th October, Surface Water, The Taylor. Shortly after sunrise
despatched Thring with McGorrerey and Nash to the Hanson. Day very hot. I
am still very ill--no improvement whatever. Wind strong from the
south-east.

Tuesday, 7th October, The Taylor. What a miserable life mine is now! I
get no rest night nor day from this terrible gnawing pain; the nights are
too long, and the days are too long, and I am so weak that I am hardly
able to move about the camp. I am truly wretched. When will this cease?
Wind, south-east.

Wednesday, 8th October, The Taylor. Wind still blowing from the
south-east; no appearance of rain.

Thursday, 9th October, The Taylor. Last night, about sundown, a native
woman and youngster came to the waterhole, rushed down, had a drink, and
were running off again, when I cooed and made signs of friendship; in a
few seconds the woman gained confidence, and, not seeing any of us
approach, went down to the hole again, and fetched up a large troughful
of water. Mr. Kekwick tried to induce her to stop, in order to gain some
information from her, but it was of no use; the faster he walked the
faster she did the same, chatting all the time, pointing to the south; so
he left her to walk at her leisure. They do not seem to be at all
frightened of us; but we cannot get any of them to come near, although we
have tried every time they have come. The day again oppressively hot. I
still feel very ill. Wind from south-east. Nothing particular has
occurred during the day. This is dreadful work to be detained here so
long. I am afraid soon I shall not be able to sit in the saddle, and then
what must I do? I feel myself getting weaker and weaker every day. I hope
the Almighty will have compassion on me, and soon send me some relief. He
is the only one that can do it--my only friend.

Friday, 10th October, The Taylor. Last night, a little before sundown,
until after dark, we were amused by a farce enacted by the natives,
apparently to keep us quiet and render us powerless, while they
approached the water hole and got what water they required. They
commenced at some distance off, raising a heavy black smoke, (by setting
fire to the spinifex), and calling out most lustily at the top of their
voices. As the sun got lower I had the party prepared for an attack; on
they came, the fire rolling before them. We could now occasionally see
them; one was an old man with a very powerful voice, who seemed to be
speaking some incantations, with the most dreadful howl I ever heard in
my life, resembling a man suffering the extremes of torture; he was
assisted in his horrid yell by some women. As the evening got darker and
they were within one hundred and fifty yards of us, and nearly opposite
our camp, the scene was very pretty--in fact grand. In the foreground was
our camp equipment with the party armed, ready to repel an attack. On the
opposite side of the creek was a long line of flames, some mounting high
in the air, others kept at a low flickering light. In the midst of the
flames the natives appeared to be moving about, performing all sorts of
antics; behind them came the old man with his women. At every high flame
he seemed to be performing some mysterious spell, still yelling in the
former horrid tone, turning and twisting his body and legs and arms into
all sorts of shapes. They appeared like so many demons, dancing,
sporting, and enjoying themselves in the midst of flames. At last they
and their fire reached the water hole after continuing this horrid noise
for nearly two hours without intermission; as soon as they came in sight
of the water, those in front rushed down into it, satisfied themselves,
filled their troughs and bags, except the old man, who kept up his howl
until he was stopped by a drink of water. This seemed to satisfy them,
for they went off from us about three quarters of a mile and camped, I
suppose thinking they had done great things in keeping us so quiet.
Shortly after this something started the horses which made them all rush
together. I kept the party under arms till nine o'clock p.m. and then,
everything appearing to be quiet, I sent them all to bed except the one
on guard. The natives were quiet during the night. This morning the
blacks watched us collecting the horses and watering them; they then very
quietly slipped down to the water, filled their troughs, etc., and in
about half an hour went off and left us in possession of the water. They
must certainly think we are very much to be frightened by fire and a
great noise, or they would never have come in the way they did last
night; they would have been rather surprised had they attacked us, to
find that we could both speak and injure by fire. I am better pleased
that they went away quietly; it is far from my wish to injure one of them
if they will let me pass peaceably through. About two o'clock p.m. Thring
returned; he had examined up the Hanson, but could not find a drop of
water, either on the surface or by digging. On his return to where he had
left the two men to dig, he found there would not be enough water for the
whole party, as it came in so slowly; it is on the top of hard burnt
sandstone; he therefore came on to inform me of the result, leaving the
two men still there. They had been visited by the natives, who appeared
to be inclined to be rather unfriendly at first, but on showing them they
were welcome to use the water as well as the party, they became friendly,
and came over night and morning to fill their troughs and bags. They
pointed to the south-south-east, and made signs, by digging with a scoop,
that there was water in that direction, but how far he could not make
out. This is a sad disappointment to me. I dare not move the party on to
where they are digging, there is too little water. To-morrow morning I
must send Thring and King on to Anna Reservoir to see if there is any
there; if that is dry I shall be locked in until rain falls, and that may
not be before the equinox, in March, a very dismal prospect to look
forward to. I shall start Thring and King to-morrow morning; they will
reach where the diggers are to-morrow night, and will rest their horses
there on Sunday. On Monday morning start for Anna Reservoir--King, with a
pack-horse carrying water, will go on one day with Thring. The water to
be given to Thring's horse night and morning. Thring will proceed to the
Reservoir. King will return to the diggers with the empty bags, have them
filled, and next morning start with fresh horses and the water to meet
Thring on his return in case the Reservoir is dry; this is the only way
that I see it can be done. I now begin to feel the want of my health
dreadfully. Although Thring is a good bushman and does his best, poor
fellow, yet he wants experience and maturer judgment; he has had hard
work of it lately, but he is always ready to start again at any moment
that I wish. Wind, south-east. A few light clouds about.

Saturday, 11th October, The Taylor. The natives camped last night at
their former place; they seem to have given up all their buffoonery. I
suppose they see it has no effect upon us. Shortly after sunrise
despatched Thring and King. The day again oppressively hot, with a few
light clouds from the south. Wind, south-east.

Sunday, 12th October, The Taylor. The natives again encamped in their
former place last night. They came in late and started early this
morning. They always seem to go off to the westward. Day again
oppressively hot. Wind, south-east.

Monday, 13th October, The Taylor. Can see nothing of the natives this
morning; they must have gone off during the early part of last night. We
tried to get near to them yesterday afternoon by making friendly signs,
etc., but the moment we approached them they ran off, and everything we
can think of will not induce them to come near us or allow us to get near
them; they are the most timid race I have ever met with, which I think is
a very bad feature--such are often very treacherous. I should have a much
higher opinion of them if they would come boldly forward and see if we
were friends or foes. Wind from the north; heavy clouds from south and
south-west.

Tuesday, 14th October, The Taylor. During the night there was a deal of
lightning in the south and south-west; clouds about, but high and much
broken. About two o'clock p.m. they collected together and gave a very
promising appearance of a heavy fall of rain; they seemed to be coming up
all round, but the heaviest from the south and south-west. At four
o'clock p.m. it began to lighten and thunder, accompanied by a shower
which did not last above a few minutes. Sundown: still the same promising
dark, heavy, gloomy appearance. Wind, south-east.

Wednesday, 15th October, The Taylor. During the night we had a terrific
storm of lightning and thunder, which continued throughout the night and
morning at intervals, but little rain has fallen, it has merely damped
the surface of the ground. At twelve o'clock to-day it has nearly cleared
all away, leaving only a few light clouds, which is another very great
disappointment. At sundown it again became overcast. Wind variable.

Thursday, 16th October, The Taylor. Still cloudy during the night and
morning, but no rain has fallen; the heavy clouds pass south of us to the
eastward. I am now nearly helpless; my legs are unable to support the
weight of my body, and, when I do walk a little way, I am obliged to have
the assistance of one of the party, and the pains caused by walking are
most excruciating. I get little sleep night or day. I must endure my
sufferings with patience, and submit to the will of the Almighty, who, I
trust, will soon send me some relief. Wind variable.

Friday, 17th October, The Taylor. Still heavy clouds during the night and
day, but no rain will fall. Still very ill. About three o'clock p.m.
Thring returned; he has been to Anna Reservoir and found plenty of water,
and a number of natives camped at it, who ran off the moment they saw
him; he watered his horse and recrossed the range, not thinking it
prudent to camp where there were so many of them. He has met with the
same description of weather that we have had up here, thunder and
lightning with a heavy, cloudy sky, but nothing but a light shower or two
of rain. I shall move the party on to the Hanson to-morrow, and, if I am
able to ride, shall push on to-morrow. Wind variable; sky still overcast.

Saturday, 18th October, The Taylor. Started at twenty minutes to eight
for the Hanson; sky still overcast with heavy clouds. We had two light
showers during the journey. I am now so helpless that I have to be lifted
into the saddle. I endured the pain of riding for the first seventeen
miles far better than I expected; after that it became almost unbearable,
and camped at twenty-four miles, having found as much water in the rocks
of the Stirling as will do for the horses to-night and to-morrow morning,
left from a shower of rain, for which I am very thankful. I could not
have gone on more than three miles. I was then enduring the greatest pain
and agony that it is possible for a man to suffer. On being lifted from
the horse, all power was gone out of my legs, and when I attempted to put
the weight of my body on them the pain was most excruciating. Still heavy
clouds about, indicating rain. Wind, south-east.

Sunday, 19th October, The Stirling. I had a few hours' sound sleep last
night, which I find has done me a deal of good. During the early part of
the night two heavy showers of rain fell, and left plenty of water for
the horses; got them up, and saddled and proceeded to the Hanson. At
eight miles arrived there, finding the party all well; they had not been
troubled with the natives except by their coming down to the water during
the night time, and bringing into the hole a quantity of sand with them.
I had to be taken from horseback nearly in the same state as yesterday.
Wind, south-east.

Monday, 20th October, The Hanson. Started early; passed the Centre;
crossed the upper part of the Hanson, and at five miles beyond it camped.
Distance, thirty-five miles. Not a drop of rain seems to have fallen for
a long time. During the whole day's journey this has been a terrible day
of agony for me; nine hours and a half in the saddle. I had to be taken
from my horse in the same helpless state as before. My feet and legs are
now very much swollen; round the ankles they are quite black, and the
pain is dreadful. I still continue to take the bicarb of potash, but it
has little or no effect. Wind variable.

Tuesday, 21st October, South of the Centre. About sunrise started for
Anna Reservoir, and at 5.30 p.m. arrived there, completely exhausted.
Wind, variable. Heavy clouds from the south-east.

Wednesday, 22nd October, Anna Reservoir. Last night I was so completely
overcome by fatigue and exhaustion that I had no sleep during the whole
of the night, which makes me feel very ill indeed this morning. I shall
be obliged to remain here to-day and to-morrow, to see if that will
recruit my strength and enable me to perform the long journeys to the
McDonnell range. About twelve o'clock heavy thundery weather to the west
and south.

Saturday, 23rd October, Anna Reservoir. I shall rest to-day and have what
shoes there are left put on the horses. I, with William Auld, will
proceed to-morrow about ten miles in advance, to divide the long journey
into two, for I have not strength to do it in one day. Wind variable.

Friday, 24th October, Anna Reservoir. Started early, taking with me
Thring, King, and Auld, with one pack-horse to carry my tent, water, etc.
Proceeded through the thick mulga scrub, and at ten miles camped, which I
find is quite as much as I am able to do. Had my tent put up, and myself
carried into it. Sent Thring and King back with the horses to the
Reservoir, keeping Auld with me. The party will start from the Reservoir
early to-morrow morning, pick me up, and proceed to Mount Harris. Wind,
east.

Saturday, 25th October, Mulga Scrub South of Anna Reservoir. A few
minutes before ten o'clock a.m. the party arrived all right. I was soon
ready and lifted up into the saddle, and started at 10.10. During the day
it has been excessively hot. At 5.45 p.m. arrived at Mount Harris, being
nearly eight hours in the saddle, which is far more than I am able to
endure in my terribly weak state. It is between my shoulder-blades and
the small of my back that I am so much affected while riding. When the
pain from them becomes unbearable I endeavour to get on as far as I can
by supporting my weight upon my arms until they give way. I arrived here
in a state of utmost exhaustion; so much so that I was quite unable to
eat a single mouthful of anything. After we had the horses unpacked, a
few natives made their appearance on the side of the mount, calling out
something and pointing to the north-east. Sent Thring and King to see if
they could make anything of them, but they soon ran down the other side
of the mount, and, when seen again, were marching off in the direction
they had pointed out. They had taken good care before leaving to use
nearly all the water in the crevices of the granite rocks; they left
about a quart. Finding it quite impossible to remain so long in the
saddle as I have done to-day, I got Mr. Kekwick and some of the others to
construct a stretcher during the night, which I hope will enable me to do
a long journey to-morrow. Wind, south-east.

Sunday, 26th October, Mount Harris. Had the stretcher placed between two
horses. Had great difficulty before we could get two that would allow it
to be passed between them. At last succeeded in getting two that we
thought would do very well, as they seemed to go very quietly with it. I
shall continue on horseback until I find that I have got enough of it.
Started a little after sunrise. I found I could continue two hours and a
half in the saddle without fatiguing myself too much. Having done this, I
sent to the rear of the party for the stretcher, when, to my great
disappointment and vexation, I found that a short time before something
had annoyed one of the horses, which set to and kicked it all to pieces,
which is a great misfortune. I continued in the saddle, and proceeded
until I was exhausted, which happened at the end of fifteen miles, when I
was compelled to stop. Keeping Auld with me, and some water, I sent on
the party and all the horses to Mount Hay. If they find water they are to
camp and return for me to-morrow; if not, they are to push on to the
Hamilton Spring; if that is gone, they will have to cross the range to
Brinkley Bluff. I find myself getting weaker and weaker every day. I am
very ill indeed. Wind, south-east.

Monday, 27th October, Hills North of Mount Hay. About 11.30 a.m. King and
Nash returned for me. Thring had found water in one of the gullies, but
the approach to it was very rough and stony indeed. Thring had gone to
see if there was any water in the clay-pans that I had camped at on my
journey up, and if there is, will take the party over there, and will
send one of the men to meet me and inform me of it. The distance from
here to the water is ten miles. Had the horses saddled; mounted, and
proceeded towards it. At the end of two hours the motion of the horse
became so dreadful to me, and the pain I was suffering from was such as
no language can describe; but I still continued in the saddle, and,
within a mile and a half of the water, met Frew, whom Thring had sent to
say that he had found plenty of water in the clay-pans, with green grass,
and that the party had moved on to it. Distance from where we were then
to the clay-pans, six miles further. I could no more sit in the saddle
that distance than I could fly; I am now already completely exhausted,
and have still a mile and a half to ride before I can reach the other
water. To that I must go, and see what a night's rest will do in the
morning. While taking a drink of water, I was seized with a violent fit
of vomiting blood and mucus, which lasted about five minutes, and nearly
killed me. Sent Frew on to the party. Went on the best way I could with
the other three to the water. Arrived there feeling worse than I have
ever done before. I have told King and Nash to remain with me in case of
my dying during the night, as it would be lonely for one young man to be
here by himself. Wind, south-east.

Tuesday, 28th October, Mount Hay. Started in the cool of the morning, and
in two hours reached where the party were camped, so much exhausted and
so completely done up that I could not speak a word--the power of speech
has completely left me. I was lifted from the saddle and placed under the
shade of a mulga bush. In about ten minutes I recovered my speech. I find
that I can no longer sit on horseback; gave orders for some of the party
to make a sort of reclining seat, to be carried between two horses, one
before the other; also gave orders that a horse was to be shot at
sundown, as we are getting rather short of meat, and I hope the change of
beef tea made from fresh meat will give me some increase of strength, for
I am now reduced to a perfect skeleton, a mere shadow. At sundown had the
horse shot; fresh meat to the party is now a great treat. I am denied
participating in that pleasure, from the dreadful state in which my mouth
still is. I can chew nothing, and all that I have been living on is a
little beef tea, and a little boiled flour, which I am obliged to
swallow. To-night I feel very ill, and very, very low indeed. Wind,
south-east, with a few clouds.

Wednesday, 29th October, Clay-pans East of Mount Hay. This morning I feel
a little relieved in comparison with my exhausted state of yesterday. I
had a very troubled night's rest. All hands cutting up the horse, and
hanging up the meat to dry. Thring and Nash out for two long poles to fix
the chair in, which they succeeded in finding. At twelve o'clock had all
the meat of the horse cut up and hung up to dry. Day oppressively hot.
Wind, south-east. Clouds.

Thursday, 30th October, Clay-pans East of Mount Hay. I think I am a
little better this morning, but still very weak and helpless. Find that
the chair will not answer the purpose, and must have a stretcher instead.
Wind, south-east.

Friday, 31st October, Clay-pans East of Mount Hay. I felt a little
improvement this morning, which I hope will continue; and I think I have
reached the turn of this terrible disease. On Tuesday night I certainly
was in the grasp of death; a cold clammy perspiration, with a tremulous
motion, kept creeping slowly over my body during the night, and
everything near me had the smell of decaying mortality in the last stage
of decomposition and of the grave. I sincerely thank the Almighty Giver
of all Good, that He, in His infinite goodness and mercy, gave me
strength and courage to overcome the grim and hoary-headed king of
terrors, and has kindly permitted me yet to live a little longer in this
world. Auld, who was in attendance upon me on that night, informed me
that my breath smelt the same as the atmosphere of a room in which a
dead body had been kept for some days. What a sad difference there is
from what I am now and what I was when the party left North Adelaide! My
right hand nearly useless to me by the accident from the horse; total
blindness after sunset--although the moon shines bright to others, to me
it is total darkness--and nearly blind during the day; my limbs so weak
and painful that I am obliged to be carried about; my body reduced to
that of a living skeleton, and my strength that of infantine weakness
--a sad, sad wreck of former days. Wind variable.

Saturday, 1st November, Clay-pans East of Mount Hay. Although in such a
weak state, I shall try if I can ride in the stretcher as far as Hamilton
Springs. Started early; found the stretcher to answer very well. On
arriving at the springs, saw that there was not sufficient water for the
horses, and, as I had stood this part of the journey so well, made up my
mind to cross the range to Brinkley Bluff. Proceeded, and arrived there
about five o'clock p.m. I have stood the long journey far better than I
expected, but feel very tired and worn out. Wind variable. Cloudy.

Sunday, 2nd November, Brinkley Bluff, The Hugh. Got a few hours' good
sleep during the night, and feel a good deal better this morning. Day
still cloudy. Wind variable.

Monday, 3rd November, Brinkley Bluff, The Hugh. Started at 7.30 a.m. for
Owen Springs. Saw where one of the horses died that I was compelled to
leave behind on coming up. As there is only the hair of his mane and tail
to be seen, and not a single bone, I am inclined to think that he has
been killed, carried off, and eaten by the natives. I expect the other
one has shared the same fate. At 2.20 p.m. arrived at the springs. Plenty
of water. I have stood the journey very well, but am very tired. Wind,
south-east.

Tuesday, 4th November, Owen Springs, The Hugh. Started at 7.20 a.m.,
passing through the gorge of the Waterhouse range. At 1.20 arrived at the
springs under the conglomerate rock, a mile and a half north-east of the
gorge in James range. I feel the shaking of the stretcher very much, and
am again very tired, but am glad to find that I am getting a little
stronger. Wind, south-east. The clouds are all gone.

Wednesday, 5th November, Spring, Conglomerate Rock, The Hugh. Started at
7.25 a.m. Passed through the gorge of James range and proceeded to the
side creek in which water was obtained on coming up. Found some still
there. Camped. Sent four of the party to clear out the hole; in the
meantime sent Thring up the side creek to see if there is any surface
water left from the showers of rain that have fallen here some short time
ago. Since leaving the McDonnell range we have had plenty of green grass,
showing that rain has fallen some time back; it has made no impression
upon the large creek, which is quite dry. In a short time Thring
returned; he has seen as much as will do for forty horses to-night, which
is a good thing. Sent him up with them, and watered the remainder at this
hole, into which the water comes very slowly, in consequence of the main
creek having none in its bed below the sand. I again feel tired from the
shaking of the horses and the stretcher. The swelling of my gums and the
black blisters, which have been so very painful for such a long time
back, are slowly giving way before some vegetable food which I have been
able to get since coming into the green, grassy country; I hope it will
soon cure me. My teeth are still loose, but it is a great thing to get a
little relief from a great mouthful of swollen, blistered, and most
painful gums. When my mouth was closed I had scarcely room for my tongue;
the blisters are now much reduced. Wind, south-east.

Thursday, 6th November, The Hugh. Started at 7.20 towards the Finke; at
five p.m. met with some water in a clay-pan, and camped. I am a little
stronger to-day, and feel that I am gradually improving. Wind,
south-east. Night and morning cool.

Friday, 7th November, North of the Finke. Proceeded to Pascoe Springs in
the Finke; found plenty of water and camped. Day oppressively hot. Wind,
south-east.

Saturday, 8th November, Pascoe Springs, The Finke. Proceeded to Sullivan
Creek and found sufficient water to do for us until Monday morning, and
this being a place for feed for the horses, I shall remain here until
that time. I feel very tired and sore after this rough week's work, and
am glad of a day's rest. I feel a gradual improvement in my health and
strength, which I hope will continue to increase. Wind variable, mostly
from south-east.

Sunday, 9th November, Sullivan Creek. During the night had a few drops of
rain; heavy clouds to the west, north-west, north, north-east, and east.
Wind blowing strong and variable. Sundown: the sky overcast with heavy
clouds.

Monday, 10th November, Sullivan Creek. Some of the horses missing this
morning. Did not get a start till nine o'clock a.m. Day oppressively hot.
Crossed the Finke three times, and arrived at Polly Springs, where there
is plenty of water. Camped. Wind, south-east.

Tuesday, 11th November, Polly Springs, The Finke. Proceeded to Marchant
Springs. Camped. The water is low and rather boggy. Dug a place about
eighteen inches deep in the firm ground, and the water came boiling up. I
am happy to find that I am gaining a little strength again. I was able to
walk two or three steps by leaning upon two of the party, but the pain
was very severe. Wind, south-east; a few clouds about.

Wednesday, 12th November, Marchant Springs, The Finke. As I am not
certain of water at the next two camps, I will rest the horses as well as
myself here to-day, for we both require it very much; it will enable them
to stand a long push if required. A number of showers of rain seem to
have fallen here this month. Wind, south-east.

Thursday, 13th November, Marchant Springs, The Finke. Started at 7.40.
Proceeded towards the Goyder, and at nine miles found myself in as dry a
country as ever; not a drop of rain seems to have fallen here for upwards
of twelve months. On arriving at the Goyder found a little moisture at
the bottom of the sand in the rocks--not enough for the horses. Pushed on
towards the Coglin, and at dark camped in the mulga scrub without water.
Day most oppressively hot. Light wind from south-east.

Friday, 14th November, Mulga Scrub. Started at six o'clock a.m. Examined
the different creeks in which I found water on my journey to the north
but there was not a drop. At twelve miles reached the Coglin--none there.
Country all in the same dry state. Proceeded on to the Lindsay, where I
am sure of water. At four o'clock arrived there and found plenty. Camped.
Thanks be to God, I am once more within the boundary of South Australia!
I little expected it about a fortnight ago. If the summer rain has fallen
to the south of this, there will be little difficulty in my getting down.
I am again suffering very much from exhaustion, caused by a severe attack
of dysentery, which has thrown me back a good deal in the strength I was
collecting so quickly, but I hope it will not continue long. Wind,
south-east.

Saturday, 15th November, The Lindsay. At day-break I have sent Thring to
the Stevenson to see if there is water there, either on the surface or by
digging in the sand; if there is I shall move the party over there
to-day, and on Monday morning start for the Hamilton (I expect no water
between); and if not, I shall remain here till that time and push for the
Hamilton. About ten o'clock a.m. he returned and reported no water, only
a little moisture on the top of the clay beneath the sand. Day very hot.
I still continue to be very unwell. Wind, south-east.

Sunday, 16th November, The Lindsay. Day oppressively hot. Light winds,
south-east.

Monday, 17th November, The Lindsay. Started soon after sunrise, crossed
the Stevenson and the Ross; both quite dry. Proceeded across Bagot range
to the gum water-hole; that is also dry. Found a little rain water in one
of the small creeks, but not enough for all the horses. The day being
excessively hot, the journey very rough and stony, and many of them lame
from want of shoes, also it being near sundown, and there being a little
green grass about, I have camped. Wind variable.

Tuesday, 18th November, The Gums, Bagot Range. Started at 5.40 a.m. to
the large waterhole in the Hamilton; in about a mile found some rain
water, which I allowed the horses to drink. At 10 a.m. arrived at the
large water-hole, and found it very low indeed; a great number of dead
fish all round it. This must certainly be a very unprecedentedly dry
season indeed; this water-hole does not seem to have received any water
for the last two years. The water being old and stagnant, I am afraid
will make us ill; we have all already been suffering much from stagnant
waters we have been compelled to use. I, however, must give the horses a
day's rest to enable them to make the next and last push, nearly a
hundred miles, to the first springs. From the dryness of the season, I
scarcely expect to find water before I reach them, which will be a severe
trial for the horses, the weather being so extremely hot. I am still
suffering very much from the effect of the stagnant waters; they have
sent me back again nearly to my former state of weakness, and have
assisted in checking my recovery from the scurvy, which is now again
gaining ground upon me since I lost the vegetable food. The country being
now so dry, there having been no late rain, there is not a blade of grass
to be seen. Hot wind from the north. This is the first and only hot wind
I have felt during the whole journey from Mount Margaret to the
sea-coast, and back to this place. In the afternoon the sky became
overcast with heavy clouds. At sundown the wind changed to west, and blew
very strong till eleven o'clock p.m.; we then had a few drops of rain,
but not enough to moisten the surface of the ground; after this it became
calm, the clouds broken, and there was no more of it.

Wednesday, 19th November, The Hamilton. This morning still cloudy, but
excessively close and hot. I am glad that I resolved to remain here
to-day, for the poor horses would have felt it very much travelling over
the high and heavy sand hills that we have to go over in the first day's
journey. In the afternoon the sky again became overcast with heavy
clouds, and there was a great deal of thunder and lightning to the west
and north, and again, at the same time as last night, we were favoured
with a few drops of rain; the result the same as it was then. Wind
variable and squally.

Thursday, 20th November, The Hamilton. This morning the clouds have
cleared away, but there is a nice cool strong breeze from the south-east
and east--a fine thing for the horses crossing the heavy sand hills.
Started at six o'clock a.m. Got over them very well, and reached the
mulga plain. About twelve the wind ceased, and it became very hot. In the
afternoon one of the horses (Trussell) began to show symptoms of being
very ill. One of the party was riding him at the time. I had him changed
immediately and allowed him to run loose, but he seemed to have lost all
spirit and soon dropped behind. I then had him led and driven for upwards
of two miles until I reached the Frew or Upper Neale. The dreadfully dry
state of the country since leaving the sand hills--it being completely
parched up--leaving me no hope of getting water until I reached the gap
in Hanson range or the Freeling Springs, and it being quite impossible
for us to drag him on there, I was compelled to abandon him, as it would
only knock up the other horses to drive him on. Proceeded through a still
parched-up country to the large dry lagoon, and at dark camped without
water. Wind, south-east.

Friday, 21st November, Large Dry Lagoon. Started at break of day through
some low sand hills, with valleys and clay-pans, all dry. At a little
more than six miles after starting, I was rather surprised to find recent
tracks of horses that had been feeding on and about our tracks. Thinking
it might be a party out looking for us, as I have now been some time
longer than I anticipated at starting, I sent Thring to examine and see
how many horses there were. In about half an hour he returned, and said
that he could only make out two, and those I immediately concluded were
two of the horses that had given in near this place on my journey to the
north. Proceeded on to the camp where I had buried the two hundred pounds
of sugar, frequently meeting their tracks, apparently in search of water.
Arrived at the camp, but there is not a drop there, and no appearance of
the two horses, but only their tracks in the bed of the creek, following
it down to the eastward, where there must be permanent water that has
supplied them during the past year. A thunder-shower must have brought
them out to visit the spot where they were first left. I should have
liked very much to have regained them, but the dry state of the country
and the want of water will not allow me to look for them. Found that the
things buried had been disturbed, and most of them carried away by the
natives--the others all destroyed--the sugar all gone, except about five
pounds, which was left in the hole and covered up. Proceeded, crossing
side branches of the Neale, but not a drop of water in any of
them--everything dried up. Went on towards the gap in Hanson range. At
about eight miles before reaching it, Frew's horse (Holland) knocked up
with him; he could not get him on a step further, and had to leave him.
On reaching the Lindsay, this horse had been allowed by Frew to drink too
much water, and had not recovered from the effects of it. At dark arrived
at the gap, and found plenty of water, for which I am very thankful, for
there are many of the horses that would not have stood another day's
journey without it. Day exceedingly hot. Wind, south-east.

Saturday, 22nd November, Gap in Hanson Range. Resting horses, etc. Sent
Frew in search of his horse shortly after sunrise. About half-past two he
returned, and reports that he cannot be found; that he had searched round
about the creeks and gullies where he had been left, but could find
nothing of him, and the country was too stony to track him. Day again
very hot.

Sunday, 23rd November, Gap in Hanson Range. Started at six o'clock a.m.,
intending to get to Freeling Springs, but one of the horses that had
eaten poison about the Roper country, and has never recovered from it,
but was always very poor, and of no use whatever, knocked up, and would
not move a step further; being only six miles from where we started, we
left him and proceeded on our journey. About this time the wind changed
to the north, and it came on to blow a fierce hot wind, and by the middle
of the day it was almost unbearable. Two more of the horses knocked up,
and being nearly opposite the McEllister Springs, I turned to them and
camped. These springs required to be dug out before we could get water
enough for all the horses. After opening two of them, we found them to
yield a sufficient supply. Still continuing to blow a terrific hot wind
from the north. A little before sundown it changed, and came on to blow
from the south, and blew the hot wind back again. For three hours it was
as hot as when coming from the north.

Monday, 24th November McEllister Springs. Proceeded to the Freeling
Springs and camped. This journey was as much as the horses are now able
to do. The stagnant and spring waters have weakened them so much that I
shall be compelled to rest them some time at Mr. Jarvis's, Levi's
station, before they will be able to perform the remainder of the journey
to Adelaide, that is, if I can get them that length.

Tuesday, 25th November, Freeling Springs. Found one of the chestnut
horses that was left here. The other one seems to have been taken on to
Mr. Jarvis's. Started shortly after sunrise. Proceeded to the Milne
Springs and camped. The day again extremely hot. Wind still from the
south-east. Twenty miles a day is now as much as my horses can
accomplish.

Wednesday, 26th November, Milne Springs. Proceeded to Mr. Jarvis's
station, Mount Margaret, which I expected to reach without losing any
more horses, but I am disappointed, for I had to leave four behind
knocked up, which I shall be able to recover to-morrow or the next day.
Mr. Jarvis being from home, we were received by his men with a hearty
welcome, and were shown every kindness and attention that was in their
power. Day again very hot. Wind, south-east.

Thursday, 27th November, Mount Margaret Station. Resting horses. Sent out
and had the one that knocked up about two miles from here brought in. I
am still very ill, but am able to walk a few yards without assistance. I
hope a few days will benefit me much. Day very hot. Wind, south-east.
Clouds.

Friday, 28th November, Mount Margaret Station. Resting horses. Still
cloudy. Promising rain. Sent out and had the other three knocked-up
horses brought in all right. Yesterday got in the other chestnut horse
left at the Freeling Springs, and brought down here by Woodforde. Clouds
breaking up. No rain. Wind, south-east.

Saturday, 29th November, Mount Margaret Station. Resting horses, etc. I
find the scurvy is fast gaining upon me, although I have had fresh meat
for the last few days. I must therefore push on as fast as possible down
the country, in order to get some vegetables. I shall start to-morrow
evening, and travel during the night to the William Spring to avoid the
great heat of the day, taking with me the stretcher (for I am not yet
able to ride), three men, and the strongest horses, leaving the rest here
for another week to recover with remainder of the party in command of Mr.
Kekwick, who, as soon as the horses are sufficiently strong, will conduct
the party to Adelaide. Clouds all gone. Wind, south-east.

Sunday, 30th November, Near Mount Margaret Station. Started at five p.m.
for the William Spring with fourteen horses, leaving the weak and done-up
ones at Mount Margaret for another week to recover. I have also brought
on with me Auld, King, and Billiatt. The others I have left with Mr.
Kekwick, to whom I have given command of the party, and who will conduct
them to Adelaide by easy stages, as soon as the horses are able to
travel. I travelled during the night, and arrived at the spring a little
before six a.m. Camped, unsaddled the horses, and turned them in amongst
the young reeds to feed, which they seemed very eager for.

Monday, 1st December, William Spring. During the day the horse that I was
compelled to leave here on my northward journey came towards the others,
but appeared very shy. I left him alone till nearly sundown, when I sent
King to see if he had joined them, and to see if the others were all
right. At dark he returned, and reported them to be all right, and that
the other had joined them. He tried to catch him, but that he would not
allow, so he left him with the others during the night. The day has been
very close and oppressive, with heavy clouds and distant thunder. I am
glad I performed this long journey during the night. Wind, south-east.
Clouds all gone.

Tuesday, 2nd December, William Spring. Got all the horses into camp, and
attempted to catch the stranger, but could not without roping him; I
therefore drove him along with the others to the Beresford Springs, and
then he allowed himself to be caught and hobbled. The journey has quieted
him. It is the longest journey he has had for nearly twelve months. I
arrived about four o'clock p.m., and there being plenty of young reeds,
camped. The day has been again very hot, but occasionally strong breezes
from the south-east and east.

Wednesday, 3rd December, Beresford Springs. Proceeded to Mount Hamilton
Station, where I received a very kind reception from Mr. Brown, and was
treated with the greatest possible kindness. Toward evening I again felt
very ill. Day very hot. Wind, south-east.

Thursday, 4th December, Mount Hamilton Station. I have been very ill
during the night, but started for Chambers Creek. Arrived there about
mid-day, where I again experienced a like hospitable reception and great
kindness from Mr. Lee. Wind variable. Day extremely hot.

Friday, 5th December, Chambers Creek. I shall require to rest my horses
here to-day. I was in great hopes that when I reached this place I should
have been again able to have ridden on horseback, but the waters of the
spring country through which I have just passed have reduced me nearly to
my former state of weakness, and I shall still be compelled to continue
in the ambulance a little longer. I feel a little better this morning--I
suppose in consequence of drinking fresh water. Hot wind from the north.
Towards evening a heavy thunderstorm coming from the westward.

Saturday, 6th December, Chambers Creek. Started at eight o'clock with the
ambulance towards Termination Hill. After crossing numerous sand hills,
we frequently found rain water. Towards sundown arrived at the south side
of Porter Hill. Found rain water, and camped, one of the horses being
nearly knocked up. I shall be compelled to take to the saddle to-morrow,
for the ambulance horses will not be able to carry me further. I must
send them back to the creek, there to rest till the others come down.
Cloudy. Wind variable.

Sunday, 7th December, Porter Hill. Mounted and started at six a.m. I find
that I can endure the motion of the horse better than I expected; but
about mid-day began to feel it very much. Towards four o'clock found some
rain water about ten miles from Termination Hill, for which I am very
thankful, for I could not have continued the journey any further. Camped.
Wind variable.

Monday, 8th December, Termination Hill. During the night had a heavy
thunderstorm and shower from the south-east. Started at six a.m. and
arrived at Mr. Glen's Station at sundown, quite done up; received a
hearty welcome. Encountered a heavy storm of thunder and lightning a few
miles from the station. Wind, south-east.

Tuesday, 9th December, Mr. Glen's Station. Proceeded to Mount Stuart
Station, where I had the pleasure of meeting Mr. John Chambers, who
received me with great kindness. There has been some heavy rain here
lately. Wind, south-east. Day hot.

Wednesday, 10th December, Mount Stuart Station. Accompanied by Mr.
Chambers, proceeded to Moolooloo, and arrived there in the afternoon
completely tired and exhausted from riding in the saddle. Day hot. Wind,
east.

In conclusion, I beg to say, that I believe this country (i.e., from the
Roper to the Adelaide and thence to the shores of the Gulf), to be well
adapted for the settlement of an European population, the climate being
in every respect suitable, and the surrounding country of excellent
quality and of great extent. Timber, stringy-bark, iron-bark, gum, etc.,
with bamboo fifty to sixty feet high on the banks of the river, is
abundant, and at convenient distances. The country is intersected by
numerous springs and watercourses in every direction. In my journey
across I was not fortunate in meeting with thunder showers or heavy
rains; but, with the exception of two nights, I was never without a
sufficient supply of water. This will show the permanency of the
different waters, and I see no difficulty in taking over a herd of horses
at any time; and I may say that one of our party, Mr. Thring, is prepared
to do so. My party have conducted themselves throughout this long and
trying journey to my entire satisfaction; and I may particularly mention
Messrs. Kekwick and Thring, who had been with me on my former expedition.
During my severe illness every attention and sympathy were shown to me by
every one in the party, and I herewith beg to record to them my sincere
thanks.

I may here mention that the accident which occurred to me at the starting
of the Expedition from Adelaide has rendered my right hand almost useless
for life.

The Journal concludes with the following letter:

To the Honourable H.B.T. Strangways, Commissioner of Crown Lands and
Immigration.

Adelaide, December 18, 1862.

Sir,

For the information of His Excellency the Governor-in-Chief, I have the
honour to report to you my return to Adelaide, after an absence of twelve
months and thirteen days; and I herewith beg to hand you my chart and
journals of the Expedition from which I have just returned.

To you, Sir, and the Government, my especial thanks are due for the
liberal manner in which the supplies were voted, and for the kind and
ready assistance I at all times experienced. Also to George Hamilton,
Esquire, Chief Inspector of Police, for the efficient manner in which my
party was fitted out. The original promoters of my various expeditions,
Messrs. James Chambers and William Finke, have always shown the most
lively interest in my success, to which they cheerfully contributed. How
much I regret the unexpected decease of the first-named gentleman I need
here hardly state, for he was indeed heart and soul in the result, and
no one would have felt so proud of my success as my much-lamented and
best friend James Chambers. To Mr. John Chambers I am also under many
obligations for assistance in many instances, and I hereby tender him my
best thanks.

I have the honour, etc.,

J.M. STUART.



APPENDIX.

[FROM THE PROCEEDINGS OF THE ZOOLOGICAL SOCIETY OF LONDON. JUNE 9,
1863.]

ON A COLLECTION OF BIRDS FROM CENTRAL AUSTRALIA.

BY JOHN GOULD, F.R.S., ETC.

The Board of Governors of the South Australian Institute having
liberally forwarded for my inspection a selection from the
ornithological collection made by Mr. Frederick G. Waterhouse during Mr.
Stuart's late Exploratory Expedition into Central Australia, I have
thought the matter of sufficient interest to bring these birds under the
notice of the Society, the more so as it will enable me to make known
through our Proceedings a new and very beautiful species of Parrakeet
pertaining to the genus Polyteles, of which only two have been hitherto
known. Every ornithologist must be acquainted with the elegant
P. melanurus and P. barrabandi, and I feel assured that the acquisition of
an additional species of this lovely form will be hailed with pleasure.
The specific appellation I would propose for this novelty is alexandrae,
in honour of that Princess who, we may reasonably hope, is destined at
some future time to be the Queen of these realms and their dependencies,
of which Australia is by no means the most inconspicuous.

Polyteles alexandrae, sp. nov.

Forehead delicate light blue; lower part of the cheeks, chin, and throat
rose-pink; head, nape, mantle, back, and scapularies olive-green; lower
part of the back and rump blue, of a somewhat deeper tint than that of
the crown; shoulders and wing-coverts pale yellowish green; spurious wing
bluish green; external webs of the principal primaries dull blue,
narrowly edged with greenish yellow; the remaining primaries olive-green,
edged with greenish yellow; under wing-coverts verditer-green; breast and
abdomen olive-grey, tinged with vinous; thighs rosy red; upper
tail-coverts olive, tinged with blue; two centre tail-feathers bluish
olive-green; the two next on each side olive-green on their outer webs
and dark brown on the inner ones; the remaining tail-feathers
tricoloured, the central portion being black, the outer olive-grey, and
the inner deep rosy red; under tail-coverts olive; bill coral red; feet
nearly brown.

Total length 14 inches; bill 1/2; wing 7; tail 9; tarsi 7/8.

Habitat. Howell Ponds, Central Australia, 16 degrees 54 minutes 7 seconds
South latitude.

Remark. This is in every respect a typical Polyteles, having the delicate
bill and elegantly striped tail characteristic of that form. It is of the
same size as P. barrabandi, but differs from that species in having the
crown blue and the lower part of the cheeks rose-pink instead of yellow.

The following is a list of the other species of birds comprised in the
collection:

Trichoglossus rubritorquis. Rare.

Aprosmictus erythropterus.

Platycercus brownii. Rare.

Struthidea cinerea.

Climacteris melanura.

Pomatorhinus rubecula. Rare.

Cincloramphus cruralis.

Artamus leucopygialis.

Artamus cinereus. Rare.

Colluricincla brunnea.

Petroica bicolor.

Pardalotus rubricatus. Extremely rare: the second specimen seen.

Graucalus melanops.

Tropidorhynchus argenteiceps.

Geopelia cuneata.

Geopelia humeralis.

Erythrogonys cinctus.

...


[FROM THE PROCEEDINGS OF THE ZOOLOGICAL SOCIETY OF LONDON, NOVEMBER 10,
1863.]

DESCRIPTIONS OF NEW SPECIES OF FRESHWATER SHELLS COLLECTED BY MR. F.G.
WATERHOUSE, DURING J. McDOUALL STUART'S OVERLAND JOURNEY FROM ADELAIDE TO
THE NORTH-WEST COAST OF AUSTRALIA. BY ARTHUR ADAMS, F.L.S., AND G. FRENCH
ANGAS, CORRESPONDING MEMBER OF THE ZOOLOGICAL SOCIETY.

1. Vivipara waterhousii, Adams & Angas.

V. testa turbinata, globoso-conica, late umbilicata, spira elatiuscula,
epidermide tenui fusco-viridi obtecta; anfractibus convexis, ad suturas
subplanatis, faciis tribus vel quatuor angustis olivaceo-viridibus
transversis ornatis; anfractu ultimo inflato, lineis duabus impressis ad
peripheriam instructo; apertura ovata, postice subangulata; labio
simplici; labro acuto.

Long. 2 inches, lat. 1 inch 8 lines.

Habitat. Newcastle Waters, Arnhem's Land (Coll. Angas):

This fine species most nearly resembles Vivipara ussuriensis, Gerst.; but
the last whorl is more inflated, and the surface of the shell is not
malleated or lirate. It is the largest species yet discovered on the
Australian continent. We have great pleasure in dedicating it to F.G.
Waterhouse, Esquire, who, under great difficulties during the expedition,
succeeded in making many valuable additions to science.

2. Vivipara kingi, Adams & Angas.

V. testa turbinata, globoso-conica, umbilicata, spira mediocri erosa
nodulosa, epidermide tenui pallide fusco-viridi obtecta, ad apicem
purpurascente; anfractibus convexis, lineolis transversis et
longitudinalibus elevatis decussatis, anfractu ultimo ad basin sulcis
impressis spiralibus instructo; apertura ovata, antice subeffusa; labio
vix reflexo.

Long. 1 inch, lat. 8 lines.

Habitat. King's Ponds, Arnhem's Land (Coll. Angas).

This is a neat, finely-decussated, concolorous species, with the upper
whorls nodulous from erosion, as in Vivipara praerosa, Gerst. It is named
after Mr. Stephen King, one of the gentlemen who accompanied the
expedition.

3. Melania (Melasma) onca, Adams & Angas.

M. testa fusiformi-turrita; spira elata, conica; epidermide pallide
olivaceo induta, rufo-fusca, pulcherrime maculata, maculis saepe in
lineis undulatis longitudinalibus dispositis; anfractibus planis,
longitudinaliter plicatis, plicis aequalibus regularibus subdistantibus,
ad suturas nodulosis; apertura oblongo-ovata, antice effusa; labio
subincrassato; labro simplici, acuto.

Long. 1 inch, lat. 4 lines.

Habitat. Tributary of Adelaide River, Arnhem's Land (Coll. Angas).

A species remarkable both for the elegance of its form and the beauty of
its painting. The whorls are plicate, with a necklace-like series of
nodules at the sutures; and the shell is covered with dark red-brown
spots, suggestive of its specific name.

4. Amphipeplea vinosa, Adams & Angas.

A. testa ovata; spira mediocri, tenui, semipellucida, vinosa; anfractu
ultimo magno, ventricoso, postice ad suturas gibboso; apertura ovata;
labio callo tenui mediocri obtecto, columella spiraliter tortuosa; labro
convexo, margine acuto.

Long. 9 lines, lat. 5 lines.

Habitat. Tributary of Adelaide River, Arnhem's Land (Coll. Angas).

This species may readily be distinguished on account of its peculiar
vinous colour. The whorls are posteriorly gibbose or tumid at the
sutures, and the callus is less spreading than in others of the genus.

5. Amphipeplea phillipsi, Adams & Angas.

A. testa ovata; spira elata, acuta, tenui, cornea; anfractu ultimo magno,
non ventricoso, transversim creberrime striato; apertura oblongo-ovali;
labio callo tenui expanso obtecto; labro simplici, acuto.

Long. 9 lines, lat. 4 lines.

Habitat. Arnhem's Land (Coll. Angas).

A neat, horn-coloured, finely transversely striated species, with an
acute elevated spire. We have named it after Mr. T. Phillips, who has
assiduously collected many new Australian shells.

6. Physa newcombi, Adams & Angas.

P. testa ovata, umbilicata; spira mediocri, acuta, ad apicem integra,
cornea, viridescente aut pallide fulva; anfractibus quinque, convexis,
saepe plus minusve transversim subliratis; apertura ovata; labio reflexo,
umbilicum partim tegente; labro vix incrassato, peristomate nigrescente.

Long. 10 lines, lat. 7 lines.

Habitat. Ponds at Mount Margaret (Coll. Angas.)

We have much pleasure in naming this noble Physa after Dr. Newcomb, the
distinguished American conchologist, who has contributed so much, by his
researches in the Sandwich Islands, to our knowledge of the genus
Helicter or Achatinella. The species is widely umbilicated, and the
peristome is usually dark-coloured.

7. Physa ferruginea, Adams & Angas.

P. testa ovata, rimata, ferruginea; spira mediocri, apice eroso;
anfractibus tribus, convexis, simplicibus, transversim crebre
crenato-striatis; apertura ovata, intus purpurascente; labio tenui, late
reflexo; labro acuto.

Long. 5 lines, lat. 4 lines.

Habitat. Arnhem's Land, North-west Australia (Coll. Angas.)

This is a small ferruginous species, with the whorls finely transversely
striated.

8. Physa badia, Adams & Angas.

P. testa elongato-ovata, imperforata, solida, badia; spira elata, apice
obtuso eroso; anfractibus quinque, convexiusculis, longitudinaliter
strigillatis; apertura elongato-ovata; labio albo, excavato, lirula
antica subspirali instructo; labro arcuato, in medio producto, intus
fusco tincto.

Long. 1 inch, lat. 6 lines.

Habitat. Tributaries of Adelaide River, Arnhem's Land (Coll. Angas.)

A fine, solid, brown species, generally more or less eroded, and with a
peculiarly strongly plicate columella.

9. Physa olivacea, Adams & Angas.

P. testa elongato-ovata, imperforata, solidiuscula, olivacea; spira
elata, attenuata, apice eroso; anfractibus quinque, convexiusculis;
apertura ovato-acuta; labio incrassato, flexuoso; labro acuto, margine
arcuato.

Long. 6 lines, lat. 3 lines.

Habitat. Arnhem's Land (Coll. Angas.)

A neat, olive-coloured species, somewhat resembling in form the British
Aplexa hypnorum, but without the polished exterior of the latter.

10. Physa concinna, Adams & Angas.

P. testa ovata, imperforata, solidiuscula, cornea; spira brevi, acuta,
apice interdum papilloso; anfractibus quinque, convexiusculis;
transversim striatis; apertura acuto-ovata; labio incrassato, spiraliter
valde tortuoso; labro intus incrassato et fusco tincto, margine acuto,
arcuato.

Long. 6 lines, lat. 3 lines.

Habitat. Arnhem's Land (Coll. Angas.)

A pale horn-coloured, somewhat solid species, with a moderately elevated
spire, acute (not eroded) at the apex, and with the terminal whorls
sometimes papillary.

11. Physa (Ameria) reevii, Adams & Angas.

P. testa ovali, postice abrupte truncata, imperforata, cornea; spira
plana, tenui; anfractibus quatuor, planis, ultimo permagno, postice acute
angulato, transversim obsolete striato; apertura oblongo-truncata; labio
antice valde tortuoso; labro postice angulato.

Long. 6 lines, lat. 4 lines.

Habitat. Arnhem's Land (Coll. Angas.)

We have much pleasure in dedicating this singular species to Mr. Lovell
Reeve, who has evinced much interest in the shells of this group. The
last whorl is acutely angulate posteriorly, and the spire is tabulated,
giving to the shell a peculiar truncate appearance.

12. Physa (Ameria) bonus-henricus, Adams & Angas.

P. testa ovata, rimata, tenui, cornea; spira vix elata, plana;
anfractibus tribus, planis, postice angulatis, ultimo magno, inflato,
ventricoso, postice subangulato, longitudinaliter plus minusve plicato;
apertura ovata; labio tenui, subtortuoso; labro simplici, margine
arcuato.

Long. 4 lines, lat. 2 1/2 lines.

Habitat. Arnhem's Land (Coll. Angas.)

This is a small inflated species, with a short truncate spire. We have
dedicated it to the founder of the section Ameria, a gentleman well known
for his deep researches in conchology.

13. Unio (Alasmodon) stuarti, Adams & Angas.

U. testa transversim elongato-ovata, tenui, compressa, epidermide
olivaceo-fusca induta, postice corrugato-plicata, latere antico breviore
rotundato, postico longiore oblique subtruncato, margine ventrali
regulariter arcuato; umbonibus parvis, erosis, dentibus cardinalibus
elongatis valde divergentibus, postico bifido, antico prominulo; intus
iridescente.

Alt. 1 1/2 inch, lat. 3 inches 2 lines.

Habitat. Lagoon, Mount Margaret, Central Australia (Coll. Angas.)

This species, which we have named after Mr. J. McD. Stuart, the leader of
the expedition, is the only Naiad, besides Alasmodon angasana of Lea, yet
discovered in the regions traversed by the explorers.

...

Description of a new Helix from the interior of Australia, by Dr. L.
Pfeiffer.

Helix perinflata, Pfr.

T. umbilicata, globosa, solida, striis incrementi rugosis et lineis
impressis antrorsum descendentibus decussata, isabellino-albida; spira
convexo-conoidea, apice obtusa; anfr. 4 1/2, ultimus magnus, ventrosus,
subtus, perinflatus, striis spiralibus obsolete sculptus, antice
deflexus; apertura diagonalis, lunari-rotundata; perist. breviter
expansum margine columellari supra umbilicum angustum fornicatim
dilatato.

Diam. mag. 23 1/2, min. 20, alt. 20 mill. (Coll. Angas.)

Habitat. McDonnell Range, Central Australia. Waterhouse, on Stuart's
expedition.


ENUMERATION OF THE PLANTS COLLECTED DURING MR. J. McDOUALL STUART'S
EXPEDITIONS ACROSS THE AUSTRALIAN CONTINENT IN 1860, 1861, AND 1862. BY
FERDINAND MULLER, M.D., Ph.D., F.R.S.

Dilleniaceae.

Pachynema macrum, F.M. Purdie Ponds. Waterhouse.
Hibbertia glaberrima, F.M. Fragmenta, Phyt. Austr. iii. 1. Brinkley
Bluff, McDonnell Range. J.M. Stuart.

Nymphaeaceae.

Nymphaea gigantea. Hook. Botanical Magazine 4647. Strangways River.

Nelumbium speciosum, W. Sp. Pl. ii. 1258. Arnhem's Land.

Capparideae.

Capparis nummularia, Cand. Prodr. i. 246. Central Australia.

Capparis lasiantha, R. Br in Cand. Prodr. i. 247. Near Central Mount
Stuart.

Busbeckea umbonata (Capparis umbonata, Lindl. in Mitch. Trop. Austr.
275). Near Newcastle Waters and Attack Creek. Flowers similar to those of
B. Mitchellii.

Droseraceae.

Drosera indica, Linn. Sp. Pl. 403. On the Bonney and Finke Rivers and
Attack Creek, also in Central Australia.

Violaceae.

Ionidium enneaspermum, Vent. Malmais. page 27. Burke Creek. An allied
species with a blue labellum occurs in the collection gathered at Purdie
Ponds.

Frankeniaceae.

Frankenia laevis, Linn. Sp. 473 var. Finke River.

Zygophylleae.

Zygophyllum apiculatum, F.M. in Linnaea, 1852, page 373. Stevenson River.

Tribulus terrestris, Linn. Sp. 554. Mount Morphett. A large flowering
variety with petals 1 inch long. At Marchant Springs, Burke River, and
Attack Creek.

Malvaceae.

Hibiscus brachysiphonius, F.M. Fragm. Phyt. Austr. i. 67. Near the
Strangways Range.

Hibiscus pentaphyllus, F.M. Fragm. Phyt. Austr. ii. 13. Newcastle Waters
and Daly Waters.

Hibiscus radiatus, Cav. Diss. iii. 150, t. 54, fig. 2. Purdie Ponds,
Newcastle Waters. Attack Creek.

Hibiscus sturtii, Hook. in Mitch. Trop. Austr. page 363. North of
McDonnell Range.

Hibiscus solanifolius, F.M. Fragm. ii. 116. Mount Denison.

Hibiscus panduriformis, Burm. Fl. Ind. page 151, t. 47, f. 2. Burke
River.

Gossypium Australe, F.M. Fragm. i. 46. Newcastle Waters, Waterhouse.
Between Mount Woodcock and the Davenport Ranges.

Gossypium Sturtii, F.M. Fragm. iii. 6. as far north as the Stevenson
River.

Abutilon tubulosum, All. Cunn. in Mitch. Trop. Austr. 390. Burke River.

Abutilon leucopetalum, F.M. Fragm. iii. 12. Daly Waters.

Sida corrugata, Lindl. in Mitch. Three Exped. ii. 12. Var. filipoda.
Attack Creek. J.M. Stuart.

Sida cryphiopetala, F.M. Fragm. ii. 4. Brinkley Bluff, McDonnell Range.
J.M. Stuart.

Tiliaceae.

Corchorus sidoides, F.M. Fragm. iii. 9. McDonnell Range. J.M. Stuart.

Triumfetta plumigera, F.M. Fragm. i. 69. Purdie Ponds. F. Waterhouse.

Buettneriaceae.

Kerandrenia nephrosperma, Benth. in Proceedings of the Linnean Society;
Seringea nephrosperma, F.M. in Hook. Kew Miscell. 1857, 15. Towards
Arnhem's Land.

Kerandrenia Hookeri, Walp. Annal. Bot. Syst. ii. 164. Near the Roper
River.

Rulingia loxophylla, F.M. Fragm. i. 68. Towards Arnhem's Land.

Melhania incana, Heyne in Wall. List. 1200. Burke River and Purdie Ponds.

Sterculiaceae.

Brachychiton ramiflorum, R. Br. in Horsf. Plant. Savan. rarior. 234. From
Burke Creek onward to Arnhem's Land.

Cochlospermeae.

Cochlospermum Gregorii, F.M. Fragm. Phyt. Austr. i. 71. Strangways River.

Cochlospermum heteronemum, F.M. in Hook. Kew Miscell. 1857, 15.
Strangways River.

Meliaceae.

Owenia acidula, F.M. in Hook. Kew Miscell. ix. 304. Central Mount Stuart.

Sapindaceae.

Thouinia variifolia, Fragm. Phyt. Austr. i. 45. Crawford Range.

Diplopeltis Stuartii, F.M. Fragm. iii. 12. Between Mount Morphett and the
Bonney River. J.M. Stuart.

Distichostemon phyllopterus, F.M. in Hook. Kew Miscell. ix. 306. Purdie
Ponds. Var. serrulatus; leaves tender, lanceolate, acute, serrulated;
stamens about 44. Burke River.

Dodonaea lanceolata, F.M. Fragm. Phyt. Austr. i. 73. Purdie Ponds,
Waterhouse. Mount Woodcock. Stuart.

Dodonaea platyptera, F.M. Fragm. i. 73. Strangways River.

Dodonaea physocarpa, F.M. Fragm. i. 74. Daly Waters.

Dodonaea microzyga, F.M. Somewhat viscid, almost glabrous; leaves with 1
to 2 pairs of small obovate-cuneate leaflets; in front rounded, or
truncate, or retuse, or sometimes 3-toothed, flat at the margin; rachis
dilated; fruit-bearing pedicels solitary; capsules 3 to 4-celled; valves
cymbeo-semiorbicular, all around broadly winged; the wing rounded-blunt
on both extremities; dissepiments persistent with the columella. On the
River Neale. J.M. Stuart.

A shrub with spreading and rigid branches. Most leaves about 1/2 an inch
long; leaflets 1 to 2 inches long; flowers unknown; capsule with the
wings added about 1/2 an inch long, shining, reddish; valves ceding from
the septa; ripe seeds unknown.

The fruit of this species is almost like that of Dodonaea viscosa.

Mollugineae.

Mollugo trigastrotheca, F.M. Plants indigenous to Victoria, i. 201.
Arnhem's Land.

Caryophylleae.

Polycarpoea corymbosa, Lam. Mount Samuel. J.M. Stuart.

Portulaceae.

Portulaca oleracea, Linn. Sp. Pl. 638. Common in the interior and in
North Australia.

Calandrinia Balonnensis, Lindl. in Mitch. Trop. Austr. page 148. River
Finke.

Phytolacceae.

Codonocarpus cotinifolius, F.M. Plants of Victoria, i. 200. From 300 to
800 miles north of Adelaide, F. Waterhouse; Central Mount Stuart, J.M.
Stuart.

Gyrostemon ramulosus, Desf. in Memoir. du Museum, vi. 17 River Finke.
J.M. Stuart.

Didymotheca pleiococca, F.M. Plants indigenous to Victoria, i. 198.
Between the River Bonney and Mount Morphett. J.M. Stuart.

Leguminosae.

Acacia retivenea, F.M. Fragm. iii. 128. Short Range.

Acacia dictyophleba, F.M. Fragm. iii. 128. Mount Humphries.

Acacia aneura, F.M. in Linnaea, xxvi. 627. Mulga. Over the whole of
Central Australia. F. Waterhouse.

Acacia tumida, F.M. in Proceedings of the Linnean Society iii. 144.
Attack Creek.

Acacia impressa, F.M. in Proceedings of the Linnean Society iii. 133.
Short Range.

Acacia lycopodifolia, A. Cunn. in Hook. Icon. ii. t. 172. Towards
Arnhem's Land.

Acacia umbellata, A. Cunn. in Hook. London Journal of Botany i. 378.
Robinson River. Stuart.

Acacia holosericea, A. Cunn. in Don. Gen. Syst. ii. 407. Near Newcastle
Waters.

Pithecolobium moniliferum, Benth. in Hook. Journal of Botany iii. 211.
Arnhem's Land.

Neptunia spicata, F.M. Fragm. Phyt. Austr. iii. 151. Arnhem's Land.

Erythrophloeum Laboucherii, Laboucheria chlorostachya, F.M. in
Proceedings of the Linnean Society  iii. 159. Newcastle Waters, Stuart;
Strangways River, Waterhouse.

Cassia venusta, F.M. Fragm. Phyt. Austr. i. 165. Newcastle Waters and
Mount Freeling. J.M. Stuart.

Cassia notabilis, F.M. Fragm. ii. 28. Between the River Bonney and Mount
Morphett.

Cassia Absus, Linn. Spec. Plant. 537. Arnhem's Land.

Cassia oligoclada, F.M. Fragm. iii. 49. Attack Creek.

Cassia desolata, F.M. in Linnaea, 1852. Central Australia.

Cassia eremophila, A. Cunn. in Sturt's Centr. Austr. Append. ii. 77.
Central Australia.

Petalogyne labicheoides, F.M. in Hook. Kew Miscell. 1856. From latitude
30 degrees South to latitude 17 degrees 58 minutes South. J.M. Stuart.
Petalogyne cassioides forms merely a variety of this species.

Erythrina biloba, F.M. in Hook. Kew Miscell. 1857, page 21. Common to
most creeks, from latitude 22 degrees to 19 degrees South. Wood soft,
corky. J.M. Stuart. Stuart's Bean-tree is a species of Erythrina.

Bauhinia Leichartdtii, F.M. in Transact. Phil. Inst. Vict. iii. 50.
Hayward Creek. J.M. Stuart.

Gastrolobium grandiflorum, F.M. Fragm. Phyt. Austr. ii. 17. Whittington
Range, J.M. Stuart; Purdie Ponds, where it attains a height of 8 feet,
Waterhouse.

Gompholobium polyzygum, F.M. Fragm. ii. 29. Between Mount Morphett and
the Bonney River.

Jacksonia odontoclada, F.M. Between Newcastle Water and Attack Creek.
J.M. Stuart.

Isotropis atropurpurea, F.M. Fragm. Phyt. Austr. ii. 16. Attack Creek,
and between Mount Morphett and the Bonney River. J.M. Stuart.

Leptosema Chambersii, F.M. Essay on the Plants of the Burdekin Expedition
page 8. Near Davenport Range, and between the Rivers Finke and Stevenson.

Crotalaria medicaginea, Lamb. Dict. ii. 201. Newcastle Waters. J.M.
Stuart.

Crotolaria dissitiflora, Benth. in Mitch. Trop. Austr. 386. Newcastle
Waters and McDonnell Range. Stuart.

Crotalaria Mitchellii, Benth. l. c. 120. Central Australia.

Crotalaria Cunninghami, R. Br. in Sturt's Central Austr. Append. 71.
Burke Creek, Waterhouse; Mount Humphries, Stuart.

Indigofera hirsuta, L. Sp. Pl. 1862. Arnhem's Land.

Indigofera viscosa, Lam. Encyl. Menth. iii. 247. Brinkley Bluff. Stuart.

Indigofera oxycarpa, F.M. Fragm. Phyt. Austr. iii. 103. Burke Creek.
Waterhouse.

Indigofera brevidens, Benth. in Mitch. Trop. Austr. 385. Central
Australia.

Indigofera lasiantha, F.M. Report on Gregory's Plants from Cooper Creek,
page 6. Denison Range. J.M. Stuart.

Swainsona phacoides, Benth. in Mitch. Trop. Austr. 363. River Neale.
Stuart.

Swainsona campylantha, F.M. Report on Gregory's Plants from Cooper Creek.
Bagot Range. J.M. Stuart.

Psoralea patens, Lindl. in Mitch. Three Exped. ii. 8. Attack Creek. Var.
cinerea. Mount Kingston.

Psoralea balsamica, F.M. in Proceed. Phil. Inst. Vict. iii. 55. Attack
Creek and McDonnell Range. J.M. Stuart.

Psoralea leucantha, F.M. l. c. iii. 54. Attack Creek.

Clianthus Dampierii, All. Cunn. in Transact. Horticult. Soc. ii. Ser.
Vol. i. 522. Near Mount Humphries.

Onagreae.

Jussioea suffruticosa, Linn. Sp. Pl. 555. Attack Creek and Strangways
River.

Rhamnaceae.

Alphitonia excelsa, Reiss. in Endl. Gen. Plant. page 1098. Daly Waters.

Euphorbiaceae.

Euphorbia hypericifolia, Linn. Sp. Plant. Attack Creek.

Flueggea leucopyris, W. Sp. Plant. McDouall Range and Roper River.

Petalostigma quadriloculare, F.M. in Hook. Kew Miscell. ix. 17. Near
Mount Blyth.

Combretaceae.

Macropteranthes Kekwickii, F.M. Fragm. iii. 151. Newcastle Waters, near
Ashburton Range.

Terminalia circumalata, F.M. Fragm. Phytogr. Austr. ii. 91. Attack Creek.

Terminalia bursarina, F.M. Fragm. Phytogr. Austr. ii. 149. Newcastle
Waters.

Rhizophoreae.

Carallia integerrima, Cand. Prodr. iii. 33. Roper River. Waterhouse.

Cucurbitaceae.

Cucumis jucunda, F.M. in Transact. Phil. Inst. Vict. iii. 45. Central
Australia.

Melastomaceae.

Osbeckia Australiana, Naudin in Annal. des Scien. Naturell. Ser. iii.
xiv. 59. Arnhem's Land.

Melastoma Novae Hollandiae, Nand. l. c. xiii. 290. Adelaide River.

Myrtaceae.

Carega arborea, Roxb. Coromand. iii. t. 218. Billiatt Springs.
Waterhouse.

Melaleuca leucadendron. L. Mant. 105. Attack Creek. Roper River.

Melaleuca dissitiflora, F.M. Fragm. iii. 153. Between the Bonney River
and Mount Morphett.

Eucalyptus setosa, Schauer in Walp. Report, ii. 926. Sandy Scrub near the
River Bonney.

Calycothrix microphylla, All. Cunn. in Botanical Magazine 3323. Sources
of the River Roper.

Boeckea polystemonea, F.M. Fragm. Phyt. Austr. ii. 124. Brinkley Bluff,
McDonnell Range.

Umbelliferae.

Didiscus glaucifolius, F.M. in Linnaea, 1852, page 395. Var.
cyanopetalus. Finke River. J.M. Stuart. The colour of the petals varies
likewise blue and white in Didiscus coeruleus and in one species of
Dimetopia.

Rubiaceae.

Canthium oleifolium, Hook. in Mitch. Trop. Austr. 397. Var. latifolium.
Central Australia, in Mulga Scrub. J.M. Stuart.

Compositae.

Calotis Waterhousii. F.M. Purdie Ponds. Waterhouse.

Eurybia Ferresii, F.M. Fragm. Phyt. Austr. iii. 18. t. xviii. Brinkley
Bluff. J.M. Stuart.

Pluchea ligulata, F.M. Enumeration of Plants of Babbage's Expedition page
12. Strangways River. Waterhouse.

Monenteles globifer, Cand. Prodr. v. 455. McDonnell Range, Stuart. Attack
Creek, Waterhouse.

Helichrysum Davenportii, F.M. Fragm. Phyt. Austr. iii. 32. (Sect.
Acroclinium.) On the River Neale.

Helichrysum Cassianum, Gaudichaud Voyage Freycenet. page 466, t. 87.
(Sect. Pteropogon.) River Finke. J.M. Stuart. The capitula are rather
smaller than those figured by Gaudichaud; but in Mr. Oldfield's
collection from the Murchison River we observe analogous specimens, with
intermediate gradations. The involucre-scales are sometimes delicately
rose-coloured.

Senecio Gregorii, F.M. Report on Gregory's Plants from Cooper Creek, page
7. Finke River. J.M. Stuart.

Goodeniaceae.

Goodenia grandiflora, Sims, Botanical Magazine 890. Mount Freeling.
Stuart.

Goodenia hirsuta, F.M. Fragm. iii. 35. Central Australia.

Goodenia heterochila, F.M. Fragm. iii. 142. Newcastle Water.

Goodenia Vilmoriniae, F.M. Fragm. Phyt. Austr. iii. 19. Between the River
Bonney and Mount Morphett. Stuart.

Goodenia Ramelii, F.M. Fragm. iii. 20. Attack Creek. Stuart.

Vellega connata, F.M. Transactions of the Phil. Soc. i. 18. Between the
River Bonney and Mount Morphett. Stuart.

Scaevola microcarpa, Cavan. Icon. vi. 6, t. 509. Towards Central
Australia.

Lobeliaceae.

Isotoma petroea, F.M. in Linnaea, 1852, page 420. James Range and Hugh
River.

Asclepiadeae.

Leichardtia Australis, R. Br. in Sturt's Central Australia. ii. Append.
page 81. Daly Water.

Apocyneae.

Carissa lanceolata, R. Br. Prodr. 468. Strangways River.

Acanthaceae.

Dipterancanthus Australasicus, F.M. Report on Gregory's Plants from
Cooper Creek, page 8. Near Anna Reservoir.

Rostellularia procumbens, Nees in Wall. Plant. Asiat. rarior. iii. 101.
Purdie Ponds.

Solaneae.

Solanum pulchellum, F.M. Transact. Phil. Soc. Vict. i. 18. Purdie Ponds.

Soluanum chenopodinum, F.M. Fragm. ii. 165. On Stuart Creek, and between
Mount Blyth and Mount Fisher. Stuart.

Scrophularineae.

Buchnera linearis, R. Br. Prodr. 437. King's Ponds.

Vandellia plantaginea, F.M. in Trans. Vict. Inst. iii. 62. Arnhem's Land.

Morgania floribunda, Benth. in Mitch. Trop. Austr. Var. glandulosa.
Central Australia.

Rhamphicarpa adenophora, F.M. Near Attack Creek.

Bignoniaceae.

Spathodea heterophylla, R. Br. Prodr. 470. King's Chain of Ponds.

Tecoma Australis, R. Br. Prodr. 471. Var. angustifolia. McDonnell Range,
and distributed over a wide range of latitude in the interior, according
to Mr. Stuart. Tecoma Oxleyi, Tecoma floribunda, and Tecoma diversifolia
are mere varieties of Tecoma Australis.

Asperifoliae.

Halgania solanacea, F.M. in Hook. Kew Miscell. 1857. page 21. Between
Bonney River and Mount Morphett.

Halgania strigosa, Schlecht. Linnaea, xx. 640. Brinkley Bluff.

Trichodesma Zeilanicum, R. Br. Prodr. 496. Newcastle Water.

Labiatae.

Prostanthera striatiflora, F.M. in Linn. 1852, page 376. Mount Morphett.

Convolvulaceae.

Evolvulus linifolius, Linn. Sp. Pl. 392. Brinkley Bluff.

Ipomoea reptans, Poir. Encycl. Suppl. iii. 460. A white-flowering
variety. Purdie Ponds.

Ipomoea pannosa, R. Br. Prodr. 487. Newcastle Water, Attack Creek, and
Strangways River.

Jasminiae.

Jasminum calcarium, F.M. Fragm. i. 212. Common to most creeks of the
interior. Stuart. The lobes of the calyx are narrower than in the
specimens from the Murchison River; the lobes of the corolla likewise
narrower, and occasionally augmented to nine. The leaflets sometimes
ovate. Transient forms are sent from Champion Bay by Mr. Walcott.

Myoporinae.

Avicennia officinalis, L. Sp. Pl. page 110. Var. angustifolia. Daly
Water.

Eremophila Goodwinii, F.M. Report on Babb. Plants, page 17. Mount
Freeling, Attack Creek, and Mount Samuel. Stuart. Var. angustifolia;
leaves linear; calyx and pedicel glabrous; corolla outside glabrous or
scantily hairy. Marchant Springs.

Eremophila Macdonellii, F.M. Report on Babb. Plants, page 18. Var.
glabra. Valley of the Elizabeth River.

Eremophila Latrobei, F.M. in Papers of Royal Society of Tasmania 1858.
Arnhem's Land, and near Anna Reservoir. J.M. Stuart.

Eremophila Brownii, F.M. in Papers of Royal Society of Tasmania 1858.
McDonnell Range. Stuart.

Eremophila Willsii, F.M. Fragm. Phyt. Austr. ii. 21, t. xx. River Finke.
Stuart.

Eremophila Sturtii, R. Br. in Sturt's Central Austr. App. page 85. Daly
Water.

Eremophila longifolia, F.M. in Papers of Royal Society of Tasmania 1858.
Strangways Range, Stuart; Billiatt Springs, Waterhouse.

Eremophila maculata, F.M. in Papers of Royal Society of Tasmania 1858.
Sandy scrub country from the south through Central Australia to Attack
Creek. Waterhouse.

Verbenaceae.

Clerodendron cardiophyllum, F.M. Fragm. iii. 144. Mulga Scrub, Stuart;
Daly Water, Waterhouse.

Newcastlia spodiotricha, F.M. Fragm. Phyt. Austr. iii. 21. Between the
Victoria River and the Gulf of Carpenteria, from 17 to 19 degrees South
latitude.

Lentibulariae.

Utricularia fulva, F.M. in Trans. Phil. Inst. iii. 63 Strangways River.

Laurineae.

Gyrocarpus sphenopterus, R. Br. Prodr. page 405. Short Range.

Thymeleae.

Pimelea sanguinea. F.M. Fragm. Phyt. Austr. i. 84. Purdie Ponds.

Proteaceae.

Grevillea mimosoides, R. Br. Prodr. page 380. Roper River.

Grevillea agrifolia, All. Cunn. in R. Br. Suppl. page 24. McDonnell
Range, Short Range. Var. lancifolia. Central Australia.

Grevillea Sturtii, R. Br. in Sturt's Centr. Austr. Append. page 24.
Central Mount Stuart. Var. pinnatisecta; segments usually five. Scrub
near Forster Range. J.M. Stuart.

Grevillea lineata, R. Br. in Sturt's Centr. Austr. Append. page 24. Scrub
near Forster Range.

Grevillea chrysodendron, R. Br. 379. Billiatt Springs. Waterhouse.

Grevillea refracta, R. Br. Prodr. page 380. Newcastle Water, Billiatt
Springs, and Short Range.

Grevillea dimidiata, F.M. Fragm. Phyt. Austr. iii. 146. Roper River.
Waterhouse.

Hakea arborescens, R. Br. Prodr. page 386. Arnhem's Land.

Hakea lorea, R. Br. Suppl. page 25. Central Australia. Bark corky.

Amaranthaceae.

Alternanthera denticulata, R. Br. Prodr. 417. Burke River.

Alternanthera nana, R. Br. Prodr. 417. Burke River.

Gomphrena humilis, R. Br. Prodr. 416. Attack Creek. The upper pair of
leaves stand either next to the flower-heads or remote from them. The
same species has been found by Dr. Muller on the Dawson River, and by Mr.
Fitzalan at Port Denison.

Gomphrena canescens, R. Br. Prodr. 416. Attack Creek. J.M. Stuart.
(Victoria River and Sturt Creek, F. Muller; Sweer's Island, Henne; Nickol
Bay, Walcot.) Capsula usually beautifully pink, sometimes purple or
white. Peduncles occasionally more than 6 inches long; the staminodia
sometimes excel the anthers in length.

Ptilotus corymbosus, R. Br. Prodr. 415. Var. spicatus. Attack Creek.

Trichinium gracile, R. Br. 415. Tropical Australia.

Trichinium nobile, Lindl. in Mitch. Three Exped. ii. 22. Short Range.

Trichinium brachytrichum, F.M. Fragm. iii. 157. Central Australia. J.M.
Stuart.

Urticeae.

Ficus Stuartii, F.M. McDonnell Range; Brinkley Bluff. Several other
undescribed species of fig-trees occur in the collection, but cannot be
satisfactorily characterised from the material extant.

Cycadeae.

A cycadeous plant, seemingly distinct from the seven Australian species,
occurs on McDonnell Range, and is mentioned as a palm in the Journal of
the explorers. Only leaves being now submitted for examination, it
remains for future researches to throw light on this plant.

Amaryllideae.

Calostemma luteum, Sims, in Botanical Magazine 2101. Mount Margaret.
Stuart. The edge of the corona is sometimes rather undulated than
toothed.

Crinum angustifolium, R. Br. 297. From latitude 22 to 32 degrees South.
J.M. Stuart.

Orchideae.

Cymbidium canaliculatum, R. Br. Prodr. 331. Strangways River.

Commelyneae.

Commelyna ensifolia, R. Br. Prodr. 269. McDonnell Range, and near Mount
Freeling. J.M. Stuart.

Commelyna agrostophylla, F.M. Arnhem's Land.

Liliaceae.

Bulbine semibarbata, Haw. Revis. 33. Thring River. Stuart.

Gramineae.

Eriachne obtusa, R. Br. Prodr. 184. Short Range.

Ectrosia leporina, R. Br. Prodr. 186. Purdie Ponds.

Perotis rara, R. Br. Prodr. 172. Purdie Ponds, Waterhouse; Short Range,
Stuart.

Andropogon bombycinus, R. Br. Prodr. 202. Central Australia, McDonnell
Range.

Chloris ventricosa, R. Br. Prodr. 186. Arnhem's Land.

Lappago racemosa, W. Sp. l. 484. Attack Creek.

Panicum decompositum, R. Br. Prodr. 191. Stevenson River.

Oryza sativa, L. Sp. Pl. Newcastle Water. J.M. Stuart.

Pappophorum commune, F.M. Enumeration of Greg. Plants from Cooper Creek,
page 10. Central Australia.

Cyperaceae.

Hypaelyptum microcephalum, R. Br. Prodr. 221. Attack Creek.

Filices.

Marsilia quadrifolia, L. Sp. Pl. Var. hirsuta. Nardoo. Through Central
and North Australia, on localities subject to inundation.

Lygodium semibipinnatum, R. Br. Prodr. 162. Roper River.

Blechnum Orientale, L. Sp. Pl. 1535. River Adelaide. This fern was not
previously recorded as existing in Australia.

Cheilanthes tenuifolia, Swartz Filic. 129. River Roper, Mount Freeling.


INDEX.

A'Beckett's Pool.

Accident:
to instruments.
to plans.
to Mr. Stuart.
to Kekwick.

Adelaide River:
Valley of the.

Alligator's skull found.

Anderson Creek.

Andamoka.

Ann Creek.

Anna:
Creek.
Mount.
Reservoir.

Ant-hills.

Ants.

Arden, Mount.

Arthur, Mount.

Ascent of:
Brinkley Bluff.
Mount Denison.
Mount Strzelecki.
Mount Primrose.
Mount Stuart.

Ashburton Range.

Attack Creek.

Attraction, Mount.

Auld's Chain of Ponds.

Auriferous appearances.

Babbage, Mr.

Bagot Range.

Baker Creek.

Bamboo.

Barker's party, Mr.

Barker Creek.

Barkley Mount.

Barrow:
Creek.
Springs.

Bectimah Gaip.

Beddome, Mount.

Beda.

Ben:
Mount.
Illness of.
Faithfulness of.

Bennett Springs.

Beresford Springs.

Billiatt Springs.

Birds.

Bishop Creek.

Blue-Grass Swamp.

Blyth:
The.
Mount.

Bonney Creek.

Bottle:
Hill.
Mr. Stuart leaves one.

Brinkley Bluff.

British Flag planted.

Brodie, Mr.

Browne, Mount.

Burial in trees.

Burke Creek.

Cabbage Palms.

Carruthers Creek.

Centre of Australia.

Central Mount Stuart.

Chambers:
Bay.
Creek.
Pillar.
River.

Charles:
Mount.
Creek.

Cockatoos.

Coffee, A substitute for.

Coglin, The.

Comet, A.

Cooper Creek.

Copper, Indications of.

Cork-tree.

Coulthard.

Coward Springs.

Crawford Range.

Cucumber:
a cure for scurvy.
boiled.

Daly:
Water.
Range.

Daniel, Mount.

Davenport:
Creek.
Range.

Deception, Mount.

Decoy Hill.

Delusion, Mount.

Denison, Mount.

Dingo, The.

Douglas, The.

Dutton, Mount.

Eclipse of sun.

Elizabeth:
The.
Springs.

Ellen Creek.

Emerald Springs.

Esther, Mount.

Expedition, Victorian.

Eyre Lake.

Fan-palm.

Fanny:
The.
Springs.

Ferguson's, Mr.:
Station.
Creek.

Fern, A New.

Figg, Mount.

Finke:
River.
Mount.

Finniss Springs.

Fish:
shooting.

Fisher Creek.

Flowers, New.

Forster Range.

Fowler Bay.

Francis Ponds.

Freeling:
Mount.
Springs.

Freemasonry among the natives.

Frew:
The.
Water-hole.

Frost.

Gairdner Lake.

George Creek.

Gibson's Station, Mr.

Gilbert Creek.

Gleeson Creek.

Glen's Station, Mr.

Goodiar Creek.

Goolong Springs.

Goose, A peculiar.

Goyder, Mount.

Grape, Wild.

Gregory:
Mr.
Creek.

Gwynne, Mount.

Hall, Mount.

Hamilton:
Mount.
Springs.

Hanson:
Creek.
Bluff.
Range.

Harvey, Mount.

Harris, Mount.

Hawker:
Creek.
Mount.
Springs.

Hay, Mount.

Hayward Creek.

Head's Range.

Helpman:
Mount.
Lieutenant.

Hergott Springs.

Hope, The Spring of.

Hostile natives.

Hot wind.

Howell Ponds.

Hugh:
Mount.
The.

Humphries, Mount.

Hunter Creek.

James Range.

Jarvis' Station.

Ice.

Illness of:
Mr. Stuart.
Kekwick.
Waterhouse.

Incantation scene.

Indiarubber-tree.

Insubordination.

Inundation, An.

John Range.

Iron tomahawk.

Kangaroo mice.

Katherine, The.

Kekwick:
Good conduct of.
Accident to.
Illness of.
Ponds.
Springs.
Large Group of Springs.

King's Chain of Ponds.

Kingston, Mount.

Lawson Creek.

Leichardt, Mount.

Levi's:
Boundary.
Station.

Levi, Mount.

Lily:
A new.
Marsh.

Lindsay Creek.

Louden Spa.

Loveday Creek.

Macaw.

Malay type of natives.

Mann, Mount.

Marchant Springs.

Margaret:
The.
Mount.

Mary, The.

Masters lost and found.

McDouall Range.

McEllister Springs.

McGorrerey Ponds.

McKinlay Creek.

McLaren Creek.

Miller's Water.

Milne Springs.

Morphett:
Creek.
Mount.

Mudleealpa.

Muller, Mount.

Murchison Range.

Mussel Camp.

Nash Spring.

Natives:
camping-place.
fondness for fishhooks.
Freemasons.

Native:
villages.
stories in Adelaide about.
weapons.
wells.

Neale:
The.
River, West.

Newcastle Water.

Nuts, Poisonous.

O'Halloran, Mount.

Opossum, Forster catches an.

Oratunga.

Owen Springs.

Palm-tree, A remarkable.

Parrots.

Parla.

Parry Springs.

Pascoe Springs.

Pasley Ponds.

Peake:
The.
Mount.

Pernatta.

Phillips Creek.

Pigeons.

Pigfaces.

Planting the flag.

Poisonous plants.

Porter Hill.

Powell Creek.

Polly:
Lameness of.
Madness of.
has a foal.
Springs.

Primrose:
Mount.
Springs.

Priscilla Creek.

Purdie Ponds.

Rennie, Mount.

Reynolds Range.

Roper River.

Rose, A new.

Ross, The.

Samuel, Mount.

Santo, Mount.

Scurvy.

Sea, First view of the.

Separation Camp.

Shells.

Shillinglaw, Mount.

Short Range.

Smith, the deserter.

Stevenson:
The.
Mount.

Stirling, The.

Stow, Mount.

Strangways:
Mount.
Range.
River.
Springs.

Streaky Bay.

Strzelecki, Mount.

Stuart, Central Mount.

Sturt:
Captain.
Mount.
Plains.
Range.

Sullivan:
lost for three days.
Creek.

Taylor, The.

Tennant Creek.

Thring Creek.

Tide Creek, Adelaide River.

Tomahawk:
Iron.
Stone.

Tomkinson Creek.

Torrens Lake.

Traces of:
a native battle.
natives.

Tree:
New.
Palm.
India-rubber.
Cork.
Orange.

Trees marked.

Turkeys.

Turtle, A live.

Vegetable, Useful.

Victoria Expedition.

Wallaby.

Warburton, Major.

Warwick Range.

Waterbags, Failure of.

Waterfall, A beautiful.

Waterfowl.

Waterhouse:
Illness of.
Range.
River.

Waterlily, A new.

Watson Creek.

Watts, Mount.

Wealaroo.

Whittington Range.

Wicksteed Creek.

William:
Creek.
Springs.

Wind, A hot.

Wingilpin.

Woodcock, Mount.

Woodforde's:
encounter with natives.
Creek.

Yarraout.

Yarra Wirta.

Yolticowrie.

Younghusband, Mount.

THE END.





*** End of this LibraryBlog Digital Book "Explorations in Australia - The Journals of John McDouall Stuart" ***

Copyright 2023 LibraryBlog. All rights reserved.



Home