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

Download this book: [ ASCII | HTML | PDF ]

Look for this book on Amazon


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

Title: Mysteries of Bee-keeping Explained
Author: Quinby, M. (Moses), 1810-1875
Language: English
As this book started as an ASCII text book there are no pictures available.


*** Start of this LibraryBlog Digital Book "Mysteries of Bee-keeping Explained" ***


in Agriculture (CHLA), Cornell University)



MYSTERIES

OF

BEE-KEEPING EXPLAINED:

BEING A COMPLETE

ANALYSIS OF THE WHOLE SUBJECT;

CONSISTING OF

THE NATURAL HISTORY OF BEES, DIRECTIONS FOR OBTAINING THE GREATEST
AMOUNT OF PURE SURPLUS HONEY WITH THE LEAST POSSIBLE
EXPENSE, REMEDIES FOR LOSSES GIVEN, AND THE SCIENCE OF
"LUCK" FULLY ILLUSTRATED--THE RESULT OF MORE
THAN TWENTY YEARS' EXPERIENCE IN
EXTENSIVE APIARIES.



BY M. QUINBY,

PRACTICAL BEE-KEEPER.



NEW YORK:

C. M. SAXTON, AGRICULTURAL BOOK PUBLISHER
152 FULTON STREET.
1853.


Entered according to the Act of Congress, in the year 1853, by
M. QUINBY,
in the Clerk's Office of the District Court of the United States
for the Southern District of New York.

E. O. JENKINS, PRINTER AND STEREOTYPER,
114 NASSAU STREET, N. YORK.



CONTENTS.


CHAPTER I.

BRIEF HISTORY.

Three kinds of Bees,                                              9
Queen described,                                                  9
Description and Duty of Workers,                                 10
Description of Drones,                                           11
Most Brood in Spring,                                            11
Their Industry,                                                  12


CHAPTER II.

HIVES.

Hives to be thoroughly made,                                     13
Different opinions about them,                                   14
The Author has no Patent to recommend,                           14
Speculators supported long enough,                               15
Prefix of Patent a bad recommendation,                           15
Ignorance of affairs and committees,                             15
Opposition to simplicity,                                        16
By gaining one point produce another evil,                       16
First Delusion,                                                  17
Chamber Hive,                                                    17
Mrs. Griffith's Hive,                                            18
Weeks' Improvement,                                              18
Inclined Bottom-Boards do not throw out all the worms,           19
Objections to suspended hives,                                   19
See bees often,                                                  20
Hall's Patent,                                                   21
Jones's Patent,                                                  21
An Experiment,                                                   21
Reason of failure in dividing hive,                              22
Cause of starving in such hives,                                 23
Advantages of the changeable hive considered,                    24
Variation of these hives,                                        25
Expense in constructing changeable hives,                        25
The surplus honey will contain bee-bread,                        26
Description of Cutting's changeable hive,                        26
First objection cost of construction,                            28
Hives can be made with less expense,                             29
Old breeding cells will last a long time,                        29
Cells larger than necessary at first,                            30
Expense of renewing combs,                                       30
Best to use old combs as long as they will last,                 31
Method for Pruning when necessary,                               31
Tools for Pruning,                                               32
Use of Tobacco Smoke,                                            33
Further objections to a sectional hive,                          34
Non-Swarmers,                                                    35
Contrast of profit,                                              35
Principle of swarming not understood,                            36
Not to be depended upon,                                         37
Hives not always full before swarming,                           37
Size of hives needed,                                            37
An Experiment,                                                   37
Bees do not increase if full after the first year in same hive,  38
Gillmore's system doubted,                                       39
Utility of moth-proof hives doubted,                             39
Instincts of the bee always the same,                            40
Profit the object,                                               41
Common hive recommended,                                         42
Size Important,                                                  42
Small hives most liable to accidents,                            42
Apt to deceive,                                                  43
Unprofitable if too large,                                       43
Correct size between two extremes,                               43
Size for warm latitudes,                                         44
Larger hives more safe for long Winters or backward Spring,      44
2,000 inches safe for this section,                              45
Kind of Wood, width of Board, &c.,                               46
Shape of little consequence,                                     46
Directions for making hives,                                     47
Size of cap and boxes,                                           48
Miner's Hive,                                                    48
Directions for making holes,                                     49
A Suggestion,                                                    50
Glass boxes preferred,                                           51
Glass boxes--how made,                                           51
Guide-combs necessary,                                           52
Wood Boxes,                                                      53
Cover for Hives,                                                 54
Jars and Tumblers--how prepared,                                 54
Perfect Observatory Hive described,                              55
One like Common Hive preferred,                                  56
What may be seen,                                                56
Directions for making Glass Hive,                                57
Plate for Hive,                                                  61


CHAPTER III.

BREEDING.

Imperfectly Understood,                                          62
Good stocks seldom without brood,                                63
How small stocks commence,                                       64
Different with larger ones,                                      65
How Pollen is stored in the breeding season,                     65
Operation of Laying, and the Eggs described,                     66
Time from the Egg to the perfect Bee,                            67
Rough treatment of the young Bee,                                67
Guess-work,                                                      68
Terms applied to  young Bees,                                    69
Discrepancy in time in rearing brood as given by Huber,          70
The number of Eggs deposited by the Queen guessed at,            71
A test for the presence of a Queen,                              73
When Drones are reared,                                          74
When Queens,                                                     74
Liability of being destroyed,                                    76
Drones destroyed when honey is scarce,                           77
Old Queen leaves with the first swarm,                           78
A young Queen takes the place of her mother in the old stock,    79
Other Theories,                                                  80
Subject not understood,                                          80
Necessity for further observation,                               84
Two sides of the question,                                       85


CHAPTER IV.

BEE PASTURAGE.

Substitute for Pollen,                                           88
Manner of packing it,                                            89
Alder yields the first,                                          89
Fruit Flowers important in good weather,                         91
Red Raspberry a favorite,                                        91
Catnip, Mother-wort and Hoarhound, are sought after,             92
Singular fatality attendant on Silkweed,                         93
Large yield from Basswood,                                       96
Garden Flowers unimportant,                                      97
Honey-dew,                                                       97
Singular Secretion,                                              98
Secretions of the Aphis,                                         98
Advantages of Buckwheat,                                        101
Amount of honey collected from it,                              101
Do Bees injure the crop?                                        102
Are not Bees an advantage to vegetation?                        103
A test for the presence of Queen doubted,                       106
An extra quantity of Pollen not always detrimental,             107
What combs are generally free from Bee-bread,                   108
Manner of packing stores,                                       108
Philosophy in filling a cell with honey,                        109
Long cells sometimes turned upward,                             110
Is a dry or wet season best for honey?                          111
How many Stocks should be kept,                                 112
Three principal sources of honey,                               112


CHAPTER V.

WAX.

Is Pollen converted into Wax?                                   115
How is it obtained?                                             115
Huber's account of a commencement of comb,                      117
Best time to witness comb-making,                               118
Manner of working Wax,                                          119
Are crooked Combs a disadvantage?                               120
Uncertainty in weight of Bees,                                  122
Some wax wasted,                                                124
Water necessary in Comb-making,                                 124
Remarks,                                                        126


CHAPTER VI.

PROPOLIS.

What used for,                                                  128
Is it an elaborate or natural substance?                        129
Huber's Opinion,                                                129
Further Proof,                                                  129
Remarks,                                                        132


CHAPTER VII.

THE APIARY.

Its location,                                                   132
Decide Early,                                                   133
Bees mark their location on leaving the hive,                   134
Changing stand attended with loss,                              134
Can be taken some distance,                                     135
Danger of setting Stocks too close,                             135
Space between Hives,                                            136
Small Matters,                                                  136
Economy,                                                        137
Cheap arrangement of stands,                                    138
Canal Bottom-board discarded,                                   139
Some advantage in being near the earth,                         139
Utility of Bee-Houses doubted,                                  141


CHAPTER VIII.

ROBBERIES.

Not properly understood,                                        142
Improper Remedies,                                              143
Difficulty in deciding,                                         144
Weak families in most danger,                                   144
Their Battles,                                                  145
Bad policy to raise in Hives,                                   146
Indications of Robbers,                                         146
A Duty,                                                         147
A Test,                                                         147
Robbing usually commences on a warm day,                        148
Remedies,                                                       149
Common Opinion,                                                 149
A case in point,                                                149
Further Directions,                                             150
Common cause of commencing,                                     151
Spring the worst time,                                          152
No necessity to have Bees plundered in the fall,                153


CHAPTER IX.

FEEDING BEES.

Should be a last resort,                                        154
Care needed,                                                    154
Apparent contradiction when feeding causes starvation,          155
How long it will do to wait before feeding,                     156
Directions for feeding,                                         157
Whole Families may desert the Hive,                             158
Objections to general feeding,                                  159
Arrangement for feeding,                                        159
Feeding to induce early swarms,                                 161
What may be fed,                                                162
Is candied honey injurious?                                     162


CHAPTER X.

DESTRUCTION OF WORMS.

Some in the best Stocks,                                        164
How Found,                                                      165
A tool for their destruction,                                   165
Mistaken Conclusions,                                           167
Objections to suspended Bottom-board,                           167
Advantage of the Hive close to the board,                       168
Objection Answered,                                             169
Insufficiency of inclined Bottom-board                          169
A Moth can go where Bees can,                                   170
Trap to catch Worms,                                            170
Box for Wren,                                                   171


CHAPTER XI.

PUTTING ON AND TAKING OFF BOXES.

Advantage of the Patent Vender,                                 172
Time of putting on--Rule,                                       172
Making holes after the Hive is full,                            174
Advantage of proper arrangement,                                174
Directions for boring holes in full Stock,                      176
To be taken off when filled,                                    177
Time taken to fill a box,                                       178
When to take off boxes part full,                               178
Tobacco Smoke preferred to Slides,                              178
Manner of disposing of the Bees in the boxes,                   179
Bees disposed to carry away honey,                              179
Not disposed to sting,                                          180
Rule,                                                           181


CHAPTER XII.

SECURING HONEY FROM THE MOTH.

Two things to be prevented,                                     181
Apt to be deceived about the Worms,                             182
Their progress described,                                       182
A Solution offered,                                             183
Method of killing Worms in boxes,                               185
Freezing destroys them,                                         186
Objection to using Boxes before the Hive is full,               187


CHAPTER XIII.

SWARMING.

Time to expect them,                                            187
All Bee-Keepers should understand it as it is,                  188
Means of understanding it,                                      188
Inverting a stock rather formidable at first,                   189
Requisites before preparation of Queen's cells,                 189
State of Queen-cell when used,                                  190
State when swarms issue,                                        190
Clustering outside not always to be depended upon,              191
Examinations--the result,                                       191
Remarks,                                                        192
Conflicting Theories,                                           192
Both Old and Young leave with swarms,                           192
Cause of the Queen's inability to fly suggested,                193
Evidence of the Old Queen's leaving,                            193
Mr. Weeks's Theory not satisfactory,                            194
Particular directions for testing the matter,                   196
Empty Hives to be ready,                                        197
Bottom-boards for hiving,                                       197
Description of swarm issuing,                                   198
Manner of hiving can be varied,                                 199
Usual Methods,                                                  199
When out of reach,                                              200
When they cannot be shaken off,                                 202
All should be made to enter,                                    203
Should be taken to the stand immediately,                       203
Protection from the Sun necessary,                              203
Clustering Bushes,                                              204
How swarms are generally managed that leave for the woods,      205
Nothing but Bees needed in a Hive,                              206
Seldom go off without clustering,                               207
Do swarms choose a location before swarming?                    207
Means of arresting a swarm,                                     208
Some Compulsion,                                                208
How far will they go in search of honey?                        209
Two or more swarms liable to unite,                             211
Disadvantage,                                                   211
Can often be prevented,                                         212
Indications of swarming inside the Hive,                        212
Preventing a swarm issuing for a time,                          213
To prevent swarms uniting with those already hived,             213
When two have united--the method of separating,                 214
No danger of a sting by the Queen,                              215
Some precautions in hiving two swarms together,                 216
How to find Queen when two strangers are together,              217
Boxes for double swarms immediately,                            218
Returning a part to the old stock,                              218
Method of uniting,                                              219
When care is necessary,                                         219
Swarm-Catcher,                                                  220
Swarms sometimes return,                                        222
Repetition prevented,                                           222
Liability to enter wrong stocks,                                223
First issues generally choose fair weather,                     224
After Swarms,                                                   225
Their Size,                                                     225
Time after the first,                                           225
Piping of the Queen,                                            225
May always be heard before and after swarm,                     226
Time of continuance varies,                                     226
Time between second and third issues,                           227
Not always to be depended upon,                                 227
A Rule for the time of these issues,                            228
When it is useless to expect more swarms,                       229
Plurality of Queens destroyed,                                  229
The Manner,                                                     230
Theory doubted,                                                 231
After-swarms different in appearance from the first,
    when about to issue,                                        232
Time of day, weather, &c.,                                      233
Swarms necessary to be seen,                                    233
Returning after-swarms to the old stock,                        235
When they should be returned,                                   235
Method of doing it,                                             235
More care needed by After-swarms when hived,                    237
Two may be united,                                              237


CHAPTER XIV.

LOSS OF QUEENS.

Of swarms that lose their Queen,                                238
A suggestion and an answer,                                     239
A disputed question,                                            240
A multitude of Drones needed,                                   241
The Queen liable to be lost in her excursions,                  243
The time when it occurs,                                        243
Indications of the loss,                                        244
The Result,                                                     245
Age of Bees indicated,                                          246
Necessity of care,                                              246
Remedy,                                                         247
Mark the date of swarms on the Hive,                            248
Obtaining a Queen from worker brood,                            249
They are poor dependence,                                       249


CHAPTER XV.

ARTIFICIAL SWARMS.

Principles should be understood,                                252
Some Experiments,                                               253
The result unsatisfactory,                                      253
Further Experiments,                                            254
A successful method,                                            256
Advantages of this method,                                      257
Artificial swarms only safe near the swarming season,           259
Sometimes hazardous,                                            259
Some Objections,                                                259
Natural and artificial swarms equally prosperous,               260
This matter too often delayed,                                  261
Is the age of the Queen important?                              261


CHAPTER XVI.

PRUNING.

Different opinions as to time,                                  262
Another time preferred,                                         263
Should not be delayed,                                          263
Objection to Pruning,                                           264
Stocks pruned now are better for winter,                        265


CHAPTER XVII.

DISEASED BROOD.

Not generally understood,                                       266
My own experience,                                              267
Description of Disease,                                         267
The cause uncertain,                                            268
Remedial Experiments,                                           268
Public inquiry and answers,                                     268
Answers not satisfactory,                                       270
A cause suggested,                                              270
Reasons for the opinion,                                        272
Cause of its spreading,                                         273
Not easily detected at first,                                   274
Symptoms to be observed,                                        274
Scalding the honey to destroy the poison for feeding,           275
When to examine stocks that have swarmed,                       275
Care in selecting stock-hives for winter,                       276
Accusations not always right,                                   276


CHAPTER XVIII.

IRRITABILITY OF BEES.

Their means of defence,                                         277
Time of greatest Irritability,                                  278
Proper Conduct,                                                 278
How to proceed when attacked,                                   279
A person's breath offensive, and other causes,                  279
Their manner of attack,                                         279
Smoker described,                                               280
Effect of Tobacco Smoke,                                        281
Sting described,                                                282
Does its loss prove fatal?                                      283
Means of protection,                                            284
Remedies for stings,                                            285


CHAPTER XIX.

ENEMIES OF BEES.

Are they all guilty?                                            286
Rats and Mice,                                                  287
Are all the Birds guilty?                                       288
King-bird--one word in his favor,                               288
Cat-bird acquitted,                                             289
Toad got clear,                                                 290
Wasps and Hornets not favored,                                  290
Ants--a word in their favor,                                    291
Spider condemned,                                               292
Wax-Moth unrivalled for mischief,                               293
Indications of their presence,                                  296
Management,                                                     296
Care in turning over Hives,                                     297
Other symptoms of Worms,                                        298
When they grow larger than usual,                               299
Time of Growth,                                                 299
Time of Transformation,                                         300
Freezing destroys Worms, Cocoon, and Moth,                      300
How they pass the Winter,                                       301
Stocks more liable to be destroyed last of Summer,              301
When Bees are safe,                                             302
Means to destroy them,                                          302
Making them drunk and their execution by Chickens,              303


CHAPTER XX.

MELTING DOWN OF COMBS.

The Cause,                                                      304
Effects,                                                        304
First Indications,                                              305
Prevention,                                                     305


CHAPTER XXI.

FALL MANAGEMENT.

First care,                                                     307
Strong Stocks disposed to plunder,                              307
Bees Changeable,                                                308
Requisites for good Stocks,                                     308
Great disadvantage of killing the Bees,                         309
Section of country may make a difference in
    what poor stocks need,                                      309
When Bees are needed,                                           310
Caution,                                                        311
Principal Difficulty,                                           311
How Avoided,                                                    311
Advantages of making one good stock from two poor ones,         312
Two families together will not consume as much as if separate,  312
An Experiment,                                                  312
Season for operating,                                           313
The Fumigator,                                                  314
Directions for uniting two families,                            315
Uniting with Tobacco Smoke,                                     317
Condition of Stocks in 1851,                                    318
How they were managed,                                          318
Cause of their superior Thrift,                                 319
Swarms partly filled pay better than to cut out the honey,      320
Advantages in transferring,                                     320
Another method of uniting two families,                         321
Uniting Comb and Honey as well as Bees,                         322
When feeding should be done for Stock Hives,                    323


CHAPTER XXII.

WINTERING BEES.

Different methods have been adopted,                            325
The idea of Bees not freezing has led to errors in practice,    326
Appearance of Bees in cold weather,                             326
How part of the swarm is frozen,                                327
How a small family may all freeze,                              327
Frost and Ice sometimes smother Bees,                           328
Frost and Ice in a Hive accounted for,                          329
The effect of Ice or Frost on Bees and Comb,                    330
Frost may cause starvation,                                     330
Other Difficulties,                                             330
Further Illustrations,                                          332
Accumulation of Fæces described by some writers as a disease,   336
The Author's remedy,                                            337
Buying Bees,                                                    337
Experiments of the Author to get rid of the Frost,              338
Success in this matter,                                         338
Bees when in the house should be kept perfectly dark,           339
A room made for wintering Bees,                                 339
Manner of stowing away Hives,                                   340
Temperature of room,                                            341
Too much Honey may sometimes be stored,                         342
Management of room towards Spring,                              342
Time for setting out Bees,                                      343
Not too many stocks taken out at once,                          343
Families may be equalized,                                      344
Snow need not always prevent carrying out Bees,                 344
Does not Analogy prove that Bees should be kept warm in Winter? 345
The next best place for wintering Bees,                         346
Evils of wintering in the open air considered,                  347
But little risk with good stocks,                               348
Effect of keeping second-rate stocks out of the sun,            348
Effects of Snow considered,                                     349
Stocks to be protected on some occasions,                       350
Do the Bees eat more when allowed to come out
    occasionally in Winter?                                     352


CHAPTER XXIII.

SAGACITY OF BEES.

Are not Bees directed alone by instinct?                        353
What they do with Propolis,                                     353
Mending broken Combs,                                           354
Making passages to every part of their Combs,                   355


CHAPTER XXIV.

STRAINING HONEY AND WAX.

Methods of removing Combs from the Hive,                        357
Different modes of straining Honey,                             358
Getting out Wax--different methods,                             360


CHAPTER XXV.

PURCHASING STOCKS AND TRANSPORTING BEES.

Why the word luck is applied to Bees,                           362
Rule in taking Bees for a share,                                364
A man may sell his "luck,"                                      364
First-rate stocks recommended to begin with,                    365
Old stocks are good as any if healthy,                          365
Caution respecting diseased brood,                              366
Result of ignorance in purchasing,                              366
Size of Hives important,                                        367
How large Hives can be made smaller,                            368
Moderate weather best to remove Bees,                           369
Preparations for transporting Bees,                             370
Securing Bees in the Hive,                                      370
Best Conveyance,                                                370
Hive to be inverted,                                            371
Conclusion,                                                     372



PREFACE.


Before the reader decides that an apology is necessary for the
introduction of another work on bees into the presence of those already
before the public, it is hoped that he will have the patience to
examine the contents of this.

The writer of the following pages commenced beekeeping in 1828, without
any knowledge of the business to assist him, save a few directions
about hiving, smoking them with sulphur, &c. Nearly all the information
to be had was so mingled with erroneous whims and notions, that it
required a long experience to separate essential and consistent points.
It was _impossible_ to procure a work that gave the information
necessary for practice. From that time to the present, no sufficient
guide for the inexperienced has appeared. European works, republished
here, are of but little value. Weeks, Townley, Miner, and others,
writers of this country, within a few years, have given us treatises,
valuable to some extent, but have entirely neglected several chapters,
very important and essential to the beginner. Keeping bees _has_ been,
and is now, by the majority, deemed a hazardous enterprise. The ravages
of the moth had been so great, and loss so frequent, that but little
attention was given to the subject for a long time. Mr. Weeks lost his
entire stock three times in fifteen years. But soon after the discovery
was promulgated, that honey could be taken from a stock without
destroying the bees, an additional attention was manifest, increasing
to a rage in many places. It seems to be easily understood, that
_profit_ must attend success, in this branch of the farmer's stock;
inasmuch as the "bees work for nothing and find themselves." This
interest in bees should be encouraged to continue till enough are kept
to collect all the honey now wasted; which, compared with the present
collections, would be more than a thousand pounds to one. But to
succeed, that is the difficulty. Some eighteen years since, after a
propitious season, an aged and esteemed friend said to me, "It is not
to be expected that you will have such luck always; you must expect
they will run out after a time. I have always noticed, when people have
first-rate luck for a time, that the bees generally take a turn, and
are gone in a few years."

I am not sure but, to the above remarks, may be traced the cause of my
subsequent success. It stimulated me to observation and inquiry. I soon
found that good seasons were the "lucky" ones, and that many lost in an
adverse season, all they had before gained. Also, that strong families
were the only ones on which I could depend for protection against the
moth. This induced the effort to ascertain causes tending to diminish
the size of families, and the application of remedies. Whether success
has attended my efforts or not, the reader can judge, after a perusal
of the work.

It is time that the word "_luck_," as applied to beekeeping, was
discarded. The prevailing opinion, that bees will prosper for one
person more than another, under the same circumstances, is fallacious.
As well might it be applied to the mechanic and farmer. The careless,
ignorant farmer, might occasionally succeed in raising a crop with a
poor fence; but would be liable, at any time, to lose it by trespassing
cattle. He might have suitable soil in the beginning, but without
knowledge, for the proper application of manures, it might fail to
produce; unless a _chance_ application _happened_ to be right.

But with the _intelligent_ farmer the case is different: fences in
order, manures judiciously applied, and with propitious seasons, he
makes a sure thing of it. Call him "_lucky_" if you please; it is his
knowledge, and care, that render him so. So with bee-keeping, the
careful man is the "lucky" one. There can be no effect without a
preceding cause. If you lose a stock of bees, there is a cause or
causes producing it, just as certain as the failure of a crop with the
unthrifty farmer, can be traced to a poor fence, or unfruitful soil.
You may rest assured, that a rail is off your fence of management
somewhere, or the proper applications have not been made. In relation
to bees, these things may not be quite so apparent, yet nevertheless
true. Why is there so much more uncertainty in apiarian science than
other farming operations? It must be attributed to the fact, that among
the thousands who are engaged in, and have studied agriculture, perhaps
not more than one has given his energies to the nature and habits of
bees. If knowledge is elicited in the same ratio, we ought to have a
thousand times more light on one subject than the other, and still
there are some things, even in agriculture, that may yet be learned.

It is supposed, by many, that we already have all the knowledge that
the subject of _bees_ affords. This is not surprising; a person that
was never furnished with a full treatise, might arrive at such
conclusions. Unless his own experience goes deeper, he can have no
means of judging what is yet behind.

In conversation relative to this work, with a person of considerable
scientific attainments, he remarked, "You do not want to give the
natural history of bees at all; that is already sufficiently
understood." And how is it understood; as Huber gives it, or in
accordance with some of our own writers? If we take Huber as a guide,
we find many points recently contradicted. If we compare authors of our
day, we find them contradicting each other. One recommends a peculiarly
constructed hive, as just the thing adapted to their nature and
instincts. If a single point is in accordance with their nature, he
labors to twist all the others to his purpose, although it may involve
a fundamental principle impossible to reconcile. Some one else succeeds
in another point, and proceeds to recommend something altogether
different. False and contradictory assertions are made either through
ignorance, or interest. Interest may blind the judgment, and spurious
history may deceive.

It is folly to expect success in bee-keeping for any length of time,
without a correct knowledge of their nature and instincts; and this we
shall never obtain by the course hitherto pursued. As much of their
labor is performed in the dark, and difficult to be observed, it has
given rise to conjecture and false reasoning, leading to false
conclusions.

When _I_ say a thing _is so_, or say it is _not so_, what evidence has
the reader that it is proved or demonstrated? _My_ mere assertions are
not expected to be taken in preference to another's; of such proof, we
have more than enough. Most people have not the time, patience, or
ability, to set down quietly with close observation, and investigate
the subject thoroughly. Hence it has been found easier to receive error
for truth, than to make the exertion necessary to confute it; the more
so, because there is no guide to direct the investigation. I shall,
therefore, pursue a different course; and for every _assertion_
endeavor to give a test, that the reader may apply and satisfy himself,
and trust to no one. As for theories, I shall try to keep them separate
from facts, and offer such evidence as I have, either for or against
them. If the reader has further proof that presents the matter in
another light, of course he will exercise the right to a difference of
opinion.

I could give a set of rules for practice, and be very brief, but this
would be unsatisfactory. When we are told a thing _must be done_, most
of us, like the "inquisitive Yankee," have a desire to know _why_ it is
necessary; and then like to know _how_ to do it. This gives us
confidence that we are right. Hence, I shall endeavor to give the
practical part, in as close connection with the natural history, that
dictates it, as possible.

This work will contain several chapters entirely new to the public: the
result of my own experience, that will be of the utmost value to all
who desire to realize the greatest possible advantages from their bees.

The additions to chapters already partially discussed by others, will
contain much original matter not to be found elsewhere. When many
stocks are kept, the chapter on "Loss of Queens," alone, will, with
attention, save to any one, not in the secret, enough in one season to
be worth more in value than many times the cost of this work. The same
might be said of those on diseased brood, artificial swarms, wintering
bees, and many others.

If such a work could have been placed in my hands twenty years ago, I
should have realized hundreds of dollars by the information. But
instead of this, my course has been, first to suffer a loss, and then
find out the remedy, or preventive; from which the reader may be
exempt, as I can confidently recommend these directions.

Another new feature will be found in the duties of each season being
kept by itself, commencing with the spring and ending with the winter
management.

In my anxiety to be understood by all classes of readers, I am aware
that I have made the elegant construction and arrangement of sentences
of secondary importance; therefore justly liable to criticism. But to
the reader, whose object is information on this subject, it can be of
but little consequence.



CHAPTER I.

A BRIEF HISTORY.


THREE KINDS OF BEES.

Every prosperous swarm, or family of bees, must contain one queen,
several thousand workers, and, part of the year, a few hundred drones.

[Illustration: QUEEN.] [Illustration: WORKER.] [Illustration: DRONE.]


QUEEN DESCRIBED.

The queen is the mother of the entire family; her duty appears to be
only to deposit eggs in the cells. Her abdomen has its full size very
abruptly where it joins the trunk or body, and then gradually tapers to
a point. She is longer than either the drones or workers, but her size,
in other respects, is a medium between the two. In shape she resembles
the worker more than the drone; and, like the worker, has a sting, but
will not use it for anything below royalty. She is nearly destitute of
down, or hairs; a very little may be seen about her head and trunk.
This gives her a dark, shining appearance, on the upper side--some are
nearly black. Her legs are somewhat longer than those of a worker; the
two posterior ones, and the under surface, are often of a bright copper
color. In some of them a yellow stripe nearly encircles the abdomen at
the joints, and meets on the back. Her wings are about the same as the
workers, but as her abdomen is much longer, they only reach about
two-thirds the length of it. For the first few days after leaving the
cell, her size is much less than after she has assumed her maternal
duties. She seldom, perhaps never, leaves the hive, except when leading
a swarm, and when but a few days old, to meet the drones, in the air,
for the purpose of fecundation. The manner of the queen's impregnation
is yet a disputed point, and probably never witnessed by any one. The
majority of close observers, I believe, are of opinion that the drones
are the males, and that sexual connection takes place in the air,[1]
performing their amours while on the wing, like the humble-bee and some
other insects. It appears that one impregnation is operative during her
life, as old queens are not afterwards seen coming out for that purpose.

      [1] The objectors to this hypothesis will be generally found
      among those who are unable to give a more plausible elucidation.
      Those who oppose the fact that one bee is the mother of the whole
      family, will probably be in the same class.


DESCRIPTION AND DUTY OF WORKERS.

As all labor devolves on the workers, they are provided with a sack, or
bag, for honey. Basket-like cavities are on their legs, where they pack
the pollen of flowers into little pellets, convenient to bring home.
They are also provided with a sting, and a virulent poison, although
they will not use it abroad when unmolested, but, if attacked, will
generally defend themselves sufficient to escape. They range the fields
for honey and pollen, secrete wax, construct combs, prepare food, nurse
the young, bring water for the use of the community, obtain propolis to
seal up all crevices about the hive, stand guard, and keep out
intruders, robbers, &c., &c.


DESCRIPTION OF DRONES.

When the family is large and honey abundant, a brood of drones is
reared; the number, probably, depends on the yield of honey, and size
of the swarm, more than anything else. As honey becomes scarce, they
are destroyed. Their bodies are large and rather clumsy, covered with
short hairs or bristles. Their abdomen terminates very abruptly,
without the symmetry of the queen or worker. Their buzzing, when on the
wing, is louder, and altogether different from the others. They seem to
be of the least value of any in the hive. Perhaps not more than one in
a thousand is ever called upon to perform the duty for which they were
designed. Yet they assist, on some occasions, to keep up the animal
heat necessary in the old hive after a swarm has left.


MOST BROOD IN SPRING.

In spring and first of summer, when nearly all the combs are empty, and
food abundant, they rear brood more extensively than at any other
period, (towards fall more combs are filled with honey, giving less
room for brood.) The hive soon becomes crowded with bees, and royal
cells are constructed, in which the queen deposits her eggs. When some
of these young queens are advanced sufficiently to be sealed over, the
old one, and the greater part of her subjects, leave for a new
location, (termed swarming.) They soon collect in a cluster, and, if
put into an empty hive, commence anew their labors; constructing combs,
rearing brood, and storing honey, to be abandoned on the succeeding
year for another tenement. One in a hundred may do it the same season,
if the hive is filled and crowded again in time to warrant it. Only
large early swarms do this.


THEIR INDUSTRY.

Industry belongs to their nature. When the flowers yield honey, and the
weather is fine, they need no impulse from man to perform their part.
When their tenement is supplied with all things necessary to reach
another spring, or their store-house full, and no necessity or room for
an addition, and we supply them with more space, they assiduously toil
to fill it up. Rather than to waste time in idleness, during a
bounteous yield of honey, they have been known to deposit their surplus
in combs outside the hive, or under the stand. This natural industrious
habit lies at the foundation of all the advantages in bee-keeping;
consequently our hives must be constructed with this end in view; and
at the same time not interfere with other points of their nature; but
this subject will be discussed in the next chapter. Those peculiar
traits in their nature, mentioned in this, will be more fully discussed
in different parts of this work, as they appear to be called for, and
where proof will be offered to sustain the positions here assumed,
which as yet are nothing more than mere assertions.



CHAPTER II.

HIVES.


HIVES TO BE THOROUGHLY MADE.

Hives should be constructed of good materials, boards of good
thickness, free from flaws and cracks, well fitted and thoroughly
nailed.

The time of making them is not very particular, providing it is done in
season. It certainly should not be put off till the swarming period, to
be made as wanted, because if they are to be painted; it should be done
as long as possible before, as the rank smell of oil and paint, just
applied, might be offensive to the bees.

But what kind of hive shall be made?

In answer, some less than a thousand forms have been given. The
advantages of bee-keeping depend as much upon the construction of
hives, as any one thing; yet there is no subject pertaining to them on
which there is such a variety of opinions, and I have but little hopes
of reconciling all these conflicting views, opinions, prejudices, and
interests.


DIFFERENT OPINIONS ABOUT THEM.

One is in favor of the old box, and the cruel practice of killing the
bees to obtain the honey, as the only means to obtain "luck;" "they are
sure to run out if they meddle with them." Another will rush to the
opposite extreme, and advocate all the extravagant fancies of the
itinerant patent-vender, as the _ne plus ultra_ of all hives, when
perhaps it would be worth more for fire-wood than the apiary.


THE AUTHOR HAS NO PATENT TO RECOMMEND.

To remove from the mind of the reader all apprehension that I am about
condemning one patent to recommend another, I would say in the
beginning, that I have _no patent to praise, no interest in deceiving_,
and I hope no prejudices to influence me, in advocating or condemning
_any_ system. I wish to make bee-keeping plain, simple, economical, and
profitable; so that when we sum up the profit "it shall not be found in
the other pocket."

It is a principle recognized by our statute, that no person is suitable
as a Juror, who is biased either by interest or prejudice. Now whether
I am the impartial Jurist, is not for me to say: but I wish to discuss
the subject fairly. I hope some few will be enabled to see their own
interest: at any rate, dismiss prejudice, as far as possible, while we
examine wherein _one class_ in community is unprofitable to
bee-keepers.


SPECULATORS SUPPORTED LONG ENOUGH.

We have faithfully supported a host of speculators on our business for
a long time; often not caring one straw about our success, after
pocketing the fee of successful "humbuggery." One is no sooner gone,
than we are beset by another, with something altogether different, and
of course the acme of perfection.


PREFIX OF PATENT A BAD RECOMMENDATION.

This has been done until the very prefix of patent, or premium,
attached to a hive, renders it almost certain that there must be
something deleterious to the apiarian; either in expense of
construction or intricate and perplexing in management, requiring an
engineer to manage, and a skilful architect to construct.

What does the American savage, who without difficulty can track the
panther or wolf, know of the principles of chemistry? What does the
Chemist know of following a track in the forest, when nothing but
withered leaves can guide him? Each understands principles, the
_minutiæ_ of which the other never dreamed.


IGNORANCE OF OFFICERS AND COMMITTEES.

Thus it appears to be with granting patents and premiums, if we take
what has been patented and praised by our committees and officers as
improvements in bee-culture. These men may be capable, intelligent, and
well fitted for their sphere, but in bee matters, about as capable of
judging, as the Hottentot would be of the merits of an intricate
steam-engine. Knowledge and experience are the only qualifications
competent to decide.


OPPOSITION TO SIMPLICITY.

I am aware that among the thousands whose direct interest is opposed to
my simple, plain manner of getting along, many will be ready to contend
with me for every departure from their patent, improved or premium
hives, as the case may be.


BY GAINING ONE POINT, PRODUCE ANOTHER EVIL.

I think it will be an easy matter to show that every departure from
simplicity to gain _one_ point, is attended in another by a
correspondent evil, that often exceeds the advantage gained. That we
have made vast improvements in art and science, and in every department
of human affairs, no one will deny; consequently, it is assumed we must
correspondingly improve in a bee-hive; forgetting that nature has fixed
limits to the instinct of the bee, beyond which she will not go!

It will be necessary to point out the advantages and objections to
these pretended improvements, and then we will see if we cannot avoid
the objections, _and retain the advantages, without the expense_, by a
simple addition to the common hive; because if we expect to encourage
bee-keeping, they must have better success than a neighbor of mine, who
expended fifty dollars for bees and a patent, and lost all in three
years! Most bee-keepers are farmers; very few are engineers sufficient
to work them successfully. I would say to all such as do not understand
the nature of bees, adhere to simplicity until you do, and then I am
quite sure you will have no desire for a change.


FIRST DELUSION.

Probably the first delusion in the patent line originated with the
idea, that to obtain surplus honey, it was absolutely necessary to have
a chamber hive. To get rid of the depredations of mice, the suspended
hive was contrived. The inclined bottom-board was then added to throw
out the worms. To prevent the combs from sliding down, the lower end
was contracted.

The principle of bees rearing queens from worker-eggs when destitute,
gave rise to the dividing hive in several forms. Comb, when used
several years, becomes thickened and black, and needs changing; hence
the changeable hives, Non-swarmers have been introduced to save risk
and trouble. Moth-proof hives to prevent the ravages of worms, &c., &c.


CHAMBER HIVE.

The chamber hive is made with two apartments; the lower and largest is
for the permanent residence of the bees, the upper or chamber for the
boxes. Its merits are these: the chamber affords all the protection
necessary for glass boxes; considered as a cover, it is never lost. Its
demerits are inconvenience in handling; it occupies more room if put in
the house in the winter; if glass boxes are used, only one end can be
seen, and this may be full when the other may hold some pounds yet, and
we cannot possibly know until it is taken out. I know we are told to
return such boxes when not full "and the bees will soon finish them,"
but this will depend on the yield of honey at the time; if abundant, it
will be filled; if not, they will be very likely to take a hint, and
remove below what there is in the box; whereas if the chamber was
separate from the hive, and was not a chamber but a loose cap to cover
the boxes, it could be raised at any time without disturbing a single
bee, and the precise time of the boxes being filled ascertained, (that
is, when they are of glass.)


MRS. GRIFFITH'S HIVE.

Mrs. Griffith, of New Jersey, is said to have invented the suspended
chamber hive with the inclined bottom-board. One would suppose this was
sufficiently inconvenient to use, and difficult and expensive to
construct.


WEEKS' IMPROVEMENT.

Yet Mr. Weeks makes an alteration, calls it an improvement, the expense
is but a trifle more; it is sufficient to be sanctioned by a patent.
From front to rear, the bottom is about three inches narrower than the
top, somewhat wedge-shape; it has the merit to prevent the combs from
slipping down, when they _happen_ to be made, to have the edges
supported. The objections are, that filth from the bees will not fall
as readily to the bottom as if every side was perpendicular, and the
extra trouble in constructing.


INCLINED BOTTOM-BOARDS DO NOT THROW OUT ALL THE WORMS.

Inclined bottom-boards form the basis of one or two patents, said to be
good to roll out the worms. I can imagine a pea rolling off such a
board; but a worm is not often found in a rolling condition. Most of us
know, that when a worm drops from the combs, it is like the spider,
with a thread attached above. The only way that I can imagine one to be
thrown out by these boards, is to have it dead when it strikes it, or
so cold that it cannot spin a thread, and wind to shake the board, till
it rolls off. The objections to these boards are coupled with the
suspended hive, with which they are usually connected.


OBJECTIONS TO SUSPENDED HIVES.

All suspended hives _must be objectionable_ to any one who wishes to
know the _true_ condition of his bees at all times. Only think of the
trouble of unhooking the bottom-board, and getting down on your back,
or twisting your neck till your head is dizzy, to look up among the
combs, and then see nothing satisfactory for want of light; or to lift
the hive from its supporters, and turn it over. The operation is too
formidable for an indolent man, or one that has much other business.
The examination would very probably be put off till quite sure it would
do no longer, and sometimes a few days after that, when you will very
often find your bees past remedy.


SEE BEES OFTEN.

"_See your bees often_," is a choice recipe,--it is worth five hundred
dollars at interest, even when you have but few stocks. How necessary
then that we have every facility for a close and minute inspection. How
much easier to turn up a hive that simply rests on a stand. Sometimes
it is necessary to turn the hive, even bottom up, and let the rays of
the sun directly among the combs, to see _all_ the particulars. By this
close inspection, I have often ascertained the cause of some
difficulty, and provided a remedy, thus saving a good many that in a
short time would have been lost; yet, with a little help, were as
valuable as any by another year.


HALL'S PATENT.

Mr. Hall has added a lower section to his hive, about four inches deep,
with two boards inside, like the roof of a house, to discharge the
worms, &c.; but as these boards would interfere with close inspection,
they are objectionable. Several other variations of inclined
bottom-boards and suspended hives have been contrived, to obtain a
patent, but the objections offered will apply to most of them. I shall
not weary the reader by noticing in detail _every_ hive that has been
patented; I think if I notice the _principles of each kind_, it will
test his patience sufficiently.


JONES' PATENT.

Jones' dividing hive was probably suggested by this instinctive
principle of the bee, viz.: when a stock by any accident loses its
queen, and the combs contain eggs or very young larvæ, they will rear
another. Now if a hive is constructed so as to divide the brood-combs,
it would seem quite certain that the half without a queen, would raise
one; and we could multiply our stocks without swarms, the trouble of
hiving, and risk of their going to the woods, &c.


AN EXPERIMENT.

Several years ago, I thought I had obtained a principle that would
revolutionize the whole system of bee management. In 1840 I constructed
such hives, and put in the bees to test by actual experiment, the
utility of what seemed so very plausible in theory. It would appear
that this principle suggested the same idea to Mr. Jones; perhaps with
this difference: I think he did not wait to test the plan thoroughly,
before obtaining his patent in '42. One vender of rights asserted that
63 stocks were made from one in three years; but somehow a great many
that obtained the rights, failed in their expectations. From my
experiments, I think I could guess at some of the reasons.

Mr. A.--"Well, what are the reasons? give us your experience, if you
please, I am interested; I had the right for such a hive, and had a lot
made to order, that cost more money in the end than I shall ever pay
again for anything about bees."

Do not be too hasty, friend, I think I can instruct you to keep bees on
principles in accordance with their nature, which is very simple, so
that if you can be induced to try again, we will have the _hives_ cost
but little, at any rate.


REASONS OF FAILURE IN DIVIDING HIVES.

The greatest difficulty with dividing hives, appeared to be here. It
must be constructed with a partition or division to keep the combs in
each apartment separate; otherwise, we make tearing work in the
division. When bees are first put into such hives, unless the swarm is
very large, and honey abundant, one apartment will be filled to the
bottom before a commencement is made in the other.

Mr. A.--"What difference can that make? It is necessary to have the
hive full; if it cannot be all filled at once, why let them fill part."

The difference is this. The first combs built by a swarm are for brood,
and store-combs afterwards, as needed; one apartment will be nearly
filled with all brood-combs, and the other with store-combs and honey.
Now in the two kinds of cells there is a great difference; those for
breeding are near half an inch in length, while those for storing are
sometimes two inches or more; totally unfit for breeding; until the
bees cut them off to the proper length, which they will not do, unless
compelled for want of room, consequently this side of store-combs is
but little used for brood. When such hive is divided, the chances are
not more than one in four, that this apartment will have any young bees
of the proper age from which to raise a queen; if not, and the old
queen is in the part with the brood-comb, where she will be ninety-nine
times in a hundred, one half of the hive is lost for want of a queen.

Mr. A.--"Ah! I think I now understand how I lost one-half of nearly
every hive I divided. I also lost some of them in the winter; there was
plenty of bees as well as honey; can you tell the cause of this?"

I will guess that they starved.

Mr. A.--"Starved! why, I said there was plenty of honey."

I understood it, but nevertheless feel quite sure.

Mr. A.--"I would like to see that made plain; I can't understand how
they could starve when there was honey!"


CAUSE OF STARVING IN SUCH HIVES.

I said one apartment would be filled with brood-combs; this will be
occupied, at least partially, with brood as long as the yield of honey
lasts; consequently, there will be but little room for storing here,
but the other side may be full throughout. The bees will take up their
winter quarters among the brood-combs. Now suppose the honey in this
apartment is all exhausted during a severe turn of cold weather, what
can the bees do? If one should leave the mass and go among the frosty
combs for a supply, its fate would be as certain as starvation. Without
frequent intervals of warm weather to melt all frost on the combs, and
allow the bees to go into the other apartment for honey, they _must_
starve.

The cost of construction is another objection to this hive, as the
labor bestowed on one is more than would finish two, that would be much
better.


ADVANTAGES OF THE CHANGEABLE HIVE CONSIDERED.

The value of changeable hives is based upon the following
principle:--Each young bee when it first hatches from the egg, is
neither more nor less than a worm; when it receives the necessary food,
the bees seal it over; it will then spin a cocoon, or line its cell
with a coating of silk, less in thickness than the thinnest paper: this
remains after the bee leaves it. It is evident, therefore, that after a
few hundreds have been reared in a cell, and each one has left its
cocoon, that such cell must be somewhat diminished, although the
thickness of a dozen cocoons could not be measured; and this old cell
needs removing, that the bees may replace it with a new one. But how
shall it be done? This is a feat for the display of ingenuity. A common
man might go about it in a very sensible, simple manner, might possibly
turn the hive over, and cut out the old combs when necessary, without
knowing perhaps that the patent-vender could _sell_ a receipt to do the
thing _scientifically_, the benefit of which would be many times on the
principle of a surgeon cutting off your head, to get a good chance to
tie a small artery according to system; or would show you a roundabout
way of half a dozen miles to accomplish what the same number of rods
would do. Had we not ocular demonstration of the fact, we could not
suppose so many variations for the same end could be invented. But if
we reward ingenuity, it will be stimulated to great exertions. Perhaps
if we describe the merits of one or two of this class, the utility of
this principle may be comprehended.


VARIATION OF THESE HIVES.

First, then, the sectional hive of various patterns has been patented;
it consists generally of about three boxes, one above another; the top
of each has one large hole, or several small ones, or cross-bars, about
an inch wide, and half an inch apart; these holes or spaces allowing
the bees to pass from one box to the other. When all are full, the
upper one is removed, and an empty one put under the bottom; in this
way all are changed, and the combs renewed in three years; very easily
and quietly done. This is as far as a patent-vender wishes the subject
investigated; and some of his customers have not gone beyond this
point. As an offset for these advantages, we will first look at the
cost of such hive.


EXPENSE IN CONSTRUCTING CHANGEABLE HIVES.

It is as much work to construct each separate section, as a common
hive; consequently, it is three times the expense to begin with. It is
objectionable for wintering bees, on the same principle as the dividing
hive. I object to it on another point: our surplus honey will never be
pure, as each section must be used for breeding, and every cell so
used, will contain cocoons corresponding to the number of bees raised.


SURPLUS HONEY WILL CONTAIN BEE-BREAD.

Also pollen, or bee-bread, is always stored in the vicinity of the
young brood; some of this will remain mixed with the honey, to please
the palate with its _exquisite flavor_. The majority will probably
prefer all surplus honey stored in pure comb, where it will be with
proper management.

I will here give a full description of a hive on this principle, as I
have the description from one of its advocates, in the Dollar
Newspaper, Philadelphia: called Cutting's Patent Changeable Hive.


DESCRIPTION OF CUTTING'S CHANGEABLE HIVE.

"The size of the changeable hive most used in this section, has an
outside shell, made of inch boards, about two feet high and sixteen and
a half inches square, with a door hung in the rear. On the inside are
three boxes or drawers, which will hold about one thousand cubic inches
each, and when filled with honey, usually weigh about thirty-five
pounds, which is a sufficient amount of honey to winter a large swarm.
The sides of these drawers are made of boards, about half an inch
thick; the tops and bottoms of the lower drawers and ends of the upper
drawers should be three-fourths of an inch, and the drawers should be
fourteen inches high, fourteen inches from front to rear, and six and
three-fourths inches wide. Two of these drawers stand side by side,
with the third placed flatwise upon the two, with a free communication
from one drawer to another, by means of thirty three-fourth inch holes
on the side of each drawer, and twenty-four in the bottom of the upper
drawer, and holes in the top and bottom of the lower drawers, to
correspond, and slides to cut off the communication when occasion may
require. Thus we see our hive may be one hive, with communication
sufficiently free throughout, or we may have three hives combined. The
drawers have tubes made in them, (for the bees to pass and repass),
which are made to go through the front side of the hive. The back-side
of the drawers are doors, with glass set in them. These drawers set up
from the bottom of the hive, and rest on pieces of wood, closely fitted
in such a way, as to make a space under the drawers for the _dirt_,
_dead bees_, and _water_, which collect in the bottom of hives in
winter; between the drawers and the outside is an air space of about
one-third of an inch.

These hives, when well made and painted, will last many years, and
those doing much in the business will find it an advantage to have a
few extra drawers. Having given you some idea of the construction of
the changeable hive, I will proceed to notice some of the most
important reasons why I prefer this hive to any I have yet seen. First
because the hive, being constructed upon the changeable principle, so
that by taking out a full drawer, and placing an empty one in its
stead, our comb is always kept new, wherefore, the size of the bee is
preserved, and kept in a more healthy, or prosperous state, or
condition, than when obliged to remain and continue to breed, in the
old comb, when the cells have become small. Secondly, because small,
late swarms may be easily united. Thirdly, because large swarms may be
easily divided. Fourthly, because however late a swarm may come off, it
may be easily supplied with honey for the winter, by taking from a full
hive a surplus drawer, and placing it in the hive of the late swarm.
Fifthly, because a column of air between the drawers and the outside of
the hive is a non-conductor of both heat and cold, preventing the
melting of the comb, and securing the bees against frost and cold."

Now here is a full description of perhaps as good a hive as any of its
class; it is given for the benefit of those who wish to go miles
instead of rods; they may know the road, especially as they can have
the privilege by paying for it: for myself, I had rather be
excused,--why, reading the description has nearly exhausted my
patience; what should I do if I attempted to make one?


FIRST OBJECTION, COST OF CONSTRUCTION.

The first obstacle in the way (after the right is obtained) is the
construction. Let's see; we want inch boards to make the shell,
three-quarter inch boards for the tops and bottoms of drawers, half
inch for sides, hinges to hang a door, glass for back of drawers, tubes
for the egress of the bees, and slides to cut off communication. It
will be necessary to get a mechanic, and a workman too. Those 108 holes
that must be bored, _must match_, or it is of no use to make them. But
few farmers would have the tools requisite, a still less number the
skill and patience to do it. What the cost might be by the time a hive
was ready to receive the bees, I could not say; but guess it might be
some three or four dollars.


HIVES CAN BE MADE WITH LESS EXPENSE.

The one I shall recommend, without paint, will not cost, or need not,
over 37-1/2 cents, with cover, etc. Now, if we wish hives for ornament,
it is well enough to expend something for the purpose; but it is well
not to refine too much, as there are limits which, if passed, will
render it unfit for bees. Therefore, when profit is an object, the
extra expense will or ought to be made up by the bees, in return for an
expensive domicil. But will they do it? The merits of the one under
consideration are fully given. "First, by taking out a full drawer and
putting in an empty one in its stead, the combs are always kept new,
and cells of full size." Now this fear of bees becoming dwarfs in
consequence of being reared in cells too small, has done more mischief
among the bees, and their owners' pockets, than if the fact had never
been thought, or heard of.


OLD BREEDING CELLS WILL LAST A LONG TIME.

These old cells do not need renewing half as often as has been
represented. It is the interest of these patent-venders to sell rights;
this interest either blinds their eyes as to facts, or lulls the
internal monitor of right, while acquisitiveness is gratified. The same
cells can be used for breeding six or eight years, perhaps longer, and
no one can tell the difference by the size of the bees; I have two
stocks now in their tenth year without renewal of comb. A neighbor of
mine kept a stock twelve years in the same combs; it proved as
prosperous as any. I have heard of their lasting twenty, and am
inclined to believe it.


CELLS LARGER THAN NECESSARY AT FIRST.

The bees seem to make a provision for this emergency, the sheets of
comb are farther apart than actually necessary at first, the diameter
of the cell is also a little larger than the size of the young bee
requires. _Of this we are certain_--great many young bees _can_ be
raised in a cell, and not be diminished in size, sufficient to be
detected. The bottom fills up faster than at the sides, and as they do
so, the bees add a little to the length, until the ends of these cells
on two parallel combs approximate too close to allow the bees to pass
freely; before which time it is unnecessary to remove comb for being
old.


EXPENSE OF RENEWING COMBS.

One important item should be considered in this matter, by those who
are so eager for new combs. It is doubtful whether one in 500 ever
thought of the expense of renewing comb. I find it estimated by one
writer,[2] that twenty-five lbs. of honey was consumed in elaborating
about half lb. wax. This without doubt is an over estimate, but no one
will deny that some is used.

      [2] See Appendix of Cottage Bee-keeper, page 118.


BEST TO USE OLD COMBS AS LONG AS THEY WILL ANSWER.

I am satisfied of this much, from actual experience, that every time
the bees have to renew their brood-combs in a hive, they would make
from ten to twenty-five lbs. in boxes, hence I infer that their time
can be more profitably employed than in constructing brood-combs _every
year_. I would also suggest that when combs have been once used for
breeding it is the best use they can be applied to, after that, as the
cocoons render it unfit for much else than a little wax.


METHOD FOR PRUNING WHEN NECESSARY.

But when the combs do actually need removing, I prefer the following
method of pruning, to driving the bees out entirely, as has been
recommended. It can be done in about an hour. As we are comparing the
merits of different methods of getting rid of old combs, I shall give
mine here, notwithstanding it may seem a little out of place.

The best time is a little before night. The first movement is to blow
under the hive some tobacco smoke (the best means of charming them I
ever found); the bees, deprived of all disposition to sting, retreat up
among the combs to get away from the smoke; now raise the hive from the
stand and carefully turn it bottom upwards, avoiding any jar, as some
of the bees that were in the top when the smoke was introduced, and did
not get a taste, will now come to the bottom to ascertain the cause of
the disturbance; these should receive a share, and they will
immediately return to the top, perfectly satisfied. When so many bees
are in the hive, as to be in the way in pruning, (which if there is not
it is not worth it,) get an empty hive the size of the old one, and set
it over, stopping the holes; now strike the lower hive with a hammer or
stick, lightly and rapidly, five or ten minutes, when nearly all the
bees will be in the upper hive, and set that on the stand. There being
now nothing in the way, except a few scattering bees, that I will
_warrant not to sting, unless you pinch or get them fast_.

[Illustration: TOOLS FOR CUTTING OUT COMB.]

The broad one is very readily made from a piece of an old scythe, about
18 inches long, by any blacksmith, by simply taking off the back, and
forming a shank for a handle at the heel. The end should be ground all
on one side, and square across like a carpenter's chisel. This is for
cutting down the sides of the hive; the level will keep it close the
whole length, when you wish to remove all the combs; it being square
instead of pointed or rounded, no difficulty will be found in guiding
it,--it being very thin; no combs are mashed by crowding.

The other tool is for cutting off combs at the top or any other place.
It is merely a rod of steel three-eighths of an inch diameter, about
two feet long, with a thin blade at right angles, one and a half inches
long, and a quarter inch wide, both edges sharp, upper side bevelled,
bottom flat, &c. You will find these tools very convenient; be sure and
get them by all means, the cost cannot be compared to the advantages.

Now with the tools just described, proceed to remove the brood-combs
from the centre of the hive. The combs near the top and outside are
used but little for breeding, and are generally filled with honey;
these should be left as a good start for refilling, but take out all
that is necessary, while you are about it; then reverse the hives,
putting the one containing the bees under the other; by the next
morning all are up; now put it on the stand, and this job is done
without one cent extra expense for a patent to help you, and the bees
are much better off for the honey left, which has to be taken away with
all patent plans that I have seen, and this, as has been remarked, is
not worth much, occupied as it is with a few cocoons and bee-bread. It
is worth much more to the bees, and they will give us pure comb and
honey for it.


USE OF TOBACCO SMOKE.

"I would not do it for fifty dollars, the bees would sting me to
death." Stop a moment, if you never tried the efficacy of tobacco
smoke, you know nothing of a powerful agent; this is the grand secret
of success; without it, I admit it would be somewhat hazardous; but
with it, I have done it time after time without receiving a single
sting, and no protection whatever, for either hands or face.

But is there no difficulty with our sectional or changeable hive, when
this feat is to be performed? The combs will be made in the two drawers
similar to the dividing hive, brood-combs in one side, and store-combs
in the other. We wish to remove the one with brood-combs of course, (as
that is the one where the combs are thick and bad, &c.) Where will the
queen be? With the brood-comb, where her duty is most likely to be;
well, this is the one we want, and we take it out. How is she to get
back? She must go back, or we have three chances in four of losing the
stock; but her majesty will remain perfectly easy, as well as some of
the workers, wherever you put the drawer.


FURTHER OBJECTIONS TO A SECTIONAL HIVE.

I can see no other way but to break the box, look her up, and help the
helpless thing home, (the chances of being stung may be here too.) Now,
for a time at least, they must use the other drawer for breeding, where
most of the cells are unfit. There is altogether too great a proportion
of drone-cells; these, as well as the other size, will nearly all be
much too long, and will have to be cut off to the proper length, a
waste of wax as well as labor. Another thing might be set down per
disadvantage of Mr. Cutting's hive; the job of getting a swarm into
such hive, at first, I fancy would not be desirable to many. Now, when
we strike the balance, putting expense, difficulties, and perplexities
on one side, and simplicity and economy on the other, it appears like a
"great cry for little wool." But stop a moment, four other advantages
are enumerated in its favor: second, third, and fourth are borrowed
from the common hive, or are all available here when required. But
fifthly, allows a "column of air between the drawers and outside of the
hive, is a non-conductor of heat and cold," &c. This is an advantage
not possessed by the common hive; neither does the common hive offer
such advantages to the moth, by affording such snug quarters for worms
to spin their cocoons, when they cannot be destroyed without
considerable trouble.


NON-SWARMERS.

Here I will endeavor to be brief; I feel anxious to get through with
this disagreeable part, where every word I say will clash with
somebody's interest or prejudice. The merits of this hive are to obtain
surplus honey with but little trouble, which often succeeds in
satisfying people of its utility. The principal objection is found on
the score of profit. Suppose we start with one, call it worth five
dollars in the beginning, at the end of ten years it is worth no more,
very likely not as much, (the chances of its failing, short of that
time, we will not take into the account;) we might get annually, say
five dollars worth of surplus honey, amounting to fifty dollars.


CONTRAST OF PROFIT.

The swarming hive, we suppose, will throw off one swarm annually, and
make us one dollar's worth of surplus honey, (we will not reckon that
yielded by the first swarm, which is often more than that from the old
stocks,) about one third of the average in good seasons. The second
year there will be two to do the same; take this rate for ten years, we
have 512 stocks, either of them worth as much as the non-swarmer, and
about a thousand dollars worth of surplus honey. Call these stocks
worth five dollars each, which makes $2,560, all added together will
make the snug little sum of about $3,500, against $55. It is not to be
expected that any of us will realize profits to this extent, but it is
a forcible illustration of the advantages of the swarming hive over the
non-swarmer.


PRINCIPLE OF SWARMING NOT UNDERSTOOD.

But many of these non-swarmers, 'tis said, can be changed to swarmers
to suit the convenience of the apiarian--Colton's is one. It is
asserted that it can be made to swarm within two days at any time,
merely by taking off the six boxes or drawers that are very ingeniously
attached; as this contracts the room, the bees are forced out. Now I
will candidly confess that I could never get this thing to work at all.
Of this I am quite positive, that he (Mr. Colton) is either ignorant of
the necessary and regular preparations that bees make before swarming,
or supposes others are. Mr. Weeks has advocated the same principle: he
says, "There is no queen in any stage of existence, in the old stock,
immediately after the first swarm leaves it." I have examined this
matter till I am satisfied I risk but little in the bold assertion,
that not one stock in fifty will cast a swarm short of a week after
commencing preparations. This opinion will be adopted by whoever will
take the trouble to investigate for themselves. (The chapter on
swarming will give the necessary instructions for examining this point,
if you wish.)


NOT TO BE DEPENDED UPON.

Further, these non-swarmers are not always to be depended upon as such.
They will sometimes throw off swarms when there is abundant room in the
hive as well as in the boxes.


HIVES NOT ALWAYS FULL BEFORE SWARMING.

I know Weeks, Colton, Miner and others, tell us the hive _must be full_
before we need expect a swarm; but experience is against them. Bees do
sometimes cast a swarm before filling the hive. From close observation,
I find when a hive is very large, say 4,000 cubic inches, and is filled
with comb, the first season, that such seldom swarm except in very good
years.


SIZE OF HIVES NEEDED.

But if such hive is only half full, or 2,000 inches, it is very common
for them to swarm without adding any new comb; proving very
conclusively that a hive that size, is sufficient for all their wants
in the breeding season. When about 1,200 inches only had been filled
the first year, I have known them to add combs until they had filled
about 1,800, and then cast a swarm, proving also that a little less
than 2,000 will do for breeding. I have tested the principle of giving
room to prevent swarming, a little further.


AN EXPERIMENT.

In the spring of '47, I placed under five full hives, containing 2,000
solid or cubic inches, as many empty ones, the same size, without the
top. I had a swarm from each; but two had added any new comb, and these
but little. If these hives had been filled to the bottom with comb in
the spring, it is very doubtful whether either of them would have
swarmed. The only place we can put a good stock and not expect it to
swarm in good seasons, is inside a building, where it is perfectly
dark, and even here a few have been known to do it. If we could manage
to get _a very large hive_ filled with combs, it would perhaps be as
good a preventive as any. All the bees that could be reared in one
season, would have sufficient room in the combs ready made for their
labors, and there would be no necessity for their emigration. "But what
becomes of all the bees raised in the course of several years?" To this
question I shall not probably be able to give a satisfactory answer at
present.


BEES DO NOT INCREASE, IF FULL, AFTER THE FIRST YEAR, IN SAME HIVE.

I only will notice the fact, that the bees somehow disappear, and there
is no more at the end of five years than at the end of one. A stock of
bees may contain 6,000 the first of May, and raise 20,000 in the course
of the year; by the first of the next May, as a general thing, not one
more will be found, even when no swarm had issued.


GILLMORE'S SYSTEM DOUBTED.

Now this fact is not known by a recent patentee from the State of
Maine, (else he supposes others do not,) as he recommends placing bees
in a house, and empty hives in connection with the one containing bees,
and in a few years all will be full. He has discovered a mixture to
feed bees, (to be noticed hereafter); this may account for an unusual
quantity being stored by an ordinary sized family. He said another
thing, that is, each of these added hives would contain a queen! This
would seem to explain away the first difficulty of the continued
increase of bees, and so it would if it did not get into another
equally erroneous; one error never made another true. This idea of bees
raising a queen, merely because they have a side box to the main hive,
is contrary to all my experience, and to the experience of all writers
(except himself) that I have consulted. If the principle is correct,
why not sometimes raise a queen in a box on the top or side for us? I
never discovered a single instance, where two perfect queens were
quietly about their duties in connection with one hive. The deadly
hostility of queens is known to all observing apiarians. Not having the
least faith in the principle, I will leave it.


UTILITY OF MOTH-PROOF HIVES DOUBTED.

As for moth-proof hives, I have but little to say, as I have not the
least faith in one of them. When I come to speak of that insect, I will
show, I think, conclusively, that no place where bees are allowed to
enter is safe from them.

Several other _perfect hives_ might be mentioned; yet I believe that I
have noticed the principles of each. Have I not said enough? Such as
are not satisfied now would not be if I filled a volume. Our view of
things is the result of a thousand various causes; the most powerful is
interest, or prejudice.

It is said that in Europe, the same ingenuity is displayed in twisting
and torturing the bee, to adapt her natural instinct to unnatural
tenements; tenements invented not because the bee needs them, but
because this is a means available for a little change. "Patent men"
have found the people generally too ignorant of apiarian science. But
let us hope that their days of prosperity in this line are about
numbered.


INSTINCTS OF THE BEE ALWAYS THE SAME.

Let us fully understand that the nature of the bee, when viewed under
any condition, climate, or circumstance, is the same. Instincts first
implanted by the hand of the Creator, have passed through millions of
generations, unimpaired, to the present day, and will continue
unchanged through all future time, till the last bee passes from the
earth. We may, we have, to gratify acquisitiveness, forced them to
labor under every disadvantage; yes, we have compelled them to
sacrifice their industry, prosperity, and even their lives have been
yielded, but never their instincts. We may destroy life, but cannot
improve or take from their nature. The laws that govern them are fixed
and immutable as the Universe.

Spring returns to its annual task; dissolves the frost, warms into life
nature's dormant powers. Flowers with a smile of joy, expand their
delicate petals in grateful thanks, while the stamens sustain upon
their tapering points the anthers covered with the fertilizing pollen,
and the pistil springs from a cup of liquid nectar, imparting to each
passing breeze delicious fragrance, inviting the bee as with a thousand
tongues to the sumptuous banquet. She does not need an artificial
stimulus from man, as an inducement to partake of the feast; without
his aid or assistance she visits each wasting cup of sweetness, and
secures the tiny drop, while the superabundant farina, dislodged from
the nodding anthers, covers her body, to be brushed together and
kneaded into bread. All she requires at the hands of man, is a suitable
storehouse for her treasures. In good seasons, her nature Will prompt
the gathering for her own use an over supply. This surplus man may
appropriate to his own use, without detriment to his bees, providing
his management is in accordance with their nature.


PROFIT THE OBJECT.

To give the bees all necessary advantages, and obtain the greatest
possible amount of profit, with the least possible expense, has been my
study for years. I might keep a few stocks for amusement, even if it
was attended with no dollar and cent profit, but the number would be
_very small_; I will honestly confess then, that _profit_ is the
actuating principle with me. I have a strong suspicion that the
majority of readers have similar motives. I am sure, then, that all of
us with these views, will consider it a pity, when a stock produces
five dollars worth of surplus honey, to be obliged to pay three or four
of it for patent and other useless fixings.


COMMON HIVE RECOMMENDED.

I would not exchange the hive I have used for the last ten years for
any patent I ever saw, if furnished gratis. I will guarantee that it
affords means to obtain surplus honey, as much in quantity and in any
way which fancy may dictate, whether in wood or glass, and what is more
than all, it shall cost nothing for the privilege of using.


SIZE IMPORTANT.

After deciding what kind of hive we want, the next important point is
the size. Dr. Bevan, an English author, recommends a size "eleven and
three-eighths inches square, by nine deep in the clear," making only
about 1,200 inches, and so few pounds necessary to winter the bees,
that when I read it, I found myself wondering if the English inch and
pound were the same as ours.


SMALL HIVES MORE LIABLE TO ACCIDENTS.

At all events, I think it too small for our Yankee bees in any place.
We must remember, that the queen needs room for all her eggs, and the
bees need space to store their winter provisions; for reasons before
given, this should be in one apartment. When this is too small, the
consequences will be, their winter supply of food is liable to run out.
The swarms from such will be smaller and the stock much more liable to
accidents, which soon finish them off.


APT TO DECEIVE.

Yet I can imagine how one can be deceived by such a small hive, and
recommend it strongly; especially if patented. Suppose you locate a
large swarm in a hive near the size of Dr. Bevan's; the bees would
occupy nearly all the room with brood-combs; now if you put on boxes,
and as soon as filled put on empty ones, the amount of surplus honey
would be great; very satisfactory for the first summer, but in a year
or two your little hive is gone. This result will be in proportion as
we enlarge our hives, until we arrive at the opposite extreme.


UNPROFITABLE IF TOO LARGE.

If too large, more honey will be stored than is required for their
winter use. It is evident a portion might have been taken, if it had
been stored in boxes. The swarms will not be proportionably large when
they do issue, which is seldom--but there is this advantage, they last
a long time, and are but little profit in surplus honey, or swarms.


CORRECT SIZE BETWEEN TWO EXTREMES.

Between the two extremes, like most other cases, is found the correct
place. A hive twelve inches square, each way, inside, has been
recommended as the correct size. Here are 1,728 cubic inches. This, I
think, is sufficient for many places, as the queen probably has all the
room necessary for depositing her eggs; and as the swarms are more
numerous, and nearly as large as from hives much larger; also, there is
room for honey sufficient to carry the bees through the winter, at
least, in many sections south of 40 degrees latitude, where the winter
is somewhat short.


SIZE FOR WARM LATITUDES.

This size will also do in this latitude (42 degrees,) in some seasons,
but not at all in others.[3] Not one swarm in fifty will consume
twenty-five lbs. of honey through the winter, that is, from the last of
_September_ to the first of April, (six months). The average loss in
that time is about eighteen lbs.; but the critical time is later; about
the last of May, or first of June, in many places.

      [3] When Mr. Miner wrote his manual recommending this size, 1,728
      inches, for all situations, it should be remembered he lived on
      Long Island. Since removing to Oneida County in this State,
      either his own experience or _some other cause_ has changed his
      views, as he now recommends my size, viz., 2,000 inches.


LARGER HIVE MORE SAFE FOR LONG WINTERS OR BACKWARD SPRING.

About the first of April they commence collecting pollen and rearing
their young; by the middle of May all good stocks will occupy nearly,
if not quite all, their brood-combs for that purpose, but _little honey
is obtained_ before fruit blossoms appear; when these are gone, no more
of any amount is obtained until clover appears, which is some ten days
later. (I am speaking now particularly of this section; I am aware it
is very different in other places, where different flowers exist.) Now
if this season of fruit flowers should be accompanied by high winds, or
cold rainy weather, but little honey is obtained; and our bees have a
numerous brood on hand that _must be fed_. In this emergency, if no
honey is on hand of the previous year, a famine ensues; they destroy
their drones, perhaps some of their brood, and for aught I know put the
old bees on short allowance. This I do know, that the whole family has
actually starved at this season; sometimes in small hives. This of
course depends on the season; when favorable, nothing of the kind
occurs. Prudence therefore dictates the necessity of a provision for
this emergency, by making the hive a little larger for northern
latitudes, as a little more honey will be stored to take them through
this critical period. From a series of experiments closely observed.


2,000 INCHES SAFE FOR THIS SECTION.

I am satisfied that 2,000 inches in the clear, is the proper size for
safety in this section, and consequently, profit. On an average, swarms
from this size are as large as any.

The dimensions should be uniform in all cases, whatever size is decided
on. It is folly to accommodate each swarm with a hive corresponding in
size; a very small family this year, may be very large next, and a very
large one, very small, &c. A queen belonging to a small swarm will be
capable of depositing as many eggs, as another belonging to a barrel
full. A small family able to get through the winter and spring, may be
expected by another year to be as numerous as any.


KIND OF WOOD, WIDTH OF BOARD, ETC.

Of the kinds of wood for hives, pine is preferable, still other kinds
will do; I have no faith in bees liking one kind better than another,
and less likely to leave on that account. Hemlock is cheaper, and used
to a great extent; when _perfectly sound_ is as good as anything, but
is very liable to split, even after the bees have been in them some
time. It should be used only when better wood cannot be obtained. Bass
wood when used for hives should _always be painted_, and then will be
very liable to warp from the moisture arising from the bees inside.
When not painted outside, and allowed to get wet, if only for a few
hours, so much moisture is absorbed that it will bend outward, and
cleave from the combs and crack them. A few days of dry weather will
relieve the outside of water, and the inside kept moist by the bees,
the bending will be reversed, and the combs pressed inward, keeping the
bees fixing that which will not "stay fixed." Perhaps there is wood as
suitable or better than pine, but it is not as common.


SHAPE OF NO CONSEQUENCE.

Boards should be selected, if possible, that will be the proper width
to make the hive about square, of the right size. Say twelve inches
square, inside, by fourteen deep. I prefer this shape to any other, yet
it is not all important. I have had some ten inches square by twenty in
length; they were awkward looking, but that was all, I could discover
no difference in their prosperity. Also, I have had them twelve inches
deep by thirteen square, with the same result. Hence, if we avoid
extremes, and give the required room, the shape can make but little
difference.

It has been recommended to plane the boards for hives, "inside and
out;" but bees, when first put into such hive, find much difficulty in
holding fast until they get their combs started, hence this trouble is
worse than useless.


DIRECTIONS FOR MAKING HIVES.

If hives are not desired of the cheapest possible construction, the
outside may be planed and painted; but it is doubtful whether strict
economy would demand it. Yet a painted hive appears so much better,
that it ought to be done, especially as the paint adds almost enough to
its durability to pay the expense. The color may be whatever fancy
dictates; the moth will not probably be attracted by one color more
than another. White is affected the least by the sun in hot weather.
Lime is put on as white-wash, annually, by many, as a protection
against insects.

When hives are not painted, the grain should never be crosswise, having
the width of boards form the height; not that the bees would have any
dislike to such, but nails will not hold firmly, they draw out in a few
years. The size, shape, materials, and manner of putting together, are
now sufficiently understood, for what I want. Sticks half an inch in
diameter, should cross each way through the centre, to help support the
combs. A hole about an inch diameter in the front side, half way to the
top, is a great convenience for the bees to enter when coming home
heavy laden.

It now remains to make the top, cover, and boxes, (the bottom-board
will be described in another chapter.) The tops should be all alike;
boards fifteen inches square are just the right size; three-fourths of
an inch is the best thickness, (inch will do;) plane the upper side,
rabbet out around the edge of the upper side one inch wide, and
three-eighths deep; this will leave the top inside the rabbeting, just
thirteen inches.


SIZE OF CAP AND BOXES.

A box for a cover or cap, that size inside, will fit any hive. The
height of this box should be seven inches. Of course other sizes will
do, but it is best to commence with one that we can adhere to
uniformly, and no vexations arise by covers not fitting exactly, &c. I
think this size is as near correct as we shall be likely to get; we
want all the room in the boxes that the majority of our stocks demand
for storing in a yield of honey,[4] at the same time not be
necessitated to give too much of the room in the height. They will
commence work in a box five inches high, much sooner than one seven or
eight. To give the requisite room, and have the boxes less than five
inches high, would require more than thirteen inches on the top, this
would make the hive too much out of shape; it would appear top-heavy.

      [4] I have added a side box occasionally, but it has seldom paid
      me for the trouble.


MINER'S HIVE.

Miner's Equilateral Hive has a cap somewhat smaller than this in
diameter; consequently, if we have the requisite room, it must be in
its height. But by making the cap of his a little larger, and a few
trifling alterations, it would do very well for a patent. And if any
one _must_ have a patent hive, my advice is to get that; it costs but
two dollars for the right of using, and is nearer what we want for
bees, than any I ever saw. I prefer rabbeting around the edge of the
top, instead of nailing on a thin board the size of the inside of the
cover, with room for a slide under it; it affords too nice a place for
worms to spin their cocoons. Also, without the rabbeting water may get
under the cap, and pass along the top till a hole lets it among the
bees. As for slides, I do not approve of them at all; in shutting off
communication, it is almost certain to crush a few bees. This makes
them irritable for a week; they are unnecessary for me, at least. We
will now finish the hive.


DIRECTIONS FOR MAKING HOLES.

After the top is got out as directed, strike a line through the centre,
three and a quarter inches from this, make another on each side, now
measure on one of the last lines, two and a half inches for the first
hole, two inches for the next, and so on till five are marked on this,
and the same number on the other side, ten in all; these holes should
be about an inch diameter, a pattern three and a quarter inches wide,
and thirteen in length, with places for holes marked on it, will save
time when many are made. When this top is nailed on, the hive is ready.
A less number of holes is often used, and one is thought by some to be
sufficient; experience has satisfied me that the more room bees have to
enter boxes, the less reluctance is manifested in commencing their work
in them; but here is another extreme to be avoided: when the holes are
much larger, or more of them, or even one very large one, the queen is
very apt to go into the boxes and deposit her eggs, which renders the
comb tough, dark, &c., also bee-bread is stored near the brood. Dr.
Bevan's and Miner's cross-bar hives are objectionable on this account,
they offer too free access to the boxes; we want all the room that will
answer, and no more.


A SUGGESTION.

Mr. Miner's cross-bar hive is intended to make the bees construct all
straight combs, and probably will do it. But the disadvantage of
bee-bread and brood in the boxes will not be made up by straight combs.

For the benefit of those who have been made to believe straight combs
_all important_, and perhaps have purchased the right to make the hive,
and had some constructed, and have found bee-bread in their surplus
honey, I would suggest an improvement, (that is, if it is thought the
straight combs will pay. If you have not the right for the cross-bar
hive, and you wish to use it, I would say, buy the right, and remove
all grounds of complaint with him.) Put in the bars and hive your bees
as he directs. After all the combs are started, instead of setting the
open bottom boxes (which are also unsuitable for sending to market)
directly on the bars as he recommends, take off the cloth, and with
screws fasten on a top with ten holes, that I have just described; and
then you will have the straight combs, and surplus honey in the boxes
pure.


GLASS BOXES PREFERRED.

Having told how I make a hive, I will now give some reasons for
preferring a particular kind of boxes. I have taken great quantities of
honey to market, put up in every style, such as tumblers, glass jars,
glass boxes, wooden boxes with glass ends, and boxes all wood. I have
found the square glass boxes the most profitable; the honey in such
appears to the best possible advantage, so much so, that the majority
of purchasers prefer paying for the box at the same rate as the honey,
than the wood box, and have the tare allowed. This rate of selling
boxes always pays the cost, while we get nothing for the wood. Another
advantage in this kind of boxes is, while being filled, the progress
can be watched, and the time they are finished known precisely, when
they should be taken off, as every day they remain after that, soils
the purity of the combs.


GLASS BOXES--HOW TO MAKE.

_Directions for making._--Select half-inch boards of pine or other soft
light wood, cut the length twelve and three-quarters inches, width six
and three-eighths inches, dress down the thickness to three-eighths or
less, two pieces for a box, top and bottom, in the bottom bore five
holes throughout the centre to match with those in the top of the hive,
(the pattern used in marking the top of hives is just the one to mark
these). Next, get out the corner posts, five-eighths of an inch square,
and five inches in length; with a saw, thick enough to fit the glass,
cut a channel length-wise on two sides, one-fourth of an inch deep,
one-eighth from the corner, for the glass. A small lath nail through
each corner of the bottom into the posts will hold them; it is now
ready for the glass--10×12 is the right size to get--have them cut
through the centre the longest way for the sides, and they are right,
and again the other way, five and five-eighths long for the ends. These
can now be slipped into the channels of the posts, and the top nailed
on like the bottom, and the box is ready.


GUIDE-COMBS NECESSARY.

It will be found a great advantage, previous to nailing on the top, to
stick fast to it some pieces of guide-combs in the direction you wish
the bees to work. They are also an inducement for them to commence
several days sooner, than if they had to start combs for themselves;[5]
a piece an inch square will do; it is well to start every comb you want
in the box; two inches apart is about the right distance to look well.
To make these pieces hold fast, melt one edge by the fire, or candle,
or melt some bees-wax, and dip one edge in that, and apply it before
cooling; with a little practice you can make them stick without
difficulty. For a supply of such combs, save all empty, clean, white
pieces you can, when removing combs from a hive.

      [5] A line of bees-wax made with a guide-plate, or other means,
      is found to be of but little use.

If you have any way superior to this for making glass boxes, so much
the better, make them so by all means: "The best way is as good as
any." I give my method to be used only when better is not convenient.
If you sell honey, I think you will find it an advantage to have glass
boxes made in some way. Two of this size when full weigh 25 lbs. If
preferred, four boxes six and three-eighths inches square, can be used
for a hive instead of two; the expense of making is a little more for
the same number of lbs., yet, when it is in market, a few customers
will prefer this size.


WOOD BOXES.

For home consumption, the wood-box will answer equally well for all
purposes of obtaining the honey, but will give no chance to watch the
progress of the bees, unless a glass is inserted for the purpose, and
then it will need a door to keep it dark, or a cover over the whole
like the one for glass boxes, may be put on. Wood boxes are generally
made with open bottom, and set on the top of the hive. A passage for
the bees out of the box to the open air is unnecessary, and worse than
useless. They like to store their honey as far from the entrance as
possible. Unless crowded for room, they will not store much there when
such entrances are made.

Whether we intend to consume our surplus honey or not, it is as well to
have the hives and covers made in a manner that we can use glass, when
we are likely to have some to spare. I am not sure, but it would pay to
make hives in this way, even if glass boxes were never used; the
rabbeting prevents light as well as water from passing under the cover;
imagine a box set on a plain board nailed on for a top, without the
rabbeting; the warping or bending admits the light and water,
especially when hives are out in the weather, (and I shall not
recommend any other way of keeping them.)


COVER FOR HIVES.

I have termed the cap or box a cover; but this should also be covered
with a board laid on, if nothing else. A good roof for each hive can be
made by fastening two boards together like the roof of a building; let
it be about 18 by 24 inches; it being loose, can be changed in
accordance with the season; in spring, let the sun strike the hive; but
in hot weather let the longest end project over the south side, &c. You
can ornament this hive, if you choose, by mouldings or dentals, under
the top, where it projects over the body of the hive, also the cap can
have the top projected a little and receive the same addition.


JARS AND TUMBLERS--HOW PREPARED.

When jars, tumblers, or other vessels, that are all glass, are used, it
is _absolutely necessary_ to fasten as many pieces of combs as you wish
made, in the top, for a beginning, or fasten a piece of wood there; as
they seldom commence building on glass, without a start.

Some of you may have seen paraded at our fairs, or in the public parts
of some of our cities, hives containing tumblers, some of them neatly
filled, others empty, and this meagre sentence written upon them, _not
to be filled_! Pretending to govern the bees, as the juggler sometimes
does his tricks, by mysterious incantations! I once encountered an
agent of this humbug, and modestly suggested to him that I had a
counter charm: that I could put a tumbler on his hive and it would be
filled if the others were, however much he might forbid it by written
charms! He saw at a glance how the matter stood; I was not the customer
he wanted, and intimated that the show was only intended for the
extreme verdancy of most visitors. It no doubt assisted in displaying
his profound knowledge in bee management, which he wished to establish,
as he had a little work on the subject to sell, also hives, and bees.
The reader no doubt will guess as I did, the reason that those tumblers
were not filled, was because no combs were put in for a start.


PERFECT OBSERVATORY HIVE DESCRIBED.

There are many things pertaining to bees that cannot be properly
examined and understood, without a glass hive of some sort. Yet a
perfect observatory hive containing but one comb, is not a perfect hive
for the bees. We can see very well what the bees are doing, but it is
not a tenement they would choose if left to themselves. It forces them
to labor in an unnatural manner, is unsuitable for wintering bees, and
otherwise but little profit. If the satisfaction of witnessing some of
their operations more perfectly than in glass hives of another kind
will not pay, it is doubtful if we get it. I will describe as briefly
as possible. Two frames or sashes about two and a half feet square,
containing glass, are so fastened together as to leave room for only
one comb between them, about an inch and three-fourths apart. A comb of
this size will not support itself by the top and edges; hence, it is
necessary to put in numerous cross-bars to assist in supporting it.
Outside the glass are doors to keep the whole dark, to be opened when
we wish to inspect proceedings. Under the bottom is a board or frame,
to keep it in an upright position, &c. Probably but few will be induced
to make one. I will therefore describe another; a hive that I think
will pay better.


ONE LIKE COMMON HIVE PREFERRED.

If we expect to know what bees are doing in ordinary hives, we must
have one similar in every respect, in size, shape, number of bees, &c.
The construction of royal cells will be watched by most observers with
the greatest interest; now these are generally on one edge of the
combs. The bees leave a space half an inch or more between the edges of
the combs and one side of the hive, near half the length of it,
apparently for no other purpose but to have room for these cells, as
the other edges of the same combs are generally attached to the hive at
the bottom.


WHAT MAY BE SEEN.

Now instead of having one piece or pane of glass in the side of several
hives, I would recommend having one or more with glass on every side;
because we might have it on three sides, and not the fourth; and this
might contain all the queen cells, and we should miss an important
sight. There are many other things to be witnessed in such a hive. The
queen may be often seen depositing her eggs! We may see the workers
detach the scales of wax from their abdomen, and apply them to the
combs during the process of construction, see them deposit pollen from
their legs, store their honey, feed the queen, each other, their young
brood, seal over cells containing brood, honey, &c. It is further
useful as a guide for putting boxes on other hives, (that is, if it is
a good one, which it should be); we can easily ascertain whether our
bees are gaining or losing.


DIRECTIONS FOR MAKING GLASS HIVE.

My method of making them is as follows: The top is like those for other
hives, fifteen inches square, adapted to boxes and cover. This hive we
want to be as profitable as any, giving us surplus honey, and swarms
like others. Four posts are then got out, two inches square, and
thirteen in length; care should be taken to have the ends perfectly
square.

A frame is then to be made, just fourteen inches square outside, for
the bottom; the pieces are one inch thick, by two in width, halved
together at the corners. A guage-mark is then made around the under
side of the top, half an inch from the edge, a post is then set inside
of each corner of this mark, and thoroughly nailed, the bottom is
nailed on with the posts even with the outside corners. Four pieces an
inch thick, and an inch and a half wide, are fitted between the posts,
even with the guage-mark on the top. Sixteen strips, about one quarter
by half an inch, are got out, eight to be ten, and eight twelve inches
long.

A gauge-mark one inch from posts, bottom, &c., is the place to nail
these strips; very small nails or tacks will hold them. The panes of
glass are to rest against them, which are held in their places by small
pieces of tin, or brads. The doors are the size of the glass, 10×12,
about three-fourths of an inch thick; these doors are cut a little too
short, and the pieces, to prevent warping, are nailed on the ends;
these are hung to a post on one side, and secured by a button on the
other. On two opposite sides inside the posts, half way up, two strips,
half an inch by three quarters, are nailed, with holes in them for the
cross-sticks; one way is enough if you have guide-combs for a start,
like those recommended for boxes, so that the sheets will be at right
angles with them; otherwise, let the sticks cross both ways, about
three each way will be needed, as the glass at the edges is not so good
a support as wood.

The cap can be made of half inch boards; the top to project over like
the hive, or let it be a little more than half an inch, it will admit a
heavier moulding, which should surround it here, as well as at the top
of the hive, or if it is prefered, dentals can be used, and look
equally well--when no ornament is wanted, omit it. But painting seems
necessary for such hives, to prevent warping, and the swelling of the
doors in wet weather; these want to open and shut without rubbing or
sticking, otherwise we disturb the bees every time a door is stirred.
Putty should not be used to hold the glass, as the bees in the course
of a few years will cover it with propolis; it is then necessary to
take it out, and scrape, clean, and return it, when, if fastened with
putty, it would be difficult; cold weather is the time for this
operation. I am aware that a hive can be more substantially made than
the one here described; but I have endeavored to make one as cheap as
possible, and if properly made, will answer. The cost will be much less
than many patents, and the satisfaction much more, at least, with many.
When our hive contains a swarm of bees, and they are thoroughly in
operation, we must not let them pass out at the bottom on every side,
as they are frequently allowed to do from other hives; because, should
one come out a little excited in consequence of a slight jar,
accidentally given the hive, on opening the door or some other way, and
should find our face within a foot of their house, peering in the
window among their works, it would be very likely to give us _a gentle
hint_ that it was a mark of low breeding, that we were not wanted there
at all, and that it was none of our business what they were doing. To
prevent this as far as possible, a bottom-board, somewhat different
from the common one, is needed. Four posts of chestnut or other lasting
wood, about two inches square, are driven into the earth in the form of
a square, far enough apart to come under the corners of the
bottom-board, (fifteen inches,) and high enough for convenience when
looking into the hive. The ends of these posts are to be perfectly
level, and to which the bottom is to be nailed fast. As the hive is to
sit perfectly close to the board, a passage must be made through it, as
well as means for ventilation in hot weather, without raising the hive
for that purpose. It requires a board about fifteen inches square,
planed smooth, the ends clamped to prevent warping or splitting; a
portion of the centre is taken out, say six inches by ten, and wire
cloth nailed over, four-ounce tacks will hold it, fasten it just enough
to keep the bees from getting through; very likely it will want to be
taken off occasionally and cleaned from the propolis that will be
spread over it. It is easiest done in freezing weather.

Take an edge in each hand, and rock the wires a few times out of
square, and it will readily crumble and fall out. In warm weather it
must be scalded or burnt off. To close this space, a moving slide is
fixed in grooves under-side, fastened to the posts or board. The slide
is to be moved in accordance with the weather, when cold, close it,
when hot, withdraw it, and give the bees as much air as possible,
without raising the hive, the whole of such space is as much
ventilation as ordinary hives raised an inch. (Wire cloth is needed for
other purposes, it is best to procure some, even at considerable
trouble and expense.) On the side of the board intended for the front,
two inches from the edge of the wire cloth, a passage is cut for the
bees, three-eights of an inch wide, by eleven in length. "But how is
the bees to get to this place, so inconvenient, something is needed to
assist them?" Certainly, Sir; an alighting board, eleven inches wide,
and about two feet long, (not planed), is placed at an angle of
forty-five degrees, between the two front posts of your stand, the
upper end passing under the bottom, far enough back; to be just even
with the back-side of the passage for the bees. The bees alight on this
board, and walk up into the hive without difficulty. When the bees are
at work pretty freely, and a door of this hive is opened, those that
are about departing will be very likely to get on the glass, instead of
through the opening at the bottom; seeing the light through the glass,
they endeavor to escape by the nearest route. When so many gather here
as to prevent a good view, and you wish to observe further, shut the
door a moment and they will leave through their own passage, when you
can open your door again, for a short time. After the hive is filled
with combs, the number attracted to the glass on opening a door will be
much less.

The plate on the preceding page represents a glass hive, cover, and
stand. The common hive can be made equally ornamental, if you choose;
this kind of stand is unnecessary for them. I use such as are
recommended on page 138.



CHAPTER III.

BREEDING.


IMPERFECTLY UNDERSTOOD.

The time that bees commence raising their young brood is but
imperfectly understood by most people. Many persons that have kept them
for years, have bestowed so little attention on this point, that they
are unable to tell at what time they commence, how they progress, or
when they cease. A kind of an idea that one swarm, and occasionally two
or three, are reared sometime in June, or fore part of summer, is about
the extent of their reflections on the subject. Whether the drones
deposit the eggs, or that a portion of the workers are females, and
each raise a young one or two, or whether the "king bee" is the chap
for laying eggs, is a matter beyond their ability to answer. It is but
a few years since, that a correspondent of a Journal of Agriculture
denied the existence of a queen bee, giving the best reasons he had, no
doubt, that is, he had never seen one. But bee-keepers of this class
are so few, it is unnecessary to waste time to convince them; suffice
it to say, that a queen exists with every prosperous swarm, and all
apiarians with much pretensions to science, acknowledge the fact, also,
that she is the mother of the whole family.

The period at which they commence depositing eggs probably depends on
the strength of the colony, amount of honey on hand, &c., and not the
time they commence gathering food.


GOOD STOCK SELDOM WITHOUT BROOD.

I once removed the bees from a hive on the tenth of January, and found
brood amounting to about five hundred, sealed over, and others in every
stage of growth down to the egg.

This hive had been in the house, and kept warm; it will doubtless be
supposed that being kept warm was the cause; but this is not a solitary
instance. A neighbor lost a hive the fourteenth February, in weather
cold enough to seal the entrance with ice, and smother the bees. I
assisted to remove the combs, and found young brood in abundance, from
the perfect bee, through all stages of growth. This stock had been in
the cold all winter. I have further noticed, when sweeping out the
litter under the hives early in spring, say the first of March, that
young bees would often be found under the best stocks. Hence it appears
there is but little time, and perhaps none, when our best stocks have
no broods. Yet stocks, when very weak, do not commence till warm
weather. It seems that a certain degree of warmth is necessary to
perfect the brood, which a small family cannot generate.


HOW SMALL STOCKS COMMENCE.

The first eggs are deposited in the centre of the cluster of bees, in a
small family; it may not be in the centre of the hive in _all_ cases;
but the middle of the cluster is the warmest place, wherever located.
Here the queen will first commence; a few cells, or a space not larger
than a dollar, is first used, those exactly opposite on the same comb
are next occupied. If the warmth of the hive will allow, whether mild
weather produces it, or the family be large enough to generate that
which is artificial, appears to make no difference; she will then take
the next combs exactly corresponding with the first commencement but
not quite as large a place is used as in the first comb. The circle of
eggs in the first is then enlarged, and more are added in the next,
&c., continuing to spread to the next combs, keeping the distance to
the outside of the circle of eggs, to the centre or place of beginning,
about equal on all sides, until they occupy the outside comb. Long
before the outside comb is occupied, the first eggs deposited are
matured, and the queen will return to the centre, and use these cells
again, but is not so particular this time to fill so many in such exact
order as at first. This is the general process of small or medium sized
families. I have removed the bees from such, in all stages of breeding,
and always found their proceedings as described.


DIFFERENT WITH LARGER ONES.

But with very large families, their proceedings are different: as any
part of the cluster of bees is warm enough for breeding, there is less
necessity for economizing heat, and having all the eggs confined to one
small spot, some unoccupied cells will be found among the brood; a few
will contain honey and bee-bread.


HOW POLLEN IS STORED IN THE BREEDING SEASON.

But in the height of the breeding season, a circle of cells nearly all
bee-bread, an inch or two wide, will border the sheets of comb
containing brood. As bee-bread is probably the principal food of the
young bee, it is thus very convenient.

When pollen is abundant, and the swarm is in prosperous condition, they
soon reach the outside sheets of comb with the brood. At this period,
when the hive is about full, and the queen is forced to the outside
combs to find a place for her eggs, it is interesting to witness
operations in a glass hive. I have seen her several times during one
day, on the same piece of comb (next the glass). The light has no
immediate effect on her "Highness," as she will quietly continue about
her duty, not the least embarrassed by curious eyes at the window.
Before depositing an egg, she enters the cell head first, probably to
ascertain if it is in proper condition to receive it; as a cell part
filled with bee-bread or honey is never used. If the area of combs is
small, or the family is small, and cannot protect a large space with
the necessary heat, she will often deposit two, and sometimes three, in
one cell (the supernumeraries I suppose are removed by the workers).
But under prosperous circumstances, with a hive of suitable size, &c.,
this emergency is avoided.


OPERATION OF LAYING AND THE EGGS DESCRIBED.

When a cell is in a condition to receive the egg, on withdrawing her
head she immediately curves her abdomen, and inserts it a few seconds.
After leaving it, an egg may be seen attached by one end to the bottom;
about the sixteenth of an inch in length, slightly curved, very small,
nearly uniform the whole length, abruptly rounded at the ends,
semi-transparent, and covered with a very thin and extremely delicate
coat, often breaking with the slightest touch.

After the egg has been about three days in the cell, a small white worm
may be seen coiled in the bottom, surrounded with a milky-like
substance, which is its food, without doubt. How this food is prepared,
is merely guess-work. The hypothesis of its being chiefly composed of
pollen, I have no objection to; as it is sufficiently proved by the
quantities that accumulate in hives that lose their queen, and rear no
brood (that is, when a requisite number of workers are so left). The
workers may be seen entering the cell every few minutes, probably, to
supply this food.[6]

      [6] When the comb in our glass hive is new, and white, these
      operations can be seen more distinctly than when very old and
      dark.


TIME FROM THE EGG TO THE PERFECT BEE.

In about six days it is sealed over with a convex waxen lid. It is now
hidden from our sight for about twelve days, when it bites off the
cover, and comes forth a perfect bee. The period from the egg to the
perfect bee varies from twenty to twenty-four days; average about
twenty-two for workers, twenty-four for drones. The temperature of the
hive will vary some with the atmosphere; it is also governed by the
number of bees. A low temperature probably retards the development,
while a high one facilitates it. You may have seen accounts of the
assiduous attentions given to the young bee when it first emerges from
the cell: 'tis said they "lick it all over, feed it with honey," &c.,
desperately pleased with their new acquisition.


ROUGH TREATMENT OF THE YOUNG BEE.

Now, if you expect to see anything of this, you must watch a little
closer than I have. I have seen hundreds when biting their way out.
Instead of care or notice, they often receive rather rough treatment:
the workers, intent on other matters, will sometimes come in contact
with one part way out the cell, with force sufficient to almost
dislocate its neck; yet they do not stop to see if any harm is done, or
beg pardon. The little sufferer, after this rude lesson, scrambles back
as soon as possible out of the way; enlarges the prison door a little,
and attempts again, with perhaps the same success: a dozen trials are
often made before they succeed. When it does actually leave, it seems
like a stranger in a multitude, with no friend to counsel, or mother to
direct. It wanders about uncared for and unheeded, and rarely finds one
sufficiently benevolent to bestow even the necessaries of life; but
does sometimes. It is _generally_ forced to learn the important lesson
of looking out for itself, the day it leaves the cradle. A cell
containing honey is sought for, where its immediate wants are all
supplied.


GUESS WORK.

The time before it is ready to leave the hive for honey, I might guess
would be two or three days. Others have said "it would leave _the day
it left the cell_;" but I guess they guess at this point. They tell us,
too, that after the bees seal over the cells containing the larvæ,
"they immediately commence spinning their cocoons, which takes just
about thirty-six hours." I think it very likely; but when I admit it, I
cannot imagine how it was ascertained;--the faculty of looking through
a mill-stone I do not possess, and it requires about the same optical
penetration to look into one of these cells after it is sealed over, as
it is all perfect darkness. Suppose we drive away the bees and open the
cell, to give us a look at the interior: the little insect stops its
labor in a moment, probably from the effect of air and light. I never
could detect one in its labor. Suppose we open these cells every hour
after sealing; can we tell anything about their progress by the
appearance of these cocoons, or even tell when they are finished? The
thickness of a dozen would not exceed common writing paper. When a
subject is obscure, or difficult to ascertain, like this, why not tell
us how they found out the particulars; and if they were guessed at, be
honest, and say so? When the bee leaves the cell, a cocoon remains, and
that is about all we _know_ about it.


TERMS APPLIED TO YOUNG BEES.

The young bee, when it first leaves the egg, is termed grub, maggot,
worm, or larva; from this state it changes to the shape of the perfect
bee, which is said to be three days after finishing the cocoon; from
the time of this change, till it is ready to leave the cell, the terms
nymph, pupa, and chrysalis, are applied. The lid of the drone's cell is
rather more convex than that of the worker's, and when removed by the
young bee to work its way out, is left nearly perfect; being cut off
around the edges, a good coat or lining of silk keeps it whole; while
the covering of the worker's cell is mostly wax, and is pretty well cut
to pieces by the time the bee gets out. The covering to the queen's
cell is like the drone's, but larger in diameter, and thicker, being
lined with a little more silk.


DISCREPANCY IN TIME IN REARING BROOD AS GIVEN BY HUBER.

We are told by most writers, the period of time necessary to perfect
from the egg, the three different kinds of bees. Huber leads the way,
and the rest, _supposing him to be right_, repeat in substance his
account as follows: That the whole time necessary to perfect a queen
from the egg is sixteen days, the worker twenty, and the drone
twenty-four days; Huber (as quoted by Harpers) gives the time of each
stage of development belonging to each kind of bee; but is rather
unfortunate in arithmetic; the items, or stages, when added together,
"do not prove," as the school-boys say; that is, he gains time by
making his bee by degrees. He says, first, of the worker, "It remains
three days in the egg, five in the grub state, it is thirty-six hours
in spinning its cocoon; in three days it changes to a nymph, passes six
in that form, and then comes forth a perfect bee." How do the items add?

    The egg,              3     days.
    Grub,                 5      "
    Spinning cocoon,      1-1/2  "
    Changing to a nymph,  3      "
    In that form,         6      "
                       -------
                        18-1/2  days.

One and a half days short. We will next see how the figures with the
royal insect match; recollect sixteen days are all she has allowed;
then, of the different stages, "three days in the egg, is five a worm,
when the bees close its cell, and it immediately begins its cocoon,
which is finished in twenty-four hours. During eleven days, and even
sixteen hours of the twelfth, it remains in a state of complete repose.
Its transformation into a nymph then takes place, in which state four
days and part of the fifth are passed." Now let us add the items:

    The egg,                                   3     days.
    A worm,                                    5       "
    Spinning a cocoon, (24 hours),             1       "
    Reposes eleven days and 16 hours,         11-2/3   "
    A nymph four days, and part of the fifth,  4-1/3   "
                                             -------
                                              25     days.

Now, reader, what do you make of such palpable blundering guess-work? A
difference of nine days--the merest school-boy ought to know better!
Can we rely on such history? Does it not prove the necessity of going
over the whole ground, applying a test to every assertion, and a
revision of the whole matter throughout? My object is not to find
fault, but to get at _facts_. When I see such guess-work as the above
published to the world, in this enlightened age, gravely told to the
rising generation, as a portion of natural history, I feel it a duty
not to resist the inclination to expose the absurdity.


THE NUMBER OF EGGS DEPOSITED BY THE QUEEN GUESSED AT.

The number of eggs that a queen will deposit is often another point of
guess-work. When the estimate does not exceed 200 per diem, I have no
reason to dispute it; the number will probably fall short in some
cases, and exceed it in others. Some writers suppose that this number
"would never produce a swarm, as the bees that are lost daily amount
to, or even exceed that number," and give us instead from eight hundred
to four thousand eggs in a day, from one queen. The only way to test
the matter accurately, is by actually counting, in an observatory hive,
or in one with sufficient empty combs to hold _all the eggs_ she will
deposit for a few days, when, by removing the bees, and counting
carefully, we might ascertain, and yet several would have to be
examined, before we could get at the average. The nearest I ever came
to knowing anything about it happened as follows: A swarm left, and the
queen from some cause was unable to cluster with it, and was found,
after some trouble, in the grass a few rods off. She was put in the
hive with the swarm about 11 o'clock, A.M.; the next morning, at
sunrise, I found on the bottom-board, among the scales of wax, 118 eggs
that had been discharged in that time. Probably a few escaped notice,
as the color is the same as wax scales; also, they might already have
had combs containing some. I have several times found a few the next
morning, under swarms hived the day previous, but never over thirty,
except in this one instance. The reason of this queen not being able to
fly well might have been an unusual burden of eggs. Perhaps it would be
as well to mention here, that in all cases where eggs are found in this
way, that they must be first swarms which are accompanied by the old
queens.

Schirach estimates "the eggs a single female will lay, from 70,000 to
100,000 in a season." Reaumer and Huber do not estimate so high.
Another writer estimates 90,000, in three months. Let the number be as
it may, probably thousands are never perfected. During the spring
months, in medium and small families, where the bees can protect with
animal heat but a few combs, I have often found cells containing a
plurality of eggs, two, three, and occasionally four, in a single cell.
These supernumeraries must be removed, and frequently may be found
amongst the dust on the bottom-board.


A TEST FOR THE PRESENCE OF A QUEEN.

If you have a hive that you suspect has lost a queen at this season,
her presence can be ascertained nine times in ten by this method. Sweep
off the board clean, and look the next day or two after for these eggs.
Take care that ants, or mice, have no chance to get them; they might
deceive you, being as fond of eggs for breakfast as anyone.[7] When one
or more is found, or any immature bees, it is sufficient, no further
proof of the presence of a queen is needed.

      [7] It is said that the bees will devour these eggs also.

Another portion of eggs is wasted whenever a supply of their food
fails; if we remove the bees from a stock during a scarcity, when the
hive is light, we will be very likely to find hundreds of eggs in the
cells, and but very few advancing from that stage towards maturity. I
have thus found it in the fall, in July, and sometimes the first of
June, or at any time when maturing the brood would be likely to exhaust
their stores, to endanger the family's supply. Now, instead of the
fertility of the queen being greater in spring and first of summer than
at other times, (as we are often told), I would suggest the probability
that a greater abundance of food at this season, and a greater number
of empty cells, may be the reason of the greater number of bees
matured.


WHEN DRONES ARE REARED.

Whenever the hive is well supplied with honey, and plenty of bees, a
portion of eggs are deposited in the drone-cells, which three or four
days more are necessary to mature than the worker.


WHEN QUEENS ARE REARED.

Also, when the combs become crowded with bees, and honey plenty, the
preparations for young queens commence: as the first step towards
swarming, from one to twenty royal cells are begun; when about half
completed, the queen (if all continues favorable) will deposit eggs in
them, these will be glued fast by one end like those for the workers;
there is no doubt but they are precisely the same kind of eggs that
produce other bees. When hatched, the little worm will be supplied with
a superabundance of food; at least, it appears so from the fact, that a
few times I have found a quantity remaining in the cell after the queen
had left. The consistence of this food is about like cream, the color
some lighter, or just tinged with yellow. If it was thin like water, or
even honey, I cannot imagine how it could be made to stay in the upper
end of an inverted cell of that size in such quantities as are put in,
as the bees often fill it near half full. Sometimes a cell of this kind
will contain this food, and no worm to feed upon it. I _guessed_ the
bees had compounded more than their present necessities required, and
that they stored it there to have it ready, also, that being there all
might know it was for royalty.

[Illustration: PLATE OF THE THREE KINDS OF CELLS.]

The taste is said to be "more pungent" than food given to the worker,
and the difference in food changes the bee from a worker to a queen. I
have nothing to say against this hypothesis; it may be so, or the young
bee being obliged to stand on its head may effect it, or both causes
combined may effect the change. I never tasted this food, or found any
test to apply.

The preceding plate represents a piece of comb containing all the
different cells--those at the left hand the size for drones. In the
centre are few that appear sealed over, others nearly covered, others
the larva in different stages of growth, as well as the eggs. _Fig. 1_
represents a queen's cell just commenced. They are usually started thus
far the first season, very frequently when the hive is only half or
two-thirds full. _Fig. 2_ is a cell sufficiently advanced to receive
the egg. _Fig. 3_ one finished, the stage when the first swarm leaves.
_Fig. 4_ when a queen has been perfected and left. _Fig. 5_ is a cell
where its occupant has been destroyed by a rival, and removed by the
workers. It will be perceived that each finished queen's cell contains
as much wax as fifty made for the workers.


LIABILITY OF BEING DESTROYED.

In any stage from the egg to maturity these royal insects are liable to
be destroyed;--if honey fails from any cause sufficient to make the
existence of a swarm any way hazardous, the preparations are abandoned,
and these young queens destroyed; (I would here request the reader not
to condemn me for telling more than I can prove, until he has had the
whole story; in the swarming season, I will give further particulars.)


DRONES DESTROYED WHEN HONEY IS SCARCE.

When an occurrence like the above happens, the drones next fall victims
to the failure of honey. A brief existence only is theirs; such as are
perfect, are destroyed without mercy; those in the chrysalis state are
often dragged out, and sacrificed to the necessities of the family.
Such as are allowed to hatch, instead of being fed and protected as
they would be if honey was abundant, are allowed, while yet weak from
the effects of hunger, to wander from the hive, and fall to the earth
by hundreds. These effects attend only a scarcity in the early part of
the season. The massacre of July and September is quite different. The
drones then have age and strength--an effort is apparently first made
by the workers to drive them out without proceeding to extremes; they
are harassed sometimes for several days; the workers feigning only to
sting, or else they cannot, as I never succeeded in seeing but very few
dispatched in that way; yet there is evidence proving beyond doubt that
the sting is used. Hundreds will often be collected together in a
compact body at the bottom of the hive; this mutual protection
affording a few hours' respite from their tormentors, who do not cease
to worry them. In a few days they are gone, and it is a hard matter to
tell what has become of them, at least the majority. If the hive in
September is well supplied with honey, a portion of the drones have a
longer lease of life given them; I have seen them as late as December.
In some seasons, when the best hives are poorly supplied with stores,
the ensuing spring the bees will rear no drones, until the flowers
yield a good supply. I have known one or two years in which no drones
appeared before the last of June; at other times, thousands are matured
by the first of May.


OLD QUEEN LEAVES WITH THE FIRST SWARM.

The old queen leaves with the first swarm; as soon as cells are ready
in the new hive she will deposit her eggs in them, at first for
workers; the number perfected will correspond with the supply of honey
and size of the swarm. When the supply fails before leaving the old
stock, she remains _there_, and continues laying throughout the season;
but the bees matured after the 20th of July (in this section) are not
more than sufficient to keep the number good. As many die, or are lost
during their excursions, as the young ones will replace; in fact, they
often lose rather than gain; so that by the next spring, a hive that
has cast no swarm, is no better for a stock than one from which a swarm
has issued. We are apt to be deceived by bees clustering outside,
towards the latter end of the season, and suppose it hardly possible
for them all to get in, when it may be caused by hot weather, full
stores, &c.


A YOUNG QUEEN TAKES THE PLACE OF HER MOTHER IN THE OLD STOCK.

In ordinary circumstances, when a swarm has left a stock, the oldest of
the young queens is ready to emerge from her cell in about eight or
nine days; if no second swarm is sent out, she will take her mother's
place, and begin to lay eggs in about ten days, or a little less. Two
or three weeks is the only time throughout the whole season, but what
eggs can be found in all prosperous hives. Whenever a copious yield of
honey occurs, drones are reared; as it becomes scarce, they are
destroyed.

The relative number of drones and workers that exist when they are most
numerous, doubtless depends on the size of the hive, whether one in
ten, or one in thirty.

When a swarm is first hived, the first cells are the size for working;
if the hive be very small, and bees numerous, it may be filled before
they are fully aware of it, and but few drone-cells constructed;
consequently, but few can be raised; whereas if the hive be large, long
before it is full, considerable honey will be stored. Cells for storing
honey are usually the size for drones; these will be made as soon as
the requisite number for workers is provided. An abundant yield of
honey during the process of filling a large hive, would therefore cause
a great proportion of these cells to be built--the amount of
drone-brood being governed by the same cause, is a strong argument
against large hives, as affording room for too many of these cells,
where an unnecessary number of drones will be reared, causing a useless
expenditure of honey, &c.


OTHER THEORIES.

Theories differing materially from the foregoing, are advanced by
nearly all writers. One says, "In spring the queen lays about 2,000
eggs of males, resumes it again in August, but during the rest of the
intervals she exclusively lays worker eggs. The queen must be at least
eleven months old before she begins to lay the eggs of males." Mr.
Townley makes the same assertion. Dr. Bevan says, "the great laying of
drone eggs usually commences about the end of April." Another author
repeats about the same, and appears to have investigated farther, as he
has found out that the eggs for the two kinds of bees are germinated
separately, and the queen knows when each kind is ready, as well as the
workers, &c. Now, I beg leave to differ a little from these authors.
Either there exists no difference in the eggs germinated, and any, or
all will produce drones or workers, just as they happen to be deposited
and fed; or else the periods of laying drone eggs are much more
frequent than any writer with which I am acquainted has been willing to
allow.


SUBJECT NOT UNDERSTOOD.

I am not anxious to establish a new theory, but to get at facts. If we
pretend to understand natural history, it is important that we have it
correct; and if we do not understand it, say so, and leave it open for
further investigation. It is my opinion that we _know_ but very little
about this point. I wish to induce closer observation, and would
recommend no _positive_ decision, until all the facts that will apply
have been examined. Whether these drone-egg theories have been too
hastily adopted, the reader can decide; I shall offer a few more facts,
somewhat difficult to reconcile with them.

First, in relation to the queen being "eleven months old" before laying
drone eggs. We _all_ agree, I believe, that the old queen goes with the
first swarm, and a young one remains in the old stock. Now suppose the
first swarm leaves in June, and the old stock yet contains a numerous
family. The flowers of buckwheat in August yield a bountiful harvest of
honey. This old stock rears a large brood of drones. Is it not proved
in this case that the queen was but two months old, instead of eleven?
We further agree that young queens accompany second or after-swarms.
When these happen to be large and prosperous, they never fail to rear a
brood of drones at this season. What is the age of these? I apprehend
that this eleven months theory originated in sections where there are
no crops of buckwheat raised, or in small quantities. Clover generally
fails in August, and May, or June, of another year comes round, before
there is a sufficient yield to produce the brood. With these
observations _only_, how very rational to conclude that it must be a
law of their nature, instead of being governed by the yield of honey,
and size of the family? If the periods of drone egg laying are limited
to only two or three, it would seem that all queens ought to be ready
with this kind of egg, about the same period of the season, but how are
the facts?

I would like to inquire what becomes of the first series of drone eggs,
the last of April, or the first of May, when the stocks are poorly
supplied with honey, or when a family is small and but little honey
through the summer? No drone brood is matured in these cases. It is not
pretended that the queen has any control over the germination of these
eggs, yet somehow she has them ready whenever the situation of the hive
will warrant it. Two stocks may have an equal number of bees the first
of May; one may have forty pounds of honey, the other four pounds; the
latter cannot afford to rear a drone, while the other will have
hundreds. Let two stocks have but four pounds each at any time in
summer when honey is scarce, now feed one of them plentifully, and a
brood of drones is sure to appear, while the other will not produce
one. Whenever stocks are well stored with honey, and full of bees, the
first of May will find drone-cells containing brood. If the flowers
continue to yield a full supply, these cells may be examined every week
from that period till the first swarm leaves, and I will engage that
drone brood may be found in all stages from the egg to maturity; and
the worker brood the same. In twenty-four days after the first swarm
leaves, the last drone eggs left by the old queen will be just about
matured. When transferring bees from old to new hives, I generally do
it about twenty-one or twenty-two days after the first swarm, (this is
the time to avoid destroying the worker-brood; the particulars will be
given in another place.) I have transferred a great many, and _never
failed_ to find a few drones about ready to leave the combs. Whether
the swarm had left the last of May, or middle of July, there was no
difference, they were on hand.

A very early swarm in good seasons, will often fill the hive, and send
out an issue in from four to six weeks: the usual amount of drone-brood
may be found in these cases. The following circumstance would appear to
indicate that all the eggs are alike, and if they are laid in
drone-cells, the bees give the proper food and make drones; if in
worker-cells, workers, just as they make a queen from a worker-egg,
when put in a royal cell.

In a glass hive, one sheet of comb next the glass, and parallel with
it, was full size; about three-quarters of this sheet was worker-cells,
the remainder drone-cells. The family had been rather small, but now
had increased to a full swarm; a few drones had matured in the middle
of the hive. It was about the middle of June, 1850, when I discovered
the bees on this outside sheet, preparing it, as I thought, for brood,
by cutting off the cells to the proper length. They had been used for
storing honey, and were much too long, being about an inch and a half
deep. In a day or two after I saw a few eggs in both worker and
drone-cells; four or five days afterwards, on opening the door, I found
her "majesty" engaged in depositing eggs in the drone cells. Nearly
every one already contained an egg; most of these she examined, but did
not use them; six or eight, it appeared, were all that were unoccupied;
in each of these she immediately deposited an egg. She continued to
search for more empty cells, and in doing so, she got on the part of
the comb containing worker-cells, where she found a dozen or more
empty, in each of which, she laid one. The whole time perhaps thirty
minutes. Query? Was her series of drone eggs exhausted just at this
time? If so, it would appear that she was not aware of it, because she
examined several drone-cells after laying the last one there, before
leaving that part of the comb, and acted exactly as if she would have
used them had they not been pre-occupied. Did the worker-cells receive
some eggs that would have produced drones, but for the circumstance of
being deposited in worker-cells? I know we are told that an egg may be
transferred from a worker-cell to one for drones, or an egg taken from
a drone-cell and deposited in a worker-cell; that the exchange will
make no difference, the bee will be just what the first deposit would
have made it. How the knowledge for this assertion was obtained, we are
not informed, at least of the practical part. That an egg was ever
detached from the bottom of one cell safely and successfully deposited
in another, without breaking or injuring it in some manner, to make the
bees refuse it, permit me at present to doubt.


NECESSITY FOR FURTHER OBSERVATION.

Cannot some experiments, practicable to all, be instituted that will
throw more light on this subject? The old hypothesis of limiting
drone-egg laying to two or three periods, is evidently at fault.


TWO SIDES OF THE QUESTION.

If we suppose that the eggs are all alike, and the subsequent treatment
makes either workers, drones, or queens, and look to analogy for
support, we shall find much against, as well as for it. For instance,
we find in almost every department of animated nature, that the sex of
the germ of a future being is decided before being separated from the
parent, as the eggs of fowls, &c. Another fact, some queens (averaging
one in sixty or eighty) deposit eggs that produce only drones,[8]
whether in worker or drone-cells, proving that sex is decided in this
case beyond controversy. Hence it would appear reasonable, if sex was
decided by the ovaries of the queen, in one case, it would be in
another.

      [8] I have had several such. It made no difference whether the
      eggs were in the worker-cells or drone-cells, the brood was all
      drones. When in the worker-cells, (and the majority was there,)
      they required to be lengthened about one-third. In an occurrence
      of this kind, the colony of workers will rapidly diminish in
      number, until too few are left to protect the combs from the
      moth. It occurs most frequently in spring, but I once had a case
      the last of summer. The first indications are an unusual number
      of caps, or covers of cells, being under and about the hive; the
      workers, instead of increasing, grow less in number. When you
      fear this state of things, make a thorough examination, blow
      under the hive some tobacco smoke, as directed in pruning, invert
      the hive, part the combs till you can see the brood; if the
      worker-cells contain drones, they are readily perceived, as they
      project beyond the usual even surface, being very irregular, here
      and there a few, or perhaps but one sticking out. The  worker-brood,
      when in their own cells, form nearly an even surface; so of the
      drones. The only remedy that I have found is to destroy this
      queen, and substitute another, which can be obtained in the
      swarming season, or in the fall, better than at other times. To
      find the queen, paralyze with puff-ball, &c. For directions see
      fall management.

To allow the bees the power of making three kinds of bees from one kind
of eggs, which would be virtually constituting a third sex, an anomaly
not often found. The drones being males, and workers imperfect females
with generative organs undeveloped, renders the anomaly of the third
sex unnecessary. On the other side it might be said in reply: That if
food and treatment would create or produce organs of generation in the
female, by making an egg destined for a worker into a queen, (a fact
which all apiarians admit,) why not food and treatment make the drone?
Is the difficulty of developing _one_ kind of sexual organs greater
than another?

Respecting the anomaly of the eggs of some queens producing only
drones, the question might be asked, Is this more of an anomaly than
that of ordinary queens which are said to germinate eggs in distinct
series? It is all out of the usual line. Other animals or insects
usually produce the sexes promiscuously. As we are ignorant of causes
deciding sex in any case, we must acknowledge mystery to belong to both
sides of the question here. The stumbling-block of more than two sexes,
which seems so necessary to make plain, is no greater here than with
some species of ants, that have, as we are told, king, queen, soldier
and laborer. Four distinct and differently formed bodies, all belonging
to one nest, and descended from one mother. Whether there are four
distinct kinds of eggs producing them, or the power is given to the
workers to develop such as are wanted, from one kind, we cannot say. If
we make two kinds of eggs, it helps the matter but very little. There
is still an anomaly. There is but one perfect female in a nest to
germinate eggs, and the myriads produced (being over 80,000 in
twenty-four hours, according to some historians) shows that the
fecundity of our queen-bee is not a parallel case by any means. And yet
they are similar, by having their offspring provided for without an
effort of their own.

I shall leave this matter for the present, hoping that _something
conclusive_ may occur in the course of my experiments, or those of
others. At present I am inclined to think that the eggs are all alike,
but am not fully satisfied.

I am aware that this matter is of but little value or interest to many,
but myself and a few others have "Yankee inquisitiveness" pretty well
developed, and would like to _know_ how it _was_ managed.

As for workers proving occasionally fertile, I have but little to say.
After years of close observation directed to this point, I have been
unable to discover anything to establish this opinion. Neither have I
found the black bees described by some authors. It is true that in the
middle or latter part of summer a portion will be much darker than
others, and perhaps rather smaller, and some of them with their wings
somewhat worn, probably the result of continued labor, peculiar food,
or some incidental circumstance.

I have a few times found a humble-bee under the hive, that had entered,
and not finding his way out readily, was speedily shorn of his
beautiful "locks," and consequently his strength--that is, every
particle of hair, down, feathers, bristles, or whatever he had been
covered with, was completely removed by the bees, who had no regard for
his beautiful alternating stripes of yellow and brown; which left him
the very picture of darkness.



CHAPTER IV.

BEE PASTURAGE.


In some seasons the earth is covered with snow much later than others.
When this occurs, a greater number of warm days are necessary to melt
it, and start the flowers, than otherwise.


SUBSTITUTE FOR POLLEN.

During these warm days, while waiting for the flowers, the bees are
anxious to do something. It is then interesting to watch them, and see
what will be used as substitutes for pollen and honey. At such times, I
have seen hundreds engaged on a heap of sawdust, gathering the minute
particles into little pellets on their legs, seeming quite pleased with
the acquisition. Rotten wood, when crumbled into powder, and dry, is
also collected. Flour, when scattered near the hive, I have known to be
taken up in considerable quantities. Some apiarians have fed it to
their bees at this season, and consider it a great advantage; I have
not tested it sufficient to give an opinion. A substitute for honey is
sap from a few kinds of trees, yet it all amounts to but very little.
All these unnatural sources are abandoned when the flowers appear.


MANNER OF PACKING IT.

The particular manner of obtaining pollen has been witnessed by but
very few persons, as it is generally brushed from their bodies and
packed on their legs, while on the wing, thereby preventing a fair
chance to inspect operations. When collecting only pollen they alight
on the flowers, passing rapidly over the stamens, detaching a portion
of the dust, which lodges on most parts of them, to be brushed together
and packed into pellets when again on the wing. Thus they keep
alternately flying and alighting until a load is obtained, when they
immediately return to the hive; each bee bringing several loads in a
day. Honey, as it is collected, is deposited in the abdomen, and kept
out of sight till stored in the hive.


ALDER YIELDS THE FIRST.

The first material gathered from flowers is pollen. Candle-alder
(_Alnus Rubra_)[9] yields the first supply. The time of flowering
varies from the 10th of March to the 20th of April. The amount afforded
is also variable. Cold, freezing weather frequently destroys a great
portion of these flowers after they are out. These staminate flowers
are nearly perfected the season previous, and a few warm days in spring
will bring them out, even before any leaves appear. When the weather
continues fine, great quantities of farina are secured.

      [9] The botanical names are from Wood's Class-Book.

The time that bees commence their labors does not govern the time of
swarming by any means; this matter depends on the weather through April
and May. These remarks apply particularly to this section, Green
County, New York, in latitude about 42 degrees. In other places many
different trees, shrubs, and herbs, may be found yielding honey and
pollen that scarcely exist here, producing far different results.

Our swamps produce several varieties of willow, (salix,) that put out
their blossoms very irregularly. Some of these bushes are a month
earlier than others, and some of the buds on the same bush are a week
or two later than the rest. These also afford only pollen, but are much
more dependence than alder, as a turn of cold weather cannot at any
time destroy more than a small part. Next comes the aspen, (_Populus
Tremuloides_); of this we have more than is necessary for any purpose.
It is not a particular favorite with the bees, as but few,
comparatively, visit it. It is followed very soon by an abundance of
the red maple (_Acer Rubrum_), that suits them better, but this, like
the others, is often lost by freezing. The first honey obtained of any
account is from the golden willow (_Salix Vitellina_); it yields no
pollen, and is seldom injured by frost. Gooseberries, currants,
cherries, pear and peach trees, add a share of both honey and pollen.
Sugar maple (_Acer Saccharinum_) now throws out its ten thousand silken
tassels, beautiful as gold. Strawberries modestly open their petals in
invitation, but, like "obscure virtues," are often neglected for the
more conspicuous Dandelion, and the showy appearance and flagrant
blossoms of the apple-trees, which now open their stores, offering to
their acceptance a real harvest.


FRUIT FLOWERS IMPORTANT IN GOOD WEATHER.

In good weather, sometimes a gain of twenty lbs. is added to their
stores, during this period of apple-tree blossoms. But we are seldom
fortunate enough to have good weather all through this period, it being
rainy, cloudy, cool, or windy, which is very detrimental. Sometimes a
frost at this time destroys all, and the gain of our bees is reversed,
that is, they are lighter at the end than at the beginning of these
flowers. Yet this is the season that decides their prosperity for the
summer, whether they do _first rate_ or otherwise. If good weather now,
we expect our first swarms about the first of June; if not, no
subsequent yield of honey will make up for this deficiency. We now have
a time of several days, from ten to fourteen, in which but few flowers
exist. If our hives are poorly supplied when this scarcity occurs, it
will so disarrange their plans for swarming, that no preparations are
again made much before July, and sometimes not at all. In sections
where the wild cherry (_Cerasus Seratina_) abounds, the flowers of this
will appear and fill this time of scarcity, which this section annually
presents.


RED RASPBERRY A FAVORITE.

The red raspberry (_Rubus Strigosus_) next presents the stamens as the
most conspicuous part of the flower, soliciting the embrace of the bee,
by pouring out bounteous libations more prized by our industrious
insect than wine. For several weeks they are allowed to partake of this
exquisite beverage; it is secreted at all hours and in all kinds of
weather. When the morning is warm we often hear their cheerful humming
among the leaves and flowers of this shrub, ere the sun appears above
the horizon. The gentle shower, sufficient to induce man to seek a
shelter, is often unheeded by the bee when luxuriating among these
flowers; even white clover, important as it is in furnishing the
greatest part of their stores, at this season, would be neglected if
there was only a full supply of this. Clover begins to blossom with the
raspberry, and continues longer. We have an insufficient supply (in
this section) in most seasons. Red clover probably secretes as much
honey as the white, but the tube of the corolla being longer, the bee
appears to be unable to reach it. Yet I have seen a few at work even
here but it appeared like slow business. Sorrel, (_Rumex Acetosella_)
the pest of many farmers, is brought under contribution, and furnishes
the precious dust in any quantity. Morning is the only part of the day
appropriated to its collection.


CATNIP, MOTHER-WORT, AND HOARHOUND ARE SOUGHT AFTER.

Catnip, (_Nepeta Cataria_,) Mother-wort, (_Leonurus Cardiaca_,) and
Hoarhound, (_Marrubium Vulgare_,) about the middle of June, put forth
their flowers, rich in sweetness, and like the Raspberry, the bees
visit them at all hours and in nearly all kinds of weather. They last
from four to six weeks; the catnip I have known to last twelve in a few
instances, yielding honey during the whole time. Ox-eye daisy,
(_Leucanthemum Vulgare_,) that beautiful and splendid flower, in
pasture and meadow, and worth but little in either, also contains some
honey. The flower is compound, and each little floret contains
particles so minute, that the task of obtaining a load is very tedious.
It is only visited when the more copious honey-yielding flowers are
scarce. Snap-dragon,(_Linaria Vulgaris_,) with its nauseous and
sickening odor, troubling the farmer with its vile presence, is made to
bestow the only good thing about it, except its beauty, upon our
insect. The flower is large and tubular, and the bee to reach the honey
must enter it; to see the bee almost disappear within the folds of the
corolla, one would think that it was about being swallowed, when the
hideous mouth was gaping to receive it; but unharmed, soon it emerges
from the yellow prison, covered with dust; this is not brushed into
pellets on its legs, like the pollen from some other flowers, but a
part adheres to its back between the wings, which it is apparently
unable to remove, as it remains there sometimes for months, making a
cluster outside the hive, appear quite speckled. Bush honey-suckle
(_Diervilla Trifida_) is another particular favorite.


SINGULAR FATALITY ATTENDANT ON SILKWEED.

Silkweed (_Asclepias Cornuti_) is also another honey-yielding
perennial, but a singular fatality attends many bees while gathering
it, that I never yet saw noticed. I had observed during the period this
plant was in bloom, that a number of the bees belonging to swarms,
before the hive was full, were unable to ascend the sides to the comb;
there would be sometimes thirty or more at the bottom in the morning.
On searching for the cause, I found from one to ten thin yellow scales,
attached to their feet, triangular, or somewhat wedge shape, in size
about the twentieth part of an inch. On the longest point or angle, was
a black thread-like point, from a sixteenth to an eighth of an inch in
length; on this stem was either hooks, barbs, or a glutinous matter,
that firmly adhered to each foot or claw of the bee, rendering it
useless as far as climbing the sides of the hive was concerned. I found
also among bees clustered outside of full hives, this ornament
attached, but to them it appeared no inconvenience. Among the scales of
wax and waste matter that accumulates about the swarms to the amount of
a handful, I found a great many of these scales, which the bees had
worked from their feet. The question then arose, were these scales a
foreign substance, accidentally entangled in their claws, or was it
something formed there by nature, or _rather_ an unnatural appendage?
It was soon decided. From the number of bees carrying it, I was
satisfied that if it was the product of any flower, it belonged to a
species somewhat abundant. I set about a close examination of all such
as were then in bloom. I found the flowers of the Silkweed, (or
Milkweed, as some call it,) sometimes holding a dead bee by the foot,
secured by this appendage. Both sepals and petals of this flower are
re-curved, that is, turned backward towards the stem, forming five
acute angles, or notches, just the thing for a trap for a bee with
_strings_ of _beads_ on its toes; when at work they are very liable to
slip a foot into one of these notches; the flower being thick and firm,
holds it fast; pulling only draws it deeper into the wedge-like cavity.
The bee must either perish or break loose; their instincts fail them in
this emergency; they know nothing about getting it out by a gentle pull
the other way. I never saw one do it except by accident. By examining
the buds of this plant just before opening, I found this fatal
appendage, by which great numbers of our bees are lost.[10] When I
point out a loss among our bees, I would like to give a remedy; but
here I am at a loss, unless all these plants are destroyed, and this is
impracticable in many places. After all I am not sure but honey enough
is obtained by such bees as do escape, to counterbalance what we lose.
This would depend on the amount of honey yielded by other flowers at
the same time.

      [10] In Wood's Class-book of Botany, "Order CII.," in a plate
      showing the parts of this plant, it is thus described: "Fig. 11,
      a pair of pollen masses suspended from the glands at an angle of
      the antheridium," &c.

      One, when reading this simple botanical description, and seeing
      the plate, or the Botanist with his glasses, when he minutely
      inspects the parts, would not suspect anything fatal to bees
      about it.

Whitewood (_Liriodendron Tulipifera_) yields something eagerly sought
for by the bees, but whether honey, or pollen, or both, I have never
been able to ascertain. All the flowers of this kind, with us, are too
high. It is very scarce, as well as Basswood, (_Tilia Americana_,)--that
in some places is abundant, and yields honey clear and transparent as
water, superior in appearance, but inferior in flavor to clover; it
also appears much thinner when first collected.


LARGE YIELD FROM BASSWOOD.

During the time this tree is in bloom, a period of two or three weeks
in many sections, astonishing quantities are obtained. A person once
assured me that he had known "ten pounds collected by one swarm in a
day, by weighing the hive in the morning and again at evening." I have
some doubt of the statement, and think half the amount would be a good
day's work; but I had but a small chance to know, as only a few trees,
as a specimen, grow in this section. I have weighed hives during
seasons of apple-tree blossoms and buckwheat, the two best yields of
honey we have, and three and a half pounds was the best for one day
that I ever had. Sumach, (_Rhus Glabra_,) in some sections, affords
considerable honey. Mustard (_Sinapis Nigra_) is also a great favorite.

I have now mentioned most of the honey-producing trees and plants that
come on before the middle of July. The course of these flowers is
termed the first yield. In sections where there are no crops of
buckwheat, it constitutes the only full one. Other flowers continue to
bloom till cold weather. Where white clover is abundant and the fields
are used for pasture, it will continue to throw out fresh flowers,
sometimes, throughout the summer; yet the bees consume about all they
collect in rearing their brood, &c. Thus it appears in some sections
six or eight weeks is about all the time they have to provide for
winter.


GARDEN FLOWERS UNIMPORTANT.

In passing along I have not mentioned garden flowers, because the
amount obtained here is a small item, compared to the forest and
fields--especially ornamental flowers. It is true that the Hollyhock,
(_Altha Rosea_,) Mallows, (_Malva Rotundifolia_) and many others yield
honey, but what does it amount to? A person expecting his hives to be
filled from such a source would very likely be disappointed, especially
when many are kept together.


HONEY-DEW.

Honey-dew is said to be a source from whence large collections are made
in some places. When or where it appears or disappears is more than I
can tell. I have seen the accounts of it, but accounts I have learned
to doubt until I find something corroborative in my own experience. I
find too many errors copied merely because they happen to be in company
with several truths. Huber discovered many important truths, and has
given them to the world; too many writers take it for granted when two
points of his are true, the third _must be also_. It is no proof that
there is no such article merely because I never discovered it. In the
many fruitless endeavors that I have made to get a view of this
substance, it may be I have lacked close observation; or possibly there
is none showered upon this region; or I may have failed to bring my
imagination to assist me to convert common dew into the real article.


SINGULAR SECRETION.

I once discovered bees collecting a secretion unconnected with flowers;
but was not honey-dew, as it has been described. I was passing a bush
of Witch-hazel, (_Hamamelis Virginiana_,) and was arrested by an
unusual humming of bees. At first I supposed that a swarm was about me,
yet it was late in the season, (it being about the 25th July.) On close
inspection, I found the bush contained numerous warty excrescences, the
size and shape of a hickory-nut. These proved to be only a shell--the
inside lined with thousands of minute insects, a species of aphis.
These appeared to be engaged sucking the juices, and discharging a
clear, transparent fluid. Near the stem was an orifice about an eighth
of an inch in diameter, out of which this liquid would gradually exude.
So eager were the bees for this secretion, that several would crowd
around one orifice at a time, each endeavoring to thrust the other
away. This occurred several years ago, and I never have been able to
find anything like it since; neither have I learned whether it is
common in other sections.


SECRETIONS OF THE APHIS.

The liquid ejected by the aphis, (plant louse,) when feeding or sucking
the juices of tender leaves, and received by the ants that are always
in attendance, is something like it; but in this case the bees were in
attendance instead of ants.

This mode of elaborating honey, although not generally collected by
bees, perhaps may not be too much out of place here. Also, it may
furnish a clue to the cause or substantiate some theory of honey-dew.

These insects (_Aphis_) have been very appropriately termed "ants'
cows," as they are regarded by them with the most tender care and
solicitude. In July or August, when the majority of the leaves of our
apple trees are matured, there is often a few sprouts or suckers about
the bottom or trunk, that continue growing and putting out fresh
leaves. On the under side of these, you will find the _aphis_ by
hundreds, of all sizes, from those just hatched to the perfect insect
with wings. All appear to be engaged in sucking the bitter juice from
the tender leaf and stalk. The ants are among them by scores. (They are
often accused by the careless observer of the injury, instead of the
_aphis_.) Occasionally there will issue from their abdomen a small,
transparent globule, which the ant is ever ready to receive. When a
load is obtained it descends to the nest; others may be seen going and
returning continually. Many other kinds of trees, shrubs and plants are
used by the ants as "cow pasture," and most kinds of ants are engaged
in this dairy business.[11] Would the bees attend on the _aphis_ for
this secretion, (for it appears to be honey,) if the ant was not there
first? Or if there were no ants or bees, would this secretion be
discharged, and falling on the leaves below them, be honey-dew? If they
were situated on some lofty trees, and it lodged on the leaves of small
bushes near the earth, it would, with some authors.

      [11] The history of insects, as published by Harpers, gives more
      particulars on this interesting subject.

These questions I shall not answer, at present. As for theory, I shall
probably have enough before I get through, where I hope the subject may
be more interesting.[12]

      [12] Since the foregoing was written, I have made some further
      observations on this subject. In August, 1852, I noticed, on
      passing under some willow trees, (_Salix Vitellina_,) that
      leaves, grass, and stones, were covered with a wet or shining
      substance. On looking among the branches, I found nearly all the
      smallest were covered with a species of large black _aphis_,
      apparently engaged in sucking the juices, and occasionally
      discharging a minute drop of a transparent liquid. I _guessed_
      this might be the honey-dew. As this was early in the morning, I
      resolved to visit this place again, as soon as the sun got up far
      enough to start out the bees, and see if they collected any of
      it. On my return I found not only bees in hundreds, but ants,
      hornets, and wasps. Some were on the branches with the _aphis_,
      others on the leaves and larger branches. Some of them were even
      on the stones and grass under the trees, collecting it.

We will now return to the flowers, and see what few there are yet to
appear, after the middle of July. The button-ball bush (_Cephalanthus
Occidentalis_) is now much frequented for honey. Also, our vines,
melons, cucumbers, squashes, and pumpkins. The latter are visited only
in the morning, and honey is the only thing obtained; notwithstanding
the bee is covered with farina, it is not kneaded into pellets on its
legs. I have seen it stated that bees never get honey early in the
morning, but pollen instead. Now it is not best always to take our
word, who pretend to know all about it, but look for yourselves into
some of these matters. Take a look some warm morning, when the pumpkins
are in bloom, and see whether it is honey or pollen they are in quest
of. Also please make an observation when they are at work on the red
raspberry, motherwort, or catnip; you will thus ascertain a fact so
easily, that you will wonder any one with the least pretension to
apiarian science could be ignorant of it. I mention this, not because
it is of much importance in itself, but to show the fallibility of us
all, as we sometimes copy the mistaken assertions of others.


ADVANTAGES OF BUCKWHEAT.

Under some circumstances, clover will continue to bloom through this
part of the season; also, a few other flowers; but I find by weighing,
a loss from one to six pounds, between the 20th July and the 10th of
August, when the flowers of buckwheat begin to yield honey, which
generally proves a second harvest. In many places it is their main
dependence for surplus honey. It is considered by many an inferior
quality. The color, when separated from comb, resembles molasses of
medium shade. The taste is more pungent than clover honey; it is
particularly prized on that account by some, and disliked by others for
the same reason. In the same temperature it is a little thicker than
other honey, and is sooner candied.


AMOUNT OF HONEY COLLECTED FROM IT.

Swarms issuing as late as the 15th July, when they commence on
buckwheat, sometimes contain not over five pounds of stores, and yet
make good stocks for winter, whereas, without this yield, they might
not live through October. It fails about once in ten years. I have
known a swarm to gain in one week sixteen pounds, and construct comb to
store it at the same time. At another time I had a swarm issue the 18th
August, that obtained thirty pounds in about eighteen days. But such
buckwheat swarms, in ordinary seasons, seldom get over fifteen pounds.
The flowers last from three to five weeks. The time of sowing the grain
varies in different sections, from the 10th of June to the 20th July.
Farmers wish to give it just time to ripen before frost, as the yield
of grain is considered better, but as the time of frost is a matter of
guess-work, some will sow several days earlier than others. Whenever an
abundant crop of this grain is realized, a proportionate quantity of
honey is obtained.


DO BEES INJURE THE CROP?

Many people contend that bees are an injury to this crop, by taking
away the substance that would be formed into grain. The best reasons
for this opinion that I have obtained are these: "I believe it, and
have thought so a long time." "It is reasonable if a portion of this
plant is taken away by the bees, there must be a less quantity of
material left for the formation of seed, &c." Most of us have learned
that a person's opinion is not the strongest kind of proof, unless he
can exhibit substantial reasons for it. Are the above reasons
satisfactory? How are the facts? The flowers expand, and a set of
vessels pour into the cup or nectary a minute portion of honey. I am
not aware that any one contends that the plant has another set of
vessels prepared to again absorb this honey and convert it into grain.
But strong testimony proves very plainly that it never again enters the
stalk or flower, but evaporates like water. We all know that animal
matter when putrid will be dissolved into particles small enough to
float in the atmosphere, too minute for the naked eye. When passing off
in this way this real flesh and blood would escape notice perhaps
altogether, and never be detected, were it not for the olfactories,
which on some occasions notify us of its presence very forcibly. In
passing a field of buckwheat in bloom, by the same means we are assured
of the presence of honey in the air. Now what is the difference whether
this honey passes off in the air, or is collected by the bees? If any
difference, the advantage appears to be in favor of the bees getting
it, for the reason that it thus answers another important end in the
economy of nature, consistent with her provisions in ten thousand
different ways of adapting means to ends. Most breeders of domestic
animals are aware of the deteriorating qualities induced by in-and-in
breeding; a change of breed is found necessary for perfection, &c.


ARE NOT BEES AN ADVANTAGE TO VEGETATION?

Vegetable physiology seems to indicate a similar necessity in that
department. The stamens and pistils of flowers answer the different
organs of the two sexes in animals. The pistil is connected with the
ovaries, the stamens furnish the pollen that must come in contact with
the pistil; in other words, it _must be impregnated_ by this dust from
the stamens, or no fruit will be produced. Now if it be necessary to
change the breed, or essential that the pollen produced by the stamens
of one flower shall fertilize the pistil of another, to prevent
barrenness, what should we contrive better than the arrangement already
made by Him who knew the necessity and planned it accordingly? And it
works so admirably, that we can hardly avoid the conclusion _that bees
were intended for this important purpose_! It is thus planned! Their
wants and their food shall consist of honey and pollen; each flower
secretes but little, just enough to attract the bee; nothing like a
full load is obtained from one; were it thus, the end in view would not
be answered; but a hundred or more flowers are often visited in one
excursion; the pollen obtained from the first may fertilize many,
previous to the bees' returning to the hive; thus a field of buckwheat
may be kept in health and vigor in its future productions. A field of
wheat produces long slender stalks that yield to the influence of the
breeze, and one ear is made to bestow its pollen on a neighboring ear
several feet distant, thereby effecting just what bees do for
buckwheat. Corn, from its manner of growth, the upright stalk bearing
the stamens some feet above the pistils, on the ears below, seems to
need no agency of bees; the superabundant pollen from the tassel is
wafted by the winds rods from the producing stalk, and there does its
office of fertilizing a distant ear, as is proved by different
varieties mixing at some distance. But how is it with our vines
trailing on the earth, a part of these flowers producing stamens, the
other only pistils? Now it _is absolutely essential_ that pollen from
the staminate flowers shall be introduced into the pistillate to
produce fruit; because if a failure occurs in this matter the germ will
wither and die. Here we have the agent ready for our purpose; these
flowers are visited by the bee promiscuously; no pollen (as was said)
is kneaded into pellets, (particularly that from pumpkins,) but it
adheres to every part of their body, rendering it next to impossible
for a bee thus covered with dust to enter the pistillated flower
without fulfilling the important duty designed, and leave a portion of
the fertilizing dust in its proper place. Hence it is reasonably
inferred by many, that if it was not for this agent among our vines,
the uncertainty of a crop from non-fertilization would render the
cultivation of them a useless task.

When the aphis is located on the stalk or leaf of a plant it is
furnished with means to pierce the surface and extract the juices
essential to the formation of the plant, thereby preventing vigorous
growth and a full development. This idea is too apt to be associated
with the bee when she visits the flower, as if she was armed with a
spear, to pierce bark or stem and rob it of its nourishment. Her real
structure is lost sight of, or perhaps never known; her slender
brush-like tongue folded closely under her neck, and seldom seen except
when in use, is not fitted to pierce the most delicate substance; all
that it can be used for is to sweep or lick up the nectar as it exudes
from the pores of the flower, secreted, it would seem, for no other
purpose but to attract her--while there she obtains nothing but what
nature has provided for her and given her the means of obtaining, and
the most delicate petal receives no injury.

During an excursion the bee seldom visits more than a single species of
flower; were it otherwise, and all kinds of flowers were visited
promiscuously, by fertilizing one species with the pollen from another,
the vegetable kingdom would be very likely to get into confusion.
Writers, when noticing the peculiarity of instinct governing the bee
here, cannot be content always, but must add other marvels. They follow
this trait into the hive, and make her store every kind by itself
there. Relative to honey it is not an easy matter to be positive; but
pollen is of a variety of colors, generally yellow, yet sometimes
pale-green, and reddish or dark-brown. Now I think a little patient
inspection would have satisfied any one that two kinds _are_ sometimes
packed in one cell, and prevented the assertion to the contrary. I will
admit that two colors are seldom found packed together, but sometimes
will be. I have thus found it, and it has entirely ruined that theory
for me.


A TEST FOR THE PRESENCE OF QUEEN DOUBTED.

It is further asserted that if a hive loses its queen "no pollen is
collected." Also, "that such quantities are sometimes collected, and
fill so many cells, that too little room is left for brood, and the
stock rapidly dwindles away in consequence." The first of these
assertions has been given as a test to decide whether the hive contains
a queen or not. Now my bees have such a habit of doing things wrong
that the above is no test whatever. It is made to appear very well in
theory, but wants the truth in practice. I will say what I have known
on this point, and perhaps clear up the difficulty of a stock
containing an unusual quantity of bee-bread with the honey, and instead
of being the cause of its having but few bees, it is the effect. Stocks
and sometimes swarms lose their queen in the swarming season, (the
particulars will be given in another place,) when, instead of remaining
idle, the usual quantity of both _pollen and honey is collected_
(unless the family is very small). There being no larvæ to consume the
bread, the consequence is, more than half the breeding cells will
contain it; they will be packed about two-thirds full, and finished out
with honey. I have known a large family left under such circumstances,
and about all the cells in the hive would be occupied. Whereas, in a
stock containing a queen and rearing brood, _a portion of the combs
will be used for this purpose until the flowers fail_, and then such
comb will be found empty.


AN EXTRA QUANTITY OF POLLEN NOT ALWAYS DETRIMENTAL.

To test whether this extra quantity of bee-bread was so _very_
detrimental, I have introduced into such hive in the fall a family with
a queen and wintered them in it, and watched their prosperity another
year, and never found them less profitable on that account. I am so
well satisfied of this, that whenever I now have a hive in such a
situation, it is a rule to introduce a swarm.

It is calculated, I believe, generally, that when medium-sized hives
are full, about seven-eighths of the cells are made the proper diameter
for raising the workers, the remainder for drones, except a few for
queens. Here is one circumstance I do not remember to have seen
mentioned, and that is, bee-bread is generally packed exclusively in
the worker cells. I would say always; but I would do better to be
careful, especially as I find my bees doing things so differently from
some others. I might as well remark here, that when taking combs from a
hive filled with honey, if such pieces were selected as contained only
the large or drone cells, but little risk of bee-bread would occur; of
the other combs, the outside sheets and the corners of the others near
the top are the next best. The sheets of comb used principally for
raising workers, and the cells next those so used, for an inch or two
in width, are nearly all packed with pollen, and much of it will
remain, when the breeding season is past. Smaller portions are found in
the worker cells in nearly all parts of the hive; even the boxes will
sometimes contain a little.


MANNER OF PACKING STORES.

In a glass hive, the bees may be seen depositing their load of pollen;
the legs holding the pellets are thrust into the cell, (not their
heads), and a motion like rubbing them together is made for a half
minute, when they are withdrawn, and the two little loaves of bread may
be seen at the bottom. This bee appears to take no farther care about
them, but another will soon come along, and enter the cell head first,
and pack it close; this cell is filled about two-thirds of its length
in this way, and when sealed over a little honey is used to fill it
out.


PHILOSOPHY IN FILLING A CELL WITH HONEY.

To witness the operation of depositing honey, a glass hive or box is
requisite; the edges of the combs will be attached to the glass--when
honey is abundant, most of these cells next the glass will contain
some. Now is the time to see the operation, glass forming one side of
such as are in contact, &c. The bee may be seen to enter the cell till
it reaches the bottom; with its tongue, the first particle is
deposited, and brushed into the corners or angles, carefully excluding
all the air from behind it--as it is filled, that next the sides of the
cell is kept in advance of the centre. The bee does not put its tongue
in the centre and pour out its load there, but carefully brushes the
sides as it fills, excluding every particle of air, and keeps the
surface concave instead of convex. This is just as a philosopher would
say it should be. If it was filled at once and no care taken to attach
it to the sides, why, the external air would never keep it there, which
it does effectually when of ordinary length. When the cell is about
one-fourth of an inch deep they often commence filling it, and as it is
lengthened they add to it, keeping it within an eighth of an inch of
the end; it is never quite full till nearly sealed over, and often not
then. In cells of the worker size, the sealing seldom touches the
honey. But in the size for drones the case is different; the honey on
the end touches the sealing, about half the diameter on the lower side;
it is kept in the same shape while being filled; but being somewhat
larger, the atmospheric pressure is less effectual in keeping the honey
in its place; consequently, when they commence sealing these cells they
begin on the lower side and finish at the top.


LONG CELLS SOMETIMES TURNED UPWARD.

When storing honey in boxes, cells of this size are usually much
longer, in which case they are crooked, the ends turning upward,
sometimes half an inch or more; this, of course, will prevent the honey
from running, but if the box is taken off and turned over before such
cells are sealed, they are very sure to spill most of their contents.
The cells in the breeding apartment, of ordinary length, will hold the
honey well enough as long as horizontal; but turn the hive on its side,
and bring the open end downward, in hot weather, or break out a piece
and hold it in that position, the air will not sustain it in them, but
will, in the size suitable for workers.

When the hive is fully supplied with bees and honey, (unless destitute
of a queen,) I never examined one, winter or summer, but it had a
number of unsealed cells containing honey, as well as pollen; it is so
when they have stored fifty pounds in boxes, even when so crowded for
room as to store honey outside or under the bottom-board; ever having
some cells open for a ready supply.

Young swarms seem unwilling to construct combs faster than needed for
use; it would appear, at first thought, to be a lack of economy. When
no honey is to be obtained and nothing to do, then it would seem to be
a fine chance for getting ready for a yield; but this is not _their_
way of doing business; whether they cannot spare the honey already
collected to elaborate the wax, or whether they find it more difficult
to keep the worms from a large quantity of comb, I shall not decide. Of
this I am satisfied, that it is better arranged by their instincts,
than we could do it. Large swarms, when first located, if honey is
abundant, will extend their combs from top to bottom in a little more
than two weeks; but such hive is not yet full; some sheets of comb may
contain honey throughout their whole length, and not a cell be sealed
over; but, however, they generally find time to finish up within a few
inches of the lower end as they proceed. Whenever unfinished cells
contain honey, it will generally be removed soon after the flowers
fail, and used before that which is sealed; and the cells will remain
empty till another year.


IS A DRY OR WET SEASON BEST FOR HONEY?

The inquiry is often made, "What kind of season is best for bees, wet
or dry?" This point I have watched very closely, and have found that a
medium between the two extremes produces most honey. When farmers begin
to express fears of a drought, then is the time (if in the season of
flowers) that most honey is obtained; but if dry weather passes these
limits, the quantity is greatly diminished. Of the two extremes,
perhaps very wet is the worst.


HOW MANY STOCKS SHOULD BE KEPT.

"What number of stocks can there be kept in one place?" is another
question often asked. This is like Mr. A. asking farmer B. how many
cattle could be pastured in a lot of ten acres. Farmer B. would first
wish to know how much pasture said lot would produce, before he could
begin to answer; since one lot of that size might produce ten times as
much as the other. So with bees, one apiary of two hundred stocks might
find honey in abundance for all, and another of forty might almost
starve. Like the cattle, it depends on pasture.


THREE PRINCIPAL SOURCES OF HONEY.

There are three principal sources of honey, viz.:--clover, basswood,
and buckwheat. But clover is the only universal dependance; as that is
almost everywhere, to some extent, in the country. Buckwheat in some
places is the main source; in others, basswood, which is of brief
duration. Where all three are abundant, there is the true El Dorado of
the apiarian! With plenty of clover and buckwheat, it is nearly as
well. Even with clover alone, enormous quantities of honey are
obtained. I have said what was our dependence in this section. I will
further say that within a circle of three or four miles, there are kept
about three hundred stocks. I have had for several years, three
apiaries about two miles apart, averaging in spring a little more than
fifty in each. When a good season for clover occurs, as many more would
probably do equally well, but in some other seasons I have had too
many; on an average nearly right. When clover furnishes too little
honey for the number, buckwheat usually supplies more than is
collected. Of surplus honey, the proportion is about fifteen pounds of
buckwheat to one of clover. I have now been speaking of large apiaries.
There can hardly be a section of country found, that man can procure
his living, but what a few stocks would thrive, even if there were no
dependence on the sources just mentioned. There will be some
honey-yielding flowers in nearly all places. The evil of over-stocking
is of short duration, and will work its own cure speedily. Some
judgment is required here as well as in other matters.

Another question of some interest, is the distance that a bee will
travel in search of honey in flowers--it is evident that it will be
farther than they will go to plunder a stock. I have heard of their
being found seven miles from home. It was said they ascertained, by
sprinkling flour on them as they left the hive in the morning, and then
saw the same bees that distance away. When we consider the chances of
finding a bee even one mile from the hive thus marked, it appears like
a "poor look;" and then pollen the color of flour might deceive us. It
is difficult to prove that bees go even two miles. Let us say we guess
at it, for the present.



CHAPTER V.

WAX.


The careless, unreflecting observer, when seeing the bees enter the
hive with a pellet of pollen on each of their posterior legs, is very
apt to conclude that it must be material for comb, as it appears unlike
honey. So little regard is paid to the matter by many people, that they
are unable to imagine any other use for it. Others suppose that it will
change from that to honey, after being stored a time in the hive, and
wonder at the curious phenomenon; but when asked how long a time must
elapse before it takes place, they cannot tell exactly, but they "have
found cells where it began to change, as a portion near the outer end
of the cell had become honey, and no doubt the remainder would in
time." It has been remarked that cells were only filled about
two-thirds full of this, and finished with honey; now when any one
finds a cell filled to the brim with pollen, and no honey, such
reasoning will apply better. If this was the case, by examining at
different periods through the summer, we certainly should find some
cells before the change had commenced, instead of their always being in
just this stage of transition.


IS POLLEN CONVERTED INTO WAX?

As for pollen being converted into wax or comb, a simple question will
show its fallacy. Do not the bees belonging to a hive that is full of
combs, and no more wax for that purpose needed, bring home as much and
often more pollen than one half full? Any person who has watched two
such hives five minutes when busily engaged at work, can answer. It is
evident, then, that pollen is for something else besides wax.


HOW IS IT OBTAINED?

The inquiry is now made, "Where do they get it from, if not from
pollen?" I might with propriety answer, they don't get it at all.
"Stop, there, if you please; if you expect us to credit you, you must
not give us too much absurdity." Well, let me ask a question. Do
cattle when grazing actually obtain flesh, bone, &c., or only the
materials from which these parts are secreted? As to the production of
wax, I believe all close observers (that I have found) agree that it is
a secretion natural only to the bee. With the ox, fruit, grain, or
grass may be converted into tallow; with the bee, honey and syrup made
of sugar may be converted into wax. These are probably the only two
substances yet discovered from which they extract it. Some writers have
pretended that pollen is also used, but they have failed to prove that
the old bees consume it at any time; which they must in this case if it
is converted into wax. From experiments related by Huber, either of
these substances, mixed with a little water, is all sufficient for its
production. From experiments of my own, I am satisfied that he is
correct. The experiment is tried by shutting up a swarm when first
hived; feeding them with honey--a few of the bees will probably have
some pollen, though not enough to make a comb three inches square, yet
it is something--and to be certain, time must be given them to exhaust
it. In three or four days take out the bees and remove the combs;
inclose them again, and feed with honey as before. Repeat the process,
until satisfied that no pollen is needed in the composition of wax.
Huber removed the combs "five times," with the same result at every
trial. Whenever bees are _confined_ in hot weather, _air and water are
absolutely necessary_.

We will now describe the first appearance of wax, and how it is
produced. When a swarm of bees is about leaving the parent stock,
three-fourths or more of them will fill their sacks with honey. When
located in their new home, of course no cells exist to hold it; it must
remain in the stomach or sack for several hours. The consequence is,
that thin white scales of wax the sixteenth of an inch in diameter,
somewhat circular, are formed between the rings of the abdomen, under
side. With the claws of one of their hind legs one of these is detached
and conveyed to the mouth, and there pinched with their forceps or
teeth, until one edge is worked somewhat rough; it is then applied to
the comb being constructed, or to the roof of the hive. The first
rudiments of comb are often applied within the first half hour after
the swarm is hived. In the history of insects before noticed, is a
minute account of the first foundation of combs, somewhat amusing, if
not instructive.


HUBER'S ACCOUNT OF A COMMENCEMENT OF COMB.

Huber, it is said, "having provided a hive with honey and water, it was
resorted to in crowds by bees, who, having satisfied their appetite,
returned to the hive. They formed festoons, remained motionless for
twenty-four hours, and after a time scales of wax appeared. An adequate
supply of wax for the construction of a comb having been elaborated,
one of them disengaged itself from the centre of the group, and
clearing a space about an inch in diameter, at the top of the hive,
applied the pincers of one of its legs to its side, detached a scale of
wax, and immediately began to mince it with the tongue. During the
operation, this organ was made to assume every variety of shape;
sometimes it appeared like a trowel, then flattened like a spatula, and
at other times like a pencil, ending in a point. The scale, moistened
with a frothy liquid, became glutinous, and was drawn out like a
riband. This bee then attached all the wax it could concoct to the
vault of the hive, and went its way. A second now succeeded, and did
the like; a third followed, but owing to some blunder did not put the
wax in the same line with its predecessor; upon which another bee,
apparently sensible of the defect, removed the displaced wax, and
carrying it to the former heap, deposited it there, exactly in the
order and direction pointed out." Now I have some objections to make to
this account. First, in the usual course of swarming, it is unnecessary
to provide the honey and water, as they come laden with honey from the
parent stock. Next, to form festoons and remain motionless twenty-four
hours to concoct the wax, is not the way they generally manage affairs.
They either swallow the honey before leaving home long enough to have
the wax ready, or less time than twenty-four hours is needed to produce
it. I have frequently found lumps, half the size of a pin's head,
attached to the branch of a tree where they had clustered, when they
had not been there over twenty-five minutes. I have had occasion a few
times to change the swarm to another tenement, an hour or two after
being hived, and found places on the top nearly covered with wax. How
it was managed to see a bee quit the "group," is more than I can
comprehend; and then the tongue to be the only instrument used to mould
the scale of wax, is another difficulty; to witness the whole process
minutely in this stage of comb-making has never been my good fortune,
and I am sometimes inclined to doubt the success of others. I have had
glass hives, and put swarms in them, and always found the first
rudiments of comb so entirely covered with bees as to prevent my seeing
anything.


BEST TIME TO WITNESS COMB-MAKING.

The only time when I have witnessed the process with any degree of
satisfaction is when the combs approach the glass, and but few bees in
the way; then, by watching patiently a few minutes, some part of the
process may be seen.


MANNER OF WORKING WAX.

Transferring the swarms to different hives from one to forty-eight
hours after being hived, will show their progress. I have found that
wax is attached to the top of the hive at first promiscuously, that is,
without the least order, until some of the blocks or lumps are
sufficiently advanced for them to begin cells. The scales of wax are
welded on the edge quite thick, without regard to the shape of the
cell, then an excavation is made on one side for the bottom of a cell,
and two others on the opposite side; the division between them exactly
opposite the centre of the first. When this piece is an inch or two in
length, two other pieces at equal distances on each side are commenced.
If the swarm is large, and honey abundant, it is common for two pieces
of comb to be started at one time on different parts of the top; the
sheets in the two places are often at right angles, or any other way,
just as chance happens to give direction. The little lumps that are
placed at random at first are all removed as they advance.

While the combs are in progress, the edges are always kept much the
thickest, and the base of the cell is worked down to the proper
thickness with their teeth, and polished smooth as glass. The ends of
the cell also, as they lengthen them, will always be found much thicker
than any other part of it when finished.

When two combs approach each other in the middle of the hive at nearly
right angles, an edge of comb is left there; but when an obtuse angle,
the edges are generally joined, making a sheet of crooked comb. It is
evident where the two combs join, there must be some irregular cells
unfit for rearing brood.


CROOKED COMBS A DISADVANTAGE.

These few irregular cells have been considered a great disadvantage. It
is thought, or pretended, that there is a vast difference between the
prosperity of a stock with straight combs and one with crooked ones. To
avoid them, or cause the bees to make them all straight, has given rise
to much contrivance, as if a few such cells could effect much. Suppose
there were a dozen sheets of comb in a hive, and each one had a row or
more of such irregular cells from top to bottom, what proportion would
they hold to those that were perfect? Perhaps not one in a thousand.
Hence we infer that in a hive of the proper size, the difference in
amount of brood never could be perceived. This is the only difference
it can make, because such cells can be used for storing honey as well
as others. But sometimes there will be corners and spaces not wide
enough for two combs, and too wide for one of the proper thickness for
breeding. As bees use all their room economically, and generally at the
best advantage, a thick comb will be the result. It is said they never
use such thick combs for breeding. How are the facts? I have just such
a space in a glass hive; one comb two inches thick. How is it managed?
Towards fall this sheet is filled with honey; the cells outside are
lengthened until there is just room for a bee to pass between them and
the glass, when they are sealed over. In spring these long cells are
all cut down (except at the top and upper corners) to the proper length
for breeding, and used for this purpose. This has been done for five
years in succession.

I will grant that there is a little waste room in such spaces, for part
of the year. It amounts to but little, as it is only outside. They are
necessitated to make such combs, because the inside combs, if built in
a breeding apartment, however crooked one may be, the next one will
generally match it, the right distance from it. But when they are built
expressly for storing honey, in such as are made in boxes, the right
distance is not so well preserved; hence it is not recommended to
compel bees to use such storing apartment for breeding. But suppose we
should compel a swarm to labor under these disadvantages, I should not
apprehend such disastrous results, (providing they have a proper
proportion of worker cells,) as no swarms, or even no surplus honey, as
has been represented. Imagine a hive filled with combs that are all too
thick, and room wasted when cut down, to the amount of one-fourth of
all that is in the hive. Now here are combs enough left to mature
three-fourths as many bees as in an ordinary hive, where all are right.
We can now suppose a good swarm will bring home the same amount of
honey as though it belonged to other hives; only three-fourths as much
can be fed to the brood, and stored in the hive; and the result ought
to be, that we get a quarter more surplus honey in boxes. Even if we
get no swarm, I cannot see how our surplus honey can be less, as in
this case there would be more bees at all times than in a hive that had
been reduced by swarming.

Does experience substantiate the theory that stocks with crooked combs
are as profitable as when they are straight? When combs are built
expressly for breeding, I could never discover any difference. Any
person can easily test it by a little observation; not by taking a
solitary instance of only one hive, because some other cause might
produce the result. Take a half-dozen at least with straight combs, and
as many with them crooked; have them all alike in other respects, and
carefully watch the result. I think you will have but little interest
which way the combs are made, providing _they are made_, as far as
profit is concerned. It is true, it would gratify order to have them
all straight, and if it was not attended with more trouble than the
result would pay for, it would be well to have them so.

In ordinary circumstances, when a swarm is first hived, they set about
comb-making immediately; yet sometimes they will remain two days, and
not make a particle. I have known them to swarm out and cluster in the
usual way, and when rehived, commence at once. This seems to prove that
they can retain the wax, or prevent secreting it, till wanted. This
seldom occurs.


UNCERTAINTY IN WEIGHT OF BEES.

A large swarm will probably carry with them some five or six pounds of
honey from the parent stock. I only guess at this, because I am
uncertain what the bees weigh exactly. "I can tell you," some one
exclaims, "I saw some weighed,--so many weigh just eight ounces." Are
you sure there was nothing but bees weighed? Was there no honey,
bee-bread, fæces, or other substance, that might deceive you? "Can't
say; I never thought of that!" Now it is important, if we weigh bees to
know _their_ weight, to be sure we weigh nothing else. It is evident,
that if five thousand weigh three pounds, when nothing is in their
sacks, they would weigh, when filled with honey, several pounds more.
Hence, the fallacy of judging of the size of a swarm by weight, as one
swarm might issue with half the honey of another. Perhaps eight pounds,
for large swarms, might be an average for bees and honey. This honey,
whatever it amounts to, cannot be stored till combs are constructed to
hold it. This principle holds good till the hive is full. That is,
whenever they have more honey than the combs will hold, if there is
room in the hive, they construct more. But they seem to go no farther
than this in comb-making. However large the swarm may be, this
compulsion appears necessary to fill the hive. Drone-cells are seldom
made in the top of the hive, but a part are generally joined on the
worker-cells, a little distance from the top; others near the bottom.
There seems to be no rule about the number of such cells. Some hives
will contain twice the number of others. It may depend on the yield of
honey at the time; when very plenty, more drone-cells, &c. If the hive
be very large, no doubt an unprofitable number would be constructed.
Where the large and small cells join, there will be some cells of
irregular shape; some with four or five angles; the distance from one
angle to the other is also varied. Even where two combs of cells the
same size join, making a straight comb, they are not always perfect.


SOME WAX WASTED.

When constructing comb, they are constantly wasting wax, either
accidentally or voluntarily. The next morning after a swarm is located,
the scales may be found, and will continue to increase as long as they
are working it; the quantity often amounts to a handful or more. It is
the best test of comb-making that I can give. Clean off the board and
look the next morning, you will find the scales in proportion to their
progress. Some will be nearly round as at first; others more or less
worked up, and a part will be like fine saw-dust.

Huber and some others have divided the working bees into different
classes, denominating some wax-workers, others nurses, and pollen
gatherers, &c. It may be partially true, but how it was found out is
the mystery.

The angles in the cells used for brood, are gradually filled, and after
a time become round, both at the ends and sides.


WATER NECESSARY TO COMB-MAKING.

Whenever bees are engaged making comb, a supply of water is absolutely
necessary. Some think it requisite in rearing brood. It may be needed
for that, or it may be required for both purposes; but yet I have
doubts if a particle is given to the young bee, besides what the honey
contains. June, and first part of July, and most part of August (the
season of buckwheat,) are periods of extensive comb-making; they then
use most water; breeding is carried on from March till October, and as
extensively in May, perhaps more so, than in August, yet not a tenth
part of the water is used in May.

I have known stocks repeatedly to mature brood from the egg to the
perfect bee, when shut in a dark room for months, when it was
impossible to obtain a drop; also stocks that stand in the cold, (if
good,) will mature some brood whether the bees can leave the hive or
not. These facts prove that some are reared without water. As they get
sufficient honey to require more comb to store it, they will at the
same time have a brood; and it is easy to guess they need it for brood
as comb, without a little investigation. This much is certain, that
they use water at such times for some purpose, and when no pond, brook,
spring, or other source is within convenient distance, the apiarian
would find it economy to place some within their reach, as it would
save much valuable time, if they would otherwise have to go a great
distance, when they might be more profitably employed; it always
happens in a season of honey. It should be so situated that the bees
may obtain it without jeopardizing their lives;--a barrel or pail has
sides so steep that a great many will slip off and drown. A trough made
very shallow, with a good broad strip around the edge to afford an
alighting place, should be provided. The middle should contain a float,
or a handful of shavings spread in the water with a few small stones
laid on them to prevent their being blown away when the water is out,
is very convenient. A tin dish an inch or so in depth, will do very
well. The quantity needed may be ascertained by what is used--only give
them enough, and change it daily. I have no trouble of this kind, as
there is a stream of water within a few rods of the hives; but I have
an opportunity to witness something of the number engaged in carrying
it. Thousands may be seen (in June and August) filling their sacks,
while a continual stream is on the wing, going and returning.


REMARKS.

The exact and uniform size of their cells is perhaps as great a mystery
as anything pertaining to them; yet, we find the second wonder before
we are done with the first. In building comb, they have no square or
compass as a guide; no master mechanic takes the lead, measuring and
marking for the workmen; each individual among them is a finished
mechanic! No time is lost as an apprentice, no service given in return
for instruction! Each is accomplished from birth! All are alike; what
one begins, a dozen may help to finish! A specimen of their work shows
itself to be from the hands of master workmen, and may be taken as a
model of perfection! He, who arranged the universe, was their
instructor. Yes, a profound geometrician planned the first cell, and
knowing what would be their wants, implanted in the sensorium of the
first bee, all things pertaining to their welfare; the impress then
given, is yet retained unimpaired! They need no lectures on domestic
economy to tell them, by using the base of one set of cells on one side
of their combs, for the base of those on the opposite, will save both
labor and wax; no mathematician that a pyramidal base, just three
angles, with just such an inclination, will be the exact shape needed,
and consume much less wax than round or square--that the base of one
cell of three angles, would form a part of the base of three other
cells on the opposite side of the comb--that each of the six sides of
one cell forms one side of six others around it--that these angles and
these only would answer their ends.

"The bees appear," says Reaumer, "to have a problem to solve, which
would puzzle many a mathematician. A quantity of matter being given, it
is required to form out of it cells, which shall be equal, and similar,
and of a determinate size, but the largest possible with relation to
the quantity of matter employed, while they shall occupy the least
possible space!"

How little does the epicure heed, when feasting on the fruits of their
industry, that each morsel tasted must destroy the most perfect
specimens of workmanship! that in a moment he can demolish what it has
taken hours, yes days, perhaps weeks, of assiduous toil and labor, for
the bees to accomplish!



CHAPTER VI.

PROPOLIS.


WHAT USED FOR.

This substance is first used to solder up all the cracks, flaws, and
irregularities about the hive. A coat is then spread over the inside
throughout; when the hive is full, and many bees cluster outside, the
latter part of summer, a coat of it is also spread there. An additional
coat it seems is annually applied, as old hives will be coated with a
thickness proportionate to its age, providing it has been occupied with
a strong family. Huber has said it was also used to strengthen the
cells when first made, by mixing it with the wax. If it was their
practice at that time, the practice has been abandoned by our bees to a
great extent. I have made examinations when comb was first made, when
it contained eggs, and when it contained larvæ, and have never been
able to find anything other than pure wax composing it. After a young
bee has matured in a cell, the coating or cocoon that it leaves is of a
dark color, somewhat resembling it, and may have given rise to the
supposition. How the article is obtained, appears to be the mystery.
This is a subject about which apiarians have failed to agree. A few
contend that it is an elaborated substance; while others assert it to
be a resinous gum, exuding from certain trees, and collected by the
bees like pollen. It differs materially from wax, being more tenacious,
and when it gets a little age, much harder.


IS IT AN ELABORATE OR NATURAL SUBSTANCE?

No modern observer has ever been able to detect the bees in the act of
gathering it.


HUBER'S OPINION.

Huber tells us, that "near the outlet of one of his hives, he placed
some of the branches of the poplar, which exuded a transparent juice,
the color of garnet. Several workers were soon seen perched upon these
branches,--having detached some of this resinous gum, they formed it
into pellets, and deposited them in the baskets of their thighs; thus
loaded, they flew to the hive, where some of their fellow-laborers
instantly came to assist them in detaching this viscid substance from
their baskets." Some of our modern apiarians have doubted this account
of Huber's. Now, in the absence of anything positive on this subject, I
am inclined to adopt this theory; that it is a resin or gum produced by
trees. (I cannot say that I am exactly satisfied with the story of
bringing the "branches and laying them by the hive," &c.) That bees
gather it in its natural state, is in accordance with my own
observation.


FURTHER PROOF.

Our first swarms that issue in May, or first of June, seldom use much
of the article pure for soldering and plastering; but instead, a
composition, the most of which is wax. I have noticed at this season,
when old pieces of boards that had been used for hives, were left in
the sun, that this old propolis would become soft in the middle of the
day. Here I have frequently seen the bees at work, packing it upon
their legs; it was detached in small particles, and the process of
packing was seen distinctly, as the bee did not fly during the
operation, as in the case of packing pollen. It is asserted that when
bees need it they always have it, indicating that they can elaborate it
like wax. I can see no reason why they do not need it in June as much
as August; yet, in the latter month, they use more than a hundred times
the quantity. At this time, they manifest no disposition to gather any
from the old boards, &c. It would seem they prefer the article new,
which they now have in abundance. Boxes filled in June contain but very
little, sometimes none. Why not, if they have enough of it? but when
filled in August, they always have the corners, and sometimes the top
and sides, lined with a good coat. Cracks, large enough for bees to
pass through, are sometimes completely filled with it. In this season,
a little before sunset of some fair day, I have frequently seen the
bees enter the hive with what I supposed to be the pure article on
their legs, like pollen, except the surface, which would be smooth and
glossy; the color much lighter than when it gets age. I have also seen
them through the glass inside, when they seemed unable to dislodge it
themselves, like pollen, and were continually running around among
those engaged in soldering and plastering; when one required a little,
it seized hold of the pellet with its teeth or forceps, and detached a
portion. The whole lump will not cleave off at once; but firmly adheres
to the leg; from its tenacity, perhaps a string an inch long will be
formed in separating, the piece obtained is immediately applied to
their work, and the bee is ready to supply another with a portion; it
doubtless gets rid of its load in this way; it is difficult to watch it
till it is freed from the whole, as it is soon lost among its fellows.
Now if this substance is not found in its natural state, how does it
happen that they pack it on their legs just as they do when getting it
from a board of an old hive, or pollen, when collected? They never take
the trouble to pack the wax there, when elaborated. Do not these
circumstances strongly favor the idea of its being a vegetable
substance? Perhaps the reason of its being collected at this season in
greater abundance, may be found in the fact, that the buds of trees and
shrubs are now generally formed. Many kinds are protected from rain and
frost, by a kind of gum or resinous coating. It may be found in many
species of Populus, particularly the balsam poplar, (_Populus
Balsamifera_) and the Balm of Gilead, (_Populus Candicans_). By boiling
the buds of these trees, an aromatic resin or gum may be obtained,
(used sometimes for making salve;) the odor is very similar to that
emitted by propolis, when first gathered by the bees, or by heating it
afterwards. In the absence of facts, we are apt to substitute theory.
This appears to me to be very plausible. Yet I am ready to yield it as
soon as facts decide differently. Perhaps not one bee in a thousand is
engaged in collecting this substance--there being so few may be one
reason why they are not often detected, yet few as they are, a few of
us should set about close observation; something certain might decide.
Apiarian science is sadly neglected; a large amount of error is mixed
up with truth, that patient, scrutinizing investigation must separate.


REMARKS.

I feel anxious to get to the practical part of this work, which I hope
will interest some readers who care but little about the natural
history. I shall begin with spring, and will now endeavor to mix more
of the practical with it, as we proceed to the end of the year. In
order to illustrate some points of practice, I may have occasion to
repeat some things already mentioned.



CHAPTER VII.

THE APIARY.


ITS LOCATION.

In the location of the apiary, one important consideration is, that it
is convenient to watch in the swarming season; that the bees may be
seen at any time from a door or window, when a swarm rises, without the
trouble of taking many steps to accomplish it; because if much trouble
is to be taken, it is too often neglected. Also, if possible, the hives
should stand where the wind will have but little effect, especially
from the northwest. If no hills or building offer a protection, a
close, high board fence should be put up for the purpose. It is economy
to do it--bees enough may be saved to pay the expense. During the first
spring months, the stocks contain fewer bees than at any other season.
It is then that a numerous family is important, for the purpose of
creating animal heat to rear the brood, if for nothing else. One bee is
of more consequence now than a half dozen in midsummer. When the hive
stands in a bleak place, the bees returning with heavy loads, in a high
wind, are frequently unable to strike the hive, and are blown to the
ground; become chilled, and die. A chilly south wind is equally fatal,
but not so frequent. When protected from winds, the hives may front any
point you choose; east or south is generally preferred. A location near
ponds, lakes, large rivers, &c., will be attended with some loss. Hard
winds will fatigue the bees when on the wing, often causing them to
alight in the water; where it is impossible to rise again until wafted
ashore, and then, unless in very warm weather, they are so chilled as
to be past the effort. I do not mention this to discourage any one from
keeping them, when so situated, because some few must keep them thus or
not at all. I am so situated myself. There is a pond of four acres,
some twelve rods off. In spring, during high winds, a great many may be
found drowned, and driven on shore. Although we cannot miss so few from
a stock, it is nevertheless a loss as far as it goes.


DECIDE EARLY.

Whatever location is chosen, it should be decided upon as early in the
spring as possible; because, when the chilling winds of winter have
ceased for a day, and the sun, unobstructed, is sending his first warm
rays to a frozen earth, the bees that have been inactive for months,
feel the cheering influence, and come forth to enjoy the balmy air. As
they come from their door, they pause a moment to rub their eyes, which
have long been obscured in darkness.


BEES MARK THEIR LOCATION ON LEAVING THE HIVE.

They rise on the wing, but do not leave in a direct line, but
immediately turn their heads towards the entrance of their tenement,
describing a circle of only a few inches at first, but enlarge as they
recede, until an area of several rods have been _viewed and marked_.


CHANGING STAND ATTENDED WITH LOSS.

After a few excursions, when surrounding objects have become familiar,
this precaution is not taken, and they leave in a direct line for their
destination, and return by their way-marks without difficulty. Man with
his reason is guided on the same principles. There are a great many
people who suppose the bee knows its hive by a kind of instinct, or is
attracted towards it, like the steel to the magnet. At least, they act
as if they did; as they often move their bees a few rods, or feet,
after the location is thus marked, and what is the consequence? The
stocks are materially injured by loss of bees, and sometimes entirely
ruined. Let us trace the cause. As I remarked, the bees have marked the
location. They leave the hive without any precaution, as surrounding
objects are familiar. They return to their old stand and find no home.
If there is more than one stock, and the removal has been from four to
twenty feet, some of the bees may find a hive, but just as liable to
enter the wrong one as the right. Probably they would not go over
twenty feet, and very likely not that, unless the new situation was
very conspicuous. If a person had but one stock, very likely the loss
would be less, as every bee finding a hive, would be sure to be home,
and none killed, as is generally the case when a few enter a strange
hive.


CAN BE TAKEN SOME DISTANCE.

When bees are taken beyond their knowledge of country, some two miles
or more, the case seems to be somewhat different, but not always
without loss, especially if many hives are set too close. They leave
the hive of course without knowing that the situation has been changed;
perhaps get a few feet before strange objects warn them of the fact.
When they return, the immediate vicinity is strange, and they often
enter their neighbors' domicil.


DANGER OF SETTING STOCKS TOO CLOSE.

A case in point occurred in the spring of '49. I sold over twenty
stocks to one person. He had constructed a bee-house, and his
arrangement brought the hives within four inches of each other. The
result was, he entirely lost several stocks; some of them were the
best; others were materially injured, yet he had a few made better by
the addition of bees from other hives; (sometimes a stock will allow
strange bees to unite with them, but it is seldom, unless a large
number enters--it is safest to keep each family by itself, under
ordinary circumstances). These stocks, before they were moved, had been
collecting pollen, and had their location well marked. Had they been
placed six feet apart, instead of four inches, he probably would not
have lost any, or even two feet might have saved them. I have often
moved them at this season, and placed them at three feet distance, and
had no bad results.

Facts like the foregoing, satisfied me long since that stocks should
occupy their situation for the summer, as early as possible in the
spring, at least before they mark the location; or if they must be
moved after that, let it be nothing short of a mile and a half, and
plenty of room between the hives.


SPACE BETWEEN HIVES.

As regards the distance between hives generally, I would say let it be
as great as convenience will allow. Want of room makes it necessary
sometimes to set them close; where such necessity exists, if the hives
were dissimilar in color, some dark, others light, alternately, it
would greatly assist the bees in knowing their own hive. But it should
be borne in mind, that whenever economy of space dictates less than two
feet, there are often bees enough lost by entering the wrong hive,
which, if saved, would pay the rent of a small addition to a garden, or
bee-yard. I have several other reasons to offer for giving plenty of
room between hives, which will be mentioned hereafter.


SMALL MATTERS.

The reader who is accustomed to doing things on gigantic principles,
will consider this long "yarn" about saving a few bees in spring, a
rather small affair, and so it is; yet small matters must be attended
to if we succeed; "a small leak will sink a ship." A grain of wheat is
a small matter; 'tis only in the aggregate that its importance is
manifest. The bee is small, the load of honey brought home by it is
still less, and the quantity secreted in the nectary of each flower,
yet _more minute_. The patient bee visits each, and obtains but a tiny
morsel; by perseverance a load is obtained, and deposited in the hive;
it is only by the accumulation of such loads that we find an object
worthy our notice: here is a lesson; look to little things, and the
manner in which they are multiplied, and preserved. It is much better
to save our bees than waste them, and wait for others to be raised; "a
penny saved is worth two-pence earned." If a stock is lost by small
means, a corresponding effort is only necessary to save it. This
trifling care is sometimes neglected through indolence. But I hope for
better things generally; I am willing to believe it is thorough
ignorance, not knowing what kind of care is necessary--how, when, and
where to bestow it. This is what now appears to be my duty to tell. You
will now sufficiently understand the cause of loss on this point;
therefore, let it be a rule to have all ready in spring, before the
bees leave their hives--the stands, bee-house, etc., and not change
them.


ECONOMY.

If we keep bees for ornament, it would be well to build a bee-house,
paint the hives, &c.; but as I expect the majority of readers will be
interested in the profit of the thing, I will say that the bees will
not pay a cent towards extra expenses; they will not do a whit more
labor in a painted house, than if it was thatched with straw. When
profit is the only object, economy would dictate that labor shall be
bestowed only where there will be a remuneration.


CHEAP ARRANGEMENT OF STANDS.

So many kinds of bee-houses and stands have been recommended--all so
different from what I prefer, that I perhaps ought to feel some
hesitancy in offering one so cheap and simple; but as profit is my
object, I shall offer no other apology. I have fifteen years'
experience to prove its efficacy, and have no fears on this score in
recommending it. I make stands in this way: a board about fifteen
inches wide is cut off two feet long; a piece of chestnut or other
wood, two inches square, is nailed on each end; this raises the board
just two inches from the earth, and will project in front of the hive
some ten inches, making it admirably convenient for the bees to alight
before entering the hive, (when the grass and weeds are kept down,
which is but little trouble). A separate piece for each hive is better
than to have several on a bench together, as there can then be no
communication by bees running to and fro. Also we are apt to give more
room between them; and a board or plank will make a stand for as many
stocks when cut in pieces, as if left whole; (and it ought to make
more).


CANAL BOTTOM-BOARD DISCARDED.

I used what is termed a canal bottom-board, until I found out it did
not pay expense, and have now discarded it, and succeed just as well.
It is generally recommended as a preventive of robberies, and keeping
out the moth. It may prevent one hive in fifty from being robbed; but
as for keeping out the moth, it is about as good an assistant for it as
can be contrived. It is a place of great convenience for the worms to
spin their cocoons, and some ingenuity of the apiarian is requisite to
get at them.


SOME ADVANTAGE IN BEING NEAR THE EARTH.

I am aware that I go counter to most apiarians, in recommending the
stands so near the earth; less than two or three feet between the bees
and the earth, it is said, will not answer any way. Mr. Miner is very
positive on this point, in his Manual. I ventured to suggest to him,
that there was more against it in theory, than in practice, and gave
him my experience. In less than two years from that time I visited him,
and found his bees close to the earth. Experience is worth a dozen
theories; in fact, it is the only test to be depended upon. I shall not
urge the adoption of any rule, that I have not proved by my own
practice. The objection raised, is dampness from the earth, when too
near; I am unable to perceive the least bad effect. Now let us compare
advantages and disadvantages a little farther. One hive or a row of
hives suspended, or standing on a bench, two or three feet from the
earth, when approached by the bees on a chilly afternoon, (and we have
many such in spring,) towards evening, even if there is not much wind,
they are very apt to miss the hive and bottom, and fall to the ground,
so benumbed with cold, as to be unable to rise again, and by the next
morning are "no use" whatever. On the other hand, if they are near the
earth, with a board as described, there is no _possibility_ of their
alighting under the hive, and if they should come short, and get on the
ground, they can always creep, long after they are too cold to fly, and
are able, and often do enter the hive without the necessity of using
their wings.

Enough may be saved in one spring, from a few hives, in this way, to
make a good swarm, which taken from several is not perceived; yet, as
much profit from them might be realized, as if they were a swarm by
themselves. A little contrivance is all that is needed to save them. To
such as _must_ and _will_ have them up away from the earth, I would
say, do suggest some plan to save this portion of your best and most
willing servants; have an alighting board project in front of the hive
at least one foot, or a board long enough to reach from the bottom of
the hive to the ground, that they may get on that, and crawl up to the
hive. Do you want the inducement? Examine minutely the earth about your
hives, towards sunset, some day in April, when the day has been fair,
with some wind, and chilly towards night, and you will be astonished at
the numbers that perish. Most of them will be loaded with pollen,
proving them martyrs to their own industry and your negligence. When I
see a bench three feet high and no wider than the bottom of the hive,
perhaps a little less, and no place for the bees to enter but at the
bottom, and as many hives crowded on as it will hold, I no longer
wonder that "bee-keeping is all in luck;" the wonder is how they keep
them at all. Yet it proves that, with proper management, it is not so
very precarious after all.

The necessary protection from the weather, for stocks, is a subject
that I have taken some pains to ascertain; the result has been, that
the cheapest covering is just as good as any; something to keep the
rain and rays of the sun from the top, is all sufficient. Covers for
each hive, like the bottom-board, should be separate, and some larger
than the top.


UTILITY OF BEE-HOUSES DOUBTED.

I have used bee-houses, but they will not pay, and are also discarded.
They are objectionable on account of preventing a free circulation of
air; also, it is difficult to construct them, so that the sun may
strike the hives both in the morning and afternoon; which in spring is
very essential. If they front the south, the middle of the day is the
only time when the sun can reach all the hives at once; this is just
when they need it least; and in hot weather, sometimes injurious by
melting the combs. But when the hives stand far enough apart, on my
plan, it is very easily arranged to have the sun strike the hive in the
morning and afternoon, and shaded from ten o'clock, till two or three,
in hot weather.

Notwithstanding our prodigality in building a splendid bee-house, we
think of economy when we come to put our hives in, and get them _too
close_. "Can't afford to build a house, and give them so much room, no
how."



CHAPTER VIII.

ROBBERIES.


Robbing is another source of occasional loss to the apiarian. It is
frequent in spring, and at any time in warm weather when honey is
scarce. It is very annoying, and sometimes gets neighbors in
contention, when perhaps neither is to blame, farther than ignorance of
the matter.


NOT PROPERLY UNDERSTOOD.

A person keeping many hives must expect to be accountable for all
losses in his neighborhood, whether they are lost by mismanagement or
want of management. Many people suppose, if one person has but one
stock, and another has ten, that the ten will combine for plundering
the one. There are no facts, showing any communication between
different families of the same apiary, that I can discover. It is true,
when one family finds another weak and defenceless, possessing
treasure, they have no conscientious scruples about carrying off the
last particle. The hurry and bustle attending it seldom escape the
notice of the other families; and when one hive has been robbed in an
apiary, perhaps two-thirds of the other families, sometimes all, have
participated in the plunder. One family, if it be large, is just as
likely, and more so, to find a weak one among the ten, and commence
plundering, as the other way.


IMPROPER REMEDIES.

Notwithstanding it is common to hear remarks like this, "I had a
_first-rate_ hive of bees," (when the fact was he had not looked
particularly at his bees for a month, to know whether it was so or not,
and if he had, very likely would not know,) "and Mr. A.'s bees began to
rob them. I tried every thing to stop it; I moved them around in
several places to prevent their finding the hive. It did no good; the
first I knew they were all gone--bees, honey, and all! The bees all
joined the robbers." Now the fact is, that not one _good_ stock of bees
in fifty, will ever be robbed, if let alone; that is, if the entrance
is properly protected. This moving the hive was enough to ruin any
stock; bees were lost at every change, until nothing was left but honey
to tempt the robbers; whereas, if left on its stand, it might have
escaped.

A great many remedies have been given me gratis, which, had one-half
been followed, would have ruined them. The fact is, with many people,
the remedies are often the cause of the disease. The most fatal is, to
move them a few rods; another, to close the hive entirely, (very liable
to smother them); or, break out some comb and set the honey to running.
There are some charms that affect them but little any way. Probably
there are but few bee-keepers able to tell at once, _when bees are
being robbed_. It requires the closest scrutinizing observation to
decide.


DIFFICULTY IN DECIDING.

There is nothing about the apiary more difficult to determine, nothing
more likely than to be deceived. It is generally supposed, when a
number are outside fighting, that it is conclusive that they are also
robbing, which is seldom the case. On the contrary, a show of
resistance indicates a strong colony, and that they are disposed to
defend their treasures. I no longer have any fears for a stock that has
courage to repel an attack.


WEAK FAMILIES IN MOST DANGER.

It is weak families, that show no resistance, where we find the most
danger. In seasons of scarcity, all _good_ stocks maintain or keep
sentinels about the entrance, whose duty it appears to be to examine
every bee that attempts to enter. If it is a member of the community,
it is allowed to pass; if not, it is examined on the spot. It would
seem that a password was requisite for admittance, for no sooner does a
stranger-bee endeavor to get in, than it is known. If without necessary
credentials, there is evidence enough against it. Each bee is a
qualified jurist, judge, and executioner. There is no delay; no waiting
for witnesses for defence. The more a bee attempts to escape, the more
likely it will be to receive a sting, unless it succeeds. How strange
bees are known, would be nothing but theory, if I should attempt to
explain. Let it suffice that they are known.


THEIR BATTLES.

I will here describe some of their battles. I have in the spring
frequently seen the whole front side of the hive covered with the
combatants, (but for such hives I have no fears; they are able to
defend themselves.) Several will surround one stranger; one or two will
bite its legs, another the wings; another will make a feint of
stinging, while another is ready to take what honey it has, when
worried sufficient to make it willing. It is sometimes allowed to go
after yielding all its honey, but at others, is dispatched with a
sting, which is almost instantly fatal. A bee is killed sooner by a
sting, than by any other means, except crushing. Sometimes a leg will
tremble, for a minute; the legs are drawn close to the body; the
abdomen contracts to half its usual size, unless filled with honey. I
have known a pint accidentally to enter a neighboring stock, and be
killed in five minutes. The only places the sting will penetrate a bee
are the joints of the abdomen, legs, the neck, &c. I have occasionally
seen one bee drag about the dead body of its victim, being unable to
withdraw its sting from a joint in the leg. During the fight, if it be
to keep off those in search of plunder, a few bees may be seen buzzing
around in search of a place unguarded to enter the hive. If such is
found, it alights and enters in a moment. At other times, when about to
enter, it meets a soldier on duty, and is on the wing again in an
instant. But another time it may be more unfortunate, and be nabbed by
a policeman, when it must either break away, or suffer the penalty of
insect justice, which is generally of the utmost severity.


BAD POLICY TO RAISE THE HIVES.

A great many apiarians raise their hives an inch from the board early
in spring. They seem to disregard the chance it gives robbers to enter
on every side. It is like setting the door of your own house open, to
tempt the thief, and then complain of depravity.

Let it be understood, then, that all good stocks, under ordinary
circumstances, will take care of themselves. Nature has provided means
of defence, with instinct to direct its use. Non-resistance may do for
highly cultivated intellect in man, but not here.


INDICATIONS OF ROBBERS.

We will now notice the appearance about a weak hive that makes no
resistance, and show the result to be a total loss of the stock,
without timely interference. Each robber, when leaving the hive,
instead of flying in a direct line to its home, will turn its head
towards the hive to mark the spot, that it may know where to return for
another load, in the same manner that they do when leaving their hive
in the spring. The first time the young bees leave home, they mark
their location, by the same process. A few of these begin to hatch from
the cells very early; in all good stocks, often before the weather is
warm enough for _any to leave the hive_. Consequently, it cannot be too
early for them at any time in spring. These young bees, about the
middle of each fair day, or a little later, take a turn of flying out
very thickly for a short time. The inexperienced observer would be very
likely to suppose such stock very prosperous, from the number of
inhabitants in motion. This unusual bustle is the first indication of
foul play, and should be regarded with suspicion; yet it is not
conclusive.


A DUTY.

It is the duty of every bee-keeper, who expects to succeed, to know
which his weak stocks are; an examination some cool morning, can be
made by turning the hive bottom up, and letting the sun among the
combs. The number of inhabitants in them is easily seen. When weak,
close the entrance, till there is just room for one bee to pass at
once. The first real pleasant days, at any time before honey is
obtained plentifully, a little after noon, look out for them to
commence robbing. Whenever a weak stock is taken with what appears to
be a fit of unusual industry, it is quite certain they are either
robbers or young bees; the difficulty is to decide which. Their motions
are alike, but there is a little difference in color--the young bees
are a shade lighter; the abdomen of the robbers, when filled with
honey, is a little larger. It requires close, patient observation, to
decide this point, and when you have watched close enough to detect
this difference, you can decide without trouble.


A TEST.

But while you are learning this nice distinction, your bees may be
ruined. We will, therefore, give some other means of protection.

Bees, when they have been stealing a sack of honey from a neighboring
hive, will generally run several inches from the entrance before
flying: kill some of these; if filled with honey, they are robbers;
because it is very suspicious, to be filled with honey when leaving the
hive; or sprinkle some flour on them as they come out, and have some
one watch by the others to see if they enter. Another way is less
trouble, but will take longer, before they are checked, if robbing.
Visit them again in the course of half an hour or more, after the young
bees have had time to get back, (if it should happen to be them); but
if the bustle continues or increases, it is time to interfere. When the
entrance has been contracted as directed, close it entirely till near
sunset. When it has been left without, it should now be done, (giving
room for only one bee at a time). This will allow all that belong to
the hive to get in, and others to get out, and materially retard the
progress of the robbers.


ROBBING USUALLY COMMENCES ON A WARM DAY.

Unless it should be cool, they will continue their operations till
evening. Very often some are unable to get home in the dark, and are
lost. This, by the way, is another good test of robbing. Visit the
hives every warm evening. They _commence_ depredations on the warmest
days; seldom otherwise. If any are at work when honest laborers should
be at home, they need attention.


REMEDIES.

As for remedies, I have tried several. The least trouble is to remove
the weak hive in the morning to the cellar, or some dark, cool place,
for a few days, until at least two or three warm days have passed, that
they may abandon the search. The robbers will then probably attack the
stock on the next stand. Contract the entrance of this in accordance
with the number of bees that are to pass. If strong, no danger need be
apprehended; they may fight, and even kill some; perhaps a little
chastisement is necessary, to a sense of their duty.


COMMON OPINION.

There is an opinion prevalent that robbers often go to a neighboring
stock, kill off the bees first, and then take possession of the
treasures. To corroborate this matter, I have never yet discovered one
fact, although I have watched very closely. Whenever bees have had all
their stores taken, at a period when nothing was to be had in the
flowers, it is evident they must starve, and last but a day or two
before they are gone. This would naturally give rise to the supposition
that they were either killed, or gone with the robbers.


A CASE IN POINT.

I have a case in point. Having been from home a couple of days, I
found, on my return, a swarm of medium strength, that had been
carelessly exposed, had been plundered of about fifteen pounds of
honey, every particle they had.[13] About the usual number of bees were
among the combs, to all appearance, very disconsolate. I at once
removed them to the cellar, and fed them for a few days. The other bees
gave over looking for more plunder, in the meantime. It was then
returned to the stand, entrance nearly closed, as directed, &c. In a
short time it made a valuable stock; but had I left it twenty-four
hours longer, it probably would not have been worth a straw.

      [13] It occurred the last of July.


FURTHER DIRECTIONS.

When a stock has been removed, if the next stand contains a weak,
instead of a strong one, it is best to take that in also; to be
returned to the stand as soon as the robbers will allow it. If a second
attack is made, put them in again, or if practicable, remove them a
mile or two out of their knowledge of country; they would then lose no
time from labor. Where but few stocks are kept, and not more than one
or two stocks are engaged, sprinkle a little flour on them as they
leave, to ascertain which the robbers are; then reverse the hives,
putting the weak one in the place of the strong, and the strong one in
the place of the weak one. The weak stock will generally become the
strongest, and put a stop to their operations; but this method is often
impracticable in a large apiary; because several stocks are usually
engaged, very soon after one commences, and a dozen may be robbing one.
Another method is, when you are _sure_ a stock is being robbed, take a
time when there are as many plunderers inside as you can get, and close
the hive at once, (wire-cloth, or something to admit air, and at the
same time confine the bees, is necessary;) carry in, as before
directed, for two or three days, when they may be set out. The strange
bees thus enclosed will join the weak family, and will be as eager to
defend what is now _their_ treasure, as they were before to carry it
off. This principle of forgetting home and uniting with others, after a
lapse of a few days, (writers say, twenty-four hours is sufficient for
them to forget home) can be recommended in this case. It succeeds about
four times in five, when a proper number is enclosed. Weak stocks are
strengthened in this way very easily; and the bees being taken from a
number of hives, are hardly missed. The difficulty is, to know when
there are enough to be about equal, to what belongs to the weak stock;
if too few are enclosed, they are surely destroyed.


COMMON CAUSE OF COMMENCING.

After all, bees being robbed is like being destroyed by worms; a kind
of secondary matter; that is, not one strong stock in a hundred will
ever be attacked and plundered on the first onset. Bees must be first
tempted, and rendered furious by a weak hive; a dish of refuse honey
set near them is sometimes sufficient to set them at work, also where
they have been fed and not had a full supply. After they have once
commenced, it takes an astonishing quantity to satiate their appetite.
They seem to be perfectly intoxicated, and regardless of danger; they
venture on to certain destruction! I have known a few instances where
good stocks by this means were reduced, until they in turn fell a prey
to others. I have for several years kept about one hundred stocks away
from home, where I could not see them much, to prevent robbing. Yet I
never lost a stock by this cause. I simply keep the entrance closed,
except a passage for the bees at work during spring. It is true I have
lost a few stocks, when the other bees took the honey, but they would
have been lost any way.


SPRING THE WORST TIME.

As I before remarked in the commencement of this chapter, bees will
plunder and fight at any time through the summer, when honey cannot be
collected; but _spring_ is the only time that such desperate and
persevering efforts are made to obtain it. It is the only time the
apiarian can be excused for having his hives plundered, or letting them
stand in a situation for it. We then often have families reduced in
winter and spring, from various causes, and when protected through this
season, generally make good stocks. It is then we wish them to form
steady, industrious habits, and not live by plunder. Prevention is
better than cure; evil propensities should be checked in the beginning.
The bee, like man, when this disposition has been indulged for a time,
it is hard breaking the habit; a severe chastisement is the only cure;
they too go on the principle of much wanting more.


NO NECESSITY TO HAVE THE BEES PLUNDERED IN THE FALL.

The apiarian having his bees plundered in the fall, is not fit to have
charge of them; their efforts are seldom as strong as in spring,
(unless there is a general scarcity,) the weak hives are usually better
supplied with bees, and consequently a less number is exposed; but yet,
when there are some very weak families, these should be taken away as
soon as the flowers fail, or strengthened with bees from another hive.
Particulars in fall management.

I have sometimes made my swarms equal, early in spring, by the
following method, and I have also failed. Bees, when wintered together
in a room, will seldom quarrel when first set out. When one stock has
an over supply of bees, and another a very few, the next day or two
after being out, I change the weak one to the stand of the strong one,
(as mentioned a page or two back,) and all bees that have marked the
location return to that place. The failure is, when too many leave the
strong stock, making that the weak one, when nothing is gained. If it
could be done when they had been out of the house just long enough for
the proper number to have marked the location, success would be quite
certain. But before an exchange of this kind is made, it would be well,
if possible; to ascertain what is the cause of a stock being weak; if
it is from the loss of a queen, (which is sometimes the case,) we only
make the matter worse by the operation. To ascertain whether the queen
be present, do not depend on the bees carrying in pollen; as most
writers assert they will not, when the queen is gone; because I have
_known_ them do it so many times without, that I can assure the reader
again, it is no test whatever. The test given in chapter III. page 73,
is always certain.



CHAPTER IX.

FEEDING.


SHOULD BE A LAST RESORT.

Feeding bees in spring is sometimes absolutely necessary; but in
ordinary seasons and circumstances, it is somewhat doubtful if it is
the surest road to success, for the apiarian to attempt wintering any
stock so poorly supplied with honey, that he feels satisfied will need
feeding in the spring or before. I will recommend in another place (in
fall management) what I consider a better disposition of such light
families. But as some stocks are either robbed, or from some other
cause, consume more honey than we expect, a little trouble and care may
save a loss. Also bees are often fed at this season to promote early
swarming, and filling boxes with surplus honey.


CARE NEEDED.

Considerable care is requisite, and but few know how to manage it
properly. Honey fed to bees, is almost certain to get up quarrels among
them. Sometimes strong stocks scent the honey given to weak ones, and
carry it off as fast as supplied.


APPARENT CONTRADICTION WHEN FEEDING CAUSING STARVATION.

It is possible that feeding a stock of bees in spring, may cause them
to starve! whereas, if let alone, they might escape. Notwithstanding
this looks like a contradiction, I think it appears reasonable.
Whenever the supply of honey is short, probably not more than one egg
in twenty which the queen deposits, will be matured--their means not
allowing the young brood to be fed. This appears from the fact that
several eggs may be found in one cell. I transferred over twenty stocks
in March, 1852--most of the cells occupied with eggs contained a
plurality; two, three, and even four, were found in one cell; it is
evident that all could not be perfected. Also, the fact of these eggs
being at this season on the bottom-board. Now suppose you give such a
stock two or three pounds of honey, and they are encouraged to feed a
large brood, and your supply fails before they are half grown. What are
they to do? destroy the brood and lose all they have fed, or draw on
their old stores for a small quantity to help them in this emergency,
and trust to chance for themselves? The latter alternative will
probably be adopted, and then, without a timely intervention of
favorable weather, the bees starve. The same effect is sometimes
produced by the changes of the weather; a week or two may be very fine
and bring out the flowers in abundance--a sudden change, perhaps frost,
may destroy all for a few days. This makes it necessary to use
considerable vigilance, as these turns of cold weather (when they
occur) make it unsafe, till white clover appears; but if the spring is
favorable, there is but little danger, unless they are robbed. If you
take the necessary care about worms, you will know which are light, and
which heavy, unless your hives are suspended; even then, it is a duty
to know their true condition, in this respect. This is another
advantage of the _simple_ hive; merely raising one edge to destroy
worms, tells you something about the honey on hand. To be very exact,
the hive should be weighed when ready for the bees, and the weight
marked on it; by weighing at any time after, tells at once within a few
pounds of what honey there is on hand. Some allowance must be made for
the age of the combs, the quantity of brood, &c. It is wrong to begin
to feed without being prepared to continue to do so, as the supply must
be kept up till honey is abundant.


HOW LONG IT WILL DO TO WAIT BEFORE FEEDING.

If it is wished to wait as long as possible, and not lose the bees, a
test will be necessary to decide how long it will do to delay feeding.
In this case, _strict attention will be necessary; they will need
examination every morning_. If a light tap on the hive is answered b; a
brisk, lively buzzing, they are not suffering yet; but if no answer is
returned to your inquiry, it indicates a want of strength. Extreme
destitution destroys all disposition to repel an attack. Sometimes a
part of the bees will be too weak to remain among the combs, and will
be lying on the bottom, and some few outside. If the weather is cool,
they appear to be lifeless; yet they can be revived, and now _must he
fed_.


DIRECTIONS FOR FEEDING.

Those among the combs may be able to move, though feebly. When this is
the condition of things, invert the hive, gather up all the scattered
bees, and put them in. Get some honey; if candied, heat it till it
dissolves; comb honey is not so good without mashing; if no honey is to
be had, brown sugar may be taken instead; add a little water, and boil
it till about the consistence of honey, and skim it; when cool enough,
pour a quantity among the combs, directly on the bees; cover the bottom
of the hive with a cloth, securing it firmly, and bring to the fire to
warm up. In two or three hours they will be revived, and may be
returned to the stand, providing the honey given is all taken up; on no
account let any honey run out around the bottom. The necessity of a
daily visit to the hives is apparent from the fact, that if left over
for one day, in the situation just described, it will be too late to
revive them. At night, if you have a box cover, such as I have
recommended, you may open the holes in the top of the hive; fill a
small baking dish with honey or syrup, and set it on the top; put in
some shavings to keep the bees from drowning, or a float may be used if
you choose; it should be made of some very light wood, very thin, and
full of holes or narrow channels, made with a saw. At the commencement
of feeding, a few drops should be scattered on the top of the hive and
trailed to the side of the dish, to teach them the way; after feeding a
few times, they will know the road. When the weather is warm enough for
them to take it during the night, it is best to feed at evening,--from
four to eight ounces daily, is sufficient. If the family is very small,
what honey is left in the morning may attract other bees; it is then
best to take it out, or carry the hive in the house to a dark room,
sufficiently warm, and feed them enough to last several days, and then
return them to the stand; keeping a good lookout that they are not
plundered, and again in a starving condition, until flowers produce
honey sufficient.


WHOLE FAMILIES MAY DESERT THE HIVE.

When you have the means to keep up a supply of food, and time requisite
to make feeding secure, perhaps it would not be advisable to wait till
the last extremity before feeding, as a small family will sometimes
entirely desert the hive, when destitute, if it occurs before they have
much brood. In these cases, they issue precisely as a swarm; after
flying a long time, they either return, or unite with some other stock.
If they return, they need attention immediately. You may be certain
there is something wrong, let the desertion take place when it may; in
spring it may be destitution, or mouldy combs; at other times the
presence of worms, diseased brood, &c. By whatever cause it is
produced, ascertain it, and apply the remedy.


OBJECTIONS TO GENERAL FEEDING.

I have known it recommended, and practised by some apiarians, to feed
bees all at once in the open air, in a large trough; but whoever
realizes much profit by this method, will be very fortunate, as every
stock in the neighborhood will soon scent it out, and carry off a good
share, and nearly every stock at home will be in contention, and great
numbers killed; the moment the honey is out, their attention is
directed to other stocks. Another objection to this general feeding is,
that some stocks are not necessitated at all, while others need it; but
the strongest stock is pretty sure to get the most. NOW, as I cannot
afford to divide with my neighbors in this way of feeding, and I
suppose but few will be found who are willing to do it, I will give my
method, which, when once arranged, is but little trouble.


ARRANGEMENT FOR FEEDING.

I got a tinman to make some dishes, two inches deep, 10×12 inches
square, and perpendicular sides. A board was then got out, fifteen
inches wide, and two feet long; two inches from one end, a hole is cut
out the longest way, just the size of the dish, so that it will set in
just even with the upper side of the board; a good fit should be made,
so that no bees can get in around it; cleats should be nailed on the
under side of the board, some over an inch thick, to prevent crowding
the dish out. This is to go directly under the hive, but it is not
ready yet, because if such dish is filled with honey under a hive, the
bees would drown; if a float is put on to keep them out, it will settle
to the bottom when the honey is out, and the bees cannot creep up the
sides of tin very easily. Another thing, there is nothing to prevent
the bees from making their combs to the bottom of this dish, two inches
below the bottom of the hive; these things are to be prevented. Get out
two pieces of half-inch board, ten inches long, one to be two inches
wide, the other one and a half inches. With a coarse or thick saw, cut
channels in the side of the strips, one-fourth inch deep, three-eighths
or half an inch apart, crosswise the whole length. You will then want a
number corresponding to the places sawed, of very thin shingles, or
strips, say one-eighth of an inch thick, and one and three-fourths
wide, and nine and a half long; these are to stand edgewise in the
dish; the first two are to hold them in the channels at the ends. The
narrow one needs a block one-half inch square, nailed on each end; on
the edge, a strip of wire cloth is then nailed on, making the whole
width just two inches. This is now put in the dish, wire cloth at the
bottom, two inches from one end; two pins to act as braces will keep it
there; the other wide one is placed against the other end, and pressed
down even with the top of the dish. The thin pieces are now slipped
into the channels even with the top; it is now ready to go under the
hive to be fed. Let the two-inch space project out on the back side of
the hive. A narrow board should be provided, some more than two inches
wide, to cover it. Let the hive stand close on this board; the hole in
the side is sufficient for the passage of bees at work, till very hot
weather. Thus you see that the hive covers all but the space behind,
which the board covers, and not a strange bee can get at the honey,
without entering the hole at the side, and passing through among the
bees belonging to the hive, which they will not often do; if the family
is numerous, it makes it as safe as feeding on the top; with this
advantage, there are no bees in the way to interfere while pouring in
the food. When the bees are to be fed, raise the board at the back and
pour in the honey; the wire-cloth in the bottom prevents all bees from
entering this space, at the same time will let the honey pass through
directly under the bees, which will take it up quicker than from any
other place that I can put it; they will work all night even when the
weather is quite cool. This board and feeder can be taken out when done
feeding, and put away till wanted again; if left under through the
summer, it affords the worms a place rather too convenient to spin
their cocoons, where they are not easily destroyed.


FEEDING TO INDUCE EARLY SWARMS.

If the object in feeding is to induce early swarms, of course the best
stocks should be chosen for the purpose; but some care is necessary not
to give too much, and fill the combs with honey, that ought to be
filled with brood, and thereby defeat your object; one pound per day is
enough, perhaps too much. The quantity obtained from flowers is a
partial guide; when plenty, feed less; when scarce, more. Begin as soon
as you can make them take it up in spring, and continue in accordance
with the weather, till white clover blossoms, or swarms issue. Another
object in feeding bees at this period, is to have the store combs all
filled with inferior honey, so that when clover appears, (which yields
our best honey,) there is no room except in the boxes to store it,
which are now put on, and rapidly filled. When this last object is
alone wished for, it is not much matter how much is given at a time,
providing it is all taken up through the night; it will then take no
time in day-light, when they might work on flowers; also, the bees
would have no trouble in repelling any attempt of others to get at it.


WHAT MAY BE FED.

Inferior honey may be used for this purpose; Southern or West India is
good, and costs but little. Even molasses sugar mixed with it will do;
but they do not relish it so well when fed without the honey. I have
usually taken about equal quantities of each, adding a pint of water to
ten pounds of this mixture, and making it as hot as it will bear
without boiling over, and skimming it.


IS CANDIED HONEY INJURIOUS?

There has an idea been advanced, that candied honey is injurious to
bees, even said to be fatal. I never could discover any thing further,
than it was a perfect waste, while in this state. When boiled, and a
little water added, it appears to be just as good as any. Nearly every
stock will have more or less of it on hand at this season; but as warm
weather approaches, and the bees increase to warm the hive, it seems to
get liquified, from this cause alone. The bees, when compelled to use
honey from these cells, thus candied, waste a large portion; a part is
liquid, and the rest is grained like sugar, which may be seen on the
bottom-board, as the bees work it out very often. Another object in
feeding bees, is to give inferior honey, mixed with sugar and flavored
to suit the taste, to the bees, and let them store it in boxes for
market. Now, I have no faith in honey undergoing any chemical change in
the stomach of the bee,[14] and cannot recommend this as the honest
course. Neither do I think it would be very profitable, feeding to this
extent, under any circumstances. I have a few times had some boxes
nearly finished and fit for market at the end of the honey season; a
little more added would make them answer. I have then fed a few pounds
of good honey, but always found that several pounds had to be given the
bees to get one in the boxes.

      [14] Mr. Gillman's patent for feeding bees, is based on the
      principle of a chemical change. It is said that the food he gives
      to the bees, when poured into the cells, becomes honey of the
      first quality. This appears extremely mysterious; for it is well
      understood that when a bee has filled its sack it will go to the
      hive, deposit its load, and return immediately for more; and will
      continue its labor throughout the day, or until the supply fails;
      each load occupying but few minutes. The time in going from the
      feeder to the hive is so short that a change so important is not
      at all probable. The nature of bees seems to be to _collect_
      honey, not _make_ it; hence we find, when bees are gathering from
      clover, they store quite a different article than when from
      buckwheat,--or when we feed West India honey, in quantities
      sufficient to have it stored _pure_ in the boxes, we find that it
      has lost none of its bad taste in passing through the sacks of
      our northern bees.

      It appears most probable that, if Southern honey and cheap sugar
      form the basis of his food, (which it is said to,) that it is
      flavored with something to disguise the disagreeable qualities of
      the compound. Should this be the secret, it would seem like a
      waste to feed it to bees--a portion would be given to the brood,
      and possibly the old bees might not always refrain from sipping a
      little of the tempting nectar. Why not, when the compound was
      ready,--instead of wasting it by this process,--put it directly
      in market? Or, is it necessary to have it in the combs to help
      psychologize the consumer into the belief that it is honey of a
      pure quality?



CHAPTER X.

DESTRUCTION OF WORMS.


I shall not give a full history of the moth in this chapter, as spring
is not the time they are most destructive. It will be further noticed
under the head of Enemies of Bees. But as this is a duty belonging to
spring, a partial history seems necessary.

As soon as the bees commence their labors, the worms are generally
ready to begin theirs.


SOME IN THE BEST STOCKS.

You will probably find some in your best stocks; but don't be
frightened; this is not the season when they often destroy your stocks,
yet they injure them some.


HOW FOUND.

In the morning, when cool, raise the hive, and you will find them on
the board. You must not suppose that these chaps are bred outside the
hive, got their growth, and are now on their way among the bees, but
the reverse. They are _bred in the hive_, and most of them are on the
way out, and this is the precise time to arrest them and bring them to
justice for their crimes.


A TOOL FOR THEIR DESTRUCTION.

I have used a simple tool, made in a few minutes, and very convenient
in this business. Any one can make it. Get a piece of narrow hoop-iron,
(steel would be better,) three-fourth inch wide, five inches long;
taper from one side three inches from the end to a point; then grind
each edge sharp; make three or four holes through the wide end, to
admit small nails through it in the handle, which should be about two
feet long and about half an inch square. Armed with this weapon, you
can proceed. Raise the hive on one edge, and with the point of your
sword you may pick a worm out of the closest corner, and easily scrape
all from under the hive with it. Now, _be sure and dispatch every one_;
not that the "little victim" will itself, personally, do much mischief;
but through its descendants the mischief is to be apprehended. Very
likely half of all you find will have finished their course of
destruction, among the combs, and have voluntarily left them for a
place to spin their cocoons. They are worried by the bees, if they are
numerous, until satisfied that it is no safe place among them to make a
shroud and remain helpless two or three weeks. Accordingly, when they
get their growth they leave, get on the board on the bottom, become
chilled and helpless in the morning, but again active by the middle of
the day. Now, if they are merely thrown on the earth, a place there
will be selected, if no better is found, for transformation; and a moth
perfected ten feet from the hive is just as capable of depositing five
hundred eggs in your hive, as if she had never left it.

Several generations are matured in the course of one summer:
consequently, one destroyed at this season, may prevent the existence
of thousands before the summer is over.

This is another subject of theoretical reasoning, and imposition, (at
least in my opinion.) I wish the reader to judge for himself; get rid
of whims and prejudice, and look at the subject candidly and fair; and
if there is no corroborative testimony comes up to confirm any position
that I assume, I shall not complain if my assertions fare no better
than some others. Only defer judgment till you _know_ for yourself.

Bees have ever received my especial regard and attention; and my
enthusiasm may blind my judgment. I may be prejudiced, but will not be
wilfully wrong. I have found so many theories utterly false, when
carried out in practice, that I can depend on no one's hypothesis,
however plausible, without facts in practice to support it. No one
should be fully credited without a test. To return to our subject.


MISTAKEN CONCLUSIONS.

It is supposed by many, when these worms are found on the board, they
get there by accident, having dropped from the combs above. They seem
not to understand that the worm generally travels on safe principles;
that is, he attaches a thread to whatever he travels over. To be
satisfied on this point, I have many times carefully detached his
foot-hold, when on the side of the hive or other place, where he would
fall a few inches, and always found him with a thread fast at the place
he left, to enable him to regain his position if he chose. Is it not
probable, then, that whenever he leaves the combs for the bottom-board,
he can readily ascend again? No doubt he often does, to be driven down
again by the bees. Now, what I wish to get at by all this preamble, is
simply this: that all our trouble and worrying to prevent the worms
from again ascending to the combs--by wire hooks, wire pins, screws,
nails, turned pins, clam-shells, blocks of wood, &c., is perfect
nonsense, when half or more of them would not harm the bees any more if
they did, and might as well go there as any where else. Besides, these
useless "fixins" are very often a positive injury to the bees.


OBJECTIONS TO SUSPENDED BOTTOM-BOARD.

Suppose, if you please, that the worm has no thread attached above, and
your board is far enough from the bottom of the hive to prevent his
reaching it. Of course, he can't get up; but how are your bees to do
any better? The worm can reach as high as they can. The bee can fly up,
you think; so it will, sometimes; but will try a dozen times first to
get up without, and when it does, it is a very bad position to start
from, being a smooth board. In hot weather it does better. Did you ever
watch by a hive thus raised, in April or May, towards night, when it
was a little cool, and see the industrious little insects arrive with a
load as heavy as they could possibly carry, all chilly, and nearly out
of breath, scarcely able to reach home, and there witness their vain
attempts to get among their fellows above them? If you never witnessed
this, I wish you would take some pains for it, and when you find them
giving up in despair, when too chilly to fly, and perishing after many
fruitless attempts for life, I think, if you possess sympathy,
benevolence, or even selfishness, you will be induced to do as I
did--discard at once wire hooks and all else from under the hive in the
spring, and give the bees, when they do get home with a load, under
such circumstances, what they richly deserve, and that is,
_protection_.


ADVANTAGE OF THE HIVE CLOSE TO THE BOARD.

An inch hole in the side of the hive, a few inches from the bottom, as
a passage for the bees, is needed, as I shall recommend letting the
hive close to the board; it is essential on account of robbing; also,
it is necessary to confine as much as possible the animal heat, in most
hives, during the season the bees are engaged in rearing young brood;
and warmth is necessary to hatch the eggs, and develop the larvæ; we
all know that when the hive is close, less heat will pass off than if
raised an inch.


OBJECTION ANSWERED.

You object to this, and tell me, "the worms will get between the bottom
of the hive and the board." Well, I think they will, and what then? Why
I expect if you intend to succeed, that you will get them out, and
crush their heads; if you cannot give as much attention as this, better
not keep them, or let some one have the care of them that will. I am as
willing to find a worm under the edge of the hive, and dispatch it, as
to have it creep into some place out of sight, and change to the moth.
I once trimmed off the bottom of my hives to a thin edge, so they did
not have this place for their cocoons, but now prefer to have them
square. _All profit_ is seldom obtained with anything. If you plant a
field with corn, you do not expect that the whole work for the crop is
finished. Neither should you expect when you set up a stock of bees,
that a full yield will be realized without something more. If you are
remunerated by keeping the weeds from your corn, be assured it is
equally profitable to weed out your bees.


INSUFFICIENCY OF INCLINED BOTTOM-BOARD.

Now do not be deceived in this matter, and through indolence be induced
to get those hives with descending bottom-boards, to throw out the
worms as they fall, and hope by that means to get rid of the trouble;
(I have already, in another chapter, expressed doubts of this). But we
will _now_ suppose such descending bottom-boards capable of throwing
every worm that touches it "heels over head" to the ground; what have
we gained? His neck is not broken, nor any other _bone_ of his body! As
if nothing extraordinary had happened, he quietly gathers himself up,
and looks about for snug quarters; he cares not a fig for the hive now;
he gormandized on the combs until satisfied, before he left them, and
is glad to get away from the bees any how. A place large enough for a
cocoon is easily found, and when he again becomes desirous of visiting
the hives, it is not to satisfy his own wants, but to accommodate his
progeny; he is then furnished with wings ample to carry him to any
height that you choose to put your bees.


A MOTH CAN GO WHERE BEES CAN.

A hive that is proof against the moth, is yet to be constructed. We
frequently hear of them, but when they come to be tested, somehow these
worms get where the bees are. When your hives become so full of bees,
that they cover the board in a cool morning, the worms will be seldom
found there, except under the edge of the hive.


TRAP TO CATCH WORMS.

You may now raise it, but you may still catch the worms by laying under
the bees a narrow shingle, a stick of elder split in two lengthwise,
and the pith scraped out, or anything else that will afford them
protection from the bees, and where they may spin their cocoons. These
should be removed every few days, and the worms destroyed, and the trap
put back. Do not neglect it till they change to the moth, and you have
nothing but to remove the empty cocoon.


BOX FOR WREN.

If you would take the trouble to put up a cage or two for the wren to
nest in, he would be a valuable assistant in this department of your
labor. He would be on the lookout when you were away, and many worms,
while looking up a hiding-place in some corner, would be relieved from
all further trouble by being deposited in his crop. The cage for him
need not be more than four inches square; it may be fastened near as
possible to the bees; to a post, tree, or side of some building a few
feet high. I have seen the skull of some animal (horse or ox) used, and
is very convenient for them, the cavity for the brains being used for
the nest. A person once told me the wren would not build in one that he
had put up. On examination, the stake to support it was found driven
into the only entrance. I mention this to show how little some people
understand what they do. It is sometimes well enough to know why a
thing is to be done, as to know it _must_ be done. I could tell you to
do a great many things, but then you would like to know _why_, then
_how_ to do it. Now if this prolixity is unnecessary for you, another
may need it. You must remember I am endeavoring to teach some few to
keep bees, who are not over supplied with ingenuity.



CHAPTER XI.

PUTTING ON AND TAKING OFF BOXES.


Putting on boxes may be considered a duty intermediate between spring
and summer management. I cannot recommend putting them on as early as
the last of April, or first of May, in ordinary circumstances. It is
possible to find a case that it would be best. But before the hive is
full of bees it is generally useless, very likely a disadvantage, by
allowing a portion of animal heat to escape that is needed in the hive
to mature the brood. Also, moisture may accumulate until the inside
moulds, &c. Some experience and judgment is necessary to know about
what time boxes are needed. That boxes _are needed_ at the proper
season, I think I shall not need an argument to convince any one, in
the present day. Bee-keepers have generally discarded the barbarous
practice of killing the bees to obtain the honey. Many of them have
learned that a good swarm will store sufficient honey for winter,
besides several dollars worth as profit in boxes.


ADVANTAGE OF THE PATENT VENDER.

Here is where the patent vender has taken the advantage of our ignorance,
by pretending that no other hive but _his ever obtained such quantities,
or so pure in quality_.


TIME OF PUTTING ON--RULE.

It is probable a great many readers will need the necessary observation
to tell precisely when the hive is full of honey; it may be full of
bees, and not of honey. And yet the only rule that I can give to be
generally applied, is, when the bees begin to be crowded out, but a day
or two before would be just the right time, that is, when they are
obtaining honey--(for it should be remembered that they do not always
get honey when beginning to cluster out). This guide will do in place
of a better one, which close observation and experience only can give.
By observing a glass hive attentively, in those cells that touch the
glass on the edge of the combs, whenever honey is being deposited here
abundantly, it is quite evident that the flowers are yielding it just
then, and other stocks are obtaining it also. Now is the time, if any
cluster out, to put on the boxes. When boxes are made as I have
recommended, that is, the size containing 360 solid inches, it is
advisable to put on only one at first; when this is full either of bees
or honey, and yet bees are crowded outside, the other can be added.
This is before swarming; too much room might retard the swarming a few
days, but if crowded outside, it indicates want of room, and the boxes
can make but little difference. It is better to have one box well
filled than two half full, which might be the case if the bees were not
numerous. The object of putting on boxes before swarming, is to employ
a portion of the bees, that otherwise would remain idly clustering
outside two or three weeks, as they often do, while preparing the young
queens for swarming. But when all the bees can be profitably engaged in
the body of the hive, more room is unnecessary.


MAKING HOLES AFTER THE HIVE IS FULL.

Whenever it is required to put boxes on a hive that has no holes
through the top, it need not prevent your getting a few pounds of the
purest honey that may be had, just as well as to have a portion of the
bees idle. I always endeavor to ascertain in what direction the sheets
of comb are made, and then mark off the row of holes on the top, at
right angles with them.


ADVANTAGE OF PROPER ARRANGEMENT.

Two inches being nearly the right distance, each one will be so made
that a bee arriving at the top of the hive between any two sheets will
be able to find a passage into the box, without the task of a long
search for it; which I can imagine to be the case when only one hole
for a passage is made, or when the row of holes is parallel with the
combs. A hive might contain eight or ten sheets of comb, and a bee
desirous of entering the box might go up between any two, many times,
before it found the passage. It has been urged that every bee soon
learns all passages and places about the hive, and consequently will
know the direct road to the box. This may be true, but when we
recollect that all within the hive is perfect darkness--that this path
must be found by the sense of feeling alone--that this sense must be
its guide in all its future travels--that perhaps a thousand or two
young workers are added every week, and these have to learn by the same
means--it would seem, if we studied our own interest, we would give
them all the facility possible for entering the boxes. What way so easy
for them as to have a passage, when they get to the top, between each
comb? That bees do not know all roads about the hive, can be partially
proved by opening the door of a glass hive. Most of the bees about
leaving, instead of going to the bottom for their exit, where they have
departed many times, seem to know nothing of the way, but vainly try to
get out through the glass, whenever light is admitted.

I am so well convinced of this, that I take some pains to accommodate
them with a passage between each comb; they will then at least lose no
time by mistakes between the wrong combs, crowding and elbowing their
way back through a dense mass of bees which impede every step, until
again at the top perhaps between the same combs, perhaps right, perhaps
farther off than at first; when I suppose they try it again; as boxes
are filled sometimes under just such circumstances.

To assist them as much as possible, when new hives are used for swarms,
I wait till the hive is nearly filled before making the holes to
ascertain the direction of the combs. We all know it is uncertain which
way the combs will be built, when the swarm is put in, unless
guide-combs are used.[15] When holes are made before the bees are put
in, guide-combs as directed for boxes should be put in; (of course they
should cross at right angles the row of holes).

      [15] Perhaps Miner's cross-bar hive would do it.


DIRECTIONS FOR BORING HOLES IN FULL STOCKS.

_To make holes in the top after the combs are made_,--Mark out the top
as directed for making hives and boxes. A centre bit or an auger bit
with a lip or barb is best, as that cuts down a little faster than the
chip is taken out, leaving it smooth; when nearly through, a pointed
knife can cut the remainder of the chip loose, and it can be taken out;
if it is between the combs, it is well; if directly over the centre of
one, it is a little better; with the knife take out a piece as large as
a walnut; even if honey is in it, no harm will be done. The bees will
then have a passage through from either side of the comb.

After you have opened one hole, very likely the bees will want to see
what is going on over head, and walk out to reconnoitre. To prevent
their interference, use some tobacco-smoke, and send them down out of
your way, till your hole is finished. Now lay over this a small stone
or block of wood, and make the others in the same way. When all are
done, blow in some smoke as you uncover them, and put on your box. This
process is not half so formidable as it appears; I have in this way
bored hundreds. You will remember my hives are not as high as many
others keep them, they are in about as convenient a position as I can
get them. This method saves me the trouble of sticking the guide-combs
in my hives; also, the necessity of covering or stopping the holes. Dr.
Bevan and some others have made a cross-bar hive, instead of nailing on
a top in the usual way; a half-inch board of the right length is cut
into strips, some over an inch wide, and half an inch apart, across the
top. It is plain that in such a hive a bee can pass into the box
whenever it arrives at the top, without difficulty. I will here repeat
the objection to allowing too much room, to pass into the boxes, that
you may see the disadvantages of the extremes of too little and too
much room. In these cross-bar hives, the animal heat rises into the box
from the main hive, making it as warm as below; the queen goes up with
the bees, and finding it warm and convenient for breeding, deposits her
eggs; and young brood as well as honey is found there. When we think it
full, it is then indispensable to return it, if taken off, till they
hatch, (otherwise they spoil it by moulding), which makes the combs
dark, tough, &c. Another objection to such open tops is, that open
bottom boxes must be used, which are not half as neat for market.


TO BE TAKEN OFF WHEN FILLED.

This advantage attends glass boxes: while being filled, the progress
can be watched till finished, when they should be taken off to preserve
the purity of the combs. Every day the bees are allowed to run over
them, renders them darker. Consequently, when our bees are a long time
filling a box, it is not as purely white as when filled expeditiously.


TIME TAKEN TO FILL A BOX.

Two weeks is the shortest time I ever had any filled and finished.
This, of course, depends on the yield of honey, and size of the swarm;
three or four weeks are usually taken for the purpose. I have before
said that the first yield of honey nearly fails in this section,
usually about the 20th of July; there are some variations, later or
earlier, according to the season. In other places it may be much later.


WHEN TO TAKE OFF BOXES PART FULL.

It can be ascertained by occasionally raising the cover to your glass
boxes. When no more is being added, all boxes that are worth the
trouble should be taken off; if left longer the comb gets darker, and
such cells of honey as are not sealed over, (and sometimes the majority
are such,) the bees generally remove down into the hive.


TOBACCO SMOKE PREFERRED TO SLIDES.

When boxes are to be taken off, if a slide of tin, zinc, &c., is used
to close the holes, some of the bees are apt to be crushed, others will
find themselves minus a head, leg, or abdomen, and all of them be
irritable for several days. A little tobacco smoke is preferable, as it
keeps all quiet. Just raise the box to be taken off sufficient to puff
under it some smoke, and the bees will leave the vicinity of the holes
in an instant; the box can then be removed, and another put on if
necessary, without exciting their anger in the least.


MANNER OF DISPOSING OF THE BEES IN THE BOXES.

Arouse the bees by striking the box lightly four or five times. If all
the cells are finished, and honey is still obtained, turn the box
bottom up, near the hive from which it was taken, so that the bees can
enter it without flying; by this means you can save several young bees,
that have never left the hive and marked the location, and a few others
too weak to fly, but will follow the others into the hive; (such are
lost when we are obliged to carry them at a distance.) Boxes can be
taken off either in the morning or evening; if in the morning, it may
stand several hours when the sun is not too hot, but on no account let
it stand in the sun in the middle of the day, as the combs will melt.
The bees will all leave, sometimes in an hour; at others they will not
be out in three. They may be taken off at evening and stand till
morning, in fair weather; if not too cool, they are generally all out;
but here is some risk of the moth finding it and depositing her eggs;
perhaps one in fifty may be thus found.


BEES DISPOSED TO CARRY AWAY HONEY.

When boxes are taken off at the end of the honey season, a different
method of getting rid of the bees must be adopted, or we lose our
honey. Unless the combs are all finished, we lose some then any way, as
most of the bees fill themselves before leaving; they carry it home and
return for more immediately, and take it all, if not prevented. It has
been recommended to take it to some dark room with a small opening to
let the bees out; in the course of the day they will sometimes all
leave; but this method I have found unsafe, as they sometimes find the
way back. When a large number of boxes are to be managed, a more
expeditious mode is, to have a large box with close joints, or an empty
hogshead, or a few barrels with one head out, set in some convenient
place; put the boxes in, one above another, but not in a manner to stop
the holes; over the top throw a sheet of one thickness, a thin one is
best, as it will let through more light. The bees will leave the boxes,
creep to the top, and get on the sheet; take this off and turn it over
a few times; in this way all may be got rid of without the possibility
of carrying off much honey. All that know the way will return to the
hive, but a few young ones are lost.


NOT DISPOSED TO STING.

They seldom offer to sting during this part of the operation, even when
the box is taken off without tobacco smoke, and carried away from the
hive; after a little time, the bees finding themselves away from home,
lose all animosity.

As honey becomes scarce, less brood is reared; a great many cells that
they occupied are soon empty; also, several cells that contained honey
have been drained, and used to mature the portion of brood just started
at the time of the failure. We can now understand, or think we do, why
our best stocks that are very heavy, that but a few days before were
crowded for room and storing in boxes, are now eager for honey to store
in the hive; as there is abundant room for several pounds. They will
quickly remove to the hive the contents of any box left exposed; or
even risk their lives by entering a neighboring hive for it; after
being allowed to make a beginning, under such circumstances.


RULE.

During a yield of honey, take off boxes as fast as they are filled, and
put on empty ones. At the end of the season take all off. Not one stock
in a hundred will starve that has worked in boxes, that is, when the
hive is the proper size, and full before adding the boxes, unless
robbed or other casualty.



CHAPTER XII.

SECURING HONEY FROM THE MOTH.


TWO THINGS TO BE PREVENTED.

When the boxes are free from the bees, two things are to be prevented,
if we wish to save our honey till cold weather. One is to keep out the
worms, the other to prevent souring. The last may be new to many, but
some few of us have had it caused by dampness in warm weather. The
combs become covered with moisture, a portion of the honey becomes thin
like water, and instead of the saccharine qualities we have the acid.
Remedy: keep perfectly dry and cool, if you can, but dry at any rate.


APT TO BE DECEIVED ABOUT THE WORMS.

But the worms, you can surely keep them out, you think, since you can
seal up the boxes perfectly close, preventing the moth or even the
smallest ant from entering! Yes, you may do this effectually, but the
worms will often be there somehow, unless in a very low temperature,
such as a very cool cellar, or in house, and then you have dampness to
guard against. I have a little experience in this matter that spoils
your theory entirely. I have taken off glass jars, and watched them
till the bees were all out, and was _certain the moth did not come
near_ them, then immediately sealed them up; absolutely preventing
access afterwards, (I could do this with a jar more effectually than a
box which is made of several pieces,) I then felt quite sure that I was
ahead, and should have no trouble with the worms, as had often been the
case before. I was sadly mistaken.


THEIR PROGRESS DESCRIBED.

In a few days, I could see at first a little white dust, like flour, on
the side of the combs, and on the bottom of the jar. As the worms grew
larger, this dust was coarser. By looking closely at the combs, a small
white thread-like line was first perceptible, enlarging as the worm
progressed.

When combs are filled with honey, they go only on the surface, eating
nothing but the sealing of the cells; seldom penetrating to the centre,
without an empty cell to give the chance. Disgusting as they seem to
be, they dislike being daubed with honey. _Wax, and not honey, is their
food._

The reader would like to know how these worms came in the jars, when,
to all appearance, it _was a physical impossibility_. I would like to
tell positively, but cannot. But I will guess, if you will allow it. I
will first premise, that I do not suppose they are generated
spontaneously! Their being found there, then, would indicate some agent
or means not readily perceived.


A SOLUTION OFFERED.

The hypothesis that I offer is original and new, and therefore open for
criticism; if there is a better way to account for the mystery, I would
be glad to know it.

From the first of June till late in the fall, the moth may be found
around our hives, active at night, but still in the day. The only
object probably is to find a suitable place to deposit its eggs, that
the young may have food; if no proper and convenient place is found,
why, I suppose it will take up with such as it _can_ find; their eggs
_must_ be deposited somewhere, it may be in the cracks in the hive, in
the dust at the bottom, or outside, as near the entrance as they dare
approach. The bees running over them may get one or more of these eggs
attached to their feet or bodies, and carry it among the combs, where
it may be left to hatch. It is not at all probable that the moth ever
passed through the hive among the bees, to deposit her eggs in the jars
before mentioned. Had these jars been left on the hive, not a worm
would have ever defaced a comb; because, when the bees are numerous,
each worm as soon as it commences its work of destruction will be
removed, that is, when it works on the surface, as in the boxes of
honey--in breeding combs, they get in the centre and are more difficult
to remove. By taking off these jars and removing the bees, it gave all
the eggs that happened to be there a fair chance. Many writers finding
the combs undisturbed when left on the hive till cold weather,
recommend that as the only safe way, preferring to have the combs a
little darker, than the risk of being destroyed by the worms. But I
object to dark combs, and leaving the boxes will effectually prevent
empty ones taking their places, which are necessary to get all the
profits. I will offer a few more remarks in favor of my theory, and
then give my remedy for the worms. I have found in all hives where the
bees have been removed in warm weather, say between the middle of June
and September, (and it has been a great many,) moth eggs enough among
the combs to destroy them in a very short time, unless kept in a very
cool place; this result has been uniform. Any person doubting this, may
remove the bees from a hive that is full of combs in July or August;
and close it to prevent the _possibility_ of a moth entering, set it
away in a temperature ranging from sixty to ninety, and if there are
not worms enough to satisfy him that this is correct, he will have
better success than I ever did. Yet, no such result will follow, when
the bees are left among the combs, unless the swarm be very small; then
the injury done will be in proportion. A strong stock may have as many
moth eggs among the combs as a weak one, yet one will be scarcely
injured, while the other may be nearly or quite destroyed.

Now, if this theory be correct, and the bees do actually carry these
eggs among the combs, is there not a great deal of lost labor in trying
to construct a moth-proof hive? The moth, or rather the worms, are ever
present to devour the combs, whenever the bees have left them in this
season.


METHOD OF KILLING WORMS IN BOXES.

Now, whether you are satisfied or not with the foregoing, we will
proceed with the remedy. Perhaps you may find one box in ten that will
have no worms about it, others may contain from one to twenty when they
have been off a week or more. All the eggs should have a chance to
hatch, which in cool weather may be three weeks. They should be
watched, that no worms get large enough to injure the combs much,
before they are destroyed. Get a close barrel or box that will exclude
the air as much as possible; in this put the boxes, with the holes or
bottom open. In one corner leave a place for a cup or dish of some
kind, to hold some sulphur matches while burning. (They are made by
dipping paper or rags in melted sulphur.) When all is ready, ignite the
matches, and cover close for several hours. A little care is required
to have it just right: if too little is used, the worms are not killed;
if too much, it gives the combs a green color. A little experience will
soon enable you to judge. If the worms are not killed on the first
trial, another dose must be administered. Much less sulphur will adhere
to paper or rags, if it is very hot, when dipped, than when just above
the temperature necessary to melt it; this should be considered, as
well as the number of boxes to be smoked, size of the vessel used in
smoking them, &c.

Whether this gas from burning sulphur will destroy the eggs of the moth
before the worm appears, I have not tested sufficiently to decide; but
I do know that it is an effectual quietus for the larvæ!


FREEZING DESTROYS THEM.

Boxes taken off at the end of warm weather, and exposed in a freezing
situation through the winter, appear to have all the worms as well as
eggs for them destroyed by the cold; consequently, all boxes so
exposed, may be kept any length of time; the only care being necessary,
to shut out the moth effectually. But don't forget to look out for all
combs from which the bees have been removed in warm weather. I prefer
taking off all boxes at the end of the first yield of honey, even when
I expect to put them on again for buckwheat honey. The bees at this
season collect a great abundance of propolis, which they spread over
the inside of the boxes as well as hive; in some instances it is spread
on the glass so thick as to prevent the quality of honey being seen.
There is no necessity for boxes on a hive at any season when there is
no yield of honey to fill them. Sometimes even in a yield of buckwheat
honey, a stock may contain too few bees to fill boxes, but just a few
may go into them and put on the propolis; this should not be allowed,
as it makes it look bad when used another year. At this season,
(August) some old stocks may be full of combs, and but few bees, but
swarms when they have got the hive full in time, are very sure to have
bees enough to go into the boxes to work. I have known them to do so in
three weeks after being hived.


OBJECTION TO USING BOXES BEFORE THE HIVE IS FULL.

Some put on boxes at the time of hiving the bees. In such cases the box
is often filled first, and nearly as often will contain brood. I
consider it no advantage, and often a damage to do so; as I want the
hive full any way--and then if they have time let them into boxes,
although it may be buckwheat, instead of clover honey that we get.



CHAPTER XIII.

SWARMING.


TIME TO EXPECT THEM.

The season for regular swarms in this section, I have known to commence
the 15th of May, and in some seasons the 1st of July. The end is about
the 15th of the latter month, with some exceptions. I have had one as
late as the 21st; also a few buckwheat swarms between the 12th and 25th
of August.

The subject now before us is one of thrilling interest. To the apiarian
the prospect of an increase of stocks is sufficient to create some
interest, even when the phenomenon of swarming would fail to awaken it.
But to the naturalist this season has charms that the indifferent
beholder can never realize.


ALL BEE-KEEPERS SHOULD UNDERSTAND IT AS IT IS.

As a guide in many cases, it is important that the practical apiarian
should understand this matter _as it is_, and not as said to be by many
authors. I shall be under the necessity of differing from nearly all in
many points.


MEANS OF UNDERSTANDING IT.

This is another case of "when doctors disagree, who shall decide?" You,
reader, are just the person. There is no need of a doctor at all in
this matter. I will endeavor to give a test for most of my assertions.
To make this subject as plain as possible in this place, I may repeat
some things said before. The facts related have come under my own
observation. I have probably taken more pains than most bee-keepers, to
understand this matter to the bottom _from the beginning_, (I mean the
bottom of the cells). But few apiarians have made the number of
examinations that I have to get at the _modus operandi_ of swarming.
Perhaps I ought not to expect full credit for veracity, when I assure
the reader that I have inverted more than one hundred stocks to get a
peep at the royal cells, some of them near a dozen times in one summer.
I have inverted them frequently for the purpose of obtaining cells. But
generally to see when such cells are being made, when they contain
eggs, when these eggs are sufficiently matured for swarming, or
abandoned and destroyed, &c.

By these signs I predict with certainty (almost) when to expect swarms,
and when to cease looking for them.


INVERTING A STOCK RATHER FORMIDABLE AT FIRST.

To a person that has never inverted a hive full of bees, even to
overflowing, or never has seen it done, it appears like a great
undertaking, as well as the probability of ruining the stock! But after
the first trial, the magnitude of the performance is greatly
diminished, and will grow less with every repetition of the feat, until
there is not the least dread attending it. Without tobacco smoke I
hardly think it practicable, but with it, there is not the least
difficulty. It would be very unsatisfactory to turn over a hive and
nothing to drive the bees away from the very places on the combs that
you wish particularly to inspect. The smoke is just the thing to do it!
As for the bad effects of such overturning and smoking, I never
discovered any.


REQUISITES BEFORE PREPARATION OF QUEEN'S CELLS.

I have found the process for all regular swarms something like this:
before they commence, two or three things are requisite. The combs must
be crowded with bees; they must contain a numerous brood advancing from
the egg to maturity; the bees must be obtaining honey either by being
fed or from flowers. Being crowded with bees in a scarce time of honey
is insufficient to bring out the swarm, neither is an abundance
sufficient, without the bees and the brood. The period that all these
requisites happen together, and remain long enough, will vary with
different stocks, and many times do not happen at all through the
season, with some.

These causes then appear to produce a few queen-cells, generally begun
before the hive is filled, (sometimes when only half full, but usually
remain as rudiments till the next year, when the foregoing conditions
of the stock may require their use).


STATE OF QUEEN'S CELL WHEN USED.

They are about half finished, when they receive the eggs; as these eggs
hatch into larvæ, others are begun, and receive eggs at different
periods for several days later. The number of such cells seem to be
governed by the prosperity of the bees: when the family is numerous,
and the yield of honey abundant, they may amount to twenty, at other
times perhaps not more than two or three; although several such cells
may remain empty. I have already said that a failure, (or even a
partial one), in the yield of honey at any time from the depositing of
royal eggs till the sealing of the cells, (which is about ten days),
would be likely to bring about their destruction. Even after being
sealed, I have found a few instances where they were destroyed.


STATE WHEN SWARMS ISSUE.

But when there is nothing precarious about the honey, the sealing of
these cells is the time to expect the first swarm, which will generally
issue the first fair day after one or more are finished. I never missed
a prediction for a swarm 48 hours, when I have judged from these signs,
in a prosperous season. When there is a partial failure of honey, the
swarm sometimes will wait several days after finishing them.


CLUSTERING OUTSIDE NOT ALWAYS TO BE DEPENDED UPON.

The clustering out of the bees I find but a poor criterion to judge
from, further than full hives do swarm--many such do not.


EXAMINATIONS--THE RESULT.

I will detail a few circumstances, that have led to these conclusions.
Some years ago the honey began to fail, when only about one third of my
good stocks had cast swarms; and all at once, the issues began to "be
few and far between." I had previously examined, and found they had
gone into preparations pretty extensively; by having not only
constructed cells, but occupied them with royal eggs and larvæ. Now I
examined again, and found five out of six had destroyed them, (at the
same time the bees clustered out extensively). This put an end to all
hopes of swarms here. Some few had finished their cells, and these, I
had some hopes, would send out the swarms; but the dry weather caused
some misgivings. After waiting three or four days and none coming, I
found these sealed cells destroyed also, and had no more swarms that
season. Subsequent observations have fully confirmed these things. One
season some of the hives commenced preparations at two different
periods, and then abandoned them without swarming at all, through the
summer. The first time it was the last of May, the next in July.


REMARKS.

The failure of honey was the cause, without any doubt. And who shall
say, these bees were not wise in their conduct? What prudent man would
emigrate with a family, if the prospect of a famine was plainly
indicated, when, by remaining at home, there was enough, at least for
the present? Who can help but admire this wise and beautiful
arrangement? The combs must contain brood; the bees must find honey
during the rearing of the queens. If a swarm were to issue the moment
of obtaining honey, the consequence might be fatal, as there would not
be a numerous brood to hatch out, and replenish the old stock with bees
sufficient to keep out the worms. Were they to issue at any time, as
soon as the bees had increased enough in numbers to spare a swarm,
without regard to the yield of honey, they might starve.


CONFLICTING THEORIES.

I find many theories conflicting with these views, which appear to call
for some remarks. It is generally supposed that a young queen must be
matured to issue with the swarms, and the old one with the old bees are
permanent residents of the old hive.


BOTH OLD AND YOUNG LEAVE WITH SWARMS.

It is probable that no rule governs the issue of workers. Old and young
come out promiscuously. That old bees come out may be known sometimes,
by so many leaving, that not a quarter as many will be left, as
commenced work in the spring. That young bees leave, any one may be
satisfied on seeing a swarm issue; a great many too young and weak to
fly will drop down in front of the hive, having come out now for the
first time, and perhaps some of them had not been out of the cell an
hour; these very young bees are known by the color.


CAUSE OF THE QUEEN'S INABILITY TO FLY SUGGESTED.

The old queen often gets down in the same way; but I would assign
another cause for her inability to fly; that is, I would suggest it to
be her burden of eggs.


EVIDENCE OF THE OLD QUEEN'S LEAVING.

That the old queen does leave with the first swarm is indicated by
several things: one is, eggs may often be found on the board the next
morning; another, when the first swarm has left, and before any of
these royal cells hatch, the bees may be driven out and no queen will
be found, or you may drive out the bees at the end of three weeks, and
the brood of workers will be about all hatched, the drone brood not
quite as near. The combs may also contain some eggs, and perhaps some
very young larvæ, that have been deposited by the young queen, which
begins to lay usually sixteen or eighteen days after the first swarm.
This shows a cessation of laying eggs for about two weeks. First swarms
will have eggs in the cells as soon as they are made to hold them,
which is often within 24 hours after being hived; occasionally a new
piece of comb will fall down, and, if the cells are deep enough, they
are almost certain to contain eggs. I could add other proof, but the
attentive observer will discover it himself.


MR. WEEKS' THEORY NOT SATISFACTORY.

Mr. J. M. Weeks, in his work on bees, says, "Two causes and two only
can be assigned why bees ever swarm: the first, the crowded state of
the hive; the second, to avoid the battle of the queens." The first
cause producing first swarms, the other second, third, &c. Mr. Colton's
patent hive, it is said, can be made to swarm "at any time within two
days," merely for want of room. By removing the six boxes attached to
it, the bees are compelled to crowd into the main body of the hive, and
swarm out in consequence. Now, if merely crowding the hive with bees is
the only cause of first swarms, how is it that half or more of mine
refused to swarm, when a great many, for want of room, were crowded
outside for weeks, and great numbers maturing every day to crowd them
still more? To me the reason is plain, that some of the
before-mentioned requisites were wanting. Mr. Weeks further says, when
the first swarm has left, "not a single queen, in any stage of
minority, is left in the old hive; the bees, destitute of a queen, set
about constructing several royal cells, take larvæ or eggs and put in
them, and feed with royal jelly, and in a few days have a queen."
Although I had not had much experience at the time of getting his work,
I had some doubts, because I found that all hives that became full and
began to run over, did not swarm, and some others swarmed before being
quite full; it seemed as if something like a preparation beforehand was
requisite. I knew of no means, for a long time, that would decide
_positively_; when it occurred to me, if I examined the old stock
immediately after the first swarm had left, I should find some
preparations if there were any; a thing so simple and easy that I felt
somewhat mortified not to have thought of it before. The first stock I
looked at revealed the secret. I examined it the evening of the day
that a swarm had left; I was gratified by finding two finished cells on
the lower edges of the combs; other cells were in different stages of
progression, from those containing an egg to the full developed larva.
Several more hives showed the same result. I now got bold enough to
examine some previous to swarming, as I have already explained.


MR. MINER NOT CORRECT.

Mr. T. B. Miner, in his work, has allowed the preparation of queen
cells previous to swarming, but he has put off the time of the swarm
issuing eight or nine days too long. That is, he has the young queen
matured so that she commences piping first, which does not occur more
than one time in fifty.

Now I think it more than probable that many readers will have some
doubts in regard to my statements about this swarming matter. Yet I
think I can give directions sufficiently particular that they may
remove them themselves. They should bear in mind that they have no
right to be _positive_ on any subject without an investigation.


PARTICULAR DIRECTIONS FOR TESTING THE MATTER.

I will now give more minute directions for an examination. Full hives
require a little more care than those containing fewer bees. Don't let
the crowded state of the hive, even if some are outside, deter you from
gratifying a laudable curiosity, (such hives are most likely to possess
these cells.) Let the satisfaction of ascertaining a few facts for
yourselves stimulate you to this exertion, the risk is not much; what I
have done you may do. This is better than to rely on any man's "_ipse
dixit_." I do it without any protection whatever for face or hands;
but, if you have too much fear of stings, a veil to protect the face
may be put on, but do without it, if you can find the courage, as you
will want a good view. The best time is, when most of the bees are out
at work near the middle of the day; but then the bees from the other
hives are sometimes cross, and interfere. On that account I prefer
morning or evening, although there are more bees to be smoked out of
the way. If you are accustomed to smoking tobacco, you will find a pipe
just the thing for making a smoke here; if not, vide a description of
an apparatus in chap. 18th, p. 281. When you are ready to proceed, some
smoke must be blown under the hive before you touch it; then raise the
front side a few inches, and blow in some more; now carefully lift the
hive from the stand, avoiding any jar, as this would arouse their
anger; turn it bottom upwards; also, be careful all the time not to
breathe among them. More smoke will now make them crowd among the combs
out of your way while you examine. It is very common for the bees to
set up a buzzing, and rush up the sides of the hive, but a little smoke
will drive them back; get them out of the way as much as possible, and
look on the edges of the combs for the queens' cells, where most of
them are. If the hive is fully supplied with honey, they will be near
the bottom, if not, farther up among the combs; in some hives they
cannot be seen even where they exist. Yet they may be found in four out
of five, by a thorough search. I have found nine within two inches of
the bottom, some on the extreme ends of the comb. I would here give a
caution about turning over hives with very new combs, before they are
attached to the sides of the hive, as they are apt to bend over.


EMPTY HIVES TO BE READY.

We will now suppose that some of your stocks are ready to cast their
swarms: we will also presume that your empty hives for the reception of
swarms are ready before this period; to prepare a hive after the swarm
has issued is bad management; negligence here argues negligence
elsewhere; it is one of the premonitions of "bad luck."


BOTTOM-BOARDS FOR HIVING.

You will want also a number of bottom-boards, expressly for hiving; get
a board a little larger than the bottom of the hive, nail strips across
the ends on the under side to prevent warping; in the middle cut out a
space five or six inches square, and cover with wire cloth. These are
for your large swarms in very hot weather, to be used for four or five
days. They are much safer than to raise the hive an inch or more for
ventilation. They are also essential for many other occasions. I would
not do without them, even if the expense was ten times what it is.


DESCRIPTION OF SWARM ISSUING.

When the day is fair and not too much wind, first swarms generally
issue from ten o'clock till three; if you are on the lookout, the first
outside indication of a swarm, will be an unusual number of bees around
the entrance, from one to sixty minutes before they start. The utmost
confusion seems to prevail, bees running about in every direction; the
entrance apparently closed with the mass of bees, (perhaps one
exception in twenty,) presently a column from the interior forces a
passage to the open air; they come rushing out by hundreds, all
vibrating their wings as they march out; and when a few inches from the
entrance, rise in the air; some run up the side of the hive, others to
the edge of the bottom-board. If you have seen the old queen come
rushing out the first one, and the rest following her, as we are often
told she does, you have seen what I never did in a first swarm! Second
and third swarms conduct themselves quite differently. I have seen the
old queen issue a few times, but not till half the swarm was out.

The bees when first rising from the hive, describe circles of but few
feet, but as they recede, they spread over an area of several rods.
Their movement are much slower than usual, in a few minutes thousands
may be seen revolving in every possible direction! A swarm may be seen
and heard, at a distance, where fifty hives, ordinarily at work, would
not be noticed! When about out of the hive, or soon after, some branch
of a tree or bush is usually selected on which to cluster. In less than
half a minute after the spot is indicated, even when the bees are
spread over an acre, they are gathered in the immediate vicinity, and
all cluster in a body from five to ten minutes after leaving the hive.
They should now be hived immediately, as they show impatience if left
long, especially in the sun; also, if another stock should send out a
swarm while they were hanging there, they would be quite sure to mix
together.


MANNER OF HIVING CAN BE VARIED.

It makes but little difference what way they are put in the hive,
providing they are all made to go in. Proceed as is most convenient; an
old table or bench is very good to keep them out of the grass if there
should happen to be any; if there is not much in the way, lay your
bottom-board on the ground, make it level, set your hive on it, and
raise one edge an inch or more to give the bees a chance to enter.


USUAL METHOD.

Cut off the branch with the bees, if it can be done as well as not, and
shake it in front of the hive, a portion will discover it, and will at
once commence a vibration of their wings; this, I suppose, is a call
for the others. A knowledge of a new home being found seems to be
communicated in this way, as it is kept up until all are in. A great
many are apt to stop about the entrance, thereby nearly or quite
closing it, and preventing others going in, when they will gather on
the outside. You can expedite the matter with a stick or quill, by
gently pushing them away; and another portion will enter. When gentle
means will not induce them to go in, in a reasonable time, and they
appear obstinate, a little water sprinkled on them will facilitate
operations greatly, when nothing else will. (Be careful and not over-do
the matter, by using too much water, they can be so wet as not to move
at all.)

When they cluster on a branch that you do not wish to cut off, place
your bottom-board as near as convenient; on it lay two sticks about an
inch in diameter, of the same length: try the hive, and see that all is
right; then turn it bottom up, directly under the main part of the
cluster; if you have an assistant, let him jar the branch sufficiently
to detach the bees; most of them will fall directly into the hive. If
no assistant is at hand it is unnecessary to wait, (I have done it a
hundred times without help); with the bottom of the hive strike the
under side of the branch hard enough to dislodge them, then turn it on
the board; the sticks will prevent the bottom crushing many bees.


WHEN OUT OF REACH.

I have gone up a ladder fifteen feet, got the bees in the hive in this
way, and backed down without difficulty. After putting the hive in its
place, sometimes a part will go back; in that case, a small branch full
of leaves should be held directly under and close to them, and as many
jarred on it as possible. Hold this still, and shake the other to
prevent their clustering there; you will soon have them all collected,
ready to bring down, and put by the hive. A handle basket or large tin
pan may be taken up the ladder instead of the hive, when they can be
readily emptied before it. But very few will fly out in coming down. If
you succeed in getting nearly all the bees in the first effort, and but
few are left, merely shaking the branch will be sufficient to prevent
their holding fast, and will turn their attention to those below, where
those which have already found a hive will be doing their best to call
them. When the hive is first turned over, most of the bees fall on the
board and rush out, but as soon as it is realized that a home is found,
a buzzing commences inside; this quickly communicates the fact to those
outside, which immediately turn about, facing the hive and hum in
concert, while marching in.

Another plan may be adopted, even if fifteen feet high; when the branch
is not too large, and there is not too much in the way below it. Have
ready two or three light poles of suitable length; select such as have
a branch at the upper end, large enough to hold a two-bushel basket.
This is raised directly under the swarm; with another pole, the bees
are all dislodged, and fall into the basket, and are quickly let down.
Now, if you have got about all, throw a sheet over for a few moments,
to prevent their escape. They soon become quiet, and may be hived
without many going back to the branch, as they do, when attempting to
hive them immediately.

I often have them begin to cluster near the ground, very conveniently
for hiving. In such a case, I do not wait for all to collect, but as
soon as such place is indicated, I get the board and hive ready. When a
quart or so are gathered, shake them in a hive, and set it up; the
swarm will now go to that, instead of the branch, especially if the
latter is shaken a little. Where many stocks are kept, it is advisable
to be as expeditious as possible. A swarm will thus hive itself much
sooner than when it is allowed to cluster.


WHEN THEY CANNOT BE SHAKEN OFF.

Swarms will sometimes get in places where it is impossible to jar them
off, or cut off a branch, such as the trunk of a tree, or a large limb
near it. In which case place the hive near, as first directed; take a
large tin dipper, a vessel most convenient for the purpose, and dip it
full of bees; with one hand turn back the hive; with the other throw
the bees into it; some of them will discover that a home is provided,
and set up the call for the rest, (by the vibration of their wings),
and the remainder may be emptied in front of the hive as you dip them
off. I have known a few instances when the first dipper full all ran
out, and joined the others without making the discovery that they were
in a hive, but this is seldom the case. When you get the queen in,
there is no trouble with the remainder, even if there are many left; as
soon as they ascertain that the queen is no longer among them, it may
be known by their uneasy movements, and they will soon leave, and join
those in the hive; but if the queen is yet on the tree, and but a dozen
with her, they will leave the hive and cluster again.


ALL SHOULD BE MADE TO ENTER.

In all cases be sure to get them all to enter; a cluster outside of it
may contain the queen, unconscious of a home so near; and the probable
consequence might be, she would leave for a miserable one in the woods.


SHOULD BE TAKEN TO THE STAND IMMEDIATELY.

When all are in, except a few that will be flying, let the hive down
close to the board; take hold of this and carry it at once to the stand
they are to occupy, and raise the front edge half an inch; let the back
rest on the board; this will give them means to re-ascend, if they
chance to drop, which large swarms often do in hot weather. If the
bottom is an inch or more from the board when the bees fall, there is
nothing to prevent their rushing out on every side--their means of
getting up again are bad--if the queen comes out with the rush, there
are some chances for their leaving.


PROTECTION FROM THE SUN NECESSARY.

Another thing is very important; _swarms should be protected from the
sun for several days, in hot weather_, from nine o'clock till three or
four; and then if the heat is very oppressive, and the bees cluster
outside, sprinkle them with water and drive them in; and by wetting the
hive occasionally, it will carry off a large portion of the heat, and
make it much more comfortable.


CLUSTERING BUSHES.

If there are no large trees in the vicinity of your apiary, all the
better, as there will then be no danger of your swarms lighting on
them; but all bee-keepers are not so fortunate, myself being one of the
number. In such a place it is necessary to provide something for them
to cluster on; get some bushes six or eight feet high (hemlock is
preferable); cut off the ends of the branches, except a few near the
top: secure the whole with strings to prevent swaying in ordinary
winds; make a hole in the earth deep enough to hold them, and large
enough to be lifted out easily. The bees will be likely to cluster on
some of these; they can then be raised out, and the bees hived without
difficulty. A bunch of dry mullein tops tied together on the end of a
pole, makes a very good place for clustering; it so nearly resembles a
swarm that the bees themselves appear to be sometimes deceived. I have
frequently known them leave a branch where they had begun to cluster,
and settle on this when held near.

The motives for immediately removing the swarm to the stand are, that
they are generally more convenient to watch in case they are disposed
to leave; also many bees can be saved. All that leave the hive, mark
the location the same as in spring; several hundreds will probably
leave the first day; a few may leave several times; when removed at
night, such will return to the stand of the previous day, and generally
are lost; whereas, if they are carried at once to a permanent stand,
this loss is avoided.

Those that are left flying at the time, return to the old stock, which
those that return from the swarm the next day will not always do. The
time for moving them now is no more than at another. It is unnecessary
to object, and say, that "it will take too long to wait for the bees to
get in;" this will not do. I shall insist on your getting all the bees
to enter before leaving any way. I consider this an essential feature
in the management. I will not say that my directions will _always_
prevent their going to the woods, but this I do say, that out of the
hundreds that I have hived, not one has ever left. It is possible
proper management has had no influence in my success, yet something
like an opinion of this kind has been indulged for a long time.


HOW SWARMS ARE GENERALLY MANAGED THAT LEAVE FOR THE WOODS.

Some of my neighboring bee-keepers lose a quarter or half of their
swarms by flight, and how do they manage? When the word is given out,
"Bees swarming," a tin-horn, tin-pan, bells, or anything to make a
"horrible din," is seized upon in the hurry of the moment, and as much
noise made as possible, to _make_ them cluster; (which they naturally
would do without the music, at least all mine have. This probably gave
rise to the opinion of one old lady, who _knew_ "drumming on a pan did
good, for she had tried it.") Very often a hive is to be constructed,
or an old one unfit to use any way, needs some sticks across, or
something to take time. When the hive is obtained, it must be washed
with something nice to make the bees like it; a little honey must be
daubed on the inside; sugar and water, molasses and water, salt and
water, or salt and water rubbed on with hickory leaves, "is the best
thing in the world;" several other things are just as good, and some
are better. Even whisky, that bane of man, has been offered them as a
bribe to stay, and sometimes they are persuaded and go to work.


NOTHING BUT BEES NEEDED IN A HIVE.

Now I cannot say positively that these things do harm, yet I am quite
sure they do no good, as nothing but bees is needed in a hive. Is it
reasonable to suppose they are fond of all the "knick-knacks" given
them? I have never used any, and could not possibly have done better. I
am careful to have the hive sweet and clean, and not too smooth inside;
an old hive that has been used before is scalded and scraped.

But to the manner they get the bees in, after the hive is ready. A
table is set out, and a cloth spread on it; sticks are put on to raise
the hive an inch or more: if they succeed in getting the swarm even on
the outside of the hive it is left; if they go in, it is well; if they
go off, why hope for "better luck next time." The hive is left
unsheltered in the hot sun and when there is no wind, the heat is soon
insupportable, or at least very oppressive; the bees hang in loose
strings, instead of a compact body, as when kept cool; they are very
apt to fall, and when they do, will rush out from every side: if the
queen chances to drop with them, they _may_ "step out." Two thirds of
all the bees that go to the woods are managed in this, or a similar
manner, and may it not be said, they are fairly driven off?


SELDOM GO OFF WITHOUT CLUSTERING.

Perhaps one swarm in three hundred will depart for the woods without
first clustering. I have had three times that number, not one of which
has ever left me thus. Yet I have evidence not to be disputed that some
will do it. Three instances have occurred near me that satisfied me of
the fact. Two were lost, the other was followed to a tree, half a mile
off; I assisted in cutting the tree, and hiving them. The cavity where
they entered was very small, and contained old comb, made by a swarm a
year or two previous, which had probably starved, as there was too
little room for storing sufficient honey for winter. This swarm, when
hived and carried home, remained perfectly contented.


DO SWARMS CHOOSE A LOCATION BEFORE SWARMING?

The inquiry is often made, Do all swarms have a place looked out before
leaving the parent stock? The answer to this must ever be guess-work. I
could offer some circumstances indicating the affirmative very
strongly, and as much for the negative; and will let it pass at that.
Yet I think if bees are properly cared for, that ninety-nine swarms in
a hundred will prefer a good clean hive to a rotten tree in the woods.


MEANS OF ARRESTING A SWARM.

I have had three swarms that were exceptions to general rules, giving
me some trouble by swarming out after being hived; the third and fourth
time they left, I threw water among them, causing quite a shower; when
my pail-full was out, I used earth; they went but a short distance, and
clustered in the usual way. Now were these bees intending to leave, and
had their designs frustrated by the water and earth? I am not quite as
sure as the old lady, who _knew_ that "drumming on a tin-pan did good,"
but I am inclined to think it had some effect. I have heard of several
instances where swarms were apparently stopped, by having earth thrown
among them, while passing over a field where men were at work. We know
they dislike being wet, as we see them hastening home on the approach
of a shower; or we can at any time drive them in the hive by sprinkling
them with water. Throwing water in the swarm is a kind of imitation
shower, and earth is something like it. Whether useful or not, these
swarms leaving the hive was rather suspicious, and I should try it
again under similar circumstances.


SOME COMPULSION.

After getting them in the hive for the fourth time, I resolved not to
be baffled or have much more such trouble, and perhaps go to the woods
at last, thereby setting a bad example. I put under the hive the
wire-cloth bottom-board, opened two or three holes on the top, and
covered these also with wire-cloth, (this was to let the air
circulate); a quantity of honey and water was given them and they were
then carried to the cellar, and kept prisoners four days, except half
an hour before sunset; when too late to leave for a journey, I set them
out to provide a few necessaries, and then returned them to the cellar.
In four days, when _honey enough_ is given them, a good swarm will half
fill an ordinary hive with combs. Some of the first eggs deposited will
be about hatching into larvæ, all of which would seem like too much to
leave. I now set them out, and gave them liberty; shading the hive,
&c., as before directed. They all proved faithful and industrious,
prospering like others. If their design was for a distant location,
they put a good face on the matter in the end.


HOW FAR WILL THEY GO IN SEARCH OF A HOME?

How far they will travel in search of a home, is also uncertain. I have
heard of their going seven miles, but could not learn how the fact was
proved. I have no experience of my own in this matter, but will relate
a circumstance that happened near me a few years since. A neighbor was
ploughing, when a swarm passed over him; being near the earth, he
"pelted them heartily" with the loose dirt he had ploughed up, which
seemed to bring them up, or rather down, as they clustered on a very
low bush; they were hived, and gave no further trouble. A man living
some three miles from this neighbor, on that day hived a swarm about
eleven o'clock, and left them to warm up in the sun as described a page
or two back; about three o'clock their stock of patience was probably
exhausted, when they resolved to seek a better shelter. They put off in
a great hurry, not even waiting to thank their owner for the spread on
his table, and the sweet-scented "yarbs" and good things with which he
had rubbed their hive. They gave him no notice whatever of their
intention to "quit," until they were moving! With all their goods ready
packed, they were soon under way, accompanied by their owner with
music; but whether they marched with martial precision, keeping time,
is uncertain. In this case the bees took the lead; the man with his
tin-pan music kept the rear, and was soon at a respectful distance.
They were either not in a mood, just then, to be charmed by melodious
sounds, or their business was too urgent to allow them to stop and
listen! Their means of locomotion being superior to his, he gave up in
despair, out of breath, after following about a mile. Another person,
about the same time in the day, saw a swarm moving in the same
direction of the first; he also followed them till compelled to yield
to their greater travelling facilities. A third discovered their flight
and attempted a race, but like the others soon came out behind. The
before-mentioned neighbor saw them, and thought of the fresh earth that
he had ploughed up, which he threw among them till they stopped. How
much farther they would have gone, if any, would be guessing. That it
was the same swarm that started three miles away, appears almost
certain; the direction was the same as seen by all, until they were
stopped; the time in the day also exactly corresponded.

We will now return to the issuing of the swarms. There will be some
emergencies to provide for, and some exceptions to notice.


TWO OR MORE SWARMS LIABLE TO UNITE.

If we expect to keep many stocks, the chances are that two or more may
issue at one time; and when they do, they nearly always cluster
together (I once knew an instance where only three stocks were kept;
they all swarmed and clustered together). It is plain that the greater
the number of stocks, the more such chances are multiplied.


DISADVANTAGE.

One first swarm, if of the usual size, will contain bees enough for
profit, yet two such will work together without quarrelling, and will
store about one-third more than either would alone; that is, if each
single swarm would get 50 lbs., the two together would not get over 70
lbs., perhaps less. Here, then, is a loss of 30 lbs., besides one of
the swarms is about lost for another year; because such double swarms
are not generally any better the next spring as a stock, and often not
as good as a single one. You will therefore see the advantage of
keeping the first swarms separate.


CAN OFTEN BE PREVENTED.

"Prevention is better than cure." We can, if we keep a good lookout,
often prevent more than one issuing at a time. This depends on our
knowledge of indications, in a great measure. I have said that before
starting to fly off, they were about the entrance in great numbers;
there may be one exception in twenty, where the first indications will
be a column of bees rushing from the hive. To take this matter a little
farther from the surface, we will take a peep at the interior; that is,
if our hives contain glass boxes, such as have been recommended. It is
an advantage to know which are about to cast their swarms, as long
beforehand as possible.


INDICATIONS OF SWARMING INSIDE THE HIVE.

These glass boxes are usually filled with bees; before leaving they may
be seen in commotion, long before any unusual stir is visible outside,
sometimes for near an hour. The same may be noticed in a glass hive.
Now, in good weather, when we have reason to expect many swarms, it is
our duty to watch closely, especially when the weather has been
unfavorable for several days previous. A number of stocks may have
finished their queen-cells during the bad weather, and be ready to come
out within the first hour of sunshine that occurs in the middle of the
day. We must expect some such occurrences, and in large apiaries there
is apt to be trouble, unless you take some precautions. If you have
taken no care (which but few will), by previous examinations, to know
which are ready, as soon as one has started or commenced flying, look
at all the rest that are in condition to swarm; or, what is much
better, look before any have started. Even if nothing unusual is seen
about the entrance, raise the cover to the boxes. If the bees in them
are all quiet as usual, no swarm need be immediately apprehended, and
you will probably have time to hive one or two first.


PREVENTING A SWARM ISSUING FOR A TIME.

But should you discover the bees running to and fro in great commotion,
although there may be but few about the entrance, you should lose no
time in sprinkling those outside with water from a watering-pot, or
other means. They will immediately enter the hive to avoid the supposed
shower. In half an hour they will be ready to start again, in which
time the others may be secured. I have had, in one apiary, twelve hives
all ready in one day, and did actually swarm; several of which would
have started at once, had they not been kept back with water, allowing
only one at a time, thus keeping them separate. They had been kept back
by the clouds, which broke away about noon.


TO PREVENT SWARMS UNITING WITH THOSE ALREADY HIVED.

When any of the subsequent swarms were disposed to unite with those
already hived, a sheet was thrown over to keep them out. I had four so
covered at once. An assistant, in such cases, is very important; one
can watch symptoms and keep them back, while the other hives the
swarms.

Occasionally, when ready for a swarm and waiting for one to start, two
may do so at once. Whenever a part have got on the wing, I never
succeeded in stopping the issue: consequently I have found it useless
trying to drive or coax them back in such cases. To succeed, the means
must be used in season, before any of the swarm leaves.


WHEN TWO HAVE UNITED, THE METHOD OF SEPARATING.

Two or more swarms will cluster together, and not quarrel, if put in
one hive; I have already told you the disadvantages. Unless business is
very urgent, your time cannot be better employed than in dividing them.
First, it is necessary to provide a good stock of patience, as it may
be a short job, or it may be a long one. Get two empty hives, and
divide the bees as nearly equal as possible. It is generally the best
way to spread a sheet on the ground, and shake the bees in the centre,
and set the hives each side of the mass, their edges raised to allow
the bees to enter; if too many are disposed to enter one hive, set it
farther off. If they cluster in a situation where they cannot be got to
the earth in a body, they must be dipped off as before directed, but,
in this case, putting a dipper full in each hive alternately, until all
are in. They should be made to hurry some in going in; keep the
entrance clear, and stir them up often; or sprinkle a very little water
on them, as they should not be allowed to stop their humming until all
are in. We have one chance in two of getting a queen in each. The two
hives should now be placed twenty feet apart; if there is a queen in
each, the bees in both will remain quiet, and the work is done; but if
not, the bees in the one destitute will soon manifest it by running
about in all directions, and, when the queen cannot be found, will
leave for the other hive, where there are probably two, a few going at
a time. Now there are two or three methods of separating these queens;
one is, to empty the bees out and proceed as before, a kind of chance
game, that may succeed at the next trial, and may have to be repeated.
Another way is, that, as soon as it is ascertained which is without a
queen, before many bees leave, spread down a sheet; set this hive on
it, and tie the corners over the top to secure the bees for the
present, turn the hive on its side for the present to give them air; or
it may be let down on a wire cloth bottom-board and the hole in the
side stopped, and this would be less likely to smother the bees, if it
could be secured to the bottom, and have the hive lie on its side; when
this division is secured, get another hive, and jar out those with the
queens; let them enter as before, and then set them apart, &c.,
watching the result; if the queens are not yet separate, it will be
known by the same appearances. The process must be continued till
separate, or the number with the queens may be easily looked over, and
one of them found; indeed, a sharp lookout should be kept up from the
beginning, and the queens caught, if possible.


NO DANGER OF A STING BY THE QUEEN.

No danger of her sting need be apprehended, as she will not demean
herself to use that for a common foe; she must have a _royal_
antagonist. When successful in obtaining one, it is sufficient; put her
in a tumbler or some safe place; then put your bees in two hives, place
them as directed, and you will soon learn where your queen is needed.
After all is done, the two hives should not be nearer than twenty feet,
at least the first day; perhaps forty would be still better. When two
swarms are mixed, and then separated, it is evident that a portion of
each swarm must be in both hives. A queen in each must of course be a
stranger to at least a part of the bees; these might, if their own
mother was too near, discover her, and leave the stranger for an old
acquaintance, and, in the act of going, call or attract the whole with
them, including the queen. I have known a few instances of the kind.


SOME PRECAUTIONS IN HIVING TWO SWARMS TOGETHER.

If you are disposed to separate them, but are afraid to work among them
to this extent in the middle of the day, or if there is danger of more
issuing, to mix with them, and add to your perplexity, of which you
already have enough, then you can hive them as a single swarm; but,
instead of a bottom-board, invert an empty hive and set the one with
the swarm on this, and insert a wedge between them, for ventilation. As
many bees are liable to drop down, in this case the lower hive will
catch them, and there is less danger of leaving. Let them remain till
near sunset, when another course may be taken to find a queen, though
by that time one is sometimes killed; yet it is well to know the fact.
Take them to some place out of the sun, as a less number will fly
during the operation.


HOW TO FIND QUEEN, WHEN TWO STRANGERS ARE TOGETHER.

First, look into the lower hive for a dead queen, and, if none is found
there, look thoroughly, as far as possible, for a little compact
cluster of bees, the size of a hen's egg, that may be rolled about
without separating. Secure this cluster in a tumbler; it is quite sure
one of the queens is a prisoner in the middle;[16] should two be seen,
get both. Then divide the bees, and give the one destitute, a queen;
or, if you have two, one to each, as the case may be. It would be well
first to see if the queen was alive, by removing the bees from about
her. But should you find nothing of the kind, spread a sheet on the
ground, shake the bees on one end of it, and set the hive on the other;
they will immediately begin a march for the hive. You may now see the
cluster, and may not; but they will spread out in marching, and give a
good chance to see her majesty, when a tumbler is the most convenient
thing to set over her. No matter if a few bees are shut up with her,
there is no risk, then, in your eagerness to get the queen, of taking
hold of a worker or two. A piece of window-glass can be slipped under,
and you have her safe, and by this time you will know what is to be
done next. This operation could not well be done in the middle of the
day, or in the sun, as too many bees would be flying, and greatly
interfere.

      [16] All stranger queens, introduced into a stock or swarm, are
      secured and detained in this manner by the workers, but whether
      _they_ dispatch them, or this is a means adopted to incite them
      to a deadly conflict, writers do not agree, and I shall not
      attempt a decision, as I never saw the bees voluntarily release a
      queen thus confined. But I have seen queens, when no bees
      interfered, rush together in a fatal rencounter, and one of them
      was soon left a fallen victim of the contest. 'Tis said it
      _never_ happens that both are killed in these battles,--perhaps
      not. As I never saw _quite all_ of these royal combats, of course
      I cannot decide.

Should you fail in finding a queen, and cannot succeed in making a
division in consequence, or should you resolve, from want of time,
patience or energy, to let them remain together in the beginning, it is
unnecessary to get a hive any larger than usual for two swarms; they
will certainly find room by cold weather: if more than two, they
_should_ be divided by all means; it will be a disadvantage for another
year. For the first four days, when two large swarms are together, it
is necessary to keep an inverted hive under them, but much longer it
would not do, as they might extend their combs into the lower hive.


BOXES FOR DOUBLE SWARMS IMMEDIATELY.

It should then be taken out, and boxes immediately put on, which should
be changed for empty ones, as fast as they are filled. Yet, this extra
honey is not quite as much advantage as increase of stocks; when that
is an object, I will recommend another disposition.


RETURNING A PART TO THE OLD STOCK.

Separate one-third or more of the two swarms, being sure there is no
queen with this part, (by the test given of setting them at a distance)
and then return them to one of the old stocks; they will immediately
enter without contention, and issue again in about nine days, or as
soon as a young queen is matured to go with them. There may be an
exception to this, of one in twenty. I would have recommended this
course in all cases of the kind, but there will be a loss of time for
the bees in the old stock; because they are apt to be rather idle, even
when they might labor in the boxes; and here there is a loss of some
eight or ten days. The collections of a good swarm may be estimated at
least one pound a day, (often two or three.) A swarm that just fills
the hive, would make at least ten pounds box-honey, if it could have
been located ten days earlier. Still another method may be adopted when
you have a very small swarm, one that is not likely to fill the hive,
and has not been hived more than two or three days. A third of your two
swarms may be put in with that; taking care, as before, not to let your
only queen go with them.


METHOD OF UNITING.

The manner of doing it is very simple; get them in a hive as before
directed, and jar them out in front of the one you wish them to enter,
or invert it, setting the other over, and let them run up.


WHEN CARE IS NECESSARY.

Except on the day of swarming, care is necessary not to introduce a
small number with a large swarm; they are liable to be destroyed. The
danger is much greater than to put together about an equal number, or a
large number put in with a few. The day that swarms issue, they will
generally mix peaceably, but in proportion as time intervenes between
the issues, so will be the liability to quarrel. Yet, I have united two
families of about equal numbers in the fall and spring, and, with a few
exceptions, have had no difficulty.


SWARM-CATCHER.

There is another method of keeping swarms separate, contrived and used
by a Mr. Loucks, of Herkimer Co., N.Y. He calls it a swarm-catcher; he
has a half dozen of them, and says he would not do without for one
season, for fifty dollars, as he has a large apiary. I made one as near
as I could from seeing his, without taking the exact measure. I got out
four light posts four and half feet long, one inch square; then twelve
pieces of one-quarter inch stuff, four inches wide; the four for the
top twelve inches long, for the bottom two were fourteen inches long,
and two were twenty. These were thoroughly nailed on the ends of the
posts, making it into an upright frame, the other four pieces were
nailed around the middle, which made the frame firmer. I made a frame
for the top, of four pieces, each an inch and a half in width, and half
inch thick, halved at the ends and nailed together, and fastened by
hinges to one side of the top, and a catch to hold it shut. The whole
was now covered with very thin cloth to admit the light, but not so
open as to let the bees through, (Mr. Loucks used cloth made for
cheese-strainers.) I now had a covered frame four and half feet high,
12 inches square at the top, at the bottom 14 by 20, with a door or lid
at the top, to let out the bees. On each side of the bottom I tacked a
piece of common muslin, near a yard in length. When a swarm is ready to
issue, the bottom of this frame is set up before the hive, one edge of
the bottom rests on the bottom-board, the other against the side of the
hive; the top sets off from the hive at an angle of about 45 degrees,
under which a brace is set to hold it. The muslin at the bottom is to
wrap around the hive at the side to prevent the escape of the bees. The
swarm rushes into this without any hesitation.

When done coming out, the muslin at the bottom is drawn over it, and
the frame is set in an upright position, and allowed to stand a few
minutes for the bees to get quiet in the top. It is now to be laid on
its side, the door opened, and the bees hived. In the few trials that I
have given it, I succeeded without difficulty. But I would remark, that
stocks from which swarms are caught in this way, must not be raised at
the back side, as a part of the swarm would issue there, and not get
into the net. Mr. Loucks had his hive directly on the board; and he
told me he kept them so through the season: the only places of entrance
was a sprout out of the bottom of the front side, about three inches
wide by half inch deep, and a hole in the side a few inches up. You
will thus perceive that stocks from which swarms are hived in this way
must be prepared for it previously. Also, it will be no use to such
bee-keepers as depend on seeing their swarms in the air. It will be
beneficial only in large apiaries, where several swarms are liable to
issue at once; the swarming indications well understood, and the
apiarian on the lookout.


SWARMS SOMETIMES RETURN.

Occasionally a swarm will issue, and in a few minutes return to the old
stock. Mr. Miner gives a cause for this, very ingenious, and romantic,
but unfortunately there are but few facts to sustain this hypothesis,
(at least I have not discovered them.) There are other causes that
appear to me more reasonable; the most common is the inability of the
old queen to fly, on account of her burden of eggs, old age, or
something else. I have sometimes, after the swarm had returned found
the queen near the stock, and put her back, and the next day she would
come out again, and fly without difficulty, (perhaps she had discharged
some of her eggs.)

Their returning is more frequent in windy weather, or when the sun is
partially obscured by clouds. About three-fourths of them will not
re-issue until a young queen is matured, eight or ten days afterwards;
and a few, not at all. But when the queen returns with the swarm, they
usually come out again the next day, or day after, and some not till
the third or fourth. I have known two instances where they issued again
the same day.


REPETITION PREVENTED.

Sometimes a swarm will issue and return three or four days in
succession, but this I generally remedy, as it is often owing to some
inability of the queen, and she may be frequently found while the swarm
is leaving outside the hive, unable to fly. In such cases it is only
necessary to have a tumbler ready, and watch for her; and as soon as
she appears, secure her, get the empty hive for the swarm, a sheet, and
put down a bottom-board a few feet from the stock. The swarm is sure to
come back; the first bees that alight on the hive will set up the call;
as soon as this is perceived, lose no time in setting the old stock on
the board, and throwing the sheet over it to keep out the bees. Put the
new one in its place on the stand, and the queen in it; in a few
minutes the swarm will be in the _new_ hive, when it can be removed,
and the old one replaced. This I have done many times. But should the
swarm begin to cluster in a convenient place, when you have so caught
the queen, by being expeditious she may be put with the swarm, before
they have missed her and may be hived in the usual way.


LIABILITY TO ENTER WRONG STOCKS.

In all cases, whether you set a new hive in place of the old one or
not, whenever a swarm returns, if other stocks stand close on each
side, they are quite sure to receive a portion of the bees--probably a
few hundreds; these are certain to be massacred. To prevent which, it
is necessary to throw sheets over them until the swarm has gathered on
their own hive. This is another reason for plenty of room between
stocks. Should no queen be discovered during their issue, or return,
she should be sought for in the vicinity of the hive, and put back if
found, and the swarm will be likely to issue several days earlier, than
to wait for a young queen.

When the old queen is actually lost, and the bees have returned to wait
for a young one, it is often ready to leave one or two days short of
the time required for second swarms. Whether a greater number of bees
in the old stock creating more animal heat, matures the chrysalis in
less time than a stock thinned by casting a swarm, or some other cause,
I cannot say. I mention it because I have known it to occur frequently,
but not invariably. A swarm flying, unaccompanied by a queen, is
scattered more than usual.


FIRST ISSUES GENERALLY CHOOSE FAIR WEATHER.

First swarms are commonly more particular as to weather than after
swarms. They have several days from which to choose, after these royal
cells are ready, and before the queens are matured; and they usually
take a fair one. But here again are exceptions. I once had two first
swarms issue in a wind that kept every branch of tree and bush in
agitation to such a degree that it was impossible to find any such
place to cluster. I expected their return to the old hive; but here
were more exceptions. After repeating a fruitless attempt at the
branches, they gave it up, and came down amongst the grass on "terra
firma." This occurred after several days of rainy weather. The next day
being pleasant, twelve issued; almost proving that the wind the
preceding day kept back a part. I also knew one to issue in a shower,
that beat many of them to the ground before they could cluster. In this
case the shower was sudden, the sun shone almost up to the time it
began to rain. About this time the swarm started when it seemed they
were unwilling to turn about.


AFTER SWARMS.

After swarms are second and third issues (or all after the first) from
a stock; and quite a different affair from the first, as also are some
first swarms, when the old queen has been lost, being led out by young
queens.


THEIR SIZE.

Second swarms are usually half as large as the first, the third half as
large as the second, the fourth still less; with some variations. I
give general features, noticing only the exceptions that occur most
frequently; others sometimes happen, but so seldom that mentioning them
is deemed unnecessary.


TIME AFTER THE FIRST.

Whenever the first swarm in a prosperous season _was not kept back by
foul weather_, the first of the young queens in the old stock is ready
to emerge in about eight days. We will suppose the first swarm issued
on Sunday; a week from the next Tuesday will be usually as soon as the
second one need be expected.


PIPING OF THE QUEEN.

On the Monday evening previous, or on Tuesday morning, by putting your
ear close to the hive, and listening attentively five minutes, you will
hear a distinct piping noise, like the word _peep, peep_, uttered
several times in succession, and then an interval of silence; two or
more may be often heard at the same time; that of one will be shrill
and fine, of another hoarse, short and quick. This piping is easily
heard by _any_ one not actually deaf, and not the least danger of its
being taken for any humming; in fact, it is not to be mistaken for
anything else _but piping_, even when you hear it for the first time.
These notes can probably never be heard except when the hive contains a
plurality of queens.


MAY ALWAYS BE HEARD BEFORE AND AFTER SWARM.

I _never failed to hear it_, previous to a second swarm, or any after
the first, whenever I listened; and whenever I have listened and not
heard it at the proper time, I never knew a second swarm to issue!


TIME OF CONTINUANCE VARIES.

The time of commencing will be later than this rule in some stocks, if
the weather is cool, or not many bees left; it may be ten or twelve
days. I once found it fourteen before I heard it. Also the swarm may
not issue in two or three days after you hear it. The longer the swarm
delays, the louder will be the piping; I have heard it distinctly
twenty feet, by listening attentively when I knew one was thus engaged;
but at first it is rather faint. By putting your ear against the hive
it may be heard even in the middle of the day, or at any time before
issuing. The length of time it may be heard beforehand seems to be
governed again by the yield of honey; when abundant it is common for
them to issue the next day; but when somewhat scarce, they will be much
longer--very often three or four days. In these cases third swarms
seldom occur.


TIME BETWEEN SECOND AND THIRD ISSUES.

Piping for third swarms (when they issue) may usually be heard the
evening after the second has left, though one day commonly intervenes
between their issues.

Here my experience is at variance with many writers, who give several
days between the second and third. I do not recollect an instance of
more than three days between, but many in less, several the next, and
one the same day of the second! I had an instance of a swarm losing its
queen (the old one) on its first sally, and returned to wait for the
young ones; when they were ready, an uncommon number of bees were
present; three swarms issued in three days! On the fourth, another came
out and returned; the fifth day it left; making four regular swarms in
five days. On the eighth, the fifth swarm left! Although I never had
five swarms from a stock before, yet I expected this, from the fact of
hearing the piping on the next evening after the fourth one had left.
The piping had continued in this hive from the evening previous to the
first swarm till the last one had left.


NOT ALWAYS TO BE DEPENDED UPON.

One stock in fifteen may commence piping, yet send out no swarm. The
bees will change their minds about coming out, and kill their queens,
or allow the eldest one of them to destroy the others, or some other
way, as they do not always swarm in such circumstances. But when the
piping continues over twenty-four hours, I never knew _but one
failure_! I have known a few (two or three) to commence this piping,
while I supposed the old queen was yet present, and had not left the
hive, on account of bad weather, but a swarm issued soon after. Also,
three instances where I supposed the old queen lost, from some other
cause than leading out a swarm, and the stock reared some young ones to
supply her place. It occurred in or near the swarming season, and one
or two issues was the consequence. One case was three weeks in advance
of the season, and the swarm was about half the usual size. When a
swarm has been out, and returned at the last of the swarming season, it
is much more probable to re-issue, than if it depended on an old queen
for a leader, that had not been out. Such will sometimes be a week or
ten days later than others. Once I had the first swarm kept back by wet
weather, and the second came out on the fifth day after; several other
instances on the seventh and eighth; and one as late as the sixteenth,
after the first.


A RULE FOR THE TIME OF THESE ISSUES.

This may be put down as a rule, that all after swarms _must_ be out by
the eighteenth day from the first. I never found an exception, unless
the following may be considered so: When a swarm left the middle of
May, and another the first of July, seven weeks after, but two cases of
this kind have come up, and these I consider rather in the light of
first swarms, as they leave under the same circumstances, leaving the
combs in the old stock filled with brood, queen-cells finished, &c. A
stock may cast swarms in June, and a buckwheat swarm in August, on the
same principle.


WHEN IT IS USELESS TO EXPECT MORE SWARMS.

Therefore, bee-keepers having but few stocks, will find it unnecessary
to watch their bees when the last of the first swarms came out sixteen
or eighteen days before. Much trouble may be thus saved by
understanding this matter. During my early days in beekeeping, I wished
for the greatest possible increase of stocks. I had some that had cast
the first swarm, and soon after clustered out again. I vainly watched
them for weeks and months, expecting another swarm. But had I
understood the _modus operandi_, as the reader may now understand it, I
should have been through with all my anxiety, as well as watching, in a
fortnight. As it was, it lasted two months. I found no one to give me
any light on this subject, or even tell me when the swarming season was
over, and I came very near watching all summer!


PLURALITY OF QUEENS DESTROYED.

When the bees, queens, or all together, decide that no more swarms are
to issue, the plurality of queens is destroyed, and but one is left. It
is probable that the oldest and strongest queen dispatches the others,
generally while in the cells.

I once had some artificial queens reared, as an experiment, from common
eggs, on the top of a hive, in a small glass box, where there was room
for but one comb, which allowed me to see all particulars.


THE MANNER.

After the first queen was matured, and had left her cell, I caught her
within six hours, taking advantage of her younger sisters, which were
yet sealed up, and of course could offer no resistance. She first made
an opening that would allow her to reach the abdomen of her competitor
(probably this is the most vulnerable). As soon as this was
sufficiently large to admit her body, she thrust it in, inflicting the
fatal sting. This was then left for another, that soon shared the same
fate. If quick and spiteful movements are any indications of hatred, it
was manifested here very plainly. The bees enlarged the orifice and
dragged out the now dead queens.

Now, if I should say that all queens were dispatched in this way,
merely because I witnessed it in this case, it would be carrying out
the principle I am endeavoring to avoid: that is, judging all cases
from one or two solitary facts. As it is, it is somewhat confirmatory
of what some others have said. I will suppose, then, until further
evidence contradicts it, that the first perfect queen leaving her cell,
makes it her business to destroy all rivals in their cradle, as soon as
it is decided that no more swarms shall issue. By keeping grass, weeds,
&c., away from about the stock, these dead queens, as they are brought
out, may be frequently found. Such as are removed during the night may
be often found on the floor-board in the morning. I have found a dozen
by one stock. Should the stock send out but one swarm, they may be
found about the time, or a little before you would listen for the
piping. But should after swarms come out, they will, or may be found
the next morning after it is decided that no more are to issue. It is
very seldom that all the queens reared are needed. They make it a rule,
as far as they have control, to go on safe principles, by having a
little more than just enough. When several such bodies are thrown out,
and no piping is heard, no further swarming need be expected. But
should you hear the piping a day or two after finding a dead queen, you
may yet look for the swarm.


THEORY DOUBTED.

It is stated that when the bees decide an after swarm shall issue, the
first queen matured is not allowed to leave her cell, but kept a
prisoner there, and fed until wanted to go forth with the swarm. This
may be true in some cases (though not satisfactorily proved), but I am
quite sure it is not in all.

When she is confined to her cell, how does she ascertain the presence
of others? By leaving the cell, this knowledge is easily obtained.
Huber says she does, and is "enraged at the existence of others, and
endeavors to destroy them while yet in the cell, which the workers will
not allow; this is so irritating to her majesty that she utters this
peculiar sound." Also second and third swarms may contain several
queens, frequently two, three, and four; even six at one time come out.
If these had to bite their way out, after the workers had decided it
was time to start (for it _must be they_ decide it when the queens are
shut up), they would hardly be in season.


AFTER SWARMS DIFFERENT IN APPEARANCE FROM THE FIRST WHEN ABOUT TO
ISSUE.

Another thing, when after swarms start, the appearance about the
entrance is altogether different from first ones, unless there is an
unusual number of bees. I have said that for a little time beforehand,
that such were in an apparent tumult, &c. But after swarms seldom give
any such notice. One or more of the young queens may sometimes be seen
to run out, and back, several times in a few minutes, in a perfect
frenzy; sometimes fly a short distance, and return before the swarm
will get started (which she could not do if confined). The workers seem
more reluctant about leaving than in first swarms, when a mother
instead of a sister is leader. Even after the swarm is in motion, she
may return and enter the hive a moment. No doubt she finds it necessary
to animate or induce as many as possible to leave with her. A person
watching the issue of a second swarm under these circumstances, for the
first time, and finding the queen leaving first, would very likely
_guess_ all must be alike. Perhaps the next one would be different; the
first thing seen might be the swarm leaving, and no queen discovered at
all. But to return to the imprisonment of the queens. I have one other
fact in objection. I once saw a queen running about in a glass hive,
while they were piping for a second swarm. She was near the glass,
appeared agitated, stopping occasionally to vibrate her wings, which
was simultaneous with the piping, and seemed to make it. The workers
appeared to take but little notice of her. The next day the swarm left.
Here was one instance, at least, of her not being confined till the
time of leaving, making an exception, if not a rule. Let this matter be
as it may, I admit it makes but little difference to the practical
apiarian, either way; but to the reader whose interest is the natural
history of the bee, the truth is important.


TIME OF DAY, WEATHER, ETC.

These after swarms are not very particular about the weather; heavy
winds, a few clouds, and sometimes a slight sprinkling of rain, will
not _always_ deter them. Neither are they very precise about the time
of day. I have known them in a warm morning to issue before seven
o'clock, and after five P.M. These things should be understood;
because, when after swarms are expected (of which the piping will give
warning), it is necessary to watch them in weather, and at times when
first ones would not venture to leave.


SWARMS NECESSARY TO BE SEEN.

It is essential that you see them, that you may know where they
cluster, otherwise it might be difficult to find them. They are apt to
go farther from the parent stock than others; sometimes fifty rods, and
then settle in two places, perhaps that distance apart, in some high or
inconvenient place to get at. (Let me not be misunderstood: I do not
say they all do so, or even the majority; but I wish to say that a
greater portion of these swarms do so than of the first.) If they
cluster in two places, a queen may be in each, and they will remain,
and when you have hived one part you may think you have all. If one
cluster is without a queen, they will join the other if near; but when
distant, will be very likely to return to the old stock soon, unless
put together. I had a swarm light in two places, in exactly opposite
directions from the stock. In one, a good swarm had clustered; in the
other, some less than a pint. The small part had one or more queens,
the other none. It was perceived at once by their movements. Now, if we
provide a hive for a swarm, and get a few to set up the call or
buzzing, they will not leave till that is stopped. There is generally
no difficulty to start it. The surest way is to jar a portion or all
directly into the hive. It takes a few minutes to get composed, and
miss the queen. In my case I got them in the hive, and before they
missed the queen, carried them to the small cluster, which I got in a
dipper and emptied in front of the hive; they entered, and all were
peaceable. You will therefore see the necessity of watching such
swarms, to see if there is no separation, if nothing else.


RETURNING AFTER SWARMS TO THE OLD STOCK.

Much has been said about returning all after swarms to the old stock;
the advantages of which will depend on the time of issuing; whether
late or early, the yield of honey, etc. It would be unusual to have
many after swarms without a liberal yield of honey, for the time being;
but to tell of its continuance is the question to be answered. Second,
and even third swarms, if early in the season, and the honey continues
plentiful, may be hived, and these, together with the old stock, will
prosper. Here the apiarian needs a little judgment and experience to
guide him.


WHEN THEY SHOULD BE RETURNED.

It is always best, if possible, to have good strong families. When
after swarms are late it is safest to return them, as the old stock
will need them to replenish the hive, and prepare for winter. Also a
less number of worms will infest it, when well provided with bees; and
the chances of box honey are greater.


METHOD OF DOING IT.

But the process of returning such requires some little patience and
perseverance. I have said there may be a dozen young queens in the old
stock. Now suppose one, two, or more leave with the swarm, and you
return the whole together, there is nothing to prevent their leading
out the swarm again the next day. Therefore it is policy to keep the
queens back. The least trouble is to hive in the usual way, and let
them stand till the next morning. It will save you the trouble of
looking for more than one, if there should be more, for all but that
are destroyed by that time. There is a chance, also, for the old stock
to decide that no more should issue, and allow all but one to be slain
there. When this is the case, and you find the one with the swarm, you
will have no further trouble by their re-issuing. They should be
returned as soon as the next morning, otherwise they might not agree,
even when put in the old home. To return them, and find a queen easily,
get a wide board a few feet long; let one end rest on the ground, the
other near the entrance, that they may enter the hive without flying;
then shake out the swarm on the lower end of the board; but few will
fly, but soon commence running up towards the hive; the first one that
discovers the entrance will set up the call for the others. If they do
not discover it, which is the case sometimes, scatter some of them near
it, and they will soon commence marching up, when you should look out
for, and secure the queen, as they spread and give a good chance. By
applying your ear to the hive, the piping will tell you if they are to
issue again. It is evident, if you follow these directions, that the
swarm cannot issue many times before their stock of royalty will be
exhausted; and when but one queen remains the piping will cease, and no
further trouble will be had. To prevent these after swarms, some
writers recommend turning over the hive and cutting out all the royal
cells but one. This I have found impracticable with a great many
stocks. Some of the cells are too near the top to be seen, consequently
this cannot always be depended upon. As for a rule about returning, it
is somewhat difficult to give one. If I should say, return all such as
issue after the 20th of June, the variation in the season might be two
or three weeks, even in the same latitude; i.e., the course of flowers
that had bloomed by that date in one season might, another year,
require two weeks more to bring out. Also, the 20th of June, in
latitude of New York City, is as late as the 4th of July in many places
further north. I once had a second swarm on the 11th of July, that
wintered well, having nearly filled the hive. Yet, in some seasons, the
first swarms, of the last of June, have failed to get enough. In
sections where much buckwheat is raised, late swarms do more towards
filling their hives than where there is none.


MORE CARE NEEDED BY AFTER SWARMS WHEN HIVED.

Should it be thought best to hive after swarms, and risk the chances,
they should receive a little extra attention after the first week or
two, to destroy the worms; a little timely care may prevent
considerable injury. They are apt to construct more combs in proportion
to the number of bees, than others; consequently, such combs cannot be
properly covered and protected. The moth has an opportunity to deposit
her eggs on them, and, sometimes, entirely destroy them.


TWO MAY BE UNITED.

Whenever these swarms issue near enough together, it is best to unite
them. I have said second swarms were generally half as large as the
first. By this rule, two second swarms would contain as many bees as a
first one, and four of the third, or one of the second issue, and two
of the third, &c. If the first and second are of the ordinary size, I
think it advisable always to return the third. But in large apiaries it
is common for them to issue without any previous warning, just when a
first one is leaving, and crowd themselves into their company, and
seeming to be as much at home as though they were equally respectable.

Whenever the hives containing our swarms are full or very near it, the
boxes should be put on without delay, unless the season of honey is so
nearly gone as to make it unnecessary.

I have found it an advantage to hive a few of these very small swarms,
on purpose to preserve queens, to supply some old stocks that sometimes
lose their own at the extreme end of the swarming season. The cases to
be mentioned at the last of the next chapter. I try and save one for
about every twenty stocks that have swarmed.



CHAPTER XIV.

LOSS OF QUEENS.


OF SWARMS THAT LOSE THEIR QUEEN.

Swarms that lose their queen the first few hours after being hived,
generally return to the parent stock; with the exception that they
sometimes unite with some other. If much time has elapsed before the
loss, they remain, unless standing on the same bench with another. On a
separate stand they continue their labor, but a large swarm diminishes
rapidly, and seldom fills an ordinary-sized hive. One singular
circumstance attends a swarm that is constructing combs without a
queen. I have never seen it noticed by any one, and may not always be
the case, but _every_ instance that has come under my notice, I have so
found it. That is, four-fifths of the combs are drone-cells; why they
thus construct them is another subject for speculation, from which I
will endeavor in this instance to refrain.


A SUGGESTION AND AN ANSWER.

It has been suggested as a profitable speculation, "to hive a large
swarm without a queen, and give them a piece of brood-comb containing
eggs, to rear one, and then as soon as it is matured, deprive them of
it, giving them another piece of comb, and continue it throughout the
summer, putting on boxes for surplus honey. The bees having no young
brood to consume any honey, no time will be lost, or taken to nurse
them, and as a consequence they will be enabled to store large
quantities of surplus honey."

This appears very plausible, and to a person without experience
somewhat conclusive. If success depended on some animal whose lease of
life was a little longer, it would answer better to calculate in this
way. But as a bee seldom sees the anniversary of its birthday, and most
of them perish the first few months of their existence, it is bad
economy. It will be found that the largest amount of our surplus honey
is obtained from our prolific stocks. Therefore it is all-important
that every swarm and stock has a queen to repair this constant loss.


A DISPUTED QUESTION.

We now approach another disputed point in natural history, relative to
the queen leaving at any time except when leading out a swarm. Most
writers say that the young queen leaves the hive, and meets her
paramour, the drone, on the wing. Others deny this _positively_, having
watched a whole summer without seeing her highness leave. Consequently
they have arrived at the very plausible and apparently consistent
conclusion, that nature never intended it to be so, since it must
happen at a time when the existence of the whole family depends
entirely on the life of the queen. The stock at such times contains no
eggs or larvæ, from which to rear another, if she should be lost. "The
chances at such times of being devoured by birds, blown away by the
winds, and other casualties, are too many, and it is not probable the
Creator would have so arranged it." But facts are stubborn things; they
will not yield one jot to favor the most "finely-spun hypothesis;" they
are most provokingly obstinate, many times. When man, without the
necessary observation, takes a survey through animated nature, and
finds with scarcely an exception that male and female are about equal
in number, he is ready, and often does conclude that one bee among
thousands cannot be the only one capable of reproduction or depositing
eggs. Why, the idea is preposterous! And yet only a little observation
will upset this very consistent and analogous reasoning. So it appears
to be with the excursions of the young queens. I was compelled, though
reluctantly, to admit that they leave the hive. That their purpose is
to meet the drones, I cannot at present contradict. Also, that, when
the queen is once impregnated, it is operative for life, (yet it is
another anomaly), as I never detected her coming out again for that
purpose. What then is the use of the ten thousand drones that never
fulfil this important duty? It seems, indeed, like a useless waste of
labor and honey, for each stock to rear some twelve or fifteen hundred,
when perhaps but one, sometimes not any of the whole number is of any
use. If the risk is great in the queen's leaving, we find it arranged
admirably in its not being too frequent.


A MULTITUDE OF DRONES NEEDED.

Instinct teaches the bee to make the matters left to them as nearly
_sure_ as possible. When they want one queen, they raise half a dozen.
If one drone or only half a dozen were reared, the chances of the queen
meeting one in the air would be very much reduced. But when a thousand
are in the air instead of one, the chances are a thousand times
multiplied. If a stock casts a swarm, there is a young queen to be
impregnated, and be got safely back, or the stock is lost. Every time
she leaves, there is a chance of her being lost, (one in fifteen). If
the number of drones was any less than it is, the queen would have to
repeat her excursions in proportion, before successful. As it is, some
have to leave several times. The chances and consequences are so great,
that on the whole no doubt but it is better to rear a thousand
unnecessarily, than to lack one just in time of need. Therefore let us
endeavor to be content with the present arrangement, inasmuch as we
could not better it, and probably had we been consulted, would have so
fixed "the thing, that it would not go at all."

But what is the use of the drones in hives that do not swarm, and do
not intend it, situated in a large room or very large hives? Such
circumstances seldom produce swarms, yet as regular as the return of
summer, a brood of drones appear. What are they for? Suppose the old
queen in such hive dies, leaving eggs or young larvæ, and a young queen
is reared to supply her place. How is she to be impregnated without the
drones? Perhaps they are taught that whenever they can afford it, they
should have some on hand to be ready for an emergency. I have already
said when bees are numerous, and honey abundant, they never fail to
provide them. I once put a swarm in a glass hive. The queen was a
cripple, having lost one of her posterior legs; in two months after she
was replaced by one young and perfect. Here was an instance of drones
being needed, when no intention of swarming was indicated; the hive was
but little more than half full.


THE QUEEN LIABLE TO BE LOST IN HER EXCURSIONS.

This excursion of the queen, whenever I have witnessed it, always took
place a little after the middle of the day, when the drones were out in
the greatest numbers. At such times I have seen them leave amid rather
more commotion than usual among the workers. I have watched their
return, which varied from three minutes to half an hour, and seen them
hover around their own hive, apparently in doubt whether they belonged
in that, or the next; in a few instances they have actually settled on
the neighboring hive, and would have there perished, but for my
assistance in putting them right.


THE TIME WHEN IT OCCURS.

Thus we see that queens are lost on these occasions from some cause,
and part of them by entering the wrong hive, perhaps most of them; if
so, it is another good reason for not packing stocks too close. The
hives are very often nearly alike in color and appearance. The queen
coming out for the first time in her life, is no doubt confused by this
similarity.

The number of such losses in a season has varied: one year the average
was one in nine, another it was one in thirteen, and another one in
twenty. The time from the first swarm also varies from twelve to twenty
days. The inexperienced reader should not forget that it is the old
stocks which have cast swarms, where these accidents happen; the old
queen having left with the first swarm. Also all after swarms are
liable to the same loss. I would suggest that these have abundant room
given between the hives; if it is necessary to pack close, let it be
the first swarms, where the old queen has no occasion to leave. Having
never seen this matter fully discussed, I wish to be somewhat
particular, and flatter myself that I shall be able to direct the
careful apiarian how to save a few stocks and swarms annually, that is,
if he keeps many. A few years ago, I wrote an article for the Albany
Cultivator. A subscriber of that paper told me a year afterwards that
he saved two stocks the next summer by the information; they were worth
at least five dollars each, enough to pay for his paper ten years or
more.

When a stock casts but one swarm, the queen having no competitors to
interfere with her movements, will leave in about fourteen days, if the
weather is fair; but should an after swarm leave, the oldest of the
young queens will probably go with that, of course: then, it must be
later before the next is ready: it may be twenty days, or even more;
those with after swarms will vary from one to six. It _always must_
occur when no eggs or larvæ exist, and no means left to repair this
loss; a loss it is, and a serious one; the bees are in as much trouble
as their owner, and a great deal more, they seeming to understand the
consequences, and he, if he knows nothing of the matter, has no
trouble. Should he now, for the first time, learn the nature of it, he
will at the same time understand the remedy.


INDICATIONS OF THE LOSS.

The next morning after a loss of this kind has occurred, and
occasionally at evening, the bees may be seen running about in the
greatest consternation, outside, to and fro on the sides. Some will fly
off a short distance and return; one will run to another, and then to
another, still in hopes, no doubt, of finding their lost sovereign! A
neighboring hive close by, on the same bench, will probably receive a
portion, which will seldom resist an accession under such
circumstances. All this will be going on while other hives are quiet.
Towards the middle of the day, this confusion will be less marked; but
the next morning it will be exhibited again, though not so plainly, and
cease after the third, when they become apparently reconciled to their
fate.

They will continue their labors as usual, bringing in pollen and honey.
Here I am obliged to differ with writers who tell us that all labor
will now cease. I hope the reader will not be deceived by supposing
that because the bees are bringing in pollen, that they _must_ have a
queen; I can assure you it is not always the case.


THE RESULT.

The number of bees will gradually decrease, and be all gone by the
early part of winter, leaving a good supply of honey, and an extra
quantity of bee-bread, as before mentioned, because there has been no
young brood to consume it. This is the case when a large family was
left at the time of the loss. When but few bees are left, it is very
different; the combs are unprotected by a covering of bees; the moth
deposits her eggs on them, and the worms soon finish up the whole. Yet
the bees from the other stocks will generally first remove the honey.


AGE OF BEES INDICATED.

Hundreds of bee-keepers lose some of their stocks in this way, and can
assign no reasonable cause. "Why," say they, "there wasn't twenty bees
in the hive; it was all full of honey," or worms, as the case may be.
"Only a short time before, it was full of bees; I got three good swarms
from it, and it always had been first rate, but all at once the bees
were gone. I don't understand it!" Such bee-keepers cannot understand
how rapidly a family of bees diminish, when there is no queen to
replenish with young this mortality of the old ones. I doubt whether
the largest and best family possibly could be made to exist six months,
without a queen for their renewal, except, perhaps, through the winter.

When standing close on one bench, they are gone sooner than if on
separate stands, as they often join a neighboring hive when they can
walk to it.


NECESSITY OF CARE.

As this tumult cannot be seen but a few days at most, it is well--yes,
it is necessary--to make it a duty to glance at the hives at this
period after swarming, _every morning_; a glance is sufficient to tell
you of the fact. Remember to reckon from the date of the first issue;
this occurs when the first royal cells are sealed over, and is the best
criterion as to when the queen will leave. If the first swarm issue and
return, it can make no difference; reckon from their first issuing.


REMEDY.

When you discover a loss, first ascertain if there is any after swarm
to be expected from another stock, (by listening for the piping); if
so, wait till it issues, and obtain a queen from that for your stock;
even if there is but one, take it, and let the bees return; they would
be likely to come out again the next day; if not, it is very often no
great loss.

Should no such swarm be indicated, go to a stock that has cast a first
swarm within a week; smoke it and turn it over, as before directed,
find a royal cell, and with a broad knife cut it out, being careful not
to injure it. This must now be secured in the other hive in its natural
position, the lower end free from any obstacle, that would interfere
with the queen leaving it. It will make but little difference whether
at the top or bottom, providing it is secure from falling.

I generally introduce it through a hole in the top, taking care to find
one that will allow the cell to pass down between two combs. It being
largest at the upper end, the combs each side will sustain it, and
leave the lower end free. In a few hours the bees will secure it
permanently to the combs with wax. This operation cannot be performed
in a chamber hive, as it is impossible to see the arrangement of the
combs through the holes. To put it in at the bottom is some more
trouble; the difficulty is, to fasten it, and prevent it resting on the
end. I have done it as follows: Get an _old_ thick piece of dry comb
some three inches square; cut out an inch of the middle. At right
angles with this, in one edge in the centre, make another to intersect
it, just the size of the cell, and have the lower end reach into the
opening. This comb will keep it in the right position, and may rest on
the floor-board. It can now be put in the hive, cutting out a piece of
comb to make room for it if necessary.

Soon after such cell is introduced, the bees are quiet. In a few days
it hatches, and they have a queen as perfect as if it had been one of
their own rearing. This queen of course will be necessitated to leave
the hive, and will be just as liable to be lost, but no more so than
others, and must be watched the same. It is unnecessary to look for a
cell in a stock that has cast its first swarm more than a week before,
as they are generally destroyed by that time, (sometimes short of it,)
unless they intend to send out an after swarm.


MARK THE DATE OF SWARMS ON THE HIVE.

Should you have so many stocks that you cannot remember the date of
each swarm without difficulty, it is a good plan to mark the date on
one side or corner of the hive, as it issues. You can then tell at once
where to look for a cell when wanted.

It will sometimes happen that a queen may be lost at the extreme end of
the swarming season, when no other stock contains such cells. I then
look around for the poorest stock or swarm that I have on hand, one
that I can afford to sacrifice, if it possesses a queen, to save the
one that has sustained this loss; this is not often the case, but is
sometimes. I have a few times put just bees enough with the queen to
keep her in a box, and kept them for this purpose, as was mentioned in
the last chapter. When introduced, the bees are generally killed, but
the queen is preserved.


OBTAINING A QUEEN FROM WORKER BROOD.

There is yet another method to be adopted, and that is, to obtain a
piece of brood-comb containing workers' eggs, or larvæ very young. You
will generally find it without much trouble, in a young swarm that is
making combs; the lower ends usually contain eggs; take a piece from
one of the middle sheets, two or three inches long, (you will probably
use smoke by this time without telling). Invert the hive that is to
receive it, put the piece edgewise between the combs, if you can spread
them apart enough for the purpose; they will hold it there, and then
there will be ample room to make the cells. They will nearly always
rear several queens. I have counted nine several times, which were all
they had room for. But yet I have very little confidence in such
queens, they are almost certain to be lost.


THEY ARE POOR DEPENDENCE.

Therefore I would recommend getting a royal cell whenever it is
practical. There is yet another advantage; you will have a queen ready
to lay eggs two or three weeks earlier, than when they are compelled to
commence with the egg. I have put such piece of brood-comb in a small
glass box on the top of the hive instead of the bottom, because it was
less trouble, but in this case the eggs were all removed in a short
time; whether a queen was reared in the hive or not I cannot say; but
this I know, I never obtained a prolific queen, after repeated
experiments in this way.

It would appear that I have been more unfortunate with queens reared in
this way, than most experimenters. I have no difficulty to get them
formed to all appearance perfect, but lose them afterwards. Now whether
this arose from some lack of physical development, by taking grubs too
far advanced to make a perfect change, or whether they were reared so
late in the season, that most of the drones were destroyed, and the
queen to meet one had to repeat her excursions till lost, I am yet
unable to _fully_ determine. To test the first of these questions, I
have a few times removed all the larvæ from the comb; leaving nothing
but eggs, that all the food given them might be "royal pap," from the
commencement, and had no better success so far. Yet occasionally
prolific queens have been reared when I could account for their origin
in no other way but from worker eggs. But you will find they are not to
be depended upon generally.

Sometimes, after all our endeavors, a stock or two will remain
destitute of a queen. These, if they escape the worms, will generally
store honey enough in this section to winter a good family. This will
have to be introduced, of course, from another hive, containing a
queen; but this belongs to Fall management.

As respects the time that elapses from the impregnation of the queen
till the commencement of egg laying, I cannot tell, but guess it might
be about two or three days. I have driven out the bees twenty-one days
after the first swarm, when no second swarm had issued--the young queen
came out on the fourteenth day. I found eggs and some very young larvæ.
When it is remembered that eggs remain three days before they hatch, it
shows that the first of these must have been deposited some four or
five days. When writers tell us the exact time to an hour (46 or 48)
from impregnation to laying, I am willing to admit the thing in this
case, but feel just as if I would like to ask how they managed to find
out the fact; by what sign they knew when a queen returned from an
excursion, whether she had been successful or not, in her amours; or,
whether another effort would have to be made; and then, how they
managed to know exactly when the first egg was laid.

Occasionally a queen is lost at other than the swarming season,
averaging about one in forty. It is most frequent in spring; at least
it is generally discovered then. The queen may die in the winter, and
the bees not give us any indications till they come out in spring.
(Occasionally they may all desert the hive, and join another.) If we
expect to ascertain when a queen is lost at this season, we must notice
them just before dark on the first warm days--because the mornings are
apt to be too cool for any bees to be outside--any unusual stir, or
commotion, similar to what has been described, shows the loss. This is
the worst time in the year to provide the remedy, unless there should
happen to be some very poor stock containing a queen, that we might
lose any way--then it might be advisable to sacrifice it to save the
other, especially if the last contained all the requisites of a good
stock except a queen. Some eight or ten, that I have managed in this
way, have given me full satisfaction. I have at other times let them go
till the swarming season, and then procured a queen, or introduced a
small swarm; at which time they are so reduced as to be worth but
little, even when not affected by the worms. To obviate this loss in
this way, it might be an advantage to transfer the bees to the next
stock, if it was not too full already; or the bees of the next stock to
this. Let the age and condition of the combs, quantity of stores, &c.,
decide.



CHAPTER XV.

ARTIFICIAL SWARMS.


PRINCIPLES SHOULD BE UNDERSTOOD.

Artificial swarms can be made with safety at the proper season. To the
bee-keeper who wishes to increase his stocks, it will be an advantage
to understand some of the principles. I have had some little experience
that has led to different conclusions from those of some others. I have
seen it stated, and found the assertion repeated by nearly every
writer, that "whenever bees were deprived of their queen, if they only
possessed eggs or young larvæ, they would not fail to rear another,"
&c. There are numerous instances of their doing this, but it is not to
be depended upon, especially when left in a hive full of combs, as the
following experiments tend to prove.


SOME EXPERIMENTS.

Several years since I had a few stocks well supplied with bees, and
every indication of swarming present, such as clustering out, &c., but
they pertinaciously adhered to the old stock, through the whole
swarming season! Others apparently not as well supplied with bees threw
off swarms. I had but few stocks, and was very anxious to increase the
number; but these were provokingly indifferent to my wishes. Taking the
assertions of these authors for facts, I reasoned thus: In all
probability there are eggs enough in each of those stocks. Why not
drive out a portion of the bees, with the old queen, and leave about as
many as if a swarm had issued? Those left will then raise a queen, and
continue the old stock, and I shall have six instead of the three, that
have been so obstinate. Accordingly, I divided each, examined and found
eggs and larvæ. Of course all _must be right_. Now, thought I, my
stocks can be doubled at least annually. If they do not swarm, I can
drive them.


THE RESULT UNSATISFACTORY.

My swarms prospered, the old stocks seemed industrious, bringing in
pollen in abundance, which to me at _that_ time, was conclusive that
they had a queen, or soon would have. I continued to watch them with
much interest, but somehow, after a few weeks, there did not seem to be
quite as many bees; a few days later, I was quite _sure_ there was not.
I examined the combs, and behold there was not a cell containing a
young bee of any age, not even an egg in any one of these old stocks.
My visionary anticipations of future success speedily retrograded about
this time.

I had, it is true, my new swarms in condition to winter, although not
quite full; but the old ones were not, and nothing was gained. I had
some honey, a great deal of bee-bread and old black comb. Had I let
them alone, and put on boxes, I should have probably obtained
twenty-five or thirty pounds of pure honey from each, worth five times
as much as what I did get; besides, the old stocks, even with the old
comb, would have been better supplied with both honey and bees;
altogether much better, as stocks for wintering. Here was a
considerable loss, merely by not understanding the matter.

I carefully looked the bees over, and ascertained to a certainty that
neither of them had a queen. I smothered what few there was left in the
fall. I then knew of no better way. I had been told that the barbarous
use of fire and brimstone was part of the "luck;" that a more
benevolent system would cause the bees "to run out," &c.


FURTHER EXPERIMENTS.

Subsequent to these experiments, I thought perhaps the jarring of the
hives in driving might have some effect on the bees, and prevent their
rearing a queen. This idea suggested the dividing hive, when the
division could be made quietly; but success was yet uncertain. I was
told to confine the bees in the old stock twenty-four hours or more,
after driving out a swarm; this I tried, with no better results. Again,
I drove out the swarm, looked out the queen, and returned her to the
old stock, compelling the new swarm to raise one. To be certain they
did so, I constructed a small box about four inches square, by two in
thickness; the sides glass. In this I put the piece of brood-comb
containing eggs and larvæ, and then put it on the hive containing the
swarm, having holes for communication, a cover to keep it dark, &c.
They were very sure to rear queens, but from some cause were lost after
they were matured.

Now, if others have been more successful in these experiments than
myself, it indicates that some favorable circumstances attended them
that did not me. I have not the least doubt but the result will be
favorable sometimes. Yet from the foregoing, I became satisfied that
not one of these methods could be relied upon. Instead of constructing
a queen's cell, and then removing the egg or larva to it from another
cell, I always found that the cell containing such egg or larva was
changed from the horizontal to the perpendicular; such cells as were in
the way below were cut off, probably using the material in forming one
for royalty, which, when finished, contains as much material as fifty
or a hundred others.

My experiments did not end here. I can now make artificial swarms, and
succeed nine times in ten with the first effort, and the reader can as
easily do the same. It must be in the swarming season, or as soon as
the first regular swarm issues. You want some finished royal cells that
any stock having cast a swarm will furnish, (unless in rare instances,
where they are too far up among the combs to be seen.)


A SUCCESSFUL METHOD.

When you are all ready, take a stock that can spare a swarm; if bees
are on the outside, raise the hive on wedges, and drive them in with a
little water, and disturb them gently with a stick. Now smoke and
invert it, setting the empty hive over. If the two hives are of one
size, and have been made by a workman, there will be no chance for the
bees to escape, except the holes in the side; these you will stop; (no
matter about a sheet tied around it.) With a light hammer or stick,
strike the hive a few times lightly, and then let it remain five
minutes. This is very essential, because most of the bees, if allowed
the opportunity, will fill themselves with honey after such
disturbance.

All regular swarms go forth so laden. A supply is necessary when bad
weather follows soon after. It is also used in forming wax, a very
necessary article in a new hive. The amount of honey carried out of a
stock by a good swarm, together with the weight of the bees (which is
not much), will vary from five to eight pounds.

This, allowing time for the bees to fill their sacks, and supplying the
old stock with a royal cell, I believe is entirely original: the
importance of which the reader can judge.


ADVANTAGES OF THIS METHOD.

It is very plain that a queen from such finished cell must be ready to
deposit eggs several days sooner than by any other method that we can
adopt. It is also clear that if we have a dozen queens depositing eggs
by the 10th of June, that our bees are increasing faster, on the whole,
than if but half that number are engaged in it for a month later. There
is yet another advantage. The sooner a young queen can take the place
of the old one in maternal duties, the less time will be lost in
breeding, the more bees there will be to defend the combs from the
moth, and the surest guaranty for surplus honey.

When the bees have filled their sacks, proceed to drive them into the
upper hive by striking the lower one rapidly from five to ten minutes.
A loud humming will mark their first movement. When you think half or
two-thirds are out, raise the hive and inspect progress. They are not
at all disposed to sting in this stage of proceeding, even when they
escape outside. If full of honey, they are seldom provoked to
resentment. The only care will be not to crush too many that get
between the edges of the hives. The loud buzzing is no sign of anger.
If your swarm is not large enough, continue to drive till it is. When
done, the new hive should be set on the stand of the old one. A few
minutes will decide whether you have the queen with the swarm, as they
remain quiet: otherwise uneasy, and run about, when it will be
necessary to drive again.

If both hives are one color, set the old one two feet in front; but if
of different colors, a little more. I prefer this position to setting
the old stock on one side, even when there is room; yet it can make but
little difference. Should you set it on one side, let the distance be
less. When the old stock is taken much farther than this rule, all the
bees that have marked the location (and all the old ones will have done
so) will go back to the old stand, and none but young bees that have
never left home will remain. The same will be the case with the new
swarm if moved off. It will not do to depend on the old queen keeping
them, as she does when they swarm out naturally. This has been my
experience. Try it, reader, and be satisfied, by putting either of the
hives fifteen or twenty feet distant.

Before you turn over the old stock, look among the combs as far as
possible for queens' cells; if any contain eggs or larvæ, you may
safely risk their rearing a queen; but otherwise wait till next
morning, or at least twenty-four hours, then go to a stock that has
cast a swarm, and obtain a finished royal cell, as before directed, and
introduce it. You will have a queen here as soon as if it had been left
in the original hive, and no risk of an after swarm, because there is
but one. But when there are young queens in the cells at the time of
driving, after swarms may issue. Should a queen-cell be introduced
immediately, it is more liable to be destroyed than after waiting
twenty-four hours; and then is not always safe. After it has had time
to hatch, (which is about eight days after being sealed), cut it out,
and examine it: if the lower end is open, it indicates that a perfect
queen has left it, and all is safe; but if it is mutilated or open at
the side, it is probable that the queen was destroyed before maturity,
in which case, another cell will have to be given them.


ARTIFICIAL SWARMS ONLY SAFE NEAR THE SWARMING SEASON.

By what I have said about artificial swarms, it would appear that it is
unsafe at any time but the swarming season; that is my opinion. It may
do a little in advance or a little after, providing royal cells can be
had. By feeding as directed, (in Chapter IX.) you may induce a stock to
send out a swarm some days in advance of the regular season, thereby
giving you a chance for these cells somewhat early.


SOMETIMES HAZARDOUS.

To make such swarms at any time when the bees are destroying drones,
would be extremely hazardous, not only on account of the young queen
being impregnated, but their massacre denotes a scarcity of honey.
Therefore I would advise never to make swarms, or drive out bees at
such periods, when it can be avoided, without spare honey is on hand to
feed them.


SOME OBJECTIONS.

It has been argued by some, and with much reason, that "nature is the
best guide, and it is better to let the bees have their own way about
swarming--if honey is abundant, and the stock is in condition to spare
a swarm, their own instincts will teach them to construct royal cells;
if it fails before they are ready, and the royal brood is destroyed, it
is because the existence of the swarm would be precarious, and it is
best not to issue." I will grant that in many instances it is better.
The chance is better for surplus honey; the stock is quite sure to be
in condition to winter; and some judgment is required to tell when a
stock can spare a swarm.

But yet, we are sometimes anxious to increase our stocks to the utmost
that safety will allow, and often have some that can spare a swarm as
well as not, but refuse to leave; perhaps commence preparations, and in
a few days abandon them. Now it is evident that as long as many
continue such preparation, that honey is sufficiently abundant to put
the safety of the swarm beyond hazard; some stocks will swarm while
these others just as good, (that had abandoned it before) and have not
now begun again, to be in time before a partial failure of honey, and
some may not have commenced in season.


NATURAL AND ARTIFICIAL SWARMS EQUALLY PROSPEROUS.

I can see no difference in artificial or natural swarms of equal size,
at the same time. By taking the matter in time into our own hands, with
the rules given, we make a sure thing of it, that is, we are sure to
get the swarms, when if left to the bees it would be uncertain, and no
greater risk afterwards than with natural issues.


THIS MATTER TOO OFTEN DELAYED.

I am aware that this matter will be apt to be put off too long; "wait
and see if they don't swarm," will be the motto of too many, and when
the season is over, drive them. Perhaps a good swarm has set outside
the hive, all through the best of the honey season, and done nothing,
while they could have half filled a hive; but this is all lost now, as
well as the best chances for getting cells. Let me impress the
necessity of doing it in season, when it will pay. If you intend to
have a swarm from every stock that can spare one, begin when nature
points out the proper time, which is, when the regular ones begin to
issue. It must, indeed, be a poor season when there are none.


IS THE AGE OF THE QUEEN IMPORTANT?

There is another object effected in this way, considered by some
apiarians as very important. It is the change of the queens in the old
stock. A young queen is thought to be "much more prolific than an old
one." They even recommend keeping none "over two or three years old,"
and give directions how they may be renewed. But as I have been unable
to discover any difference in relation to the age in this respect, I
shall not at present take much time to discuss it. It is well enough,
when we can take our choice without trouble, to preserve a young queen.
When we consider that there are but few queens but what will deposit
three times as many eggs in a season as are matured, it looks as if it
would hardly pay to take much trouble to change them.  At what time the
queen becomes barren from old age, I presume has never yet been fully
determined.

A friend of mine has had a stock in a large room eight years, that has
never swarmed, and is still prosperous! I think it very probable that
this queen will gradually decay, and possibly become barren, some weeks
before she dies; if so, this stock will soon die off. A few such cases
will probably occur in swarming hives, perhaps one in fifty, but
generally such old and feeble queens are lost when they leave with the
swarm, especially in windy weather. As long as they are able to go with
the swarm, and sometimes when they are not, I have found them
sufficiently prolific for all purposes. I would rather risk their
fecundity, and hive the swarm, than to allow the bees to return to the
parent stock, and wait eight or nine days for a young queen to mature.
A great many will remain idle, even if there is room to work in the
boxes.



CHAPTER XVI.

PRUNING.


Notwithstanding I have given the method of pruning in the chapter on
hives, (page 23, Chapter II.) it will be necessary to give the tyro in
bee-culture a few more particulars. The season for doing it is of
importance.


DIFFERENT OPINIONS AS TO TIME.

The month of March has been recommended by several; others prefer
April, August, or September. Here, as usual, I shall have to differ
from them all, preferring still another period, for which I offer my
reasons, supposing, of course, that the reader is conscious of a
freeman's privilege, that is, to adopt whatever method he thinks
proper, on this, as on any other point.


ANOTHER TIME PREFERRED.

There is but one period from February till October, when prosperous
stocks are free from young brood in the combs. If combs are taken out
when occupied, there must be a loss of all the young bees they contain;
which may be avoided. The old queen leaves with the first swarm; all
the eggs she leaves in the worker-cells will be matured in about
twenty-one days, consequently this is the time to clear out the old
combs with the least waste. A few drones will be found in the cells,
that would require a few days more to hatch, but these are of no
account. Also a few very young larvæ and some eggs may be sometimes
found, the product of the young queen; these few must be wasted, but as
the bees have expended no labor upon them as yet, it is better to
sacrifice these than the greater number left by her mother, which have
consumed their portion of food; the bees have sealed them up, and now
only require the necessary time to mature, to make a valuable addition
to the stock.


SHOULD NOT BE DELAYED.

Should this operation be put off for a time much longer than three
weeks, the young queen will so fill the combs again as to make it a
serious loss. Therefore, I wish to urge strongly attention to this
point at the proper season. If you think it unimportant to mark the
date of your first swarms for the purposes mentioned in another place,
it will be found very convenient here, for those that need pruning.

It is also recommended by some, to take only a part, say one-third or
half, in a season; thereby taking two or three years to renew the
combs. This is advisable only when the family is very small. As this
space made by pruning cannot be filled without wax and labor, our
surplus honey will be proportionate to its extent. Now suppose we take
out half the old combs, and get half a yield of box honey this year,
and the same next, or make a full operation of it and get none this
year, and a full one next. What is the difference? There is none in
point of honey, but some in trouble, and that is in favor of a full
operation at once. We have to go through with about the same trouble to
get one-third or half as to take the whole.


OBJECTION TO PRUNING.

The objection to this mode of renewing combs generally, will be the
fear of getting stung. But I can assure you there is but little danger,
not as much as to walk among the hives in a warm day. Only begin right,
use the smoke, and work carefully, without pinching them, and you will
escape unhurt generally.


STOCKS PRUNED NOW ARE BETTER FOR WINTER.

Besides the advantage of saving a large brood by pruning at this
season, such stocks will usually refill before fall, and are much
better for wintering, which is not the case when it is done later. We
must of necessity then waste the brood, and have a large space
unoccupied with combs through the winter. But few combs can then be
made, and those few must be at the expense of their winter stores,
unless we resort to feeding.

These objections apply with greater force to pruning in March or April.
The loss of brood is of much more consequence now, than in mid-summer,
or even later, and a space to be filled with combs is a serious
disadvantage. It is important that the bees should devote their whole
attention now to rearing brood, and be ready to cast their swarms as
early as possible. One _early_ swarm is worth two late ones. Suppose a
stock, instead of collecting food and nursing its young, is compelled
to expend its honey and labor in secreting wax and constructing combs
before it can proceed with breeding advantageously, it _must of
necessity_ be some weeks later.

Further, I have always found it best to have the bees out of the way,
during this operation. It will be found much more difficult to drive
the bees out of a hive in the cool weather of March or April, than in
summer, as they seem unwilling to shift their warm quarters and go into
a cold hive.

It is presumed the reader will bear in mind the disadvantages already
given of too frequently renewing combs; the little value of combs for
storing honey, _for our use_, after being once used for breeding; the
necessity of the bees using them as long as they possibly will answer;
and not compel them to be filling the hive, when they might be storing
honey of the purest quality in boxes, &c.

Vide remarks on this subject on page 22, Chapter II.



CHAPTER XVII.

DISEASED BROOD.


This, like many other chapters in this work, is probably new, as I,
never saw one thus headed. A few newspaper discussions are about all
that have yet appeared on this subject.


NOT GENERALLY UNDERSTOOD.

This disease is probably of recent origin. Mr. Miner, it appears, knew
nothing of it until he moved from Long Island to Oneida County, in this
State. Mr. Weeks, in a communication to the N.E. Farmer, says, "Since
the potato rot commenced, I have lost one-fourth of my stocks annually,
by this disease;" at the same time adds his fears, that "this race of
insects will become extinct from this cause, if not arrested." (Perhaps
I ought to mention, that he speaks of it as attacking the "chrysalis"
instead of the larva; but as every thing else about it agrees exactly,
there is but little, doubt of its being all one thing.)


MY OWN EXPERIENCE.

My first experience will probably go back to a date beyond many others;
it is almost twenty years since the first case was noticed. I had kept
bees but four or five years when I discovered it in one of my best
stocks; in fact, it was No. 1 in May and first of June. It cast no
swarm through the summer; and now, instead of being crowded with bees,
it contained but very few; so few, that I dared not attempt to winter
it. What was the matter? I had then never dreamed of ascertaining the
condition of a stock while there were bees in the way, but was like the
unskilful physician who is obliged to wait for the death of his
patient, that he may dissect and discover the cause. I accordingly
consigned what few bees there were to the "brimstone pit."


DESCRIPTION OF DISEASE.

A "_post mortem_" examination revealed the following circumstances:
Nine-tenths of the breeding-cells were found to contain young bees in
the larva state, stretched out at full length, sealed over, dead,
black, putrid, and emitting a disagreeable stench. Now here was one
link in the chain of cause and effect. I learned why there was a
scarcity of bees in the hive. What should have constituted their
increase, had died in the cells; none of them were removed,
consequently but few cells, where any bees could be matured, were left.


THE CAUSE UNCERTAIN.

But when I attempted the next link in the chain (to wit) What caused
the death of this brood just at this stage of development? I was
obliged to stop. Not the least satisfaction could be obtained. All
inquiries among the bee-keepers of my acquaintance were met with
profound ignorance. They had "never heard of it!" No work on bees that
I consulted ever mentioned it.

Subsequently, I had more stocks in the same situation. I found,
whenever the disease existed to any extent, that the few bees matured
were insufficient to replace those that were lost; that the colony
rapidly declined, and _never afterwards cast a swarm_!


REMEDIAL EXPERIMENTS.

As for remedies, I tried pruning out all those combs containing brood,
leaving only such as contained honey, and let the bees construct new
for breeding. It was "no use," these new combs were invariably filled
with diseased brood! The only thing effectual was to drive out the
bees, into an empty hive. In this way, when done in season, I generally
succeeded in rearing a healthy stock. But here was a loss of all
surplus honey, and a swarm or two that might have been obtained from a
healthy one.


PUBLIC INQUIRY AND ANSWERS.

I had so many cases of the kind, that I became somewhat alarmed, and
made inquiry through the Cultivator, (an agricultural paper,) as to a
cause, and remedy, offering a "reward for one that would not fail when
thoroughly tested," &c. Mr. Weeks, in answer, said, "that cold weather
in spring chilling the brood was the cause." (This was several years
prior to his article in the N.E. Farmer.) Another gentleman said, "dead
bees and filth that accumulated during winter, when suffered to remain
in the spring, was the cause." A few years after, another correspondent
appeared in the Cultivator, giving particulars of his experience,
proving very conclusively to himself and many others, that cold was the
cause. Having mislaid the paper containing his article, I will endeavor
to quote correctly from memory. He had "three swarms issue in one day;
the weather during the day changed from very hot to the other extreme,
producing frost in many places the next morning. These swarms had left
but few bees in the old stocks, and the cold forced them up among the
combs for mutual warmth; the brood near the bottom, thus left without
bees to protect it with animal heat, became chilled, and the
consequence was diseased larvæ." He then reasoned thus: "If the eggs of
a fowl, at any time near the end of incubation, become chilled from any
cause, it stops all further development. Bees are developed by
continued heat, on the same principle, and a chill produces the same
effect, &c.; afterwards, other swarms issued under precisely similar
circumstances; but these old stocks were covered with a blanket through
the night, which enabled the bees to keep at the bottom of the hive. In
a few days, enough were hatched to render this trouble unnecessary.
These last remained healthy." He further says, that "last spring was
the first time I ever knew them to become diseased before swarming had
thinned the population. The weather was remarkably pleasant through
April. The bees obtained great quantities of pollen and honey, and by
this means extended their brood further than usual at this season.
Subsequent chilly weather in May, caused the bees to desert a portion
of brood, which were destroyed by the chill."

Now this is reasoning from cause to effect very consistently.


ANSWERS NOT SATISFACTORY.

Had I no experience further than this, I should, perhaps, rest
satisfied as to the cause, and should endeavor to apply the remedy.
Several other writers have appeared in different papers, on this
subject, and nearly all who assign a cause have given this one as the
most probable. Now I have known the chrysalis in a few stocks to be
chilled and destroyed by a sudden turn of cold weather, yet these were
removed by the bees soon after, and the stocks remained healthy. To me
the cause assigned appears inadequate to produce _all_ the results with
the larvæ. After close, patient observation of fifteen years, I have
never yet been wholly satisfied that any one instance among my bees,
was thus produced.


A CAUSE SUGGESTED.

We are all familiar to some extent with the contagious diseases of the
human family, such as small-pox, whooping-cough, and measles, and their
rapid spread from a given point, &c. We must also admit that some cause
or causes, adequate to the effect, must have produced the first case.
To contagion, then, I would attribute the spread of this disease of our
bees, at least nineteen cases in twenty. I will admit, if you please,
that one stock in twenty or fifty may be somewhat affected by a chill
to a small extent. It is only a portion of the brood that is in
danger--only such as have been sealed over, and before they have
progressed to the chrysalis state, are attacked. How many then can
there be in a hive at any one time, in just the right stage of
development to receive the fatal chill? Of course there will be some;
but they should be confined to the cells near the bottom, where the
bees had left them exposed. These should be all; and these few would
never seriously damage the stock. Why then does this disease, when
thoroughly started, spread so rapidly throughout all the combs in the
hive? Will it be said that the chill is repeated every few days through
the summer? Or will it be admitted that something else may continue it?

I think there must be other causes, besides the chill, even to start
it, in most cases. As our practice will be in accordance with the view
we take of this matter, and the result of our course will be somewhat
important, I will give some of the reasons that have led to this
conclusion.


REASONS FOR THE OPINION.

For instance, I had all the bees of a good swarm leave the hive in
March; after flying a time, they united with another good stock, making
double the usual number of bees at this season; enough to keep the
brood sufficiently warm at any time; if other stocks with half or a
quarter of the number could. By the middle of June, the bees were much
reduced, and had not cast a swarm. It was examined, and the brood was
found badly diseased. My best and most populous stocks, in spring, are
just as liable, and I might add more so, than smaller or weaker
families. I have had two large swarms unite, and were hived together,
that were diseased the next autumn. These cases prove strongly, if not
conclusively, that animal heat is not the only requisite. The fact that
when I had pruned out all affected comb from a diseased stock, and left
honey in the top and outside pieces, and the bees constructed new for
breeding, and the brood in such were invariably affected, though only a
few at first, and increasing as the combs were extended; led me to
suppose that it was a contagious disease, and the virus was contained
in the honey. Some of it had been left in these stocks, and very
probably the bees had fed it to the brood. To test this principle still
further, I drove all the bees from such diseased stocks, strained the
honey, and fed it to several young healthy swarms soon after being
hived. When examined a few weeks after, every one, without an
exception, had caught the contagion.

Here then is a clue to the cause of this disease spreading, whether we
have its origin or not. We will now see if we can trace it through, if
there is any consistency in its transfer from one stock to another.


CAUSE OF ITS SPREADING.

Suppose one stock has caught the infection, but a small portion of the
brood is dead. In the heat of the hive, it soon becomes putrid; other
cells adjoining with larvæ of the right age are soon in the same
condition. All the breeding combs in the hive become one putrid mass,
with an exception, perhaps, of one in ten, twenty or a hundred, that
may perfect a bee. Thus the increase of bees is not enough to replace
the old ones that are continually dying off. It is plain, therefore,
that this stock _must_ soon dwindle down to a very small family. Now
let a scarcity of honey occur in the fields, this poor stock cannot be
properly guarded, and is easily plundered of its contents by the
others. Honey is taken that is in close proximity to dead bodies,
corrupting by thousands, creating a pestilential vapor, of which it has
probably absorbed a portion. The seeds of destruction are by this means
carried into healthy stocks. In a short time, these in turn fall
victims to the scourge; and soon dwindle away, when some other strong
stock is able to carry off _their_ stores; and only stop, perhaps, at
the last stock! The moth is ever ready with her burden of eggs, which
she now without hindrance deposits directly on the combs. In a short
time the worms finish up the whole business, and are judged guilty of
the whole charge; merely because they are found carrying out effects
that speedily follow such causes.

Let the reader who doubts this theory, simply strain out honey,
vitiated in this way, and feed it to a few stocks or swarms, that are
healthy; and if they escape, communicate the fact to the public. But
should he become satisfied that such honey is poison to his bees, he
will with me, and all others interested, wish to stop this growing
evil.


NOT EASILY DETECTED AT FIRST.

It is very difficult to detect the first hundred or two that die in a
stock. But when nine-tenths of the breeding cells hold putrid larvæ,
there is but very little trouble in making out a correct diagnosis. The
bees are few and inactive. When passing the hive our olfactories are
saluted with a nauseous effluvia, arising from this corrupting mass.
Now, if we wish, or expect to escape, the most severe penalty, our
neglect must never allow this extent of progression before such a stock
is removed. Therefore, we must watch symptoms--ascertain the presence
of the disease _at the earliest moment possible_.


SYMPTOMS TO BE OBSERVED.

As no part of the breeding season is exempt, the stocks should be
carefully observed during spring, and fore part of summer, relative to
increase of bees. When one or more is much behind others in this
respect, make an examination immediately. (I would here urge again the
convenience of the simple, common hive, over those more complicated, or
suspended, and difficult to turn over. In one case we might make an
examination in season; in the other, too much trouble and difficulty
might cause it to be put off too long.) The hive must be inverted, and
the bees smoked out of the way. Our attention is to be directed to the
breeding cells; with a sharp-pointed knife, proceed to cut off the ends
of some of them that appear to be the oldest; bearing in mind that
young bees are always white, until some time after they take the
chrysalis state. Therefore, if a larva is found of a dark color, it is
dead! Should a dozen such be found, the stock should be condemned at
once, and all the bees driven into an empty hive. (The directions for
this have been given, see page 31.) If honey should be scarce, at the
time, they should be fed.


SCALDING THE HONEY TO DESTROY THE POISON FOR FEEDING.

The honey from the old hive may be used, if you will only first destroy
the virus. This, I have ascertained, may be done by scalding: add a
half-pint of water to about ten lbs.; stir it well, and heat it to the
boiling point, and carefully remove all the scum.

Stocks in which the disease has not progressed too far, will generally
swarm.


WHEN TO EXAMINE STOCKS THAT HAVE SWARMED.

Three weeks from the first swarm, will be the time to examine them. I
make it a rule to inspect all my stocks at this period. It is easily
done now, as about all the healthy brood (except drones) should be
matured in that time. By perseverance in these rules, I allow no stocks
to dwindle away until they are plundered by others. If all my neighbors
were equally careful, this disease would probably soon disappear. This
is like one careless farmer allowing a noxious weed to mature seeds, to
be wafted by winds on the lands of a careful neighbor, who must fortify
his mind to continual vigilance, or endure the injury of a foul pest.
So with the successful apiarian; in sections where the disease has
appeared (it has not in all), he must be continually on the watch; it
is the price of success.


CARE IN SELECTING STOCK HIVES FOR WINTER.

Again, after the breeding season is over, in the fall, _every stock
should be thoroughly inspected, and all diseased ones condemned for
stock hives_. It is better to do it, even if it should take the last
one. It would pay much better to procure others instead, that are
healthy.

Persons wishing to eat the honey from such hives, will experience no
bad effects from it, if they are careful to remove all the dead brood,
as they take it out of the hive.

The greatest distance that I ever knew bees to go, and plunder a
defenceless stock of its contents, was three-fourths of a mile. Very
likely they would go farther on some occasions, but not often.


ACCUSATIONS NOT ALWAYS RIGHT.

Careless bee-keepers, when their hives are thus robbed, feel regret, or
are more often vexed at somebody--at the result of their carelessness.
The person, keeping most bees in a neighborhood, must expect to be
accountable for all effects of their ignorance, mismanagement, or
carelessness, and consequent "bad luck;" when all the honey thus
obtained, probably carries with it more mischief than can be eradicated
in a twelvemonth, thereby giving the real cause of complaint to the
other party.



CHAPTER XVIII.

IRRITABILITY OF BEES.


Keeping bees good-natured, offers a pretty fair subject for ridicule:
it seems rather too absurd to teach _a bee_ anything! Nevertheless, it
is worth while to think of it a little. Most of us know that by
injudicious training, horses, cattle, dogs, &c., may be rendered
extremely vicious. If there is no perceptible analogy between these and
bees, experience proves that they may be made ten times more irritable
than they naturally would be.


THEIR MEANS OF DEFENCE.

Nature has armed them with means to defend their stores, and provided
them with combativeness sufficient to use them when necessary. This
could not be bettered. If they were powerless to repel an enemy, there
are a thousand lazy depredators, man not excepted, who would prey upon
the fruits of their industry, leaving them to starve. Had it been so
arranged, this industrious insect would probably have long since been
extinct.


TIME OF GREATEST IRRITABILITY.

The season of their greatest caution, in this section, is August,
during the flowers of buckwheat. It is then their stores are greatest.
As soon as a stock is pretty well supplied with this world's goods,
like some bipeds, they become very haughty, proud, aristocratic, and
insolent. A great many things are construed into insults, that in their
days of adversity would pass unnoticed; but now it is becoming and
proper for their honor to show a "just resentment." It behooves us,
therefore, to ascertain what are considered insults.


PROPER CONDUCT.

First, all quick motions, such as running, striking, &c., about them,
are noticed. If our movements among them are slow, cautious, humble,
and respectful, we are often let to pass unmolested, having manifested
a becoming deportment. Yet the exhalations from some persons appear
very offensive, as they attack them much sooner than others; though I
apprehend there is not so great a difference as many suppose. Whenever
an attack is made, and a sting follows, the venom thus imparted to the
air, if by only one, is perceived by others at some distance, which
will immediately approach the scene, and more stings are likely to
follow than if the first had not been.


HOW TO PROCEED WHEN ATTACKED.

Striking them down renders them ten times more furious. Not in the
least daunted, they return to the attack. Not the least show of fear is
perceived. Even after losing their sting, they obstinately refuse to
desist. It is much the best way to walk as quietly as possible to the
shelter of some bush, or to the house. They will seldom go inside of
the door.


A PERSON'S BREATH OFFENSIVE, AND OTHER CAUSES.

The breath of a person inside the hive, or among them, when clustered
outside, is considered in the tribunals of their insect wisdom as the
greatest indignity. A sudden jar, sometimes made by carelessly turning
up the hive, is another. After being once thoroughly irritated in this
way, they remember it for weeks, and are continually on the alert; the
moment the hive is touched, they are ready to salute a person's face.
When slides of tin or zinc are used to cut off the communication
between the hives and boxes, some of the bees are apt to be crushed or
cut in two. This they remember, and retaliate, as occasion offers; and
it may be when quietly walking in the apiary.


THEIR MANNER OF ATTACK.

I must disagree with any one who says we always have warning before
being stung. I have been stung _a few times_ myself. Two-thirds of them
were received without the least notice--the first intimation was the
"blow." At other times, when fully determined on vengeance, I have had
them strike my hat and remain a moment endeavoring to effect their
object. In this case, I have warning to hold down my face to protect it
from the next attempt, which is quite sure to follow. As they fly
horizontally, the face held in that position is not so liable to be
attacked. When they are not so thoroughly charged with anger, they
often approach in merely a threatening attitude, buzzing around very
provokingly for several minutes in close proximity to our ears and
face, apparently to ascertain our intentions. If nothing hostile or
displeasing is perceived, they will generally leave; but should a quick
motion or offensive breath offend them, the dreaded result is almost
sure to follow. Too many people are apt to take these threatening
manifestations as positive intentions to sting. When these things can
be quietly endured, and at the same time leave their vicinity, it
generally ends peaceably. They never make an attack while away from
their home in quest of honey, or on their return, until they have
entered the hive. It is only in the hive and its vicinity that we
expect to meet this irascible temperament, which should not be allowed,
or at least may be subdued in a great measure, if not entirely, by
doing things in a quiet manner, and, by the use of tobacco smoke. Any
person having the care of bees should go armed with this powerful
weapon. As bees are not much affected with smoke, while flying in the
air, but will have their own way, we must take them in the hive as the
place to teach _them_ a proper deportment!

Those who are accustomed to smoking will find a pipe or segar very
convenient here. But such as are not would do better, perhaps, not to
learn a bad habit. I will therefore give a simple substitute.


SMOKER DESCRIBED.

Get a tube of tin about five-eighths of an inch diameter, five or six
inches in length; make stoppers of wood to fit both ends, two and a
half or three inches long; with your nail-gimlet make a hole through
them lengthwise: when put together it should be about ten inches. The
ends may be tapered. On one end leave a notch, that it may be held with
the teeth, which is the most convenient way, as you will often want to
use both hands: it is also always ready, without any trouble to blow
through, and also to keep the tobacco burning. When ready to operate,
fill the tube with tobacco, ignite it, and put in the stoppers; by
blowing through it you keep the tobacco burning while the smoke issues
at the other end.


EFFECT OF TOBACCO SMOKE.

We can now subdue these combative propensities, or render them
harmless; turn their anger to submission, and make them yield their
treasures to the hands of the spoiler without an effort of resistance!
When once overpowered, they seem to lose all knowledge of their
strength, and no slave can be more submissive! After the effects of the
smoke have passed off, their former animosity will return. Should any
resentment be shown on raising a hive, blow in the smoke; they
immediately retreat, "begging pardon." After a few times, they learn
"it's no use," and allow an inspection. If you wish to take off a box,
raise it just enough to blow under the smoke; there is no trouble; you
can replace it with another; the bees are kept out of the way with a
little more smoke, _and no anger created about it to be remembered_.
Those in the box are all submission; they can be carried away and
handled as you please, without a possibility of getting them irritated,
until they once more get home, and then are much more "amiable" than if
the box had been taken without the smoke. They seem to forget, or do
not realize anything of the transaction. When bees are to be
transferred to a new hive, it is unnecessary to be so very particular
about the escape of a single bee; no fears need be entertained of such
as get out. In driving, the loud humming indicates their submission;
the upper hive can then be safely raised at any time. After being thus
driven out, they may be pushed about with impunity, and still be quiet!
In short, by using smoke on all occasions where they would be likely to
be disturbed without it by our meddling with them, it has a tendency to
keep dormant their combative propensities. When these have never been
aroused, there is much less danger from their attacks while walking or
looking among them. Any one wishing further proof, I would recommend
the experiment of managing one year with smoke, and the next without.


STING DESCRIBED.

Their sting, as it appears to the naked eye, is but a tiny instrument
of war; so small, indeed, that its wound would pass unheeded by all the
larger animals, if it was not for the poison introduced at the same
instant. It has been described as being "composed of three parts, a
sheath and two darts. Both the darts are furnished with small points or
barbs like a fishhook," that hold it when introduced into the flesh;
the bee being compelled to leave it behind.


DOES ITS LOSS PROVE FATAL?

It is said "to the bee itself this mutilation proves fatal." This last
is another assertion for fact, so often repeated, that perhaps we might
as well admit it; seeing the difficulty we should have in disproving
it. Only think of the impossibility of keeping our eye, for five
minutes, on a bee that is flying about, after it has left its sting.
Yet there are some persons so very particular about what they receive
as facts, that they would require this very unreasonable thing of
watching a bee till it died, before they could be _positively sure_
that the loss of its sting caused its death. (It is much easier to
guess.) They might even take analogy, and say that other insects
possess so little sensation that they have been known to recover after
much more extensive mutilation--that beetles have lived for months
under circumstances that would have instantly killed some of the higher
animals--that spiders often reproduce a leg, even lobsters can replace
a lost claw, &c. I have put off describing any protection against their
attacks, because I wish to get up a little more courage in our doings
among them. Yet it is folly to expect all will manage successfully
without something for defence.


MEANS OF PROTECTION.

The face and hands are most exposed; for the latter, thick woollen
mittens or gloves are best; the sting is generally left when thrust
into a leather glove. For the face procure one and a half yards of thin
muslin or calico, sew the ends together, the upper end gathered on a
string small enough to prevent it slipping over the head when put on.
An arm-hole is to be cut out on each side; below is another string to
gather it close to the body. As I do not expect you to work in the
dark, we will have a place cut out in front, and a piece of coarse lace
inserted; that which will just prevent a bee from passing, is best, as
it gives us a better chance to see. To keep it from falling against the
face, a wire is bent around and sewed fast. Any person that knows how
to put on a shirt will manage this. When thus equipped, and other
garments of proper thickness, the most timid ought not to hesitate to
venture among them, when necessary. I cannot avoid cautioning you again
to beware of irritating your bees, until this protection is necessary,
as it is a rather bad state of things. With this on, you cannot
conveniently use any smoke. To put this on and off is considerable
trouble, and every time you go among them, if you have to resort to
this, I fear some necessary duties will be neglected. Whenever a
partial protection will do, I would recommend a handkerchief; it is
always at hand, and can be put on in a moment; throw it over the head,
letting the ends fall around the neck and shoulders, covering all but
the face. The hat can come on over it. As for the face, whenever a bee
comes around in a menacing attitude, hold it down--unless he stings at
the first onset, there is not much risk.


REMEDIES FOR STINGS.

Concerning the remedies for stings, it is a hard matter to tell which
is the best. There is so much difference in the effect in different
individuals, and the different parts of the body, as well as the depth
the sting reaches, that a great variety of remedies are recommended.

A person is slightly stung, and applies something as an antidote; the
effect of the sting is trifling, as perhaps it would have been without
anything, and the medicine is forthwith extolled as a sovereign remedy.
I have been thus deceived; when slightly stung applied what I thought
cured in one case, when in the next the sting might have penetrated
deeper, or in some other place, and the remedy would seem to have no
effect. For the last few years, I have not made any application
whatever for myself, and the effect is no worse, nor even as bad as
formerly. (This, I am told, is because the system is hardened, and now
can resist or throw off the effects.) Among the remedies recommended,
are saleratus and water, salt and water, soft-soap mixed with salt, a
raw onion cut in two and one-half applied, mud or clay mixed pretty wet
and changed often, tobacco wet and rubbed thoroughly to get at the
strength, and cold water constantly applied. To cure the smart, the
application of tobacco is strongly urged, and cold water is spoken of
with equal favor to prevent the swelling.

When stung in the throat, drinking often of salt and water is said will
prevent serious consequences.

Whether any of these remedies are applied or not, I suppose it is
unnecessary to say that the sting should be pulled out as soon as
practicable.



CHAPTER XIX.

ENEMIES OF BEES.


Among the enemies of bees, there are included rats, mice, birds, toads,
and insects.


ARE THEY ALL GUILTY?

But some of these are probably clear of any actual mischief. I strongly
suspect that the spirit of destructiveness with many people is
altogether too active. There are some farmers, with this principle
predominant, so short-sighted, that if it was in their power they would
destroy a whole class of birds, because some of them had picked a few
cherries, or dug out a few hills of corn, when, at the same time, they
are indebted to their activity in devouring worms, insects, &c., that
would otherwise have destroyed entire crops! It will be well,
therefore, before condemnation, to see if on the whole we are to be
gainers or losers by an indiscriminate slaughter, without judge or
jury.


RATS AND MICE.

Rats and mice are never troublesome, except in cold weather. The
entrances of all hives standing out are too small to admit a rat. It is
only when in the house that much damage need be apprehended. They
appear to be fond of honey, and when it is accessible will eat several
pounds in a short time.

Mice will often enter the hive when standing on the bench, and make
extensive depredations. Sometimes, after eating a space in the combs,
they will there make their nest. The animal heat created by the bees
will make a snug, warm place for winter quarters. There are two kinds:
one the common class, belonging to the house; the other called
"deer-mouse"--the under side perfectly white, the back much lighter
than the other kind. The latter seems to be particularly fond of the
bees, while the first appears to relish the honey. Whether they take
bees that are alive, or only such as are already dead, I cannot say.
Only a part of the bee is eaten; and if we take the fragments left to
judge of the number consumed, the circumstance will go some ways to
prove the sacrifice of quite a number. Whether bees or honey is wasted,
a little care to prevent their depredations is well worthy of bestowal.
As rats and mice have so long since been condemned and sentenced for
being a universal plague, and without a redeeming trait, I will say
nothing in their favor, and am perfectly willing they shall be hanged
till dead.


ARE ALL THE BIRDS GUILTY?

But for some of the birds accused of preying upon bees, I would say a
word.


KING-BIRD--ONE WORD IN HIS FAVOR.

The king-bird stands at the head of the list of depredators! With a
fair trial he will be found guilty, though not so heinously criminal as
many suppose. I think we shall find him guilty of taking only the
drones. In the afternoon of a fair day he may be seen perched upon some
dry branch of a shrub or tree near the apiary, watching for his
victims, occasionally darting to seize them. I have shot him down and
examined his crop, after seeing him devour a goodly number; but in
every instance the bees were so crushed to pieces, that it was
impossible to distinguish workers from drones. We are told of great
numbers of workers being counted. It may be so, or it may be thus
represented by a spice of prejudice. I have found the brutal
gratification of taking life so strong with some, that a natural
antipathy is allowed to take the place of justice, and a proper defence
is not allowed in such cases where the suffering party has not the
power to enforce it. If he was satisfied with workers as well as
drones, why does he not visit the apiary long before noon, and fill his
crop with them? But instead, he waits till afternoon for the drones;
and if none are flying, he watches quietly till one appears, although
workers may be out by hundreds continually. If the question is asked,
how they tell the difference in the two kinds of bees, I might suggest
that _instinct_ has taught most animals the proper kind of food, and
might direct the birds in this case. If it was not sufficient, a little
experience in catching bees provided with stings, might impart the
important difference, in one or two lessons. I once had a chicken that
knew the difference by some means, and would stand by the hive and
devour every drone, the moment it touched the board, while the workers
would pass by him in scores untouched!

Now, whether this taking the drones is a disadvantage or otherwise,
would depend entirely upon circumstances. If honey was a little scarce,
the less we had of them the better; it would also save the bees some
trouble in dispatching them. It is probably a matter of so little
moment to our bees, that it will not pay for powder to shoot them.

Martins, and a kind of swallows, are said to be guilty of taking bees
on some occasions; but as they pursue them on the wing (if they do),
the same remarks will apply as to the king-bird.


CAT-BIRD ACQUITTED.

The cat-bird also comes in for a share of censure. It is said "they
will get right down by the hive, and pick up bees by the hundred." Yet,
right in the face of this charge, I am disposed to acquit him. With the
closest observation, I find him about the hive, picking up _only_ young
and immature bees, such as are removed from the combs and thrown out.
They may be seen as soon as the first rays of light make objects
visible about the apiary, looking for their morning supply, as well as
frequent visits during the day. Should an unlucky worm be in sight just
then, while looking up a place for spinning a cocoon, or a moth
reposing on some corner of the hive, their fate is at once decided.
Before destroying this bird, it would be well to judge by actual
observation as to facts; otherwise we might "destroy a friend instead
of a foe."


TOAD GOT CLEAR.

A toad is discovered near the hives, and forthwith he is executed as a
bee-eater. "He ought to be killed for his looks, if nothing else!" He
is thus often sacrificed _really_ on account of his appearance, while
pretending he is a villain. It is true his "feathers" will not vie in
brilliancy with the plumage of the humming-bird, and do not gratify
ideality--therefore he is dispatched. The next week the complaint is
made that the little bugs, that he might have destroyed, "have eaten up
all the little cucumbers and cabbages." His food is probably small
insects. Whoever has seen him swallow bees, must have watched closer
than I ever did.


WASPS AND HORNETS NOT FAVORED.

As for the frequent visits of the black-wasp in the sunny days of
spring, but little can be said in their favor--they seem to have no
other object but to tease and irritate the bees. I never could discover
that they entered the hive for the purpose of plunder. They have
frequent battles with the bees, but I never saw any bees devoured or
carried off, nor even killed. After the first of June they are seldom
troublesome. The yellow wasp or hornet, that is around in autumn, is of
but little account; their object is honey, which they take when they
can get it, but are not apt to enter the hive among the bees.


ANTS--A WORD IN THEIR FAVOR.

Ants come in for a share of condemnation. This little industrious
insect shall have my endeavors for a fair hearing; I think I can
understand why they are so frequently accused of robbing bees. Many
bee-keepers are wholly ignorant, most of the time, of the real
condition of their stocks. Many causes independent of ants, induce a
reduction of population. Suppose the bees are so reduced as to leave
the combs unprotected, and the ants enter and appropriate some of the
honey to themselves, and should the owner come along just then and see
them engaged, "Ha! you are the rascals that have destroyed my bees,"
without a thought of looking for causes, beyond present appearances.
They are often unjustly accused by the farmer of injuring the growth of
his little trees, by causing the tender leaves to curl and wither.
Inquiries are often made in some of the agricultural papers for means
to destroy them, merely because they are found on them; when the real
cause of the mischief is with the plant louse, (aphis) that is upon the
leaves or stalk in hundreds, robbing them of their important juices,
and secreting a fluid greatly prized by the ants. By destroying the
lice, you remove all the attraction of the ants. The peculiar habits of
the small black ants, probably give rise to a suspicion of mischief in
this way. They live in communities of thousands--their nests are
usually in old walls, in old timber, under stones, and in the earth.
From their nests a string may be traced sometimes for rods, going
after, and returning laden with food. During a spell of wet weather,
such as would make the earth and many other places too damp and cold
for a nest, they look out for better quarters. The top or chamber of
our bee-hives affords shelter from rain. The animal heat from the bees
renders it perfectly comfortable. How then can we blame them for
choosing such a location, so completely answering all their wants? As
long as the bees are not disturbed, we can put up with it better. But
the careless observer having discovered their train to and fro from
their nest on the hive, exclaims: "Why, I have seen them going in a
continual stream to the hive after honey;" when a little scrutiny into
the matter would show that only the nest was on the top of the hive,
and they were going somewhere else for food; not one to be seen
entering the hive among the bees for honey, (at least I never could
detect it.)

When honey is unprotected by bees, or boxes of it placed where they can
have access, as a natural consequence, they will carry off some; but it
is easily secured.


SPIDER CONDEMNED.

Spiders are a source of considerable annoyance to the apiarian, as well
as to the bees; not so much on account of the number of bees consumed,
as their habit of spinning a web about the hive, that will occasionally
take a moth, and will probably entangle fifty bees the whilst. They are
either in fear of the bees, or they are not relished as food;
particularly, as a bee caught in the morning is frequently untouched
during the day. This web is often exactly before the entrance,
entangling the bees as they go out and return; irritating and hindering
them considerably. They often escape after repeated struggles. I have
removed a web from the same place every morning, for a week, that was
renewed at night with astonishing perseverance! I can generally look
out his hiding-place, which is in some corner near by, and dispatch
him. His redeeming qualities are few, and are more than balanced by the
evil, as far as I have discovered. Their sagacity in some instances
will find a place of concealment not easily discovered. At the approach
of cold weather, the box or chamber of the hive being a little warmer
than other places, will attract a great many there to deposit their
eggs. Little piles of webbing or silk may be seen attached to the top
of the hive, or sides of boxes. These contain eggs for the next year's
brood. This is the time to destroy them and save trouble for the
future.

If we combine into one phalanx all the depredators yet named, and
compare their ability for mischief with the wax moth, we shall find
their powers of destruction but a small item! Of the moth itself we
would have nothing to fear were it not for her progeny, that consist of
a hundred or a thousand vile worms, whose food is principally wax or
comb.

As the instinct of the flesh-fly directs her to a putrid carcass to
deposit her eggs, that her offspring may have their proper food, so the
moth seeks the hive containing combs, and where its natural food is at
hand to furnish a supply. During the day a rusty brown miller, with its
wings wrapped close around the body, may be often seen lying perfectly
motionless on the side of the hive on one corner, or the under edge of
the top, where it projects over--they are more frequent at the corners
than anywhere else, one-third of their length projecting beyond it;
appearing much like a sliver on the edge of a board that is somewhat
weather-beaten. Their color so closely resembles old wood, that I have
no doubt their enemies are often deceived, and let them escape with
their lives. As soon as daylight shuts out the view, and no danger of
their movements being discovered by their enemies, they throw off their
inactivity, and commence searching for a place to deposit their eggs,
and woe to the stock that has not bees sufficient to drive them from
the comb. Although their larvæ has a skin that the bee cannot pierce
with its sting, in most cases, it is not so with the moth, and of this
fact they seem to be aware, for whenever a bee approaches they dart
away with speed ten times greater than that of any bee, disposed to
follow! They enter the hive and dodge out in a moment, having either
encountered a bee, or fear they may do so. Now it needs no argument to
prove that when all our stocks are well protected, that it must be a
poor chance to deposit eggs, on the combs of such hives, where their
instinct has taught them is the proper place. But they _must_ leave
them somewhere. When driven from all the combs within, the next best
place is the cracks and flaws about the hive, that are lined with
propolis; and the dust and chips that fall on the floor-board of a
young swarm not full will be used. This last material is mostly wax,
and answers very well instead of comb. The eggs will here hatch and the
worms sometimes ascend to the combs; hence the necessity of keeping the
bottom brushed off clean. It will prevent those that are on the bottom
from going up; also the bees from taking up any eggs, if this should
happen to be the method. I can conceive of no other way by which they
get among the combs of a populous stock; where they are often detected,
having been deposited by some means. A worm lodged in the comb, makes
his way to the centre, and then eats a passage as he proceeds, lining
it with a shroud of silk, gradually enlarging it, as he increases in
size. (When combs are filled with honey, they work on the surface,
eating only the sealing.) In very weak families this silken passageway
is left untouched,--but removed by all the stronger ones. I have found
it asserted that "the worms would be all immediately destroyed by the
bees, were it not for a kind of dread in touching them until compelled
to by necessity." As the facts which led to this conclusion are not
given, and I can find none confirming it, perhaps I shall be excused if
I have no faith. On the contrary, I find to all appearance an
instinctive antipathy to all such intruders, and are removed
immediately when possessing the power.

When a worm is in a comb filled with brood, its passage being in the
centre, it is not at first discovered. The bees, to get it out, must
bite away half the thickness, removing the brood in one or two rows of
cells, sometimes for several inches. This will account for so many
immature bees found on the bottom board at morning, in the spring; as
well as in stocks and swarms but partially protected after the swarming
season.


INDICATIONS OF THEIR PRESENCE.

Sometimes a half dozen young bees, nearly mature, will be removed
alive, all webbed together, fastened by legs, wings, &c. All their
efforts for breaking loose prove unavailing. Also others that are
separate may be seen running about with their wings mutilated, or part
of their legs eaten off, or tied together! These generally are the
first symptoms of worms in our stock at this season. Although
unfavorable, it might be worse. It shows that the bees are not
discouraged yet,--that when finding the worms present, have sufficient
energy left to make an effort to rid themselves of the nuisance.


MANAGEMENT.

Should the apiarian now give them a little assistance for a few days,
they will soon be in a prosperous condition. The hive should be
frequently raised, and everything brushed out clean. If it is a new
swarm half full, that presents these indications, it should be turned
over a few times, perhaps once a week, till the worms are mastered; and
the corners below the bees examined for the cocoons, that will very
often be found there, and are easily detached and destroyed. In turning
over a hive part full, in warm weather, you should first observe the
position of the combs, and let the edges rest against the side of the
hive, otherwise they might bend over and break loose when the hive was
again set up, (by simply making a pencil mark across the top in the
direction of the combs, you may know any time after first looking).


CARE IN TURNING OVER HIVES.

When a hive is full of combs, the edges are usually attached sufficient
to steady them, and it is of less consequence which way it is turned,
yet in very warm weather the honey will run out of drone cells if
perpendicular.

In _very_ small swarms, hundreds of the young brood may be frequently
seen with their heads out of the cells, endeavoring to escape, but are
firmly held inside by these webs. I have known a few instances in such
circumstances, where it appeared as if the bees had cut off the whole
sheet of comb and let it drop, thereby ridding themselves of all
further trouble (or would be rid of it, if their owner only did his
part by taking out what fell down.)


OTHER SYMPTOMS OF WORMS.

But when the bees make no effort to dislodge the enemy or his works in
old stocks, the case is somewhat desperate! Instead of the foregoing
symptoms we must look for something entirely different. But few young
bees will be found. In their place we may find the fæces of the worms
dropped on the board. During winter and spring the bees, in biting off
the covering of cells to get at the honey, drop chips closely
resembling it. To detect the difference and distinguish one from the
other requires a little close inspection. The color of the fæces varies
with the comb on which they feed, from white to brown and black. The
size of these grains will be in proportion to the worm--from a mere
speck to nearly as large as a pin-head: shape cylindrical, with obtuse
ends: length about twice its diameter. By the quantity we can judge of
the number. If the hive is full of combs the lower ends may appear
perfect, while the middle or upper part is sometimes a mat of webs!

Whenever our stocks have become reduced from over-swarming or other
cause, this is the next effect in succession that we must expect. Here
is another important reason that we know the _actual_ condition of our
bees at all times; we can then detect the worms very soon after they
commence. In some instances we might save the stock by breaking out
most of the combs, leaving just enough to be covered by the bees. When
success attends this operation, it _must_ be done before the worms have
progressed to a thorough lodgment. When the stock is weak, and
appearances indicate the presence of many, it is generally the safest,
and will be the least trouble in the end, to drive out the bees at once
and secure the honey and wax. The bees when put into a new hive _may_
do a little, but if they should do nothing, it would be no worse. It
cannot be as bad any way as to have left them in the old hive till the
worms had destroyed all and matured a thousand or two moths in addition
to those otherwise produced, thereby multiplying the chances of damage
to other stocks a thousand-fold. It is probably remembered that I said
when bees are removed from a hive in warm weather, if it was not
infested with worms at the time, it soon would be, unless smoked with
sulphur.


WHEN THEY GROW LARGER THAN USUAL.

In a hive thus left without bees to interfere, the worms will increase
to one-half or two-thirds larger than where their right to the combs is
disputed. In one case they often have their growth, and actually wind
up in their cocoon when less than an inch in length: in the other they
will quietly fatten till they are an inch and a half long and as large
as a pipe-stem.


TIME OF GROWTH.

When first hatched from the egg, it requires very close inspection to
see them with the naked eye. The rapidity of growth depends on the
temperature in which they are, as much or more than their good living.
A few days in hot weather might develop the full-grown worm, while in a
lower temperature it would require weeks and even months in some cases,
perhaps from fall till spring.


TIME OF TRANSFORMATION.

The worm, after spinning its cocoon, soon changes to a chrysalis, and
remains inactive for several days, when it makes an opening in one end
and crawls out. The time taken for this transformation is also governed
by the temperature, although I think but few ever pass the winter in
this state. It is a rare thing to find a moth before the end of May,
and not many till the middle of June; but after this time they are more
numerous till the end of the season.


FREEZING DESTROYS WORMS, COCOON AND MOTH.

It is pretty well demonstrated that the moth, its eggs, larvæ and
chrysalis cannot pass the winter without warmth of some kind to prevent
their freezing to death. The following facts indicate this. I have
taken all the bees out of a hive in the fall, and without disturbing
the comb or honey, put it in a cold chamber where it could freeze
thoroughly. In the following March bees were again introduced, and when
not on a bench with some other stock that had worms, not a single
instance in forty cases has ever produced a worm before the middle of
June, or until the eggs of some moth matured in another hive has had
time to hatch. I have sometimes, instead of putting bees in these in
March, kept them till June for swarms, perfectly free from any
appearance of worms!


HOW THEY PASS THE WINTER.

But it is altogether a different thing with our hives in which bees are
wintered; they are seldom or never entirely exempt! Perhaps it is
impossible to winter bees without preserving some eggs of the moth or a
few worms at the same time. The perfect moth perhaps never survives the
winter; the only place that the chrysalis would be safe, I think must
be in the vicinity of the bees--and a good stock will never allow it
there--but eggs, it would appear, are suffered to remain. In the fall,
at the approach of cold weather, the bees are apt to leave the ends of
the combs exposed; the moth can now enter and deposit her eggs directly
upon them; these, together with what are carried in by means before
suggested, are enough to prevent losing the breed. The warmth generated
by the bees will keep these eggs from freezing and preserve their
vitality. When warm weather approaches in the spring, those nearest the
bees are probably hatched first, and commence depredations and are
removed by the bees. As the bees increase and occupy more comb, more
are warmed up and hatched. In this way, even a small family of bees
will hatch, and get rid of all the eggs that happen to be in their
combs, and not be destroyed. This is the time that the apiarian may be
of service in destroying the worms, as the bees get them on the floor.


STOCKS MORE LIABLE TO BE DESTROYED LAST OF SUMMER.

But in July and August it is different in this respect; a single moth
may enter the hive when exposed, and deposit her whole burden of
several hundred eggs, as in the other case, but the heat from the bees
is now unnecessary to hatch them. The weather at this season will make
any part of the hive warm enough to set her whole brood at work at
once, and in three weeks all may be destroyed! This, and the fact that
more moths exist now than before, may account for the greater number of
stocks being destroyed at this season. Yet it is considered extremely
bad management to allow honey or combs to be devoured by this
disgusting creature. A little care to know the condition of the stocks
_is necessary_ to prevent their getting the start. These duties should
be fully considered before we take the responsibility of the care of
bees.


WHEN BEES ARE SAFE.

The only condition when we can rest and feel safe is when _we know all
our stocks are full of bees_. Even the "moth-proof" hive containing
combs will be scented out by the moth, when there are no bees to guard
it. An argument to show that a moth can enter where a bee can go is
unnecessary, and a little observation, I think, will prove that her
eggs sometimes go where she is not allowed.


MEANS TO DESTROY THEM.

At this season, (July and August), it is a good plan to put a few
pieces of old dry combs near the hives, in a box, or other place, as a
decoy, where the moth may have access. She will deposit a great many of
her eggs here, instead of the hive, and can be easily destroyed. As we
cannot always have our bees in a situation to feel safe, it will be
well to adopt some of the means recommended to diminish the number of
moths. First destroy all the worms that can be found at any time,
particularly in spring; second, all cocoons that can be got at. A great
many worms can be enticed to web up, under a trap of elder, &c., when
it is an easy matter to dispatch them. Thirdly, destroy all the moths
possible that can be seen about the hive. They are very much like the
flea, "when you put your finger on him he is not there;" a careful move
is necessary to crush him at once, otherwise he darts away at the least
disturbance. Probably the most expeditious mode is to make them drunk.


MAKING THEM DRUNK, AND THEIR EXECUTION BY CHICKENS.

Mix with water just enough molasses and vinegar to make it palatable;
this is to be put in white saucers or other dishes, and set among the
hives at night. Like nobler beings, if not wiser, when once they have
tasted the fatal beverage, they seem to lose all power to leave the
fascinating cup; but give way to appetite and excitement till a fatal
step plunges them into destruction! The next morning finds them yet
wallowing in filth, weak and feeble. Whether they would recover from
the effects of their carousal if lifted out of the mire, and carefully
nursed like other specimens of creation, I never ascertained. With but
little trouble a chicken or two will learn to be on hand, and greedily
devour every one. Hundreds are caught in this way, although many other
kinds besides the bee-moth will be mixed with them. This drink may be
used till dried up, occasionally adding a little water; perhaps it is
better after fermenting. This recipe appeared some years ago in some
paper; I have forgotten where. Salt has been recommended to prevent the
mischief of the worms, as well as a benefit to the bees. I used it
pretty extensively for several years, as I thought without much
benefit, and got tired. I then tried salting a part, and let the rest
do entirely without, and found no difference in their prosperity. Since
then, some ten years ago, I abandoned its use altogether, and succeed
just as well.



CHAPTER XX.

MELTING DOWN OF COMBS.


THE CAUSE.

When extreme hot weather occurs immediately after the bees have been
gathering from a plentiful harvest for two or three weeks, or even
during the yield, the wax composing new combs is very liable to be
softened, till they break loose from their fastenings and settle to the
bottom.


EFFECTS.

Sometimes the injury is trifling, only a piece or two slipping down; at
other times the whole contents fall in a confused and broken mass, the
weight pressing out the honey, and besmearing the bees, which in that
situation creep out, and away, from the hive in every direction.

I once had some new stocks ruined, and several others injured by hot
weather, in this way, about the first of September, immediately after
the flowers of buckwheat. The bees, or most of them, being covered with
honey, together with what ran out of the hive, at once attracted bees
from the others to the spot, which carried off the entire contents in a
few hours. This was an uncommon occurrence; I have known but one season
in twenty-five years when it occurred after the failure of honey in the
flowers. It usually happens during a plenteous yield, and then other
stocks are not apt to be troublesome.


FIRST INDICATIONS.

The first indications of such an accident will be, the bees outside in
clusters, when the hive is perhaps only half or two-thirds full, and
the honey running out from the bottom, (this is when part has fallen.)


PREVENTION.

To prevent such occurrences as far as possible, ventilate by raising
the hives on little blocks at the corners, and _effectually protect
them from the sun_; and if necessary, wet the outside with _cold_
water. At the time of losing those before mentioned, I kept all the
rest of the young swarms wet through the middle of the day, and I have
no doubt but I saved several by this means. I had some trouble with
such as had only a piece or two come down, and started just honey
enough to attract other bees. It was not safe to close the hive to
prevent the robbers, as this would have made the heat still greater,
and been certain destruction.

The best protection I found, was to put around the bottom of the hive a
few stems of asparagus; this gave a free circulation of air, and at the
same time, made it very difficult for the robbers to approach the
entrance, without first creeping through this hedge and encountering
some bees that belonged to the hive; which, with this assistance, were
enabled to defend themselves till all wasting honey was taken up.

When the hive is nearly full, and but one or two sheets come down, the
lower edge will rest on the floor, and the other combs will keep it in
an upright position, until the bees fasten it again. It is generally as
well to leave such pieces as they are. If the hive is but half full or
little more, and such pieces are not kept perpendicular by the
remaining combs, they are apt to be broken and crushed badly, by
falling so far; and most of the honey will be wasted. To save this, it
will be necessary to remove it, (unless a dish can be made to catch
it). Be careful not to turn the hive on its side, and break the
remaining combs, if any are left. Such combs as contain brood and but
little honey, might be left for the brood to mature. Should the bees be
able to take the honey or not waste much, it might be advisable to
leave it, till the contents were taken up; it would greatly assist in
filling up. But these broken pieces should be removed before they
interfere with the combs extending to the bottom. A part of the bees
are generally destroyed, but the majority will escape; even such as are
covered with honey, (if they are not crushed) will clean it off and
soon be in working order, when others do not interfere officiously,
assisting to remove it. A good yield of honey is the best protection
against this disposition to pillage. After the first year combs become
thicker, and are not so liable to give way.



CHAPTER XXI.

FALL MANAGEMENT.


FIRST CARE.

When the flowers fail at the end of the season, the first thing
necessary is to ascertain which are the weakest stocks, and all that
cannot defend themselves should either be removed or reinforced. The
strength of all stocks is pretty thoroughly tested within a few days
after a failure of honey. Should any be found with too few bees for
defence, they are quite sure to be plundered. Hence the necessity of
action in season, that we may secure the contents in advance of the
robbers.


STRONG STOCKS DISPOSED TO PLUNDER.

Strong stocks, that during a yield have occupied every cell with brood
and honey, when it fails, will soon have empty cells left by the young
bees, hatching. These empty cells, without honey to fill them, appear
to be a source of much uneasiness. Although such hive and caps may be
well stored, I have ever found them to be the worst in the apiary, much
more disposed to plunder, than weaker ones with half the honey. As weak
stocks cannot be bettered now, it is best to remove them at once, and
put the temptation out of the way. Carelessness is but a sorry excuse,
for letting bees establish this habit of dishonesty. Should any stocks
be weak from disease, the consequences would be even more disastrous
than bad habits; the reasons why such impure honey should not go into
thrifty stocks, have already been given. If we want the least possible
trouble with our bees, none but the best should be selected for winter.
But what constitutes a good stock, seems to be but partially
understood; if we judge from the number lost annually, too many are
careless, or ignorant in the selection; supposing, perhaps, because a
stock has been good one winter and swarmed well, it must of course be
right; the mistake is often fatal.


BEES CHANGEABLE.

Bees are so changeable, especially in the summer and swarming season,
that we can seldom be certain what they are, by what they have been. It
is safest, therefore, _to know what they are now_.


REQUISITES FOR GOOD STOCKS.

The proper requisites for a good stock are a full hive of proper shape
and size, (viz., 2,000 inches,) well stored with honey; a large family
of bees, and in a healthy condition, which must be ascertained by
actual inspection. The age is not important till over eight years old.
Stocks possessing these points, can be wintered with but little
trouble. But it cannot be expected that all will be in this condition.
Many bee-keepers will wish to increase their stocks and keep all that
is practicable, by supplying any deficiency. I shall endeavor to make
it appear profitable to do so, until bees enough are kept in the
country, to get all the honey that is now wasted.

All can understand why it is a loss to have bees eat honey part of the
winter and then die--that the honey consumed might have been
saved--that it makes no great difference to the bees whether they are
killed in the fall or sacrificed in the winter. I am not an advocate
for fire and brimstone as the reward of all unfortunate stocks, and
shall recommend it only when its use will make it no worse. We will see
how far it can be dispensed with.


GREAT DISADVANTAGE OF KILLING THE BEES.

Those rustic bee-keepers who are in the habit of making their hives
very large, such as will hold from 100 to 140 lbs., and killing the
bees in the fall, and sending the honey to market, will probably
continue the use of sulphur, unless we can convince them of the greater
advantage of making the hive smaller and have fifty or eighty lbs. of
this honey in boxes which will sell for more than can be realized for
their larger hive full, and at the same time, save their bees for a
stock-hive, making a better return in the long run, than one hundred
dollars at interest. When hives are made the proper size, the honey
will not be an object sufficient to pay for destroying the bees.


SECTION OF COUNTRY MAY MAKE A DIFFERENCE IN WHAT POOR STOCKS NEED.

The kind of requisite to be supplied to our deficient stocks, will
probably depend on the section of country. Where the principal source
is clover and basswood, it will fail partially, at least, before the
end of warm weather.

Some poor or medium stocks will continue to rear brood too extensively
for their means, and exhaust their winter stores in consequence; such
will need a supply of honey. But where great quantities of buckwheat
are sown, cold weather follows almost immediately after this yield, and
stops the breeding. Consequently a scarcity of bees is more frequent
than honey. There are exceptions, of course; I am speaking of these
cases generally. My experience has mostly been in a section where this
crop is raised, and will say that there is not more than one season in
ten, but that the honey will be in proportion with the bees the first
of September; that is, if there are bees enough, there will be honey
enough.


WHEN BEES ARE NEEDED.

I have frequently had stocks with stores amply sufficient to carry a
good family through the winter, and yet too few bees to last till
January, or even to defend themselves from the robbers. Hence I am in
the habit of supplying bees oftener than honey.

I usually have some few hives with too little honey, as well as too few
bees. Now it is very plain if the bees of one or more of this class
were united with the first successfully, we should have a respectable
family. I have made additions to stocks in this way that proved
first-rate.


CAUTION.

Whenever we make additions in this manner, it would be well first to
ascertain what was the cause of a scarcity of bees; if it was
over-swarming or loss of queen, it is well enough--but if from disease,
reject them, unless the bees are to be transferred the next spring, and
then, when too many cells are occupied with dead brood, as the bees
cannot be successfully wintered.


PRINCIPAL DIFFICULTY.

The greatest difficulty in uniting two families or more in this manner,
is where they have to be taken from different places in the same
apiary; where the locations have been marked. It is sufficiently shown
that bees return to the old stand.

To prevent these results, it has been recommended "to set an empty hive
with some pieces of comb, fastened in the top in the place of the one
removed, to catch the bees that go back to the old stand, and remove
them at night for a few times, when they remain." This should be done
only when we cannot do better; it is considerable trouble; besides
this, we do not always succeed to our satisfaction.


HOW AVOIDED.

I like the plan of bringing them a mile or more for this purpose, and
have no after trouble about it. Two neighbors being that distance
apart, each having stocks in this condition might exchange bees, making
the benefit mutual. I have done so, and considered myself well paid for
the trouble. But latterly I have had several apiaries away from home,
and now manage without difficulty.


ADVANTAGES OF MAKING ONE GOOD STOCK FROM TWO POOR ONES.

This making one good stock out of two poor ones, cannot be too highly
recommended; aside from its advantages, it relieves us from all
disagreeable feelings in taking life, that we can with but little
trouble preserve.


TWO FAMILIES TOGETHER WILL NOT CONSUME AS MUCH AS IF SEPARATE.

Even when a stock already contains bees enough to make it safe for
winter, another of the same number of bees may be added, and _the
consumption of honey will not be five lbs. more than one swarm would
consume alone_. If they should be wintered in the cold, the difference
might not be one pound. Why more bees do not consume a proportionate
quantity of honey, (which the experience of others as well as myself
has thoroughly proved), is a mystery, unless the greater number of bees
creates more animal heat, and being warm, eat less, is a solution,
(which if it is, is a strong reason for keeping bees warm in winter.)


AN EXPERIMENT.

Notwithstanding all this, I cannot recommend making a _good_ stock
better by adding the bees from another good one as a source of profit.
I tried it a few times. I had purchased some large hives for market,
and wished to dispose of the bees without sulphur, and try the
experiment of uniting two or more. The next spring when they commenced
work such double stocks promised much; but when the swarming season
arrived, the single swarms, such as were good and had just about bees
enough, were in the best condition, in ordinary seasons. Whether this
was owing to the circumstance of there being already bees enough that
were beginning to crowd and interfere with each other's labors, and
less brood raised in consequence, or to some other reason, I cannot
say. I have often noticed, (as others have), that stocks which have
cast no swarms, are no better the next spring than others. The same
cause might operate in both cases. Therefore it would appear
unnecessary to unite two or more _good swarms_, unless it is to spare
our feelings in destroying the bees. The two extremes may generally be
avoided, and not have too many or too few bees.


SEASON FOR OPERATING.

The season for operating is, generally, when all the brood has matured
and left the cells. The exceptions are when there are not bees enough
to protect the stores; it may then be necessary, immediately after the
failure of honey.

Col. H. K. Oliver, of Salem, Mass., is said to be the inventor of the
fumigator, an instrument to burn fungus (_puff-ball_). By the aid of
this the smoke is blown in the hive, paralyzing the bees in a few
minutes; when they fall to the bottom, apparently dead, but will
recover in a few minutes, on receiving fresh air.


THE FUMIGATOR.

I am indebted to a communication from J. M. Weeks, published on page
151 of the Cultivator for 1841, for this method. The description of the
fumigator that I constructed will vary a trifle from his, but will
retain the principle.  I obtained a tin tube four inches long, and two
in diameter. Next, I made a stopper of soft wood, three inches long, to
exactly fit one end of the tube when driven in half an inch, and
secured it by little nails driven through the tin. Through the centre
of this stopper I made a hole one-fourth of an inch in diameter. To
prevent this hole filling up, the end in the tube was covered over with
wire cloth, made a little convex. The end of this stopper was cut down
to about half an inch, tapering it from the tin. For the other end a
similar piece of wood is fitted, though a little longer, and not to be
fastened, as it must be taken out for every operation. The outer end of
this is cut down into a shape to be taken into the mouth, or attached
to the pipe of a bellows.  (I fitted them in the turning lathe, but
have seen them fixed very nicely without.) It could all be made of tin;
but then it is necessary to use solder, which is liable to melt and
cause leaks.

[Illustration: FUMIGATOR.]

"The puff-balls must not be too much injured by remaining in the
weather, and should be picked, if possible, just before they are ripe
and burst open. When not thoroughly dry, put them in the oven after the
bread is out." When used, the cuticle or rind must be carefully
removed; ignite it by a lamp or coal (it will not blaze in burning),
blow it, and get it thoroughly started, before putting it in the tube.
Put in the stopper, and blow through it; if it smokes well, you are
ready to proceed. When it does not burn freely, unstop and shake it
out. The dry air is much better than moist breath at the commencement.


DIRECTIONS FOR UNITING TWO FAMILIES.

The hive to receive the bees is inverted, the other set over it right
end up, all crevices stopped to prevent the escape of the smoke. Now
insert the end of the fumigator into a hole in the side of the hive
(which if not made before will need to be now); blow into the other
end, this forces the smoke into the hive; in two minutes you may hear
the bees begin to fall. Both hives should be smoked; the upper one the
most, as we want all the bees out of that. The other only needs enough
to make the scent of the bees similar to those introduced. At the end
of eight or ten minutes, the upper hive may be raised, and any bees
sticking between the combs brushed down with a quill. The two queens in
this case are of course together; one will be destroyed, and no
difficulty arise. But if either of them is a young one, and you have
been convinced by some "bee-doctor" that such are much more prolific,
and happen to know which hive contains her, and wish that one to be
preserved, you can do so by varying the process a little. Instead of
inverting one hive, set them both on a cloth right side up, and smoke
the bees; the queens are easily found, while they are all paralyzed;
then put the bees all together. The hive should now have a thin cloth
tied over the bottom, to prevent the escape of the bees. Before they
are fully recovered, they seem rather bewildered, and some of them get
away. Set the hive right end up, and raise it an inch; the bees drop on
the cloth, and fresh air passing under soon revives them. In from
twelve to twenty-four hours, they may be let out.

Families put together in this way will seldom quarrel (not more than
one in twenty), but remain together, defending themselves against
intruders as one swarm.

I once had a stock nearly destitute of bees, with abundant stores for
wintering a large family. I had let it down on the floor-board, and was
on the lookout for an attack. The other bees soon discovered this
weakness, and commenced carrying off the honey. I had brought home a
swarm to reinforce them only the day before, and immediately united
them by means of the fumigator. The next morning I let them out,
allowing them to issue only at the hole in the side of the hive. It was
amusing to witness the apparent consternation of the robbers that were
on hand for more plunder; they had been there only the day before, and
had been allowed to enter and depart without even being questioned. But
lo! a change had come over the matter. Instead of open doors and a free
passage, the first bee that touched the hive was seized and very rudely
handled, and at last dispatched with a sting. A few others receiving
similar treatment, they began to exercise a little caution, then tried
to find admission on the back side, and other places; and attempted one
or two others on either side, perhaps thinking they were mistaken in
the hive; but these being strong, repulsed them, and they finally gave
it up. I mention this to show how easy it is, with a little care, to
prevent robberies at this season. Too many complaints are made about
bees being robbed; it is very disagreeable. Suppose that _none were
plundered through carelessness_; this complaint would soon be a rare
thing.


UNITING WITH TOBACCO SMOKE.

By the use of tobacco smoke, bees may be united with nearly the same
success. First, smoke the two to be united, thoroughly; disturb them
and smoke again, that all may become partially drunk, and acquire the
same scent. Then invert both hives, and with your pruning tools, cut
the combs down on the sides of the hive, and across the top, and take
out one comb at a time with the bees on it, and brush them with a quill
into the other hive; they immediately go down among the combs, without
once thinking it necessary to sting you. When done, the bees are to be
confined, the same as in the other method. I do not like this method as
well as the first, and do not resort to it when I can get the
puff-ball. The bees are more liable to disagree, and it compels me to
take out the comb, which I do not always like to do at the time. To
avoid it, I have tried to drive them, but when the hive is only part
full of combs, or contains but few bees, it is a slow job; and more so
in cool weather.


CONDITION OF STOCKS IN 1851.

The latter part of the summer of 1851 was very dry and cold; the yield
of buckwheat honey was not a tenth of the usual quantity; the
consequence was, that none but early swarms had sufficient honey for
winter; twenty-five pounds is required to make it _safe_ in this
section. I had over thirty young swarms with less than that quantity.
Feeding for winter I avoid when I can; they would not winter as they
were; and yet I made the most of them good stocks for the next summer
by the following plan.


HOW THEY WERE MANAGED.

I had about twenty old stocks with diseased brood, and but few bees,
yet _honey enough_. Now this honey appears healthy enough for the old
bees, and fatal only to the young brood.

I transferred the bees of these new swarms to the old stocks with black
comb and diseased brood. The bees were thus wintered on honey of but
little account any way, and all that was in the others, new and
healthy, was saved. These new hives were set in a cold dry place for
winter; _right end up_, to prevent much of the honey from dripping out
of the cells; some will leak then, but not as much as when the hive is
bottom up. Honey that runs out, when the hive is bottom up, will soak
into the wood at the base of the combs; this will have a tendency to
loosen the fastenings, and render them liable to fall, &c.

The next March the bees were again transferred from the old to the new
hives. My method is as follows: As the combs in the hive to receive the
bees are rather cold, I set them by the fire, or in a warm room, for
several hours previous. I take a warm room before a window, and as some
few bees fly off, they will collect there. The new hive is turned
bottom up on the floor; the old one on a bench by the side of it,
having smoked the bees to keep them quiet. One comb at a time is taken
out, and the bees brushed into the new hive; (a little smoke will keep
them there). When through, I get the few on the window, and tie over a
cloth to confine them, and keep them warm for a few hours longer.
Paralyzing with puff-ball will answer instead, but they do not always
all fall out of the combs when the hive is filled to the bottom, and it
is possible that if a few were left, the queen might be one. Also a
very few bees are worth saving at this season, and the combs might have
to be broken out at last, for this purpose.

When a good-sized family is put in a hive containing fifteen or twenty
pounds of honey, and near half full of clean new comb, they are about
as sure to fill up and cast a swarm, as another that is full and has
wintered a swarm.


CAUSE OF THEIR SUPERIOR THRIFT.

One cause of superior thrift may be found in the circumstance, that all
moth eggs and worms are frozen to death, and the bees are not troubled
with a single worm before June. No young bees have to be removed to
work them out. Nearly every young bee that is fed and sealed up, comes
forth perfect, and of course makes a vast difference in the increase.


SWARMS PARTLY FILLED PAY BETTER THAN TO CUT OUT THE HONEY.

Any person wishing to increase his stocks to the utmost, will find this
plan of saving all part-filled hives, of much more advantage than to
break it out for sale. Suppose you have an old stock that needs
pruning, and have neglected it, or it has refused to swarm, and give
you a chance without destroying too much brood. You can let it be, and
put on the boxes; perhaps get twenty-five pounds of cap honey; and then
winter the bees as described, and in the spring transfer them to the
new combs. Again, if there is no stocks to be transferred in the
spring, keep them till the swarming season. If a swarm put into an
empty hive would just fill it, the same swarm put into one containing
fifteen pounds of honey, it seems plain, would make that number of
pounds in boxes. The advantage is, in the comparative value of box or
cap honey over that stored in the hive; the difference being from
thirty to a hundred per cent.


ADVANTAGES IN TRANSFERRING.

I would now like to show the advantages I derived in transferring the
twenty swarms before mentioned. We will suppose that each family, from
the first of October till April, consumed twenty pounds of honey. That
in the centre combs, where there is most bee-bread, &c., is eaten
first; if any is left, it is at the top and outside. If I had attempted
to take out and strain this twenty pounds in the fall, it would have
been so mixed with dead brood, and bee-bread, that I probably should
have rejected most of it. The remainder, when strained, might have been
five pounds, not more. The market price for it is about ten cents per
pound; amount fifty cents. We will say the new hive kept through the
winter to receive the bees in the spring contained fifteen pounds; this
would also have averaged about ten cents per pound, amounting to $1.50.
All that a stock of this kind costs me appears to be just $2.00, and
worth at least $5.00. The advantage in changing twenty would be $60.00.
The labor of transferring will offset against the trouble of straining,
preparing, and the expense of getting the honey to market.


ANOTHER METHOD OF UNITING TWO FAMILIES.

I have occasionally adopted yet another method of making a good stock
from two poor ones, which the reader may prefer. When all your old
stocks have been reinforced that need it, and you still have some
swarms with too few bees and too little honey for safety as they are,
two or more can be united. The fact, which has been thoroughly tested,
that two families of bees, when united and wintered in one hive, will
consume but little, if any more, than each of them would separately, is
a very important principle in this matter. If each family should have
fifteen pounds of honey, they would consume it all, and probably starve
at last, after eating thirty pounds. But if the contents of both were
in one hive, it would be amply sufficient, and some to spare in the
spring.


UNITING COMB AND HONEY AS WELL AS BEES.

The process of uniting them is simple. Smoke both the stocks or swarms
thoroughly, and turn them over. Choose the one with the straightest
combs, or the one nearest full, to receive the contents of the other;
trim off the points of the combs to make them square across, and this
one is ready; remove the sticks from the other, and with your tools
take out the combs with the bees on as before directed, one at a time,
and carefully set them on the edges of the other; if the shape will
admit it, let the edges match; if not, let them cross. Small bits of
wood or rolls of paper will be needed between them, to preserve the
right distance. When both hives are of one size, the transferred combs
will exactly fit, if you are careful to place them as they were before.
You will now want to know, "what is to prevent these combs from falling
out when the hive is turned over?" This hive is to remain bottom up in
some dark place for some time, or till spring. (See method of wintering
bees.) The bees will immediately join these combs fast; the hive being
inverted, the honey in these combs will be consumed first; and when the
hive is again set out in spring, it will be a rare occurrence for any
pieces to drop out. Should any pieces project beyond the bottom of the
hive, they may be trimmed off even after they are fastened, any time
before setting out. An additional cross-stick may pass under the bottom
of the combs, to assist in holding them, if you desire. You will
probably never discover any difference in the subsequent prosperity in
consequence of the joining or crossing of the combs in the middle. I
have had them in this way, when they were among the most prosperous of
my stocks. As this operation is to be put off till November, it will be
an advantage in another way; that is, families of the same apiary can
be united, and will mostly forget the old location by spring, and no
difficulty arise by returning to the old stand, etc.


WHEN FEEDING SHOULD BE DONE FOR STOCK HIVES.

In some sections of country the _honey_ is more frequently wanting than
bees, or comb, and some seasons in this; in such cases, it will be
found an advantage to feed, until enough is stored for winter. This
should be done in September or October. But if they lack comb as well
as honey, and you wish to try feeding, (which I seldom do lately,) it
should be done if possible in warm weather, as they cannot work combs
to advantage in the cold. While feeding bees, it requires a great deal
of caution to prevent others from scenting the honey, and their
contentions about it. The safest place is on the top of the hive, with
a good cap over; but they will not work quite as fast, especially if
the weather is cool. The next best place is under the bottom in the
manner described in Chapter IX.

Setting out honey to feed all at once, I condemn wholly. These
disadvantages attend it: strong stocks that do not need an ounce, will
get two or three pounds, while those weaker ones, needing it more, will
not get one. Nearly every stock, in a short time, will be fighting.
Probably the first bee that comes home with a load, will inform a
number of its fellows that a treasure is close at hand. A number will
sally out immediately, without waiting for particular directions for
finding it; and mistaking other hives for the place, alight there, are
seized and probably dispatched. As soon as the honey given them is
gone, the tumult is greatly increased, and great numbers are destroyed.
If any of your neighbors near you have bees, you must expect to divide
with them.

If the honey to be fed is in the comb, and your hives are not full, and
they are to be wintered in the house, bottom up, it may be done at any
time through the winter, merely by laying pieces with honey on those in
the hive. The bees readily remove the contents into their own combs;
when empty, remove them and put in more until they have a full supply.
They will join such pieces of comb to their own; yet there will be no
harm in breaking them loose. The principal objection to feeding in this
way, will be found in the tendency to make them uneasy and disposed to
leave the hive, when we want them as quiet as possible, A thin muslin
cloth, or other means, will be necessary to confine them to the hive.

I have now given directions to avoid killing any family of bees worth
saving, if we choose.

When such as need feeding have been fed, and all weak families made
strong by additions, etc., but little more fall work is needed in the
apiary. It is only when you have weak stocks, unfit to winter, that it
is necessary to be on the lookout every warm day to prevent pillage.



CHAPTER  XXII.

WINTERING BEES.


There is almost as much diversity of opinion with respect to wintering
bees as in the construction of hives, and about as difficult to
reconcile.


DIFFERENT METHODS HAVE BEEN ADOPTED.

One will tell you to keep them warm, another to keep them cold; to keep
them in the sun, out of the sun, bury them in the ground, put them in
the cellar, the chamber, wood-house, and other places, and no places at
all; that is, to let them remain as they are, without any attention.
Here are plans enough to drive the inexperienced into despair. Yet I
have no doubt but that bees have been sometimes successfully wintered
by all these contradictory methods. That some of these methods are
superior to others, needs no argument to illustrate. But what method
_is best_, is our province to inquire. Let us endeavor to examine the
subject without prejudice to bias our judgment.


THE IDEA OF BEES NOT FREEZING HAS LED TO ERRORS IN PRACTICE.

By close observation we shall probably discover that the assertion so
often repeated, that bees have never frozen except when without honey,
has led to an erroneous practice.


APPEARANCE OF BEES IN COLD WEATHER.

We will first endeavor to examine the condition of a stock left to
nature, without any care, and see if it affords any hints for our
guidance, when to assist and protect with artificial means.

Warmth being the first requisite, a family of bees at the approach of
cold weather crowd together in a globular form, into a compass
corresponding to the degree of cold; when at zero it is much less than
at thirty above. Those on the outside of this cluster are somewhat
stiffened with cold; while those inside are as brisk and lively as in
summer. In severe weather every possible space within their circle is
occupied; even each cell not containing pollen or honey will hold a
bee. Suppose this cluster is sufficiently compact for mutual warmth,
with the mercury at 40, and a sudden change brings it down to zero, in
a few hours, this body of bees, like most other things, speedily
contracts by the cold. The bees on the outside, being already chilled,
a portion of them that does not keep up with the shrinking mass, is
left exposed at a distance from their fellows, and receive but little
benefit of the warmth generated there; they part with their vitality,
and are lost.


HOW PART OF THE SWARM IS FROZEN.

A good family will form a ball or circle about eight inches in
diameter, generally about equal every way, and must occupy the spaces
between four or five combs. As combs must separate them into divisions,
the two outer ones are smallest, and most exposed of any; these are
often found frozen to death in severe weather. Should evidence be
wanting from other sources to show that bees will freeze to death, the
above would seem to furnish it. It is said, "that in Poland bees are
wintered in a semi-torpid state, in consequence of the extreme cold."
We must either doubt the correctness of this relation, or suppose the
bee of that country a different insect from ours--a kind of semi-wasp,
that will live through the winter, and eat little or nothing.  The
reader can have no difficulty in deciding which is the most probable,
whether _bees are bees_ throughout the world, endowed with the same
faculties and instincts, or that the facts as they are, are not
precisely given, especially when we see what our own apiarians tell us
about their never freezing.

Here I might use strong language in contradiction; but as I am aware
that such a course is not always the most convincing, I prefer the test
of close observation. If bees will freeze, it is important to know it,
and in what circumstances.


HOW A SMALL FAMILY MAY ALL FREEZE.

Suppose a quart of bees were put in a box  or hive where all the cells
were filled and lengthened out with honey; the spaces between the combs
would be about one-fourth of an inch--only room for one thickness of
bees to spread through. The combs would perhaps be one and a half or
two inches thick. All the warmth that could be generated then, would be
by one course or layer of bees, an inch and a half apart. Although
every bee would have food in abundance without changing its position,
the first turn of severe weather would probably destroy the whole.
This, it may be said, "is an unnatural situation." I will admit that it
is; the case was only supposed for illustration. I know that their
winter quarters are among the brood combs, where the hatching of the
brood leaves most of the cells empty; and the space between the combs
is half an inch; a wise and beautiful arrangement; as ten times the
number of bees can pack themselves within a circle of six inches, as
can in the other case; and in consequence the same number of bees can
secure much more animal heat, and endure the cold much better; but a
_small_ family, even here, will often be found frozen, as well as
starving.


FROST AND ICE SOMETIMES SMOTHER BEES.

Besides freezing, there are other facts to be observed in stocks which
stand in the cold. If we examine the interior of a hive containing a
medium-sized swarm, on the first severely cold morning, except in the
immediate vicinity of the bees, we shall find the combs and sides of
the hive covered with a white frost. In the middle of the day, or as
soon as the temperature is slightly raised, this begins to melt,--first
next to the bees, then at the sides. A succession of cold nights will
prevent the evaporation of this moisture; and this process of freezing
and thawing, at the end of a week or two, will form icicles sometimes
as large as a man's finger, attached to the combs and the sides of the
hive. When the bottom of the hive is close to the floor, it forms a
sealing around the edges, perfectly air-tight, and your bees are
smothered. I have frequently heard bee-keepers say in these cases, "The
storm blew in, and formed ice all round the bottom, and froze my bees
to death." Others that have had their bees in a cold room, finding them
thus, "could not see how the water and ice could get there any way;
were quite sure it was not there when carried in," &c. Probably they
never dreamed of its being accounted for philosophically, and to
analyze anything pertaining to bees would be rather small business. But
what way can it be accounted for?


FROST AND ICE IN A HIVE ACCOUNTED FOR.

Physiologists tell us "that innumerable pores in the cuticle of the
human body are continually throwing off waste or worn out matter; that
every exhalation of air carries with it a portion of water from the
system, in warm weather unperceived, but will be condensed into
particles large enough to be seen in a cold atmosphere." Now, if
analogy be allowed here, we will say the bee throws of waste matter and
water in the same way. Its food being liquid, nearly all will be
exhaled--in moderate weather it will pass off, but in the cold it is
condensed--the particles lodge on the combs in form of frost, and
accumulate as long as the weather is very severe, a portion melting in
the day, and freezing again at night.


THE EFFECT OF ICE OR FROST ON BEES AND COMB.

When the bees are not smothered, this water in the hive is the source
of other mischief. The combs are quite certain to mould. The water
mould or dampness on the honey renders it thin, and unhealthy for the
bees, causing dysentery, or the accumulation of fæces that they are
unable to retain. When the hive contains a very large family, or a very
small one, there will be less frost on the combs,--the animal heat of
the first will drive it off; in the latter there will be but little
exhaled.


FROST MAY CAUSE STARVATION.

This frost is frequently the cause of medium or small families starving
in cold weather, even when there is plenty of honey in the hive.
Suppose all the honey in the immediate vicinity of the cluster of bees
is exhausted, and, the combs in every direction from them are covered
with frost; if a bee should leave the mass and venture among them for a
supply, its fate would be as certain as starvation. And without timely
intervention of warmer weather, they _must_ perish!


OTHER DIFFICULTIES.

Should they escape starving, there is another difficulty often
attending them in continued cold weather. I said that small families
exhaled but little. Let us see if we can explain the effect.

There is not sufficient animal heat generated to exhale the aqueous
portion of their food.  The philosophy that explains why a man in warm
blood and in profuse perspiration would throw off or exhale more
moisture than in a quiet state, will illustrate this. The bees in these
circumstances must retain the water with the excrementitious part,
which soon distends their bodies to the utmost, rendering them unable
to endure it long. Their cleanly habits, that ordinarily save the combs
from being soiled, is not a sure protection now, and they are compelled
to leave the mass very often in the severest weather, to expel this
unnatural accumulation of fæces. It is frequently discharged even
before leaving the comb, but most of it at the entrance; also some
scattered on the front side of the hive, and a short distance from it.
In a moderately warm day, more bees will issue from a hive in this
condition than from others; it appears that a part of them are unable
to discharge their burden--their weight prevents their flying--they get
down and are lost. When cold weather is too long continued, they cannot
wait for warm days to leave, but continue to come out at any time; and
not one of such can then return. The cluster inside the hive is thus
reduced in numbers till they are unable to generate heat sufficient to
keep from freezing. With the indications attendant upon such losses, my
own observation has made me somewhat familiar, as the following
conversation will illustrate.


FURTHER ILLUSTRATIONS.

A neighbor who wished to purchase some stock hives in the fall,
requested my assistance in selecting them. We applied to a perfect
stranger; his bees had passed the previous winter in the open air. I
found on looking among them that he had lost some of them from this
cause, as the excrement was yet about the entrance of one old
weather-beaten hive, that was now occupied by a young swarm, and was
about half filled with combs.

I saw at once what had been the matter, and felt quite confident that I
could give its owner a correct history of it. "Sir," said I, "you have
been unfortunate with the bees that were in this hive last winter; I
think I can give you some particulars respecting it."

"Ah, what makes you think so? I would like to hear you guess; to
encourage you, I will admit that there has been something rather
peculiar about it."

"One year ago you considered that a good stock-hive; it was well filled
with honey, a good family of bees, and two or three years old or more.
You had as much confidence in its wintering as any other; but during
the cold weather, somehow, the bees unaccountably disappeared, leaving
but a very few, and they were found frozen to death. You discovered it
towards spring, on a warm day. When you removed the combs, you probably
noticed a great many spots of excrement deposited on them, as well as
on the sides of the hive, particularly near the entrance. Also one-half
or more of the breeding cells contained dead brood, in a putrid state;
and this summer you have used the old hive for a new swarm."

"You are right, sir, in every particular. Now, I would like to know
what gave you the idea of my losing the bees in that hive? I can see
nothing peculiar about that old hive, more than this one," pointing to
another that also contained a new swarm. "You will greatly oblige me if
you will point out the signs particularly."

"I will do so with pleasure" (feeling quite willing to give him the
impression that I was "posted up" on this subject, notwithstanding it
savored strongly of boasting).

I then directed his attention to the entrance in the side of the hive,
where the bees had discharged their fæces, on the moment they issued,
until it was near the eighth of an inch thick, and two or three inches
broad; that yet remained, and just began to cleave off. "You see this
brown substance around this hole in the hive?"

"Yes, it is bee-glue (_propolis_); it is very common on old hives."

"I think not; if you will examine it closely, you will perceive it is
not so hard and bright; it already begins to crumble; bee-glue is not
affected by the weather for years."

"Just so, but what is it, and what has that to do with your
guess-work?"

"It is the excrement of the bees. In consequence of a great many cells
containing dead brood, which the bees could not enter, they were unable
to pack themselves close enough to secure sufficient animal heat to
exhale or drive off the water in their food, it was therefore retained
in their bodies till they were distended beyond endurance--they were
unable to wait for a warm day--necessity compelled them to issue daily
during the coldest weather, discharging their fæces the moment of
passing the entrance, and part of them before. They were immediately
chilled, and could not return; the quantity left about this entrance
shows that a great many must have come out. That they came out in cold
weather is proved by its being left on the hive, because in warm
weather they _leave_ the hive for this purpose."

"This is a new idea; at present it seems to be correct; I will think it
over. But how did you know that it was not a new swarm; that it was
well filled?"

"When looking under it just now, I saw that combs of a dark color had
been attached to the sides near the bottom, below where those are at
present; this indicates that it had been full, and the dark color that
it was not new. Also, a swarm early and large enough to fill such a
hive the first season, would not be very likely to be affected by the
cold in this way."

"Why not? I think this hive was crowded with bees as much as any of my
new swarms."

"I have no doubt they appeared so; but we are very liable to be
deceived in such cases, by the dead brood in the combs. A
moderate-sized family in such a hive will make more show than some
larger ones that have empty cells to creep into, and can pack closer."

"But how did you know about the dead brood?"

"Because old stocks are thus often reduced and lost."

"What were the indications of its being filled with honey?"

"Combs are seldom attached to the side of the hive farther down than
they are filled with honey. In this hive the combs had been attached to
the bottom, consequently must have been full. Another thing, unless the
family is very much reduced, the hive is generally well stored, even
when diseased."

"Why did you suppose it was near spring before I discovered it?"

"I took the chances of guessing. The majority of bee-keepers, you know,
are rather careless, and when they have fixed their bees for winter,
seldom give them much more attention, till they begin to fly out in the
spring."

"But what should I have done had I discovered the bees coming out?"

"As it was affected with dead brood, it was but little use to do
anything; you would have lost it eventually. But if it had been a stock
otherwise healthy, and was affected in this way only because it was a
small family, or the severity of the weather, you could have taken it
to a warm room, and turned it bottom up; the animal heat would then
convert the most of the water contained in their food into vapor; that
would rise from the hive, and the bees could retain the excrementitious
portion without difficulty till spring."

"I suppose you must get along without losing many through the winter,
if I may judge by your confident explanations."

"I can assure you I have but little fear on this head. If I can have
the privilege of selecting proper stocks, I will engage not to lose one
in a hundred."

"How do you manage? I would be glad to obtain a method in which I could
feel as perfectly safe as you appear to."

"The first important requisite is to have all good ones to start with.
Enough weak families are united together till they are strong, or some
other disposition made of them." I then gave him an outline of my
method of wintering, which I can confidently recommend to the reader.


ACCUMULATION OF FÆCES DESCRIBED BY SOME WRITERS AS A DISEASE.

This accumulation of fæces is considered by many writers as a
disease--a kind of dysentery. It is described as affecting them towards
spring, and several remedies are given. Now if what I have been
describing is not the dysentery, why I must think I never had a case of
it; but I shall still persist in guessing it to be the same, and
suppose that inattention with many must be the reason that it is not
discovered in cold weather, at the time that it takes place. Some
stocks may be badly affected, yet not lost entirely, when moderate
weather will stop its progress. When a remedy is applied in the spring,
long after the cause ceases to operate, it would be singular if it was
not effectual. I have no doubt but some have taken the natural
discharge of fæces, that always takes place in spring when the bees
leave the hive, for a disease. Others, when looking for a cause for
diseased brood, and found the combs and hive somewhat besmeared, have
assigned this as sufficient; but according to my view, have reversed
it, giving the effect before the cause.


THE AUTHOR'S REMEDY.

For a time, I supposed that this moisture on the combs gradually mixed
with the honey, making it thin, and that the bees eating so much water
with their food, would affect them as described. Some experiments that
followed, induced me to assign cold as the cause, as I always found,
when I put them where it was sufficiently warm, that an immediate cure
was the result, or at least, it enabled them to retain their fæces till
set out in the spring.


BURYING BEES.

Burying bees in the earth below the frost, has been recommended as a
superior method of wintering, for small families. I have known it
confidently asserted, that they would lose nothing in weight, and no
bees would die. I found, in testing it, that a medium quantity of honey
sufficed, and but very few were lost, perhaps less than by any other
method. Yet the combs were mouldy, and unfit for further use. There was
no escape for the vapor and dampness of the earth. This did not satisfy
me; it only cured "one disease by instituting another." I saved the
bees, (and perhaps some honey), but the combs were spoiled.


EXPERIMENTS OF THE AUTHOR TO GET RID OF THE FROST.

I wished to keep them warm, and save the bees as well as honey, and at
the same time, get rid of the moisture. I found that a large family
expelled it much better than small ones; and if all were put together
in a close room, the animal heat from a large number combined, would be
an advantage to the weak ones, at least,--this proved of some benefit.
Yet I found on the sides of a glass hive, that large drops of water
would stand for weeks.


SUCCESS IN THIS MATTER.

The following suggestion then came to my relief. If this hive was
bottom up, what would prevent all this vapor as it arises from the bees
from passing off? (It always rises when warm, if permitted.) The hive
was inverted; in a few hours the glass was dry.

This was so perfectly simple, that I wondered I had not thought of it
before, and wondered still more that some one of the many intelligent
apiarians had never discovered it. I immediately inverted every hive in
the room, and kept them in this way till spring; when the combs were
perfectly bright, not a particle of mould to be seen, and was well
satisfied with the result of my experiment. Although I was fearful that
more bees would leave the hives when inverted, than if right side up,
yet the result showed no difference. I had now tried both methods, and
had some means of judging.


BEES WHEN IN THE HOUSE SHOULD BE KEPT PERFECTLY DARK.

When not kept perfectly dark, a few would leave the hives in either
case. I have found it much better to make the room dark to keep the
bees in the hive, than to tie over them a thin muslin cloth, as that
prevents a free passage of the vapor, and a great number of full stocks
were not at all satisfied in confinement; and were continually
worrying, and biting at the cloth, till they had made several holes
through it for passages out. Thus the little good was attended by an
evil, as an offset. Even wire cloth put over to confine them, which
would be effectual, would not save bees enough to pay expense. I have
thus wintered them for the last ten years, and am extremely doubtful if
a better way can be found.[17] For several years I made use of a small
bed-room in the house, made perfectly dark, in which I put about 100
stocks. It was lathed and plastered, and no air admitted, except what
might come through the floor. It was single, and laid rather close,
though not matched.

      [17] I was so well pleased with my success, especially with small
      families, that I detailed the most important points in a
      communication to the Dollar Newspaper, Philadelphia, published
      November, 1848.


A ROOM MADE FOR WINTERING BEES.

In the fall of 1849 I built a room for this purpose; the frame was
eight by sixteen feet square, and seven high, without any windows. A
good coat of plaster was put on the inside, a space of four inches
between the siding and lath was filled with saw-dust; under the bottom
I constructed a passage for the admission of air, from the north side;
another over head for its exit, to be closed and opened at pleasure, in
moderate weather, to give them fresh air, but closed when cold, and so
arranged as to exclude all the light.

A partition was extended across near the centre. This was to prevent
disturbing the whole by letting in light when carrying them out in the
spring. By closing the door of this partition, those in one room only
need be disturbed at once.


MANNER OF STOWING AWAY BEES.

Shelves to receive the hives were arranged in tiers one above the
other; they were loose, to be taken down and put up at pleasure.
Suppose we begin at the back end: the first row is turned directly on
the floor, a shelf is then put across a few inches above them, and
filled, and then another shelf, still above, when we again begin on the
floor, and continue thus till the room is full; or if the room is not
to be filled, the shelves may be fixed around the sides of the room in
two or three courses. This last arrangement will make it very
convenient to inspect them at any time through the winter, yet they
should be disturbed as little as possible. The manner of stowing each
one is to open the holes in the top, then lay down two square sticks,
such as are made by splitting a board, of suitable length, into pieces
about an inch wide. The hive is inverted on these; it gives a free
circulation through the hive, and carries off all the moisture as fast
as generated.


TEMPERATURE OF ROOM.

The temperature of such a room will vary according to the number and
strength of the stocks put in; 100 or more would be very sure to keep
it above the freezing point at all times. Putting a very few into such
a room, and depending on the bees to make it warm enough, would be of
doubtful utility. If these means will not keep the proper temperature,
probably some other method would be better. All full stocks would do
well enough, as they would almost any way. Yet I shall recommend
housing them whenever practicable. If the number of stocks is few, let
the room be proportionably small.[18] It is the smallest families that
are most trouble: if they are too cold, it may be known by bees leaving
the hive in cold weather, and spots of excrement on the combs; they
should then have some additional protection; close part or all of the
holes in the top, cover the open bottom partially or wholly, and
confine to the hive as much as possible the animal heat; when these
means fail, it may be necessary to take them to a warm room, during the
coldest weather.

      [18] As an additional proof that this method of inverting hives
      in the house for winter is valuable, I would say that Mr. Miner,
      author of the American Bee-Keeper's Manual, seems fully to
      appreciate it. In. the fall of 1850, I communicated to him this
      method; giving my reasons for preferring it to the cold method
      recommended in his Manual. The trial of one winter, it appears,
      satisfied him of its superiority, so much so that within a year
      from that time he published an essay recommending it; but advised
      confining the bees with muslin, &c.


TOO MUCH HONEY MAY SOMETIMES BE STORED.

After the flowers fail, and all the brood has matured and left the
combs, it sometimes happens that a stock has an opportunity of
plundering, and rapidly filling all those cells that had been occupied
with brood during the yield of honey, and which then effectually
prevents their storing in them. This, then, prevents close packing,
which is all-important for warmth. Although a large family, as much
care is needed as with the smaller ones. Also such as are affected with
diseased brood should receive extra attention for the same reason.

Some bee-keepers are unwilling to risk the bold measure of inverting
the hive, but content themselves by merely opening the holes in the
top; this is better than no ventilation, but not so effectual, as all
of the moisture cannot escape. There are some who cannot divest
themselves of the idea, that if the hive is turned over, the bees must
also stand on their heads all winter!

Rats and mice, when they find their way into such room, are less bold
with their mischief than if the hive is in its natural position.


MANAGEMENT OF ROOM TOWARDS SPRING.

A few warm days will often occur, towards spring, before we can get our
bees out. In these cases, a bushel or two of snow or ice pounded up
should be spread on the floor; it will absorb and carry off as it melts
much of the heat, that is now unnecessary, and will keep them quiet
much longer than without it; (provision for getting rid of this water
should be made when putting down the floor.)


TIME FOR SETTING OUT BEES.

The time for carrying out bees is generally in March, but some seasons
later. A warm pleasant day is the best, and one quite cold, better than
one only _moderately_ warm.

After their long confinement, the light attracts them out at once,
(unless very cold air prevents), and if the rays of a warm sun do not
keep them active, they will soon be chilled and lost.

Some bee-keepers take out their stocks at evening. If we could be
always sure of having the next day a fair one, it would probably be the
best time; but should it be only moderate, or cloudy, it would be
attended with considerable loss--or if the next day should be quite
cold, but few would leave, and then the only risk would be to get _a
good day_, before one that was just warm enough to make them leave the
hive, but not quite enough to enable them to return.


NOT TOO MANY STOCKS TAKEN OUT AT ONCE.

When too many are taken out at once, the rush from all the hives is so
much like a swarm, that it appears to confuse them. Some of the stocks
by this means will get more bees than actually belong to them, while
others are proportionably short, which is unprofitable, and to equalize
them is some trouble; yet it may be done. Being all wintered in one
room, the scent or the means of distinguishing their own family from
strangers, becomes so much alike, that they mix together without
contention.


FAMILIES MAY BE EQUALIZED.

By taking advantage of this immediately, or before the scent has again
changed, and each hive has something peculiar to _itself_, you can
change the stands of very weak and very strong families.

To prevent, as far as possible, some of these bad effects, I prefer
waiting for a fair day to begin, and then not until the day has become
sufficiently warm to make it safe from chill.


SNOW NEED NOT ALWAYS PREVENT CARRYING OUT BEES.

I am not particular about the snow being gone--if it has only lain long
enough to have melted a part of it, it is "terra firma" to a bee, and
answers equally well as the bare earth. When the day is right, about
ten o'clock I put out twelve or fifteen, taking care that each hive
occupies its old stand, at the same time endeavoring to take such as
will be as far apart as possible; (to make this convenient, they should
be carried in in the manner that you wish them to come out.) When the
rush from these hives is over, and the majority of the bees has gone
back, I set out as many more about twelve o'clock, and when the day
continues fair, another lot about two. In the morning, while cool, I
move from the back to the first apartment, about as many as I wish to
set out in a day, except a few at the last.

To do this in the middle of the day, while warm, would induce a good
many bees to leave the hive, while the light was admitted, and which
would be lost. It will be supposed generally that their long
confinement makes them thus impatient to get out; but I have frequently
returned stocks during a cold turn of weather after they had been out,
and always found such equally as anxious to come out, as those which
had been confined throughout the winter; without the airings, I have
kept them thus confined, for five months, without difficulty! The
important requisites are, sufficient warmth and perfect darkness.


DOES NOT ANALOGY PROVE THAT BEES SHOULD BE KEPT WARM IN WINTER?

Opposition to this method of wintering will arise with those who have
always thought that bees must be kept cold; "the colder the better." I
would suggest for their consideration the possibility of some analogy
between bees and some of the warm-blooded animals--the horse, ox, and
sheep, for instance, that require a constant supply of food, that they
may generate as much caloric as is thrown off on the cold air. This
seems to be regulated by the degree of cold, else why do they refuse
the large quantity of tempting provender in the warm days of spring,
and greedily devour it in the pelting storm? The fact is pretty well
demonstrated, that the quantity of food needed for the same condition
in spring, is much less when protected from the inclemency of the
weather, than when exposed to the severe cold. The bee, unlike the
wasp, when once penetrated with frost, is dead--_their temperature must
be kept considerably above the freezing point, and to do this, food is
required_. Now if the bees are governed by the same laws, and cold air
carries off more heat than warm, and their source of renewing it is in
the consumption of honey in proportion to the degree of cold, common
sense would say, keep them warm as possible. As a certain degree of
heat is necessary in all stocks, it may take about such a quantity of
honey to produce it, and this may explain why a small family requires
about the same amount of food as others that are very large.


THE NEXT BEST PLACE FOR WINTERING BEES.

A _dry_, warm cellar is the next best place for wintering them; the
apiarian having one perfectly dark, with room to spare, will find it a
very good place, in the absence of a room above ground. If a large
number was put in, some means of ventilation should be contrived for
warm turns of weather. I know an apiarian, who by my suggestion has
wintered from sixty to eighty stocks in this way, for the last six
years, with perfect success, not having lost one. Another has wintered
thirty with equal safety.

As for burying them in the earth, I have not the least doubt, if a dry
place should be selected, the hive inverted, and surrounded with hay,
straw, or some substance to absorb the moisture, and protected from the
rain, at the top of the covering, that perfect success would attend the
experiment. But this is only theory; when I tried the experiment of
burying, and had the combs mould, the hives ware right side up.


EVILS OF WINTERINGS IN THE OPEN AIR CONSIDERED.

As a great many bee-keepers will find it inconvenient, or be unable to
avail themselves of my method of wintering, it will be well enough to
see how far the evils of the open air, which we have already glanced
at, may be successfully avoided. I am told by those who have tried
wintering them in straw hives, that in this respect they are much safer
than those made of boards; probably the straw will absorb the moisture.
But as these hives are more trouble to construct, and their shape will
prevent the use of suitable boxes for surplus honey, this one advantage
will hardly balance the loss. They are said also to be more liable to
injury from the moth. We want a hive that will unite advantageously as
many points as possible.

It should be remembered that bees always need air, especially in the
cold.[19] With this in view, we will try to dispose of the vapor or
frost. If the hive is raised sufficient to let it out, it will let in
the mice; to prevent which, it should be raised only about one-fourth
of an inch. The hole in the side should be nearly covered with wire
cloth to keep out the mice; but give a passage for the bees; otherwise
they collect here, endeavoring to get out, and remain till chilled, and
thus perish by hundreds. The boxes on the top must be removed, but not
the cap or cover; the holes all opened, to let the vapor pass up into
the chamber; if this is made with perfectly close joints, so that no
air escapes, it should be raised a very little; otherwise not. The
moisture will condense on the sides and top, when it melts will follow
the sides to the bottom, and pass out; the rabbeting around the top of
the hive will prevent its getting to the holes, and down among the
bees. It will be easily comprehended, that a hole between each two
combs at the top, (as mentioned in the subject of putting on the
boxes,) will ventilate the hive much better than where there is but one
or two, or where there is a row of several, and all are between two
combs.

      [19] It is presumed that the inexperienced will soon learn to
      distinguish such bees, as die from old age or natural causes,
      from those affected by the cold.


BUT LITTLE RISK WITH GOOD STOCKS.

All _good stocks_ may be wintered in this way, with but little risk in
most situations. Whether in the bleak north-wind, buried in a
snow-bank, or situated warm and pleasant, it will make no great
difference. The mice cannot enter; the holes give them air, and carry
off moisture, &c. But second-rate stocks are not equally safe in cold
situations.


EFFECT OF KEEPING SECOND-RATE STOCKS OUT OF THE SUN.

It has been strongly urged, without regard to the strength of the
stock, to keep them all out of the sun; because an occasional warm day
would call out the bees, when they get on the snow, and perish; this is
a loss, to be sure, but there is such a thing as inducing a greater one
by endeavoring to avoid this. I have said in another place that second
rate or poor stocks might occasionally starve, with plenty of stores in
the hive, on account of frosty combs. If the hive is kept from the sun,
in the cold, the periods of temperate weather might not occur as often,
as the bees would exhaust the honey within their circle or cluster. But
on the contrary, when the sun can strike the hive, it warms up the
bees, and melts the frost more frequently. The bees may then go among
their stores and obtain a supply, generally, as often as needed. We
seldom have a winter without sunny days enough for this purpose; but
should such an one occur, stocks of this class should be brought into a
warm room, once in four or five days, for a few hours at a time, to
give them a chance to get at the honey. Stocks much below second-rate
cannot be wintered successfully in this climate; the only place for
them is the warm room. I have known bees thoroughly covered in a
snow-drift, and their owner was at considerable trouble to shovel the
snow away, fearing it would smother them. This is unnecessary, when
protected from the mice and ventilated as just directed; a snow-bank is
about as comfortable a place as they can have, except in the house.
When examined a short time after being so covered, the snow for a space
of about four inches on every side of the hive is found melted, and
none but quite poor stocks would be likely to suffer with this
protection. A little snow around the bottom, without a vent in the side
of the hive, might smother them.


EFFECTS OF SNOW CONSIDERED.

As for bees getting on the snow, I apprehend that not many more are
lost there, than on the frozen earth; that is, in the same kind of
weather. I have seen them chilled, and lost on the ground by hundreds,
when a casual observer would not have noticed them; whereas, had they
been on the snow, at the distance of several rods, every bee would have
been conspicuous. Snow is not to be dreaded as much as chilly air.
Suppose a hive stands in the sun throughout the winter, and bees are
allowed to leave when they choose, and a portion are lost on the snow,
and that it was possible to number all that were lost by getting
chilled, throughout the season, on the bare earth--the proportion (in
my opinion) lost on the snow would not be one in twenty. A person that
has not closely observed during damp or chilly weather, in April, May,
or even the summer months, has no adequate conception of the number.
Yet, I do not wish to be understood that it is of no consequence what
are lost on the snow, by any means. On the contrary, a great many are
lost, that might be saved with proper care. But I would like to impress
the fact, that frozen earth is not safe without warm air, any more than
snow, when crusted, or a little hard. Even when snow is melting, it is
solid footing for a bee; they can and do rise from it, with the same
ease as from the earth. Bees that perish on snow in these
circumstances, would be likely to be lost if there was none.


STOCKS TO BE PROTECTED ON SOME OCCASIONS.

The worst time for them to leave the hive is immediately after a new
snow has fallen, because if they light on it then, it does not sustain
their weight; and they soon work themselves down out of the rays of the
sun, and perish. Should it clear off pleasant, after a storm of this
kind, a little attention will probably be remunerated. Also, when the
weather is moderately warm, and not sufficiently so to be safe, they
should be kept in, whether snow is on the ground or otherwise.

For this purpose, a wide board should be set up before the hive to
protect it from the sun, at least above the entrance in the side. But
if it grows sufficiently warm so that bees leave the hive when so
shaded, it is a fair test by which to tell when it will do to let them
have a good chance to sally out freely, except in cases of a new snow,
when it is advisable to confine them to the hive. The hive might be let
down on the floor-board, and the wire-cloth cover the passage in the
side, and made dark for the present; raising the hive at night again,
as before. I have known hundreds of stocks wintered successfully
without any such care being taken, and the bees allowed to come out
whenever they chose to do so. Their subsequent health and prosperity
proving that it is not altogether ruinous. It has been recommended to
enclose the whole hive by a large box set over it, and made perfectly
dark, with means for ventilation, &c. (A snow-bank would answer equally
well, if not better.) For large families it would do well enough, as
would also other methods. But I would much rather take the chances of
letting them all stand in the sun, and issue as they please, than to
have the warmth of the sun entirely excluded from the moderate-sized
families. I never knew a whole stock lost by this cause alone.[20] Yet,
I have known a great many starved, merely because the sun was not
allowed to melt the frost on the combs, and give them a chance to get
at their stores.

      [20] Vide other causes of loss, a few pages back.


DO THE BEES EAT MORE WHEN ALLOWED TO COME OUT OCCASIONALLY IN WINTER?

Besides the loss of bees on the snow when standing in the sun, and
taking an airing occasionally, there are some economical bee-keepers
who urge this disadvantage, "that every time bees come out in winter
they discharge their excrement, and eat more honey in consequence of
the vacant room." What a ridiculous absurdity it would be to apply this
principle to the horse, whose health, strength, and vital heat is
sustained by the assimilation of food! and the farmer is not to be
found who would think of saving his provender by the same means. That
bees are supported in cold weather on the same principle is indicated
strongly, if not conclusively.

Is it not better (if what has been said on the subject of wintering
bees is correct) to keep our bees warm and comfortable when
practicable, as a means of saving honey?

To winter bees in the best manner, considerable care is required.
Whenever you are disposed to neglect them, you should bear in mind that
one early swarm is worth two late ones; their condition in spring will
often decide this point. Like a team of cattle or horses when well
wintered, they are ready for a good season's work, but when poorly
wintered have to recruit a long time before they are worth much.



CHAPTER XXIII.

SAGACITY OF BEES.


ARE NOT BEES DIRECTED ALONE BY INSTINCT?

On this subject I have but little to say, as I have failed to discover
anything uncommonly remarkable, separate and distinct in one swarm,
that another would not exhibit. I have found one swarm guided alone by
instinct, doing just what another would under the same circumstances.

Writers, not contented with the astonishing results of instinct, with
their love of the marvellous, must add a good share of reason to their
other faculties,--"an adaptation of means to ends, that reason alone
could produce." It is very true, without close inspection, and
comparing the results of different swarms in similar cases, one might
arrive at such conclusion. It is difficult, as all will admit, "to tell
where instinct ends, and reason begins." Instances of sagacity, like
the following, have been mentioned. "When the weather is warm, and the
heat inside is somewhat oppressive, a number of bees may be seen
stationed around the entrance, vibrating their wings. Those inside will
turn their heads towards the passage, while those outside will turn
theirs the other way. A constant agitation of air is thus created,
thereby ventilating the hive more effectually." _All full stocks do
this in hot weather._


WHAT THEY DO WITH PROPOLIS.

"A snail had entered the hive and fixed itself against the glass side.
The bees, unable to penetrate it with their stings, the cunning
economists fixed it immovably, by cementing merely the edge of the
orifice of the shell to the glass with resin, (propolis), and thus it
became a prisoner for life." Now the instinct that prompts the
gathering of propolis in August, and filling every crack, flaw, or
inequality about the hive, would cement the edges of the snail-shell to
the glass, and a small stone, block of wood, chip, or any substance
that they are unable to remove, would be fastened with it in the same
manner. The edges or bottom of the hive, when in close proximity to the
bottom, is joined to it with this substance. Whatever the obstacle may
be, it is pretty sure to receive a coating of this. The stoppers for
the holes at the top are held in their places on the same principle;
and the unaccountable sagacity that once fastened a little door, might
possibly be nothing more than the same instinct.

Another principle, I think, will be found to be universal with them,
instead of sagacious reasoning.

Whenever the combs in a hive have been broken, or when combs have been
added, as was mentioned in the chapter on fall management, the first
duty of the bees appears to be to fasten them as they are; when the
edges are near the side of the hive, or two combs in contact, a portion
of wax is detached and used for joining them together, or to the side.


MENDING BROKEN COMBS.

Where two combs do not touch, and yet are close together, a small bar
is constructed from one to the other, preventing any nearer approach.
(This may be illustrated by turning the hive a few inches from the
perpendicular after being filled with combs in warm weather.)


MAKING PASSAGES TO EVERY PART OF THEIR COMBS.

Should nearly all the combs in the hive become detached from any cause,
and lie on the bottom in one "grand smash of ruin," their first steps
are, as just described, pillars from one to the other to keep them as
they are. In a few days, in warm weather, they will have made passages
by biting away combs where they are in contact, throughout every part
of the mass; little columns of wax below, supporting the combs
above,--irregular, to be sure, but as well as circumstances admit. Not
a single piece can be removed without breaking it from the others, and
the whole will be firmly cemented together. A piece of comb filled with
honey, and sealed up, may be put in a glass box with the ends of these
cells so sealed, touching the glass. The principle of allowing no part
of their tenement to be in a situation inaccessible, is soon
manifested. They immediately bite off the ends of the cells, remove the
honey that is in the way, and make a passage next to the glass, leaving
a few bars from it to the comb, to steady and keep it in its position.
A single sheet of comb lying flat on the bottom-board of a populous
swarm is cut away under side, for a passage in every direction,
numerous little pillars of wax being left for its support. How any
person in the habit of watching their proceedings, with any degree of
attention, could come at the conclusion that the bees raised such comb
by mechanical means and then put under the props for its support, is
somewhat singular. Their efforts united for such a purpose like
reasonable beings, I never witnessed.

These things, considered as the effect of instinct, are none the less
wonderful on that account. I am not sure but the display of wisdom is
even greater than if the power of planning their own operations had
been given them.

I have mentioned these, to show that a course of action called forth by
the peculiar situation of one family, would be copied by another in a
similar emergency, without being aware of its ever being done before.
Were I engaged in a work of fiction, I might let fancy reign and
endeavor to amuse, but this is not the object. Let us endeavor then to
be content with truth, and not murmur with its reality. When we take a
survey of the astonishing regularity with which they construct their
combs without a teacher, and remember that the waxen material is formed
in the rings of their body, that for the first time in life, without an
experienced leader's direction, they apply a claw to detach it, that
they go forth to the fields and gather stores unbidden by a tyrant's
mandate, and throughout the whole cycle of their operations, one law
and power governs. Whoever would seek mind as the directing power, must
look beyond the sensorium of the bee for the source of all we behold in
them!



CHAPTER XXIV.

STRAINING HONEY AND WAX.


When about to remove the contents of a hive, I have never found it
necessary to use all the precautions often recommended to prevent the
access of bees. I have seen it stated that a room in which there was a
chimney open, would be unsuitable, as the bees would scent the honey,
and thus find their way down into the room. I never was thus troubled
by their perpendicular travelling. It is true, if the day was warm, and
a door or window was standing open, the bees would find their way in
during a scarcity of honey. But with doors and windows closed no
difficulty need be apprehended.


METHODS OF REMOVING COMBS FROM THE HIVE.

The most convenient way to remove combs from the hive is to take off
one of its sides, but this is apt to split the boards, if it was
properly nailed, and injure it for subsequent use. With tools such as
have been described, it may be done very nicely, and leave the hive
whole. The chisel should have the bevel all on one side, like those
used by carpenters. When you commence, turn the flat side next the
board of the hive, and the bevel crowded by the combs will follow it
close the whole length; with the other tool they are cut across the
top, and readily lifted out. If preferred, they may be cut across near
the centre and take out half a sheet at a time; this is sometimes
necessary on account of the cross-sticks.


DIFFERENT METHODS OF STRAINING HONEY.

Such combs as are taken from the middle or vicinity of brood-cells, are
generally unfit for the table; such should be strained. There are
several methods of doing it. One is, to mash the comb and put it in a
bag, and hang it over some vessel to catch the honey as it drains out.
This will do very well for small quantities in warm weather, or in the
fall before there is any of it candied. Another method is to put such
combs into a colander, and set this over a pan, and introduce it into
an oven after the bread is out. This melts the combs. The honey and a
portion of the wax run out together. The wax rises to the top and cools
in a cake. It is somewhat liable to burn, and requires some care. Many
prefer this method, as there is less taste of bee-bread, no cells
containing it being disturbed, but all the honey is not certain to
drain out without stirring it. If disposed, two qualities may be made,
by keeping the first separate. Another method is merely to break the
combs finely, and put them into a colander, and allow the honey to
drain out without much heat, and afterwards skim off the small
particles that rise to the top, or when very particular, pass the honey
through a cloth, or piece of lace. But for large quantities, a more
expeditious mode is to have a can and strainer, made for the purpose,
where fifty pounds or more can be worked out at once. The can is made
of tin, twelve or fourteen inches deep, by about ten or twelve
diameter, with handles on each side at the top, for lifting it. The
strainer is just enough smaller to go down inside the can; the height
may be considerably less, providing there are handles on each side to
pass out at the top; the bottom is perforated with holes like a
colander, combs are put into this, and the whole set into a kettle of
boiling water, and heated without any risk of burning, until all the
wax is melted, (which may be ascertained by stirring it,) when it may
be taken out. All the wax, bee-bread, &c., will rise in a few minutes.
The strainer can now be raised out of the top and set on a frame for
the purpose, or by merely tipping it slightly on one side it will rest
on the top of the can. It might be left to cool before raising the
strainer, were it not liable to stick to the sides of the can; the
honey would be full as pure, and separate nearly as clean from the wax
and bee-bread, &c. When raised out before cooling, the contents should
be repeatedly stirred, or considerable honey will remain. Two qualities
may be made by keeping the first that runs through separate from the
last, (as stirring it works out the bee-bread). Even a third quality
maybe obtained by adding a little water, and repeating the process.
This is worth but little. By boiling out the water, without burning,
and removing the scum, it will do to feed bees. By adding water until
it will just bear a potato, boiling and skimming, and letting it
ferment, it will make metheglin, or by letting the fermentation proceed
it will make vinegar. Honey that has been heated thoroughly, will not
candy as readily as when strained without heat. A little water may be
added to prevent its getting too hard; but should it get so in cold
weather, it can at any time be warmed, and water added until it is of
the right consistence.


GETTING OUT WAX--DIFFERENT METHODS.

Several methods have been adopted for separating the wax. I never found
any means of getting out the _whole_. Yet I suppose I came as near it
as any one. Some recommend heating it in an oven, similar to the method
of straining honey through the colander, but I have found it to waste
more than when melted with water. A better way for small quantities, is
to half fill a coarse stout bag with refuse comb and a few
cobble-stones to sink it, and boil it in a kettle of water, pressing
and turning it frequently till the wax ceases to rise. When the
contents of the bag are emptied, by squeezing a handful, the particles
of wax may be seen, and you may thereby judge of the quantity thrown
away. For large quantities the foregoing process is rather tedious. It
can be facilitated by having two levers four or five feet long and
about four inches wide, and fastened at the lower end by a strong
hinge. The combs are put into a kettle of boiling water, and will melt
almost immediately; it is then put into the bag, and taken between the
levers in a wash-tub or other large vessel and pressed, the contents of
the bag shaken, and turned, several times during the process, and if
need be returned to the boiling water and squeezed again. The wax, with
a little water, is now to be remelted and strained again through finer
cloth, into vessels that will mould it into the desired shape. As the
sediment settles to the bottom of the wax when melted, a portion may be
dipped off nearly pure without straining.

Wax in cool weather may be whitened in a short time in the sun, but it
must be in very thin flakes; it is readily obtained in this shape by
having a very thin board or shingle, which should be first thoroughly
wet, and then dipped into pure melted wax; enough will adhere to make
it the desired thickness, and will cool instantly on being withdrawn.
Draw a knife along the edges, and it will readily cleave off. Exposed
to the sun in a window or on the snow, it will become perfectly white,
when it can be made into cakes for market, where it commands a much
higher price than the yellow. It is said there is a chemical process
that whitens it readily, but I am not acquainted with it.



CHAPTER XXV.

PURCHASING STOCKS AND TRANSPORTING BEES.


If the reader has no bees, and yet has had interest or patience to
follow me thus far, it is presumptive evidence that he would possess
the requisite perseverance to take charge of them. It would be well,
however, to remember the anxieties, perplexities, and time necessary to
take the proper care, as well as the advantages and profit.

But if you are disposed to try the experiment, very likely some
directions for a commencement would be acceptable.


WHY THE WORD LUCK IS APPLIED TO BEES.

There has been so much uncertainty in stock of this kind, that the word
_luck_ has been made to express too much. Some have been successful,
while others have failed entirely; this has suggested the idea that
_luck_ depended on the manner that the stocks were obtained; and here
again there seems to be a variety of opinions, as is the case always,
when a thing is guessed at. One will assert that the "fickle dame" is
charmed into favor by stealing a stock or two to begin with, and
returning them after a start. Another, (a little more conscientious,
perhaps) that you must take them without _liberty_, to be sure, but
leave an equivalent in money on the stand. Another, that the only way
to get up an effectual charm, is to exchange sheep for them; and still
another says, that _bees must always be a gift_. I have had all these
methods offered me gratis, with gravity, suitable to make an
impression. And, finally, there has yet another method been found out,
and that is, when you want a few stocks of bees go and buy them, yes,
and pay for them too, in dollars and cents, or take them for a share of
the increase for a time, if it suits your pecuniary resources best. And
you need not depend on any _charm_ or mystic power for your success--if
you do, I cannot avoid the unfavorable prediction of a failure. It is
true that a few have accidentally prospered for a few years; I say
accidentally, because when they have no true principles of management,
it must be the result of accident. It is a saying with some, that "one
man can't have luck but few years at once," and others none at all,
although he tries the whole routine of charms. Nearly twenty years ago,
when my respected neighbor predicted a "turn in my luck, because it was
always so," I could not understand the force of this reasoning, unless
it belonged to the nature of bees to deteriorate, and consequently run
out. I at once determined to ascertain this point. I could understand
how a farmer would often fail to raise a crop, if he depended on chance
or luck for success, instead of fixed natural principles. It was
possible that bees might be similar. I found that in good seasons the
majority of people had luck, but in poor seasons, the reverse, and when
two or three occurred in succession, then was the time to lose their
luck. It was evident, then, if I could pass in safety the poor seasons
by any means, I should do well enough in good ones.[21] The result has
given me but little reason to complain. My advice therefore is, that
reliance should be placed on proper management, instead of luck,
arising from the manner the first stock was obtained. Should any one
feel disposed to make you a present of a stock or two of bees, I would
advise you to accept the offer and be thankful, discarding all
apprehension of a failure on that account. Or if any one is willing you
should take some on shares, this is a cheap way to get a start, and you
have no risk of loss in the old stock. Yet if bees prosper, the
interest on the money that stocks cost is a mere trifle in comparison
to the value of increase, and you have the same trouble. On the other
hand, the owner of bees can afford to take care of a few hives more,
for half the profits, which he has to give if another takes them; this
is apt to be the case, especially, with such as have no faith in charms.

      [21] There are sections of country where the difference in
      seasons is less than in this.


RULE IN TAKING BEES FOR A SHARE.

The rule generally adopted for taking bees is this. One or more stocks
are taken for a term of years, the person taking them finding hives,
boxes, and bestowing whatsoever care is necessary, and returning the
old stocks to the owner with half the increase and profits.


A MAN MAY SELL HIS "LUCK."

There are yet a few persons who refuse to sell a stock of bees, because
it is "bad luck." There is often some grounds for this notion. It might
arise under the following circumstances. Suppose a person has a half
dozen hives, three extra good, the others of the opposite extreme. He
sells for the sake of the better price his three best; there is but
little doubt but his best "luck" would go too! But should his poorest
be taken, the result would be different, without doubt.

But there are cases where an apiarian has more stocks than he wishes to
keep. (It has been the case with myself frequently.) Persons wishing to
sell, are the proper ones of which to buy. Purchasers seldom want any
but first-rate stocks, they are generally cheapest in the end. There is
usually a difference of about a dollar in the spring and fall prices,
and five and six dollars are common charges. I have known them sell at
auction at eight, but in some sections they are less.


FIRST-RATE STOCKS RECOMMENDED TO BEGIN WITH.

For a beginning then, I would recommend purchasing none but first-rate
stocks; it will make but little difference in the risk, whether you
obtain them in the spring, or fall, if you have read my remarks on
winter management with attention; I have already said the requisites
for a good stock for winter, were a numerous family and plenty of
honey, and that the cluster of bees should extend through nearly all
the combs, &c. To avoid as far as possible diseased brood, find an
apiary where it has never made its appearance, to make purchases. There
are some who have lost bees by it, and yet are totally ignorant of the
cause. It would be well, therefore, to inquire if any stocks have been
lost, and then for the cause--be careful that secondary are not
mistaken for primary causes.


OLD STOCKS ARE GOOD AS ANY, IF HEALTHY.

When it appears that all are exempt, (by a thorough examination, if not
satisfied without,) you need not object to stocks two or three years
old; they are just as good as any, sometimes better, (providing they
have swarmed the season previous, according to one author; because such
always have young queens, which are more prolific than old ones, that
will be in all first swarms).

Old stocks are as prosperous as any, as long as they are healthy, yet
they are more liable to become diseased.


CAUTION RESPECTING DISEASED BROOD.

When no apiary from which to purchase can be found, but where the
disease _has made_ its appearance, and you are necessitated to purchase
from such, or not at all, you cannot be too cautious about it. It would
be safest in this case to take none but young swarms, as it is not so
common for them to be affected the first season, yet they are not
always exempt. But here, again, you may not be allowed to take all
young stocks; in which case let the weather be pretty cold, the bees
will be further up among the combs, and give a chance to inspect the
combs. At this season, say not earlier than November, all the healthy
brood will be hatched. Sometimes, a few young bees may be left that
have their mature shape, and probably had been chilled by sudden cold
weather--these are not the result of disease, the bees will remove them
the next season, and no bad results follow. In warm weather a
satisfactory inspection can be had no other way, but by the use of
tobacco smoke. Be particular to reject all that are affected with the
disease in the least; better do without, than take such to begin with.
(A full description has been given of this disease in another place.)


RESULT OF IGNORANCE IN PURCHASING.

A neighbor purchased thirteen stock-hives; six were old ones, the
others swarms of the last season. As the old hives were heavy, he of
course thought them good; either he knew nothing of the disease, or
took no trouble to examine; five of the six old ones were badly
affected. Four were lost outright, except the honey; the fifth lasted
through the winter, and then had to be transferred. He had flattered
himself that they were obtained very cheaply, but when he made out what
his good ones cost, he found no great reason, in this respect, for
congratulation.


SIZES OF HIVES IMPORTANT.

Another point is worthy of consideration: endeavor to get hives as near
the right size as possible, _viz._, 2,000 cubic inches; better too
large than too small. If too large, they may be cut off, leaving them
the proper size. But yet, it often makes an ungainly shape, being too
large square for the height. As the shape probably makes no difference
in the prosperity of the bees, the appearance is the principal
objection, after being cut off.

An acquaintance had purchased a lot of bees in very large hives, and
called on me to know what to do with them, as he feared such would not
swarm well in consequence; I told him it would be doubtful, unless he
cut them off to the right size.

"Cut 'em off! how can that be done? there is bees in 'em."

"So I expected, but it can be done nearly as well as if empty."

"But don't you get stung dreadfully?"

"Not often: if it is to be done in warm weather, I smoke them well
before I begin; _in very cold weather_ is the best time, then it is
unnecessary; simply turn the hive bottom up, mark off the proper size,
and with a sharp saw take it off without trouble."

"Some are filled with combs; you don't cut off such, do you?"

"Certainly; I consider all the room for combs in a hive over 2,000
inches as worse than lost."

"What will you ask to cut mine off? If I could see it done once, I
might do it next time."

"The charge will be light; but if you intend to keep bees, you should
learn to do everything pertaining to them, and not be dependent on any
one; I did it before I ever saw or heard of its being done." I then
gave him full directions how to manage, but could not persuade him to
undertake.


HOW LARGE HIVES CAN BE MADE SMALLER.

A short time after, I attended, on a cold day, with a sharp saw,
square, &c. I found his hives fourteen inches square inside, and
eighteen deep, holding about 3,500 inches. Of this square, a little
more than ten inches in height, would make just the right size. To work
convenient, I inverted the hive on a barrel, set on end, marked the
length, and sawed it off, without a bee leaving. It was very cold,
(mercury at 6 deg.) The bees came to the edges of the combs, but the
cold drove them back. In a short time I had taken off six; four when
done were just about full; the other two were so when I began, but they
were marked and sawed like the rest; when the combs were attached, they
were severed with a knife, and the piece of the hive thus loose, was
raised off, leaving several inches of the combs projecting out of the
hive. I now cut off the first comb, even with the bottom of the hive.
On the next comb there were a few bees; with a quill these were brushed
down into the hive; this piece was then removed, and the bees on the
other side of it were brushed down also. In this way all others were
removed, and left the hive just full. The other full hive, after it was
sawed on each side, a small wire was drawn through, parallel with the
sheets, and severed all the combs at once; each piece was taken out,
and the bees that were clustered on them brushed back; removing the
loose part of the hive, was the last thing to be done. This last method
was preferred to the other by my employer; yet it was all performed to
his satisfaction, no sting or other difficulty about it, except the
trouble of warming fingers rather frequently. Tobacco smoke would have
kept them quiet during the operation, nearly as well. If preferred, a
hive may stand right side up while sawing it.


MODERATE WEATHER BEST TO REMOVE BEES.

In transporting your bees, avoid if possible the two extremes of very
cold, or very warm weather. In the latter the combs are so nearly
melted, that the weight of the honey will bend them, bursting the
cells, spilling the honey, and besmearing the bees. In very cold
weather, the combs are brittle, and easily detached from the sides of
the hive. When necessitated to move them in very cold weather, they
should be put up an hour or so before starting. The agitation of the
bees after being disturbed will create considerable heat; a portion of
this will be imparted to the combs, and add to their strength.


PREPARATIONS FOR TRANSPORTING BEES.

To prepare for moving them, pieces of thin muslin about half a yard
square is as good as anything, secured by carpet tacks.


SECURING BEES IN THE HIVE.

The hive is inverted, and the cloth put over, neatly folded, and
fastened with a tack at the corners, and another in the middle.  The
tack is crowed in about two-thirds of its length, it then presents the
head convenient to pull out. If the bees are to go a great distance,
and require to be shut up several days, the muslin will be hardly
sufficient, as they would probably bite their way out. Something more
substantial would then be required. Take a board the size of the
bottom, cut out a place in the middle, and cover with wire cloth, (like
the one recommended for hiving,) and fasten it with tacks. This board
is to be nailed on the hive. After the nails are driven, with the
hammer start it off about the eighth of an inch; it will admit a little
air around the sides as well as the middle, quite necessary for heavy
stocks. But very small families might be safe without the wire cloth;
air enough would pass between the hive and board, except in warm
weather. New combs break easier than old.


BEST CONVEYANCE.

Probably the best conveyance is a wagon with elliptic springs. But a
wagon without springs is bad, especially for young stocks. Yet I have
known them moved safely in this way, but it required some care in
packing with hay, or straw, under and around them, and careful driving.
Good sleighing will answer very well, and by some thought to be the
best time.


HIVE TO BE INVERTED.

Whatever conveyance is employed, the hive should be inverted. The combs
will then all rest closely on the top, and are less liable to break
than when right end up, because then the whole weight of the combs must
depend upon the fastenings at the top and sides for support, and are
easily detached and fall. When moving bees, so reversed, they will
creep upward; in stocks part full, they will often nearly all leave the
combs, and get upon the covering. In a short time after being set up,
they will return, except in very cold weather, when a few will
sometimes freeze; consequently a warm room is required to put them in
for a short time.

After carrying them a few miles, the disposition to sting is generally
gone, yet there are a few exceptions. In moderate weather, when bees
are confined, they manifest a persevering determination to find their
way out, particularly after being moved, and somewhat disturbed. I have
known them to bite holes through muslin in three days. The same
difficulty is often attendant on attempting to confine them to the hive
by muslin when in the house in the winter, except when kept in a cold
situation. Should any combs become broken, or detached from their
fastenings, in hives not full, by moving or other accident, rendering
them liable to fall when set up, the hive may remain inverted on the
stand till warm weather, if necessary, and the bees have again fastened
them, which they do soon after commencing work in the spring. If they
are so badly broken that they bend over, rolls of paper may be put
between them to preserve the proper distance till secured. When they
commence making new combs, or before, it is time to turn the right end
up. While the hive is inverted, it is essential that a hole is in the
side, through which the bees may work. A board should fit close over
the bottom, and covered, to effectually prevent any water from getting
among the bees, &c.


CONCLUSION.

In conclusion I would say, that the apiarian who has followed me
attentively, and has added nothing of value to his stock of
information, possesses an enviable experience that all should strive to
obtain.

It has been said that "three out of five who commence an apiary must
fail;" but let us suppose it is through ignorance or inattention, and
not inherent with the bees. To the beginner then I would say,--if you
expect to succeed in obtaining one of the most delectable of sweets for
your own consumption, or the profit in dollars and cents, you will find
something more requisite than merely holding the dish to obtain the
porridge. "SEE YOUR BEES OFTEN," and know at all times their actual
condition. This one recipe is worth more than all others that can be
given; it is at the head of the class of duties; _all others begin
here_. Even the grand secret of successfully combating the worms,--KEEP
YOUR BEES STRONG, must take its rise at this point. With the above
motto acted upon, carried out fully, and with perseverance, you cannot
well fail to realize all reasonable expectations. Avoid over-anxiety
for a rapid increase in stocks; try and be satisfied with one good
swarm from a stock annually, your chances are better than with more; do
not anticipate the golden harvest too soon. You will probably be
necessitated to discard some of the _extravagant_ reports of profits
from the apiary. Yet you will find one stock trebling, perhaps
quadrupling its price or value in products, while the one beside it
does nothing. In some seasons particularly favorable your stocks
collectively will yield a return of one or two hundred per cent.--in
others, hardly make a return for trouble. The proper estimate can be
made only after a number of years, when, if they have been judiciously
managed, and your ideas have not been too extravagant, you will be
fully satisfied. I have known a single stock in one season to produce
more than twenty dollars in swarms and honey, and ninety stocks to
produce over nine hundred dollars, when a few of the number added not a
farthing to the amount. I do not wish to hold out inducements for any
one to commence bee-keeping, and end it in disgust and disappointment.
But I would encourage all suitable persons to try their skill in bee
management. I say suitable persons, because there are many, very many,
not qualified for the charge. The careless, inattentive man, who leaves
his bees unnoticed from October till May, will be likely to complain of
ill success.

Whoever cannot find time to give his bees the needed care, but can
spend an hour each day obtaining gossip at the neighborhood tavern, is
unfit for this business. But he who has a home, and finds his
affections beginning to be divided between that and his companions of
the bar-room, and wishes to withdraw his interest from unprofitable
associates, and yet has nothing of sufficient power to break the bond,
to what can he apply with a better prospect of success, than to engage
in keeping bees? They make ample returns for each little care.
Pecuniary advantages are not all that may be gained--a great many
points concerning their natural history are yet in the dark, and many
are disputed. Would it not be a source of satisfaction to be able to
contribute a few more facts to this interesting subject, adding to the
science, and holding a share in the general fund? Supposing all the
mysteries pertaining to their economy discovered and elucidated,
precluding all chance of further additions, would the study be dry and
monotonous? On the contrary, the verification witnessed by ourselves
would be so fascinating and instructive, that we cannot avoid pitying
the condition of that man who finds gratification only in the gross and
sensual. It has been remarked, that "he who cannot find in this and
other branches of natural history a salutary exercise for his mental
faculties, inducing a habit of observation and reflection, a pleasure
so easily obtained, unalloyed by any debasing mixture--tending to
expand and harmonize his mind, and elevate it to conceptions of the
majestic, sublime, serene, and beautiful arrangements instituted by the
God of nature, must possess an organization sadly deficient, or be
surrounded by circumstances indeed lamentable." I would recommend the
study of the honey-bee as one best calculated to awaken the interest of
the indifferent. What can arrest the attention like their
structure--their diligence in collecting stores for the future--their
secretion of wax and moulding it into structures with a mathematical
precision astonishing the profoundest philosophers--their maternal and
fraternal affection in regarding the mother's every want, and assiduous
care in nursing her offspring to maturity--their unaccountable display
of instinct in emergencies or accidents, filling the beholder with
wonder and amazement? The mind thus contemplating such astonishing
operations, cannot well avoid looking beyond these results to their
divine Author. Therefore let every mind that perceives one ray of light
from nature's mysterious transactions, and is capable of receiving the
least enjoyment therefrom, pursue the path still inviting onward in the
pursuit. Every new acquisition will bring an additional satisfaction,
and assist in the next attempt, which will be commenced with a renewed
and constantly increasing zest; and will arise from the contemplation a
wiser, better, and a nobler being, far superior to those who have never
soared beyond the gratifications of the mere animal, grovelling in the
dark. Is there, in the whole circle of nature's exhaustless storehouse,
any one science more inviting than this? What more exalting and
refining, and at the same time making a return in profits as a
pecuniary reward?

What would be the result in the aggregate of all the honey produced in
the flowers of the United States annually? Suppose we estimate the
productions of one acre to be one pound of honey, which is but a small
part of the real product in most places; yet, as a great many acres are
covered with water and forest,[22] this estimate is probably enough for
the average. This State (New York) contains 47,000 square miles; 640
acres in a square mile will multiply into a little more than
30,000,000, and each acre producing its pound of honey, we have the
grand result of 30,000,000 lbs. of honey. If we add the States of
Pennsylvania, Ohio and Michigan, we have an amount of over 126,000,000
lbs. What it might be by including all the States, those disposed may
ascertain. Enough for our purpose is made clear, and that is, a small
item only of an enormous amount is now secured.

      [22] It should not be forgotten that forest trees are valuable,
      especially when there is basswood, or even maple.



                     *      *      *      *      *



BOOKS PUBLISHED BY
J. M. SAXTON, AGRICULTURAL BOOK PUBLISHER,
152 FULTON STREET, NEW YORK,

Suitable for

School, Town, Agricultural and Private Libraries.

                           *      *      *

FOR SALE BY

A. S. BARNES & CO.,

Corner of Dutch and John-streets,

NEW YORK.

                           *      *      *

The American Farm Book;
    The American Farm Book; or, a Compend of American Agriculture,
    being a Practical Treatise on Soils, Manures, Draining, Irrigation,
    Grasses, Grain, Roots, Fruits, Cotton, Tobacco, Sugar-Cane, Rice,
    and every staple product of the United States; with the best
    methods of Planting, Cultivating, and Preparation for Market.
    Illustrated by more than 100 engravings. By R. L. Allen.

American Poultry Yard;
    The American Poultry Yard; comprising the Origin, History and
    Description of the different Breeds of Domestic Poultry, with
    complete directions for their Breeding, Crossing, Rearing,
    Fattening, and Preparation for Market; including specific
    directions for Caponizing Fowls, and for the Treatment of the
    Principal Diseases to which they are subject; drawn from authentic
    sources and personal observation. Illustrated with numerous
    engravings. By D. J. Browne.

The Diseases of Domestic Animals;
    Being a History and Description of the Horse, Mule, Cattle,
    Sheep, Swine, Poultry, and Farm Dogs, with Directions for their
    Management, Breeding, Crossing, Rearing, Feeding, and Preparation
    for a profitable Market; also, their Diseases and Remedies;
    together with full Directions for the Management of the Dairy, and
    the Comparative Economy and Advantages of Working Animals, the
    Horse, Mule, Oxen, &c. By R. L. Allen.

American Bee Keeper's Manual;
    Being a Practical Treatise on the History and Domestic Economy of
    the Honey Bee, embracing a full illustration of the whole subject,
    with the most approved methods of Managing this Insect, through
    every branch of its Culture, the result of many years' experience.
    Illustrated with many engravings. By T. B. Miner.

The Modern Stair Builder's Guide;
    Being a Plain, Practical System of Hand Railing, embracing all its
    necessary Details, and Geometrically Illustrated by Twenty-two
    Steel Engravings; together with the Use of the most important
    Principles of Practical Geometry. By Simon De Graff, Architect.

Prize Essay on Manures.
    An Essay on Manures, submitted to the Trustees of the Massachusetts
    Society for Promoting Agriculture, for their Premium. By Samuel L.
    Dana.

American Bird Fancier.
    Considered with reference to the Breeding, Rearing, Feeding,
    Management, &c, of Cage and House Birds. Illustrated with
    engravings. By D. J. Browne. Cloth, 50 cts.; mail edition, paper,
    25 cts.

American Architect.
    The American Architect; comprising Original Designs of cheap
    Country and Village Residences, with Details, Specifications,
    Plans, and Directions, and an estimate of the Cost of each Design
    By John W. Ritch, Architect. First and Second Series quarto, bound
    in 2 vols., sheep. $6. Mail edition, paper, $5.

Domestic Medicine.
    Gunn's Domestic Medicine; or, Poor Man's Friend in the Hours of
    Affliction, Pain, and Sickness. Raymond's new revised edition,
    improved and enlarged by John C. Gunn 8vo. Sheep. $3.

Saxton's American Farmer's Almanac for 1852.
    Per 100, $3.

Family Kitchen Gardener.
    Containing Plain and Accurate descriptions of all the Different
    Species and Varieties of Culinary Vegetables; with their Botanical,
    English, French, and German names, alphabetically arranged, and the
    best mode of cultivating them in the garden, or under glass; also,
    Descriptions and Character of the most Select Fruits, their
    Management Propagation, &c. By Robert Buist, author of the American
    Flower Garden Directory, &c., cloth or sheep, 75 cts.; mail edition,
    paper, 50 cts.

Practical Agriculture.
    Being a Treatise on the General Relations which Science bears to
    Agriculture  Delivered before the New York State Agricultural
    Society, by James F. W. Johnston, F.R.S.S.S. and K., Professor
    of Agricultural Chemistry in Durham University, and author of
    Lectures on Agricultural Chemistry, with Notes and Explanations by
    an American Farmer. Cloth, 75 cts.; mail edition, paper, 50 cts.

Elements of Agricultural Chemistry and Geology.
    By J. F. W. Johnston, M.A., F.R.S. 50 cts.

Youatt and Martin on Cattle:
    Being a Treatise on their Breeds, Management, and Diseases:
    comprising a full History of the Various Races; their Origin,
    Breeding, and Merits; their capacity for Beef and Milk. By W.
    Youatt and W. C. L. Martin. The whole forming a complete Guide for
    the Farmer, the Amateur, and the Veterinary Surgeon, with 100
    illustrations. Edited by Ambrose Stevens. $1 25.

Youatt on the Horse.
    Youatt on the Structure and Diseases of the Horse, with their
    Remedies. Also, Practical Rules for Buyers, Breeders, Breakers,
    Smiths, &c. Edited by W. C. Spooner, M.R.C.V.S. With an account of
    the Breeds in the United States, by Henry S. Randall. $1 25.

Youatt on Sheep:
    Their Breed, Management, and Diseases, with illustrative
    engravings; to which are added Remarks on the Breeds and Management
    of Sheep in the United States, and on the Culture of Fine Wool in
    Silesia. By Wm. Youatt. 75 cts.

Hoare on the Grape Vine.
    A Practical Treatise on the Cultivation Of the Grape Vine on open
    Walls, with a Descriptive Account of an improved method of Planting
    and Managing the Roots of Grape Vines. By Clement Hoare. With an
    Appendix on the Cultivation of the same in the United States. 50
    cts.

The American Agriculturist:
    Being a Collection of Original Articles on the Various Subjects
    connected with the Farm, in ten vols. 8vo., containing nearly four
    thousand pages. $10.

Johnston's Agricultural Chemistry.
    Lectures on the Application of Chemistry and Geology to
    Agriculture. New edition with an Appendix. $1 25.

Stephens' Book of the Farm.
    A Complete Guide to the Farmer, Steward, Plowman, Cattleman,
    Shepherd, Field-Worker, and Dairy Maid. By Henry Stephens. With
    Four Hundred and Fifty Illustrations; to which are added
    Explanatory Notes, Remarks, &c., by J. S. Skinner. Really one of
    the best books for a Farmer to possess. Cloth, $4; leather, $4 50.

The Complete Farmer and American Gardener,
    Rural Economist, and New American Gardener, containing a
    Compendious Epitome of the most Important Branches of Agricultural
    and Rural Economy; with Practical Directions on the Cultivation of
    Fruits and Vegetables; including Landscape and Ornamental
    Gardening. By Thomas G. Fessenden  2 vols. in one. $1 25.

Allen's Treatise on the Cultivation of the Grape
    $1 25.

Brandy and Salt,
    A Remedy for various Internal as well as External Diseases,
    Inflammation and Local Injuries. By Rev. Samuel Fenton. 12-1/2 cts.

Southern Agriculture.
    Comprising Essays on the Cultivation of Corn, Hemp, Tobacco, Wheat,
    &c. $1.

The Cottage and Farm Bee Keeper:
    A Practical Work, by a Country Curate. 50 cts.

A Book for Every Boy in the Country.
    Elements of Agriculture. Translated from the French, and adapted to
    General Use, by F. G. Skinner. 25 cts.

Rural Architecture;
    Comprising Farm Houses, Cottages, Carriage Houses, Sheep and Dove
    Cotes, Pigeries, Barns, &c. &c. By Lewis F. Allen. $1 25.

The American Muck Book.
    The American Muck Book; treating of the Nature, Properties,
    Sources, History, and Operations of all the principal Fertilizers
    and Manures in Common Use, with Specific Directions for their
    Preservation, and Application to the Soil and to Crops; drawn from
    Authentic Sources. Actual Experience, and Personal Observation, as
    Combined with the leading Principles of Practical and Scientific
    Agriculture. By J. D. Browne. $1 25

Youatt on the Pig.
    A Treatise on the Breeds, Management, and Medical Treatment of
    Swine; with direction for Salting Pork, Curing Bacon and Hams. By
    Wm. Youatt, R.S. Illustrated with engravings drawn from life. 60
    cts.

Youatt on the Dog.
    By Wm. Youatt, Splendidly illustrated. Edited, with Additions, by
    E. J. Lewis, M.D. $1 50.

The Poultry Book.
    By John C. Bennett, M.D. 84 cts.

The American Poulterer's Companion,
    With illustrations. By C. N. Bement. $1.

American Poultry Book.
    By Micajah Cock. 38 cts.

The Rose Culturist.
    A Practical Treatise on its Cultivation and Management. 50 cts.

A Practical Treatise on Honey Bees,
    Their Management, &c. By Edward Townley. 50 cts.

The American Fruit Book.
    By S. W. Cole. 50 cts.

The American Veterinarian.
    By S. W. Cole. 50 cts.

The Gardener's Text Book.
    By Peter Adam Schenck. 50 cts.

The American Gardener.
    By William Cobbett. 50 cts.

The Farmer's Land Measurer.
    By James Pedder. 50 cts.

New England Fruit Book.
    By John M. Ives. 56 cts.

Practical Treatise on Fruits,
    Adapted to New England Culture. By George Jaques 50 cts.

Farmer and Emigrant's Hand Book.
    A Guide to Clearing the Forest and Prairie Land, &c, &c. By Josiah
    T. Marshall. $1.

Farmer's Barn Book.
    By Youatt, Clater, Skinner and Mills. $1 25.

Hind's Farriery and Stud Book.
    Edited by J. S. Skinner. $1.

Mason's Farrier and Stud Book.
    Edited by J. S. Skinner. $1 25.

Stewart's Stable Economy.
    A Treatise on the Management of Horses.  Edited by A. B. Allen. $1.

Sugar Planter's Manual.
    By W. S. Evans, M.D. $1 25.

Treatise on Hothouses and Ventilation.
    By R. B. Suckara. $1 25.

Ornamental and Domestic Poultry.
    By Rev. Edmund Saul Dixon, A.M. With Large Additions by J. J. Kerr,
    M.D. With illustrations  $1.

Canfield on Sheep,
    Their Breeds Management, Structure, and Diseases.  With
    Illustrative Engravings and an Appendix  Edited by H. J. Canfield.
    $1.

Book of Flowers,
    In which are described the various Hardy Herbaceous Perennials,
    Annuals, Shrubby Plants and Evergreen Trees desirable for
    Ornamental Purposes. By Jos. Breck. 75 cts.

Experimental Researches on the Food of Animals,
    The Fattening of Cattle, and Remarks on the Food of Man. By Robert
    Dundas Thompson, M.D. 75 cts.

The American Flower Garden Companion,
    Revised and enlarged. By Edward Sayres. 75 cts.

The Farmer's Treasure.
    A Treatise on the Nature and Value of Manures, and Productive
    Farming. By F. Faulkner and Joseph A. Smith. 75 cts.

The Practical Farrier.
    By Richard Mason. 75 cts.

The American Farrier.
    By Barnum. 75 cts.

Principles of Practical Gardening.
    By Geo. W. Johnston, Esq. $1 25.

Buchannan on the Grape.
    50 cents.

Spooner on the Grape.
    The Cultivation of American Grape Vines, and making of Wine. By
    Alden Spooner. 38 cts.

The Young Gardener's Assistant.
    By Thomas Bridgeman. $1 50.

The Florist's Guide.
    By Thos. Bridgeman. 50 cts.

The Kitchen Gardener's Instructor.
    By Bridgeman. 50 cts.

The Fruit Cultivator's Manual.
    By Bridgeman. 50 cts.

The Horse,
    Its Habits, Diseases and Management, in the Stable and on the Road,
    &c. 25 cts.

The Fruit, Flower, and Kitchen Garden.
    By Patrick Neill, LL.D., F.R.S., adapted to the United States. $1 25.

Ladies' Companion to the Flower Garden.
    By Mrs. Loudon. Edited by A. J. Downing. $1 25.

The Fruits and Fruit Trees of America.
                               By A. J. Downing. $ 1 50.
    Do.    do.    do.    do.            colored.  15 00.

Dictionary of Modern Gardening.
    By Geo. W. Johnston. Edited by David Landreth. $1 50.

The Rose Fancier's Manual.
    By Mrs. Gore. $1 50.

Parsons on the Rose.
    The Rose: its History, Poetry, Culture, and Classification. By S.
    B. Parsons. $1 50.

Hovey's Fruits of America.
    Containing richly colored Figures and full Descriptions of all the
    Choicest Varieties cultivated in the United States, in 12 numbers.
    $12.

History, Treatment and Diseases of the Horse,
    With a Treatise on Draught, and Copious Index. $2.

Rural Economy,
    n its Relations with Chemistry, Physics, and Meteorology. By J. B.
    Boussingault. Translated, &c, by George Law. $1.

Liebig's Agricultural Chemistry.
    Edited by Lyon Playfair, Ph.D, F.G.S., and William Gregory, M.D.,
    P.R.S.E. $1.

The Modern System of Farriery,
    As Practised at the Present Time at the Royal Veterinary College,
    and from Twenty Years' Practice of the Author, George Skevington,
    M.R.V.C. $5.

Ewbank's Hydraulics:
    A Descriptive and Historical Account of Hydraulic and other
    Machines for Raising Water. $2 50.

The Fruit Garden.
    By P. Barry. $1 25.

The American Fruit Culturist;
    Containing Directions for the Culture of Fruit Trees in the
    Nursery, Orchard, and Garden. By John J. Thomas. $1 25

The Rose Manual.
    By Robert Buist. 75 cts.

The Plants of Boston and Vicinity.
    By Jacob Bigelow, M.D., $1 50

The Indian Meal Book;
    Comprising the best Receipts for the Preparation of that Article.
    By Miss Leslie. 25 cts.

The Horse's Foot,
    And How to Keep it Sound. By William Miles. 25 cts.

Catechism of Agricultural Chemistry and Geology.
    By J. F. W. Johnston. 25 cts.

Chemistry Applied to Agriculture.
    By Le Count Chaptal. 50 cts.

British Husbandry.
    Three Vols. and Supplement. $5.

Loudon's Arboretum.
    Eight Vols. $25.

Loudon on Gardening.
    Loudon's Encyclopedia of Gardening. $10

Loudon on Agriculture.
    Loudon's Encyclopedia of Agriculture. $10.

Loudon on Trees, &c.
    Loudon's Encyclopedia of Trees, Shrub, &c.

Loudon on Plants, &c.
    Loudon's Encyclopedia of Plants, &c.

The Farmer's Library.
    Two vols. 8vo. English. $5.

The Farmer's Dictionary.
    By D. P. Gardner. $1 50.

Practical Treatise on the Grape Vine.
    By J. Fisk Allen. Boards, $1.

Practical Treatise on the Veterinary Art.
    By J. Bridden. 75 cts.

Sheep Husbandry.
    By Henry S. Randall. $1 50.

Agricultural Chemistry.
    By Justus Liebig. Cloth, $1; cheap edition, 25 cts.

Animal Chemistry.
    By J. Liebig. Cloth, 50 cts.; cheap ed. paper, 25 cts.

Liebig's Complete Works,
    In one vol. 8vo. $1.

Cottage and Farm Houses.
    By A. J. Downing. $2.

Country Houses.
    By A. J. Downing. $4.

Sportsman's Library.
    By T. B. Johnson. English edition. $5.

Landscape Gardening.
    By A. J. Downing. $3 50.

Cottage Residences.
    By A. J. Downing. $2.

Chaptal's Agricultural Chemistry,
    With Notes. $1.

American Husbandry.
    By Gaylord and Tucker. $1.

Gardeners Dictionary.
    By Geo. Don, F.L.S. 4 vols., quarto, $10.

Journal of Agriculture.
    Edited by John S. Skinner. 3 vols. $6.

Downing's Horticulturist.
    Half morocco. Per Vol. yearly Vols. $3 75.
         Do.      do. half yearly   "    2 00.

The Complete Produce Reckoner,
    Showing the Value by Pound or Bushel. By R. Robbins. 75 cts.

The American Shepherd.
    By L. A. Morrill. $1.

The Principles of Agriculture.
    By Albert D. Thaer. $2 50.

Lectures to Farmers on Agricultural Chemistry.
    By Alexander Petzholdts. 75 cts.

The Complete Farrier.
    By John C. Knowlson. 25 cts.

The Complete Cow Doctor.
    By J. C. Knowlson. 15 cts.

Milch Cows.
    By Guenon. 38 cts.

A Home for All;
    Or a New, Cheap, and Superior mode of Building, By O. S. Fowler. 50
    cts.

The Poultry Breeder.
    By George F. Burnham. 25 cts.

The American Fowl Breeder. 25 cts.

The Farmer's Companion.
    By Judge Buel. 75 cts.

The Farmer's Instructor.
    By Judge Buel. $1.

European Agriculture,
    From Personal Observation.  By Henry Coleman. 2 vols.  $5 00.
           Do.           do.            do.       1 vol.    4 00.

The Gardener and Florist. 25 cts.

The Honey Bee.
    By Bevan. 31 cts.

Elements of Practical Agriculture.
    By John P. Norton. 50 cts.

Rogers' Scientific Agriculture. 75 cts.

Mills' Sportsman's Library. $1.

Stable Talk and Table Talk. $1.

Hawker and Porter on Shooting. $2 75.

Field Sports.
    By Frank Forrester. 2 vols. $4

Fish and Fishing.
    By Frank Forrester. $2 50.

The American Angler's Guide.
    By J. J. Brown. $1 50.

Johnson's Farmer's Encyclopedia.
    Edited by G. Emerson, M.D. $4.

Scientific and Practical Agriculture.
    By Alonzo Gray. 75 cts.

Theory and Practice of Agriculture.
    By A. Partridge. 12 cts.

Armstrong on Agriculture. 50 cts.

Hovey's Magazine of Horticulture.
    Published monthly. Per annum $2

Downing' Horticulturist.
    Published monthly. Per annum $3

Gilpin's Landscape Gardening.
    English edition. $2 50.

The Gardener's Calendar.
    By M. Mahon. $3 50.

Agriculture for Schools.
    By Rev. J. L. Blake, D.D. $1.

Text Book of Agriculture.
    By Davis. 50 cts.

The American Agriculturist and Farmer's Cabinet.
    Published monthly. Per annum $1.

Weeks on the Honey Bee.

Cottages and Cottage Life.
    By Elliott. $2 25.

Chemical Analysis.
    By Fresinus and Bullock. $1.

Applied Chemistry.
    By A. Parnell. $1.

The Vegetable Kingdom,
    Or Handbook of Plants. By L. D. Chapin. $1 25.

The Muck Manual.
    A new edition. By Samuel L. Dana. $1.

Youatt on the Horse.
    Edited by J. S. Skinner. $1 50.

Clater's Farrier. 50 cts.

The Dog and Sportsman.
    By J. S. Skinner. 50 cts.

The Bird Keeper's Manual. 50 cts.

The American Herd Book.
    By Lewis F. Allen. $

The American Orchardist.
    By J. Kenrick. 75 cts.

Blacklock's Treatise on Sheep. 50 cts.

Guide to the Orchard and Fruit Garden.
    By G. Lindley. $1 50.

The Trees of America.
    By D. J. Brown. $4 50.

American Flower Garden Directory.
    By Robert Buist. $1 25.

The American Cattle Doctor.
    By G. H. Dodd. 1 vol. $1.

Maury's Navigation.
    A New Theoretical and Practical Treatise on Navigation, in which
    the Auxiliary Branches of Mathematics and Astronomy are treated of,
    and the Theory and most simple Methods of Finding Time, Latitude,
    and Longitude, by Chronometers, Lunar Observations, Single and
    Double Altitudes, are taught. Third edition, enlarged and improved.
    By M. F. Maury, Lieut. U.S. Navy. 8vo. sheep, library style. $3 50.

Works of Thomas Dick.
    The Works of Thomas Dick, LL.D. 10 vols., in 5, 12mo. Embellished
    by a Portrait of the Author, from an engraving on steel, and
    illustrated by woodcuts representing more than 500 different
    objects. Neat half muslin binding, $3 25; morocco backs, $4.

Compendium of English Literature.
    A Compendium of English Literature, chronologically arranged, from
    Sir John Mandeville (14th century) to William Cowper (close of the
    18th century); consisting of Biographical Sketches of the Authors,
    choice selections from their works; with Notes explanatory and
    illustrative, and directing to the best editions, and to various
    criticisms. Designed as a text-book to the higher classes in
    Schools and Academies, as well as for private reading. By Charles
    D. Cleveland. $1 50.

An Elementary Treatise on Statics.
    By Gaspard Monge. With a Biographical Notice of the Author.
    Translated from the French by Woods Baker, A.M., of the U.S. Coast
    Survey. $1 25.

Harrison on the English Language.
    The Rise, Progress and present Structure of the English Language,
    by the Rev. Matthew Harrison, A.M., Rector of Church Oakley, Hants.
    and Late Fellow of Queen's College, Oxford. $1.

The Progressive Farmer.
    By J. A. Nash. 50 cents.

The American Florist's Guide.
    75 cents.

Agricultural Dynamics.
    By J. J. Thomas.



                     *      *      *      *      *



THE AMERICAN BEE-KEEPER'S MANUAL.

BY T. B. MINER.

350 pp. 12mo. 35 ENGRAVINGS. PRICE $1.

PUBLISHED BY C. M. SAXTON, 152 FULTON ST., N.Y.

                           *      *      *

OPINIONS OF THE PRESS.

"The most complete work on the Bee and Bee-keeping we have yet
seen."--_N.Y. Tribune._

"Mr. Miner has handled this subject in a masterly manner."--_N.Y. True
Sun._

"He has written a work of the most fascinating interest."--_N.Y. Sunday
Dispatch._

"It will interest the general reader. It is indeed a charming
volume."--_Commercial Advertiser._

"This is a truly valuable work, and very interesting."--_Morning Star._

"It is decidedly the best work we have ever seen."--_Boston Daily
Mail._

"Mr. Miner has performed his task with signal ability."--_Scientific
American._

"It does high credit to the observation and intelligence of the
author."--_Christian Intelligencer._

"This is the most comprehensive and valuable work on the Honey-bee that
has ever come under our notice."--_Journal of Commerce._

"To appreciate the value of the honey-bee one must get this book and
read it attentively."--_Noah's Messenger._

"We like it for its independent tone, and the amount of practical
information that it contains."--_Literary World._

"We have been greatly edified and entertained by this book, from which
the reader will collect a great deal of excellent information."--_The
Independent._

"This is probably the most complete manual of the kind ever published.
It will richly repay the general reader, too, by the variety of
interesting facts it contains."--_Boston Traveller._

"It is a most excellent and useful treatise, and happily supplies a
vacuum that had long existed."--_Boston Times._

"This volume has all the charm of a romance and admirably displays the
habits of this insect."--_Organ._

"This volume is what it pretends to be, (more than can be said of many
works) and all who want a full and thorough history of the nature and
management of the bee should have it in their possession."--_Scientific
American._

"It is neatly printed, well illustrated and clearly written and
contains a great deal of practical information."--_Mirror._

"This work probably contains better instructions in regard to bees than
any which have ever appeared."--_Sun._

"The practical directions are the result of evident experience, and
being plainly and concisely stated, are excellent, It is so much better
than can be obtained elsewhere that we commend it to favor."--_N.Y.
Evangelist._

"It is an excellent book and the best published on the
subject."--_Boston Olive Branch._



                     *      *      *      *      *



ILLUSTRATED TREATISE ON DOMESTIC ANIMALS,

Being a history and description of the Horse, Mule, Cattle, Sheep
Swine, Poultry, and Farm Dogs; with Directions for their Management,
Breeding, Crossing, Rearing, Feeding, and preparation for a profitable
market. Also, their Diseases and Remedies; together with Full
Directions for the Management of the Dairy, and the Comparative Economy
and Advantages of Working Animals, the Horse, Mule, Oxen, &c., by R. L.
ALLEN, _Author of "Compend of American Agriculture"_ &c.

The above work contains more than 40 Engravings and Portraits of
Improved Animals illustrative of the different breeds and various
subjects treated in it.

The most minute as well as general principles for Breeding, Crossing,
Rearing, Feeding, and Management of all Domestic Animals, are herein
given, to produce the utmost marketable value for the food and
attention bestowed on them; as well as to prevent disease, and save the
immense losses which annually occur from this source. It can be sent by
Mail, in Cloth Binding, for 75 Cents--Paper, 50 Cents. Published by C.
M. SAXTON, 152 Fulton St. New York. For sale by all the Booksellers
throughout the country.

Agents wanted for _every county_ in every state. Address, _post paid_,
the Publisher.


OPINIONS OF THE PRESS.

The Compactness yet completeness will make it a favorite with
agriculturists.--_Chronicle, Philadelphia._

Its greatest worth is, as a complete farrier, showing the diseases of
animals, their treatment, and cure.--_Far. & Mec._

The portion which relates to the dairy alone, is worth the cost of the
book.--_Worcester Transcript._

It is every way adapted to be serviceable in every household which has
domestic animals.--_D. Adv.; Newark._

We believe it a complete guide for the farmer and dairyman in the
purchase, care, and use of animals.--_Jeffersonian._

Here is a work which should be in the hands of every farmer.-_-Highland
Courier._

We can confidently recommend this work as a very instructive one to
those engaged in farming, raising stock, or husbandry.--_Northampton
Courier._

The author is a practical farmer and stockbreeder, and is able to vouch
for the correctness of the remedies for diseases of Domestic Animals,
as well as the best mode of managing them.--_Huron, O. Reflector._

It costs but _seventy-five cents_, and cannot fail to be worth _ten
times_ that amount to any farmer.--_Summit S.C. Beacon._

It is the best of that character we have yet seen; no farmer should be
without it--_Democrat, Carlisle, Pa._

This is just such a book as every owner of stock should be possessed
of.--_Easton Md. Star._

Here is a book which all--those who follow the plow, and those who
direct it--can read to profit. It is a library of knowledge, presenting
the latest improvements and discoveries, on all the topics treated of;
and illustrated by a great variety of cuts. The "Allens," one of whom
is the author of the work before us, are quite famous in their especial
_role_, so that what proceeds from them may be confidently credited at
all events. The present book is a most interesting and instructive one,
and must meet with a great sale.--_Sciota Gazette._

This work, to the farmer and stock raiser, will be useful, instructive,
and profitable, enabling them to improve the breed of their stock,
preserve them from sickness, and cure them when infected with
disease.--_Herald, Morrisville, Pa._

The time has gone by when farmers can expect to _succeed_ without
giving some attention to Book Farming, and we trust they begin to see
it for themselves. We should like to hear that this work was in the
hands of every farmer in the county.--_Mercury, Potsdam, N.Y._

The title page of this work gives a good idea of its scope and intent.
It is a comprehensive summary of farm operations, and will prove very
acceptable to the great mass of our farming population. We are informed
that 3,000 copies of this work have been sold since the first of
January. It is well printed and profusely illustrated--_N.Y. Tribune._

It is furnished with numerous illustrating cuts, and will form a
complete "vade mecum" for The agriculturist, convenient for reference,
and to be relied on when consulted---_Baltimore American._

This is a practical book by a practical man, and will serve extensive
practical ends. It is a companion which every farmer will feel that he
cannot well be without.--_N.Y. Observer._

We cheerfully recommend this work to farmers.--_Signal, Juliett, Ill._

We anticipate an extensive sale for this work--_Ohio Cultivator._

This work ought to be in the hands of every planter.--_N.O. Delta._

The author is a gentleman of fine attainments, and who ranks as one of
the most accomplished writers on agricultural subjects in the
country.--_Ala. Planter._

Many a valuable animal is lost, every year, for want of the knowledge
here conveyed.--_Eagle Brattleboro, Vt._

The author (Mr. Allen), is a practical man, and everything from his
pen, on subjects connected with agriculture and cattle breeding, is
valuable to those who prefer matter of fact to mere theory--_Maine
Farmer._



                     *      *      *      *      *



_Published by C. M. Saxton._

                           *      *      *

AMERICAN MUCK BOOK,

Treating of the Nature, Properties, Sources, Operations, &c.

OF ALL THE PRINCIPAL FERTILIZERS AND MANURES IN COMMON USE, WITH
SPECIFIC DIRECTIONS FOR THEIR

PREPARATION, PRESERVATION, AND APPLICATION

to Soil and Crops; with the leading principles of Practical and
Scientific Agriculture, &c., &c. By D. J. Browne. 420 pp. 12mo. price
$1.

                           *      *      *

Opinions of the Press.

"The Muck Book contains a great deal of valuable matter. This has been
drawn from a large number of the best authorities on the subjects
indicated in the title; the numerous analyses of plants and manures are
particularly valuable, and are not to be found in any other single
treatise. Those who wish to advance towards perfection in the saving,
manufacturing, and judging of the comparative value of manures, and in
applying them with the least possible waste to crops, will find in this
book a vast magazine of suggestions and advice, worth many times its
cost and the labor of perusal."--_Albany Cultivator._

"The title 'Muck Book' inadequately describes the character of the
work; for it treats of all kinds of fertilizers, animal, vegetable and
mineral, and in a style to instruct without perplexing. The manner in
which the various manures operate, and the means whereby any required
deficiency in the soil can be supplied, are plainly given; and none
need waste a horn-pith or an old shoe, as many do, for want of
knowledge how to turn it to advantage.

"We recommend the work to intelligent and inquiring farmers, who desire
to make everything tell in the manure heap, and who would keep their
soil in good heart."--_Journal of Agriculture._

"From an attentive examination of the pages of this book, I have come
to the conclusion that it is one of the best works extant, on the
principles of scientific agriculture, and the best compendium of our
most recent knowledge of the nature of manures and their adaptation to
particular soils and crops."--_N.E. Farmer._

"Mr. Browne was, we believe, bred and educated a practical farmer
himself, and having a general knowledge of geology, chemistry, &c., and
extensive personal knowledge of farming, gardening, &c, in almost every
soil and climate, having been for five years a traveller and resident
in America, Europe, Western Africa, and the West Indies, his
observation and experience combined, would render him eminently
qualified for the task. This he has accomplished with credit to
himself, and no doubt the result will prove it highly advantageous to
the farming community. It is just such a work as is needed by every
agriculturist, and the very neat and excellent style in which the
enterprising publisher has issued it, will we are very sure commend it
to every friend of the farming interest in the country."--_N.Y. Farmer
and Mechanic._

"This is a well-written work of over four hundred pages, printed and
bound in the usual handsome and permanent style of Mr. Saxton. The
importance to every farmer and horticulturist of the great subject of
which it treats cannot fail to make this work invaluable to the library
of every man who tills the soil. One feature of this work which pleases
us, and which will make it universally acceptable is, that the subjects
are treated in such a manner as to be easily understood by the 'working
farmer,' who knows little or nothing of chemical science and learned
technicalities. With such a work as this in his hands, the farmer is
enabled to reclaim his lands, impoverished by his own or his ancestors'
mismanagement, and realize abundant crops where nothing would grow to
reward his toil in the ordinary mode of culture."--_Phrenological
Journal._

_The following is from. Dr. C. T. Jackson, of Boston, the best
Agricultural Chemist in the United States_:


    BOSTON, NOVEMBER 6th, 1851.

    DEAR SIR: I have the pleasure of acknowledging the receipt of a
    copy of the "American Muck Book," recently published by you, and
    edited by Mr. D. Jay Browne.

    From an attentive examination of the pages of this book, I have come
    to the conclusion that it is one of the best works extant, on the
    principles of scientific agriculture, and the best compendium of our
    most recent knowledge of the nature of manures and their adaptation
    to particular soils and crops. It cannot be expected that a single
    volume could possibly contain the whole sum of chemical knowledge
    applicable to the science of agriculture; but, on looking over the
    closely-printed and compact tables of analyses, and the abundant
    formulas, which this publication contains, I could not fail to be
    surprised at the industry manifested in preparing it. I was also
    gratified to find it so well adapted to the American system of
    husbandry, and so practical, in its character. Its copious and
    accurate index adds not a little to its value.

    I shall certainly recommend it to my agricultural friends as a very
    useful book, and one necessary to every scientific farmer.

    I am, respectfully, your ob't servant,

    CHARLES T. JACKSON, State Assayist, &c., &c.

    To C. M. SAXTON, Esq., New York.



                     *      *      *      *      *



THE

AMERICAN FARM BOOK,

OR

COMPEND OF AMERICAN AGRICULTURE

CONTAINING A CONCISE AND PLAINLY-WRITTEN EXPOSITION OF DUTIES PERTAINING
TO THE CULTIVATION OF THE EARTH, THE MANAGEMENT OF THE FARM, &C., &C.,
ON PRACTICAL SCIENTIFIC PRINCIPLES.

BY R. L. ALLEN.

_The cheapest and most valuable book for a farmer ever printed; being
a complete guide, both practical and scientific, for the_

MANAGEMENT OF THE FARM.

                           *      *      *

Besides the varied practical knowledge which this book imparts, and
which is indispensable to the proper management of every department of
agriculture, it gives the elements of other information highly
necessary to a successful farmer, as History, Geology, Chemistry,
Botany, Physiology, and Mechanics. These branches of knowledge are
given as applicable to agricultural pursuits, and when properly
understood will essentially aid and assist the farmer. In fact, a
knowledge of these sciences is a sure key to wealth for any
agriculturist. It gives the modes of preparation, and the effects of
all kinds of manures; the origin, texture, divisions, and description
of every variety of soil; the economy of sowing, reaping, and mowing,
irrigation, and draining; cultivation of the grasses, clovers, grains,
and roots; _Southern_ and miscellaneous products, as cotton, hemp,
flax, the sugar cane, rice, tobacco, hops, madder, woad, &c.; the
rearing of fruit--apples, pears, peaches, plums, grapes, &c.; farm
buildings, hedges, &c.; with the best methods of planting, cultivating,
and preparation for market. Illustrated by 100 engravings.

The reader can form some idea of the above work, from the fact that it
treats of 800 _different subjects_ important to a farmer. It contains
354 pages, and is beautifully bound in cloth, suitable for a library.
_Price only One Dollar._


NOTICES OF THE PRESS.

The author has been one of the most able contributors to the
agricultural press for the last ten years; aside from this, he is a
practical farmer and stock-breeder, and consequently knows from his own
experience what he is writing about.--_Commercial Advertiser._

This work is by a gentleman of known experience; the work is
exceedingly cheap, and the farmer will find it a valuable book of
reference.--_N.Y. Express._

It is in fact a brief encyclopedia on the subjects treated, and the
farmer will find appropriate Information on almost any subject coming
within his reach.--_N.Y. Observer._

Here is a book for the million, precisely what its title indicates.
Compassed within its pages, the reader will find the subject of soils,
manures, crops, and animals, treated in a style easily comprehended.--_N.Y.
Spirit of the Times._

This work is what might be expected from one so well qualified for the
undertaking.--_Boston Cultivator._

Why shall not every good farmer economise his muscles by storing his
mind? We hope this book will find its way into many family and school
libraries.--_N.Y. Tribune._

We think that Mr. Allen's volume, the basis of which is good practical
farming, as practised by the best cultivators in the United States,
with an intelligent reference to those principles of science which lie
at the root of all successful practice, is likely to be of as muck or
more real service to us, than any work on agriculture yet issued from
the press, and we gladly commend it to the perusal of every one of our
readers engaged in the cultivation of the land.--_Horticulturist._



                     *      *      *      *      *



THE

FAMILY KITCHEN GARDENER;

CONTAINING

PLAIN AND ACCURATE DESCRIPTIONS

OF ALL THE

DIFFERENT SPECIES AND VARIETIES

OF

CULINARY VEGETABLES:

BY ROBERT BUIST,

AUTHOR OF THE AMERICAN FLOWER-GARDEN DIRECTORY, ROSE MANUAL, ETC.

                           *      *      *

NEW  YORK:

C. M. SAXTON, 152 FULTON STREET,

ALSO, STRINGER & TOWNSEND, H. LONG & BROTHER, W. F. BURGESS,
DEWITT & DAVENPORT, WILSON & CO., DEXTER &
BROTHER, BOSTON: REDDING & CO. PHILADELPHIA:
W. B. ZIMMER, LINDSAY & BLAKISTON.



                     *      *      *      *      *



THE

AMERICAN BIRD FANCIER;

CONSIDERED WITH REFERENCE TO THE
BREEDING, REARING, FEEDING, MANAGEMENT, AND PECULIARITIES
OF

CAGE AND HOUSE BIRDS.

Illustrated with Engravings

BY D. J. BROWNE,

AUTHOR OF THE SYLVA AMERICANA, THE AMERICAN POULTRY YARD, ETC.

NEW YORK:

C. M. SAXTON, 152 FULTON STREET.

ALSO, STRINGER & TOWNSEND, H. LONG & BROTHER, W. F. BURGESS,
DEWITT & DAVENPORT, WILSON & CO., DEXTER & BROTHER
PHILADELPHIA: W. B. ZIEBER, LINDSAY & BLAKISTON.
BOSTON: REDDING & CO.



                     *      *      *      *      *



LECTURES

ON THE

GENERAL RELATIONS WHICH SCIENCE

BEARS TO

PRACTICAL AGRICULTURE,

DELIVERED BEFORE THE

NEW-YORK STATE AGRICULTURAL SOCIETY.

BY

JAMES F. W. JOHNSTON, F.R.S.S. L. & E.

WITH NOTES AND ADDITIONS

                           *      *      *

NEW YORK:

C. M. SAXTON, AGRICULTURAL BOOKSELLER.



                     *      *      *      *      *



AN

ESSAY ON MANURES,

SUBMITTED TO THE TRUSTEES OF

THE MASSACHUSETTS SOCIETY

FOR

PROMOTING  AGRICULTURE,

FOR THEIR PREMIUM.

BY SAMUEL L. DANA.

_From the New York Observer:_

ESSAY ON MANURES. By SAMUEL L. DANA

This Essay contains much useful information for the practical farmer, in
a small compass, in reference to the nature and management of manures
immediately under his control; the knowledge and practice of which will
amply compensate for the expense of ascertaining its value.

                           *      *      *

NEW YORK:

C. M. SAXTON, 152 FULTON STREET.

ALSO, STRINGER & TOWNSEND, H. LONG & BROTHER, W. F. BURGESS,
DEWITT & DAVENPORT, WILSON & CO., DEXTER &
BROTHER. BOSTON: REDDING & CO. PHILADELPHIA:
W. B. ZIEBER, LINDSAY & BLAKISTON.



                     *      *      *      *      *



A PRACTICAL TREATISE

ON THE

CULTIVATION OF THE GRAPE VINE

ON OPEN WALLS.

WITH A DESCRIPTIVE ACCOUNT OF AN

IMPROVED METHOD OF PLANTING AND MANAGING

THE

ROOTS OF GRAPE VINES.

BY CLEMENT HOARE

TO WHICH IS ADDED,

AN APPENDIX,

CONTAINING

REMARKS ON THE CULTURE OF THE GRAPE VINE IN

THE UNITED STATES.

                           *      *      *

NEW YORK:

C. M. SAXTON, 152 FULTON STREET.





*** End of this LibraryBlog Digital Book "Mysteries of Bee-keeping Explained" ***

Copyright 2023 LibraryBlog. All rights reserved.



Home