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: The Cultivation of The Native Grape, and Manufacture of American Wines
Author: Husmann, George
Language: English
As this book started as an ASCII text book there are no pictures available.
Copyright Status: Not copyrighted in the United States. If you live elsewhere check the laws of your country before downloading this ebook. See comments about copyright issues at end of book.

*** Start of this Doctrine Publishing Corporation Digital Book "The Cultivation of The Native Grape, and Manufacture of American Wines" ***

This book is indexed by ISYS Web Indexing system to allow the reader find any word or number within the document.

in Agriculture (CHLA), Cornell University)







Art, Architectural and Rural Books,

ORANGE JUDD CO., 245 Broadway.

Entered according to Act of Congress, in the year 1866,
by GEO. E. & F. W. WOODWARD,
In the Clerk's Office of the District Court of the United States,
for the Southern District of New York.




INTRODUCTION                                                        9


Remarks on its History in America, especially at the West;
its Progress and its Future,                                       13


  I.--From Seed                                                    27

 II.--By Single Eyes                                               30
      The Propagating House                                        31
      Mode of Operating                                            32

III.--By Cuttings in Open Air                                      37

 IV.--By Layering                                                  39

  V.--By Grafting                                                  40


Location and Soil                                                  43
Preparing the Soil                                                 45


Choice of Varieties                                                47
The Concord                                                        48
Norton's Virginia                                                  48
Herbemont                                                          49
Delaware                                                           49
Hartford Prolific                                                  49
Clinton                                                            50


Planting.                                                          51
Treatment of the Vine the First Summer                             56
Treatment of the Vine the Second Summer                            57
Treatment of the Vine the Third Summer                             63
Treatment of the Vine the Fourth Summer                            69
Training the Vines on Arbors and Walls                             71
Other Methods of Training the Vine                                 75
Diseases of the Vine                                               78
Insects Injurious to the Grape                                     80
Birds                                                              84
Frosts                                                             85
Girdling the Vine to Hasten Maturity                               86
Manuring the Vine                                                  91
Thinning of the Fruit                                              91
Renewing Old Vines                                                 92
Pruning Saws                                                       93
Preserving the Fruit                                               95
Gathering the Fruit to Make Wine                                   96


Concord (Description)                                              97
Concord (Plate)                                                   111
Norton's Virginia (Description)                                    98
Norton's Virginia (Plate)                                          87
Herbemont (Plate)                                                  99
Herbemont (Description)                                           101
Hartford Prolific (Description)                                   101
Hartford Prolific (Plate)                                         105
Clinton                                                           102
Delaware (Description)                                            102
Delaware (Plate)                                                   81


Cynthiana                                                         103
Arkansas                                                          104
Taylor                                                            104
Martha                                                            107
Maxatawney (Description)                                          107
Maxatawney (Plate)                                                177
Rogers' Hybrid, No. 1                                             107
Creveling (Description)                                           108
Creveling (Plate)                                                 117
North Carolina Seedling                                           108
Cunningham                                                        109
Rulander                                                          109
Louisiana                                                         110
Alvey                                                             110
Cassady                                                           110
Blood's Black                                                     113
Union Village (Description)                                       113
Union Village (Plate)                                             167
Perkins                                                           113
Clara (Description)                                               114
Clara (Plate)                                                     127
Ive's Seedling                                                    114


Minor Seedling                                                    116
Mary Ann                                                          119
Northern Muscadine                                                119
Logan                                                             119
Brown                                                             119
Hyde's Eliza                                                      119
Marion Port                                                       120
Poeschel's Mammoth                                                120
Cape                                                              120
Dracut Amber                                                      120
Elsinburgh                                                        120
Garber's Albino                                                   121
Franklin                                                          121
Lenoir                                                            121
North America                                                     121


Catawba                                                           121
Diana                                                             122
Isabella                                                          122
Garrigues                                                         123
Tokalon                                                           123
Anna                                                              123
Allen's Hybrid                                                    123
Cuyahoga                                                          123
Devereux                                                          124
Kingsessing                                                       124
Rogers' Hybrid, No. 15                                            124


Oporto                                                            124
Massachusetts White                                               125


Gathering the Grapes                                              131
The Wine Cellar                                                   133
Apparatus for Wine Making.--The Grape Mill and Press              136
Fermenting Vats                                                   137
The Wine Casks                                                    138
Making the Wine                                                   140
After Treatment of the Wine                                       146
Diseases of the Wine and their Remedies                           147
Treatment of flat and Turbid Wine                                 147
Use of the Husks and Lees                                         148
Dr. Gall's and Petoil's Method of Wine Making                     148
The Must Scale or Saccharometer                                   150
The Acidimeter and Its Use                                        151
The Change of the Must, by Fermentation, into Wine                157
Normal Must                                                       161
The Must of American Grapes                                       162
Wine Making Made Easy                                             173


Cost of Establishing A Vineyard                                   179
Cost of an acre of Concord                                        179
Cost of an acre of Herbemont                                      179
Cost of an acre of Norton's Virginia                              180
Cost of an acre of Delaware                                       180
Cost of an acre of Catawba                                        180
Product                                                           181
Produce Fifth Year                                                182
Yield of Mr. MICHAEL POESCHEL'S Vineyard                          184
New Vineyard of Mr. M. POESCHEL, Planted in 1861; First Partial
  Crop, 1863; Second Crop, 1864; Third Crop, 1865,           184, 185
Yield of Vineyard of Mr. WILLIAM POESCHEL, 1857, 1858,
  1859, 1860                                                      185
Yield of Vineyard of Mr. WILLIAM POESCHEL 1861, 1862,
  1863, 1864                                                      186
Yield of Vineyard of Mr. WILLIAM POESCHEL 1865                    187
Yield of Delaware Vineyard of JOHN E. MOTTIER                     189


It is with a great deal of hesitation I undertake to write a book about
Grapes, a subject which has been, and still is, elucidated every day;
and about which we have already several works, which no doubt are more
learned, more elaborate, than anything I may produce. But the subject
is of such vast importance, and the area suitable for grape culture so
large, the diversity of soil and climate so great, that I may be
pardoned if I still think that I could be of some use to the beginner;
it is for them, and not for my brethren of the craft more learned than
I am, that I write. If they can learn anything from the plain talk of a
practical worker, to help them along in the good work, I am well

Another object I have in view is to make grape growing as easy as
possible; and I may be pardoned if I say that, in my opinion, it is a
defect in all books we have on grape culture, that the manner of
preparing the soil, training, etc., are on too costly a plan to be
followed by men of little means. If we are first to trench and prepare
the soil, at a cost of about $300 per acre, and then pay $200 more for
trellis, labor, etc., the poor man, he who must work for a living, can
not afford to raise grapes. And yet it is from the ranks of these
sturdy sons of toil that I would gain my recruits for that peaceful
army whose sword is the pruning-hook; it is from their honest,
hard-working hands I expect the grandest results. He who has already
wealth enough at command can of course afford to raise grapes with
bone-dust, ashes, and all the fertilizers. He can walk around and give
his orders, making grape culture an elegant pastime for his leisure
hours, as well as a source of profit. But, being one of the first class
myself, I had to fight my way up through untold difficulties from the
lowest round of the ladder; had to gain what knowledge I possess from
dear experience, and can therefore sympathize with those who must
commence without means. It is my earnest desire to save _them_ some of
the losses which _I_ had to suffer, to lighten their toil by a little
plain advice. If I can succeed in this, my object is accomplished.

In nearly all our books on grape culture I notice another defect,
especially in those published in the East; it is, that they contain a
great deal of good advice about grape culture, but very little about
wine-making, and the treatment of wine in the cellar. For us here at
the West this is an all-important point, and even our Eastern friends,
if they continue to plant grapes at the rate they have done for the
last few years, will soon glut the market, and will be forced to make
them into wine. I shall therefore try to give such simple instructions
about wine-making and its management as will enable every one to make a
good saleable and drinkable wine, better than nine-tenths of the
foreign wines, which are now sold at two to three dollars per bottle. I
firmly believe that this continent is destined to be the greatest
wine-producing country in the world; and that the time is not far
distant when wine, the most wholesome and purest of all stimulating
drinks, will be within the reach of the common laborer, and take the
place of the noxious and poisonous liquors which are now the curse of
so many of our laboring men, and have blighted the happiness of so many
homes. Pure light wine I consider the best temperance agent; but as
long as bad whisky and brandy continue to be the common drink of its
citizens we can not hope to accomplish a thorough reform; for human
nature seems to crave and need a stimulant. Let us then try to supply
the most innocent and healthy one, the exhilarating juice of the grape.

I have also endeavored throughout to give plain facts, to substantiate
with plain figures all I assert; and in no case have I allowed fancy to
roam in idle speculations which cannot be demonstrated in practice. I
do not pretend that my effort is "the most comprehensive and practical
essay on the grape," as some of our friends call their productions, but
I can claim for it strict adherence to truth and actual results.

I have not thought it necessary to give the botanical description of
the grape-vine, and the process of hybridizing, etc.; this has already
been so well and thoroughly done by my friend FULLER, that I can do no
better than refer the scientific reader to his book. I am writing more
for the practical farmer, and would rather fill what I think a vacancy,
than repeat what has been so well said by others.

With these few remarks, which I thought due to the public and myself, I
leave it to you, brother-winegrowers, to say whether or not I have
accomplished my task. To all and every one who plants a single vine I
would extend the hand of good fellowship, for he is a laborer in the
great work to cover this glorious land of the free with smiling
vineyards, and to make its barren spots flow with noble grape juice,
one of the best gifts of an all-bountiful Creator. All hail to you, I
greet you from _Free_ Missouri.



In an old chronicle, entitled, "The Discovery of America in the Tenth
Century," by CHARLES C. PRASTA, published at Stralsund, we find the
following legend:

"LEIF, son of ERIC the Red, bought BYARNES' vessel, and manned it with
thirty-five men, among whom was also a German, TYRKER by name, who had
lived a long time with LEIF'S father, who had become very much attached
to him in youth. And they left port at Iceland, in the year of our Lord

But, when they had been at sea several days, a tremendous storm arose,
whose wild fury made the waves swell mountain high, and threatened to
destroy the frail vessel. And the storm continued for several days, and
increased in fury, so that even the stoutest heart quaked with fear;
they believed that their hour had come, and drifted along at the mercy
of wind and waves. Only LEIF, who had lately been converted to CHRIST
our Lord, stood calmly at the helm and did not fear; but called on Him
who had walked the water and quieted the billows, with firm faith, that
He also had power to deliver them, if they but trusted in Him. And,
behold! while he still spoke to them of the wonderful deeds of the
Lord, the clouds cleared away, the storm lulled; and after a few hours
the sea, calmed down, and rocked the tired and exhausted men into a
deep and calm sleep. And when they awoke, the next morning, they could
hardly trust their eyes. A beautiful country lay before them, green
hills, covered with beautiful forests--a majestic stream rolled its
billows into the ocean; and they cast the anchor, and thanked the Lord,
who had delivered them from death.

A delightful country it seemed, full of game, and birds of beautiful
plumage; and when they went ashore, they could not resist the
temptation to explore it. When they returned, after several hours,
TYRKER alone was missing. After waiting some time for his return, LEIF,
with twelve of his men, went in search of him. But they had not gone
far, when they met him, laden down with grapes. Upon their enquiry,
where he had stayed so long, he answered: "I did not go far, when I
found the trees all covered with grapes; and as I was born in a
country, whose hills are covered with vineyards, it seemed so much like
home to me, that I stayed a while and gathered them." They had now a
twofold occupation, to cut timber, and gather grapes; with the latter,
they loaded the boat. And Leif gave a name to the country, and called
it Vinland, or Wineland."

So far the tradition. It is said that coming events cast their shadows
before them. If this is so, may we not recognize one of those shadows
in the old Norman legend of events which transpired more than eight
hundred years ago? Is it not the foreshadowing of the destiny of this
great continent, to become, in truth and verity, a _Wineland_. Truly,
the results of to-day would certainly justify us in the assertion, that
there is as much, nay more, truth than fiction in it. Let us take a
glance at the first commencement of grape culture, and see what has
been the progress in this comparatively new branch of horticulture.

From the very first settlement of America, the vine seems to have
attracted the attention of the colonists, and it is said that as early
as 1564, wine was made from the native grape in Florida. The earliest
attempt to establish a vineyard in the British North American Colonies
was by the London Company in Virginia, about the year 1620; and by
1630, the prospect seems to have been encouraging enough to warrant the
importation of several French vine-dressers, who, it is said, ruined
the vines by bad treatment. Wine was also made in Virginia in 1647, and
in 1651 premiums were offered for its production. BEVERLY even
mentions, that prior to 1722, there were vineyards in that colony,
producing seven hundred and fifty gallons per year. In 1664, Colonel
RICHARD NICOLL, Governor of New York, granted to PAUL RICHARDS, a
privilege of making and selling wine free of all duty, he having been
the first to enter upon the cultivation of the vine on a large scale.
BEAUCHAMP PLANTAGENET, in his description of the province of New
Albion, published in London, in 1648, states "that the English settlers
in Uvedale, now Delaware, had vines running on mulberry and sassafras
trees; and enumerates four kinds of grapes, namely: Thoulouse Muscat,
Sweet Scented, Great Fox, and Thick Grape; the first two, after five
months, being boiled and salted and well fined, make a strong red
Xeres; the third, a light claret; the fourth, a white grape which
creeps on the land, makes a pure, gold colored wine. TENNIS PALE, a
Frenchman, out of these four, made eight sorts of excellent wine; and
says of the Muscat, after it had been long boiled, that the second
draught will intoxicate after four months old; and that here may be
gathered and made two hundred tuns in the vintage months, and that the
vines with good cultivation will mend." In 1633, WILLIAM PENN attempted
to establish a vineyard near Philadelphia, but without success. After
some years, however, Mr. TASKER, of Maryland, and Mr. ANTIL, of
Shrewsbury, N.J., seem to have succeeded to a certain extent. It seems,
however, from an article which Mr. ANTIL wrote of the culture of the
grape, and the manufacture of wine, that he cultivated only foreign

In 1796, the French settlers in Illinois made one hundred and ten
hogsheads of strong wine from native grapes. At Harmony, near
Pittsburgh, a vineyard of ten acres was planted by FREDERIC RAPP, and
his associates from Germany; and they continued to cultivate grapes and
silk, after their removal to another Harmony in Indiana.

In 1790, a Swiss colony was founded, and a fund of ten thousand dollars
raised in Jessamine county, Kentucky, for the purpose of establishing a
vineyard, but failed, as they attempted to plant the foreign vine. In
1801, they removed to a spot, which they called Vevay, in Switzerland
County, Indiana, on the Ohio, forty-five miles below Cincinnati. Here
they planted native vines, especially the Cape, or Schuylkill Muscadel,
and met with better success. But, after about forty years' experience,
they seem to have become discouraged, and their vineyards have now
almost disappeared.

These were the first crude experiments in American grape culture; and
from some cause or another, they seem not to have been encouraging
enough to warrant their continuation. But a new impetus was given to
this branch of industry, by the introduction of the Catawba, by Major
ADLUM, of Georgetown, D.C., who thought, that by so doing, he conferred
a greater benefit upon the nation than he would have done, had he paid
the national debt. It seems to have been planted first on an extensive
scale by NICHOLAS LONGWORTH, near Cincinnati, whom we may justly call
one of the founders of American grape culture. He adopted the system of
leasing parcels of unimproved land to poor Germans, to plant with
vines; for a share, I believe, of one-half of the proceeds. It was his
ambition to make the Ohio the Rhine of America, and he has certainly
done a good deal to effect it. In 1858, the whole number of acres
planted in grapes around Cincinnati, was estimated, by a committee
appointed for that purpose, at twelve hundred acres, of which Mr.
LONGWORTH owned one hundred and twenty-two and a half acres, under
charge of twenty-seven tenants. The annual produce was estimated by the
committee at no less than two hundred and forty thousand gallons, worth
about as many dollars then. We may safely estimate the number of acres
in cultivation there now, at two thousand. Among the principal grape
growers there, I will mention Messrs. ROBERT BUCHANAN, author of an
excellent work on grape culture, MOTTIER, BOGEN, WERK, REHFUSS, DR.
MOSHER, etc.

Well do I remember, when I was a boy, some fourteen years old, how
often my father would enter into conversation with vintners from the
old country, about the feasibility of grape culture in Missouri. He
always contended that grapes should succeed well here, as the woods
were full of wild grapes, some of very fair quality, and that this
would indicate a soil and climate favorable to the vine. They would
ridicule the idea, and assert that labor was too high here, even if the
vines would succeed, to make it pay; but they could not shake his faith
in the ultimate success of grape culture. Alas! he lived only long
enough to see the first dawnings of that glorious future which he had
so often anticipated, and none entered with more genuine zeal upon the
occupation than he, when an untimely death took him from the labor he
loved so well, and did not even allow him to taste the first fruits of
the vines he had planted and fostered. Had he been spared until now,
his most sanguine hopes would be verified.

I also well remember the first cultivated grape vine which produced
fruit in Hermann. It was an Isabella, planted by a Mr. FUGGER, on the
corner of Main and Schiller streets, and trained over an arbor. It
produced the first crop in 1845, twenty years ago, and so plentifully
did it bear, that several persons were encouraged by this apparent
success, to plant vines. In 1846, the first wine was made here, and
agreeably surprised all who tried it, by its good quality. The Catawba
had during that time, been imported from Cincinnati, and the first
partial crop from it, in 1848, was so plentiful, that every body,
almost, commenced planting vines, and often in very unfavorable
localities. This, of course, had a bad influence on so capricious a
variety as the Catawba; rot and mildew appeared, and many became
discouraged, because they did not realize what they had anticipated. A
number of unfavorable seasons brought grape growing almost to a stand
still here. Some of our most enterprising grape growers still
persevered, and succeeded by careful treatment, in making even the
Catawba pay very handsome returns.

It was about this time, that the attention of some of our grape-growers
was drawn towards a small, insignificant looking grape, which had been
obtained by a Mr. WIEDERSPRECKER from Mr. HEINRICHS, who had brought it
from Cincinnati, and, almost at the same time, by Dr. KEHR, who had
brought it with him from Virginia. The vine seemed a rough customer,
and its fruit very insignificant when compared with the large bunch and
berry of the Catawba, but we soon observed that it kept its foliage
bright and green when that of the Catawba became sickly and dropped;
and also, that no rot or mildew damaged the fruit, when that of the
Catawba was nearly destroyed by it. A few tried to propagate it by
cuttings, but generally failed to make it grow. They then resorted to
grafting and layering, with much better success. After a few years a
few bottles of wine were made from it, and found to be very good. But
at this time it almost received its death-blow, by a very unfavorable
letter from Mr. LONGWORTH, who had been asked his opinion of it, and
pronounced it worthless. Of course, with the majority, the fiat of Mr.
LONGWORTH, the father of American grape-culture, was conclusive
evidence, and they abandoned it. Not all, however; a few persevered,
myself. We thought Mr. LONGWORTH was human, and might be mistaken; and
trusted as much to the evidence of our senses as to his verdict,
therefore increased it as fast as we could, and the sequel proved that
we were right. After a few years more wine was made from it in larger
quantities, found to be much better than the first imperfect samples;
and now that despised and condemned grape is _the_ great variety for
red wine, equal, if not superior to, the best Burgundy and Port; a wine
of which good judges, heavy importers of the best European wines too,
will tell you that it has not its equal among all the foreign red
wines; which has already saved the lives of thousands of suffering
children, men, and women, and therefore one of the greatest blessings
an all-merciful God has ever bestowed upon suffering humanity. This
despised grape is now the rage, and 500,000 of the plants could have
been sold from this place alone the last fall, if they could have been
obtained. Need I name it? it is the Norton's Virginia. Truly, "great
oaks from little acorns grow!" and I boldly prophecy to-day that the
time is not far distant when thousands upon thousands of our hillsides
will be covered with its luxuriant foliage, and its purple juice become
one of the exports to Europe; provided, always, that we do not grow so
fond of it as to drink it all. I think that this is pre-eminently a
Missouri grape. Here it seems to have found the soil in which it
flourishes best. I have seen it in Ohio, but it does not look there as
if it was the same grape. And why should it? They drove it from them
and discarded it in its youth; we fostered it, and do you not think,
dear reader, there sometimes is gratitude in plants as well as in men?
Other States may plant it and succeed with it, too, to a certain
extent, but it will cling with the truest devotion to those localities
where it was cared for in its youth. Have we not also found, during the
late war, that the Germans, the adopted citizens of this great country,
clung with a heartier devotion to our noble flag, and shed their blood
more freely for it, than thousands upon thousands of native-born
Americans? And why? Because here they found protection, equal rights
for all, and that freedom which had been the idol of their hearts, and
haunted their dreams by night; because they had been oppressed so long
they more fully appreciated the blessings of a free government than
those who had enjoyed it from their birth. But you may call me
fantastical for comparing plants to human beings, and will say, plants
have no appreciation of such things. Brother Skeptic, have you, or has
any body, divined _all_ the secrets of Nature's workshop? Truly we may
say that we have not, and we meet with facts every day which are
stranger than fiction.

The Concord had as small a beginning with us. In the winter of 1855 a
few eyes of its wood were sent me by Mr. JAS. G. SOULARD, of Galena,
Ill. I grafted them upon old Catawba vines, and one of them grew. The
next year I distributed some of the scions to our vine-growers, who
grafted them also. When my vine commenced to bear I was astonished,
after what I had heard of the poor quality of the fruit from the East,
to find it so fine, and so luxurious and healthy; and we propagated it
as fast as possible. Now, scarcely nine years from the time when I
received the first scions, hundreds of acres are being planted with it
here, and one-third of an acre of it, planted five years ago, has
produced for me, in fruit, wine, layers, cuttings, and plants, the
round sum of ten thousand dollars during that time. Its wine, if
pressed as soon as the grapes are mashed, is eminently one of those
which "maketh glad the heart of man," and is evidently destined to
become one of the common drinks of our laboring classes. It is light,
agreeable to the palate, has a very enlivening and invigorating effect,
and can be grown as cheap as good cider. I am satisfied that an acre
will, with good cultivation, produce from 1,000 to 1,500 gallons per
year. My vines produced this season at the rate of 2,500 gallons to the
acre, but this may be called an extra-large crop. I have cited the
history of these two varieties in our neighborhood merely as examples
of progress. It would lead too far here, to follow the history of all
our leading varieties, though many a goodly story might be told of
them. Our friends in the East claim as much for the Delaware and
others, with which we have not been able to succeed. And here let me
say that the sooner we divest ourselves of the idea that one grape
should be _the_ grape for this immense country of ours; the sooner we
try to adapt the variety to the locality--not the locality to the
variety--the sooner we will succeed. The idea is absurd, and unworthy
of a thinking people, that one variety should succeed equally well or
ill in such a diversity of soil and climate as we have in this broad
land of ours. It is in direct conflict with the laws of vegetable
physiology, as well as with common sense and experience. In planting
our vineyards we should first go to one already established, which we
think has the same soil and location, or nearly so, as the one we are
going to plant. Of those varieties which succeed there we should plant
the largest number, and plant a limited number also of all those
varieties which come recommended by good authority. A few seasons will
show which variety suits our soil, and what we ought to plant in
preference to all others. Thus the Herbemont, the Cynthiana, Delaware,
Taylor, Cunningham, Rulander, Martha, and even the Iona, may all find
their proper location, where each will richly reward their cultivator;
and certainly they are all too good not to be tried.

Now, let us see what progress the country at large has made in
grape-growing during, say, the last ten years. _Then_, I think I may
safely assert, that the vineyards throughout the whole country did not
comprise more than three to four thousand acres. _Now_ I think I may
safely call them over two millions of acres. _Then_, our whole list
embraced about ten varieties, all told, of which only the Catawba and
Isabella were considered worthy of general cultivation; _now_ we count
our native varieties by the hundreds, and the Catawba and Isabella will
soon number among the things which have been. Public taste has become
educated, and they are laid aside in disgust, when such varieties as
the Herbemont, Delaware, Clara, Allen's Hybrid, Iona, Adirondac, and
others can be had. _Then_, grape-growing was confined to only a few
small settlements; _now_ there is not a State in the Union, from Maine
to California, but has its vineyards; and especially our Western States
have entered upon a race which shall excel the other in the good work.
Our brethren in Illinois bid fair to outdo us, and vineyards spring up
as if by magic, even on the prairies. Nay, grape-culture bids fair to
extend into Minnesota, a country which was considered too cold for
almost anything except oats, pines, wolves, bears, and specimens of
daring humanity encased in triple wool. We begin to find out that we
have varieties which will stand almost anything if they are only
somewhat protected in winter. It was formerly believed that only
certain favored locations and soils in each State would produce good
grapes--for instance, sunny hillsides along large streams; now we begin
to see that we can grow some varieties of grape on almost any soil. One
of the most flourishing vineyards I have ever seen is on one of the
islands in the Missouri river, where all the varieties planted
there--some six or seven--seemed perfectly at home in the rich, sandy
mould, where it needs no trenching to loosen the soil. _Then_,
grape-growing, with the varieties then in cultivation, was a problem to
be solved; _now_, with the varieties we have proved, it is a certainty
that it is one of the most profitable branches of horticulture, paying
thousands of dollars to the acre every year. _Then_, wine went begging
at a dollar a gallon; _now_ it sells as fast as made at from two
dollars to six dollars a gallon. Instead of the only wine then
considered fit to drink, we number our wine-producing varieties by the
dozen, all better than the Catawba; among the most prominent of which I
will name--of varieties producing white wine, the Herbemont, Delaware,
Cassidy, Taylor, Rulander, Cunningham, and Louisiana; of light-red
wines, the Concord; of dark-red wines, the Norton's Virginia,
Cynthiana, Arkansas and Clinton; so that every palate can be suited.
And California bids fair to outdo us all; for there, I am told, several
kinds of wine are made from the same grape, in the same vineyard, and
in fabulous quantities. To cite an example of the increase in planting:
in 1854 the whole number of vines grown and sold in Hermann did not
exceed two thousand. This season two millions of plants have been grown
and sold, and not half enough to meet the demand. It is said that the
tone of the press is a fair indication of public sentiment. If this is
true what does it prove? Take one of our horticultural periodicals, and
nine-tenths of the advertisements will be "Grape-vines for sale," in
any quantity and at any price, from five dollars to one hundred dollars
per 100, raised North, East, South, and West. Turn to the reading
matter, and you can hardly turn over a leaf but the subject of grapes
stares you in the face, with a quiet impunity, which plainly says, "The
nation is affected with grape fever; and while our readers have grape
on the brain there is no fear of overdosing." Why, the best proof I can
give my readers that grape fever does exist to an alarming degree, is
this very book itself. Were not I and they affected with the disease, I
should never have presumed to try their patience.

But, fortunately, the remedy is within easy reach. Plant grapes, every
one of you who is thus afflicted, until our hillsides are covered with
them, and we have made our barren spots blossom as the rose.

Truly, the results we have already obtained, are cheering enough. And
yet all this has been principally achieved in the last few years, while
the nation was involved in one of the most stupendous struggles the
world ever saw, while its very existence was endangered, and thousands
upon thousands of her patriotic sons poured out their blood like water,
and the husbandman left his home; the vintner his vineyard, to fight
the battles of his country. What then shall we become now, when peace
has smiled once more upon our beloved country; and the thousands of
brave arms, who brandished the sword, sabre, or musket, have come home
once more; and their weapons have been turned into ploughshares, and
their swords into pruning hooks? When all the strong and willing hands
will clear our hillsides, and God's sun shines upon _one_ great and
united people; greater and more glorious than ever; because now they
are _truly free_. Truly the future lies before us, rich in glorious
promise; and ere long the words and the prophecy contained in the old
legend will become sober truth, and America will be, from the Atlantic
to the Pacific _one_ smiling and happy _Wineland_; where each laborer
shall sit under his own vine, and none will be too poor to enjoy the
purest and most wholesome of all stimulants, good, cheap, native
_wine_. Then drunkenness, now the curse of the nation, will disappear,
and peace and good will towards all will rule our actions. And we,
brother grape growers? Ours is this great and glorious task; let us
work unceasingly, with hand, heart, and mind; truly the object is
worthy of our best endeavors. Let those who begin to-day, remember how
easy their task with the achievements and experiments of others before
them, compared with the labors of those who were the pioneers in the
cultivation of the vine.



This would seem to be the most natural mode, were not the grape even
more liable to sport than almost any other fruit. It is, however, the
only method upon which we can depend for obtaining new and more
valuable varieties than we already possess, and to which we are already
indebted for all the progress made in varieties, a progress which is,
indeed, very encouraging; for who would deny that we are to-day
immeasurably in advance of what we were ten years ago. Among the
innumerable varieties which spring up every day, and which find ready
purchasers, just because they _are new_, there are certainly some of
decided merit. But those who grow seedlings, should bear in mind, that
the list of our varieties is already too large; that it would be better
if three-fourths of them were stricken off, and that no new variety
should be brought before the public, unless it has some decided
superiority over any of the varieties we already have, in quality,
productiveness and exemption from disease. It is poor encouragement to
the grape growing public, to pay from two to five dollars a vine for a
new variety, with some high-sounding name, if, after several years of
superior cultivation and faithful trial, they find their costly pet
inferior to some variety they already possessed, and of which the
plants could be obtained at a cost of from ten to fifty cents each.

The grapes from which the seed is to be used, should be fully ripe, and
none but well developed, large berries, should be taken. Keep these
during the winter, either in the pulp, or in cool, moist sand, so that
their vitality may remain unimpaired. The soil upon which your seed-bed
is made, should be light, deep and rich, and if it is not so naturally,
should be made so with well decomposed leaf-mould. As soon as the
weather in spring will permit, dig up the soil to the depth of at least
eighteen inches, pulverising it well; then sow the seed in drills,
about a foot apart, and about one inch apart in the rows, covering them
about three-quarters of an inch deep. It will often be found necessary
to shade the young plants when they come up, to prevent the sun from
scalding them, but this should not be continued too long, as the plants
will become too tender, if protected too long. When the young plants
have grown about six inches, they may be supplied with small sticks, to
which they will cling readily; the ground should be kept clean and
mellow, and a light mulch should be applied, which will keep the soil
loose and moist. The young plants should be closely watched, and if any
of them show signs of disease, they should at once be pulled up; also
those which show a very feeble and delicate growth; for we should only
try to grow varieties with good, healthy constitutions. In the Fall,
the young plants should be either taken up, and carefully heeled in, or
they should be protected by earth, straw, or litter thrown over them.
In the Spring, they may be transplanted to their permanent locations;
the tops shortened in to six inches, and the roots shortened in to
about six inches from the stem. The soil for their reception should be
moderately light and rich, and loosened up to the depth of at least
eighteen inches.

Make a hole about eight inches deep, then throw in soil so as to raise
a small mound in the centre of the hole, about two inches high; on this
place the young vine, and carefully spread the roots in all directions;
then fill up with well pulverized soil, so that the upper eye or bud is
even with the surface of the ground; then press the soil down lightly;
place a good stake, of about four feet high, with the plant, and allow
but one shoot to grow, which should be neatly tied to the stake as it
grows. The vines may be planted in rows six feet apart, and three feet
apart in the rows, as many of them will prove worthless, and have to be
taken out. Allow all the laterals to grow on the young cane, as this
will make it short-jointed and stocky. Cultivate the ground well,
stirring it freely with plough, cultivator, hoe, and rake, which
generally is the best mulch that can be applied.

With the proper care and attention, our seedlings will generally grow
from three to four feet, and make stout, short-jointed wood this second
season. Should any of them look particularly promising, fruit may be
obtained a year sooner by taking the wood of it, and grafting strong
old vines with it. These grafts will generally bear fruit the next
season. The method to be followed will be given in another place.

At the end of the second season the vines should be pruned to about
three eyes or buds, and the soil hilled up around them so as to cover
them up completely. The next spring take off the covering, and when the
young shoots appear allow only two to grow. After they have grown about
eighteen inches, pinch off the top of the weakest, so as to throw the
growth into the strongest shoot, which keep neatly tied to the stake,
treating it as the summer before, allowing all the laterals to grow.
Cultivate the soil well. At the end of this season's growth the vines
should be strong enough to bear the following summer. If they have made
from eight to ten feet of stocky growth, the leading cane may be pruned
to ten or twelve eyes, and the smaller one to a spur of two eyes. If
they will fruit at all, they will show it next summer, when only those
promising well should be kept, and the barren and worthless ones


As this method is mostly followed only by those who propagate the vine
for sale in large quantities, and but to a limited extent by the
practical vineyardist, I will give only an outline of the most simple
manner, and on the cheapest plan. Those wishing further information
will do well to consult "The Grape Culturist," by Mr. A. S. FULLER, in
which excellent work they will find full instructions.

The principal advantages of this mode of propagation are the following:
1st. The facility with which new and rare kinds can be multiplied, as
every well ripened bud almost can be transformed into a plant. 2d. As
the plants are started under glass, by bottom heat, it lengthens the
season of their growth from one to two months. 3d. Every variety of
grape can be propagated by this method with the greatest ease, even
those which only grow with the greatest difficulty, or not at all, from
cuttings in open ground.

As to the merits or demerits of plants grown under glass from single
eyes, to those grown from cuttings or layers in open ground, opinions
differ very much, and both have their advocates. For my part, I do not
see why a plant grown carefully from a single eye should not be as good
as one propagated by any other method; a poor plant is not worth
having, whether propagated by this or any other method, and,
unfortunately, we have too many of them.


I will only give a description of a lean-to of the cheapest kind, for
which any common hot-bed sash, six feet long, can be used.

Choose for a location the south side of a hill, as, by making the house
almost entirely underground, a great deal of building material can be
saved. Excavate the ground as for a cellar--say five feet deep on the
upper side, seven feet wide, and of any length to suit convenience, and
the number of plants you wish to grow. Inside of the excavation set
posts or scantlings, the upper row to be seven feet long above the
ground, and two feet below the ground; the lower row four and one-half
feet above the ground, so that the roof will have about two and
one-half feet pitch. Upon these nail the rafters, of two-inch planks.
Then take boards, say common inch-plank, and set them up behind the
posts, one above the other, to prevent the earth from falling in. This
will make all the wall that is needed on both sides. On the ends,
boards can be nailed to both sides of the posts, and the intervening
space tilled with spent tan or saw-dust. Upon the rafters place the
sash on the lower side; the upper side may be covered with boards or
shingles, where also the ventilating holes can be left, to be closed
with trap-doors. The house is to be divided into two compartments--the
furnace-room on one end, about eight feet long, and the propagating
house, The furnace is below the ground, say four feet long, the flue to
be made of brick, and to extend under the whole length of the bench. To
make the flue, lay a row of bricks flat and crosswise; on the ends of
these place two others on their edges, and across the top lay a row
flat, in the same way as the bottom ones were placed. This gives the
flue four inches by eight in the clear. The flue should rise rather
abruptly from the furnace, say about a foot; it can then be carried
fifty feet with, say six to nine inches rise, and still have sufficient
draft. Inside of the propagating room we have again two
compartments--the propagating bench, nearest to the furnace, and a
shelf for the reception of the young plants, after their first
transplanting from the cutting-pots or boxes. Make a shelf or table
along the whole length of the house; at the lower end it should be
about eighteen inches from the glass, and five feet wide. To a house
of, say fifty feet, the propagating bench may be, say twelve feet long,
and the room below it and around the flue should be inclosed with
boards, as it will keep the heat better.


The wood should be cut from the vines in the fall, as soon as the
leaves have dropped. For propagating, use only firm, well-ripened wood
of the last season's growth, and about medium thickness. These are to
be preferred to either very large or very small ones. The time to
commence operating will vary according to climate; here it should be
the early part of February. The wood to be used for propagating can be
kept in a cool cellar, in sand, or buried in the ground out doors. Take
out the cuttings, and cut them up into pieces as represented in Figure

[Illustration: FIG. 1.]

Throw these into water as they are cut; it will prevent them from
becoming dry. It will be found of benefit with hard-wooded varieties to
pack them in damp moss for a week or so before they are put into the
propagating pots or boxes; it will soften the alburnous matter, and
make them strike root more readily. They should then be put into, say
six-inch pots, filled to about an inch of the top with pure coarse
sand, firmly packed. Place the cuttings, the buds up, about an inch
apart, all over the surface of the pot; press down firmly with thumb
and forefinger until the bud is even with the surface; sift on sand
enough to cover the upper point of the bud about a quarter of an inch
deep; press down evenly, using the bottom of another pot for the
purpose, and apply water enough to moisten the whole contents of the
pot. Instead of the pots, shallow boxes of about six inches deep, can
also be used, with a few holes bored in the bottom for drainage.

After the pots have been filled with cuttings they are placed in a
temperature of from 40° to 45°, where they remain from two to three
weeks, water being applied only enough to keep them moist, not wet. As
roots are formed at a much lower degree of temperature than leaves,
they should not be forced too much at the beginning, or the leaves will
appear before we have any roots to support them. But when the cutting
has formed its roots first, the foliage, when it does appear, will grow
much more rapidly, and without any check. Then remove them to another
position, plunging the pots into sand to the depth of, say three
inches, and raise the temperature at first to 60° for the first few
days, then gradually raise it to 80°. When the buds begin to push,
raise the temperature to 90° or 95°, and keep the air moist by frequent
waterings, say once a day. The best for this purpose is pure
rain-water, but it should be of nearly the same temperature as the air
in the house, for, if applied cold, it would surely check the growth of
the plants. The young growth should be examined every day, to see if
there is any sign of rotting; should this be the case, give a little
more air, but admit no sudden cold currents, as they are often fatal.
The glass should be whitewashed, to avoid the direct rays of the sun.

When the young vines have made a growth of two or three inches shift
them into three-inch pots.

So far we have used only pure sand, which did not contain much plant
food, because the growth was produced from the food stored up in the
bud and wood, and what little they obtained from the sand, water, and
air. Now, however, our young vines want more substantial food. They
should therefore be potted into soil, mixed from rotten sod,
leaf-mould, and well-decomposed old barnyard manure. This should be
mixed together six months before using; add, before using, one-quarter
sand, then mix thoroughly, and sift all through a coarse sieve. In
operating, put a quantity of soil on the potting bench, provide a
quantity of broken bricks or potsherds for drainage, loosen the plants
from the pots by laying them on their side, giving them a sudden jar
with the hand, to loosen the sand around them; draw out the plant
carefully, holding it with one hand, while with the other you place a
piece of the drainage material into the pot; cover it with soil about
an inch; then put in the plant, holding it so that the roots spread out
naturally; fill in soil around them until the pot is full; press the
soil down firmly, but not hard enough to break the roots. When the
plants are potted give them water to settle the earth around the roots,
and keep the air somewhat confined for a few days, until they have
become established, when more air may be given them. Keep the
temperature at 85° to 95° during the day, and 70° to 80° during the

When the plants have made about six inches of growth they can either be
placed in another house, or in hot-bed frames, if they are to be kept
under glass. The usual manner of keeping them in pots during summer,
shifting them into larger and larger sizes, I consider injurious to the
free development of the plants, as the roots are distorted and cramped
against the sides of the pots, and cannot spread naturally. I prefer
shifting them into cold frames, in which beds have been prepared of
light, rich soil, into which the young plants can be planted, and kept
under whitewashed hot-bed sashes for a while, which, after several
weeks, may be removed, and only a light shading substituted in their
place, which, after several weeks more, can also be removed. Thus the
young plants are gradually hardened, their roots have a chance to
spread evenly and naturally, without any cramping; and such plants,
although they may not make as tall a growth as those kept under glass
all the season, will really stand transplanting into the vineyard much
better than those hot-house pets, which may look well enough, but
really are, like spoiled and pampered children, but poorly fitted to
stand the rough vicissitudes of every-day life.

The young plants should be lightly tied to small sticks provided for
the purpose, as it will allow free circulation of air, and admit the
sun more freely to the roots. In the fall, after their leaves have
dropped, they should be carefully taken up, shortened to about a foot
of their growth, and they are then ready either to sell, if they are to
be disposed of in that way, or for planting into the vineyard. They
should, however, be carefully assorted, making three classes of
them--the strongest, medium, and the smallest--each to be put separate.
The latter generally are not fit to transplant into the vineyard, but
they may be heeled in, and grown in beds another year, when they will
often make very good plants. Heeling in may be done as shown in Figure
2, laying the vines as close in the rows as they can conveniently be
laid, and then fill the trench with well-pulverized soil. They can thus
be safely kept during the winter.

[Illustration: FIG. 2.]

I have only given an outline of the most simple and cheapest mode of
growing plants from single eyes, such as even the vineyardist may
follow. For descriptions of more extensive and costly buildings, if
they desire them, they had better apply to an architect. I have also
not given the mode of propagating from green wood, as I do not think,
plants thus propagated are desirable. They are apt to be feeble and
diseased, and I think, the country at large would be much better off,
had not a single plant ever been produced by that method.

Plants from single eyes may also be grown in a common hot-bed; but as
in this the heat can not be as well regulated at will, I think it, upon
the whole, not desirable, as the expense of a propagating house on the
cheap plan I have indicated, is but very little more, and will
certainly in the long run, pay much better. Of course, close attention
and careful watching is the first requisite in all the operations.


This is certainly the easiest and most simple method for the
vineyardist; can be followed successfully with the majority of
varieties, which have moderately soft wood, and even a part of the hard
wood varieties will generally grow, if managed carefully.


There are several methods, which are followed with more or less
success. I will first describe that which I have found most successful,
namely, short cuttings, of two or three eyes each, which are made of
any sound, well ripened wood, of last season's growth. Prune the vines
in the fall or early winter, and make the cuttings as soon as
convenient; for if the wood is not kept perfectly fresh and green, the
cuttings will fail to grow. Now, cut up all the sound, well-ripened
wood into lengths of from two to four eyes each, making them of a
uniform length of say eight inches, and prepare them as shown in Figure

[Illustration: FIG. 3.]

These should be tied into convenient bundles, from 100 to 250 in each,
taking care to even the lower ends, and then buried in the ground,
making a hole somewhat deeper than the cuttings are long, into which
the bundles are set on their lower ends, and soil thrown in between and
over them. In spring, as soon as the ground is dry enough, the
cutting-bed should be prepared. Choose for this a light, rich soil,
which should be well pulverized, to the depth of at least a foot, and
if not light enough, it should be made so by adding some leaf mould.
Now draw a line along the whole length of the bed; then take a spade
and put it down perpendicular along the line or nearly so, moving it a
little backwards and forwards, so as to open the cut. Now take the
cutting and press it down into the cut thus made, until the upper bud
is even with the surface of the soil. The cuttings may be put close in
the rows, say an inch apart, and the rows made two feet apart. Press
the ground firmly down with your foot along the line of cuttings, so as
to pack it closely around the cutting. After the bed is finished, mulch
them with straw, or litter, spent tan or saw-dust, say about an inch
thick, and if none of these can be had, leaves from the forest may be
used for the purpose. This will serve to protect the young leaves from
the sun, and will also keep an even moisture during the heat of summer,
at the same time keeping the soil loose and porous. If weeds appear,
they should be pulled up, and the cuttings, kept clean through the
summer. They will generally make a firm, hardy growth of from one to
four feet, have become used to all the hardships and changes of the
weather; and as they have formed their roots just where they ought to
be, about eight inches below the ground, will not suffer so much from
transplanting, as either a single eye or a layer, whose roots have to
be put much deeper in transplanting, than they were before, and thus,
as it were, become acclimated to the lower regions. For these reasons,
I think, that a good plant grown from a cutting is preferable to that
propagated by any other method. In the Fall, the vines are carefully
taken up, assorted and heeled in, in the same manner as described, with
single eyes, and cut back to about three inches of their growth. They
are then ready for transplanting into the vineyard.


This is a very convenient method of increasing such varieties as will
not grow readily from cuttings; and vines thus propagated will, if
treated right, make very good plants. To layer a vine, shorten in its
last season's growth to about one-half; then prepare the ground
thoroughly, pulverizing it well; then, early in spring make a small
furrow, about an inch deep, then bend the cane down and fasten it
firmly in the bottom of the trench, by wooden hooks or pegs, made for
the purpose. They may thus be left, until the young shoots have grown,
say six inches; then fill up with finely pulverized soil or leaf-mould.
The vines will thus strike root generally at every joint. The young
shoots may be tied to small sticks, provided for the purpose, and when
they have grown about a foot, their tips should be pinched off to make
them grow more stocky. In the Fall they are taken up carefully,
commencing to dig at the end furthest removed from the vine, and
separate each plant between the joints, so that every shoot has a
system of roots by itself. They are then either planted immediately, or
heeled in as described before.


The principal advantages to be gained by this method are: 1st. The
facility by which new and rare kinds may be increased, by grafting them
on strong stocks of healthy varieties, when they will often grow from
ten to twenty feet the first season, producing an abundance of wood to
propagate. 2d. The short time in which fruit can be obtained from new
and untried varieties, as their grafts will generally bear the next
season. 3d. In every vineyard there are, in these days of many
varieties, vines which have proved inferior, yet by grafting into them
some superior variety, they may be made very valuable. 4th. The
facility by which vines can be forced under glass, by grafting on small
pieces of roots, and the certainty with which every bud can thus be
made to grow.

[Illustration: FIG. 4.]

The vine, however, does not unite with the same facility as the pear
and apple, and, to ensure success, must be grafted under ground, which
makes the operation a difficult and disagreeable one. It will therefore
hardly become a general practice; but, for the purposes above named, is
of sufficient importance, to make it desirable that every vineyardist
should be able to perform it. I have generally had the best success in
grafting here about the middle of March, in the following manner: Dig
away the ground around the vine you wish to graft, until you come to a
smooth place to insert your scion; then cut off the vine with a sharp
knife, and insert one or two scions, as in common cleft-grafting,
taking care to cut the wedge on the scion very thin, with shoulders on
both sides, as shown in Figure 4, cutting your scion to two eyes, to
better insure success. Great care must be taken to insert the scion
properly, as the inner bark or liber of the vine is very thin, and the
success of the operation depends upon a perfect junction of the stock
and scion. If the vine is strong enough to hold the scion firmly, no
further bandage is necessary; if not, it should be wound firmly and
evenly with bass bark. Then press the soil firmly on the cut, and fill
up the hole with well pulverized earth, to the top of the scion.
Examine the stock from time to time, and remove all wild shoots and
suckers, which it may throw up, as they will rob the graft of
nourishment and enfeeble it.

Others prefer to graft in May, when the leaves have expanded, and the
most rapid flow of sap has ceased, keeping the scions in a cool place,
to prevent the buds from starting. The operation is performed in
precisely the same manner, and will be just as successful, I think, but
the grafts that have been put in early, have the advantage of several
weeks over the others, and the latter will seldom make as strong a
growth, or ripen their wood as well as those put in early.

Mr. A. S. FULLER performs the operation in the fall, preventing the
graft from freezing by inverting a flower-pot over it, and then
covering with straw or litter. He claims for this method--1st. That it
can be performed at a time when the ground is more dry, and in better
condition, and business not so pressing as in spring.--2d. That the
scion and stock have more time to unite, and will form their junction
completely during the winter, and will therefore start sooner, and make
a more rapid growth than in spring. It certainly looks feasible enough,
and is well worth trying, as, when the operation succeeds, it must
evidently have advantages over any of the other modes.

Vines I had grafted in March have sometimes made twenty to thirty feet
of growth, and produced a full crop the next season. This will show one
the advantage to be derived from it in propagating new and scarce
varieties, and in hastening the fruiting of them. Should a seedling,
for instance, look very promising in foliage and general appearance,
fruit may be obtained from it from one to two seasons sooner by
grafting some of the wood on strong stocks, than from the original
plant. Hence the vast importance of grafting, even to the practical



As the selection of a proper location is of vast importance, and one of
the main conditions of success, great care and judgment should be
exercised in the choice. Some varieties of grapes may be grown on
almost any soil, it is true; but even they will show a vast difference
in the quality of the fruit, even if the quantity were satisfactory; on
indifferent soil, and in an inferior location. Everybody should grow
grapes enough for his own use, who owns an acre of ground, but every
one cannot grow them and make the most delicious wine.

The best locations are generally on the hillsides, along our larger
rivers, water-courses, and lakes, sloping to the East, South, and
Southwest, as they are generally more exempt from late spring frosts
and early frosts in fall. The location should be sheltered from the
cold winds from the north and northwest, but fully exposed to the
prevailing winds in summer from the south and southwest. If a hill is
chosen at any distance from a large body of water, it should be high
and airy, with as gentle a slope as can be obtained. The locations
along creeks and smaller water-courses should be particularly avoided,
as they are subject to late spring frosts, and are generally damp and

The soil should be a dry, calcareous loam, sufficiently deep, say three
feet; if possible, draining itself readily. Should this not be the case
naturally, it should be done with tiles.

I was much struck by the force of a remark made by medical friend last
summer, when, in consequence of the continual rains, the ague was very
prevalent. It was this: wherever you will find the ague an habitual
guest with the inhabitants you need not look for healthy grapevines.
Wherever we find stagnant water let us avoid the neighboring hillsides,
for they would not be congenial to our grape-vines. But on the bluffs
overhanging the banks of our large streams, especially on the northern
and western sides, where the vines are sheltered from the north and
west winds, and fully exposed to the warm southern winds of our summer
days, and where the fogs arising from the water yet give sufficient
humidity to the atmosphere, even in the hottest summer days, to refresh
the leaf during the night and morning hours; where the soil on the
southern and eastern slopes is a mixture of decomposed stone and
leaf-mould, and feels like velvet to the feet--there is the paradise
for the grape; and the soil is already better prepared for it than the
hand of man can ever do. Such locations should be cheap to the
grape-grower at _any_ price. We find them very frequently along the
northern banks of the Missouri and Mississippi rivers, and they will no
doubt become the favored grape regions of the country. The grape grows
there with a luxuriance and health which is almost incredible to those
living in less favored locations.

But the question may be asked here, what shall be done by those who do
not live in these favored regions, and yet would like to grow grapes? I
answer, let them choose the best location they have, the most free and
airy, and let them choose only those sturdy varieties that withstand
everything. They cannot grow the most delicate varieties--the
Herbemont, the Delaware, the Clara, are not for them; but they can grow
the Concord, Hartford Prolific, and Norton's Virginia, and they at
least are "very good," although they may not be the "best." There is no
excuse for any one in this country why he should not grow his own
grapes, for the use of his family at least, if he has any ground to
grow them on.


In this, the foundation of all grape-growing, the vineyardist must also
look to the condition in which he finds the soil. Should it be free of
stones, stumps, and other obstructions, the plough and sub-soil plough
will be all-sufficient.

Should your soil be new, perhaps a piece of wild forest land, have it
carefully grubbed, and every tree and stump taken out by the roots.
After the ground is cleared take a large breaking-plough, with three
yoke of sturdy oxen, and plough as deep as you can, say twelve to
fourteen inches. Now follow in the same furrow with an implement we
call here a sub-soil stirrer, and which is simply a plough-share of
wedge shape, running in the bottom of the furrow, and a strong coulter,
running up from it through the beam of the plough, sharp in front, to
cut the roots; the depth of the furrow is regulated by a movable wheel
running in front, which can be set by a screw. With two yoke of oxen
this will loosen the soil to the depth of, say twenty inches, which is
sufficient, unless the sub-soil is very tenacious. In land already
cultivated, where there are no roots to obstruct, two yoke of oxen or
four horses attached to the plough, and one yoke of oxen or a pair of
horses or mules to the sub-soil plough, will be sufficient. In stony
soil the pick and shovel must take the place of the plough, as it would
be impossible to work it thoroughly with the latter; but I think there
is no advantage in the common method of trenching or inverting the
soil, as is now practiced to a very great extent. If we examine the
growth of our native vines we will generally find their roots extending
along the surface of the soil. It is unnatural to suppose that the
grape, the most sun-loving of all our plants, should be buried with its
roots several feet below the surface of the soil, far beyond the reach
of sun and air. Therefore, if you can afford it, work your soil deep
and thoroughly; it will be labor well invested; is the best preventive
against drouth, and also the best drainage in wet weather; but have it
in its natural position--not invert it; and do not plant too deep.
Should the soil be very poor it may be enriched by manure, ashes,
bone-dust, etc.; but it will seldom be found necessary, as most of our
soil is rich enough; and it is not advisable to stimulate the growth
too much, as it will be rank and unhealthy, and injurious to the
quality and flavor of the fruit.

Wet spots may be drained by gutters filled with loose stones, or tiles,
and then covered with earth. Surface-draining can be done by running a
small ditch or furrow every sixth or eighth row, parallel with the
hillside, and leading into a main ditch at the end or the middle of the
vineyard. Steep hillsides should be terraced or benched; but, as this
is very expensive, they should be avoided.



It is a very difficult matter, in a vast country like ours, where the
soil and climate differ so much, to recommend any thing; and I think it
a mistake, into which many of our prominent grape-growers have fallen,
to recommend _any_ variety, simply because it succeeded well _with
them_, for _general_ cultivation. Grape-growing is, perhaps, more than
any other branch of horticulture or pomology, dependent upon soil,
location and climate, and it will not do to dictate to the inhabitants
of a country, in which the "extremes meet," that they should _all_
plant one variety. Yet this has been done by some who _pretend_ to be
authorities, and it shows, more than any thing else, that they have
more arrogance than knowledge. I, for my part, have seen such widely
different results, from the same varieties, under the same treatment,
and in vineyards only a few miles apart, but with a different soil and
different aspect, that I am reluctant to recommend to my next neighbor,
what he shall plant.

But, while the task is a difficult one, yet we may lay down certain
rules, which can govern us in selection of varieties to a certain
extent. We should choose--1st. The variety which has given the most
general satisfaction in the State or county in which we live, or the
nearest locality to us. 2d--Visit the nearest accessible vineyard in
the month of August and September, observe closely which variety has
the healthiest foliage and fruit; ripens the most uniformly and
perfectly; and either sells best in market, or makes the best wine, and
which, at the same time, is of good quality, and productive enough.
Your observations, thus taken, will be a better guide than the opinion
of the most skillful grape grower a thousand miles off.

I will now name a few of the most prominent varieties which should at
least be tried by every grape grower.


This grape seems to have given the most general satisfaction all over
the country, and seems to be _the_ "grape for the million." Wherever
heard from, it seems to be uniformly healthy and productive. Our
Eastern friends complain of its inferior quality; this may be owing
partly to their short seasons, and partly to the too early gathering of
the fruit. It is one of those varieties which color early, but should
hang a long time after coloring, to attain its full perfection. Here it
is at least _very_ good; makes an excellent wine, and, if we take into
consideration its enormous productiveness, its vigor and adaptability
to all soils and climates, we must acknowledge that as yet it stands
without a rival, and will be a safe investment almost anywhere. Our
long summers bring it to a perfection of which our Eastern friends have
no idea, until they try it here. It will do well in almost any soil.


This, so far, is the leading grape for red wine, and its reputation
here and in the entire West is now so fully established, that it would
be difficult indeed to persuade our people into the belief, that any
other grape could make a better red wine. It is healthy and uniformly
productive, and will be safe to plant, I think, in nearly all the
Western States. I rather doubt that our Eastern friends will succeed in
making a first class wine from it, as I think their summers are too
short, to develop all its good qualities. Will succeed in almost any
soil, but attains its greatest perfection in southern slopes with
somewhat strong soil.


This is a truly delicious grape, but somewhat tender, and wants a long
season to fully ripen its fruit and bring out all its good qualities.
Will hardly do much further north than we are here, in Missouri, but
is, I think, destined to be one of the leading grapes for the Southern
States. If you have a warm, southern exposure, somewhat stony, with
limestone foundation, plant the Herbemont, and you will not be
disappointed. It is healthy and very productive; more refreshing than
the Delaware, and makes an excellent wine.


Is much recommended by Eastern authorities, and where it succeeds, is
certainly a fine grape and makes a delicious wine. Here at the West, it
has proved a failure in most locations, being subject to leaf-blight,
and a feeble grower. There are some locations, however, where it will
flourish; and whoever is the fortunate possessor of such a one should
not forget to plant it. It seems to flourish best in light, warm,
somewhat sandy soil.


This is immensely productive; of very fair quality here; hardy and
healthy; and if planted for early marketing, will give general
satisfaction. It hangs well to the bunch, and even makes a very fair
wine. Will flourish in almost every soil.


Hardy, healthy and productive; will make a fair wine, but is here not
equal even to the Concord, and far behind the Norton's Virginia in
quality. May be desirable further north.


The distance at which the vines may be planted will of course vary
somewhat with the growth of the different varieties. The rows may all
be six feet apart, as this is the most convenient distance for
cultivating, and gives ample space for a horse and man to pass through
with plough or cultivator. Slow-growing varieties, such as the Delaware
and Catawba, may be planted six feet apart in the rows, making the
distance six feet each way; but the Concord, Norton's Virginia,
Herbemont, Hartford Prolific, Cunningham, and all the strong growers,
will need more room, say ten feet in the rows, so as to give the vines
ample room to spread, and allow free circulation of air--one of the
first conditions of health in the vines, and quality of the fruit.

The next question to be considered is: Shall we plant cuttings or
rooted plants? My preference is decidedly for the latter, for the
following reasons: Cuttings are uncertain, even of those varieties
which grow the most readily; and we cannot expect to have anything like
an even growth, such as we can have if the plants are carefully
assorted. Some of the cuttings will always fail, and there will be gaps
and vacancies which are hard to fill, even if the strongest plants are
taken for replanting. Therefore, let us choose plants.

But we should not only choose rooted plants, but the best we can get;
and these are good one year old, whether grown from cuttings, layers or
single eyes. A good plant should have plenty of strong, well-ripened
roots; not covered with excrescences and warts, which is always a sign
of ill health; but smooth and firm; with well-ripened, short-jointed
wood. They should be of uniform size, as they will then make an even
stand in the vineyard, when not forced by the propagator into an
unnaturally rank growth by artificial manures. This latter
consideration, I think, is very important, as we can hardly expect such
plants, which have been petted and pampered, and fed on rich diet, to
thrive on the every-day fare they will find in the vineyard. Do not
take second or third rate plants, if you can help it; they may live and
grow, but they will never make the growth which a plant of better
quality would make. We may hear of good results sometimes, obtained by
planting second-rate plants, but certainly the results would be better
if better plants had been chosen. Especially important is the selection
of good plants with those varieties which do not propagate and
transplant readily, such as the Norton's Virginia, Delaware, and other
hard-wood varieties. Better pay double the price you would have to give
for inferior plants; the best are the cheapest in the end, as they will
make the healthiest vines, and bear sooner.

But I would also caution my readers against those who will sell you
"extra large layers, for _immediate_ bearing," and whose "plants are
better than those whom anybody else may grow," as their advertisements
will term it. It is time that this humbug should cease; time that the
public in general should know, that they cannot, in nature and reason,
expect any fruit from a plant transplanted the same season; and that
those who pretend it can be done, without vital injury to the plant,
are only seeking to fill their pockets at the cost of their customers.
They know well enough themselves that it cannot be done without killing
or fatally injuring the plant, yet they will impose upon the credulity
of their confiding customers; make them pay from $3 to $5 a piece for a
plant, which these good souls will buy, with a vision of a fine crop of
grapes before their eyes, plant them, with long tops, on which they may
obtain a few sickly bunches of fruit the first season; but if they do
the vines will make a feeble growth, not ripen their fruit, and perhaps
be winter-killed the next season. It is like laying the burden of a
full grown man on the shoulders of a child; what was perhaps no burden
at all to the one, will kill the other. Then, again, these "plants,
superior to those of every one else." It is the duty of every
propagator and nursery-man to raise good plants; he can do it if he
tries; it is for his interest as much as for the interest of his
customers to raise plants of the best quality; and we have no reason to
suppose that we are infinitely superior to our neighbors. While the
first is a downright swindle, the latter is the height of arrogance. If
we had a good deal less of bombast and self laudation, and more of
honesty and fair dealing in the profession, the public would have more
confidence in professional men, and would be more likely to practice
what we preach. Therefore, if you look around for plants, do not go to
those who advertise, "layers for immediate bearing," or "plants of
superior quality to all others grown;" but go to men who have honesty
and modesty enough to send you a sample of their best plants, if
required, and who are not averse to let you see how they grow them.
Choose their good, strong healthy, one year old plants, with strong,
firm, healthy roots, and let those who wish to be humbugged buy the
layers for _immediate_ bearing. You must be content to wait until the
third year for the first crop; but, then, if you have treated your
plants as you ought to do, you can look for a crop that will make your
heart glad to see and gather it. You cannot, in reason and nature
expect it sooner. If your ground has been prepared in the Fall, so much
the better, and if thrown into ridges, so as to elevate the ground
somewhat, where the row is to be, they may be planted in the Fall. The
advantages of Fall planting are as follows: The ground will generally
work better, as we have better weather in the Fall; and generally more
time to spare; the ground can settle among the roots; the roots will
have healed and callused over, and the young plant be ready to start
with full vigor in spring.

[Illustration: FIG. 5.]

Mark your ground, laying it off with a line, and put down a small stick
or peg, eighteen inches long, wherever a plant is to stand. Dig a hole,
about eight to ten inches deep, as shown in Figure 5, in a slanting
direction, raising a small mound in the bottom, of well-pulverized,
mellow earth; then, having pruned your plant as shown in Figure 6, with
its roots and tops shortened in, as shown by the dotted lines, lay it
in, resting the lower end on the mound of earth, spread out its roots
evenly to all sides, and then fill in among the roots with rich,
well-pulverized earth, the upper bud being left above the ground. When
planted in the fall, raise a small mound around your vine, so that the
water will drain off, and throw a handful of straw or any other mulch
on top, to protect it. Of course, the operation should be performed
when the ground is dry enough to be light and mellow, and will readily
work in among the roots.

[Illustration: FIG. 6.]


The first summer after planting nothing is necessary but to keep the
ground free from weeds, and mellow, stirring freely with hoe, rake,
plough, and cultivator, whenever necessary. Should the vines grow
strong they may be tied to the stakes provided in planting, to elevate
them somewhat above the ground. Allow all the laterals to grow, as it
will make the wood stronger and more stocky. They may even be
summer-layered in July, laying down the young cane, and covering the
main stem about an inch deep with mellow soil, leaving the ends of the
laterals out of the ground. With free-growing kinds, such as the
Concord and Hartford Prolific, these will generally root readily, and
make very good plants, the laterals making the stems of the layers.
With varieties that do not root so readily, as the Delaware and
Norton's Virginia, it will seldom be successful, and should not be
practiced. The vineyard may thus be made to pay expenses, and furnish
the vines for further plantations the first year. They are taken up and
divided in the fall, as directed in the chapter for layers. In the
fall, prune the vine to three buds, if strong enough, to one or two if
it has only made a weak growth. A fair growth is from four to five feet
the first summer. During the winter, trellis should be provided for the
vines, as we may expect them to grow from twelve to fifteen feet the
coming summer. The cheapest and most economical are those of strong
upright posts, say four inches in diameter, made of red cedar if it can
be had, if not, of any good, durable timber--mulberry, locust, or white
oak--and seven feet long, along which No. 10 wire is stretched
horizontally. Make the holes for the posts with a post-hole auger, two
feet deep; set in the posts, charred on one end, to make them durable.
If wire is to be used, one post every sixteen feet will be enough, with
a smaller stake between, to serve as a support for the wires. Now
stretch your wire, the lowest one about two feet from the ground, the
second one eighteen inches above it, and the third eighteen inches
above the second. The wires may be fastened to the posts by nails,
around which they can be twisted, or by loops of wire driven into the
post. Where timber is plenty, laths made of black oak may be made to
serve the same purpose; but the posts must then be set much closer, and
the wire will be the cheapest and neatest in the end. A good many
grape-growers train their vines to stakes, believing it to be cheaper,
but I have found it more expensive than trellis made in the above
manner, and it is certainly a very slovenly method, compared with the
latter. Trellis is much more convenient for tying the vines, the canes
can be distributed much more evenly, and the fruit and young wood, not
being huddled and crowded together as on stakes, will ripen much more
evenly, and be of better quality, as the air and sun have free access
to it.


We find the young vine at the commencement of this season pruned to
three buds of the last season's growth. From these we may expect from
two to three strong shoots or canes. Our first work will be to
cultivate the whole ground, say from four to six inches deep, ploughing
between the rows, and hoeing around the vines with a two-pronged German
hoe, or _karst_. Figure 7 shows one of these implements, of the best
form for that purpose. The ground should be completely inverted, but
never do it in wet weather, as this will make the ground hard and

[Illustration: FIG. 7.]

Of the young shoots, if there are three, leave only the two strongest,
tying the best of them neatly to the trellis with bass, or pawpaw bark,
or rye straw. If a Catawba or Delaware, you may let them grow
unchecked, tying them along the uppermost wire, when they have grown
above it. The Concord, Herbemont, Norton's Virginia, and other
strong-growing varieties, I treat in the following manner: When the
young shoot has reached the second wire I pinch off its leader. This
has the tendency to force the laterals into stronger growth, each
forming a medium-sized cane. On these we intend to grow our fruit the
coming season, as the buds on these laterals will generally produce
more and finer fruit than the buds on the strong canes. Figure 8 will
show the manner of training the second summer, with one cane layered,
for the purpose of raising plants. This is done as described before;
only, as the vine will make a much stronger growth this season than the
first, the layering maybe done in June, as soon as the young shoots are
strong enough. Figure 9 shows the vine pruned and tied, at the end of
the second season. Figure 10 illustrates the manner of training and
tying the Catawba or Delaware.

[Illustration: FIG. 8.    FIG. 9.]

[Illustration: FIG. 10.]

The above is a combination of the single cane and bow system, and the
horizontal arm training, which I first tried on the Concord from sheer
necessity; when the results pleased me so much that I have adopted it
with all strong-growing varieties. The circumstances which led me to
the trial of this method were as follows: In the summer of 1862, when
my Concord vines were making their second season's growth, we had, in
the beginning of June, the most destructive hail storm I have ever seen
here. Every leaf was cut from the vines, and the young succulent shoots
were all cut off to about three to three and a half feet above the
ground. The vines, being young and vigorous, pushed out the laterals
vigorously, each of them making a fair-sized cane. In the fall, when I
came to prune them, the main cane was not long enough, and I merely
shortened in the laterals to from four to six buds each. On these I had
as fine a crop of grapes as I ever saw, fine, large, well-developed
bunches and berries, and a great many of them, as each had produced its
fruit-bearing shoot. Since that time I have followed this method
altogether, and obtained the most satisfactory results.

The ground should be kept even and mellow during the summer, and the
vines neatly tied to the trellis with bast or straw.

There are many other methods of training; for instance, the old bow and
stake training, which is followed to a great extent around Cincinnati,
and was followed to some extent here. But it crowds the whole mass of
fruit and leaves together so closely that mildew and rot will follow
almost as a natural consequence, and those who follow it are almost
ready to give up grape-culture in despair. Nor is this surprising. With
their tenacious adherence to so fickle a variety as the Catawba, and to
practices and methods of which experience ought to have taught them the
utter impracticability long ago, we need not be surprised that
grape-culture is with them a failure. We have a class of grape-growers
who never learn, nor ever forget, anything; these we cannot expect
should prosper. The grape-grower, of all others, should be a close
observer of nature in her various moods, a thinking and a reasoning
being; he should be trying and experimenting all the time, and be ready
always to throw aside his old methods, should he find that another will
more fully meet the wants of his plants. Only thus can he expect to

There is also the arm system, of which we hear so much now-a-days, and
which certainly looks very pretty _on paper_. But paper is patient, and
while it cannot be denied that it has its advantages, if every spur and
shoot could be made to grow just as represented in drawings, with three
fine bunches to each shoot; yet, upon applying it practically, we find
that vines are stubborn, and some shoots will outgrow others; and
before we hardly know how, the whole beautiful system is out of order.
It may do to follow in gardens, on arbors and walls, with a few vines,
but I do not think that it will ever be successfully followed in
vineyard culture for a number of years, as it involves too much labor
in tying up, pruning, etc. I think the method described above will more
fully meet the wants of the vinyardist than any I have yet seen tried;
it is so simple that every intelligent person can soon become familiar
with it, and it gives us new, healthy wood for bearing every season.
Pruning may be done in the fall, as soon as the leaves have dropped.


At the commencement of the third season, we find our vine pruned to two
spurs of two eyes each, and four lateral canes, of from four to six
eyes each. These are tied firmly to the trellis as shown in Figure 12,
for which purpose small twigs of willows (especially the golden willow,
of which every grape-grower should plant a supply) are the most
convenient. The ground is ploughed and hoed deeply, as described
before, taking care, however, not to plough so deep as to cut or tear
the roots of the vine.

Our vines being tied, ploughed, and hoed, we come to one of the most
important and delicate operations to be performed; one of as
great--nay, greater--importance than pruning. I mean summer-pruning, or
pinching, _i.e._ thumb or finger pruning. Fall-pruning, or cutting
back, is but the beginning of the discipline under which we intend to
keep our vines; summer-pruning is the continuation, and one is useless,
and cannot be followed systematically without the other.

Let us look at our vine well, before we begin, and commence near the
ground. The time to perform the first summer-pruning is when the young
shoots are about six to eight inches long, and when you can see plainly
all the small bunches or buttons--the embryo fruit. We commence on the
lower two spurs, having two buds each. From these two shoots have
started. One of them we intend for a bearing cane next summer;
therefore allow it to grow unchecked for the present, tying it, if long
enough, to the lowest wire. The other, which we intend for a spur again
next fall, we pinch with thumb and finger to just beyond the last bunch
or button, taking out the leader between the last bunch and the next
leaf, as shown in Figure 11, the cross line indicating where the leader
is to be pinched off. We now come to the next spur, on the opposite
side, where we also leave one cane to grow unchecked, and pinch off the
other. We now go over all the shoots coming from the arms or laterals
tied to the trellis, and also pinch them beyond the last bunch. Should
any of the buds have pushed out two shoots, we rub off the weakest; we
also take off all barren or weak shoots. If any of them are not
sufficiently developed we pass them over, and go over the vines again,
in a few days after the first pinching.

[Illustration: FIG. 11.]

This early pinching of the shoot has a tendency to throw all the vigor
into the development of the young bunch, and the leaves remaining on
the shoot, which now grow with astonishing rapidity. It is a gentle
checking, and leading the sap into other channels; not the violent
process which is often followed long after the bloom, when the wood has
become so hardened that it must be cut with a knife, and by which the
plant is robbed of a large quantity of its leaves, to the injury of
both fruit and vine. Let any of my readers, who wish to satisfy
themselves, summer-prune a vine, according to the method described
here, and leave the next vine until after the bloom, and he will
plainly perceive the difference. The merit of first having practised
this method here, which I consider one of vast importance in
grape-culture, belongs to Mr. WILLIAM POESCHEL, of this place, who was
led to do so, by observing the rapid development of the young bunches
on a shoot which had accidentally been broken beyond the last bunch.
Now, there is hardly an intelligent grape-grower here, who does not
follow it; and I think it has added more than one-third to the quantity
and quality of my crop. It also gives a chance to destroy the small,
white worm, a species of leaf-folder, which is very troublesome just at
this time, eating the young fruit and leaves, and which makes its web
among the tender leaves at the end of the shoot.

The bearing branches having all been pinched back, we can leave our
vines alone until after the bloom, only tying up the young canes from
the spurs, should it become necessary. But do not tie them over the
bearing canes, but lead them to the empty space on both sides of the
vine; as our object must be to give the fruit all the air and light we

By the time the grapes have bloomed, the laterals will have pushed from
the axils of the leaves on the bearing shoots. Now go over these again,
and pinch each lateral back to one leaf, as shown in Figure 12. This
will make the leaf which remains grow and expand rapidly, serving at
the same time as a conductor of sap to the young bunch opposite, and
shading it when it becomes fully developed. The canes from the spurs,
which we left unchecked, and which we design to bear fruit the next
season, may now also be stopped or pinched, when they are about three
feet long, to start their laterals into stronger growth. Pinch off all
the tendrils; this is a very busy time for the vine-dresser, and upon
his close attendance and diligence now, depends, in a great measure,
the value of his crop. Besides, "a stitch in time saves nine," and he
can save an incredible amount of labor by doing everything at the
proper time.

[Illustration: FIG. 12.]

In a short time, the laterals on the fruit-bearing branches which have
been pinched will throw out suckers again. These are stopped again,
leaving one leaf of the young growth. Leave the laterals on the canes
intended for next years' fruiting to grow unchecked, tying them neatly
with bass, or pawpaw bark, or with rye straw.

This is about all that is necessary for this summer, except an
occasional tying up of a fruiting branch, should its burden become more
than it can bear. But the majority of the branches will be able to
sustain their fruit without tying, and the young growth which may yet
start from the laterals may be left unchecked, as it will serve to
shade the fruit when ripening. Of course, the soil must be kept clean
and mellow, as in the former summer. This short pruning is also a
partial preventative against mildew and rot, and the last extremely wet
season has again shown the importance of letting in light and air to
all parts of the vine; as those vineyards, where a strict system of
early summer pruning had been followed, did not suffer half as much
from rot and mildew as those where the old slovenly method still

My readers will perceive, that Fall-pruning, or shortening-in the
ripened wood of the vine, and summer-pruning, shortening in and
thinning out the young growth, have one and all the same object in
view, namely, to keep the vine within proper bounds, and concentrate
all its energies for a two-fold object, namely, the production and
ripening of the most perfect fruit, and the production of strong,
healthy wood for the coming season's crop. Both operations are, in
fact, only different parts of one and the same system, of which
summer-pruning is the preparatory, and fall pruning the finishing part.

If we think that a vine is setting more fruit than it is able to bear
and ripen perfectly, we have it in our power to thin it, by taking away
all imperfect bunches, and feeble shoots. We should allow no more wood
to grow than we need for next season's bearing; if we allow three canes
to grow where only two are needed, we waste the energies of the vine,
which should all be concentrated upon ripening its fruit in the most
perfect condition, and producing the necessary wood for next season's
bearing, and that of the best and most vigorous quality, but no more.
If we prune the vine too long, we over-tax its energies; making it bear
more fruit than it can perfect, and the result will be poor,
badly-ripened fruit, and small and imperfect wood. If, on the contrary,
we prune the vine too short, we will have a rank, excessive growth of
wood and leaves, and encourage rot and mildew. Only practice and
experience will teach us the exact medium, and the observing vintner
will soon find out where he has been wrong, better than he can be
taught by a hundred pages of elaborate advice. Different varieties will
require different treatment, and it would be foolishness to suppose
that two varieties so entirely different, as for instance, the Concord
and the Delaware, could be pruned, trained and pinched in the same
manner. The first, being a rank and vigorous grower, with long joints,
will require much longer pruning than the latter, which is a
slow-growing, short-jointed vine. Some varieties, the Taylor for
instance, also the Norton, will fruit better if pruned to spurs on old
wood, than on the young canes; it will therefore be the best policy for
the vintner in pruning these, to retain the old arms or canes, pruning
all the healthy, strong shoots they have to two buds, as long as the
old arms remain healthy; always, however, growing a young cane to fall
back upon, should the old arm become diseased; whereas, the Catawba and
Delaware, being only moderate growers, will flourish and bear best when
pruned short, and to a cane of last season's growth. The Concord and
Herbemont, again, will bear best on the laterals of last season's
growth, and should be trained accordingly. Therefore it is, because
only a few of the common laborers will take the pains to think and
observe closely, that we find among them but few good vine-dressers.

At the end of this season, we find our Concords or Herbemonts, with the
old fruit-bearing cane, and a spur on each side, from which have grown
two canes; one of which was stopped, like all other fruit-bearing
branches, and which we now prune to a spur of two eyes; and another,
which was stopped at about three feet, and on which the laterals were
allowed to grow unchecked. We therefore have one of these canes, with
its laterals, on each side of the vine. These laterals are now pruned
precisely as the last season, each being cut back to from four to six
eyes, and the old cane, which has borne fruit, is cut away altogether.
With Norton's Virginia, Taylor, and some others, which will bear more
readily on spurs from old wood, the old cane is retained, provided the
shoots on it are sound and healthy, with well developed buds; the weak
ones are cut away altogether, and the others cut back to two eyes each.
One of the canes is pruned, as in the Concord, to be tied to one side
of the trellis, the next spring. This closes our summer and fall
pruning for the third year. Of the gathering of the fruit, as well for
market as for wine, I shall speak in another chapter.


We may now consider the vine as established, able to bear a full crop,
and when tied to the trellis in spring, to present the appearance, as
shown in Fig. 13. The operations to be performed are precisely the same
as in its third year.

[Illustration: FIG. 13.]

In addition, I will here remark, that in wet seasons the soil of the
vineyard should be stirred as little as possible, as it will bake and
clog, and in dry seasons it should be deeply worked and stirred, as
this loose surface-soil will retain moisture much better than a hard
surface. Should the vines show a decrease in vigor, they may be manured
with ashes or compost, or still better, with surface-soil from the
woods. This will serve to replenish the soil which may have been washed
off and is much more beneficial than stable manure. When the latter is
applied, a small trench should be dug just above the vine, the manure
laid in, and covered with soil. But an abundance of fresh soil, drawn
up well around the vine, is certainly the best of all manures.

Where a vine has failed to grow the first season, replant with extra
strong vines, as they will find it difficult to catch up with the
others; or the vacancy can be filled up the next season, by a layer
from a neighboring vine, made in the following manner: Dig a trench
from the vine to the empty place, about eight to ten inches deep, and
bend into it one of the canes of the vine, left to grow unchecked for
that purpose, and pruned to the proper length. Let the end of it come
out to the surface of the ground with one or two eyes above it, at the
place where the vine is to be, and fill up with good, well pulverized
earth. It will strike roots at almost every joint, and grow rapidly,
but, as it takes a good deal of nourishment from the parent vine, that
must be pruned much shorter the first year. When the layer has become
well established, it is cut from the parent vine; generally the second

Pruning is best done in the fall, but it can be done on mild days all
through the winter months, even as late as the middle of March.
Fall-pruning will prevent all flow of sap, and the cuttings are also
better if made in the fall, and buried in the ground during winter. All
the sound, well-ripened wood of last season's growth may be made into
cuttings, which may be either planted, as directed in a former chapter,
or sold; and are an accession to the product of the vineyard not to be
despised, for they will generally defray all expenses of cultivation.


This is altogether different from the treatment in vineyards; the first
has for its object to grow the most perfect fruit, and to bring the
vine, with all its parts, within the easy reach and control of the
operator; in the latter, our object is to cover a large space with
foliage, for ornament and shade, fruit being but a secondary
consideration. However, if the vine is treated judiciously, it will
also produce a large quantity of fruit, although not of as good quality
as in the vineyard.

[Illustration: FIG. 14.    FIG. 15.]

Our first object must be to grow very strong plants, to cover a very
large space. Prepare a border by digging a trench two feet deep and
four feet wide. Fill with rich soil, decomposed leaves, burnt bones,
ashes, etc. Into this plant the strongest plants you have, pruned as
for vineyard planting. Leave but one shoot to grow on them during the
first summer, which, if properly treated, will get very strong. Cut
back to three buds the coming fall. These will each throw out a strong
shoot, which should be tied to the arbor they are designed to cover, as
shown in Figure 14, and allowed to grow unchecked. In the fall
following cut each shoot back to three buds, as our first object must
be to get a good basis for our vines. These will give us nine canes the
third summer; and as the vine is now thoroughly established and strong,
we can begin to work in good earnest. It will be perceived that the
vine has three different sections or principal branches, each with
three canes. Cut one of these back to two eyes, and the other two to
six or eight buds each, according to the strength of the vine, as shown
in Figure 15. The next spring tie these neatly to the trellis, and when
the young shoots appear thin out the weakest, and leave the others to
grow unchecked. The next fall cut back as indicated by the black cross
lines, the weakest to be cut back to one or two eyes, and the stronger
ones to three or four, the spurs at the bottom to come in as a reserve,
should any of the branches become diseased. Figure 16 shows the manner
of pruning.

[Illustration: FIG. 16.]

In this manner a vine can be made, in course of time, to cover a large
space, and get very old. The great vine at Windsor Palace was planted
more than sixty years ago, and in 1850 it produced two thousand large
bunches of magnificent grapes. The space covered by the branches was
one hundred and thirty-eight feet long, and sixteen feet wide, and it
had a stem two feet nine inches in circumference. This is one of the
largest vines on record. They should, however, be strongly manured to
come to full perfection.

Other authorities prefer the Thomery system of training, but I think it
much more complicated and difficult to follow. Those wishing to follow
it will find full directions in DR. GRANT'S and FULLER'S books, which
are very explicit on this method.


There are many other systems in vogue among vine-dressers in Germany
and France, but as our native grapes are so much stronger in growth,
and are in this climate so much more subject to mildew and rot, I think
these methods, upon the whole, but poorly adapted to the wants of our
native grapes, however judicious they may be there. I will only mention
a few of them here; one because it is to a great extent followed in
Mexico and California, and seems to suit that dry climate and arid soil
very well; and the other, because it will often serve as a pretty
border to beds in gardens. The first is the so-called buck or stool
method of training. The vine is made to form its head--_i.e._, the part
from which the branches start--about a foot above the ground, and all
the young shoots are allowed to grow, but summer-pruned or checked just
beyond the last bunch of grapes. The next spring all of the young
shoots are cut back to two eyes, and this system of "spurring in" is
kept up, and the vine will in time present the appearance of a bush or
miniature tree, producing all its fruit within a foot from the head,
and without further support than its own stem. Very old vines trained
in this manner often have twenty to twenty-five spurs, and present,
with their fruit all hanging in masses around the main trunk, a
pleasing but rather odd aspect. This method could not be applied here
with any chance of success only to those varieties which are slow
growers, and at the same time very hardy. The Delaware would perhaps be
the most suitable of all varieties I know for a trial of this method;
such strong growers as the Concord and Norton's Virginia could never be
kept within the proper bounds, and it would be useless to try it on
them. It might be of advantage on poor soil, where there is at the same
time a scarcity of timber. Figure 17 shows an old vine pruned after
this method.

[Illustration: FIG. 17.]

The other method of dwarfing the grape is practiced to make a pretty
border along walks in gardens, and is as follows: Plant your vines
about eight feet apart; treat them the first season as in common
vineyard planting, but at the end of the first season cut back to two
eyes. Now provide posts, three to three and a half feet long; drive
them into the ground about eighteen inches to two feet, which can be
easily done if they are pointed at one end, and nail a lath on top of
them. This is your trellis for the vines, and should be about eighteen
inches above the ground when ready. Now allow both shoots which will
start from the two buds to grow unchecked; and when they have grown
above the trellis, tie one down to the right, the other to the left,
allowing them to ramble at will along it. The next fall they are each
cut back to the proper length, to meet the next vine, and in spring
tied firmly to the lath, as shown in Figure 18. When the young shoots
appear, all below the trellis are rubbed off, but all those above the
trellis are summer-pruned or pinched immediately beyond the last bunch
of grapes, as in vineyard culture, and the trellis, with its garland of
fruit, will present a very pretty appearance throughout the summer. In
the fall all of these shoots are pruned to one bud, from which will
grow the fruit-bearing shoot for the next season, as shown in Figure
19; and the same treatment is repeated during the summer and fall.

[Illustration: FIG. 18.    FIG. 19.]


I cannot agree with Mr. FULLER that the diseases of the vine are not
formidable in this country. They are so formidable that they threaten
to destroy some varieties altogether; and the Catawba, once the glory
and pride of the Ohio vineyards, has for the last fifteen years
suffered so much from them, that many of the grape-growers who are too
narrow-minded to try anything else are about giving up grape-growing in

It is very fortunate, therefore, that we have varieties which do not
suffer from these diseases, or only in a very slight degree; and my
advice to the beginner in grape-culture would be, "not to plant largely
of any variety which is subject to disease." Men may talk about
sulphuring, and dusting their vines with sulphur through bellows; but I
would rather have vines which will bear a good crop without these windy
appliances. We can certainly find some varieties for _every_ locality
which do not need them, and these we should plant.

The mildew is our most formidable disease, and will very often sweep
away two-thirds of a crop of Catawbas in a few days. It generally
appears here from the first to the fifteenth of June, after abundant
rains, and damp, warm weather. It seems to be a parasitic fungus, and
sulphur applied by means of a bellows, or dusted over the fruit and
vine is said to be a partial remedy. Close and early summer-pruning
will do much to prevent it, throwing, as it does, all the strength of
the vine into the young fruit, developing it rapidly, and also allowing
free circulation of air. In some varieties--for instance, the
Delaware--it will only affect the leaves, causing them to blight and
drop off, after which the fruit, although it may attain full size, will
not ripen nor become sweet, but wither and drop off prematurely. In
seasons when the weather is dry and the air pure, it will not appear.
It is most prevalent in locations which have a tenacious subsoil, and
under-draining will very likely prove a partial preventive, as excess
of moisture about the roots is no doubt one of its causes.

The gray rot, or so-called grape cholera, generally follows the mildew,
and I think that the latter is the principal cause of it, as I have
generally found it on berries whose stems have been injured by the
mildew. The berry first shows a sort of gray marbling; in a day or two
it turns to a grayish-blue color, and finally withers and drops from
the bunch. It will continue to affect berries until they begin to
color, but only attack a few varieties--the Catawba, To Kalon,
Kingsessing, and sometimes the Diana.

The spotted, or brown rot, will also attack many of our varieties; it
is very destructive to the Isabella and Catawba, and even the Concord
is not quite free from it. But it is, after all, not very destructive,
and not half as dangerous as the mildew or gray rot.

Early and close summer-pruning is a partial preventative against all
these diseases, as it will hasten the development of the fruit, allow
free circulation of air, and the young leaves which appear on the
laterals after pinching seem to be better able to withstand the effects
of the mildew, often remaining fresh and green, and shading the fruit,
when the first growth of leaves have already dropped.

But "an ounce of prevention is better than a pound of cure," and our
best preventive is to plant none but healthy varieties. A grape,
however good it may be in quality, is not fit for general cultivation
if seriously affected with any of these diseases. Nothing can be more
discouraging to the grape-grower than to see his vines one day rich in
the promise of an abundant crop, and a few days afterwards see
two-thirds or three-fourths swept away by disease. It is because I have
so often felt this bitter disappointment, that I would warn my readers
against planting varieties subject to them. I would save _them_ from
the discouragement and bitter losses which I have experienced, when it
was out of my power to prevent it. They _can_ prevent it, for the
grape-growing of to-day is no longer the same uncertain occupation it
was ten years ago. We of to-day have our choice of varieties not
subject to disease; let us make it judiciously, and we may be sure of a
paying crop every year.


The grape has many enemies of this kind, but if they are closely
watched from the beginning their ravages are easily kept within proper

The common gray cut-worm will often eat the young tender shoots of the
vine, and draw them into the ground below. Wherever this is perceived
the rascal can easily be found by digging for him under some of the
loose clods of ground below the vine, and should be destroyed without

[Illustration: FIG. 20.
DELAWARE.--_Berries 1/2 diameter_.]

Small worms, belonging to the family of leaf-folders, some of them
whitish gray, some bluish green, will in spring make their webs among
the young, downy leaves at the end of the shoots, eating the young
bunches or buttons, and the leaves. These can be destroyed when summer
pruning for the first time. Look close for them, as they are very
small; yet very destructive if let alone.

A small, gray beetle, of about the size and color of a hemp-seed, will
often eat a hole into the bud, when it is just swelling, and thus
destroy it. He is very shy, and will drop from the vine as soon as you
come near him. It is a good plan to spread a newspaper under the vine,
and then shake it, when he will drop on the paper and can be caught.

Another bug, of about the size of a fly, gray, with round black specks,
will sometimes pay us a visit. They will come in swarms, and eat the
upper side of the leaves, leaving only the skeletons. They are very
destructive, devouring every leaf, as far as they go; they can also be
shaken off on a paper or sheet spread under the vine.

The thrip, a small, rather three-cornered, whitish-green insect, has of
late been very troublesome, as they eat the under side of the leaves of
some varieties, especially the Delaware and Norton's Virginia, when the
leaf will show rusty specks on the surface, and finally drop off. It
has been recommended to go through the vineyard at night, one man
carrying a lighted torch, and the other beating the vines, when they
will fly into the flame, and be burnt. They are a great annoyance, and
have defoliated whole vineyards here last fall.

Another leaf-folder makes his appearance about mid-summer, making its
web on the leaf, drawing it together, and then devouring his own house.
It is a small, greenish, and very active worm, who, if he "smells a
rat," will drop out of his web, and descend to the ground in
double-quick time. I know of no other plan, than to catch him and crush
his web between the finger and thumb.

The aphis, or plant louse, often covers the young shoots of the vine,
sucking its juices. When a shoot is attacked by them, it will be best
to take it off and crush them under your feet, as the shoot is apt to
be sickly afterwards, any way.

The grape vine sphynx will be found occasionally. It is a large, green
worm, with black dots, and very voracious. Fortunately, it is not
numerous, and can easily be found and destroyed.

There are also several caterpillars--the yellow bear, the hog
caterpillar, and the blue caterpillar, which will feed upon the leaves.
The only remedy I know against them is hand picking, but they have not
as yet been very numerous, nor very destructive.

Wasps are sometimes very troublesome when the fruit ripens, stinging
the berries and sucking the juice. A great many can be caught by
hanging up bottles, with a little molasses, which they will enter, and
get stuck in the molasses.


These are sometimes very troublesome at the time of ripening, and
especially the oriole is a "hard customer," as he will generally dip
his bill into every berry; often ruining a fine bunch, or a number of
them, in a short time. I have therefore been compelled to wage a war
upon some of the feathered tribe, although they are my especial
favorites, and I cannot see a bird's nest robbed. However, there are
some who do not visit the vineyard, except for the purpose of
destroying our grapes, and these can not complain if we "won't stand it
any longer," but take the gun, and retaliate on them. The oriole, the
red bird, thrush, and cat bird are among the number, and although I
would like to spare the latter three, in thankful remembrance of many a
gratuitous concert, the first must take his chance of powder and lead,
for the little rascal is too aggravating. A few dry bushes, raised
above the trellis will serve as their resting place before they
commence their work of destruction, where they can be easily killed.


Although our winters are seldom severe enough to destroy the hardy
varieties, yet they will often fatally injure such half hardy varieties
as the Herbemont and Cunningham, and the severe winter of 1863,-'64,
killed even the Catawba, down to the snow line, and severely injured
the Norton's Virginia, and even the Concord. Fortunately, such winters
occur but rarely, and even in localities where the vines are often
destroyed by the severe cold in winter, this should deter no one from
growing grapes, as, with very little extra labor he can protect them,
and bring them safely through the winter. I always cover my tender
varieties, in fact, all that I feel not quite safe to leave out, even
in severe winters, in the following manner: The vines are properly
pruned in the fall; then select a somewhat rainy day, when the canes
will bend more easily. One man goes through the rows, and bends the
canes to the ground along the trellis, while another follows with the
spade, and throws earth enough on them to hold them in their places.
Afterwards, I run a plough through the rows, and cover them up
completely. In the spring when all danger from frost is over, I take a
so-called spading fork, and lift the vines. The entire cost of covering
an acre of grape vines and taking them up again in spring, will not
exceed $10; surely a trifling expense, if we can thereby ensure a full

We have thus a protection against the cold in winter, but I know none
against early frosts, in fall, and late spring frosts; and the grape
grower should therefore avoid all localities where they are prevalent.
The immediate neighborhood of large streams, or lakes, will generally
save the grape grower from their disastrous influence; and our summers,
here, along the banks of the Missouri river, are in reality full two
months longer than they are in the low, small valleys, only four to six
miles off. Let the grape grower, in choosing a locality, look well to
this, and avoid the hills along these narrow valleys. Either choose a
location sufficiently elevated, to be beyond their influence, or, what
is better still, choose it on the bluffs above our large streams; where
the atmosphere, even in the heat of summer, will never become too dry
for the health of the vine. It is a sad spectacle to see the hopes of a
whole summer frustrated by one cold night; to see the vines which
promised an abundant crop but the day before, browned and wilted beyond
all hopes of recovery, and the cheerless prospect before you, that it
may occur every spring; or to see the finest crop of grapes, when just
ripening, scorched and wilted by just one night's frost, fit for
nothing but vinegar. Therefore, look well to this, when you choose the
site of your vineyard, and rather pay five times the price for a
location free from frost, than for the richest farm along the so-called
creek bottoms, or worse still, sloughs of stagnant water.


The practice of girdling to induce early ripening is supposed to have
been invented by Col. BUCHATT, of Metz, in 1745. He claimed for it that
it would also greatly improve the quality of the fruit, as well as
hasten maturity. That it accomplishes the latter, cannot be denied; it
also seems to increase the size of the berries, but I hardly think the
fruit can compare in flavor with a well developed bunch, ripened in the
natural way. As it may be of practical value to those who grow grapes
for the market, enabling them to supply their customers a week earlier
at least, and also make the fruit look better, and be of interest to
the amateur cultivator, I will describe the operation for their

[Illustration: FIG. 21
NORTON'S VIRGINIA--_Berries 1/3 diameter._]

It can be performed either on wood of the same season's growth, or on
that of last year, but in any case only upon such as can be pruned away
the next fall. If you desire to affect the fruit of a whole arm or
cane, cut away a ring of bark by passing your knife all around it, and
making another incision from a quarter to half an inch above the first,
taking out the intermediate piece of bark clean, down to the wood. It
should be performed immediately after the fruit is set. The bunches of
fruit above the incision will become larger, and the fruit ripen and
color finely, from a week to ten days before the fruit on the other
canes. Of course, the cane thus girdled, cannot be used for the next
season, and must be cut away entirely. The result seems to be the
consequence of an obstruction to the downward flow of the sap, which
then develops the fruit much faster.

Ripening can also be hastened by planting against the south side of a
wall or board fence, when the reflection of the rays of the sun will
create a greater degree of warmth.

But nothing can be so absurd and unnatural than the practice of some,
who will take away the leaves from the fruit, to hasten its ripening.
The leaves are the lungs of the plants; the conductors and elevators of
sap; and nothing can be more injurious than to take them away from the
fruit at the very time when they are most needed. The consequence of
such an unwise course will be the wilting and withering of the bunches,
and, should they ripen at all, they will be deficient in flavor. Good
fruit must ripen _in the shade_, only thus will it attain its full

Another practice very injurious to the vines is still in practice in
some vineyards, and cannot be too strongly condemned. It is the
so-called "cutting in" of the young growth in August. Those who
practice it, seem to labor under the misapprehension that the young
canes, after they have reached the top of the trellis, and are of the
proper length and strength for their next year's crop, do not need that
part of the young growth beyond these limits any more, and that all the
surplus growth is "of evil." Under the influence of this idea they arm
themselves with a villainous looking thing called a bill-hook, and cut
and slash away at the young growth unmercifully, taking away one-half
of the leaves and young wood at one fell swoop. The consequence is a
stagnation of sap: the wood they have left, cannot, and ought not to
ripen perfectly, and if anything like a cold winter follows, the vines
will either be killed entirely, or very much injured at least. The
intelligent vine dresser will tie his young canes, away from the
bearing wood as much as he can, to give the fruit the fullest
ventilation; but when they have reached the top of the trellis, tie
them along it and let them ramble as they please. They will thus form a
natural roof over the fruit, keep off all injurious dews, and shade the
grapes from above. There is nothing more pleasing to the eye than a
vineyard in September, with its wealth of dark green foliage above, and
its purple clusters of fruit beneath, coyly peeping from under their
leafy covering. Such grapes will have an exquisite bloom, and color, as
well as thin skin and rich flavor, which those hanging in the scorching
rays of the sun can never attain.


As remarked before, this will seldom be necessary, if the vintner is
careful enough to guard against washing of the top-soil, and to turn
under all leaves, etc., with the plow in the Fall. The best manure is
undoubtedly fresh surface soil from the woods. Should the vines,
however, show a material decrease in vigor, it may become necessary to
use a top-dressing of decomposed leaves, ashes, bone-dust, charcoal,
etc. Fresh stable-yard manure I would consider the last, and only to be
used when nothing better can be obtained. Turn under with the plow, as
soon as the manure is spread. Nothing, I think, is more injurious than
the continual drenching with slops, dish-water, etc., which some good
souls of housewives are fond of bestowing on their pet grape vines in
the garden. It creates a rank, unwholesome growth, and will cause
mildew and rot, if anything can.


This will sometimes be necessary, to more fully develop the bunches.
The best thinning is the reduction of the number of bunches at the time
of the first summer pruning. If a vine shows more fruit, than the vine
dresser thinks it can well ripen, take away all weak and imperfect
shoots, and also all the small and imperfect bunches. If the number of
bunches on the fruit bearing branches is reduced to two on each, it
will be no injury, but make the remaining number of bunches so much
more perfect. Thinning out the berries on the bunches, although it will
serve to make the remaining berries more perfect and larger, is still a
very laborious process, and will hardly be followed to any extent in
vineyards, although it can well be practised on the few pet vines of
the amateur, and will certainly heighten the beauty of the bunches and


Should a vine become old and feeble, it can be renewed by layering. The
vine is prepared in the following manner: Prune all the old wood away,
leaving but one of the most vigorous of your canes; then dig a trench
from the vine along the trellis, say three feet long, eight inches
deep; into this bend down the old vine, stump, head and all, fastening
it down with a strong hook, if necessary, letting the end of the young
cane come out about three eyes above the ground, and fill up with rich,
well pulverized soil. The vine will make new roots at every joint, and
become vigorous, and, so to say, young, again. Some recommend this
process for young vines, the first year after planting; but if good
plants have been chosen and planted, it will not be necessary. Feeble
and poor plants may need this process, but if plants have good strong
roots when planted, (and _only_ such should be planted when they can be
obtained), they will not be benefited by it.


_Pruning Shears._ These are very handy, and with them the work can be
done quicker, and with less labor, as but a slight pressure of the hand
will cut a strong vine. Fig. 22 will show the shape of one for heavy
pruning. They are made by J. T. HENRY, Hampden, Connecticut, and can be
had in almost all hardware stores. The springs should be of brass, as
steel springs are very apt to break. A much lighter and smaller kind,
with but one spring, is very convenient for gathering grapes, as it
will cut the stem easily and smoothly, and not shake the vine, as
cutting with the knife will do. They are also handy to clip out unripe
and rotten berries, and should be generally used instead of knives.

[Illustration: FIG. 22]

_Pruning Saws._ It will sometimes be necessary to use these, to cut out
old stumps, etc., although, if a vine is well managed, it will seldom
be necessary. Fig. 23 will show a kind which is very convenient for the
purpose, and will also serve for orchard pruning; the blade is narrow,
connected with the handle, and can be turned in any direction.

[Illustration: FIG. 23.]


In this, the vineyardist, of course, only aims at profit, and for that
purpose the grapes are often gathered when they are hardly
colored--long before they are really ripe--because the public will
generally buy them at a high price. Let us hope, however, that better
taste will in time prevail, and that even a majority of the public will
learn to appreciate the difference between ripe and unripe fruit. I
would advise my readers at least to wait until the fruit is fully and
evenly colored; for it is our duty to do all we can to correct this
vicious leaning towards swallowing unripe fruit, which is so prevalent
in this nation, and the producer will not lose anything either, because
his fruit will look much better, it will therefore bring the same price
which half ripened fruit would have brought, even a week sooner, and
will weigh heavier. Every grape will generally color full two weeks
before it is fully ripe; and as they are one of the fruits that will
not ripen _after_ they are gathered, they will shrivel and look
indifferent if gathered before.

To ship them to market any distance, they should be packed in low,
shallow boxes, say six inches high, so that they will hold about two
layers of grapes. Cut the branches carefully, with as long a stem as
possible, for more convenient handling, taking care to preserve all the
bloom, and clipping out all the unripe berries. They are generally
weighed in the basket before packing. Now put a layer of vine leaves on
the bottom of the box; then make a layer of grapes, laying them as
close as possible; then put a layer of leaves over them; on them put
another layer of grapes, filling up evenly; then spread leaves rather
thickly over them, and nail on the cover. The box should be perforated
with holes, to admit some air. The grapes must be perfectly dry when
gathered, and the box should be well filled to prevent shaking and


For this purpose, the fruit must be thoroughly ripe. When fully ripe,
the stem will turn brown, and shrivel somewhat. The fruit is then
carefully gathered, and laid upon a dry floor, or shelves, for a day or
two, so that some of the moisture will evaporate. They can then be
packed in boxes, in about the same manner as described before, but
paper will be better than leaves for this purpose. They are then put
away on shelves, in an airy room, which must, however, be free from
frost, in an even temperature of from 30° to 40°. They should be
examined from time to time, and the decayed berries taken out. They may
thus be kept for several months.


For this purpose, the grapes should hang as long as it is safe to allow
them; for it will make a very material difference in the quality of the
wine, as the water will evaporate, and only the sugar remain; and the
flavor or the bouquet will only be fully developed in fully ripened
fruit. For gathering, use clean tin or wooden pails; cut the stems as
short as possible, and clip or pinch out all unripe or rotten berries,
leaving none but fully ripe berries on the bunch. The further process
will be described under "wine making."


I would here, again remark, that I consider the question of "what to
plant" as chiefly a local one, for which I do not presume to lay down
fixed rules; but which every one must, to a certain extent, determine
for himself, by visiting vineyards as nearly similar in soil and
location to the one he intends to plant, and then closely observing the
habits of the varieties after planting. Only thus can we obtain certain
results; not by following blindly in the footsteps of so-called
authorities, who may live a hundred, or a thousand miles from us, and
whose success with certain varieties, on soil entirely different from
ours, under different atmospheric influences, can by no means be taken
by us as evidence of our success under other circumstances.

CLASS 1.--_Varieties most generally used._


Originated with Mr. E. BULL, of Concord, Mass. This variety seems to be
the choice of the majority throughout the country, and however much
opinions may differ about its quality, nobody seems to question its
hardiness, productiveness, health and value as a market fruit. Here it
is of very good quality--and our Eastern brethren have no idea what a
really well ripened Missouri grown Concord grape is. It seems to become
better the further it is grown West and South; an observation which I
think applies with equal force to the Hartford Prolific, Norton's
Virginia, Herbemont and others.

Bunch large, heavy shouldered--somewhat compact; berries large, round,
black, with blue bloom; buttery, sweet and rich _here_, when well
ripened; with very thin skin and tender pulp. A strong and vigorous
grower; with healthy, hardy foliage; free from mildew, and but slightly
subject to rot; succeeds well in almost any soil; and is, so far, the
most profitable grape we grow. A fine market fruit, and also makes a
fine, light red wine, which is generally preferred to the Catawba. Can
be easily grown from cuttings.


Originated by DR. N. NORTON, of Richmond, Virginia. This grape has
opened a new era in American grape culture, and every successive year
but adds to its reputation. While the wine of the Catawba is often
compared to Hock, in the wine of Norton's Virginia, we have one of an
entirely different character; and it is a conceded fact that the best
red wines of Europe are surpassed by the Norton as an astringent, dark
red wine, of great body, fine flavor, and superior medical quality.
Vine vigorous and hardy, productive; starting a week later in the
Spring than the Catawba, yet coloring a week sooner; and will succeed
in almost any soil, although producing the richest wine in warm,
southern aspects. Bunches medium, compact; berries small, black, sweet
and rich; with dark bluish red juice; only moderately juicy. Healthy in
all locations, as far as I know, but I doubt its utility in the East,
as I do not think the summers warm and long enough. Seems to attain its
greatest perfection in Missouri, but is universally esteemed in the
West. Very difficult to propagate, as it will hardly grow from cuttings
in open air.

[Illustration: FIG. 24.
HERBEMONT.--_Berries 1/3 diameter._]


Origin uncertain. Wherever this noble grape will succeed and fully
ripen, it is hard to find a better, for table, as well as for wine. Its
home seems to be the South; and I think it will become one of the
leading varieties, as soon as the new order of things has been fully
established, and free, intelligent labor has taken the place of the
drudging, dull toil of the slave. It is particularly fond of warm,
southern exposures, with light limestone soil, and it would be useless
to plant it on soil retentive of moisture. Bunch long, large shouldered
and compact; berry medium, black, with blue bloom--"bags of wine," as
Downing fitly calls them; skin thin, sweet flesh, without pulp, juicy
and high flavored, never clogs the palate; fine for the table, and
makes an excellent wine, which should be pressed immediately after
mashing the grapes, when it will be white, and of an exquisite flavor;
generally ripens about same time as Catawba. A very vigorous and
healthy grower, but tender in rich soils, and should be protected in
winter. Extremely productive.


Raised by Mr. STEEL, of Hartford, Conn.: hardy, vigorous and
productive; bunch large, shouldered, rather compact; berry full medium,
globular, with a perceptible foxy flavor; skin thick, black, covered
with blue bloom; flesh sweet, juicy; much better here than at the East;
of very fair quality for its time of ripening; hangs well to the bunch
here, although said to drop at the East. For market, this is perhaps as
profitable as any variety known, as it ripens very early and uniformly,
producing immense crops. I have made wine from it, which, although not
of very high character, yet ranks as fair.


Origin uncertain; from Western New York; vigorous, hardy and
productive; free from disease; bunch medium, long and narrow, generally
shouldered, compact; berry medium, roundish oblong, black, covered with
bloom; juicy; somewhat acid; colors early, but should hang late to
become thoroughly ripe; brisk vinous flavor, but somewhat of the aroma
of the frost grape; makes a dark red wine, of good body, and much
resembling claret, but not equal to Norton's Virginia, or even the
Concord, in my estimation. Although safe and reliable, I think it has
lately been over praised as a wine grape, and as it is a very long,
straggling grower, it is one of the hardest vines to keep under
control. Propagates with the greatest ease.


First disseminated and made known to the public by Mr. A. THOMPSON, of
Delaware, Ohio. This is claimed by many to be the best American grape;
and although I am inclined to doubt this, and prefer, for my taste, a
well ripened Herbemont, it is certainly a very fine fruit.
Unfortunately, it is very particular in its choice of soil and
location, and it seems as if there are very few locations at the West
where it will succeed. Whoever has a location, however, where it will
grow vigorously and hold its leaves, will do well to plant it almost
exclusively, as it makes a wine of very high character, and is very
productive. A light, warm soil seems to be the first requisite, and the
bluffs on the north side of the Missouri river seem to be peculiarly
adapted to it, while it will not flourish on those on the south side.
Bunch small, compact, and generally shouldered; berry below medium,
round; skin thin, of a beautiful flesh-color, covered with a lilac
bloom; very translucent; pulp sweet and tender, vinous and delicious;
wood very firm; short-jointed; somewhat difficult to propagate, though
not so much so as Norton's Virginia. Subject in many locations, to
leaf-blight, and is _there_ a very slow grower. Fine for the table, and
makes an excellent white wine, equal to, if not superior, to the best
Rhenish wines, which sells readily at from five to six dollars per
gallon. Although I cannot recommend it for general cultivation, it
should be tried every where, and planted extensively where it will
succeed. Ripens about five days later than Hartford Prolific.

CLASS 2.--_Healthy varieties promising well_.


Origin unknown--said to come from Arkansas. This grape promises fair to
become a dangerous rival to Norton's Virginia, which variety it
resembles so closely in wood and foliage, that it is difficult if not
impossible to distinguish it from that variety. The bunch and berry are
of the same color as Norton's Virginia, but somewhat larger, and more
juicy; sweeter, with not quite as much astringency, and perhaps a few
days earlier. Makes an excellent dark red wine, with not as much
astringency, but even more delicate aroma, and was pronounced the "best
red wine on exhibition," at the last meeting of the State Horticultural
Society, where it was in competition with eight samples of the Norton's
Virginia. A strong grower, and productive; as difficult to propagate as
the Norton. Mr. FULLER evidently has not the true variety, when he
calls it worthless, and identical with the Chippewa and Missouri, from
both of which it is entirely distinct.


Closely resembles the foregoing, and will also make an excellent wine
of a similar character. I consider both of these varieties as great
acquisitions, as they are perfectly healthy, very productive, and will
make a wine unsurpassed in merit by any of their class.


This grape, under proper treatment, has proved very productive with me,
and will make a wine of very high quality. The bunches and berries are
small, it is true; but not much more so than the Delaware; it also sets
its fruit well, and as it is hardy, healthy, and a strong grower, it
promises to be one of our leading wine grapes. Bunches small, but
compact, shouldered; berry small; white at the East; pale flesh-color
here; round, sweet, and without pulp; skin very thin. Requires long
pruning on spurs, to bring out its fruitfulness.

[Illustration: FIG. 25.
HARTFORD PROLIFIC.--_Berries 1/2 diameter._]


This new grape, grown from the seed of the Concord, by that
enthusiastic and warm-hearted horticulturist, SAMUEL MILLER, of
Lebanon, Pa., promises to be one of the greatest acquisitions to our
list of really hardy and good grapes, which have lately come before the
public. It has fruited with me the last extremely unfavorable season,
and has stood the hardest test any grape could be put to, without
flinching. Bunch medium, but compact and heavy, shouldered; berry pale
yellow, covered with a white bloom; perhaps a trifle smaller than the
Concord; round; pulpy, but sweet as honey, with only enough of the foxy
aroma to give it character; juicy--very good. I esteem it more highly
than any other white grape I have, as it has the healthy habit and
vigorous growth of its parent, and promises to make an excellent white
wine. Hangs to the bunch well, and will ripen some days before the


Another very promising white grape--a strong grower, and healthy; may
be somewhat too late in the east, but will, I think, be valuable at the
West and South. Bunch medium to large---not shouldered; berry above
medium; oval; pale yellow, with a slight amber tint on one side; pulp
tender, sweet and sprightly; few seeds; fine aroma; quality, best.
Ripens about same time as Catawba; seems to be productive.


This variety, which is also too late in ripening for the East, to be
much esteemed there, fruited with me last season, and more than
fulfilled all the expectations I entertained of it. It is the best of
Mr. ROGERS' Hybrids, which I have yet tasted; and its productiveness,
healthy habit, large berry, and good quality, makes it one of the most
desirable of all the grapes we raise here, for the table and market.
Bunch medium, loose, shouldered; berry very large, oblong, pale
flesh-color; skin thin; pulp tender; few seeds, separating freely from
the pulp; sweet, vinous and juicy; quality very good. Ripens about same
time as Catawba. It is to be regretted that Mr. ROGERS has not named
some of the best of his hybrids, as the numbers give rise to many
mistakes, and a great deal of confusion. It would be in the interest of
grape-growing if this was avoided, by naming at least the best of them.


This grape, although not quite perhaps so early as has been claimed for
it--ripening about five days after Hartford Prolific--is yet of much
better quality; and if it only should prove productive enough, will no
doubt make an excellent wine. Bunch long, loose, shouldered; berry full
medium, black, round, with little bloom; pulp tender; dark juice, sweet
and very good--seems to be hardy and healthy.


Bunch large, shouldered, compact; berry large, oblong, black, with blue
bloom; pulpy, but sweet and good; ripens only a few days after Hartford
Prolific--very productive, hardy and healthy; strong grower. One of the
most showy market grapes we have--not much smaller than Union
Village--and as it ripens evenly, and is of very fair quality, is quite
a favorite in the market. Makes also a wine of very fair quality.


For the West, and very likely further South, this is a very desirable
grape for wine, of the Herbemont class. Bunch compact and heavy,
sometimes shouldered; berry rather small, black, without pulp, juicy
sweet and good; productive, but somewhat tender; strong grower; should
be covered in Winter; makes a very delicious wine, of the Madeira
class, which very often remains sweet for a whole year. Ripens late,
about a week after the Catawba.


Mr. FULLER evidently does not know this grape, as he says it is the
same as Logan. The Rulander we have here, is claimed to be a true
foreign variety. I am inclined to think, however, that it is either a
seedling from foreign seed, raised in the country, or one of the
Southern grapes of the Herbemont class. Be this as it may however, it
certainly bears no resemblance to the Logan, which is a true Fox, of
the Labrusca family. Vine a strong, vigorous, short-jointed grower,
with heart-shaped, light green, smooth leaves; very healthy, and more
hardy than either the Herbemont or Cunningham. Bunch rather small, very
compact, shouldered; berry small, black, without pulp, juicy sweet and
delicious; not subject to rot or mildew: makes a delicious, high
flavored wine, but not a great deal of it. The wine of this variety is
certainly one of the most delicate and valuable ones we have yet made
here and on the soil around Hermann, it will, I think, take preference
over the Delaware. Ripens a few days later than Concord.


Introduced here by Mr. F. MUENCH, who received it from Mr. THEARD, of
Louisiana, where it has been cultivated for some time. Some claim that
it is the grape which makes the famous white Burgundy wine of Europe. I
am inclined to think it is also a native, grown from foreign seed, like
the foregoing, which it closely resembles in foliage and wood; but
will, I think, make a wine of still higher quality, perhaps the most
delicate white wine we yet have. It can hardly be distinguished from
the Rulander in appearance, but has a more sprightly flavor. Ripens at
the same time.


This nice little grape will certainly make one of the most delicious
red wines we have, if it can only be raised in sufficient quantity. It
is healthy and moderately productive, but a slow grower. Bunch loose,
small, shouldered; berry small, black, without pulp, juicy, sweet and
delicious; quality best. Ripens about the same time as the Concord.


Bunch medium, very compact, shouldered; berry medium, round,
greenish-white, covered with white bloom; thick skin, pulpy, but very
sweet, and of fine flavor; makes an excellent white wine; very
productive, but somewhat subject to leaf-blight in wet seasons; does
not rot or mildew.

[Illustration: FIG. 26.
CONCORD.--_Berries 1/2 diameter._]


Has often been confounded with Mary Ann, as both varieties were
disseminated here, by different persons, under the same name. The true
Blood's Black is a few days later than Hartford Prolific; bunch heavy
and compact, shouldered; berry round, black, full medium, of very fair
quality, and an excellent early market grape. The vine is healthy,
hardy, and enormously productive.


Perhaps the largest native grape, of fair quality; bunch large, heavy
and compact, shouldered; berry very large, oval, black, with blue
bloom, pulpy, but juicy, sweet and good. Of better quality here than
Isabella; tolerably free from disease, and a splendid market and table
fruit. Ripens rather late.


For those who do not object to a good deal of foxy flavor, this will be
a valuable market grape, on account of its earliness, beautiful color,
and great productiveness. Mr. FULLER has evidently not the true
variety, as he describes it as a "black grape, sour and worthless."

Bunch medium, compact, shouldered; berry full medium, oval,
flesh-color, with a beautiful lilac bloom; very sweet, pulpy and foxy.
Ripens at same time with Hartford Prolific. Vine a strong grower,
healthy and hardy.


For family use, there is at present no grape here at the West, which is
superior to this in quality; and although it will not pay to plant
largely, either for market or wine, yet no one who can appreciate a
really good grape, should be without a few vines of it at least.

Bunch long, rather loose, shouldered; berry medium, pale yellow,
translucent, without pulp, sweet, juicy, and of excellent flavor; vine
moderately productive and healthy. Ripens with Catawba.


This variety is recommended so much lately, as a superior grape for red
wine, that I will mention it here, although I have not yet fruited it.
It was first introduced by Col. WARING, of Hamilton County, Ohio, and
is said to be free from rot, healthy and vigorous, and to make an
excellent red wine, the must having sold from the press at $4 to $5 per
gallon. The following description is from bunches sent me from Ohio
last fall:

Bunch medium, compact, shouldered; berry rather below medium, black,
oblong, juicy, sweet and well flavored; ripens about the time of the
Concord. Vine vigorous and healthy; said to propagate with the greatest
ease; evidently belonging to the Labrusca species.

We have a seedling here of the Norton's Virginia, raised by Mr. F.
LANGENDORFER, of this neighborhood, which promises to be a valuable
wine grape for this location. It has not yet been named, and the owner
says will never receive a name, unless it proves, in some respect,
superior to anything we have yet. He has fruited it twice, and made
wine from it the last season, which is of a very high character,
resembling Madeira, of a brownish-yellow color; splendid flavor, and of
great body. The vine is a strong grower, healthy and very productive;
bunch long, seldom shouldered, very compact; berry small, black, with
blue bloom; only moderately juicy, and ripens a week later than its
parent. I am inclined to think that it will be of great value here and
further south as a wine grape, although it would ripen too late to suit
the climate further north.

It may be expected here that I should speak of the Iona, Israella, and
Adirondac, as many, and good authorities too, think they will be very
valuable. The Iona and Israella have fruited but once with me, last
summer, and my experience, therefore, has not been long enough to
warrant a decided opinion. As far as it goes, however, it has been
decidedly unfavorable. My Iona vine set about twenty five bunches, but
mildewed and rotted so badly, that I hardly saved as many berries. It
may improve in time, but I hardly think it will do for our soil;
whatever it may do for others--and I cannot put it down as "promising
well." It is a grape of fine quality, _where it will succeed_. The
Israella stood the climate and bad weather bravely, but ripened at
least five days later than the Hartford Prolific close by, and was not
as good in quality as that grape; in fact, the most insipid and
tasteless grape I ever tried. They may both improve, however, upon
closer acquaintance, or be better in other locations. Here, I do not
feel warranted in praising them, and a description will hardly be
needed, as their originator has taken good care to so fully bring their
merits, real or imaginary, before the grape-growing community, that it
would be superfluous for me to describe them.

The Adirondac I saw and much admired at the East, in 1863; and if its
originator, Mr. BAILEY, had only been liberal enough to furnish me with
a scion of two eyes, for which I offered to pay him at the rate of a
dollar per eye, I would, perhaps, be able to report about it. Instead
of the scion, he sent me a dried up vine, which had no life in it when
I received it, and in consequence of these disadvantages, I have not
been able to fruit it yet. It seems to be healthy and vigorous,
however; and should the quality of the fruit be the same as at the
East, may be a valuable acquisition.

On this list I have only mentioned those which have fruited here from
four to five years, with very few exceptions, and which have generally,
during that time, proved successful. To fully warrant the
recommendation of a grape for general cultivation I think, we should
have fruited it at least five or six years; and although there are many
on this list which I should not hesitate to plant largely, yet I have
preferred to be rather a little over cautious than too sanguine.

CLASS 3.--_Healthy varieties, but inferior in quality._


This grape has attracted some attention lately--some persons claiming
for it superior qualities as a _wine_ grape, even classing it with the
Delaware, a statement which I cannot believe. It is a rank Fox, and I
can therefore hardly think it will make a wine to suit a fastidious

Bunch medium, very compact, sometimes shouldered; berry full medium,
pale red, round, sweet, but very pulpy and foxy. Ripens later than
Catawba; is very productive, vigorous and healthy--not subject to rot.

[Illustration: FIG. 27.
CREVELING.--_Berries 1/2 diameter._]


The earliest grape we have--healthy, hardy and productive--but in point
of quality, a rather poor Isabella, which it much resembles.

Bunch full medium, moderately compact, shouldered; berry medium, oval,
black, pulpy, with a good deal of acidity, and strong flavor. Ripens
about four to five days before the Hartford Prolific, but is much
inferior to that variety in quality.


Very productive and healthy, but too foxy, and liable to drop from the
bunch when ripe.

Bunch medium, compact, sometimes shouldered; berry round, brown, sweet,
very foxy--pulpy. Ripens about five days later than Hartford Prolific.


Ripens about same time with Hartford Prolific--but rather inferior in
quality. Bunch long, loose, shouldered; berry medium, oval; resembling


Resembling Isabella, but more free from disease; good grower and
productive; will suit those who like the Isabella.


Bunch medium, compact; berry medium, round, black, juicy; rather
pleasant, but unproductive, and of little value, where better varieties
can be had.


Resembles the foregoing; may, perhaps, make a better wine, but cannot
be recommended.


Grown here, from seed of the Mammoth Catawba, by Mr. MICHAEL POESCHEL.

Bunch medium, compact, sometimes shouldered; berry very large, round,
pale red, pulpy; rather deficient in flavor, but very large; free from
disease. Ripens a week later than Catawba.


Bunch rather small, compact; berry medium, black, round, pulpy, rather
sweet, dark juice. Said to make a good red wine, but my experience has
not been favorable. Ripens late--a week after the Catawba.


A Fox Grape, pale red, pulpy, inferior in quality and color to Perkins,
which it closely resembles; ripens about same time.


This old variety was largely disseminated under the latter name, by
NICHOLAS LONGWORTH, of Cincinnati. It is a nice little grape; but too
unproductive to be of any value here, although it makes a very superior
wine. Bunch long and loose, shouldered; berry small, round, black,
moderately juicy, with little pulp, sweet and good. Ripens a week
before the Catawba.


A grape of very fair quality, and rather early, but a shy bearer. Bunch
small, rather loose; berry medium, pale yellow, sweet and good.


A strong grower; said to be very productive; resembling Clinton in
foliage and general habit. Bunch small, compact; berry below medium,
black, juicy, with a marked frost grape flavor, and hardly worthy of


Of the Herbemont class, but about a week earlier; of good quality, but
too unproductive to be recommended. Bunch medium, compact, shouldered;
berry small, round, black, sweet and good.


Early and hardy, but too unproductive, and bunch too small. Bunch
small, shouldered; berry round; of very good quality for its season;
black, juicy. Ripens as early as Hartford Prolific.

CLASS 4.--_Varieties of good quality, but subject to disease._


This well known grape was brought into notice by Major ADLUM, of
Georgetown, D.C., who thought he had, by its introduction, conferred a
greater boon upon the American people, than if he had paid the national
debt. For the last ten years, it has been so much subject to disease,
that it cannot be recommended any longer, except for some peculiar
locations. It is said to be healthy in northern Illinois and Iowa,
where it will not stand the winter, however, without protection.

Bunch large, moderately compact, shouldered; berry medium, red, covered
with lilac bloom; juicy, pulpy, sweet, somewhat astringent, of good
flavor. A fair grape for the table, and makes a good wine, resembling
Hock, but subject to mildew, rot and leaf-blight.


A seedling of the foregoing, raised by Mrs. DIANA CREHORE. Perhaps one
of the most variable of all the grapes, being very fine one season, and
very indifferent the next. Bunch large and long, compact, shouldered;
berry pale red, round, somewhat pulpy; thick skin; juicy and sweet,
with a peculiar flavor, which DR. WARDER very aptly calls "feline;"
others call it "delicate." Very productive, but subject to leaf-blight,
mildew and rot; although perhaps not so much as the Catawba. Ripens
about a week earlier.


Unworthy of cultivation here, but said to be better at the North. Bunch
long, loose, shouldered; berry medium, oval, black; tough pulp, with a
good deal of acidity, juicy, and a peculiar flavor. Ripens irregularly.
Subject to rot and leaf-blight.


Closely resembling the Isabella, but ripens more evenly, and is of
somewhat better quality.


Bunch large, loose, shouldered; berry black, large, sweet and buttery;
of very good quality, but very much subject to disease. Ripens somewhat
later than Catawba.


Bunch large and loose; berry pale amber, covered with white bloom;
sweet, tolerable flavor, but poor bearer, and subject to mildew. Ripens
about same time as Catawba.


Bunch large and loose, shouldered; berry medium, nearly round; white,
without pulp, juicy and delicious; quality very good, but variable;
sometimes best. Said to be a hybrid of Vitis Labrusca and a foreign
grape, raised by J. F. ALLEN, Salem, Massachusetts, and is really a
fine grape, although too tender and variable for extensive vineyard
culture. Ripens about two weeks before Catawba.


Much recommended in Ohio, where it originated, but unworthy of culture
here, being a poor grower, a shy bearer and very much subject to
leaf-blight. Bunch medium, compact; berry dirty greenish-white; thick
skin; pulpy, and insipid.


This is, in dry seasons, a really fine grape, but subject to
leaf-blight and mildew in hot seasons. Bunch often a foot long, loose,
shouldered; berry below medium, round, black, juicy; without pulp,
sweet and vinous. Belonging to the Herbemont family; is a strong
grower; very productive, and rather tender. May be valuable in well
drained soils, and southern climate, as it undoubtedly will make a fine


Bunch long and loose, large, shouldered; berry medium, round, pale red,
with fine lilac bloom; pulpy; of fair quality, but subject to
leaf-blight, and mildew.


Bunch large, loose, shouldered; berry above medium, red with blue
bloom, roundish-oblong, pulpy, with peculiar flavor, sweet and juicy. A
showy grape, but not very good in quality, and much subject to mildew
and rot. Ripens at the same time with Catawba.

CLASS 5.--_Varieties unworthy of cultivation._


Of all the humbugs ever perpetrated upon the grape-growing public, this
is one of the most glaring. The vine, although a rank and healthy
grower, is unproductive; seldom setting more than half a dozen berries
on a bunch, and these are so sour, have such a hard pulp, with such a
decided frost-grape taste and flavor, and are so deficient in juice,
that no sensible man should think of making them into wine, much less
call it, as its disseminator did, "the true port wine grape."


This was sent me some eight years ago, by B. M. WATSON, as "the best
and hardiest white grape in cultivation," and he charged me the
moderate sum of $5 each, for small pot plants, with hardly two eyes of
ripened wood. After careful nursing of three years, I had the pleasure
of seeing my labors rewarded by a moderate crop of the vilest _red_ Fox
Grapes it has ever been my ill luck to try.

The foregoing have all been tried by me, and have been characterized
and classified as I have found them _here_. The following are varieties
I have not fruited yet, although I have them on trial.

Varieties highly recommended by good authorities: Telegraph, Black
Hawk, Rogers' Hybrids, Nos. 3, 4, 6, 9, 12, 13, 19, 22, 33, Hettie,
Lydia, Charlotte, Mottled, Pauline, Wilmington, Cotaction and Miles.

There are innumerable other varieties, for which their originators all
claim peculiar merits, and some of whom may prove valuable. But all who
bring new varieties before the public, should consider that we have
already names enough, nay, more than are good for us, and that it is
useless to swell the list still more, unless we can do so with a
variety, superior in some respects to our best varieties. A new grape,
to claim favor at the hands of the public, should be healthy, hardy, a
good grower, and productive; and of superior quality, either for the
table or for wine.

There are some varieties circulated throughout the country as natives,
which are really nothing but foreign varieties, or, perhaps, raised
from foreign seed. They will not succeed in open air, although now and
then they will ripen a bunch. The Brinkle, Canadian Chief, Child's
Superb, and El Paso belong to this class.

A really good _table_ grape should have a large amount of sugar, but
tempered and made more agreeable by a due proportion of acid, as, if
the acid is wanting, it will taste insipid; a tender pulp, agreeable
flavor, a large amount of juice, a good sized bunch, large berry, small
seeds, thin skin, and hang well to the bunch.

A good _wine_ grape should have a large amount of sugar, with the acid
in due proportion, a distinctive flavor or aroma; though not so strong
as to become disagreeable, and for red wines a certain amount of
astringency. It is an old vintner's rule, that the varieties with small
berries will generally make the best wine, as they are generally richer
in sugar, and have more character than varieties with larger berries.

[Illustration: FIG. 28.
CLARA.--_Berries 1/2 diameter._]



Although I have described the process already, I will here again
reiterate that the grapes should be thoroughly _ripe_. This does not
simply mean that they are well colored. The Concord generally begins to
color here the 5th of August, and we could gather the majority of our
grapes, of that variety, for market, by the 15th or 20th of that month;
but for wine-making we allow them to hang until the 15th or 20th of
September, and sometimes into October. Thus only do we get the full
amount of sugar and delicacy of aroma which that grape is capable of
developing, as the water evaporates, and the sugar remains; it also
loses nearly all the acidity from its pulp; and the latter, which is so
tough and hard immediately after coloring, nearly all dissolves and
becomes tender. The best evidences of a grape being thoroughly ripe
are: 1st. The stem turns brown, and begins to shrivel; 2nd, the berry
begins to shrivel around the stem; 3d, thin and transparent skin; 4th,
the juice becomes very sweet, and sticks to the finger like honey or
molasses, after handling the grapes for some time.

It is often the case that some bunches ripen much later on the vines.
In such a case, the ripest should be gathered first, and those that are
not fully ripe remain on the vines until mature. They will ripen much
quicker if the ripest bunches have been removed first.

The first implements needed for the gathering are clean wooden and tin
pails and sharp knives, or better still, the small shears spoken of in
a former part of this work. Each gatherer is provided with a pail, or
two may go together, having a pail each, so that one can empty and the
other keep filling during the time. If there are a good many unripe
berries on the bunches, they may be put into a separate pail, and all
that are soft will give an inferior wine. The bunch is cut with as
short a stem as possible, as the stem contains a great deal of acid and
astringency; every unripe or decayed berry is picked out, so that
nothing but perfectly sound, ripe berries remain.

[Illustration: FIG. 29.]

The next implement that we need is a wooden tub or vat, to carry the
grapes to the mill; or the wagon, if the vineyard is any distance from
the cellar. This is made of thin boards, half-inch pine lumber
generally; 3 feet high inside, 10 inches wide at the bottom, 20 inches
wide at the top, being flat on one side, where it is carried on the
back, and bound with thin iron hoops. It is carried by two
leather-straps running over the shoulders, as shown in Fig. 29, and
should contain about eight to ten pails, or a little over two bushels
of grapes. The carrier can pass easily through the rows with it to any
part of the vineyard, and lean it against a post until full. If the
vineyard is close to the cellar or press-house, the grapes can be
carried to it directly; if too far, we must provide a long tub or vat,
to place on the wagon, into which the grapes are emptied. I will here
again repeat that the utmost cleanliness should be observed in _all_
the apparatus; and no tub or vat should be used that is in the least
degree mouldy. Everything should be perfectly sweet and clean, and a
strict supervision kept up, that the laborers do not drop any crumbs of
bread, &c., among the grapes, as this will immediately cause acetous
fermentation. The weather should be dry and fair, and the grapes dry
when gathered.


As the wine-cellar and press-house are generally built together, I will
also describe them together. A good cellar should keep about an even
temperature in cold and warm weather, and should, therefore, be built
sufficiently deep, arched over with stone, well ventilated, and kept
dry. Where the ground is hilly, a northern or northwestern slope should
be chosen, as it is a great convenience, if the entrance can be made
even with the ground. Its size depends, of course, upon the quantity of
wine to be stored. I will here give the dimensions of one I am
constructing at present, and which is calculated to store from 15,000
to 20,000 gallons of wine. The principal cellar will be 100 feet long,
by 18-1/2 feet wide inside, and 12 feet high under the middle of the
arch. This will be divided into two compartments; the back one, at the
farthest end of the cellar, to be 40 feet, which is destined to keep
old wine of former vintages; as it is the deepest below the ground, it
will keep the coolest temperature. It is divided from the front
compartment by a wall and doors, so that it can be shut off should it
become necessary to heat the other, while the must is fermenting. The
other compartment will be 60 feet long, and is intended for the new
wine, as the temperature will be somewhat higher, and, therefore,
better adapted to the fermentation of the must. This will be provided
with a stove, so that the air can be warmed, if necessary, during
fermentation. This will also be closed by folding doors, 5-1/2 feet
wide. There will be about six ventilators, or air-flues, on each side
of these two cellars, built in the wall, constructed somewhat like
chimneys, commencing at the bottom, whose upper terminus is about two
feet above the arch, and closed with a grate and trap-doors, so that
they can be closed and opened at will, to admit air and light. Before
this principal cellar is an arched entrance, twenty feet long inside,
also closed by folding doors, and as wide as the principal cellar. This
will be very convenient to store empty casks, and can also be used as a
fermenting room in Fall, should it be needed. The arch of the principal
cellar will be covered with about six feet of earth; the walls of the
cellar to be two feet thick. The press-house will be built above the
cellar, over its entire length, and will also be divided into two
rooms. The part farthest from the entrance of the cellar, to be 60 feet
by 18, will be the press-house proper, with folding doors on both
sides, about the middle of the building, and even with the surface
ground, so that a wagon can pass in on one side and out on the other.
This will contain the grape-mill, wine-presses, apparatus for stemming,
and fermenting vats for white or light-colored wine. The other part, 40
feet long, will contain an apparatus for distilling, the casks and vats
to store the husks for distilling, and the vats to ferment very dark
colored wines on the husks, should it be necessary. It will also be
used as a shop, contain a stove, and be floored, so that it will be
convenient, in wet and cold weather, to cut cuttings, &c. A large
cistern, to be built on one side of the building, so that the necessary
water for cleaning casks, &c., will be handy; with a force-pump, will
complete the arrangement. I need hardly add here, that the whole cellar
should be paved with flags or brick, and well drained, so that it will
be perfectly dry.

This cellar is destined to hold two rows of casks, five feet long, on
each side. For this purpose layers of strong beams are provided, upon
which the casks are laid in such a manner that they are about two feet
from the ground, fronting to the middle, and at least a foot or
eighteen inches of space allowed between them and the wall, so that a
man can conveniently pass and examine them. This will leave five and
a-half to six feet of space between the two rows, to draw off the wine,
move casks, _&c_.

This cellar will, at the present rates of work, cost about $6,000. Of
course, the cellar, as before remarked, can be built according to the
wants of the grape-grower. For merely keeping wine during the first
winter, a common house cellar will do; but during the hot days of
summer wine will not keep well in it.


This mill can be made very simple, of two wooden rollers, fastened in a
square frame, running against each other, and turned with a crank and
cog-wheel. The rollers should be about nine inches in diameter, and set
far enough apart to mash the berries, but not the seeds and stems. A
very convenient apparatus, mill and press, is manufactured by Geiss &
Brosius, Belleville, Ill., and where the quantity to be made does not
exceed 2,000 gallons, it will answer every purpose. The mill has stone
rollers, which can be set by screws to the proper distance, with a
cutting apparatus on top, for apples in making cider, which can be
taken off at will. The press is by itself, and consists of an iron
screw, coming up through the platform, with a zinc tube around it to
prevent the must from coming in contact with it. The platform has a
double bottom, the lower one with grooves; the upper consists simply of
boards, with grooves through it to allow the must to run through. These
boards are held in their places by wooden pegs, and can be taken off at
will. A circular hopper, about a foot in diameter, and made of laths
screwed to iron rings, with about a quarter of an inch space between
them, encloses the zinc tube. The outer frame is constructed in the
same way, is about 2-1/2 feet in diameter, and bound with strong wooden
and iron hoops. The mashed grapes are poured into the frame, a
close-fitting cover is put on, which is held down by a strong block,
and the power is applied by an iron nut just on the top of the screw,
with holes in each end to apply strong wooden levers. The apparatus is
strong, simple, and convenient, and presses remarkably fast and clean,
as the must can run off below, on the outside and also on the inside.
The cost of mill and press is about $90, but each can be had separately
for $45.

If a large amount of grapes are to be pressed, the press should be of
much larger dimensions, but may be constructed on the same principle--a
strong, large platform, with a strong screw coming through the middle,
and a frame made of laths, screwed to a strong wooden frame, through
which the must can run off freely, with another frame around the
outside of the platform. The must runs off through grooves to the lower
side, where it is let off by a spout. It may be large enough to contain
a hundred bushels of grapes at a single pressing, for a great deal
depends upon the ability of the vintner to press a large amount just at
the proper time, when the must has fermented on the husks just as long
as he desires it to do.


These should correspond somewhat with the size of the casks we intend
to fill; but they are somewhat unhandy if they hold more than, say four
hundred gallons. They are made of oak or white pine boards, 1-1/2 inch
thick, bound securely by iron hoops, about three feet high, and, say,
five feet wide. The bottom and inside must be worked clean and smooth,
to facilitate washing. When the must is to ferment a longer time on the
husks, as is often the case in red wines, a false bottom should be
provided, for the purpose of holding the husks down below the surface
of the must. It is made to fit the size of the vat, and perforated with
holes, and held in its place by sticks of two inches square, let into
the bottom of the vat, and which go through the false bottom. A hole is
bored through them, and the bottom held down by means of a peg passed
through this hole. The vat is closed by a tight-fitting cover, through
which a hole is bored, large enough to admit a tin tube of about an
inch in diameter, to let off the gas. The vats are set high enough
above the ground to admit drawing off the must through a faucet near
the bottom of the vat. For those grapes which are to be pressed
immediately we need no false bottoms or covers for the vats. As
fermentation generally progresses very rapidly here, and it is not
desirable with most of our wines to ferment them on the husks very
long, as they generally have astringency enough, operations here are
much more simple than in Europe.

The must is generally allowed to run into a large funnel, filled with
oat straw, and passes through a hose into the casks in the cellar. A
hole can be left through the arch for that purpose, as it is much more
convenient than to carry the must in buckets from the press into the

It is sometimes desirable to stem the grapes, although it is seldom
practiced in this country. This can be easily done by passing the
bunches rapidly over a grooved board, made somewhat in the form of a
common washboard, only the grooves should be round at the bottom and
the edges on top. It is seldom desirable here.


These should be made of well-seasoned white oak staves, and can, of
course, be of various sizes to meet the wants of the vintner. The best
and most convenient size for cellar use I have found to be about 500
gallons. These are sufficiently large to develop the wine fully, and
yet can be filled quick enough to not interrupt fermentation. Of
course, the vintner must have some of all sizes, even down to the
five-gallon keg; but for keeping wine, a cask of 500 gallons takes less
room comparatively, and the wine will attain a higher degree of
perfection than in smaller casks. The staves to make such a cask should
be about 5 feet long, and 1-1/2 to 2 inches thick, and be the very best
wood to be had. The cask will, when ready, be about as high as it is
long, should be carefully worked and planed inside, to facilitate
washing and have a so-called door on one end, 12 inches wide and 18
inches high, which is fastened by means of an iron bolt and screw, and
a strong bar of wood. This is to facilitate cleaning; when a cask is
empty, the door is taken out, and a man slips into the cask with a
broom and brush, and carefully washes off all remnants of lees, etc.,
which, as the lees of the wine are very slimy and tenacious, cannot be
removed by merely pouring in water and shaking it about. It is also
much more convenient to let these large casks remain in their places,
than to move them about. The casks are bound with strong iron hoops.

To prepare the new casks, and also the vats, etc., for the reception of
the must, they should be either filled with pure water, and allowed to
soak for several days, to draw out the tannin; then emptied, scalded
with hot water, and afterwards steamed with, say two or three gallons
of boiling wine; or they can be made "wine-green," by putting in about
half a bushel of unslaked lime, and pouring in about the same quantity
of hot water. After the lime has fallen apart, add about two quarts of
water to each pound of lime, put in the bung, and turn the cask about;
leaving it lie sometimes on one side, sometimes on the other, so that
the lime will come in contact with every part of the cask. Then pour
out the lime-water; wash once or twice with warm water, and rinse with
a decoction of vine leaves, or with warm wine. Then rinse once more
with cold water, and it will be fully prepared to receive the must.
This is also to be observed with old casks, which have become, by
neglect or otherwise, mouldy, or have a peculiar tang.


As we have our apparatus all prepared now, we can commence the
operation itself. This can be done in different ways, according to the
class of wine we are about to make.

To make white, or light-colored wine, the grapes which were gathered
and mashed during the day, can be pressed and put into the cask the
following night. To mash them, we place the mill above one of the
fermenting vats, mashing them as quick as they are carried or hauled to
the press-house. The vat is simply covered with a cloth during the day.
If the season has been good, the must will make good wine without the
addition of anything else. In poor seasons it will be necessary to add
water and sugar, to improve its quality, but I will speak of this
method in a separate chapter. In the evening, the must which will run
off, is first drawn from the vat, and by some kept separate; but I
think, it makes, upon the whole, a better wine, if the pressing is
added to it. The husks, or mashed grapes, are then poured upon the
press, and pressed until fully dry. To accomplish this the press is
opened several times, and the edges of the cake, or "cheese," as some
call it, are cut off with an axe or cleaver and put on top, after which
they are pressed down again. The casks are then filled with the must;
either completely, if it is intended that the must should ferment
_above_, as it is called, or _under_, when the cask is not completely
filled, so that the husks, which the must will throw up, will remain in
the cask. Both methods have their advantages, but I prefer the former,
with a very simple contrivance, to exclude the air, and also prevent
waste. This is a siphon or tin tube, bent in the form of a double
elbow, of which one end fits tightly in the bung hole, and the other
empties into a dish of water, to be set on one end of the cask, through
which the gas escapes, as shown in Fig. 30.

We should, however in pressing, be guided somewhat by the weather. In
warm weather fermentation will commence much sooner, and be more
violent, than when the weather is cold. Consequently we should press
much sooner in warm weather, than when the air is cool. Late in the
fall, it is sometimes advisable to leave the must a day longer on the
husks, than indicated below. The cellar should be kept at an even
temperature of about 60° during the first few weeks, and if it does not
naturally attain this temperature, then it should be warmed by a stove,
as much of the quality of the wine depends upon a thorough fermentation
during the first ten days.

[Illustration: FIG. 30.]

When violent fermentation has ceased, say after about ten or twelve
days, and the must has become quiet, the cask should be closed with a
tight bung, and the wine is left until it is clear. In about two to
three months it ought to be perfectly clear and fine--is then racked,
_i.e._, drawn from the lees, by means of a faucet, and put into clean,
sweet casks. It is very important that the casks are "wine-seasoned,"
that is, have no other tang than of wine. For must, fresh brandy or
whiskey casks may be used, but after the wine has fermented, it will
not do to use such, as the wine will acquire the smell and taste of the
liquor. When a cask has been emptied, it should be carefully cleaned,
as before described, by entering at the door, or with smaller casks, by
taking out the head. After it is thoroughly cleansed, it may be
fumigated slightly, by burning a small piece of sulphured paper, or a
nutmeg in it, and then filled. To keep empty casks in good condition
they should, after cleaning, be allowed to become thoroughly dry, when
they are sulphured, closed tightly, and laid away in the cellar. The
operation of sulphuring should be repeated every six weeks. If wanted
for use, they are simply rinsed with cold water.

[Illustration: FIG. 31.]

For racking the wine, we should have: 1st a large brass faucet. 2d.
Pails of a peculiar shape, wider at the top, to prevent wastage. 3d. A
wooden funnel, as shown in Fig. 31, to hold about six gallons. In
racking--first carefully lift the bung of the cask, as the exclusion of
air from above would cause a gurgling motion in the cask, if tapped
below, which would stir up the lees in the bottom. Then, after having
loosened with a hammer the wooden peg, closing the tap hole, let your
assistant hold the pail opposite the hole, hold the faucet in your
right hand, and with the left, withdraw the plug, inserting the faucet
quickly. Drive it in firmly with a hammer, and you are ready for the

Do not fully open the faucet at first, because the first pailful is
generally not quite clear, and should run slowly. You can keep this by
itself; and this, and the last from the lees, is generally put into a
cask together and allowed to settle again. It will make a good, clear
wine after a few weeks. As soon as the wine runs quite clear and
limpid, it can be put into the cask destined to receive it, and you can
let it run as fast as it can be emptied. When the wine has run off down
to the tap hole, the cask may be carefully raised on the other end, one
inserting a brick or piece of board under it, while the other lifts
gently and slowly. This may be repeated several times, as long as the
wine runs clear; and even the somewhat cloudy wine may be put with the
first pailful into a separate cask. As soon as it comes thick or muddy,
it is time to stop. The lees are emptied out, and will, if distilled,
make a fine flavored and very strong brandy.

This treatment can be applied to all white and light-colored wines,
when it is not desirable to have a certain astringency in the wine. The
Catawba, Concord, Herbemont, Delaware, Rulander, Cassady, Taylor,
Louisiana, Hartford Prolific, and Cunningham should all be treated in a
similar manner. The Concord, although it will, under this treatment,
make only a light red wine, of which the color can be changed to dark
red by fermenting on the husks, is not desirable if treated in the
latter manner; as the peculiar foxy aroma of the grape will be imparted
to the must to such a degree, as to make the flavor disagreeable, I
shall recur to the subject of flavor in wines in another chapter.

To make red wine, the must should be fermented on the husks, as
generally the darkest color is desired, and also, a certain
astringency, which the wine will acquire principally from the seeds,
skins, and stems of the grapes, which contain the tannin. The grapes
are mashed, and put into the fermenting vat, of the kind described
before, with false bottoms. After the vat is filled about three-fourths
the false bottom is put on, the husks are pressed down by it, until
they are covered about six inches by the must, and the cover put on. It
is seldom desirable here to ferment longer than three days on the
husks, if the weather is warm--in a temperature of 60°--two days will
often be enough, as the wine will become too rough and astringent by an
excessively long fermentation. Only experience will be the proper guide
here, and also the individual taste. It will be generally time to
press, when the must has changed its sweet taste, and acquired a
somewhat rough and bitter one. Where it is desired to make a very dark
colored wine, without too much astringency, the grapes should be
stemmed, as most of the rough and bitter taste is in the stems; and it
can then be fermented on the husks for six or eight days. In this
manner the celebrated Burgundy wines are made; also most of the red
wines of France and Germany. Many of them are even allowed to go
through the whole process of fermentation, and the husks are filled
into the cask with the must, through a door, made in the upper side of
the cask; and it there remains, until the clear wine is drawn off. This
is seldom desirable here, however, as our red wine grapes have
sufficient astringency and color without this process. The treatment
during fermentation, racking, etc., is precisely the same as with white
wine, with only this difference, that the red wine is generally allowed
to stay longer on the lees; for our object in making this class of wine
is different than in making white, or so-called Schiller or light red
wine. In white and light colored wines we desire smoothness and
delicacy of bouquet and taste; in dark red wines, we desire astringency
and body, as they are to be the so-called stomach or medical wines. It
is therefore generally racked but once, in the latter part of February
or March, and the white and light colored wines are racked in December
or January, as soon as they have become clear--and again in March. We
also use no sulphur in fumigating the casks, as it takes away the color
to a certain extent. We generally do not use anything, but simply clean
the casks well, in racking red wine.

I will say a few words in regard to _under_ fermentation. If this
method is to be followed, the casks are not filled, but enough space
left to allow the wine to ferment, without throwing out lees and husks
at the bung. The bung is then covered, by laying a sack filled with
sand over it, and when fermentation is over--as well by this as by the
other method--the casks are filled with must or wine, kept in a
separate cask for the purpose. The casks should always be kept well
filled, and must be looked over and filled every two or three weeks, as
the wine will continually lose in quantity, by evaporation through the
wood of the casks. The casks should be varnished or brushed over with
linseed oil, as this will prevent evaporation to some extent.

In wine making, and giving the wine its character, we can only be
guided by practice and individual taste, as well as the prevailing
taste of the consuming public. If the prevailing taste is for light
colored, smooth and delicate wines, we can make them so, by pressing
immediately, and racking soon, and frequently. If a dark colored,
astringent wine is desired, we can ferment on the husks, and leave it
on the lees a longer period. There is a medium course, in this as in
everything else; and the intelligent vintner will soon find the rules
which should guide him, by practice with different varieties.

Among the wines to be treated as dark red, I will name Norton's
Virginia, Cynthiana, Arkansas, and Clinton, and, I suppose, Ives'
Seedling. It would be insulting to these noble wines to class with them
the Oporto, which may make a very dark colored liquid, but no _wine_
worth the name, unless an immense quantity of sugar is added, and
enough of water to dilute the peculiar vile aroma of that grape.


Even if the wine was perfectly fine and clear, when drawn off, it will
go through a second fermentation as soon as warm weather sets it--say
in May or June. If the wine is clear and fine, however, the
fermentation will be less violent, than if it is not so clear, as the
lees, which the wine has never entirely deposited; act as they ferment.
It is not safe or judicious, therefore, to bottle the wine _before_
this second fermentation is over. As soon as the wine has become
perfectly clear and fine again--generally in August or September--it
can be bottled. For bottling wine we need: 1st. clean bottles. 2d. good
corks, which must first be scalded with hot water, to soften them, and
draw out all impurities, and then soaked in cold water. 3d. a small
funnel. 4th. a small faucet. 5th. a cork-press, of iron or wood. 6th. a
light wooden mallet to drive in the corks.

After the faucet has been inserted in the cask, fill your bottles so
that there will be about an inch of room between the cork and the wine.
Let them stand about five minutes before you drive in the cork, which
should always be of rather full size, and made to fit by compressing it
with the press at one end. Then drive in the cork with the mallet, and
lay the bottles, either in sand on the cellar floor, or on a rack made
for that purpose. They should be laid so that the wine covers the cork,
to exclude all air.

The greater bulk of the wine, however, if yet on hand; can be kept in
casks. All the wine to be kept thus, should be racked once in about six
months, and the casks kept well filled. Most of our native wines,
however, are generally sold after the second racking in March, and a
great many even as soon as clear--in January.


These will seldom occur, if the wine has been properly treated. Cases
may arise, however, when it will become necessary to rack the wine, or
fine it by artificial means.


The cause of this is generally a want of Tannin. If the wine has a
peculiar, flat, soft taste, and looks cloudy, this is generally the
case. Draw the wine into another cask, which has been well sulphured,
and add some pulverized tannin, which can be had in every drug store.
The tannin may be dissolved in water--about an ounce to every two
hundred gallons of wine--and the wine well stirred, by inserting a
stick at the bung. Should it not have become clear after about three
weeks, it should be fined. This can be done, by adding about an ounce
of powdered gum-arabic to each forty gallons, and stirring the wine
well when it has been poured in. Or, take some wine out of the
casks--add to each forty gallons which it contains the whites of ten
eggs, whipped to foam with the wine taken out--pour in the mixture
again--stir up well, and bung up tight. After a week the wine will
generally be clear, and should then be drawn off.


These should be distilled, and will make a very strong, fine flavored
brandy. The husks are put into empty barrels or vats--stamped down
close, and a cover of clay made over them, to exclude the air. They
will thus undergo a fermentation, and be ready for distillation in
about a month. They should be taken fresh from the press, however; for
if they come into contact with the air, they will soon become sour and
mouldy. The lees can be distilled immediately. Good fresh lees, from
rather astringent wines are also an excellent remedy when the wine
becomes flat, as before described.


The process of wine making before described, however, can only be
applied in such seasons, and with such varieties of grapes, that
contain all the necessary elements for a good wine in due proportion.
For unfavorable seasons, with such varieties of grapes as are deficient
in some of the principal ingredients, we must take a different
course--follow a different method. To see our way clearly before us in
this, let us first examine which are the constituent parts of must or
grape juice. A chemical analysis of must, shows the following result:

Grape juice contains sugar, water, free acids, tannin, gummy and mucous
substances, coloring matter, fragrant or flavoring substances, (aroma
bouquet). A good wine should contain all these ingredients in due
proportion. If there is an excess of one, and a want of the other, the
wine will lose in quality. Must, which contains all of these, in due
proportion, we call _normal_ must, and only by determining the amount
of sugar and acids in this so-called normal must, can we gain the
knowledge how to improve such must, which does not contain the
necessary proportion of each. The frequent occurrence of unfavorable
seasons in Europe, when the grapes did not ripen fully, and were sadly
deficient in sugar, set intelligent men to thinking how this defect
could be remedied; and a grape crop, which was almost worthless, from
its want of sugar, and its excess of acids, could be made to yield at
least a fair article, instead of the sour and unsaleable article
generally produced in such seasons. Among the foremost who experimented
with this object in view I will here name CHAPTAL, PETIOL; but
especially DR. LUDWIG GALL, who has at last reduced the whole science
of wine-making to such a mathematical certainty, that we stand amazed
only, that so simple a process should not have been discovered long
ago. It is the old story of the egg of Columbus; but the poor vintners
of Germany, and France, and we here, are none the less deeply indebted
to those intelligent and persevering men for the incalculable benefits
they have conferred upon us. The production of good wine is thus
reduced to a mathematical certainty; although we cannot in a bad
season, produce as high flavored and delicate wines, as in the best
years, we can now always make a fair article, by following the simple
rules laid down by DR. GALL. When this method was first introduced, it
was calumniated and despised--called adulteration of wine, and even
prohibited by the governments of Europe; but, DR. GALL fearlessly
challenged his opponents to have his wines analyzed by the most eminent
chemists; which was repeatedly done, and the results showed that they
contained nothing but such ingredients which pure wine should contain;
and since men like VON BABO, DOBEREINER and others have openly endorsed
and recommended gallizing, prejudice is giving way before the light of
scientific knowledge.

[Illustration: FIG. 32.]

But to determine the amount of sugar and acids contained in the must we
need a few necessary implements. These are:


The most suitable one now in use is the _Oechsle's_ must scale,
constructed on the principle that the instrument sinks the deeper into
any fluid, the thinner it is, or the less sugar it contains. Fig. 32
shows this instrument, "which is generally made of silver, or German
silver, although they are also made of glass. A, represents a hollow
cylinder--best made of glass, filled with must to the brim, into which
place the must scale B. It is composed of the hollow float _a_, which
keeps it suspended in the fluid; of the weight _c_, for holding in a
perpendicular position; and of the scale _e_ divided by small lines
into from fifty to one hundred degrees. Before the gauge is placed in
the must, draw it several times through the mouth, to moisten it--but
allow no saliva to adhere to it. When the guage ceases to descend, note
the degree to which it has sunk; after which press it down with the
finger a few degrees further, and on its standing still again, the line
to which the must reaches, indicates its so-called weight, expressed by
degrees." The must should be weighed in an entirely fresh state, before
it shows any sign of fermentation, and should be free from husks, and

This instrument, which is indispensable to every one who intends to
make wine, can be obtained in nearly every large town, from the
prominent opticians. JACOB BLATTNER, at St. Louis keeps them for sale.

The saccharometer will indicate the amount of sugar in the must, and
its use is so simple, that every one can soon become familiar with it.
The next step in the improvement of wines was to determine the amount
of acids the must contained, and this problem has also been
successfully solved by the invention of the acidimeter:


"The first instrument of this kind which came into general use, was one
invented by DR. OTTO, and consists of a glass tube, from ten to twelve
inches in length, half an inch in width, and closed at the lower end.
Fig. 33 shows OTTO'S Acidimeter.

"The tube is filled to the partition line _a_, with tincture of litmus.
The must to be examined, before it has begun to ferment is then poured
into the tube, until it reaches the line 0. The blue tincture of
litmus, which would still be blue, if water had been added, is turned
into rose-color by the action of the acids contained in the must.

"If a solution of 1,369 per cent, of caustic ammonia is added to this
red fluid, and the tube is turned around to effect the necessary
mixture, keeping its mouth closed with the thumb, after the addition of
more or less of the ammonical fluid, it will change into violet. This
tinge indicates the saturation of the acids, and the height of the
fluid in the tube now shows the quantity of acid in the must, by whole,
half and fourth parts per cent. The lines marked 1, 2, 3, 4, indicate
whole per cents.; the short intermediate lines, one-fourth per cents."

[Illustration: FIG. 33.]

When DR. GALL, shortly before the vintage of 1850, first publicly
recommended the dilution of the acids, he was obliged to refer to this
instrument, as already known, and everywhere at hand, which was at the
same time cheap, and simple in its use. "It is true, however, that if
must is examined by this instrument, the quantity of acids contained in
it, is really somewhat larger than indicated by the instrument; because
the acids contained in the must require for their saturation a weaker
solution of ammonia than acetic acid." As however, OTTO'S acidimeter
shows about one eighth of the acids less than the must actually
contains, and about as much acids combined with earths is removed
during fermentation, DR. GALL recommends that the quantity of acids be
reduced to 6-1/2, or at most 7 thousandths of OTTO'S acidimeter, and
the results have shown that this was about the right proportion; as the
wines in which the acids were thus diluted were in favor with all

"The acidimeter referred to was afterwards improved, by making the tube
longer and more narrow, and dividing it into tenths of per cents,
instead of fourths; thus dividing the whole above 0 into thousandths.
But although by this improved acidimeter the quantity of acids could be
ascertained with more nicety, there remained one defect, that in often
turning the glass tube for mixing the fluids, some of the contents
adhered to the thumb in closing its mouth. This defect was remedied in
a new acidimeter, invented by Mr. GEISLER, who also invented the new
vaporimeter for the determination of the quantity of alcohol contained
in wine. It is based on the same principle as OTTO'S, but differs
altogether in its construction. It is composed of three parts, all made
of glass; the mixing bottle, Fig. 34; the Pipette, Fig. 35; and the
burette, Fig 36. Besides, there should be ready three small
glasses--one filled with tincture of litmus, the second with a solution
of 1,369 per ammonia, and the third with the must or wine to be tested;
also, a taller glass, or vessel, having its bottom covered with cotton,
in which glass the burette, after it has been filled with the solution
of ammonia, is to be placed in an upright position until wanted.

[Illustration: FIG. 34.]

[Illustration: FIG. 35.]

[Illustration: FIG. 36.]

"To use this instrument the must and the tincture of litmus, having
first received the normal temperature of 14° Reaumer, are brought into
the mixing bottle by means of the pipette, which is a hollow tube of
glass, open on both ends. To fill it, place its lower end into the
tincture or must, apply the mouth to the upper end, and by means of
suction fill it with the tincture of litmus to above the line indicated
at A. The opening of the top is then quickly closed with the thumb; by
alternately raising the thumb, and pressing it down again, so much of
the tincture is then allowed to flow back into the glass so as to lower
the fluid to the line indicated at A. The remainder is then brought
into the bottle, and the last drops forced out by blowing into the

"In filling it with must, raise the fluid in the same way, until it
comes up to the line indicated at B, and then empty into the mixing

"The burette consists of two hollow tubes of glass. In filling it, hold
the smaller tube with the right hand into the glass containing the
solution of ammonia, apply the mouth to the larger one, and by drawing
in the fluid the tube is filled exactly to the line indicated at 0 of
the tube.

"Holding the mixing bottle by the neck between the thumb and forefinger
of the left hand, place the smaller tube of the burette into the mouth
of the mixing bottle, which must be constantly shaken; let enough of
the solution of ammonia be brought drop by drop, into the mixture in
the bottle, till the red has been changed into the deep reddish blue of
the purple onion. This is the sign of the proper saturation of the
acids. To distinguish still better, turn the mixing bottle upside down,
by closing its mouth with the thumb, and examine the color of the fluid
in the tube-shaped neck of the bottle, and afterwards, should it be
required, add another drop of the ammonia. Repeat this until the proper
tone of color has been reached, neither red nor blue. After thus fixing
the precise point of the saturation of the acids, the burette is held
upright, and the quantity of the solution of ammonia consumed is
accurately determined,--that is, to what line on the scale the burette
has been emptied. The quantity of the solution so used corresponds with
the quantity of acids contained in the must--the larger division lines
opposite the numbers indicating the thousandths part, and the smaller
lines or dots the ten thousandths part.

"Until the eye has learned by practice to recognize the points of
saturation by the tone of color, it can be proven by means of litmus
paper. When the mixture in the bottle begins to turn blue, put in the
end of a slip of litmus paper about half an inch deep, and then draw
this end through your fingers, moistened with water. So long as the
ends of the blue litmus paper become more or less reddened, the acids
have not been completely saturated. Only when it remains blue, has the
point of saturation been reached.

"In examining _red_ must, the method should be modified as
follows:--Instead of first filling the pipette with tincture of litmus,
fill it with water to the line A, and transfer it into the bottle.
After the quantity of must has been added, drop six-thousandths of the
solution of ammonia into the mixture, constantly shaking it while
dropping, then test it, and so on, until, after every further addition
required with litmus paper, it is no longer reddened after having been
wiped off."

DR. GALL further gives the following directions, as a guide, to
distinguish and determine the proportion of acids which a must should
contain, to be still agreeable to the palate, and good:

"Chemists distinguish the acid contained in the grape as the vinous,
malic, grape, citric, tannic, gelatinous and para-citric acids. Whether
all these are contained in the must, or which of them, is of small
moment for us to know. For the practical wine-maker, it is sufficient
to know, with full certainty, that, as the grape ripens, while the
proportion of sugar increases, the quantity of acids continually
diminishes; and hence, by leaving the grapes on the vines as long as
possible, we have a double means of improving their products--the must
or wine.

"All wines, without exception, to be of good and of agreeable taste,
must contain from 4-1/2 to 7 thousandths parts of free acids, and each
must containing more than seven thousandths parts of free acids may be
considered as having too little water and sugar in proportion to its
quantity of acids.

"In all wine-growing countries of Germany, for a number of years past,
experience has proved that a corresponding addition of sugar and water
is the means of converting the sourest must, not only into a good
drinkable wine, but also into as good a wine as can be produced in
favorable years, _except_ in that peculiar and delicate aroma found
only in the must of well-ripened grapes, and which must and will always
distinguish the wines made in the best seasons from those made in poor

"The saccharometer and acidimeter, properly used, will give us the
exact knowledge of what the must contains, and what it lacks; and we
have the means at hand, by adding water, to reduce the acids to their
proper proportion; and by adding sugar, to increase the amount of sugar
the must should contain; in other words, we can change the poor must of
indifferent seasons into the normal must of the best seasons in
_everything_, _except_ its bouquet or aroma, thereby converting an
unwholesome and disagreeable drink into an agreeable and healthy one."


Let us glance for a few moments at this wonderful, simple, and yet so
complicated process, to give a clearer insight into the functions which
man has to perform to assist Nature, and have her work for him, to
attain the desired end. I cannot put the matter in a better light for
my readers than to quote again from DR. GALL. He says:--"To form a
correct opinion of what may and can be done in the manufacture of wine,
we must be thoroughly convinced that Nature, in her operations, has
other objects in view than merely to serve man as his careful cook and
butler. Had the highest object of the Creator, in the creation of the
grape, been simply to combine in the juice of the fruit nothing but
what is indispensable to the formation of that delicious beverage for
the accommodation of man, it might have been still easier done for him
by at once filling the berries with wine already made. But in the
production of fruits, the first object of all is to provide for the
propagation and preservation of the species. Each fruit contains the
germ of a new plant, and a quantity of nutritious matter surrounding
and developing that germ. The general belief is, that this nutritious
matter, and even the peculiar combination in which it is found in the
fruit, has been made directly for the immediate use of man. This,
however, is a mistake. The nutritious matter of the grape, as in the
apple, pear, or any similar product, is designed by Nature only to
serve as the first nourishment of the future plant, the germ of which
lies in it. There are thousands of fruits of no use whatever, and are
even noxious to man, and there are thousands more which, before they
can be used, must be divested of certain parts, necessary, indeed, to
the nutrition of the future plant, but unfit, in its present state, for
the use or nourishment of man. For instance, barley contains starch,
mucilaginous sugar, gum, adhesive matter, vegetable albumen, phosphate
of lime, oil, fibre and water. All these are necessary to the formation
of roots, stalks, leaves, flowers and the new grain; but for the
manufacture of beer, the brewer needs only the first three substances.
The same rule applies to the grape.

"In this use of the grape, all depends upon the judgment of man to
select such of its parts as he wishes, and by his skill he adapts and
applies them in the best manner for his purposes. In eating the grapes,
he throws away the skins and seeds; for raisins, he evaporates the
water, retaining only the solid parts, from which, when he uses them,
he rejects their seeds. If he manufactures must, he lets the skins
remain. In making wine, he sets free the carbonic acid contained in the
must, and removes the lees, gum, tartar, and, in short, everything
deposited during, and immediately after fermentation, as well as when
it is put into casks and bottles. He not only removes from the wine its
sediments, but watches the fermentation, and checks it as soon as its
vinous fermentation is over, and the formation of vinegar about to
begin. He refines his wine by an addition of foreign substances if
necessary; he sulphurizes it; and, by one means or another, remedies
its distempers.

"The manufacture of wine is thus a many-sided art; and he who does not
understand it, or knows not how to guide and direct the powers of
Nature to his own purposes, may as well give up all hopes of success in

So far DR. GALL; and to the intelligent and unbiased mind, the truth
and force of these remarks will be apparent, without further extending
or explaining them. How absurd, then, the blind ravings of those who
talk about "natural" wines, and would condemn every addition of sugar
and water to the must by man, when Nature has not fully done her part,
as adulteration and fraud. Why, there is no such thing as a "natural
wine;" for wine--good wine--is the product of art, and a manufacture
from beginning to end. Would we not think that parent extremely cruel,
as well as foolish, who would have her child without clothing, simply
because Nature had allowed it to be born without it? Would not the
child suffer and die, because its mother failed to aid Nature in her
work, by clothing and feeding it when it is yet unable to feed and
clothe itself? And yet, would not that wine-maker act equally foolish
who has it within his power to remedy the deficiencies of Nature with
such means as she herself supplies in good season, and which ought and
would be in the must but for unfavorable circumstances, over which we
have no control? Wine thus improved is just as pure as if the sugar and
water had naturally been in the grapes in right proportions; just as
beneficial to health; and only the fanatical "know-nothing" can call it
adulterated. But the prejudices will disappear before the light of
science and truth, however much ignorance may clamor against it.
GALILEO, when forced to abjure publicly his great discovery of the
motion of the earth around the sun as a heresy and lie, murmured
between his teeth the celebrated words, "And yet it moves." It _did_
move; and the theory is now an acknowledged truth, with which every
schoolboy is familiar. Thus will it be with improved wine-making. It
will yet be followed, generally and universally, as sure as the public
will learn to distinguish between good and poor wine.

Let us now observe for a moment the change which fermentation makes in
converting the must into wine. The nitrogeneous compounds--vegetable
albumen, gluten--which are contained in the grape, and which are
dissolved in the must as completely as the sugar, under certain
circumstances turn into the fermenting principle, and so change the
must into wine. This change is brought about by the fermenting
substance coming into contact with the air, and receiving oxygen from
it, in consequence of which it coagulates, and shows itself in the
turbid state of must, or young wine. The coagulation of the lees takes
place but gradually, and just in the degree the exhausted lees settle.
The sugar gradually turns into alcohol. The acids partly remain as
tartaric acid, are partly turned into ether, or settle with the lees,
chrystallize, and adhere to the bottom of the casks. The etheric oil,
or aroma, remains, and develops into bouquet; also the tannin, to a
certain degree. The albumen and gluten principally settle, although a
small portion of them remains in the wine. The coloring matter and
extractive principle remain, but change somewhat by fermentation.

Thus it is the must containing a large amount of sugar needs a longer
time to become clear than that containing but a small portion of it;
therefore, many southern wines retain a certain amount of sugar
undecomposed, and they are called _sweet_, or liqueur wines; whereas,
wines in which the whole of the sugar has been decomposed are called
_sour_ or _dry_ wines.

I have thought it necessary to be thus explicit to give my readers an
insight into the general principles which should govern us in
wine-making. I have quoted freely from the excellent work of DR. GALL.
We will now see whether and how we can reduce it to practice. I will
try and illustrate this by an example.


"Experiments continued for a number of years have proved that, in
favorable seasons, grape juice contains, on the average, in 1,000 lbs.:

Sugar,   240 lbs.
Acids,     6  "
Water,   754  "
       1,000  "

This proportion would constitute what I call a normal must. But now we
have an inferior season, and the must contains, instead of the above
proportions, as follows:

Sugar,   150 lbs.
Acids,     9  "
Water,   841  "
       1,000  "

What must we do to bring such must to the condition of a normal must?
This is the question thus arising. To solve it, we calculate thus: If,
in six pounds of acids in a normal wine, 240 pounds of sugar appear,
how much sugar is wanted for nine pounds of acids? Answer, 360 pounds.
Our next question is: If, in six pounds of acids in a normal must, 754
pounds of water appear, how much water is required for nine pounds of
acids? Answer, 1,131 pounds. As, therefore, the must which we intend to
improve by neutralizing its acids, should contain 360 pounds of sugar,
nine pounds of acids, and 1,131 pounds of water, but contains already
150 pounds of sugar, 9 pounds of acids, and 841 pounds of water, there
remain to be added, 210 pounds of sugar, no acids, and 290 pounds of

By ameliorating a quantity of 1,000 pounds must by 210 pounds sugar,
and 290 pounds water, we obtain 1,500 pounds of must, consisting of the
same properties as the normal must, which makes a first-class wine."

This is wine-making, according to GALL'S method, in Europe. Now, let us
see what we can do with it on American soil, and with American grapes.


If we examine the must of most of our American wine grapes closely, we
find that they not only contain an excess of acids in inferior seasons,
but also a superabundance of flavor or aroma, and of tannin and
coloring matter. Especially of flavor, there is such an abundance that,
were the quantity doubled by addition of sugar and water, there would
still be an abundance; and with some varieties, such as the Concord, if
fermented on the husks, it is so strong as to be disagreeable. We must,
therefore, not only ameliorate the acid, but also the flavor and the
astringency, of which the tannin is the principal cause. Therefore it
is, that to us the knowledge of how to properly gallize our wines is
still more important than to the European vintner, and the results
which we can realize are yet more important. By a proper management, we
can change must, which would otherwise make a disagreeable wine, into
one in which everything is in its proper proportion, and which will
delight the consumer, to whose fastidious taste if would otherwise have
been repugnant. True, we have here a more congenial climate, and the
grapes will generally ripen better, so that we can in most seasons
produce a drinkable wine. But if we can increase the quantity, and at
the same time improve the quality, there is certainly an inducement,
which the practical business sense of our people will not fail to
appreciate and make use of.

There is, however, one difficulty in the way. I do not believe that the
acidimeter can yet be obtained in the country, and we must import them
direct from the manufacturers, DR. L. C. MARQUART, of Bonn, on the
Rhine; or J. DIEHN, Frankfort-on-the-Main.

However, this difficulty will soon be overcome; and, indeed, although
it is impossible to practice gallizing without a saccharometer, we may
get at the surplus of acids with tolerable certainty by the results
shown by the saccharometer. To illustrate this, I will give an example:

Last year was one of the most unfavorable seasons for the ripening of
grapes we have ever had here, and especially the Catawba lost almost
nine-tenths of its crop by mildew and rot; it also lost its leaves, and
the result was, that the grapes did not ripen well. When gathering my
grapes, upon weighing the must, I found that it ranged from 52° to 70°;
whereas, in good seasons, Catawba must weighs from 80° to 95°. I now
calculated thus: if normal must of Catawba should weigh at least 80°,
and the must I have to deal with this season will weigh on an average
only 60°, I must add to this must about 1/2 lb. of sugar to bring it up
to 80°. But now I had the surplus acid to neutralize yet. To do this, I
calculated thus: If, even in a normal Catawba must, or a must of the
best seasons, there is yet an excess of acid, I can safely count on
there being at least one-third too much acid in a must that weighs but
60°. I, therefore, added to every 100 gallons of must 40 gallons of
soft water, in which I had first dissolved 80 lbs. of crushed sugar,
which brought the water, when weighed after dissolving the sugar in it,
up to 80°. Now, I had yet to add 50 lbs., or half a pound to each
gallon of the original must, to bring _this_ up to 80°. I thus pressed,
instead of 100 gallons, 150 gallons, from the same quantity of grapes;
and the result was a wine, which every one who has tasted it has
declared to be excellent Catawba. It has a brilliant pale yellow color,
was perfectly clear 1st of January, and sold by me to the first one to
whom I offered it, at a price which I have seldom realized for Catawba
wine made in the best seasons, without addition of sugar or water.
True, it has not as strong an aroma as the Catawba of our best seasons,
nor has it as much astringency; but this latter I consider an
advantage, and it still has abundant aroma to give it character.

Another experiment I made with the Concord satisfied me, without
question, that the must of this grape will always gain by an addition
of water and sugar. I pressed several casks of the pure juice, which,
as the Concord had held its leaves and ripened its fruit very well,
contained sugar enough to make a fair wine, namely, 75°. This I
generally pressed the day after gathering, and put into separate casks.
I then took some must of the same weight, but to which I had added, to
every 100 gallons, 50 gallons of water, in which I had diluted sugar
until the water weighed 75°, or not quite two pounds of sugar to the
gallon of water, pressed also after the expiration of the same time,
and otherwise treated in the same manner. Both were treated exactly
alike, racked at the same time; and the result is, that every one who
tries the two wines, without knowing how they have been treated,
prefers the gallized wine to the other--the pure juice of the grape. It
is more delicate in flavor, has less acidity, and a more brilliant
color than the first, the ungallized must. They are both excellent, but
there is a difference in favor of the gallized wine.

DR. GALL recommends grape sugar as the best to be used for the purpose.
This is made from potato starch; but it is hard to obtain here, and I
have found crushed loaf sugar answer every purpose. I think this sugar
has the advantage over grape sugar, that it dissolves more readily, and
can even be dissolved in cold water, thus simplifying the process very
much. It will take about two pounds to the gallon of water to bring
this up to 80°, which will make a wine of sufficient body. The average
price of sugar was about 22 cents per pound, and the cost of thus
producing an additional gallon of wine, counting in labor, interest on
capital, etc., will be about 60 cents. When the wine can be sold at
from $2 to $3 per gallon, the reader will easily perceive of what
immense advantage this method is to the grape-grower, if he can thereby
not only improve the quality, but also increase the quantity of the

The efforts made by the Commissioner of Patents, and the contributors
to the annual reports from the Patent Office, to diffuse a general
knowledge of this process, can therefore not be commended too highly.
It will help much to bring into general use, among all classes, good,
pure, native wines; and as soon as ever the poorer classes can obtain
cheap agreeable wines, the use of bad whiskey and brandy will be
abandoned more and more, and this nation will become a more temperate

But this is only the first step. There is a way to still further
increase the quantity. DR. GALL and others found, by analyzing the
husks of the grape after the juice had been extracted by powerful
presses, that they not only still contained a considerable amount of
juice, but also a great amount of extracts, or wine-making principles,
in many instances sufficient for three times the bulk of the juice
already expressed. This fact suggested the question: As there are so
many of these valuable properties left, and only sugar and water
exhausted, why cannot these be substituted until the others are
completely exhausted? It was found that the husks still contained
sufficient of acids, tannin, aroma, coloring matter, and gluten. All
that remained to be added was water and sugar. It was found that this
could be easily done; and the results showed that wine made in this
manner was equal, if not superior, to some of that made from the
original juice, and was often, by the best judges, preferred to that
made from the original must.

I have also practiced this method extensively the last season; and the
result is, that I have fully doubled the amount of wine of the Norton's
Virginia and Concord. I have thus made 2,500 gallons of Concord, where
I had but 1,030 gallons of original must; and 2,600 gallons of Norton's
Virginia, where I had but 1,300 gallons of must. The wines thus made
were kept strictly separate from those made from the original juice,
and the result is, that many of them are better, and none inferior, to
the original must; and although I have kept a careful diary of
wine-making, in which I have noted the process how each cask was made,
period of fermentation on the husks, quantity of sugar used, etc., and
have not hesitated to show this to every purchaser after he had tasted
of the wine, they generally, and with very few exceptions, chose those
which had either been gallized in part, or entirely.

[Illustration: FIG. 37.
UNION VILLAGE.--_Berries 1/3 diameter._]

My method in making such wines was very simple. I generally took the
same quantity of water, the husks had given original must, or in other
words, when I had pressed 100 gallons of juice, I took about 80 gallons
of water. To make Concord wine, I added 1-3/4 lbs. of sugar to the
gallon, as I calculated upon some sugar remaining in the husks, which
were not pressed entirely dry. This increased the quantity, with the
juice yet contained in the husks to 100 gallons, and brought the water
to 70; calculating that from 5° to 10° still remained in the husks, it
would give us a must of about 80°. The grapes, as before remarked, had
been gathered during the foregoing day, and were generally pressed in
the morning. As soon as possible the husks were turned into the
fermenting vat again, all pulled apart and broken, and the water added
to them. As the fermentation had been very strong before, it
immediately commenced again. I generally allowed them to ferment for
twenty-four hours, and then pressed again, but pressed as dry as
possible this time. The whole treatment of this must was precisely
similar to that of the original.

In making Norton's Virginia, I would take, instead of 1-3/4 lbs., 2
lbs. of sugar to the gallon--as it is naturally a wine of greater body
than the Concord--and I aimed to come as near to the natural must as
possible. I generally fermented this somewhat longer, as a darker color
was desired. The time of fermentation must vary, of course, with the
state of the atmosphere; in cooler weather, both pressings should
remain longer on the husks. The results, in both varieties were wines
of excellent flavor, good body, a brilliant color, with enough of
tannin or astringency, and sufficient acid--therefore, in every way

The experiments, however, were not confined to these alone, but
extended over a number of varieties, with good results in every case.
Of all varieties tried, however, I found that the Concord would bear
the most of gallizing, without losing its own peculiar flavor; and I
satisfied myself, that the quantity in this grape can safely be
increased _here_, from 100 gallons of must to 250 gallons of wine, and
the quality yet be better, than if the must had been left in its normal

And it is here again where only experience can teach us _how far_ we
can go with a certain variety. It must be clear and apparent to any one
who is ever so slightly acquainted with wine-making, how widely
different the varieties are in their characteristics and ingredients.
We may lay it down as a general rule, however, that our native grapes,
with their strong and peculiar flavors, and their superabundance of
tannin and coloring matter, will admit of much more gallizing, than the
more delicately flavored European kinds.

I have thus tried only to give an outline of the necessary operations,
as well as the principles lying at the foundation of them. I have also
spoken only of facts as I have found them, as I am well aware that this
is a field in which I have much to learn yet, and where it but poorly
becomes me to act the part of teacher. Those desiring more detailed
information, I would refer to the Patent Office Reports of 1859-60,
where they will find valuable extracts from the works of DR. GALL; and
also to the original works.

If we look at the probable effect these methods of improving wines are
likely to have upon grape-culture, it is but natural that we should ask
the question: Is there anything reprehensible in the practice--any
reason why it should not become general? The answer to this is very
simple. They contain nothing which the fermented grape juice, in its
purest and most perfect state does not also contain. Therefore, they
are as pure as any grape juice can be, with the consideration in their
favor, that everything is in the right proportion. Therefore, if wine
made from pure grape juice can be recommended for general use, surely,
the gallized wines can also be recommended. DR. GALL has repeatedly
offered to pay a fine for the benefit of the poor, if the most critical
chemical analysis could detect anything in them, which was injurious to
health, or which pure wines ought not to contain, and his opponents
have always failed to show anything of the kind.

I know that some of my wine-making friends will blame me for thus
"letting the cat out of the bag." They seem to think that it would be
better to keep the knowledge we have gained, to ourselves, carefully
even hiding the fact that any of our wines have been gallized. But it
has always been a deep-seated conviction with me, that knowledge and
truth, like God's sun should be the common property of all His
children--and that it is the duty of every one not to "hide his light
under a bushel," but seek to impart it to all, who could, perhaps, be
benefitted by it. And why, in reality, should we seek to keep as a
secret a practice which is perfectly right and justifiable? If there is
a prejudice against it, (and we know there is), this is not the way to
combat it. Only by meeting it openly, and showing the fallacy of it,
can we hope to convince the public, that there is nothing wrong about
it. Truth and justice need never fear the light--they can only gain
additional force from it. I do not even attempt to sell a cask of
gallized wine, before the purchaser is made fully acquainted with the
fact, that it has been gallized.

It is a matter of course, that many, who go to work carelessly and
slovenly, will fail to make good wine, in this or any other way. To
make a good article, the nature of each variety and its peculiarities
must be closely studied--we must have as ripe grapes as we can get,
carefully gathered; and we need not think that water and sugar will
accomplish _everything_. There is a limit to everything, and to
gallizing as well as to anything else. As soon as we pass beyond that
limit, an inferior product will be the result.

But let us glance a moment at the probable influence this discovery
will have on American grape culture. It cannot be otherwise than in the
highest degree beneficial; for when we simply look at grape-culture as
it was ten years ago, with the simple product of the Catawba as its
basis; a variety which would only yield an average of, say 200 gallons
to the acre--often very inferior wine--and look at it to-day, with such
varieties as the Concord, yielding an average of from 1,000 to 1,500
gallons to the acre, which we can yet easily double by gallizing, thus
in reality yielding an average of 2,500 gallons to the acre of
uniformly good wine; can we be surprised if everybody talks and thinks
of raising grapes? Truly, the time is not far distant--of which we
hardly dared to dream ten years ago--and which we _then_ thought we
would never live to see; when _every_ American citizen can indulge in a
daily glass of that glorious gift of God to man, pure, light wine; and
the American nation shall become a really _temperate_ people.

And there is room for all. Let every one further the cause of
grape-culture. The laborer by producing the grapes and wine; the
mechanic by inventions; the law-giver by making laws furthering its
culture, and the consumption of it; and _all_ by drinking wine, in wise
moderation of course.


Some of my readers may think I did not look much to this, which I told
them was one of the objects of this little work. To vindicate it and
myself I will here state, that our object should always be to attain
the highest perfection in everything. But, while I am aware that I have
generally given the outline of operations on a large scale, I have
never for a moment lost sight of the interests of those, who, like
myself, are compelled, by bitter necessity, to commence at the lowest
round of the ladder. And how could I forget the bitter experience of my
first years, when hindered by want of means; but also the feelings of
sincere joy, of glad triumph, when I had surmounted one more obstacle,
and saw the path open wider before me at every step; and I can,
therefore, fully sympathize with the poor laborer, who has nothing but
his industrious hands and honest will to commence with.

While, therefore, it is most advantageous to follow grape-growing and
wine-making with all the conveniences of well prepared soil,
substantial trellis, a commodious wine cellar and all its
appurtenances; yet, it is also possible to do without most of these
conveniencies in the beginning, and yet succeed. If the grape-grower
has not capital to spare to buy wire, he can, if he has timber on his
land, split laths and nail them to the posts instead of wire. He can
layer his plants even the first summer, and thus raise a stock for
further planting; or dispose of them, as already mentioned in the
beginning of this work. Or he can lease a piece of land from some one
who wishes to have a vineyard planted on it, and who will furnish the
plants to him, besides the necessary capital for the first year or so.
I have contracted with several men without means in this manner,
furnished them a small house, the necessary plants, and paid them $150
the first two years, they giving me half the returns of the vineyards,
in plants and grapes; and they have become wealthy by such means. One
of my tenants has realized over $8,000 for his share the last season,
and will very likely realize the same amount next season.

And if he cannot afford to build a large cellar in the beginning, he
can also do with a small one, even the most common house cellar will do
through the winter, if it is only kept free from frost. One of our most
successful wine-growers here, commenced his operations with a simple
hole in the ground, dug under his house, and his first wine press was
merely a large beam, let into a tree, which acted as a lever upon the
grapes, with a press-bed, also of his own making. A few weeks ago the
same man sold his last year's crop of wine for over $9,000 in cash, and
has raised some $2,000 worth more in vines, cuttings, etc. Of course,
it is not advisable to keep the wine over summer in an indifferent
cellar, but during fermentation and the greater part of winter, it will
answer very well, and he can easily dispose of his wine, if good, as
soon as clear. Or he can dispose of his grapes at a fair price, to one
of his neighbors, or take them to market.

But there is another consideration, which I cannot urge too strongly
upon my readers, and which will do much to make grape-growing and
wine-making easy. It is the forming of grape colonies, of
grape-growers' villages. The advantages of such a colony will be easily
seen. If each one has a small piece of suitable land, (and he does not
need a large one to follow grape-growing), the neighbors can easily
assist each other in ploughing and sub-soiling; they will be able to do
with fewer work animals, as they can hitch together, and first prepare
the soil for one and then for the other; the ravages of birds and
insects will hardly be felt; they can join together, and build a large
cellar in common, where each one can deliver and store his wine, and of
which one perhaps better acquainted with the management of wine than
the others, and whom all are willing to trust, can have the management.
If there should be no such man among them, an experienced cooper can be
hired by all, who can also manufacture the necessary casks. An
association of that kind has also, generally, the preference in the
market over a single individual, and they are able to obtain a higher
price for their products, if they are of good quality.

There are thousands upon thousands of acres of the best grape lands yet
to be had in the West, especially in Missouri, at a merely nominal
price, which would be well adapted for settlements of that kind; where
the virgin soil yet waits only the bidding of intelligent labor--of
enterprising and industrious men--to bring forth the richest fruits.
There is room for all--may it soon be filled with willing hearts to
undertake the task.

And how much easier for you to-day, men with the active hand and
intelligent brain, to commence--with the certainty of success before
you--with varieties which will yield a large and sure return _every_
year; with the market open before you, and the experience of those who
have commenced, to guide you; with the reputation of American wines
established; with double the price per gallon--and ten times the
yield--compared with the beginner of only ten years ago, with nothing
but uncertainty; uncertainty of yield, uncertainty of quality, of
price, and of effecting a sale.

It took a brave heart _then_, and an iron will; the determination to
succeed,--succeed against _all_ obstacles. And yet, hundreds have
commenced thus, and have succeeded. Can _you_ hesitate, when the future
is all bright before you, and the thousand and one obstacles have been
overcome? If you do, you are not fit to be a grape-grower. Go toil and
drudge for so many cents per day, in some factory, and end life as you
have begun it. God's free air, the cultivation of one of His noblest
gifts, destined to "make glad the heart in this rugged world of ours,"
is not for you. I may pity you, but I cannot sympathize with nor assist
you, except by raising a cheap glass of wine to gladden even _your_
cheerless lot.

[Illustration: FIG. 38.
MAXATAWNY.--_Berries 1/2 diameter._]



In this, of course, allowances must be made for soil, locality, cost of
plants, cost of timber, etc., which will vary with the locality. The
estimation given here is about what it would cost _here_, with the
leading varieties.


Preparing ground by ploughing, laying off, etc.,              $ 50 00
700 first-class yearling plants, to be planted 6×10,
  $12 per hundred,                                              84 00
450 posts, 15 feet apart, 10 cents each,                        45 00
450 intermediate stakes, 3     "                                13 50
600 lbs. No. 12 wire, 16 cents per lb.,                         96 00
Cost of erecting trellis,                                       50 00
Attendance, labor, etc., during first year,                     50 00
Interest on capital,                                            20 00
                                                              $408 50

The following year the vineyard can be made to pay all expenses, by
layering, etc.


Preparing ground,                                               50 00
700 first class plants, 6×10, $25 per hundred,                 175 00
450 posts, 10 cents each,                                       45 00
450 stakes, 3  "                                                13 50
600 lbs. wire, 16 cents per lb.,                                96 00
Cost of erecting trellis,                                       50 00
Attendance, labor, during first two years,                     125 00
Interest on capital during first two years,                     66 00
                                                              $620 50


Preparation of soil, etc.,                                      50 00
850 plants, first class, to be planted 6×8, $25 per hundred,   212 50
450 posts, 10 cents each,                                       45 00
450 stakes, 3   "                                               13 50
600 lbs. No. 12 wire, 16 cents per lb.,                         96 00
Cost of erecting trellis,                                       50 00
Attendance, labor, etc., during first two years,               125 00
Interest on capital during first two years, at
  6 per cent. per annum,   70 00                               ------
                                                              $662 00


Cost of preparing ground,                                       50 00
1,200 first-class plants, planted 6×6,                         400 00
450 posts, 10 cents each,                                       45 00
450 stakes, 3   "                                               13 50
600 lbs. No. 12 wire, 16 cents per lb.                          96 00
Cost of erecting trellis,                                       50 00
Cost of cultivation two first years,                           125 00
Interest on capital two years,                                  92 00
                                                              $871 50


Preparing ground,                                               50 00
Cost of 1,200 plants, 6×6,                                      45 00
450 posts, 10 cents each,                                       45 00
450 stakes, 3   "                                               13 50
600 lbs. wire, 16 cents per lb.,                                96 00
Cost of erecting trellis,                                       50 00
Attendance during two years,                                   125 00
Interest on capital two years,                                  39 00
                                                              $463 50


The following has been the produce of a vineyard of Catawba, now under
my management, since 1849:

      Bearing        Vines           Gallons of            Yield per
      season.       bearing.           Wine.      Price.     acre.

1849,  1st year,    1,500              750       $1.25      $600 00
1850,  2d   "       2,000              150        1.25        95 00
1851,  3d   "       2,000              500        1.25       300 00
1852,  4th  "       1,800              210        1.25       120 00
1853.  5th  "       1,500              580        1.25       500 00
1854,  6th  "       2,500              750        1.50       600 00
1855,  7th  "       3,000              230        2.00       150 00
1856   8th  "       4,000              150        2.00        75 00
1857   9th  "       4,000            2,000        1.20       600 00
1858, 10th  "       4,000              210        1.20        60 00
1859, 11th  "       4,200            1,200        1.20       360 00
1860, 12th  "       4,200            1,300        1.25       405 00
1861, 13th  "       4,200              150        1.00        37 50
1862, 14th  "       4,200               20        2.00        10 00
1863, 15th  "       4,200              150        2.00        75 00
1864, 16th  "       4,200              150        2.00        75 00
1865, 17th  "       4,200              500        2.00       250 00

Which will show the average yield per acre,
  to have been somewhat over                                 250 00
Deduct from this cost of labor per year,
  per acre,                         50 00
Interest on capital,                40 00-90 00
Would leave a clear profit, per acre, of                     160 00

The poor returns were nearly all occasioned by mildew and rot, with the
exception of 1862, when a very destructive hail-storm swept away almost
the entire crop; and in 1864, when the vines were all killed down to
the snow line by frost the preceding winter.

The following is the cost of a vineyard planted by me in May, 1861,
containing about 3,000 vines, on 2-1/2 acres of ground. The ground
could not be made ready until late in the season, consequently many of
the vines failed to grow, and had to be replanted the second season:

1700 Norton's Virginia, $20.00 per hundred,                    340 00
 400 Concord (small),       25       "                         100 00
 350 Delaware,              50       "                         175 00
 150 Herbemont,             25       "                          37 50
  50 Cunningham,            50       "                          25 00
Other varieties assorted,                                      100 00
Cost of clearing, ploughing, and planting, $50 per acre,       125 00
Putting up trellis, $150 per acre,                             375 00
Interest on capital,                                           100 00
                                                            $1,377 50


For layers and cuttings made 1st year,                         339 00
      "           "          2d    "                          1200 00
      "           "          3d    "                          2500 00
Concord grapes sold, 2,000 lbs., net 16 cents,                 320 00
Plants and cuttings fourth year,                              4000 00
2,040 lbs. of grapes (Concord), marketed at
  24 cents per lb., net    489 60
                                                            $8,848 60


1,030 gallons Concord at        $2.50                       $2,575 00
1,300    "    Norton's Virginia  4.00                        5,200 00
  125    "    Herbemont          3.00                          375 00
   30    "    Cunningham         4.00                          120 00
   40    "    Delaware           6.00                          240 00
   10    "    Clinton            3.00                           30 00
   50    "    Other Varieties    3.00                          150 00
  336    "    Hartford Prolific Grapes 20 cts. per lb.          67 20
57,000 Plants from cuttings and layers, average
  price $100 per thousand                                    5,700 00
                                                           $14,457 20

Leaving the product of the first five years                $23,305 80

From which deduct expenses for plants,
  trellis, etc.,                       1,277
Interest on capital at 5 per cent.       500
Cost of labor 1st. year,                 150
              2d.   "                    300
              3d.   "                    400
              4th.  "                    500
              5th.  "                    500
Total Cost                                                     $3,627
Leaves clear profit for first five years of                $19,679 80

The fourth year, nearly all the fruit buds of the vines had been killed
above the snow line, but I made, besides the grapes sold, about $1,500
worth of wine, which was emptied by the rebels in their raid that fall,
and consequently lost. The vines were not all in bearing this last
season, for reasons already given; and the whole amount of vines
bearing, was not more than 2,200--hardly two acres. If my readers will
contrast this with the yield of the Catawba vineyard, they will see the
difference in yield between varieties suited to the climate and soil,
and those unused to it.

The last season--although unfavorable to the Catawba--produced an
enormous yield of Concord and Norton's Virginia, and cannot be taken as
an average crop. I think about 700 gallons of Norton's Virginia, and
1,200 gallons of Concord would be a fair average estimate per
year--which the vines can easily produce, and remain healthy and


Year after planting.      Acres in Vines.        Yield.        Price.

1847,           2d            5-6                24 gallons      2.00
1848,           3d            3-6             1,000   "          2.00
1849,           4th           2                 600   "          1.50
1850,           5th           2                 350   "          1.25
1851,           6th           2-1/2             450   "          1.75
1852,           7th           2-1/2             500   "          1.50
1853,           8th           2-1/2             350   "          2.00
1854,           9th           3-1/2             800   "          2.00
1855,          10th           3-1/2              50   "          1.50
1856,          11th           3-1/2           1,000   "          1.25
1857,          12th           6               4,500   "          1.50
1858,          13th           6               1,100   "          1.75
1859,          14th           6               1,500   "          1.50
1860,          15th           6               2,000   "          1.25
1861,          16th           6                 250   "          1.00
1862,          17th           6                 300   "          1.50
1863,          18th           8               2,000   "          1.15


500 Gallons Norton's Virginia--2 acres, at $3 per gallon    $1,500 00
Grapes sold from 1/2 acre of Concords                          400 00
Plants from cuttings and layers sold                         2,000 00
                                                            $3,900 00


2 Acres of Norton's Virginia produced 600 gallons,
  at $4 50                                                  $2,700 00
2-1/2 Acres of Catawba, produced 400 gallons, at $2 15         850 00
Grapes sold from 1/2 acre of Concord                           400 00
Plants sold                                                  1,500 00
                                                            $5,450 00


2-3/4 Acres of Norton's Virginia, produced
    2,000 gallons at $4                                      8,000 00
2-1/2 Acres Catawba, produced 450 gallons at $1 75             787 50
1-1/4 Acres Concord, produced 1,000 gallons, at $250         2,500 00
  1/2 acre Herbemont produced 400 gallons,
    at $3 per gallon,                                        1,200 00
  1/2 acre Rulander produced 50 gallons, at $5                 250 00
Plants sold,                                                 1,500 00
                                                           $14,237 50

This vineyard was trenched at an average cost of $120 dollars to the
acre, and most of the vines are planted 5×5, evidently too close. They
are trained to wire trellis, as described in a former part of this
work, and receive close attention, and the very best cultivation.


1-1/2 acres of Catawba produced 1,050 gallons of wine;
  sold at                                                    1,402 50


1-3/4 acres of Catawba produced 250 gallons;
  sold at $1.10 per gallon,                                    275 00


1-3/4 acres Catawba produced 300 gallons;
  sold at $1.25 per gallon,                                    375 00


2 acres of Catawba produced 8,843 lbs. of grapes;
  sold at 10c. per lb.,                                        884 30
120 gallons of wine, at $1.20 per gallon,                      144 00
230    "                 0.95    "                             218 50
Plants sold,                                                   600 00
                                                            $1,846 80


2 acres of Catawba produced 270 gallons,
  at $1.05 per gallon,                                         283 50
Plants sold,                                                   500 00
                                                              $783 50


2 acres Catawba produced 6,718 lbs. of grapes;
  sold at 9 cents per lb.,                                     604 62
225 gallons of wine, sold at $1.25 per gallon,                 281 25
 75   "     of Norton's Virginia, from about
  1-10th of an acre, at $2.75 per gallon,                      206 25
Plants sold,                                                   650 00
                                                            $1,742 12

1863--2-1/4 ACRES IN ALL.

720 gallons of Catawba, at $1.85 per gallon,                 1,332 00
 60   "        Concord, at $2.00       "                       120 00
 70   "        Herbemont, at $2        "                       140 00
 40   "        Norton's Virginia, $3   "                       120 00
Plants sold,                                                   800 00
                                                            $2,512 00


45 gallons Catawba, $2.00 per gallon,                           90 00
42   "     Concord,  2.50       "                              105 00
20   "     Norton's Virginia and Delaware mixed,
  at $5.25 per gallon,                                         105 00
10   "     Norton's Virginia, second class, at $3               30 00
Plants sold,                                                   300 00
                                                              $630 00


2-1/2 acre Catawba produced 900 galls., at $1.75,            1,575 00
  1/2  "   Concord     "    700   "         2.50,            1,750 00
1      "   Norton's Vir. "  600   "         4.00,            2,400 00
  1/3  "   Delaware    "    120   "         5.00,              600 00
  1/2  "   Herbemont   "    350   "         2.50,              875 00
Balance in other varieties,                                    150 00
Plants sold,                                                   940 00
                                                            $8,290 00

This vineyard has one of the best locations for Catawba and Delaware in
the neighborhood, and its proprietor one of the most intelligent and
industrious cultivators and wine-manufacturers in the vicinity.

The following are copied from the report of a special committee
appointed by the Cincinnati Horticultural Society, to inquire into the
condition of vineyards, and report whether or not grape-growing was
still profitable. I regret to say that our Cincinnati friends have not,
generally speaking, paid as much attention to the introduction and
testing of better varieties--and there are but few vineyards in that
neighborhood--where any other variety than the Catawba has been planted
to any extent. It is to be hoped that the signal failure of that variety
last season will do much to open their eyes to the full importance of
the subject, and to abandon the Catawba, which evidently will not pay
any longer.

But, as we have already said, there are other varieties of grapes being
successfully grown in this vicinity, and we have extended our researches
to some of those vineyards, and give the results as follows:--

Ives' Seedling is a grape of much promise, not addicted to mildew and
rot. Col. WAHRING, of Indian Hill, in this county, has a small vineyard,
only two acres in bearing, which made, the past season, 650 gallons of
wine. The season previous, only one acre in bearing, yielded 560
gallons. The Colonel makes his account for the past season's business
stand as follows:--

650 gallons of wine, sold at $4.10 per gallon,    $2,665 00
Sale of cuttings,                                  1,500 00
                                                  $4,165 00
Deduct cost of taking care of vineyard,              100 00
Leaving net product of vineyard,                  $4,065 00
  Or over $2,000 per acre.

Norton's Virginia is another promising grape that is being grown
considerably hereabouts.

The Messrs. BOGEN have given us their figures for the product of this
grape, as follows:

1863--From 1-1/2 acres, first year in bearing,
  they made 500 gallons, sold at $3 per gallon,   $1,500 00
Sale of cuttings,                                    400 00
Sale of roots from layers,                           800 00
                                                  $2,700 00
Deduct from this, for cost of culture,               100 00
Leaves net,                                       $2,600 00
  Or $1,733 per acre.

1864--Yield of same in wine and cuttings,          2,300 00
  Or about $1,500 per acre.

Delaware is another grape of very great promise and profit, now being
extensively grown throughout the country. The Messrs. BOGEN, from
one-third of an acre, first bearing year, give us the following figures
for the past season:

87 gallons of wine, sold at $6 per gallon,           522 00
Sold cuttings,                                       450 00
Sold roots from layers,                            2,050 00
                                                  $3,022 00
Deduct cost of culture,                               22 00
                                                  $3,000 00
  Or $9,000 per acre.

Mr. J. E. MOTTIER gives us, as the result of his Delaware vineyard for
the past two years, as follows:

1863--FROM 1-1/2 ACRES.

165 gallons of wine, sold at $5 per gallon,         $825 00
Sale of cuttings,                                  1,630 00
                                                   2,455 00
Deduct expenses,                                     200 00
Leaving net,                                      $2,255 00
  Or $1,504 per acre.


200 gallons of wine, at $6 per gallon,            $1,200 00
Sold roots from layers,                            1,835 00
Sales of cuttings,                                 2,360 00
                                                   5,395 00
Deduct expenses,                                     200 00
Leaves net,                                       $5,195 00
  Or $3,562 per acre

Mr. MOTTIER says he might have obtained a larger yield of wine, but
his vineyard being young, he would not allow it to overbear.

Your committee, therefore, take pleasure in submitting the foregoing
facts, in refutation, in part, of the loose and reckless statements of
Mr. YEATMAN, and take this method of entering their protest against the

(Signed), E. A. THOMPSON.
          JOHN E. MOTTIER.

The foregoing contains some valuable facts, but it would seem to me
that our Cincinnati friends have hardly estimated labor and expenses
high enough. We cannot begin to cultivate our vineyards at as low an

The following is a rough estimate of the last season's crop around
Hermann. It may be rather inaccurate, but it is about as near as I
could come to the result. There are now, I suppose, something like
1,000 acres planted in grapes, of which about 400 may be in bearing.
Unfortunately, nearly all the old vineyards are planted with the
Catawba, which was almost an entire failure this season, the average
crop being only about 75 gallons to the acre. Most of the later
planting has been done with the Concord and Norton's Virginia, but
these vineyards are not bearing yet. Of the Norton's Virginia, the
average crop the last season may have been about 600 gallons to the
acre; of the Concord, 1,000 gallons per acre. The Herbemont may have
yielded about 800 gallons to the acre.

Grapes marketed, mostly Concord, 20,000 lbs.
  average price, 15c. per lb.,                              $3,000 00
Catawba wine made, about 25,000 gallons;
  average value, $1.50 per gallon,                          37,500 00
Norton's Virginia wine made, about 10,000 gallons;
  average value, $4 per gallon,                             40,000 00
Concord wine made, about 5,000 gallons;
  average value, $2.50 per gallon,                          12,500 00
Herbemont wine made, about 1,500 gallons;
  average value, $3 per gallon,                              4,500 00
Other varieties made, about 1,000 gallons;
  average value, $3 per gallons,                             3,000 00
Grape roots, cuttings, etc., grown and sold,                50,000 00
                                                          $150,500 00

I think the above is rather below the real amount; and the value of the
crop may come up even as high as $200,000.

Although grape culture is followed to a larger extent around Hermann
than anywhere in the State, yet there are also a great many grapes
grown and wine made around Boonville, in Cooper County; and Augusta,
St. Charles County; also, Hannibal, on the Mississippi river; and St.
Joseph, on the Missouri; and there is hardly a county in the State now
but has some flourishing vineyards.

The above facts may serve to give my readers a clearer insight into the
cost and profits of grape-growing, and also the comparative varieties.
In every case, the figures given can be relied on as actual facts.

In our neighboring States, Illinois and Iowa, grape-growing is
progressing rapidly. There are already a number of vineyards established
in the neighborhood of Alton, Belleville, Mascoutah, Warsaw, and Nauvoo,
in Illinois; and in the neighborhood of Burlington and Davenport, in
Iowa. I am told that in the neighborhood of Makanda alone, in Jackson
County, Illinois, at least 70,000 vines of the Concord will be planted
the coming spring.

Our sister State, Kansas, is also progressing bravely in the good work;
and I do not think that, although our propagators throughout the
country have done their best, there will be half the number of vines
for sale that are wanted to meet the demand.

But, while I am fully aware of the importance of grape-culture
_everywhere_, I cannot help but believe that the southwest will
take the preference in grape-growing over the eastern and northern
States. We have the advantages of longer seasons and a warmer climate,
generally of richer soil, of cheaper lands; we can cultivate varieties
which cannot be grown by our eastern brethren, and therefore all the
chances are on our side. The mountainous regions of Tennessee, Georgia,
Arkansas, Texas, and Alabama may, perhaps, rival and even surpass us in
the future, but their inhabitants at present are not of the clay from
which grape-growers are formed. They still cling to the demon of
slavery, and their hatred of northern industrious _freemen_ seems
to be stronger than their love of prosperity. Let us hope that a better
spirit may prevail, that they will in time begin to see their own
interest, and welcome with open arms every one who can assist them in
developing the natural advantages of their lands. The grape can only
flourish on _free_ soil, and by _free_ intelligent labor.

*** End of this Doctrine Publishing Corporation Digital Book "The Cultivation of The Native Grape, and Manufacture of American Wines" ***

Doctrine Publishing Corporation provides digitized public domain materials.
Public domain books belong to the public and we are merely their custodians.
This effort is time consuming and expensive, so in order to keep providing
this resource, we have taken steps to prevent abuse by commercial parties,
including placing technical restrictions on automated querying.

We also ask that you:

+ Make non-commercial use of the files We designed Doctrine Publishing
Corporation's ISYS search for use by individuals, and we request that you
use these files for personal, non-commercial purposes.

+ Refrain from automated querying Do not send automated queries of any sort
to Doctrine Publishing's system: If you are conducting research on machine
translation, optical character recognition or other areas where access to a
large amount of text is helpful, please contact us. We encourage the use of
public domain materials for these purposes and may be able to help.

+ Keep it legal -  Whatever your use, remember that you are responsible for
ensuring that what you are doing is legal. Do not assume that just because
we believe a book is in the public domain for users in the United States,
that the work is also in the public domain for users in other countries.
Whether a book is still in copyright varies from country to country, and we
can't offer guidance on whether any specific use of any specific book is
allowed. Please do not assume that a book's appearance in Doctrine Publishing
ISYS search  means it can be used in any manner anywhere in the world.
Copyright infringement liability can be quite severe.

About ISYS® Search Software
Established in 1988, ISYS Search Software is a global supplier of enterprise
search solutions for business and government.  The company's award-winning
software suite offers a broad range of search, navigation and discovery
solutions for desktop search, intranet search, SharePoint search and embedded
search applications.  ISYS has been deployed by thousands of organizations
operating in a variety of industries, including government, legal, law
enforcement, financial services, healthcare and recruitment.