Historic, archived document

Do not assume content reflects current
scientific knowledge, policies, or practices

U.S. DEPARTMENT OF AGRICULTURE.

division of estomolooy.

Bulletin No. 24.

THE

BOLL WORM OF COTTON

A

REPORT OF PROGRESS

IN A

SUPPLEMENTARY INVESTIGATION OF THIS INSECT.

MADE UNDER THE DIRECTION OF THE ENTOMOLOGIST
BY

F. W. MAIXY.

(PUBLISHED BY AUTHORITY OF THE SECRETARY OF AGRICULTURE.)

WASHINGTON:

GOVERNMENT PRINTING OFFICE,
I 89 I .

TABLE OF CONTENTS.

Page.

Letter of submittal 3

Introduction 5

Letter of transmittal 7

The boll worm of cotton 9

Destructiveuess 9

Food plants other than cotton 12

Characters and transformation 14

The egg 14

The larva 14

The pupa 20

The imago 22

Number of broods and hibernation 25

Natural enemies 27

Vertebrates: Butcher Bird ; Crows; Quails.

Invertebrates: Podisus spinosus : Lepioierna ; Erax lateralis; Ants
Trichogramma pretiosa; Hexaplasta zigzag; Tachiua; Euplectrus
comstockii.

Insect ravages easily mistaken for those of the boll worm 28

Euphoria melancholica 29

Platynota sentana , 29

Caccccia rosacean a 30

Prodenia linen tella 30

Noctuid (undetermined) 30

Plant lice (Aphis gossypii and Aphis sp. ) 30

ThripidsB 30

Remedies 31

Topping of cotton and rotation of crops 31

Fall plowing 31

Corn as protection to cotton 32

Lights for attracting the moths 33

Poisoned sweets 38

Pyrethrum 39

Experiments with dry powder 39

Decoctions of Pyrethrum 42

Other vegetable insecticides 44

Meteorological considerations 45

Insect, diseases 48

2

LETTER OF SUBMITTAL

U. S. Department of Agriculture,

Division of Entomology,
Washington, D. (7., February 27, 1891.
Sir : I have the honor to submit for publication Bulletin No. 24 of
this Division. It consists of a preliminary report upon the special in-
vestigation of the Cotton Boll worm (Heliothis armigera Htibn.) which
was authorized by Congress in the bill making appropriations for the use
of the Department for the fiscal year 1890-91.
Respectfully,

C. V. ElLEY.

Entomologist.
Hon. J. M. Rusk,

Secretary of Agriculture.

INTRODUCTION.

The present bulletin consists of a report made by Mr. F. TV7. Mally
upon the progress of the special investigation of the Cotton Boll-worm
which has been carried on under the Division since the appropriation
became available, July 1, 1890. Mr. Mally has had charge of the minor
details of the investigation, and has been constantly in the field since
last July. He also summarizes the results obtained by Messrs. Mc-
Neill and Booth. The Boll Worm was treated at some length in the
Fourth Report of the U. S. Entomological Commission, and the chief
object of the present investigation was to conduct further experiments
with remedies, as well as to verify the value of those already employed.
A thorough series of experiments has been planned with the diseases
of Heliothis and allied insects, in the hope of being able to practically
utilize them. Incidentally I have desired to ascertain new facts, if
possible, and to verify or disprove what has been previously written in
connection with the life history and habits of the species.

The observers have all been hampered in their work by the unexpected
lack of material. The funds were not available until the season was
three-fourths spent. The observations so far made will, therefore, have
to be supplemented the coming spring and summer. It transpires that
the ravages of the Boll Worm have been overestimated, and that while
from 20 to 30 per cent of the bolls are damaged in an average season in
Mississippi, onlv about one-third of this damage is done by this insect.
Several other species which do work somewhat similar to that of the
Boll Worm are treated in this report. Some new food-plants have been
found, and a careful study has been made of the habits and life history
which are here treated with more care and detail than has heretofore
been given to the subject. Two new parasites have been discovered, and
observations have been made which show that the egg parasite (Trich-
ogramma pretiosa Riley) is an extremely important factor in the economy
of this insect, as it is, also, in that of the Cotton Worm (Aletia xylina
Say), and the Grass Worm or Fall Army Worm (Laphygma frugiperda
Smith & Abbott). A careful count shows that 84 per cent, of the eggs
were destroyed by this useful parasite. All of the old remedies have
been once more tested, and the use of corn as a trap crop is again shown
to be one of the most satisfactory means of protecting the cotton crop.
The old subjects of attracting the moths to lights and poisoned sweets
have once more been carefully considered, aud my former conclusions
have been confirmed, that there is little to be hoped for from either of
these methods. The pyrethrum experiments, from which I had much

hope, have not proved very favorable, while experiments with a large
series of other vegetable insecticides have given no practical results
as yet.

The experiments with contagious diseases can not be reported upon
in any detail at the present time; but a large number of cultures of
several diseases of the Imported Cabbage Worm, the Bronzy Cutworm
and of two other Noctuids have been secured and carried through the
winter. What may prove to be a specific disease of the Boll Worm
has also been discovered, and cultures have been obtained. It results
from the few experiments made that the Boll Worm is probably sus-
ceptible to the Cabbage Worm disease, but positive statements can
not be made until these experiments are confirmed by those of another
season. A bacteriological laboratory has been established at Shreve-
port, Louisiana, and has been well fitted out with the necessary
apparatus, so that work in this direction the coming season will not
be hampered, except in the case of an unexpected paucity of Boll
Worms.

C. V. R.

LETTER OF TRANSMITTAL.

Shreveport, Louisiana, February 19, 1891.
Sir: In compliance with your request I have made out a report of
progress of an investigation of the History and Habits of the Boll Worm
(Heliothis armigera Hiibner), carried on under your instructions since
July, 1890, and submit the same herewith. The treatment of the various
subjects is not at all in detail aud has only been made complete enough
to give an adequate conception of what has been done, the present
status of the investigation, and what remains to be accomplished in
the future.

Very respectfully yours,

F. W. Mally,
^1 ssis to n t Entomo logist.
Dr. C. V. Eiley,

United States Entomologist.

THE BOLL WORM OF COTTON.

DESTRUCTIVENESS.

The damage to corn by the Boll Worm is difficult to estimate, owing
to the nature of the attack. Its ravages in the "bud" of the young
plants and later in the ends of the ears taken collectively no doubt are
considerable, though no definite per cent can be given. Tomatoes,
cucumbers, and melons also suffer more or less seriously from its rav-
ages. It is the attack upon cotton which is considered most serious
and supposed to be of great proportions. To determine the amount of
damage to cotton in the regions visited the past season the following
studies were made. The first was made August 14, in a large field of
upland cotton surrounded by woods. Two rows were taken at random
in the field ; the first was rank high cotton, the second a smaller growth.
About 10 feet of each row were marked off and all the bolls on the
plants in each counted. (See Table I.)

Table I.

Row.

Good bolls.

Loss by
Boll Worm.

Loss
by other
causes.

Total.

1

2 ...

290
270

2
1

95
43

389
314

The next study was made September 16, in a small field of rank
bottom-land cotton. The first five plants were taken at random, the
next fifteen successively in one row. (See Table II).

Table II.

Plant.

Good bolls.

Loss by
Boll Worm.

Loss
by other
causes.

Total.

l

2

40

21
4
30
34
19
18
35
17
70
33
42
49
3
10
33
25
72
17
7

7

2

3
10
11
2
9
9
C
7
4
0
0
0
2
2
2
0
0

13

0

20

34

'JO

0

15

12

9

1

9

4

0

0

2

11

24

3

0

52
3G
0
53
78
50
20
59
38
85
41

53
3
10
37

38
98

20

7

3

4

5

0

7

8

9

10

11

12

13

14

15

16

17

18

19...

20

Total . . .

579

78

182 839

10

September 17, a similar study of twenty-one successive plants was
made in another portion of the same field. (See Table III).

Table III.

Plants.

Worms.

G„o<,b„llJBJ-»-X.

Loss by
other
causes.

Total.

1

2

3 .

1
1
o

0
2
0
1
0
0
2
1
0
0
0
0
3
0
0
0
0
0

33
23
CO
30
41
65
11
29
53
53
24
63
21
19
23
88
22
36
14
20
49

4.
1
3
3
3
3
1
1
3
5
5
G
0
8
9
13
4
0
0
0
G

1

2

1

8

0

4

1

1

1

14

25

5

3

12

30

5

2

0

11

39

50
25
65
34
52
68
16
31
57
59
43
94
29
30
44
131
31
38
14
31
94

1,036

4

5

6...

7

8

9

10

11

12

13

14.

15

16

17

18

19

20

21

Total.

13

780

78

178

Table I should not be included in the table of percentages, since its
data were obtained early in the season, before the Boll Worm had really
become well established in cotton. Omitting table I we have the fol-
lowing table :

Table IV.

Percentages from Tables II and III.

Table.

Good bolls.

Loss by
Boll Worm.

Loss
by other
causes.

Total

loss.

II

Per cent.
.690
. 753

Per cent.
.092
.075

Per cent.
.218
.172

i
Percent.
.310
.247

Ill

Average .

.7215

.0835

.195

. 2785

The four preceding studies were made by a count of what was actu-
ally found on the plants at the time of observation. September 18 only
bolls and forms which had fallen were collected and examined. The
result is given below :

Number bored by Boll Worm 1G7

Number shed from other causes 3(52

Total 529

Taking the average of the total loss found in the same field on the
two preceding days, and tabulated as Tables II and III, and again in
Table IV, these 529 bolls maybe considered as equivalent to the count-

11

ing of 1,900 bolls by the method of Tables II and III.
therefore be given as below :

Table V.

This study may

1 Number.

Per cent.

Good bolls ] 371

.722
.088
.190

Loss by Boll Worm . „ . , i 1 67

Loss by otber causes 362

1,900 1.000

1

Averaging this result with that of Table IV we have the table given

below as the result :

Table VI.

Data.

Good boiis. L^rwofi ^*£«

causes.

Total
loss.

Table V

Study V

Average...

Per cent. Per cent. Per cent.
. 7215 . 0835 . 195
. 722 . 088 . 190

Per cent.
. 2785
.2780

. 72175

.08575 ; .1925 \ .27825

The above calculations certainly give the Boll Worm as much credit
as it deserves, and for the following reasons : T^e observations were
made after the cotton bad been "laid by v late in July, therefore the
fallen bolls collected from the ground in September covered what had
fallen during August and September. This is the period of greatest
damage to the cotton. No cornfields near by to lessen and detract from
the egg deposition on cotton. This in addition to the consideration of
the injured fruit actually on tlie plants but which was likely to shed,
certainly does not make the results arrived at much below the entire
damage done during that period.

From the results given above and from subsequent observation it is
evident that bottom-land cotton is worse infested than the "hill-coun-
try " cotton. Further, even in the same field, as is shown by the record
of plants 5 and 10 of Table II and plants 12, 16, and 21 of Table III
large, rank, leafy cotton plants, bearing a great number of forms and
bolls, are subject to much more serious attack.

The number of forms and bolls which one worm may destroy during
its period of existence can only be approximated. From the rate of
feeding during favorable conditions and when the larval state is about
15 days the number eaten into may range from ten to twenty. During
the longer periods of larval existence caused by unfavorable conditions,
the worms are inclined to move about more and perhaps injure more in-
dividual fruits, though the absolute amount eaten is not much greater.

What the extent of injury due to Boll Worm over the entire State of
Mississippi was the past year will be seen from the closing weather
and crop report of Prof. R. B. Fulton, observer, U. S. Signal Service,

12

University, Mississippi, from which is quoted the following : " In the
southern part of the State reports show that on account of injury by
Boll Worms and shedding, due to wet weather, the cotton crop will
be short from 30 to 40 percent, of last year's yield. * * * In the
northern section the Boll Worm did no material damage." It may fur-
ther be added that upon application to Mr. George E. Hunt, chief sig-
nal officer U. S. Signal Service, New Orleans, Louisiana, for weather
crop bulletins and the names of observers who had reported much
damage to cotton by Boll Worm last season, he replied that no material

Fig. 1. Heliothis armigera ; full-grown larva eating into a to-
mato—nat. size (after Riley).

damage was done from that source and none had been reported by any
of the volunteer or other observers. From this information as also the
facts shown by the preceding tables it is quite evident that the depre-
dations upon cotton by Boll Worms have been greatly overestimated.
If the statements of planters living in the regions where observations
were made are accepted, the Boll Worm was fully as numerous if not
more so than in previous years. Nearly all agreed that the damage was
fully up to the average, others thought it above the normal, but none
estimated it as being lower than usual. These statements together
with the almost daily reports obtained from interviews that " one-fourth
or one-third of the crop was being ruined" (?) led to above careful stud-
ies. The results simply show that on the whole the planters fail to dis-
tinguish the Boll Worm ravages from those of other insects, from phys-
iological phenomena of the cotton plant, and lastly, from some of the
fungoid diseases. In order to assist the planters in this matter a few
observations have been added at the close on " Other insect ravages
easily confused with those of the Boll Worm."

FOOD PLANTS OTHER THAN COTTON.

Com. — The habits of the Boll Worm when feeding on corn have been
so fully presented in the Fourth Beport IT. S. Entomological Commis-
sion, pp. 359-361, that only such observations will be given as verify

13

important points or add to our knowledge of the species. The manner
of attacking the ears of corn and the semi-solid excrement of the worm
which is left behind along its path into the ear, paves the way and pro-
vides a fertile soil for the germinatiou and subsequent growth of all
kinds of molds. The additional decay resulting in this way, aided by
the ravages of Dipterous and other larvre which revel in such matter,
perhaps fully equals the actual damage done directly by the worm. In
large fields of corn not often more than one large worm is found in a sin-
gle ear, but when the fields are small, and especially when surrounded
by cotton fields, there are often three or four nearly grown worms in a
single ear, and perhaps as many more newly hatched ones. This is a
direct result of the preference of the moth for the corn when the lat-
ter is in close proximity to cotton fields and of suitable age. When the
field is small the female often passes through it several times during a
single Might, depositing eggs as she does so. In this way I have often
observed the same female deposit eggs three times on the silks of
an ear of corn during a single visit. This of course is not the normal
method of deposition, and occurs only under the circumstances men-
tioned. If it were so the numbers of worms would be greatly reduced
through the agency of their cannibalistic habits, to be discussed here-
after in considering the history of the worms. Other females visiting
the field may also deposit on the same ears of corn, and so on. As a
result of this as high as fifteen to twenty-five eggs have been found on
the silk of a single ear of corn, and in addition as many more on the
husks and leaves.

In regions where corn is cultivated extensively a second crop is
planted late in July to produce a fodder crop by the end of the season.
These fields are invariably badly infested. When these are near cot-
ton fields they afford a great protection to that crop.

Except on cotton, hereafter considered, no observations of special in-
terest were made on any of the other well-known food plants, though it
was noted that the tomato crop suffered severely from Boll Worm dep-
redations.

Other Plants. — As additional food plants of economic importance
which are attacked by Boll Worm are to be mentioned the muskmelon,
watermelon, and cucumber. The cucumber is attacked usually by eat-
ing a hole near the base from below up into the center and then tunnel-
ing the length of it to the anterior end. The melons are usually bored
from the under side near the base, but occasionally at almost any other
point. The female was seen depositing on the following weeds : Relen-
ium tenuifolium, Amarantus retroflexus, A. sjrinosus, probably Erigeron
canadense, and one undetermined species of Panicum.

It was evident, however, that the young worms did not relish any of
the above weeds as food plants and left soon after hatching. This was
further verified by taking some branches of the plants just named to
the laboratory and placing newly hatched worms upon them. They

14

fed sparingly upou the small flowers aucl tender stems, but soon left
the branches and could not be induced to remain long. The female
shows no inclination whatever to deposit her eggs upon the last-named
host-plants except as they may be found in corn fields or near by
From here the young worms can easily migrate to the corn plants near
at hand, and from observations already cited it is quite probable that
they do so. These last observations have, furthermore, led me to sus-
pect that the female may occasionally deposit upon all weeds or other
plants indiscriminately growing in a corn field and suitable for this
purpose.

CHARACTERS AND TRANSFORMATIONS,

THE EGG.

The egg is oval, the greatest diameter being very near the base. It
tapers but little from the point of greatest diameter to the base, but
slants much more towards the apex. The vertical diameter averages
0.375 millimetres, the horizontal and greatest diameter 0.5 millimetres.
The sculpture of the eggs consists of polar ribs with cross bars, giving
them a checkered-appearing surface. When first deposited the egg
appears nearly a pure white, but soon turns yellowish as the growth of
the embryo begins, and deepens as the latter develops. After about
25 or 30 hours that part of the embryo at the apex of the egg is notice-
ably darker, and between it and the center of the egg a reddish or brown-
ish band is formed. The latter so far as can be seen extends only part
way round the egg.

This band is later absorbed into the body of the worm and the
darker spot at the apex is found to be the head of the developing larva.
At this stage the body of the worm can be quite definitely seen through
the eggshell.

The duration of the egg state varies somewhat, as will be seen hereafter,
with the meteorological conditions prevailing at and immediately fol-
lowing the time of deposition. One lot of eggs deposited in confine-
ment at night and followed by two very hot days began hatching within
45 hours. But of a number of lots of eggs deposited in confinement
from time to time, the duration of the egg state was usually from 2h to
3 or 3 J days. This may be considered about the normal duration of the
egg state. Several lots, however, which had been deposited dining un-
favorable weather did not hatch until after 4 days 5 in a few instances
a few hours over 5 days.

THE LARVA.

The newly hatched larvae, before they have taken any food, average 1.54
millimetres in length, are slightly larger anteriorly, tapering gradually,
as is shown by measurements of the diameters of a number of worms at
the first, middle, and last segments, whose averages were 0.23, 0.20, and

15

0.14 millimetres, respectively. The general color of the body is white,
with a yellowish tinge j head, black ; a black or brownish shield-shaped
spot on the dorsal surface of the first segment.

Soon after they begin feeding the larvae turn darker and before the
first molt are usually of a deep rose or brownish color. The piliferous
tubercles are not yet very prominent. The true legs at first are slightly

Fig. 2. — Heliothis armigera : a, egg from side ; b, do. from
top — enlarged ; c, full-grown larva ; d, earthen cell with con-
tained pupa; e, moth -with wings spread; /, do. at rest— nat-
ural size (after Riley).

dusky, but soon turn much darker, are hairy and provided with a small
bifid claw. At first the first pair of prolegs seem to be a little less
robust than the others, and hence may be slightly weaker. Obscure
dark lateral patches are found on the prolegs, which at this stage are
further provided with fine small hooks.

By the time of the first molt the worm has attained a length of 5.62
millimetres and is slightly larger in the middle.

After the first molt the larva is at first of a yellowish color, but again
turns darker rapidly when it begins feeding. The true prolegs become
much darker, and at their insertion next the body a small dusky spot,
both anteriorly and posteriorly, is found. The dark lateral patches on
the prolegs are also more distinct, each proleg being now provided with
nine small hooks.

By the time of the second molt the larva measures 7.75 millimetres
in length and is still a little wider in the middle.

Soon after the second molt the worm measures 8.75 millimetres in
length. The most noticeable changes are, that the piliferous tubercles
are now much more prominent ; that the very small tubercles found
thickly scattered all over the body first become quite discernable to
the naked eye.

16

With the three subsequent molts there are no marked changes except
in size and the distinctness in definition of the various colorings and
markings of the body of the larva.

The mature worm varies in length from 31 to 36 millimetres, with
diameters of about 4, 5, and 4 millimetres at anterior, middle, and pos-
terior regions, respectively. The head of the mature Boll Worm is never
darker than a light brown, or, in the darker colored worms, mahogany,
but may be paler according as the specimen is a lighter colored one.
The true legs are dusky or blackish, as also the spots on the j)rolegs.
The latter are now each provided with fifteen small hooks.

The color of the body of the mature worms varies from all gradations
of the darker or rose-colored specimens to those which are light-greenish
with a faint rose tint or entirely light-greenish. The darker colored
ones greatly predominate from about August on through the remainder
of the season. The markings of the worms which are most frequently
met with at this time are as follows :

Along the median line of the dorsal region is a brownish or black-
ish stripe containing in its center an interrupted white line. Next, the
subdorsal stripe, which is lighter colored, and along which is found the
first subdorsal row of piliferous tubercles. Below this is a subdorso-
lateral stripe, which is usually about the same color as the dorsal one.
The subdorso-lateral stripe is slightly wider at the center of each seg-
ment, and within its borders are found two more rows of piliferous tuber-
cles. Next comes the lateral or stigmata stripe, which is usually pure
white. Along this stripe are found the spiracles and one row of pilif-
erous tubercles. Between the lateral stripe and the prolegs is a stripe,
Avhichis usually of the same color as the ventral surface, which latter
is a uniform whitish, This sublateral stripe contains two rows of
smaller tubercles. In many of the darker rose tinted specimens this
stripe is often nearly a pure rose color, in which case the stigmatal
stripe is also more or less tinted. Throughout all the stripes except
the lateral and sublateral ones, and in these where they are colored
other than white, are found numerous interrupted, irregular, white lines
and spots. The stripes are ail of a uniform width throughout, with the
single exception mentioned. In the lighter colored specimens none but
the dorsal and subdorso-lateral stripes can be distinguished at all, and
often only the dorsal one.

The first segment is provided dorsally with a denser, calloused pare
irregularly shield-shaped and which is pale brown or black, according
as that is the general color of the worm. In the newly hatched worm,
however, it nearly always appears as a very distinct black patch. This
calloused portion has a distinct median groove dividing it into two
symmetrical halves, each of which bear similar markings as follows :

Four small pits in the form of a trapezoid, the posterior pair being
nearest together; from each of these pits extends a fine short hair. In
the center and extending each side of the median line are two deep

17

transverse parallel grooves. At the curved lateral regions of the shield
are found slight depressions, and again one at the posterior end of the
median line. The shield is further traversed by irregular lines or
wrinkles, and in addition contains short white interrupted lines and
spots much the same as those found in the stripes of the body. Dor-
sally the second and third segments each have a transverse row of four
piliferous tubercles. The fourth to tenth segments, inclusive, have each
four piliferous tubercles, slightly larger than those of the second and
third, and are arranged in the form of a trapezoid, the anterior pair be-
ing nearest together. On the eleventh segment the four tubercles are
arranged in the form of a square ) on the twelfth the trapezoid is re-
versed, the posterior pair of tubercles being nearest together.

Laterally, the first segmeut is provided with a spiracle about which
are two piliferous tubercles. Below there are two smaller tubercles
which are usually contiguous and appear much like one tubercle from
which two bristles project. The second and third segments each have
two large piliferous tubercles, two slightly smaller ones and two still
smaller. The fourth to eleventh segments inclusive each bear three
tubercles arranged in triangular form about the spiracles with a fourth
smaller one below them. The twelfth segment usually has two placed
contiguously, or nearly so. Centrally the tubercles are all small. The
fourth segment bears six small tubercles arranged so as to form the arc
of a circle. The fifth segment has two transverse rows of tubercles,
four in each, the ones in the anterior row being wider apart than those
in the posterior. The tenth, eleventh, and twelfth segments each bear
a transverse row of four tubercles. All the tubercles throughout have
projecting from them a short, stout hair, or bristle.

The growth of the larvae is somewhat slower during their earlier
stages than when half grown and approaching maturity.

AVhen neariug maturity the molts occur at shorter intervals. The
intervals between molts become longer later in the season or with cold
and unfavorable weather. Thus in August the first molt occurred on
the fifth day after hatching, the second on the tenth, the third on the
fourteenth, and the fourth on pupation, seven days later. September
25 some eggs began hatching; the first molts occurred on the-seventh
and eighth days thereafter, the second on the fifteenth and sixteenth
days 5 at this time the particular worms under observation made their
escape and the interval of subsequent molts could not be recorded.

The length of the larval state is longer later in the season. Thus at
SliL'eveport, Louisiana, Dr. A. R. Booth reports that during August a
worm matured and was preparing to pupate after having fed 15 days 5
another after 18 days. At Holly Springs, Mississippi, for the same
month worms matured within 18 to 20 days after hatching. For Sep-
tember worms hatched about the first of the month, matured in from
21 to 26 days, while larva3 hatched September 25 escaped after 18 days
days and had only just molted the second time.
23024— No." 21—2

18

The first food of the worm is the eggshell from which it has just
issued. No sooner is the young larva fully out of its shell than it crawls
away once or twice its length, turns around, and eats the shell either
entirely or but a portion of it. This done it crawls about a short time
until it finds a spot suited to its taste, usually first spreading a few
threads of a frail web. Under this the worm feeds on the epidermis of
whatever surface it may have selected. Often they crawl but a short
distance, drop themselves down by a slender thread until another sur-
face is struck, when they crawl away as before and begin feeding.

Before the first molt the larvae feed principally outside of the young
forms and squares on whatever surface they may be when hatched.
At about the time of the first molts, or soon thereafter, they begin seek-
ing out the forms, blossoms, and squares and begin their destructive
work of boring into them. It must be stated, however, that from the
first a few may be found to go directly to a small form or square and
either boring through the involucre and thence into the blossom or
fruit, or else first crawling between the involucre and fruit and then
boring into the latter. Often, too, they at first simply hide behind the
involucre, eating the epidermis from the inside and boring into the fruit
later. Preferring the tender growing portions of the branches, the
youuger worms after finding them secrete themselves in the opening
clusters of leaf and flower buds and begin boring.

Their first feedings upon the leaves, whether on the upper or lower
surface, are principally confined to the epidermis. About the time of
the first molt they begin to eat small holes through the leaves. This
is not continued long, however, as at this time the worms begin their
search for the forms and squares. During the youuger period the worms
feed most on the reprodutive organs of young forms or such as are
nearly grown and opening. Some time later, after about the second molt,
they begin boring more plentifully into the larger bolls. The older
worms, while feeding principally in larger bolls, do not confine them-
selves to them and are often found devouring a form or opening blos-
soms. Due to their choice of food (speaking only of the bulk) the
younger worms are found mostly on the younger tender-growing por-
tions of the cotton plant, and the more nearly grown ones on the more
mature portions. For a change the full-grown worm sometimes feeds
on a leaf or eats through the green stem of a newly formed branch.
Stems which were .25-.30 inch in diameter and having above them a
number of forms have been observed to be eaten nearly through by
them. A day or two after the branch would be found wilted or nearly
broken off.

The habits of the worm, as to the manner in which a boll is entered
and the extent to which it feeds upon the same after it has entered, are
also rather variable. The worm may pass in behind the involucre and
bore into the boll from the inside, or it may choose to do so from the
outside, eating through the involucre and then into the boll. The hole
is usually bored from the bottom and passes towards the apex; fre-

19

quently, however, tbe bole is bored about midway of the boll and passes
straight iii. As the worm tbus feeds on the inside of the boll it may
pass straight towards the apex, devouring only the section of the fruit
which it has entered and then retreat, and attack another boll, or it
may bore through the partition into a second section, devouring it,
or the passage of the worm may be a slanting one from the first, in
which case two or three sections may be entered before retreating. The
worm seldom continues until it has destroyed all of the sections of the
boll, though the remaining sections are usually made worthless by the
decay which arises and spreads from the injured portions.

The notion which some planters have that the worm eats in at one
point on the boll and passes out at another is wrong, for if it ever does
so it is certainly the exception. Occasionally, however, there are two
holes in a boll similar to those which the Boll Worm makes and in-
deed the Boll Worm has been the culprit. Observation, however, has
proven that the Boll Worm occasionally begins boring atone point, eats
into it but a short distance, retreats, and begins a second hole at another
point on the same boll. Sometimes, also, two worms are found attack-
ing the same boll, which, when they have finished and left it, appears
as if a worm had entered at one point and passed out at another.

It is further believed by some that the Boll Worm travels only at
night and feeds only late in the afternoon and evening. Concerning
both it must be said that the worms avoid the extremely hot sunshiny
part of the day, and prefer traveling and feeding during the cooler
parts; but at the same time considerable range must be allowed for the
whims of the worm, and other conditions which may arise to induce
the worm to go from boll to boll during mid-day and often at high
temperatures. As bearing on this it may be stated that often while
making morning observations in the cotton fields worms were found
in bolls. Without any disturbance these branches were marked so that
they could be easily found for experimental purposes in the afternoon.
In the afternoon it was always found that a number of worms had gone
to other bolls or even branches on the same plant. Occasionally, too,
one would be found to have left the plant entirely and could not be
found. The changing of plants was not often done during the day, but
was oftener found to be the case with those specimens which had been
marked the preceding evening and were looked for the next day. The
changing from boll to boll on the same plant may be frequently ob-
served during the day. The time of day when the worms feed most
vigorously is during the cooler portions of the afternoon and evening
and in the morning before the sun shines so hotly.

After the killing frosts in late October and November the worms
which had not yet matured were found to feed as best they could upon
the berries of SSolanum carolinensc, and possibly, if forced to, would
feed upon any other available green plant found in the cotton fields.
Extensive observations on this point could not be made, since the worms

20

were not at all plentiful. It remains to be added that the larger
Heteropterous insects, mostly Metapodius, were abundant upon Sola-
num at this time, and that probably great numbers of the remaining
immature worms fell a prey to them.

The cannibalistic habits of the Boll Worm are also of importance and
must be considered here. First to be mentioned in this connection is
the fact that the Boll Worm devours its larval skin as soon as molted
and does so before it begins feeding on vegetable matter again. When
collecting worms from the fields considerable care must be exercised to
have a box or basket large enough to receive some foliage, in order that
the worms may not iujure and destroy each other before reaching the
laboratory. In the breeding cages the Boll Worms not only attack
each other, but almost any other larva placed in the arena. In a state
of nature they have not been observed to deliberately attack each other
except in corn, especially when the field was a small one, and great
numbers of eggs are deposited and hatched on the same plant, so that
an unnatural number of worms expect to feed upon the same ear of
corn. The same may be said concerning the "bud," if the plant be
young yet.

In several instances where such small fields of corn were found near
large cotton fields, examination of the ears resulted in finding from three
to six worms of various sizes in them. A day or two later when the
same ears were examined some of the smaller ones were found to be
badly bitten (enough to cause death), and some dead ones whose bodies
gave evidence of a violent death. None of the worms had been de-
voured, and it seems that the warfare had been brought about by the
worms intruding on each other's territory, as feeding progressed, and
that nothing more was done than would maintain their positions.

THE PUPA.

A number of Boll Worms which had been fed on bolls in breeding
cages matured and were placed in a fruit jar about two-thirds full of
earth. The worms at once entered the earth, formed their cells, and
pupated about 2 or 3 days after. The pupa is a beautiful green at first,
but soon turns to a light brown or mahogany. When all the worms
had pupated the earth was examined to determine something about
the depth of the burrows and cells and the manner in which they were
made. The earth had been recently placed in the jar and was quite
loose when the worms were placed on it. The depth of the burrows
varied from 2J to 5 inches and throughout their course were partially
filled with loose earth. The cells were all much larger than any part
of the burrows and extended upward from the end of the same. One
worm had come to the surface of the glass in making its burrow,
and was observed while making its cell. The worm seemed to test
carefully every part of the wall of the cell to see that all was firm. The
whole surface was then thinly coated with a sticky fluid, at the same

21

time adding a small amount of webbing. Tbe latter was plainly to be
seen on the glass of the jar.

Some of the mature Boll Worms which had been left in the breed-
ing cage without earth pupated unprotected on the floor of the same
and later issued as moths. In another instance, however, the mature
worm had been left in a newly-made breeding cage. The next morn-
ing the worm was found to have nicely webbed together the sawdust
accidentally left in the corner, and had formed a neat little cell, which
might almost be called a cocoon. This shows plainly that the Boll
Worm is capable at least of spinning a certain amount of web for its
cell, whether it always does so to so great an extent or not.

It was stated above that the worms pupated about 2 or 3 days
after having entered the earth. This, however, applies only to the
months of August, September, and October. Later the time is longer,
as is shown by the following observations : On November 1 a mature
Boll Worm from a breeding cage was taken to a cotton field and placed
on solid earth, a large open tin can being placed about the worm so
as to compel it to make its burrow under observation. The work of
digging its burrow was begun at once, and by the next day it had dis-
appeared below the surface. A small conical-shaped mound of loose
earth was formed about and over the opening of the burrow. On No-
vember 4 two other mature Boll Worms were similarly placed in open
cans, the one on solid earth, the other on earth which had been dug up
and made very loose. So far as could be seen without digging the bur-
rows were made exactly as the one just mentioned.

On November 14 all the burrows were carefully followed up with a
small trowel, to determine depth, condition of the burrows and cells.
The worm placed on solid earth, November 1, had proceeded as follows :
Down for an inch, then slanting at about an angle of 120 degrees for
2 inches more when the cell had been made .upward. The cell was
about an inch and a half long, and was therefore within an inch or less
of the surface. Very little webbing was noticeable along the burrow,
and but little in the cell. The burrow was about 0.25 inch in diameter,
larger at the distal end, aiid contained a small amount of loose earth
along its entire length ; also a plug of loose earth about 0.25 inch long
at the distal end next the cell. Having been in the earth 2 weeks, I
was surprised to find that the worm had not yet pupated. The burrows
and cells of the other two worms were examined, but no special differ-
ence worthy of note was found. The peculiar facts in all were : (1)
That the cells were all inclined and higher than the lowest part of the
burrow, thus bringing the pupa above the latter; (2) that the cells
were so near the surface; (3) that the larva3 were all found with their
heads at the upper end of the cell, wherefore the pupae would have
been found with the anterior portion highest and resting on posterior
end ; (4) that they had not pupated after having been in the earth so
long a time.

22

The duration of the pupal state as noted by Dr. Booth for August to
September 2, was 10 to 11 days. At Holly Springs, Mississippi, a
uuraber of worms pupated between August 28 aud 31. Some of the
pupae issued after 15 days and others not uutil after 27 days. Two
others which had pupated September 4 and 7 issued September 20 and
30, making 16 and 22 days for the pupal state, respectively. Another
worm pupated August 3L. This pupa was alive, but had not issued at
last observation, October 20, when it was injured and died later.

For description of pupa see Fourth Report, U. S. Entomological Com-
mission, p. 371.

THE IMAGO.

For description of the moth see Fourth Report, U. S. Entomological
Commission, p. 371.

The sexes of the Boll Worm moth can usually be readily recognized,
especially if but a short time has elapsed since their issuance. After
the females have deposited most of their eggs and their wings have-
become worn and battered the sex is less easily distinguished. The
body of the female is noticeably more robust than the male; especially
the abdomen, which is distended somewhat by the eggs which are being
matured. The end of the abdomen is ovoid, acute, the tip not provided
with so large a tuft of thick hairs as is that of the male. The abdomen
of the male tapering more gradually is slightly longer, of less diameter,
and more cylindrical.

When feeding or ovipositing, the flight of the moth is much slower
than when flying long distances. The moth approaches a flower or
gland, often steadying itself with the fore legs, in any case fluttering
its wings rapidly, with antennae in constant motion. Sometimes they
alight to sip sweets, or perhaps to rest ; in either case the wings are not
closed down upon the body, but are partially spread and elevated, leav-
ing bare the abdomen. If, however, the moth alights to hide, the wings
are folded down closely upon the body. Much the same flight is ob-
served in the female when ovipositing, sometimes alighting to do so,
but usually only steadying herself with the fore legs. In the act of
oviposition, the abdomen is bent forward sickle-shaped, bringing the
apex squarely upon the surface to be deposited on. The time occupied
in depositing an egg in this manner is equivalent to the time it requires
to count three or four slowly. At this rate, and in the interval of flying
about from plant to plant, some half dozen eggs are deposited, when
the moth is seen to fly away. So far as can be determined these inter-
vening flights are for the purpose of feeding and rest from labor.

Though the habits of the Boll Worm moth have been classed as noc-
turnal, aud tbey are principally so, yet its diurnal habits are perhaps of
geater importance than has hitherto been supposed. During continued
daily observations on the Boll Worm in the field the moth was frequently
seen flying about, and at times observed to feed. Upon closer obser-

23

vation it was found that on pleasant sunshiny afternoons the moths
flew about quite plentifully, feeding freely during their flight. Often
while standing in a patch of cowpeas, from about 3 p. m., have I ob-
served the moths, without any previous disturbance, rise here and
there, fly about the pea blossoms or the glands at the base of the young
pods, sip their exudations for a few minutes at a time, and then fly away
a short distance, alight to rest and hide. In the early part of the
afternoon the length of their visits to the pea blossoms are short and
the time of their hiding longer than later. About 5 to G p. m., when
the sun is yet quite high, the moths begin to fly for a longer time and
their hidings are of shorter duration.

My own experience with the moth has been that it feeds freely from
about 4 p.m. until suudowu, when the females begin depositing their
eggs, feeding being apparently a minor matter at this time. The moths
thus seen flying about in the afternoon are not confined to one sex,
though the males predominate during the earlier period until an hour
or so before sundown, when both sexes appear presumably in about
equal numbers. The favorite food of the moths at daytime are the
blossoms and other secreting glands of the cowpeas ; they also feed freely
on clover and Selenium tenuifolium. During the day the moth is seldom
met with in the cotton fields either as feeding or by being flushed. It
therefore appears that during the day the moth prefers to hide and
feed upon plants some distance away from the cotton fields. During
twilight ami night the moths are found abundantly in corn and cotton
fields, and feed almost entirely upon the exudations of the various
glands found on the cotton plant.

But not only are the feeding habits of the moth partly diurnal but
also those of the deposition of eggs. Though on several occasions a
moth was seen depositing eggs on corn, and once on cowpeas in mid-
afternoon, the habit is not one of frequent occurrence.

These diurnal habits of feeding and occasional deposition are of great
importance in the consideration of the utility of lights as traps for
catching the moths at night, and will be taken up more fully in that
connection.

The number of eggs which a female may be capable of depositing
is difficult of absolute determination, but has been shown to be much
greater than was supposed. A female which had issued in confinement
was placed in a breeding cage with a male which had issued the suc-
ceeding day. Five days after deposition of eggs began, and continued
for 7 days. During this time C>87 eggs were laid. Unfortunately,
the female which was thus under observation after having died became
mixed with other dead specimens on the table and hence no dissections
were made with a view of determining whether any well developed
eggs remained in the abdomen or whether the number of potential ova
was great. A female captured August 5, and confined in a box de-
posited 627 eggs in one night. Another, captured August 8, and

24

kept in the same maimer, deposited 468 in one night. Another, cap-
tured August 14, deposited 505 eggs the first night, was kept in a
tin box without food during the next day and deposited 125 eggs more
on the second night, making 630 in all. The first female spoken of
above as having deposited her 687 eggs on seven successive nights did
so as follows, begiuning with the first night and continuing in order:
49, 5, 10, 436, 147, 22, 18 : averaging 98 per night.

The life of the moth was probably shortened by confinement, and
therefore the average number of eggs deposited each night is entirely too
high for oviposition under normal conditions. The record, however,
shows that a climax in egg deposition is reached after a certain period.
(The possible importance of this fact is considered under the head of
lights as traps for the moth.) The total number of eggs deposited by
the four females above noted was 2,413, averaging 603 per moth. But
the number of eggs deposited in one night by the moths captured and
confined show that their period of greatest egg deposition had already
been reached by them. Judging from the record of the moth whose
deposition of eggs was observed from the first, it seems safe to suppose
that probably each of the other moths had deposited about 50 eggs
previous to being captured. This would raise the average to 653 eggs
per female. But again, from dissections of the females thus observed in
confinement it was found that a number of well developed eggs re-
mained, and usually also a great number of potential ova. The aver-
age of 653 eggs per female is therefore certainly not too great and proba-
bly much too small.

What the number of eggs deposited in one night in a free state of
nature are can only be approximated. As has been stated the female
deposits four, five, or more eggs in succession, then flies away, feeds
or rests a time, afterwards repeating the process. The interval of
nondeposition is necessarily variable, though as near as can be deter-
mined about 5 to 15 minutes. How long deposition is continued dur-
ing an evening is also not to be definitely stated, though it is noticeable
that the moths begin to decrease greatly in numbers soon after 8 o'clock,
thus approximately the time is about 2 hours. From these data it may
be approximated that from about 30 to 60 eggs are normally deposited
in a single night. For the nights during the period of greatest deposi-
tion, the number deposited is probably much larger.

Upon the number and distribution of the eggs upon the various parts
of the host plants the following data are collated : By actual count the
number of eggs found on five corn plants is as follows, in averages: per
plant, 74 distributed as follows: tassel, 10; leaf sheaths, 10; leaves,
14; husks of ears, 15 ; silks, 25. The above count was made in a small
patch of corn surrounded by cotton fields with no other corn near,
wherefore the number of eggs per plant was perhaps greater than on
plants in larger fields of corn. (This point will be considered more fully
under corn as a protection to cotton, which see.) Of five cotton plants

25

the number and distribution of eggs in averages was as follows: per
plant, 7 ; leaves, 4 ; involucre, 1 ; stem and petiole, each, 1. These are
the data for plants examined at random at different times while mak-
ing observations. Bearing upon the choice of the place of deposition
the record of the female already spoken of as depositing in confinement
is interesting, and for the whole period of deposition was as follows:
on leaves, upper side, 37 ; under side, 110 ; stem, 23; petiole, 51; in-
volucre, 4 ;

The eggs are deposited upon quite a number of host plants. Corn is
unquestionably preferable if not too near maturity. Cotton perhaps
ranks next, though they deposit freely upon cowpeas. In addition to
these the moth was observed to deposit on the flower heads of Helenium
tenuifoliuni) Amarantus spinosus, and A. retroflexus with Datura stra-
monium, upon which the moth feeds occasionally, Erigeron canadense,
and a species of Panicum, on the doubtful list. Deposition was not
actually observed on any other host plants, though presumably the moth
will deposit on those plauts ivpou which the worm is known to feed.

The food of the moth is quite diversified. During the night, so far
as observed, the secretions of the various glands of cotton seem to be
their main food, while during the day the same may be said of cow-
peas, Helenium tenuifolium, and to a lesser degree of newly protruding
corn tassels ; occasionally, Amarantus retroflexus, A spinosus, Datura
stramonium, and perhaps some of the grasses. There may be many
other flowers, glands, possibly also fruits, which are visited, but which
did not come under observation.

The regular flight of the moth is very swift and never very high. In
flying some distance in a cotton field it seldom rises to the level of the
tops of the cotton plants, but flies lower, darting this way and that be-
tween the plants and foliage in the rows, and in this way from row to
row across the field. This manner of flight is also of importance in ad-
justing lights as traps for them, and is referred to its appropriate
heading.

During the greater portion of the day the moths remain hid. If in
corn field they are found down behind the sheath of the blades of the
stalks, about two or three feet from the ground. But most of the moths
hide outside of both corn and cotton fields, around the edges in the
weeds, under dried grass and rubbish, or in adjoining fields of clover
or cowpeas. When found hiding in these places they are usually upon
or near the ground, wings folded upon the body, aud so located that a
dried blade of grass or other object quite completely hides them from
view.

NUMBER OF BROODS AND HIBERNATION.

Observations having only begun in August, the notes taken begin
with the fourth brood, which is the one which first begins to deposit
freely on cotton.

26

The broods overlap each other mostly as a result of a difference in
the rapidity of growth of many of the worms. One lot of Boll Worms
obtained from eggs deposited in a breeding cage by a single female in
a single night, and later hatched on the same date, were reared under
the same conditions. Some of these worms matured and entered the
earth for pupation, while others were yet but half or two-thirds grown.
Due to such great irregularity in the length of the larval state fresh
females of the fourth brood may be found along with the first to issue
of the fifth brood in September. Not much work of the Boll Worm in
cotton is noticed until August or early in September. From about the
middle of September the moths of the fifth brood begin appearing,
and continue to the last of the mouth, or even the first of October,
after which time they are not often met with. At least a partial sixth
brood begins appearing late in September and early in October, and
consist principally no doubt of those individuals that have undergone
their transformations rapidly. That all of the sixth brood does not
appear is evident from the fact that they are fewer in number than any
of the earlier broods. ' Hence many of the pupa? of the fifth brood of
moths pass through the winter as such and form a part of the first
brood in spring. The moths of the last brood appearing so irregularly,
worms hatched from eggs of this brood are found in all stages as late
as November 20 to December 1. At Shreveport, Louisiana, during
this period Boll Worms were found on cotton which had only molted
the second time and were therefore only about one-third grown.

But this irregularity in duration of certain stages of the insect is not
confined to the larva?, but to the pupae as well. Of a number of pupae
which had been kept over from September and October one issued at
Shreveport, Louisiana, December 12. Whether the moths, if there be
many which issue at this time, hibernate as such or deposit their eggs
at once and die soon after, has not been determined positively. If the
latter be the case, the progeny will certainly be entirely lost, since no
living food plants are found at this period. The Boll Worms which were
yet immature at the time of the killing frost early in December were
quite certainly destroyed, as nothing remained for them to feed upon.
As to whether the moths hibernate, I can only say that close and con-
tinued search during December has failed to discover the moth. This
may not be surprising, however, since certainly the moths which issue
at so late a date, and which would therefore be likely to hibernate, are
very few in number, and hence would be met with perhaps only acci-
dentally during the winter season. Though a few moths issue at so
late a time as has been mentioned, the other extreme is also met with
in the pupal state. Evidence of this is the fact that several pupa?
which were obtained from breeding cages late in August had not issued
up to November, when they were still alive, but were accidentally in-
jured and died.

It must be kept in mind throughout in speaking of certain stages of

27

the species that a majority only of that stage is referred to. It should
further be remembered that these observations apply only to the north-
ern region of "the cotton belt" and doubtless can be much enlarged
upon by observations in more southern portions.

NATURAL ENEMIES.

The fact that the Boll Worm was so scarce during the past season
precluded making extensive observations along this liue.

Among the vertebrates only circumstantial evidence was obtained.
In one instance where the dissevered wings and torn bodies of Helio-
this were found under and near a large tree in a cotton field it was
also found that a " butcher bird " had its nest on one of the upper
branches. Another was the case of a negro tenant, who complained
about the crows lighting on his corn plants in the field aud eating into
the end of the ears. Upon examination it was found that the corn was
badly infested with Boll Worm. This alone could not serve to estab-
lish the fact that crows picked into the ears for the primary purpose of
feeding on the worms. Feeding on the tender grains of corn beneath
the husk quite probably an occasional small Boll Worm was eating. It
is also probable that still others will be injured by the pecking into the
ends of the ears. The crops and stomachs of a number of quails were
examined and though they had them shot about cotton fields no Boll
Worms were found in these parts of their digestive organs. Upon vis-
iting the fields about which they had been shot no Boll Worms could
be found and the negative result has therefore no great significance.

A common species of Soldier bug (Podisus sjrinosus) was found de-
vouring a large full-grown Boll Worm. An immature capsid (near
Leptoterna) was overlooked and left on a branch of cotton placed in a
breeding cage for a female to deposit upon. Soon after deposition
some of the eggs showed signs of shriveling and were supposed to be
sterile. Close examination, however, led to the discovery of the de-
stroyer, which was as yet but a pupa. The eggs being nearly empty
it was evident that the pupa had punctured the eggs and sucked their
conteuts. The same pupa was then placed on a branch of cotton with
some newly-hatched Boll Worms, all of which fell victims to its beak.
A common species of the robber flies (Erax lateralis) was also seen to
catch the moth while on the wing.

No -observations could be made upon ants in relation to the Boll
Worm, since the latter were not abundant enough for that purpose.
The auts have been watched on corn for an hour without noticing an
attack upon the eggs found deposited there. They are occasionally
seen to enter the holes through the husks into the ears, but I did not
observe that they went in for the purpose of attacking the Boll Worms.
They only sipped freely of the juices and ferments of the injured kernels
of corn and the excrement of the worms. Sometimes dead worms are

28

found in the ends of ears into which ants have entered, but the condi-
tion of the worms plainly indicates that they had not been bitten or
tormented to death. In fact, worms under similar conditions, except
the absence of the ants, are often found, but from which parasites are
usually bred. A nearly grown Boll Worm was placed in the path of a
great army of ants, but was not caused any great incouveuience by
them. Sometimes an ant would ruu up on the back of the worm, but
the twisting, jerking, and rolling of the worm soon displaced the in-
truder and the worm escaped uninjured.

Of the three parasites, the one attacking the eggs (Triclwgramma
pretiosa) is most important, though there are at least three others at-
tacking the worms. A small Chalcid* was bred in great numbers from
a Boll Worm captured in the held and transferred to a breeding cage
to rear. The worm had been dead for a day or two before the para-
sitic larva? issued from its body. These did not form silken cocoons
but pupated nakedly on the side of the glass bottle. At least two
species of Tachina deposit their eggs on the backs of the worms. The
one deposits a pure white egg, the other a deep brown or black one.t

Both kinds are of the usual form and size of Tachina eggs. Great
difficulty has been experienced in rearing the dipterous larvoe after
issuing from the dead body and I have thus far obtained no adults.

As already stated, the most important parasite is the small Tricho-
gramma of the egg. The number of eggs which were found to be de-
stroyed by this parasite was simply amazing. In small patches of corn
near cotton fields it was noticed that of the many eggs found on the
husks and blades but a few retained their normal color, but soon turned
dark or entirely black. Of the 57 eggs taken from some 8 or 10 corn
silks from this field October 18 only 7 hatched. The remainder were
kept in a vial for a time, when later the parasites issued in abundance.
In this instance 84 per cent of the eggs had been destroyed by the
parasites. This per cent may be a little too high for the average, but
judging from the large majority of eggs seen on the plants, which were
black and evidently parasitized, it is certainly conservative to say that
during the Fall season 75 per cent of the eggs are destroyed through its
agency.

INSECT RAVAGES EASILY MISTAKEN FOR THOSE OF THE BOLL

WORM.

Owing to the fact that many planters attribute all of the shed forms
or boils which show any signs of insect attack to the work of the Boll
Worm, it seems advisable to treat briefly of a few other insect depreda-
tions which are not well understood by them, and whose marks upon the
fallen squares may readily be mistaken by an inexperienced eye.

*This was Hexaplasta zigzag, and is a parasite of Phora and not of Aletia, Phora
being a scavenger on dead larvae of all kinds in the South. — C. V. K.

tNo black Tachiuid eggs are known, and these were doubtless the eggs of Eu-
jjJectrns eomstockii. — C. V. R.

29

EUPHORIA. MELANCHOLICA.

These beetles, together with the four species of larvae immediately
following, are perhaps of greatest importance in this connection. The
first observation upon this species was made at Lamar, Mississippi.
It led me to believe that the beetles did original boring into the bolls in
order to reach the soft parts and their juices inside. Subsequent ob-
servations have not verified this opinion. The beetles observed at
Lamar were found on a boll with their heads inserted into a small per-
fectly round hole about an eighth of an inch deep, or just deep enough to
reach the soft parts beneath the pericarp. Few Boll Worms had been
found in the field, and it seemed quite probable that the beetle had eaten
out the cavities themselves. At Holly Springs, Mississippi, where the
beetles were found quite plentifully in some fields, none were ever again
seen under similar circumstances.

Flying about among the cotton plants during an afternoon they would
be seen to alight on some boll which had been recently bored by the
Boll Worm, but which had already been deserted by it. Here the
beetle would sip of whatever juices there might be coming out of the
injured boll. This is quite profuse at times, especially from those bolls
of which the Boll Worm has but partially destroyed a certain section.
From these proceeds a profuse frothing ferment, highly relished by the
beetles, for occasionally two or three may be found at such bolls.
When no boll with this tasteful exudate is found, they often alighi on
the tender-growing portions of a branch where leaf and flower buds
may as yet be found but partially developed. They crowd down be-
tween these and puncture the tender and juicy peduncles, nearly
always attacking those bearing flower buds. The small form supported
by this peduncle dries up just as those bored by the newly hatched
Boll Worm, and when dried enough to fall readily can not easily be
distinguished from young Boll Worm work. With a view of determin-
ing whether the beetle ever did original boring upon cotton bolls if left
to its choice, a number were placed on branches of cotton in a breeding
cage, so as to be kept under observation. The results of these studies
during confinement showed plainly that the beetle did its most injurious
work by puncturing peduncles bearing forms or puncturing the very
small bolls ; in either case they were always shed. It therefore appears
that if the beetle bores or eats into bolls at all, it certainly is an excep-
tional method of attack.

PLATYNOTA SENT AN A.

The larva of this Tortricid moth is a small, green, slender, hairy
worm, having a brown head, and is about half an inch long. It attacks
forms and squares much the same as the young Boll Worm does. After
the work is done and the worm has gone, its work can not be distin-
guished from young Boll- Worm ravages. These larvae continue their

30

habit of feeding on forms or young bolls until about half grown, when
they often migrate to the leaves, fold a portion of them together, and
feed under cover. Many, however, remain with the young bolls, and
reach maturity by feeding on them. They have been observed to bore
half grown bolls and destroy their contents.

CACCECIA ROSACE ANA.

This Tortrieid attacks the cotton in much the same way as the pre-
ceding species, and for that reason its depredations may- be mistaken
for traces of the Boll Worm. The worm differs from the preceding in
that the head, dorsal surface of the first segment, and the legs are black.

PRODENIA LLNEATELLA.

This fleshy worm was observed entering into nearly grown bolls and
feeding on their contents. Its ravages are exactly like those of a nearly
grown Boll Worm, and the two can not be distinguished.

nocttjid (undetermined).

A cutworm, lookiug much like Agrotis c-nigrum, was found in a large
breeding cage which had been placed over some cotton plants in the
field. When placed over the plants, none of the forms or bolls had been
injured, and no Boll Worms were found on the plants. Some time later
several large bolls had been bored, and this worm was the only one
which could be found in the cage. The evidence is therefore only cir-
cumstantial.

PLANT LICE.

(Aphis gossypii and Aphis sp.)

These small, greenish, mostly wingless, insects were especially abun
dant during the past season. Earlier in the season they are found prin-
cipally on the leaves and younger growing portions of the branches,
but frequently also on the young bolls between them and the involucre.
Later in the season they are found most abundantly in the last-named
localities, and in such great numbers on a single form or young boll
that the latter soon fall off as a result of their puucturiugs.

In many cases the fruit thus injured sitnply dries and adheres to the
branch. This fact often serves to distinguish it from Boll Worm work.
Even wdien this is not the case their work is readily distinguished in
that the form or square contains numerous small punctures.

THRIPID^E.

These small brownish insects during August were found in great
numbers in the forming blossoms of the cotton plant. The feeding of
these insects causes the form to drop soon after the blossom falls, if

31

not before. Such forms often present small black spots looking like
small borings, but which are so numerous that they need not be con-
fused with young Boil- Worm ravages. It must be noted that these
signs of mechanical injury are not to be attributed to the Thrips. The
shedding of these bolls is probably due to the fact that the work of the
Thrips on the essential organs prevents fertilization. This insures the
dropping of the fruit.

Mauy other species of the suborder Heteroptera probably puncture the
pericarp of the very young bolls or their peduncles, in (uther case caus-
ing the shedding of the fruit. Careful examination will show that the
injury is a puncture and should not be mistaken.

Neither of the first four species mentioned are numerous enough to
cause alarm or extended damage, and are only mentioned to show that
there is a certain small per cent of injury easily attributed to the Boll
Worm which does not justly belong to that species.

REMEDIES.
TOPPING OF COTTON AND ROTATION OF CROPS.

These have both been justly pronounced inefficient as a means of
fighting the Boll Worm (see Fourth Report IT. S. Entomological Com-
mission). It may be stated, however, that numerous interviews with
farmers verified the opinion that topping did no harm, and that if " you
could strike it right" it was an advantage. Experimentation is first
necessary to show that it is practical and profitable to practice topping
of cotton as an additional means of cultivation and the proper time to
do so determined. When this is done it will depend largely upon
whether that time falls within the period of greatest deposition by the
moths. If so, no doubt some additional benefit will be derived by the
destruction of the eggs deposited on the parts cut away in topping.
But since the moth has been found to have such a wide range of de-
position and the portion cut off in topping is so small in proportion to
the whole surface of the plant exposed and suitable for deposition, it is
not to be recommended to incur the expense of topping when nothing
more is to be accomplished than the destruction of the few eggs which
are likely to be found on the parts cut away. The rotation of crops
can be of no avail against the insect, since it feeds equally well upon the
corn or cowpeas, which are most likely to be rotated with the cotton.

FALL PLOWING.

This is to be urged for several reasons. It has been my experience,
that where the cells of the Boll Worm pupae are broken up and placed
in loose, moist earth, which is allowed to be moist continuously and
possibly to excess, that the pupae die in a majority of cases even with-
out freezing. It therefore appears that actual contact of the pupa

32

with the cold moist earth soouer or later may cause its death. From
this fact and the long continuous rainy season of the winter here it
seems probable that great numbers of the pupa? will be destroyed if
the soil be plowed late in December, so as to allow the loose earth to
become well drenched by the almost continuous January rains. Sub-
sequent rains will keep it quite wet, often perhaps, to excess. The
papal cells having been broken up, the wet earth directly affects the
wellbeing of the pupa?. Even though the exposure to moisture alone
should not prove entirely efficient, a light frost or the sudden cold
wave changes of the atmosphere would greatly aid in the work of
destruction. Certain it is, that a heavy frost occurring when the pupa?
are in such condition would destroy all thus exposed. For this reason
if the soil could be plowed in November so that the first black frost
of the winter season could be utilized in killing exposed pupae, great
benefit would certainly be derived.

CORN AS PKOTECTlOZs TO COTTON.

Cornfields planted in July or August were always found to be badly
infested with worms. Especially was this the case where the fields
were small and near cotton fields. At the same time corn fields no
larger but greater distances away from cotton fields were less infested
and the cotton more so. As has previously been noted, this is explained
by the fact that the moths feed mostly on cotton at night, but leave it
to deposit on corn if found suitable and near by. Even late spring
planting was found suitable for deposition in August, though maturing
rapidly and having nearly grown worms in the ears. Especially sug-
gestive .were the observations made in cotton fields where a poor stand
had been obtained and where corn had been planted in the u skips."
In all cases the moths deposited freely upon the corn, though it was
fast reaching maturity. Several of these cotton fields were carefully ex-
amined. The most extended search for worms revealed very few in-
deed, and the only possible conclusion to be arrived at was on the
whole that the damage to the cotton was not so great as in those fields
without the corn distributed through them. It is therefore evident that
by the proper management of the planting of corn the latter could be
made to answer as a great protection to the cotton against Boll Worm
ravages. This management must consist in arranging the crops on the
plantation so that green corn suitable for egg deposition shall be kept
near or in the cotton fields in range of the moths.

It must be stated, however, that the corn, which is intended to act as
a trap for the deposition of the eggs, and hence of the worms as soon
as hatched, must be planted with a view of being cut as fodder as soon
as a sufficient number of worms are found in the plants and before the
worms begin maturing. From this it follows that the corn, which is to
be allowed to mature and produce corn must be planted further away
from the cotton fields in order that it may become infested as little as

33

possible. The importance of this will be appreciated when it is remem-
bered that all the worms which mature on this corn and produce moths
will furnish an additional supply to infest cotton and at the same time
reduce the number trapped by the corn planted for that purpose. The
corn cut for fodder should be disposed of in such a way as to insure the
destruction of the worms found in the plants when cut. Each planting
of corn which is to act as a trap should be planted soon enough to be
in good condition for deposition as each brood of moths makes its appear-
ance. This would require about three plantings for the northern por-
tions of the cotton belt, -and probably four in the southern.

The three plantings should occur about the first days of June,' July,
and August, respectively, and be cut whenever the worms are nearing
maturity, to be sure to prevent their escape. By this method the
least possible number of worms reach maturity. This consequently
reduces their ravages on cotton later in the season to a minimum. This
minimum is the most that can be hoped for, no matter what may be the
remedial or preventive measures resorted to. This method would
furnish a great source of fodder, and would tend to diversify Southern
agriculture, a result greatly to be desired.

LIGHTS FOR ATTRACTING THE MOTHS.

Most of the experiments with lights for trapping the moths have
proven unsatisfactory so far as economic results are concerned, but
have been suggestive in that they have clearly marked out what will be
necessary to make the use of lights more efficient.

While at Shreveport, Louisiana, in company with Dr. A. R. Booth,
two kinds of patented lamps were taken into a large field of cotton to
test their relative values and also to determine if possible the ease
with which the moth could be attracted to lights. The lamps were
lighted at 6:30 p. m. Quite a number of moths were seen flying about
in the field as we passed through it, but up to half an hour after sun-
down but few insects of any kind were attracted. From that time
until 8 to 9 p. m. insects of nearly every description were captured, but
no Boll Worm moths. Sometimes a moth would be seen to approach
the lamps but was more interested in feeding and depositing, always
passing by or around without apparently noticing the lamps.

The lamps had been placed on pedestals high enough to bring them
above the level of the top of the cotton plants, hoping thereby to
attract moths from greater distances. In the mean time it was noted
that the moth seldom attained to such a height during her flights about
the cotton plants. Accordingly, on the evening of August 8, the same
lamps were taken into the field, but placed so as to meet the habits of
flight of the moth. This placed them at least on a level or a little
below the plane of the top of the cotton plants. It was found that
more moths approached and came nearer the lamp, and one was caught.
In most cases, judging from the flight and actions of the moths, the lamps
23024— ]NTo. 24 3

34

were simply met with in their regular flight through the field for the
evening, and that their course had not been materially influenced by
the lights. Despite this fact, it was evident that the probabilities of
trapping the moths at this height were increased. To vary the experi-
ment, the lamp was carried through the field at about the height just
mentioned, and one person walked along on each side some distance
from the lamps so as to disturb the moths in that vicinity. In this way
still more of the moths came near the lamps, and another one was
caught. This is impractical, however, since the expense of labor is
too great, at least until some means of making the lights more efficient
is effected.

One of the lamps was provided with a shield constructed so that it
would revolve with the wind, and thus prevent the lights from being
blown out. This is entirely wrong, since the moth usually flies with
the wind, in which case the light is of course shut oft from view en-
tirely. This defect renders the lamp entirely worthless for the end de-
sired. In our experiments this shield was held or made stationary, and
hence the defect did not enter or vary the significance of the results.

At Holly Springs, Mississippi, lamps were placed in cotton fields at
various times during August. Some consisted merely of beer bottles
filled with kerosene and a piece of unraveled cotton rope for a wick, and
others of more powerful lamps. In all cases the lamps were placed in
pans containing an inch or so of water, with a little oil on the surface,
the whole being placed on a supporting pole or pedestal. The beer-
bottle lamps burned satisfactorily, produced a good light, and attracted
insects of nearly every order and kind except the Boll Worm moth,
though these had been seen flying about in the field late in the after-
noon.

On the night of September 9 these lamps were again taken into a
cotton field having a small patch of corn near by ; also a large patch
of weeds, principally sneeze weed {Helenium tenuifolium). A lamp was
placed in each at about the level of the tops of the plants, except the
one in corn which was placed on about the same level as the ears on
the plants. By 5 o'clock the moths were seen flying plentifully and
ovipositing freely. The lamps were lighted at 6:15 o'clock. No insects
of any kind were attracted until 7 o'clock, when moths of all kinds be-
gan flying near the lamps. By 7:15 a Boll Worm moth had been
caught at the lamp in corn. At 7:25 at the lamp in cotton a Boll
Worm moth flew near the lamp, alighted on the pedestal and rested.
From here it flew up to a small boll in the direct light of the lamp, de-
posited an egg and flew oft'. From this time on many Ichneumonids and
other Hymenoptera, as also great numbers of Microlepidoptera were
caught. At 7:40 a Boll Worm moth was seen to fly through the flame of
the lamp but was not captured. In corn at 7:50 a moth flew about the
lamp and alighted on a blade of corn less than 2 feet away. For this
act of defiance it was introduced to the cyanide bottle. Not many

35

moths were seen between 8 and 9 o'clock but the lamps were left burn-
ing all night to determine what would be the nature of the catch by
the next morning. At this time the catchwas examined and the results
are tabulated below. For convenience the lamps in Helenium, corn, and
cotton will be numbered 1, 2, and 3, respectively.

Table VII.

Hymenoptera.

Ichneunjo-
nidae.

Miscella-
neous.

Total.

1

10

9

12

i

2

14
12
14

40

2 -.-

3..

Total for all .

Lepidoplera.

Lamp.Lgrotis. ^

Spilo-
soma.

Miscel-
laneous
Noctuids.

Georne-
tridte.

Pyralidae.

Pteropho- Miscel-
ridse. laneous.

Total

1

17
1

15
30
25

300

332
361
323

1,016

2.

3

2 i
1

27 300
96 950

1

20

T(

Diptera.

Lamp.

Tipulidae. !

Miscella-

40 !

Total for all

Total.

33
50

75

158

Coleoptera.

Lamp.

Epicauta.

Scara-
baeidse.

ei»«->»- ?sr

Carabidse.

Miscella-
neous.

Total.

1

2

2 25

8 ! 25
2 1 50

62
59
9

130

2

3

7
2

1

6

Tot

Hemiptera .

Lamp.

Homoptera. Heteroptera.

Total.

1

2

3

To

200 100
200 100
200 150

300
300
350

950

36

Table YII— Continued.
Neuroptera.

Lamp.

Chrysopa.

Total.

1

2

3

1
1

1
1

Total for all....

2

OrtJioptera.

Lamp. Mantis.

(Ecanthus.

Locus tidse.

Total.

i

1 1

2 1

3 1 2

2!

2

5
3
2

10

Total for all

The above results show that the direct benefit to cotton plants, so far
as known, is very small. At the same time such beneficial insects as
the Ichneumonids, predaceous beetles, and the Praying Mantis arede-
stroyed, and if trapping be systematically followed up the loss may in-
deed be considered greater than the gain.

These experiments led to the belief that the lights used were not bril-
liant enough for the intended purpose. Accordingly, an electric lamp
was rented. The lamp is provided with a round burner and the flame
produced is about 5 inches in circumference. When the chimney is
placed over the flame the lamp is said to give a light equal to 100 can-
dles. Experiments with this lamp in connection with the others were
continued in the cotton fields during September. On the evening of the
13th both kinds were placed out. The night proved to be a damp cold
one, and the dew fell early in the evening. As a result no Boll Worm
moths and but few insects of any kind were trapped. Of this small
number the parasitic Hymenoptera and predaceous beetles were greatly
in the majority. The insects caught were such as are easily attracted
to lights, and were quite equally distributed between the three lamps
(two beer-bottle ones and the other the electric lamp). A rainy season
began at this time and coutinued so that no further experiments with
lamps could be made until October 4. At this time the electric lamp
was placed in a cotton field to determine what would be the nature of
the catch. Heliothids had been seen flying about in the evening while
making some other observations, but none were captured by the lamp
at night. The other insects trapped were about the same in kind as
those already tabulated for September 9, only that the quantity cap-
tured by the electric latnp was about equal to that of all three of the
others, and it had only been left burning until 10 o'clock.

October 20 the lamps were again placed out, this time one in a patch
of cowpeas and the other some distance away in a small patch of corn
which had been planted, in July. During the day Boll Worm moths

37

were seen hiding behind the sheaths of the corn blades, while a number
were also seen flying about the cowpeas. The one in the cowpeas was a
beer- bottle lamp and placed about the height of the plants. The one in
the corn was the electric lamp, and was placed about the height of the
ears of corn. They were lighted at 6 p. m. At this time a few of the fe-
males were depositing on corn. At 7 o'clock some were seen to fly by
the lamp but were not trapped. The parasitic Ilymenoptera and smaller
Lepidoptera had been caught in great abundance at both lamps. He-
turning at 10 o'clock to further examine the catch of the lamps, they
were found to have been stolen and no further notes could be taken.
The moths at this time were not very abundant and doubtless were but
a portion of the last brood of the season. The moths seen flying by the
electric lamp were near enough to have been stopped had the lamp been
provided with long projecting wings and a larger pan to receive the
moths as they fell, ^"o Heliothids were observed at the beer-bottle
lamp.

These light experiments, as wili be seen from the record, were begun
at the time that the midsummer brood was issuing abundantly, and
hence also during the period of greatest egg deposition a little later.
During this period, as has been stated, the provoking observation was
made of seeing the female near the lights, deposit an egg in plain view,
fly away and continue her work. It is evident, therefore, that the
female is not easily diverted from the work of depositing eggs by the
ordinary lights used. Later, when the experiments show that a few
moths were trapped, it is also true that the period of greatest deposi-
tion had passed, and that, though dissections showed that a few eggs
still remained together with a number of potential ova, the females had
passed their prime. As bearing on this the following may be drawn
from the observations of Dr. Booth : The insect contents of a globe of a
2,000candle power arc light were examined continuously from Septem-
ber 3 to 13, inclusive. An average of 40 Heliothids were found for each
night. Of these 1 in 6 or 8 were females, containing on an average
from 30 to 40 eggs in the oviducts. The lamp tender reported that after
September 26 no more moths were caught.

The fact that the moth was frequently seen to fly near the light,
often as near as 2 or 3 feet, suggests that the lamps to be efficient not
only must be brilliant, but must also have some wide and extensive
wings extending from it in. such a way as not to throw a shadow and
to arrest, temporarily at least, the flight of the moth passing near by.
If now the large pan and the lamp be provided with an additional in-
ducement in the way of some strong smelling sweets, the moth thus
arrested in its flight and its attention diverted from its evening work,
if not falling into the pan, may be attracted a second time and be cap-
tured.

Unless it is found that the earlier broods are more easily attracted to
lights it is questionable whether the inefficient lights so commonly used

38

by planters are to be at all recommended if nothing more is to be accom-
plished than the trapping of the Boll Worm moth, and for the following
reasons : (1) Bat a small per cent are caught ; (2) of these the great ma-
jority are males ; (3) while some females are caught before having
deposited many eggs, the greater per cent have passed their prime;
(4) beneficial insects being more easily trapped are destroyed in too
great numbers in proportion to the benefit derived from the destruction
of obnoxious insects to warrant such inefficient warfare. These may
all be included in the one general reason that the lights are only strong
enough to readily attract beneficial insects but are powerless to attract
the obnoxious insects desired until its most important work (deposition
of eggs) has almost been completed.

POISONED SWEETS.

No field experiments were made with poisoned sweets, but a number
were made with moths in the laboratory. The mixture was composed
of 1 part of white arsenic dissolved in 20 parts boiling water; 4 parts
of this solution were added to 3 parts of ordinary table sirup. The
mixture was placed in a watch glass under a bell jar or sprayed upon
cotton branches in a breeding cage. When the moths were placed in
they always soon found the sweets and sipped of them. The result of all
the experiments showed that the moths readily partook of the sweetened
liquids. Those having sipped of the poisoned solutions died, on an
average, within 30 minutes ; the shortest time being 15, the longest 45
minutes. Experiments were also made upon a few other insects, mostly
such as were considered beneficial. They were placed in the cages just
as the moths had been and were found to partake of the sweets quite as
readily and died as certainly. Thus in field experiments doubtless
many beneficial insects will also be destroyed by the extensive use of the
poisoned sweets.

The moths kept in cages for experimental purposes were fed by spray-
ing unpoisoned solutions of the sweets upon the cotton branches. The
moths fed readily and lived usually from 5 to 8 days.

The poisoned sweets used in the experiments in the laboratory con-
tained no liquids which could liberate a strong odor such as is neces-
sary in field experiments. These may be added in the form of beer or
vinegar or perhaps any other liquid having similar properties.

It was demonstrated by the experiments in the laboratory that newly
issued and old moths were alike easily induced to feed on the poisoned
drops of sweets sprayed on the branches in the cages. It may there-
fore follow that if these poisoned liquids can be properly applied to the
plants upon which the moths feed freely both at night and during the
day, that females may be readily attracted to feed, and hence killed,
during their entire period of deposition. It therefore appears probable
that if some practical means is employed to apply these poisoned sweets
properly and abundantly as food for attracting the moths that such

39

method of warfare against the adults will prove more efficient than any
other alternative yet resorted to against them. It is evident, however,
that to be most efficient, the poisoned sweets must be applied from the
time when the moths begin feeding freely, and in such a way that they
may meet with them readily in their flights about their food plants.
The first will be accomplished if applied as early as 4 o'clock in the
afternoon, in which case the poisoned liquids would also be exposed to
their visits during the evening and night. The second can be attained
by spraying the poisoned liquid upon the food plants. For those moths
feeding during the day this must be applied principally to cow peas, for
those feeding at night upon cotton.

The practicability of this method is yet somewhat questionable since
probably one application of the poisoned liquids would be efficient only
for a few days. It may further be questioned in that, as has been noted,
the moths of any given brood issue quite scatteringly. At the same
time it may be that applications of the poisoned sweets at intervals of 3
or 4 days will prove to be as practical as arranging for, and attending to,
light trapping properly. My own efforts to experiment fully along this
line were rather frustrated by rainy weather during September. This
made experiments difficult and more or less indecisive.

The possible utility of combining poisoned sweets with lights has al-
ready been noted. The fact that females are readily attracted by sweets
before many eggs have been deposited by her may become a sufficient
additional inducement to entice those flying so near the lamps to linger
a few moments longer and probably result in her capture. With these
probabilities in mind it is to be hoped that the approaching season may
be more propitious for experimental work and the Boll Worm more
abundant.

PYRETHRUM.

Experiments with the dry powder. — The lirst of the following series of
experiments with pyrethrum were made upon infested corn. The patch
of corn was about two rods square and located near the center of the town
(Holly Springs, Mississippi). It was a second planting and was only
knee high at this time, August 19. The middle rows of the patch were
selected and one row for each experiment taken. The powder was
dusted from above down into the bud of the corn by means of a small
cheese-cloth sack, double thickness.

Experiment 1.

August 19, 2:30 p. in. Mixture, equal parts lime dust and pyrethrum. The plants
iu the row by actual count contained 43 worms of various sizes.

Result. — Soon after dusting a few acted uneasily, began to crawl, and finally drop-
ping to the ground, hid in the loose earth. This note applies more or less to all the
experiments made with the powders. August 20, 10 a. m., 17 worms alive and feed-
ing, 10 dead, 16 not present. In percentages this is 39.5, 23.3, 37.2, respectively. The
living worms were mostly nearly mature ones which had penetrated far into the
center of the bud and may not all have come in contact with the powder. The dead

40

ones were mostly composed of half or two-thirds grown worms. These notes again
apply equally well to nearly all the other experiments with the powder. The follow-
ing experiments will therefore be given more concisely.

Experiment 2.

August l'Jj 3 p. m. Mixture. "2 parts lime dust, 1 part pyrethrum. Xumber of
worms in plants. 43.

Result. — August 30. 11:30 a, m . '20 living. 10 dead. 13 not present, or 46.5. "23.3. 30.2
per cent, respectively.

Experiment 3.

August 19. 3:30 p. m. Mixture pyrethrum fall strength. Xumber of worms 54.
Result. — August 20. 12 in.. 25 living. 8 dead. 21 not present, or 40.3. 14.8, 35.9 per
cent, respectively.

Experiment 4.

August 19, 4 p. m. Mixture equal parts lime and pyrethrum. Worms not counted.
Result. — August 20, 2 p. ni., 19 living. 12 dead, or 61.3 and 35.7 per cent, respec-
tively.

Experiment 5.

August 19, 4:30 p. m. Mixture 2 parts lime, 1 part pyrethrum. Worms not
counted.

Result. — August 20, 2:30 p. m., 26 living, 8 dead, or 76. 5 and 23.5 per cent, respec-
tively.

Experiment 6.

August 19, 5 p m. Full strength pyrethrum. Worm? not counted and only a part
of the row dusted.

Result. — August 20, 3 p, in.. 5 living. 6 dead, or 45.5 and 54.5 per cent, respectively.

The dead worms of experiments 1 to 6 were kept, for raising any possible parasites,
until September 3. when they were found to be perfectly dry and were thrown away.
No parasites had issued from them.

As checks on experiments 1 to 6 it may be stated that in examining
the rows carefully to count the actual number of worms in the plants
no dead worms were found. Numerous other observations upon corn
of a similar age, and which had not been dusted, verified the one made
while counting the worms.

The first 6 experiments and their results may be tabulated for con-
venience as follows :

Table VIII,

Experi-
ment.

-:ancc used.

Living. Dead.

,k,mr Probable ,
Absent- benefit.

1

2

3

4

5

6

Lime and pyrethrum, equal parts

L;imr 2 parts, pyrethrum 1 part

Pyrethrum. full strength

Same as experiment 1

39.5

46.5 23.3

46.3 14.5

61.3

76. 5 23. 5

45.5 54.5

37. 2 60. 5
30. 2 53. 5

38. 9 53. 7
3S.7

23.5

Same as experiment 3

Averages

54.5

52.6 29.7 17.7 47.4

41

Experiment 7,

August 20, 10:45 a. m. Dusted the 17 living uearly-grown worms of experiment 1
with full strength pyrethrum, and placed them in a closed mailing box without food.

Record — 11a.m., some are beginning to be restless; 12:30 p. m., all are quiet;
1:45 p. m., one almost dead, others jump when touched ; 4:20, p. m., one dead, others
as before.

August 21, 11:30 a. m. Somehave become quite active again, and as a result 4 were
bitten, 3 are dead, 6 others alive and active, while the rest were probably eaten.
The active ones were placed in a breeding cage, and provided with branches of cot-
ton having leaves and bolls. Some died later and a few matured.

Experiment 8.

August 30, 4:30 p. m. Eighteen of the living worms from experiments 4, 5, and 6 were
well dusted with lime. Nine were placed in each of two closed mailing boxes. This
^experiment was for the purpose of having a check on any possible effect of the air-
slacked lime on the worms.

August 21, 11:45 a. m. In one box two were injured and died. The rest alive and
active. In the other box all are well and active. Both lots were then placed on
branches of cotton in breeding cages. All began feeding ; a few died later, but most
of them matured.

Experiment 9.

September 11. Boll Worm in a boll with posterior segments protruding. Dusted
this with full strength pyrethrum at 9:15 a. m. ; no immediate effect was noticeable;
11:30 a. m. ; worm has turned round ; head almost protruding ; not feeding ; 3 a. in.,
bas turned round now to position same as when dusted, and is feeding. With a
pair of forceps placed some pyrethrum powder on the body of the worm in the hole.
Did not effect the worm noticeably for five to ten minutes.

September 12, 8:30 a. m. Has left the boll and is in the upper corner of the cage
still alive and active.

September 13, 8 a. m. Has returned to the boll it had left, and is feeding. Dusted
the protruding portion with pyrethrum. In the afternoon the worm was found
crawling about in the cage, but appeared to be full grown and searching for a place
to pupate. Placed back on the branch again.

September 14. Worm still on the boll and is active, but not feeding: Particles of
pyrethrum are found adhering to the body. The anus is swollen and inflamed, pro-
ducing a watery exudate. This may be due to the effect of the pyrethrum, for this is
the portion of the body which was usually protruding from the hole in the boll and
received most of the dusting. The worm was placed in a partially opened form.

September 15. Has eaten the form almost entirely. '

September 16. Crawling about in cage. Placed on a branch.

September 17. Crawling about in cage. Has shortened some and is preparing to
pupate. Placed in a tin can with earth to allow it to do so. Pupated September 21.
Pupa still alive October 3, when it was placed in alcohol as a specimen.

In one instance, when a worm had been experimented with in a similar way to Ex-
periment 9, the worm went down to corner of the cage (which was a newly made one),'
webbed together the loose sawdust found there, and pupated in the cell thus formed.

Experiment 10.

September 20, 5:30 p. m. Marked five bolls in which Boll Worms were feeding and
a, portion of the body protruding. At 5:40 dusted profusely with pyrethrum, full
strength.

42

6 p. rn. The first worm ceased feeding and left ; can not be found. The second was
entirely in boll, but came out and is twisting about uneasily; finally it fell to the
ground in convulsions, tumbling over on its back as if to scrub off the powder. The
third ceased feeding. The fourth was a young worm and is not to be found. The fifth
still feeding.

6:15. The second is still in convulsions and can not crawl well; it is one about
two-thirds grown. The third has fallen to the ground, but is crawling into loose
earth. The fifth continues feeding. The last two are nearly grown worms, and
doubtless will take some time to become badly paralyzed. At 6:30, while making;
other observations, it became dark, and could not find the worms again.

A Dumber of individual experiments with, grown worms, both in the
laboratory and open air, were made. The worms were well dusted with,
full-strength pyrethrum and were allowed their pleasure as to their
abode afterwards. They always crawled into the loose earth as soon
as possible, and as long as they were observed showed signs of recovery ..
Other experiments similar to Experiment 9 are omitted because their
results were practically the same in all cases.

DECOCTIONS OF PYRETHRUM.

Seven pints of rain water were brought to boiling in an open pan j
12 grains of pyrethrum were then stirred in and boiled for 15 minutes.
The whole was then strained so as to get out most of the powder. This
decoction was made on the afternoon of September 19, but owing to a
threatening rain was kept in sealed Mason jars until the next day, when
the decoction was sprayed on bolls containing Boll Worms. The fol-
lowing strengths were used: Full, two-thirds, and half.

Experiment 15.

September 20. The bolls with worms in had all been found and marked during
the forenoon. The day was warm and sunshiny. In the afternoon it was found that
one of the worms had changed bolls since morning observation. At 2:40 p. m. full
strength of the decoction was sprayed on each of six bolls containing Boll Worms.
Four of the six were not in bolls, but between them and their involucres. The
greater portion of the plants surrounding the boll was also sprayed.

2:50. No uneasiness manifested by any.

3:15. Five as before ; one half-grown worm has moved and can not be found.

3:45. No change ; has not affected the worms yet.

4:10. None feeding ; no change.

5:00. One feeding; others as before.

5:50. None feeding.

September 21. Two worms still in place ; one feeding, the other just molted ; two-
others finished the bolls in which they were found and have disappeared ; the other-
one is in boll in laboratory.

September 23. All have gone but one; this one went to another boll, fed, and has
just molted ; an hour later it was found devouring molted skin.

Experiment 16.

September 20. Equal parts decoction and rain water. At 3 o'clock sprayed five
bolls, each containing a Boll Worm. Four were not feeding, but resting between boll
and involucre; the other was in boll feeding. None had changed position since
morning observation.

43

3:25. One seems to be a little uneasy ; others manifest no anxiety.

3:42. The one in boll has turned around and is poking its head out of the bolls j
another has moved and gone into a blossom; the others same as before.

4:10. One is feeding ; all the others quiet and not feeding.

5:00. One feeding ; others no change.

5:50. None feeding.

September 21. Four still in place, all feeding; the other has left the plant.

September 23. One still in place, but has about destroyed its boll. The others-
have done so and are gone.

The record of the experiment with two parts decoction and one part rain water is
omitted because of the similarity of results to those of Experiment 16.

Checks on Experiments 15 and 16.

September 20. At first three, but later five more worms were marked as checks.

2:30. One entirely in boll, feeding ; another, nearly mature, resting and not feeding;
a third, very young one, is feeding.

5:00. Checks all in place feeding.

5:50. All but very young one feeding. Made a search for more worms ; found fivey
all of which are in bolls feeding.

September 21. One still in boll, but not feeding.

The small worm has bored through a form and is feeding. The. nearly matured
worm has destroyed its boll and has gone away.

The result of all the experiments with pyrethrum is, on the whole,,
negative. Before treating more fully of the results of the experiments
it must be stated that the com plants, cotton bolls, and Boll Worms were
more thickly and thoroughly dusted or sprayed than it would have
been possible to do by dry method of application which would be inex-
pensive enough to be practical. There is a special difficulty in the case
of cotton. At the time when the powder would be most efficient, that
is, when the worms are yet less than half grown, they are found princi-
pally at work in forming blossoms and very young bolls. In these the
involucre so completely and effectually inclose the portions in which
the worms are at work that it is practically impossible to reach them*
It is well known that the young form or boll is sensitive to excessive
rains, and their involucres, it seems, are to a great extent a provision of
nature to protect the tender young bolls from such injury. To what-
ever extent this may b e the case, it is certain that their involucres make
it exceedingly difficult to reach the forms and bolls beneath them by
any of the methods of spraying, and therefore also to all decoctions or
solutions of whatever kind.

Upon corn before it has tasseled the powder may be used with greater
success, as will be seen from a study of Experiments 1 to 6. From these
we find that a certain benefit of about 30 per cent, is obtained, with a
possible benefit of about 47 per cent. This last is too high, however, as
some of the worms which leave do so only temporarily and to recover,
after which they return. We also find that the young worms are much
more susceptible, or at any rate less able to resist the effect of the pyre-
thrum. Consequently of the worms killed, the great majority were half
or less than half grown. From the behavior of the grown or nearly

44

mature worms in all the experiments, it is evident that they strongly
resist the effects of the powder, and if ample opportunity is given to
escape to the ground or loose earth, may often entirely overcome its
influence and recover. Whether on corn or cotton, it must be admitted
that the protection is only temporary. This is shown by the fact that
in some of the experiments undisturbed individuals entered bolls with
impunity soon after dusting and after the first worm had retreated, or
even the same worm going back and feeding upon the boll from which
it had been driven, presumably, by the pyrethrum.

As has been noted, there is a certain benefit derived from the appli-
cation of the powder to young corn before tasseling. It is just to con-
sider that the pyrethrum was at a disadvantage, in that it was not ap-
plied early enough to catch the worms before they had become so
nearly grown or had entered far into the bud. If it had been applied
earlier a much greater per cent of the worms then present would
doubtless have been destroyed. Such being the case, the use of pyre-
thrum may prove to be a decided advantage in cooperation with the plan
of planting corn as traps for egg deposition, and hence the worms
when these are hatched. This can be done by thoroughly applying
pyrethrum of about one-half or third dilution with lime to the corn
plants at a time when the worms are found to be about half grown.
By doing this the time of cutting out corn to destroy the worms it con-
tains will be delayed for a time longer, and hence also be exposed to the
depositions of the moth for a greater period. Experiments in this
direction will be taken up extensively this season.

The powder being thus limited in its efficacy, especially on cotton, it
is not surprising that decoctions of the powder prove to be even less
effective. As will be noted from the experiments with the decoctions
when compared with the record of the checks upon the same, little
more was accomplished than to temporarily arrest the feeding of the
worms. It is true some of the worms changed bolls during the after-
noon, and others which were in bolls came out, but it must also be
noted that the same action was taken by other worms which were un-
der observation and which had not been sprayed. There is some ques-
tion, therefore, that the decoction was directly accountable for the
action of the worms upon which it was sprayed.

This doubt is further increased from the fact that it was often noted
in worms which had been marked for observation that they very fre-
quently changed bolls or even plants during midday or afternoon.

OTHER VEGETABLE INSECTICIDES.

The work upon vegetable insecticides was assigned almost entirely to
Prof. Jerome McNeill, Fayetteville, Arkansas. He has been as unfor-
tunate as myself in being unable to obtain plenty of Boll Worms to ex-
periment with. Progress was further impeded by unpropitious weather.
For this reason the greater portion of the time was occupied in collecting

45

such roots, plauts, flowers, and fruits as might, upon experimentation,
prove to have insecticidal properties. This was undertaken with a view
of discovering if possible some product easily grown in the infested
regions through the cultivation of which it might be possible to pro-
vide for an insecticide which would be cheap and accessible to all.
From these various collections Professor McNeill has made numerous
extracts, emulsions, and decoctions, some of which he informs me are
quite promising, and which are on hand to experiment with when oppor-
tunity offers. As this part of the work has, therefore, not been com-
pleted for the reasons stated, I shall at present give only a summary of
Professor McNeill's letters and report of progress during the past sea-
son.

(1) Alcoholic extracts and decoctions have thus far been, on the
whole, unsatisfactory.

(2) Extracts and extract emulsions of the various vegetables or parts
thereof seem to be promising. Of these kerosene, kerosene ether, gas-
oline and benzine extracts, and emulsions of pyrethrum are perhaps
most important.

(3) Of the plants experimented with, Lobelia syphilitica, L. cardi-
nalis, probably L. inflata, and Ariscema triphyllum are among the more
important as giving promise of good results. They have been shown
to possess insecticide properties, but to what extent and how best
utilized remains an open question.

(4) An exceedingly dilute solution of potassium cyanide is an effi-
cient insecticide, but its effects on the cotton plants has not yet been
determined.

METEOROLOGICAL CONSIDERATIONS.

Of these rain, humidity, and temperature are the principal phenom-
ena to consider. What relation these may have to the various stages
of the transformations of Heliotliis, the following tabulated data may
serve to indicate. The averages of humidity and temperature are given
for the entire period covered by each example :

Table IX.

EGG.

When deposited.

When hatched.

Dura-
tion.

Rain,
number
of days.

Temperature.

Humidity.

Max.

Min.

Max.

Min.

Night, Aug. 5-6

Night, Aug. 8-9

Nio-ht, Au°\ 14-16

Aug. 10, 10 a. m

H

2

5
3i

3
0
1
0
5
5
4

0

93

94
P4
84
79

78

77

c

74
73
72
72
67
64
67

o

88
87

{* )

o
56

63

Night, Aug. 15-16

Night, Sept. 19-20

Night, Sept. 20-21

Night, Sept. 21-22

Aug. 1 8, 9 a. m

\f

(*)

* Data for humidity at Holly Springs, Miss , could not be obtained. The first two are from Shreve-
port, La., where eggs were under observation.

46

Table IX— Continued,

LAEYA.

"When hatched.

When matured.

Dura-
tion.

Rain, Temperature^ Humidity.

number ; —

of days. Max. I Min. Mas. Min.

August 9

August 10 —
August 18

August 18

September 25.

August 23 i 15

August 25 (died) 16

September?.., 21

September 12

October 12 (two molts)

PUPA.

When pupated.

When issued.

Dura-
tion.

Eain, [Temperature.; Humidity,
number

August 9 —
August 23 . .
August 28 . .
August 28 ..
August 28 . .
August 28 . .
August 31 . .
September 4
September 7

August 20 : 11

September 2 '■ 10

September 12 ' 15

September 13 16

September 16 1 19

September 17 i 20

September 27 , 27

September 20 16

September 30 22

These data may be studied in the order given.

Egg.— For the first two lots the temperature is the same with only a
slight difference in humidity, but during the period of the first lot rain
fell for a part of the time on 3 successive days ; during the second
none fell at all. Under these conditions the duration of the first lot
was 1J days longer. The period covered by the next two lots of Au-
gust 15 and August -16, furnish about the same conditions with the same
results. The next three lots were deposited much later in the season,
had lower temperature and excessive rains, 6.37 inches having fallen
from September 22 to 25 inclusive. The duration of the egg state as a
result was much prolonged. There is no check on this lot, however,
since no eggs under direct observation were hatched during that sea-
son with the same low temperature but without the excessive rains.
From general observations, however, there is no question but that low
temperatures also prolong the duration of the egg state, the same as
the rains seem to have done in each of the two first lots mentioned.

Larva. — The first two larvae were reared under almost exactly similar
conditions and, as will be seen, matured almost at the same time. As
compared with those that follow the duration is of interest, as there
was but little rain and a high temperature. The next two were worms
hatched from the same lot of eggs, and, as is seen from the table, were
reared under exactly similar conditions. Despite this, the difference in
time of maturing is 5 days. This can only be accounted for by the pe-
culiarities of the species, such as have been previously discussed. The
difference, as compared with the two preceding, was principally due to
the much lower temperature. If, with the abundant rain during that

47

period, the temperature had been maintained as high as in the first, the
worms would have matured more rapidly. This is verified by noticing
the retardation of growth of the last worm recorded in the table. This
worm had only molted twice after 18 days. Thetemperature during this
period was 18° lower than that of the first two and 9° lower than that
of the second two. General observations established this fact concern-
ing the feeding of the worms, viz : that a moderate amount of rain
with high temperature was least suited to their most vigorous feeding
and growth, and consequently their earliest maturity. The same
amount of rain, however, with a much lower temperature, is as much
a disadvantage, and increases the retarding effect, which the lower tem-
perature itself would have had. But again, high, dry temperatures are
avoided by the worms, which during that time feed less vigorously,
and thereby prolong their larval existence some.

Light frosts began (both in Mississippi and Louisiana) as early as
October 27, and were more or less continuous from that time on. At
Holly Springs, Mississippi, a killing frost occurred October 31, which
froze and entirely blackened the cotton plants. At Shreveport, Louisi-
ana, however, the cotton was not entirely frozen and blackened until
about December 4. As has been previously noted, worms of nearly all
stages were found at both localities a short time previous to the killing
frosts, by which latter the younger ones were quite certainly killed.

Pupa. — For the first of the pupse recorded it is found that a mod-
erate amount of rain with high temperature shortens the duration of
the pupal state. From the remaining ones it is found that with but
little variation in the low temperature, which alone would have pro-
longed the duration, the excessive rains greatly added to the delay.

In genera], then, it may be stated that the duration of the various
stages of Heliothis are shortest under high temperatures with moder-
ate rainfall ; longer, except in egg and pupa, when a high, dry tempera-
ture is maintained; longer still with much lower temperature; and yet
again longer with lower temperature and excessive rains.

Some atmospheric conditions also noticeably influence the behavior
of the moths. The hot weather, dry, or somewhat rainy, seems to have
but little diverting effect on the habits of the moths. When the tem-
perature is much lower, and is accompanied with much rain, the moth,
adapts itself to the condition of things. The excessive rains last sea-
son continued late in the evening and into the night. This of course
covers the period of feeding and deposition. This seemed to have the
effect of inducing the moths to fly and feed more freely during the mid-
dle of the afternoon, when it was clear and warmer. When the rainy
spell began to be a protracted one, the females were frequently seen at
3 o'clock during the warm sunshiny afternoons busily engaged in de-
positing their eggs. The instances in which deposition was observed
in daytime were confined mostly to this period, though some were ob-
served under normal conditions. From this it follows that to a certain

48

extent, at least, the imagos adapt themselves to unfavorable conditions,
and that their period of egg deposition, on the whole, is not much in-
fluenced by such conditions. Their progeny, however, as has already
been noted, suffers materially.

As bearing on the abundance (or rather scarcity) of the Boll Worm
the past year, I quote from the report of Professor Fulton for 1890 as
follows: "The most important irregularities of the year were the un-
usually high temperature in January and February, with a marked
deficiency of temperature in March." During the period of high tem-
perature in January, and especially February, it may be that many
of the moths issued. If so, the cold period in March quite likely
killed many of those which had issued. In the Eed Eiver section of
Louisiana the Eed Eiver overflowed badly in spring, and planting of
both corn and cotton was delayed until late in May and some in June.
This necessarily delayed finding suitable host plants for the moths
which had issued during April and May to deposit on, and doubtless a
large per cent of their progeny failed to survive. In some localities
also corn and cotton had been planted and was large enough for the
moths which had issued to deposit upon when the river overflowed. As
a result the corn and cotton both were drowned, or at any rate stunted
so that it was all plowed under and planted a second time. By this
process doubtless many of the first brood of worms were destroyed.
From these reasons the second brood and consequently all subsequent
broods were in all probability greatly reduced.

From all the information gathered through observers of the U. S.
Signal Service it is certain that the boll-worm depredations are much
more extensive in the southern portion of the cotton belt. There is,
therefore, no question but that the future work on the Boll Worm
should be carried on principally in that region.

INSECT DISEASES.

The work upon insect diseases has formed an important part of the
investigation. At the present time, however, it would be unwarranted
and hazardous to enter largely into a report upon the work done and in
^contemplation, or to draw conclusions. I shall therefore give but little
more than a synopsis of the present condition of the work, and will re-
serve acknowledgments to those who have contributed in any way for a
more detailed report in the future.

The first thing to be done in preparing for such work was to equip
and arrange for a bacteriological laboratory. Some time was spent at
Shreveport, Louisiana, in cooperation with Dr. Booth (who assumed
charge of the work for the season) towards accomplishing this end.
Hot-air and steam sterilizers were designed and a good workman soon
had them in condition for use. The other supplies immediately neces-
sary were ordered. These have been added to as the progress of the

49

work demanded, until now quite a complete laboratory has been fitted
up, sufficient to carry on to a finish ail the work and experimentation
which it will be possible to execute.

The diseased insects and worms from which the cultures On hand
have been made were obtained from various sources from entomological
workers throughout the country.

Extensive and conclusive experiments with theinsect diseases on hand
were not made for the same reasons stated by Professor McNeill. The
status of this portion of the work is, therefore, much the same as the
latter, viz, ready for extensive and thorough work during the approach-
ing season. The few observations made are encouraging, but do not
warrant any definite and positive statements at this time.

It seems highly probable that the Boll Worm is readily susceptible to
the cabbage- worm disease. Dr. Booth in one instance fed Boll Worms
upon diseased cabbage worms, which Boll Worms later died. Cul-
tures were obtained from these dead Boll Worms. Mounts from the
cultures were made later and studied with a microscope. Micrococci
were present in great abundance. At Holly Springs, Mississippi, some
Boll Worms were accidentally placed in a breeding cage in which dead
cabbage worms had been temporarily placed. A number of these Boll
Worms died at various intervals. The dead worms were sent to Dr.
Booth, who made cultures from their dead and decaying bodies. Ex-
amination of mounts made from these cultures again showed micro-
cocci in abundance. The above evidence is not direct and positive ; is
merely indicative, and at best unscientific. It consists simply of obser-
vations noted during the progress of the work, and simply indicates that
scientific experiments may prove successful.

Though no experiments could be made upon the Boll Worms with
other insect diseases, the interesting and important discovery was made
at Holly Springs, Mississippi, that the Boll Worm itself is subject to a
disease. The disease is not confined to the larval stage, but has been
obtained from all the stages of the species. Two females issued on the
night of September 14. On the second day, it was noticed that the
moths were rather sluggish and that the abdomen was greatly dis-
tended. By the next day the females were absolutely helpless, and the
abdomen so decomposed that it barely held together while pinning the
moth. The last signs of life of the moth consist of peculiar alternate
openings and closings, contracting and expanding of the anus and gen-
ital organs. At the time it did not occur to me that it was a disease of
the species, and it was only the peculiar manner of the dying of the
moth which had attracted my attention. Hence it was that the moths
were simply pinned and placed in insect boxes. This was done Septem-
ber 17. November 28, the abdomen of the moth was accidentally
broken off and the internal parts were found to be partially liquid.
Erom this partially liquid portion tubes of beef broth were inoculated,
as also from a whitish, waxy, gelatinous substance in the extreme poste-
23024— No. 24 4

50

rior end of the abdomen. A culture was obtained from the waxy portion,,
and the microbe is different from any of the others studied. It is pos-
sible that the cultures obtained from the moths after so long a time are
non-pathogenic microbes, instead of the one which produced the disease
of the imago. No positive statements will therefore be given until ex-
periments have been made. Cultures from the egg, larvae and pupae are
in stock, and, so far as examined, are all exactly alike. This disease
can not have been mistaken for any other, since it was noted before any
of the others were on hand. Thus, having probably found the Boll
Worm subject to a disease perhaps peculiar to itself, it remains to be
seen whether it is contagious and easily disseminated for infection.

In addition to this, a disease of each of the two larvae whose ravages
are easily mistaken for those of the Boll Worm was also discovered at
Holly Springs, Mississippi. The two species are Prodenia UneateUa and
the undetermined Noctuid spoken of. In fact the disease of each was
so prevalent, that but few of the worms were found, and of those found
all but one, which had been placed in alcohol, died of the disease. No
great apprehensions need therefore be had concerning these two species.

Diseases of Agrotis messoria, Nephelodes minians, as also of the large
tomato worm, are at hand in the form of cultures.

Cultures from all these sources were begun in August, 1890. By the
regular methods for such work pure cultures have now been obtained
and are transferred from time to time to fresh media, in order to con-
tinue the healthy growing germs through the winter and in good con-
dition for the approaching season's experimental work. In this way a
vast number of cultures in fine condition are on hand, and it is to be
hoped that abundant opportunity may be offered this season to execute
extensive and thorough experiments.