Historic, archived document

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

U. s. ‘DEPARTMENT OF AGRICULTURE,

BUREAU OF. ENTOMOL0GY—BULLETIN No. 100.
Bi 0. HOWARD, Entomologist and Chick of DH

INSECT ENEMIES OF THE
COTTON BOLL WEEVIL.

Aas

W. DWIGHT PIERCD, Ge ne eS
Agent and ABE es :
eaeaey ASSISTED. BY VS Nien nee
R.A. CUSHMAN anp C. E. HOOD,
Agents and Experts, mM
UNDER THE DIRECTION OF
W. D, HUNTER,

Agent and Expert, ie ein
ain Charge of Southern Field Eee Insect 1 Investigations : poe ee

IssuED APRIL: s LOI

i

a ee
A

“WASHINGTON:
GOVERNMENT PRINTING OFFIOE.
1912.

U.S. DEPARTMENT OF AGRICULTURE,

BUREAU OF ENTOMOLOGY—BULLETIN No. 100.
L. O. HOWARD, Entomologist and Chief of Bureau. ‘

THE INSECT ENEMIES OF THE
COTTON BOLL WHEVIL.

BY

W. DWIGHT PIERCE,
Agent and Expert,

ASSISTED BY
R. A. CUSHMAN anp C. E. HOOD,
Agents and Experts,
UNDER THE DIRECTION OF

W. D. HUNTER,

Agent and Expert,
In Charge of Southern Field Crop Insect Investigations.

Issurep APRIL 3, 1912.

i

Ra
il

re. Hie
oon li DRO
Ie ie
PL Lt ni

o = ny

= <
area

Sen
Thy liss S

WASHINGTON:

GOVERNMENT PRINTING OFFIOE.
1912,

BUREAU OF ENTOMOLOGY.

L. O. Howarp, Entomologist and Chief of Bureau.
C. L. Martatt, Entomologist and Acting Chief in Absence of Chief.
R. 8. Currton, Executive Assistant.
W. F. Tastet, Chief Clerk.

F. H. CHITTENDEN, 77 charge of truck crop and stored product insect investigations.
A. D. Hopxtns, in charge of forest insect investigations.

W. D. Hunter, in charge of southern field crop insect investigations.

F. M. WEBSTER, 2n charge of cereal and forage insect investigations.

A. L. QUAINTANCE, in charge of deciduous fruit insect investigations.

E. F. Pures, 77 charge of bee culture.

D. M. RoceErs, 77 charge of preventing spread of moths, field work.

Rouua P. Currig, in charge of editorial work.

MABEL COLcorD, in charge of library.

SOUTHERN FIELD Crop INSEcT INVESTIGATIONS.

W. D. Hunter, in charge.

W. D. Pierce, J. D. Mitcwett, G. D. Surru, E. A. McGrecor, Harry PInxvs,
W. A. THomas, D.C. Parman, B. R. Coan, engaged in cotton boll weevil inves-
tigations.

F. C. Bisuorp, A. H. Jennines, H. P. Woop, W. V. Kine, G. N. Wotcorr, SS
in tick life-history investigations.

A. C. Morean, G. A. RunNER, S. E. Crump, engaged in tobacco insect investigations.

J. L. Wess, T. E. Hottoway, E. R. BARBER, engaged in sugar cane and rice insect
investigations.

R. A. Cootzy, D. L. Van Dine, Witmon NEWELL, A. F. Conrapt, C. C. KRUMBHAAR,
collaborators.

2

LETTER OF TRANSMITTAL.

U. 5. DEPARTMENT OF AGRICULTURE,
BUREAU OF ENTOMOLOGY,
Washington, D. C., September 28, 1911.

Str: I have the honor to transmit herewith a manuscript entitled
‘The Insect Enemies of the Cotton Boll Weevil,’ by Messrs. W.
Dwight Pierce, R. A. Cushman, and C. E. Hood, agents and experts
engaged in cotton boll weevil investigations. The present manu-
script contains a complete summary of the studies of the boll-weevil
parasites conducted since 1905. The boll weevil is now known to
be attacked by 29 species of parasites and 26 species of predatory
insects, most of which are indigenous to the United States. It is the
purpose of this manuscript to show the sources and value of these
enemies and to indicate how they may be utilized to the advantage of
the farmers of the cotton belt. J recommend the publication of this

manuscript as Bulletin No. 100 of the Bureau of Entomology.

Respectfully,
L. O. Howarp,
Entomologist and Chef of Bureau.
Hon. JAMES WILSON,
Secretary of Agriculture.

SCONTENTS:

Hiralr@ Gul GEO Me ey ee ieee, ahora eer e sade sacra a eel tal ar choles area yarae ee wea See ones aoc oc ceia lal a
Conduct omtheyparashbe: PLOJCCUsser cass terse eae Clowes ose eee es
astormealidata meme sk ctos ok ae, sasha eel Aa se. . sik wo eaten eee an
SCOPCOMmPRCSEMUTCDOLG. cose oe Moki cetiasaiepermers cere vice js le alate shana, tarsi a)a

art L.

SOON Dar wN HE

a
a)

13.
14.
15.
16.

The status of the cotton boll weevil and its enemies.

. A general chronological study of the insect control of the boll weevil. .
. Nature and sources of the material examined............-------------
. Seasonal studies of insect control, by class of infested material... ..--
. A geographic study of the statistics of insect control...........---..-.-
. A study of the share of insect control in the mortality of immature boll

WCE Val Greet Neel MPU igh aa glne nn acre a AMiN EE Lu ag sr aN Csr i naan

. A study of how agriculture modifies insect control.................----
MiG himatie considera trom sassy ene Ge Nia ela a ia ik PD pay Meese
. How insect control follows the dispersion of the boll weevil.......-...-
. The status of the boll weevil and its control by insects...............
. A brief statement of the various classes of control exercised upon the

Na Oy. eval eco se ee ste pun pec NR LM MAIL OSTA Ciee, MAC Ve US Wa a nti Se a

. Practical conclusions derived from statistical studies..............----
Part i.
. A list of the insect enemies of the cotton boll weevil................--
-hnerhosts of bolloweevalparasites:- 2.5 ces. oon soe ea ee ee
savatesswnieh attack the bollweejvall se). sees see en aa ec
Plies swihieh, parasitize the boll weevils). 00286 ooo Vo ec a ae

Biological complex.

The hymenopterous parasites of the boll weevil...................-..

. Biological notes upon the parasites of the weevil .......-.......-.-..--
Mmiherdevelopmentor the, parasites 0h c0een. Leeper oe
Pebherdistuipmtion ot the parasites: :2 2 os sur. oe Se See
PUNO PALASILO SEASONS sole aise ie is ciel, Bee cies Se See tga ae en aN
POA CIUSEM CIE COMM W MOSES 2/2.) otic ci oS Sie les tN one, mate yee
J beetles which prey upon the bolliweevil.. 2... 22-2225.) . 622s ee
. Lepidopterous larva which are incidentally predatory upon the boll

WiC ON testes CCE eyes cays Shue Lec c Ie NOS se, 9) URE ee re ape
Ants which prey upon the boll weevil.. ste Saree Bea ely gs cman Saipan
Biology of the cohosts of the ‘elie aennaties Ae psa a ce ce ee OS eae
A list of the host plants of the cohost weevils..-...-.-.00.-22.....2...
A summary of the more important biological facts..................-.-

Part III. The economic application.

lls

“IO Or P & LO

8.

he economic principles imvolved 26/1 e yaa yo 4) ee yee

Helmterpretation on parasite statistics. 225 is seme sa cepa sani ei
MAuterpretatom onthe biolosicali complex. s.0. gags! 22s eee ee ee
EHOW etOLPrObt DyvexIstInONCONGIGIONS! aie. sya lex dees Style yee
. How to plan for the greatest possible control...........---.-----.----
PEropacaloneand antiicial amtrodimctionsss .5)o%e <4) 2) wile oye
: Objectionable PEACCICER sesso amie aera

The economic significance of the ee one

J oZal] NUS Koyea zy OV ON hsb SNR i MRR LN ae eA RL ee ne es URC ae ena hg URE RS rye

EES IE IOL RES:

PLATES.

Pirate I. Shedding and retention of forms on the cotton plant. Fig, 1.—Normal
scars left by shed forms. Fig. 2.—Abnormal scars with forms
retained. Fig. 3.—Longitudinal section through base of retained

IT. Eggs of boll-weevil parasites. Fig. 1.—Microdontomerus anthonomi.
Fig. 2.—Unidentified egg. Fig. 3.—Cerambycobius cyaniceps.
Fig. 4.—Eurytoma tylodermatis. Fig. 5.—Catolaccus hunteri. Fig.
6.=Umidertitied ega re Re Orn oun ec oe aes Were nee eae
III. Parasites of weevils. Fig. 1.—Hurytoma tylodermatis, pupa. Fig. 2.—
Catolaccus incertus, pupa. Fig. 3.—Cerambycobius cyaniceps, pupa.
Fig. 4.—Microdontomerus anthonomi, pupa. Fig. 5.—Larva of
Microbracon. Fig. 6.—Microbracon mellitor, pupa. Fig. 7.—Larva
ofrchaleidoid {a2 72 eee ee ee ee

TEXT FIGURES.

Fic. 1. Diagram illustrating the monthly percentage of mortality of immature
boll weevils due'to inseetenemies: <0... 2 eas oe ee

2. Diagram illustrating the average climatic and insect control of the
immature boll weevils during 1906, 1907, 1908, and 1909, in hanging
BQWAKES es chs). eelace es same weet ak at ecient Mea gee eee ae eee

3. Diagram illustrating the average climatic and insect control of the
immature boll weevils during 1906, 1907, 1908, and 1909, in fallen
BOMUIAT CS sero ter hcrcere e ere eee Sek Soke eens eae ene en aan

4. Diagram illustrating Texas climatic variations in 1907, 1908, and 1909.
5. Diagram illustrating Louisiana climatic variations in 1907, 1908, and
WOOO oak monet o Sotk waysiels oe fee ote erersye Seat aren ee neers

6. Diagram illustrating the boll-weevil complex......- Oe ia alee eee eee
7. Diagram giving the parasites of the boll weevil and their other hosts.
8. Pediculoides ventricosus: Adult female, before and after inflation of
abdomen: withteges:and’ young. oc ee sae ie Soe ee eee

9s Microdontomerus.anthonomi: NGuUlt 222), eee ee
Oca brocyius: pierces ke Wp aes Se ce ae ree eee Ae age
11. Sigalphus curculionis: Male, female, antenna................:-...---
12 Microbracon: mellitor: Adulto bo on sae ee
13. Diagram illustrating yearly rank of the boll-weevil parasites, 1906,
OO T VOOS am NOOO ee eae saree are aes eee hee tea

14. Map showing the distribution of the more important parasites of the
boll weevil ee oe ee Ne se pe rae a Noa ae

15. Diagram illustrating the seasonal rotation of hosts of Catolaccus huntert
and’ Cerambycobius cyanicepsc. = ace: ee eee a

16: The * fire-ant”” (Solenopsis:geminata):| Worker. _.2. 22 =e) see
17. The little black ant (Monomorium minimum): Adult, egg, larva, pupa -

Page.

16

56

56

19

2

Fig. 18

19.
20.
21.
22.
23.
24.
20.
26.

INSECT ENEMIES OF THE BOLL WEEVIL.

. The little red ant (Monomorium pharaonis): Female and worker. ....
The coffee-bean weevil (Arexcerus fasciculatus): Adult, larva, pupa...
The bloodweed weevil (Lizus scrobicollis): Adult. .....-.......--..-
The ironweed weevil (Desmoris scapalis): Adult..............-....--
The pepper weevil (Anthonomus eugeni): Adult.......-.-.....----
The cowpea weevil (Chalcodermus xneus): Adult... .......-........
The plum curculio (Conotrachelus nenuphar): Larva, adult, pupa...--
The potato-stalk weevil (Trichobaris trinotata): Adult, larvee, pupee- -
Mhe rice weevil (Calandraoryza): Adultssoes oases cece ss see

ot baAaeath at oe Aya me. WS

we, SY

tees
sighbies:
LT. Say

f :
Hats:
- a

THE INSECT ENEMIES OF THE
COTTON BOLL WEEVIL.

INTRODUCTION.

When the cotton boll weevil first entered the United States it
appeared to have almost undisputed sway. It did, in fact, escape
most of its enemies. But whether parasites were introduced with it
or not, we now know that within the first three years of its existence in
Texas it was attacked by three important species of parasites. Year
after year new factors in its control are becoming apparent, although
some have probably been concerned since the beginning. On the other
hand, it is certain that entirely new elements are entering the struggle
as rapidly as the weevil enters new biological complexes. Among the
most striking of these new elements in the control is the recent adjust-
ment of Microdontomerus anthonom Crawford (fig. 9, p. 49). This spe-
cies was unknown until 1906 when a very few were taken in material
reared at Cuero, Goliad, Hallettsville, Victoria, and Waco, Tex.
In 1907 it was found to predominate in a portion of central Texas.
In 1908 its range was found to extend northward to the Red River.
In 1909 it was found as far east as the Mississippi River in Louisiana.
Only a single record of the occurrence of Sigalphus curculionis Fitch
(fig. 11, p. 53), the common parasite of the plum curculio, had been
made prior to 1908. In that year it began to make its presence felt
in northeastern Louisiana and western Mississippi. In 1908 a new
chalcidoid parasite, recently described as Tetrastichus huntert Crawford,
was found to be the leading parasite of the boll weevil in northern
Louisiana and western Mississippi. In 1909 this species was found
as far west as Arlington, Tex.

With the advent of each new enemy and its more complete adjust-
ment, the power for damage possessed by the weevil is by so much
diminished. On the other hand, every factor which checks these
enemies without also checking the weevil benefits the weevil.

If the solution of the boll-weevil problem consisted merely in adding,
one by one, factors which would cut off a given percentage of the
weevils, the time would not be distant when that might be accom-
plished. The problem is far more complicated. The various factors
do not discriminate against one another, for while heat, ants, cold
weather, and excessive moisture may remove many parasites from
the struggle, it also follows that heat, cold weather, heavy rains,

9

10 INSECT ENEMIES OF THE BOLL WEEVIL.

predators, and other pests remove many ants. Proliferated plant
tissue frequently acts as an important check upon the weevil, and yet
in many cases it serves as a food for the weevil rather than as a con-
trolling force. Moist weather aids the weevil. The leaf-worm, while
cutting off most of the weevil food, finally is checked by parasites.
Light, heat, and dryness favor certain parasites, whereas shade and
moisture favor others. Parasites that are normally primary fre-
quently waste their energies by accidentally becoming secondary,
and normally secondary parasites, and probably tertiary as well, also
enter the consideration. Predatory enemies fail to distinguish the
parasites from the weevils, but are in turn held in check by their own
enemies. The birds devour weevils, predators, and parasites, but
they in turn are kept down by other birds and even man himself,
and thus the complexity grows.

There are 49 species of insects which attack the immature stages of
the boll weevil. Of these insects 29 species may be classed as para-
sitic, 5 as predatory larve, and 15 as predatory adults. They are
divided among the orders, with 3 in the Acarina, 4 in Coleoptera, 36
in Hymenoptera, 5 in Diptera, and 1 in Lepidoptera. One of the
acarians, 1 coleopteron, 15 Hymenoptera, and 1 dipteron, may be
considered as quite important. A weighted average mortality of
3.93 per cent of all immature stages may be accredited to the com-
bined factors of all parasitic enemies; the predators are responsible
for 15.93 per cent, while climate is responsible for 24.45 per cent, and
plant proliferation 12.42 per cent. This makes the total natural
control of immature stages 56.73 per cent.

In addition to the insects attacking the boll weevils in the squares,
there must be considered the insects which prey on the adults. These
include one praying mantis, one predaceous bug, two beetles, and
two ants—six species in all.

This bulletin is not concerned with the mortality of the weevils
after they become adult because reliable figures can not be gathered
upon this point. It is certain that many weevils fall prey to preda-
ceous insects and to birds, and the well-known habits of the horned
lizard would include it in the list of possible enemies. The important
fact is that 56 per cent of the weevil eggs fail to produce adult weevils.
The remainder is still a formidable number but the many adverse
influences continue to operate upon the adults and likewise upon
their progeny.

CONDUCT OF THE PARASITE PROJECT.
The investigation of the insect enemies of the cotton boll weevil

was initiated in 1905. It has been conducted from the beginning by
the senior author under the direction of Mr. W. D. Hunter, and with

HISTORICAL DATA. 11

the direct influence and encouragement of Dr. L. O. Howard, Chief
of the Bureau of Entomology. Messrs. R. A. Cushman and C. E.
Hood have been intimately associated in the preparation of the
material for this bulletin since 1907. The collecting, examining, and
recording of the immense mass of material has involved in addition
the services of Messrs, E. A. Schwarz, J. D. Mitchell, W. E. Hinds,
Wilmon Newell, F. C. Bishopp, A. C. Morgan, F. C. Pratt, C. R.
Jones, W. W. Yothers, G.D.Smith, A. H. Rosenfeld, H.S. Smith, E.S.
Tucker, T.C. Barber, S. Goes, C.S. Spooner, C. W. Flynn, T.C. Paulsen,
J. A. Hyslop, V. I. Safro, T. E. Holloway, H. Pinkus, W. H. Hoffman,
¥. L. Elliott, and O. M. Lander. Considerable credit is due Messrs.
K. A. Schwarz, J. C. Crawford, W. M. Wheeler, D. W. Coquillett, and
C. H. T. Townsend for determination of the insects concerned. The
weevils entering the biological complex have been determined by the
senior author. In short, 33 entomologists have directly contributed
the data which are herewith presented.

HISTORICAL DATA.

The first definite records of the parasites of the cotton boll weevil
were made by C. H. T. Townsend in 1895 when he mentioned a small
hymenopterous parasite and also recorded the suspicious occurrence
of several species of Scymnus in the squares, and mentioned that a
fungoid parasite, a species of Cordyceps, ‘‘was found growing out of a
dead pupa in its cell in a boll, November 26, in a field in San Juan
Allende, Mexico.”’ (Townsend, 1895.) In 1901 Profs. Herrera and
Rangel published notes concerning the parasitic attack of Pedi-
culoides ventricosus Newport (fig. 8, p. 46) upon the boll weevil
(Rangel, 1901b, 1901c).

In 1902 Dr. Wm. H. Ashmead described Bruchophagus herrere from
Coahuila, Mex., as a primary parasite of the boll weevil (Ashmead,
1902a, 1902b; Herrera, 1902a). Thisis probably Hurytoma tylodermatis
Ashmead. Prof. Herrera also recorded the activities of a predaceous
ant, Formica fusca Linn., subspecies subpolita Mayr, variety perpilosa
Wheeler, (Herrera, 1902b; Wheeler, 1902). In the same year Prof.
F. W. Mally recorded the fact that Bracon (now Microbracon) mellitor
Say (fig. 12, p.54) and Hupelmus (now Cerambycobius) cyaniceps Ash-
mead had been reared by him, since 1899, in considerable numbers
from the boll weevil. He also recorded a species of Hurytoma.
(Mally, 1902.)

In 1904 Hunter and Hinds recorded additional primary parasites
as follows: Sigalphus curculioms Fitch (fig. 11, p. 53), Catolaccus
incertus Ashmead, Urosigalphus (robustus Ashmead), Bracon (dorsator
Say), and Hurytoma tylodermatis Ashmead, as well as an entomog-
enous fungus, Aspergillus (Hunter and Hinds, 1904). The Urosi-

12 INSECT ENEMIES OF THE BOLL WEEVIL.

galphus has since heen described as Urosigalphus anthonomi Crawford
(Crawford, 1907a).

In 1906 Mr. Nathan Banks described a mite, Tyroglyphus breviceps,
collected at Victoria, Tex., from boll-weevil larve (Banks, 1906).

In 1907 the senior author of this bulletin added Hydnocera pubescens
LeConte as a predaceous enemy of the boll weevil (Pierce, 1907a, 1907b,
1907c). In the same year Dr. Hinds published two papers in which
the work of Solenopsis geminata Fab. (fig. 16, p. 70), variety xylone
McCook, as a predator on the boll weevil was fully discussed, and
considerable statistical data on the parasitic control of the weevil
were presented (Hinds, 1907a, 1907b). Mr. Morgan published a
brief account of the predatory attack of a bug, Apiomerus spissipes
Say (Morgan, 1907). Mr. J. C. Crawford described as parasites of
the boll weevil Torymus anthonomi, Urosigalphus anthonomi, and
Urosigalphus schwarzi (Crawford, 1907a).

In 1908 the senior author of this report recorded Catolaccus antho-
nomi Ashmead as a boll-weevil parasite and Cathartus cassiz Reiche
(gemellatus Duval) as a predator (Pierce, 1908a, 1908b, 1908c, 1908d).
Mr. Crawford described Cerambycobius cushmani and Catolaccus
huntert as new parasites of the boll weevil (Crawford, 1908). During
the same year two new predaceous enemies of the boll weevil were
recorded from Louisiana, namely, Hvarthrus sodalis LeConte and
Evarthrus sp. (Newell and Trehearne, 1908). . Mr. Townsend, in a
paper on the muscoidean flies, recorded Ennyomma globosa Townsend
as a parasite of the boll weevil (Townsend, 1908). During 1909 Mr.
Crawford described Tetrastichus hunteri as a parasite of the boll weevil
(Crawford, 1909b).

SCOPE OF PRESENT REPORT.

The present report is supplementary to a former bulletin which
was based on investigations prior to 1907 (Pierce, 1908a). The
matter contained herein has mainly been gathered during the years
1907, 1908, and 1909. Only such notes as are of value for the sake of
comparison have been repeated from the previous report.

The work is divided into three parts:

I. The status of the cotton boll weevil and its enemies.

II. The biological complex.

III. The economic application.

PART I. THE STATUS OF THE COTTON BOLL WEEVIL AND ITS
ENEMIES.

Part I of this bulletin shows the large mass of statistical material
gathered during the four years of the parasite investigation, and
attempts to place this material in such form as to show its economic
value and significance.

CHRONOLOGICAL STUDY OF INSECT CONTROL. 1h)

The matter is arranged topically as follows:

1. A general chronological study of the insect control of the boll
weevil.

2. Thenature and sources of the material examined.

3. Seasonal studies of the insect control by class of infested material.

4. A geographical study of the statistics.

5. A study of the share of insect control in the mortality of imma-
ture weevils.

6. A study of how agriculture modifies insect control.

7. Climatic considerations.

8. How insect control follows the weevil dispersion.

9. The status of the boll weevil and its control by insects.

10. A brief statement of the various classes of control exercised
upon the weevil.

11. Practical conclusions derived from statistical studies.

1. A GENERAL CHRONOLOGICAL STUDY OF THE INSECT CONTROL OF
THE BOLL WEEVIL.

Records upon parasitic control of the boll weevil begin in July,
1902, and occur more or less scatteringly until June, 1906. The
records of 1906 were very extensive, but, as will be shown, they were
merely preliminary.

Table I is arranged to show the extent of the examinations made
since 1902. This table should not be used to compare the records
of the various years, as the manner of investigation in each year has
been different, and the sources of the material have varied greatly.
The table is only intended to show the total of the examinations and
how these were distributed from year to year. A careful analysis of
the figures has been given in the various sections of Part I.

TaBLE I.—Insect control of the boll weevil, by years.

Per cent mortality due to—
Weevil | Preda- Para-
YEAR. 5 :
stages. tors. sites. Predas Barat All
tors. sites. insects.
SPAS AA See oe cise ree eS FRE Serre ee ee 602 (69) Ul (?) 1.16 (?)
QOS os eres etal eer Rs atch oi 819 (?) 59 (?) 7. 20 (?)
QOS; Mian Cheyer acs shal eres ie) aie ae 1,005 (?) 32 (?) 3.18 (?)
1905, August. SE OHS BORG TeS Cs SESE 1,702 (?) 21 (?) 1.23 (?)
1c SUS SSS eo SSS SSO eyo eee mines ra ae 40,073 10, 547 1, 728 26.31 4.31 30. 62
MO ret ener Ne LEE Ra LES gd ol. SA 13, 602 2,279 aaa 16.75 8.24 24.99
OOS were strc alarctt rorejelniaiobene seer nareie: tod pe inis: aaa 29, 349 3, 862 2,952 13.15 10. 05 23. 20
O09 Barctvsiaeaiseineriec ae ote faci ce acd eee 11, 653 1,231 620 10. 56 5. 32 15. 88
Totalsiand averages../-.--.......-.- G8ES805 hee see eels 6N540) Ee o ee eee LGHOV TS Cees:

1 This table should not be construed as indicating that parasite control has been falling off, because
the table is based upon different kinds of examinations in different years. The detailed analysis of these
records will be found on subsequent pages,

14 INSECT ENEMIES OF THE BOLL WEEVIL.

The figures of 1902 are based on the total number of stages reared.
The data of 1903 are based on the total number of stages reared, but
include both stages in hanging and fallen forms. There were 654
stages in fallen forms of which 14, or 2.14 per cent, were parasit-
ized, and 165 stages in hanging forms of which 45, or 27.27 per cent,
were parasitized. The figures of 1905 were separated to show the
investigations of March and August because of the great difference
in the mortality from parasites in these two months. Between 1906
and 1909 the data represent all classes of infested material and all
infested regions. It will be noticed that for the last four years the
total insect control of the immature weevils has fluctuated between
15 and 30 per cent.

2. NATURE AND SOURCES OF THE MATERIAL EXAMINED.

During the four years 1906-1909 examinations of mortality have
been made of material collected at 6 places in Arkansas, 26 in Louisi-
_ ana, 6 in Mississippi, 7 in Oklahoma, and 65 in Texas, making a total
of 110 places. These examinations are based upon 94,677 stages,
involving an individual examination of over 222,700 cotton forms
(squares, blooms, and bolls). Many other collections and exami-
nations were made, but because of incomplete records are excluded
from the accompanying tables.

During the four years there has been the equivalent of examina-
tions in 176 localities, or an average of 44 localities per year.

3. SEASONAL STUDIES OF INSECT CONTROL, BY CLASS OF INFESTED
MATERIAL.

Very shortly after the work began in 1906 it became evident that
the activity of the parasites and other insect enemies of the weevil
was very different in squares and bolls, and in fallen or hanging
squares or bolls, and also that the highest control by parasites was
in hanging squares. |

An examination of the squares of various varieties of cotton plants
will show the observer that certain ones have a transverse attachment
of the pedicel to the stem. In all cases where this attachment is
perfectly transverse, the square when injured by any insect is caused
to drop because of the separation of the infested part from the main
stalk by the growth of an absciss layer at the point of attachment.
(Pl. I, fig. 1; fig. 3, a.) Certain other varieties indicate a long diag-
onal attachment to the stem. When these squares are injured, a
diagonal absciss layer is formed which runs down the stem from one-
half to three-fourths of an inch or even more. This layer is gen-
erally incomplete at the lower point and consequently the square

SEASONAL STUDIES OF INSECT CONTROL. ‘15

hangs by a few threads and dries on the plant. - (Pl. I, fig. 2; fig. 3, 6.)
To these squares and bolls which thus hang, we have applied the
terms “‘hanging squares”? and “hanging bolls.”’

Tables II and III, which are arranged to show the monthly per-
centages of control by parasites, by predators, and by all insects,
illustrate the differences in the control of the weevil in the four
principal classes of infested material, namely, fallen and hanging dry
squares, and fallen and hanging dry bolls.

A few words of explanation of these tables are necessary in order
to show what is meant by the different classes of mortality. It has
been found that a large number of stages are destroyed as eggs or
young larve by the proliferation of the plant tissues. At the time
of the examination for mortality of the weevil, all evidence of the
weevils destroyed in these stages has disappeared; consequently the
percentages of mortality given in the following tables are the per-

~ centages of stages found which are killed by the causes enumerated,
and the mortality from proliferation is entirely ignored.

TasiE IIl.— Monthly mortality of the boll weevil due to insects in fallen squares.

t

Stages killed by— Percentage of stages killed.

Squares Stages.

Month. exam | found
ined. y Cli- | Preda-| Para- Total Preda-| Para- All
mate. | tors. sites. een aEtOrS® sites. | insects.
1906.

NUTT Geto scree seein ss 4,476 3, 831 1,797 914 118 73. 8 23.8 ual 26. 9
Tulyee ease eee ee sek Se Ae 7,265 | 4,552] 1,676} 1,079 207 65. 1 23. 6 4,7 28. 3
ATI OUST See eee Nome 19,305 | 11,186 2, 828 4,097 180 63.5 36. 6 1S A 38. 3
Septembenay. iy wees 10,709 | 5,365 | 1,475 | 1,625 28511) e684 3002 5.4 35. 6
October ches eee eee 1,981 690 205 133 33 61.8 22): 5. 6 OHNE

Totals and averages for
06

1907.
BIULTI OS cise ee reso sepe Lea 2,074 | 1,261 339 198 76 48. 6 15.6 7.0 22. 6
TRO ose gots peg a ape ep a 5,192 | 3,608 778 433 TBS Rese odd eae) 3.8 15.7
PAVIBUS Gah rset isicrete nes 7,400 | 5,058] 2,156 | 1,372 155 72. 8 PHA 3. 0 30. 1
INOVemMbeRt eos -eeee cee 15 93 9 0 1 97. 7 0 1.0 1.0

Totals and averages for

OO (amie cio eaten ise 14,816 | 10,020 | 3,363 | 2,003 365 57. 1 19.9 Sali 23. 6

1908.
IME ycater ta er yen ee re eR 100 56 4 0 2 10. 7 0 305 Bho)
TUTE Sc bee oes lee ieee mm 7,808 | 5,285 | 1,169 866 324] 446] 16.4 6.1 22. 5
AGI fis) ined see S a ea 7,437 | 4,690 1,047 902 219 46. 2 19. 2 4.7 23.9
PANIDUS Hamre Ne tienewls, ce eee 1,941 1, 208 540 294 63 74, 2 24. 3 5. 2 29. 5
HEDLEMbeh ais hs pee Sea 7,189 | 3,894 578 815 136 39. 2 20. 9 By i) 24, 4
OCTODER AE See emer aes aer 9,678 | 5,342 807 289 659 32.8 5. 4 12.3 LEE
INOVeTMDCR? 432 «2 rine oe es 604 369 $0 1 88 Sit Odi 22. 6 25. 3

Totals and averages for

QOS te eee eee 34,757 | 20,844 | 4,235] 3,167] 1,491 42. 7 IP} (al 22. 3

1909.
JANUAR see see ee 50 23 0 0 0 0 0 0 0
BULLS arate rece naa oe 7,677 | 4,334] 1,318 507 131 45.1 1h 3. 0 14.7
PANEL DUS tees ere galas crys 5,507 | 2,866 680 428 38 39. 9 14.9 1) 574
September: cess 750 364 19 5 1 6.8 1.4 wo se

OOS sree raya ae 13,984 | 7,587 | 2,017 940 170 41.2 12. 4 2. 2 14.6

16 INSECT ENEMIES OF THE BOLL WEEVIL.

A study of Table II, on the mortality due to insects in fallen
squares, shows that the principal insect work is that of predatory
insects and, furthermore, that the total insect control is, as a rule,
less than the eimatic control. The table embraces the examination
of 63,985 weevil stages of which 17,596 were killed by climate, 13,958
by predators, and 2,849 by parasites. In other words, the average
percentage of control in fallen squares by all kinds of insects is 26.2
per cent, or 21.8 per cent by predators and 4.4 per cent by parasites.

A further study of Table II shows that the predators have in each
year done their most valuable work in the month of August. The
parasites, however, have shown considerable variation in the month
of their best work. In 1906 the highest average percentage was in
October; in 1907 it was in June; in 1908 in November; and in 1909
in July.

Tasie III.— Monthly mortality of the boll weevil due to insects in hanging squares.

Stages killed by— Percentage of stages killed.
Squares Stages

Months. exam | sound
ined. *' Cli- | Preda-| Para- Total Preda-| Para- | All in-
; mate. | tors. | sites. ‘| tors. | sites. | sects.
1906
Jilly one See ees ees 474 247 47 20 76 57.9 8.0 21.0 29.0
INUICUS Geen ee eee 4,165 | 2,566 539 544 337 55.3 PAS Heyl 34.3
Septemipersea--- eee ease 2,032 | 1,285 245 269 125 49.7 20.9 9.7 30.6
October se -se oesee 33 20 : 10 0 70.0 50.0 0 50.0
Totals and averages for
1GQG Ro ee Sess 6,704] 4,118 835 843 538 | 53.8 20.5 13.0 3a50
1907 |
DUNC es See ec ee ce ees cree 150 88 26 5 13 50. 0 Sarl 14.7 20. 4
MW yess ese ea ee eee ee 956 513 57 16 103 34.3 ail 20.1 23.2
AVISUS Feet se eases 3,543. |. 2,011 296 175 580} 52.2 8.7 28.8 387.5
Totals and averages for
LOO Steee Rees 4,649 | 2,612 379 196 696 48.6 Uo®) 26.6 34.1
1908
H DOL ee pees ara Serres ee eRe 2,955 | 2,177 395 179 479 48.4 8.2 22.0 30.2
AIS ISU Heer sn oe sees sees 1,393 842 141 91 279 60.7 10.8 33.1 43.9
September= 32-2 35-6 .2 2-5 3,922 | 2,340 476 259 410 48.9 11.0 IW(3e) 28.5
OClOberas haces es eee 1,192 553 116 53 113 52.9 10.0 2D, Sue
INOVeMbeLate sce tee oes 1,125 10 8 0 0 80. 0 0 0 0
Totals and averages for
1QOS eee eee ees 10,587 | 5,922] 1,136 582 | 1,281 50.9 9.8 21.7 IES
1909
FaNwany sce scc awe shoes Saseiee 80 3 0 0 0 0 0 0 0
Dilys os eee acer oe eee 630 383 36 31 117 48.0 8.1 30.5 38.6
SALIIOUS tate ease ae eee ee 1, 290 856 96 72 88 29.9 8.4 10.3 18.7
Septembermeae ae see eeceses 745 496 61 21 76 31.9 j 4.2 15.3 19.5
November yee. se ae 136 52 14 0 30 | 100.0 0 68. 0 68.0
December ase sees se 515 169 47 1 71 70. 4 6 42.0 42.6
Totals and averages for

GW) Cegscbebhasssoase 3,396 | 1,959 254 | 125 382 38.8 6.4 19.5 25.9

Table IIL shows that the principal insect work in hanging squares
is that of parasitic insects and, furthermore, that the total insect
control in hanging squares is each year higher than the climatic
control. This table embraces the examination of 14,611 weevil

Bul, 100, Bureau of Entomology, U. S. Dept. of Agriculture. PATE T

SHEDDING AND RETENTION OF FORMS ON THE COTTON PLANT.

Fig. 1.—a, b, c, Normal sears left by shed forms. Fig. 2.—a, b,c, Abnormal sears with forms
retained. Fig.3.—a, Longitudinal section through base of shed form; 6, Longitudinal
section through base of retained form. Fig. 1, natural size; fig. 2, somewhat reduced;
fig. 3, enlarged two diameters. (After Hinds.)

SEASONAL STUDIES OF INSECT CONTROL. 17

stages of which 2,604 were killed by climate, 1,746 by predators,
and 2,897 by parasites. In other words the average percentage
of control by all kinds of insects is 31.61 per cent, or 11.93 per cent
by predators and 19.68 per cent by parasites. It appears that July
and August are usually the best months for parasite control in hang-
ing squares. :

Taste IV.—Monthly mortality of the boll weevil due to insects in hanging bolls.

Stages killed by— Percentage of stages killed.

Months. * exam- Binees
ined. i Cli- | Preda-| Para- Total Preda-| Para- | All in-
mate. tors. sites. Hh GabOrss sites. | sects.

1906

APIO S c's Soe gees SHES ESSE SOeee 145 0 0 0 0 0.0 0.0 0.0 0.0

MUlyes eee ee te aseeeee 2,947 290 28 14 0 14.5 4.8 0 4.8

ATT OUSt see te ee eee emcee 23,454 | 2,977 416 742 19 39.5 24.9 6 25.5

Septemibenix Ssa-cacecsssce 3 6,210 | 1,555 198 188 23 26.3 12.1 5G) 13.6

WClOMe esse ee ass el 399 147 21 11 3 23.8 7.5 2.0 9.5
Totals and averages for

LONG ee sates ee 33,155 | 4,969 663 955 45 35.9 21.9 0.9 22.8

1907 a

NUIT reper ou ete Shares, Uo aig eek 50 5 1 0 1 40.0 0 20.0 20.0

Uy PEE eae ise SS eee 50 d 0 0 14.3 0 0

JUD ies Sone ae SoSH ena aaers 1,272 330 115 51 4 ole5 15.4 1 16.6
Totals and averages for

MOOV lays site rere os 1,372 342 117 51 5 50.6 14.9 1.4 16.3

1908 a,

UNE ee ce Nee a aeiaeiaas eee 2,227 238 34 6 21 29.7 2.5 8.8 ible}

Uy ee ee ones cee ok eee 4, 450 405 125 26 13 40.5 6.4 3.2 9.6

IANTOUST EA niee ee aera teil 1123 200 22 20 0 21.0 10.0 0 10.0

Septem eres soem sesso ae 933 98 8 4 1 13.2 4.1 1.0 5.1
Totals and averages for

OO eae seers sete 8, 733 941 189 56 35 29.7 5.9 3.7 9.6

1909 i

OT AT We eras ayers eS a ere ics i, 616 402 133 25 11 42.0 6.2 2.7 8.9

Mebruaryeec den coos oee cto: 716 115 12 49 2 54.7 42.6 Ise 44.3

Mar Gh ect sae, ees sare 65 seeks 608 56 14 12 1 48.2 21.4 lave 23.1
Totals and averages for

1909 ise eeepc 2,940 573 159 86 14 45,2 15.0 2.4 17.4

In fallen bolls the principal insect work is that accomplished by
predatory insects, and the total insect control has been less than
the climatic control except in the year 1906. Table IV covers an
examination of 6,825 weevil stages, of which 1,128 were killed by
climate, 1,148 by predators, and only 99 by parasites. This means
that 18.2 per cent of all the stages were killed by insects, or 16.8
per cent by predators and 1.4 per cent by parasites. In this class
of infested forms it is also noticeable that the principal work by
the predators is accomplished during the month of August.

16844°—Bull. 100—12——2

18 INSECT ENEMIES OF THE BOLL WEEVIL.

TaBLE V.— Monthly mortality of the boll weevil due to insects in hanging bolls.

Stages killed by— Percentage of stages killed.

Month. exam- | pound
ined. R Cli- Pred- | Para- Total Pred- | Para- All
mate. | ators. | sites. ‘ i

1906.
SIU Ve oa eee ee a ye ener 434 22 2 3 0 22.6 13.6 0.0 13.6
JNU EA DIST Pes sites co UR Pe ene 8,762 | 2,444 281 340 155 31.7 13.9 6.3 20.2
Septemibenesees-ceeceee aes 4,224 | 1,627 130 292 101 32.1 UAE!) 6. 2 24.
OCtOberse ai oo ise 8,629 | 1,359 186 266 66 38. 2 19.6 4.8 24.
Totals and averages for
EGO Gresser oe Scie Sse 17,049 | 5,452 599 901 322 33.4 16.5 5.9 22.4
1907
thy eee eee ee oa tase ae. 460 38 2 0 0 5.3 0 0
PATIO UISh eee se eee here eee 683 393 35 13 50 25.0 3.6 112505) 16.1
Totals and averages for
1 OT Se 1, 143 431 37 13 50 23.2 3.0 11.6 14.6
1908. | ;
MebrUaAnys.e es c ee eeee sees 12, 451 515 424 0 54 92.8 0 10.5 10.
Manchaeesasaen es ecm oa: 1, 239 22 21 0 1} 100.0 0 4.5 :
ACG Sa Ss ee AG es ee eee ee ee 2 492 63 37 58 32.0 7.5 11S 7 1
INUSUS Gece eee ae ae a eo 720 191 23 7 22 27.2 3.6 11.5 1
Septembers- 5-6 .+-ns22 = 425- 664 83 7 17 2 31.3 20.4 2.4 2
October esas eee sae 432 262 36 11 5 19.8 4.2 1.9
INOVembers2a ee. Stee sae 519 274 134 1 8 52. 2 3 2.9
Totals and averages for
QOS esate eae ed 18,157 | 1,839 708 73 | 150 50.6 3.9 8.1 12.0
1909.
JANwMaLy: ssa se scan eee 3, 941 857 335 38 43 48.5 4.4 5.0 :
MO DRUAT Yas cee sot cetera 430 30 13 11 0 68. 6 31.4 0 31.
Marche iio. ee an 653 81 16 16 0 39.5 19.7 0 19.
ATI OLISt ates lace Sime Ce ers 365 42 1 2 0 8.1 5.4 0 :
December. a4 hese ee 2,148 519 217 13 11 46.4 25.0 Qe 27.
Totals and averages for
1S) 08 Aa cm ar ae 7,587 | 1,534 | 582 80 54 46.7 5.2 3.5
: }

a

SEASONAL STUDIES OF INSECT CONTROL. 19

Table V was based, as may be seen, upon the examination of 9,256
weevil stages, of crit 1,926 were killed by climate, 1,067 by pred-
ators, and 576 by sometices In other words, shee meee enemies
killed 17.74 per cent, of which 11.52 per cent were killed by predators
and 6.22 per cent by parasites. July and August are the principal
months for attack upon hanging bolls.

Summarizing the four preceding tables, Table VI is presented to
show the monthly rate of mortality in all classes of infested forms.

|

| /
2 ae
| Ce

8

oad aay

x of

«90

ky

: spec haor ae aS - ae
hs

Fic. 1. oh ee illustrating the ie percentage of mortality of immature boll weevils due to
insect enemies. (Original.)

It will be noticed that in each year certain months have been omitted
and it must be explained that the reason therefor has been the
necessity of making these examinations only when other work was
not pressing. An analysis of this table is more readily made by
reference to the accompanying diagram (fig. 1), which demonstrates
conclusively that August is the month during which the insect enemies
of the boll weevil are most active.

20 INSECT ENEMIES OF THE BOLL WEEVIL.

TaBLeE VI.— Monthly rate of mortality of the boll weevil in infested forms of all classes

Percentage of stages killed by—

Totals and averages for 1909 27, 857 11, 653 41.73 25.84 10. 56 5. 32 15.88

Forms
Month. exam- Siages
ined S : All Glmste Preda- Para- All
causes. ; tors. sites. insects.

1906.
aU Cy psa a ee ee Oe NEE 4,621 3, 831 73. 80 46. 90 23. 80 3.10 26. 90
oe a ee ea tere ie ee 11, 120 5,111 61. 67 34. 29 21.83 5.53 27. 36
IASI CUS tyre see ea en ee Aree meee 55, 686 19,173 54. 64 21.19 29. 32 3. 61 32.93
Neprembens moc noses eee 23,175 9, 832 50. 40 20. 80 24.14 5. 43 29.57
Octobersseees ssec ee eee ak 6, 042 2, 126 44.12 19. 56 19.75 4.79 24.54
Totals and averages for 1906} 100, 644 40,073 55. 81 25.15 26. 31 4.31 30. 62

1907
Unde) VR a oe be eae eee 2, 274 1, 354 48. 67 27.03 14.99 6. 64 21.63
Surly eee ace ok ee i ae Re ee 6, 658 4,166 36. 55 20.11 10. 77 5. 66 16. 43
CATES ES es eM Se eater 12,898 7,792 64.19 33. 39 20. 67 10.12 30. 79
Nowemberc.ccc site l eee ce 150 93 97.70 96. 70 0 1.00 1.00
Totals and averages for 1907} 21,980 13, 405 54. 27 29. 06 16. 88 8. 32 25. 20

1908
IBENRUALY S38 sok ote aekicas ois Se 12, 451 515 92. 81 82. 33 0 10. 48 10. 48
MATCH ssn Bam = = kee ine ome 1, 329 22 100. 00 95.50 0 4.50 4.50
WEN AE See ee On peian oem arene 100 56 10.7 7. 20 0 3. 50 3.50
FUNC He sec cee eee tae ees 10, 035 5, 523 43.81 15.7 15. 78 6. 24 22.02
SRE yee ee Meenas eee SS aie 16, 974 7, 764 45.63 20. 99 14.73 9. 90 24. 63
PAULO UIS tre teats a the Wit Vp 5,177 2,441 61.53 29.79 16. 87 14.91 31.78
Seplembersse ase eee ee 12, 708 6, 415 42.29 16. 66 17.06 8.55 25.61
Octo bert ee meee Le se 11, 302 6,157 33. 92 15. 57 5.73 12. 61 18. 34
INO VEIM DED o a. 25x cette. Oe ee f 2, 248 653 50. 53 39.52 3.06 14.70 17.76
Totals and averages for 1908 72, 234 29, 546 44. 34 21.21 13.12 10. 00 23.12

1909.
PAMUAD Yio a ace Tee oe 5, 687 1, 285 45.52 36. 42 4.90 4.20 9.10
INC DEUREY Aes cscs ose eee ae 1,146 150 58. 00 16. 66 40. 00 1.33 41.33
IMT CH eee ae eee a neers rag 1, 261 137 43.06 21.89 20. 43 Vf? 21.15
url yee ee ee ett ay 8, 307 4,717 45. 36 28.7 TMU, 3. 25 16. 34
SAUL UES Gaetan one en RS mr ee nee a 7, 162 3, 764 37. 32 20. 64 13. 33 3. 34 16. 67
Septenierwee oes ees ee 1, 495 860 21.25 9. 30 3. 02 8. 95 Teer
OV CINDER ate. eee ae eee: 136 52 100. 00 32. 00 0 68. 00 68. 00
December sasha dagen ool: 2, 663 688 52. 32 38. 37 2.03 11.91 13. 94

|
|

4. A GEOGRAPHIC STUDY OF THE STATISTICS OF INSECT CONTROL.

A study of these same statistics, when arranged to show the insect
control by States, has given much interesting light upon the subject
of the control of the weevil.

In fallen squares we find an average for total insect control of 26.8
per cent in Oklahoma, 25.9 per cent in Mississippi, 24.5 per cent in
Texas, 20.6 per cent in Louisiana, and 12.5 per cent in Arkansas.
Analyzing these figures from another standpoint, we find that the
State of Mississippi leads in parasite control with 14.27 per cent,
Oklahoma standing next with 4.71 per cent, Texas with 3.9 per cent,
Louisiana with 2.52 per cent, and Arkansas with 0.71 per cent. The —
relative rank of the States for predatory control is quite different.
Oklahoma leads with 22.16 per cent, Texas comes next with 20.6 per
cent, Louisiana with 18.1 per cent, Arkansas with 11.82 per cent,
and Mississippi with 11.63 per cent. In climatic control Texas leads _

GEOGRAPHIC STUDY OF STATISTICS. 91

with 37.9 per cent, Oklahoma comes next with 30.8 per cent, Arkan-
sas with 25.65 per cent, Louisiana with 12.5 per cent, and Mississippi
with 11.7 per cent. Thus it may be seen that the dry, prairie States
of Texas and Oklahoma lead in the climatic and predatory control
of the weevil and also in the total amount of control, and that the
climatic control in each of these States is greater than the total insect
control. This latter fact is also true of Arkansas. In Louisiana
and Mississippi, States which are naturally more humid, the climate
has less influence and the greater proportion of the control is by the
insect enemies.

In hanging squares the conditions are entirely reversed. It is
noticeable that Oklahoma leads in parasitism with an average of 31.74
per cent, Texas averages 26.6 per cent, Arkansas 24.16 per cent, Mis-
sissip1 21.2 per cent, and Louisiana 12.07 per cent. In predatory
control Louisiana leads with 12.9 per cent, Texas comes next with 10.9

FER CENT
/0 20 30 FO 50_ 60
se se: Za
LOUISIANA ‘ies ” YY ma 40,86 2 sat sires
MISSISSIPP!
OG
TEXAS (eee “Gy LD LL DATE 54.3

ee aa

OKLAHOMA Tage ee BESS Sop ces See
Bat pos

Fig. 2.—Diagram illustrating the average climatic and insect control of the immature boll weevils
during 1906, 1907, 1908, and 1909, in hanging squares. (Original.)

per cent, Mississippi with 6.98 per cent, and Arkansas with 2.53 per
cent. We have no record of predatory control in Oklahoma. In all
five States insect control in hanging squares is greater than climatic
control. With regard to climatic control Arkansas leads with 26.69
per cent, Texas has 16.8 per cent, Louisiana 15.89 per cent, Oklahoma
9.53 per cent, and Mississippi 8.12 per cent. These statistics are
sraphically shown in figures 2 and 3.

A brief comparison of the condition in hanging and fallen squares will
show that the States of Texas and Oklahoma have a higher average
percentage of control from all factors in fallen squares than in the
hanging squares; the States of Louisiana and Arkansas have a higher
average percentage of control from all factors in hanging squares than
in fallen squares, and in the State of Mississippi the difference is very
slight, although in favor of the fallen squares. This illustrates the

22 INSECT ENEMIES OF THE BOLL WEEVIL.

difficulty of giving any single recommendation for the control of the
boll weevil which would apply to all regions. This point will be
brought out more fully in other sections of this bulletin.

5. A STUDY OF THE SHARE OF INSECT CONTROL IN THE MORTALITY OF
IMMATURE BOLL WEEVILS.

The condensed tables which have been presented are likely to give
the impression that the parasite control of the weevil is on an average
very low, but it must be remembered that the examinations have
been made in all parts of the infested region whether the weevil has
been present 17 years or only a few months, and whether the weevil
damage amounts to less than 1 per cent of the crop or to almost 100

BM PER CENT
-10 S20. 30 40

Pyse| | anes
/1,82 25.65
eee eat

ARKANSAS LZZZZZA__|
2.52

= PREDATORS

22.16

cos gee ar —

Fic. 3.—Diagram ——L the average climatic and insect control of the immature boll weevils
during 1906, 1907, 1908, and 1909, in fallen squares. (Original.)

per cent. This great difference in the sources of the material exam-

ined has necessarily lowered the average mortality to its minimum.

The following records show some of the cases of very high mortality

due to parasites:

Highest records of parasitism of the boll weevil.
IN FALLEN SQUARES.

ee eee
ane umber} ageo
Locality. Date. of stages. | parasit-
ism.
EEC OTD SOM SL ey eae a yada NS re Cees Pye ey INOWt0 OO Taso 53 77.36
CornpusiChristi Mexce. 8 cee sor Cem Tes 2eeN ale, Weenie June) 20190 fees nea 92 36. 95
SING CHG SMI SS a9 1) Ae Bie aoe, ya ra aan flee aN IO NR OCES23 1908 anaes 157 28.6
BL) MUL AS Un sep ease onc eyes) ep co sree peice op cy ae pyar ger Rey eye A olO QOS eee see 18 27.78
CONTE ST es ps eae i ep ae en a ota Sr See fe Ny Jialye28 lOO Sees aa 114 26. 63
INE CHEZ IMSS esp emcee msc ra ca vay ean ehe ph slo Rete pay Adee ey al ee oa OC TGSIS0 St seus reese 230 OAT
UTE T Ne etsy pay yh OO N ates wat un POV nn ie aegaE CNL Y SACU ESL il 90) (payee ee 105 19. 04
TNE H HG) Oe As Dope LA aa ft el So voir aa eae So aes July, 1909s hee oes 200 18.5
SHre Ve Writs ss ies Po oe Fala RSA SR IS ing mua UU ae OCESZ9SIO0 Sa ee Wea 624 15.8
SW SIGH 0 ch YS DY ails aU EN BN a a a 1 A nN re June 19, (90825 ae 513 14.5
EUOOSE VEN GC ea een se ee SARI ean ted aur 8 gout gl Sept. 24, 1906.......... 69 14.4
va Ned ira 2a OT NY SE BY I ST 8 Me bk al ace ural aay so July 7, 1908.22 ee 382 13.35
Brownsville excuse eke 2 iiose ee Me ews AL A ee ap Sept); 1906s22222 ee 1,147 12.4
DenISON MOR ce ene ese PAS Ns ea ates oN NO A SR eR ao July, 1900s ee a A494 8.5
ATOKA 5 Olas Siete eee et OR ae deep I pe a ei Se Sept 201 90S ses at ae 100 8.0

SHARE OF INSECT CONTROL IN WEEVIL MORTALITY. Dace}
Highest records of parasitism of the boll weevil—Continued.
IN HANGING SQUARES.
a y ergent
: umber} ageo
Locality. Date. of stages. | parasit-
ism,
VVC O seh sity ayaa ap ah EMEA IC tes tale el Ci tas ha July. 231909 eee 39 66.6
PATI COMO Xepam ence a Mel cr pe male cos aiaee mt ne ein Ae Ue July TOOO Re ee ee 55 63. 63
AVA CEOTL BUY Le xk eer e aie Oe ee ET Uae ues eNO ANI EO LOOT eae 26 61.5
Malas Mesa ee ae sass bo EE BONS OPE ALES Beta Clete Ae et | IND A Oa Sse rioe 82 59.7
PALIT COT GRO xe ara acta Se ae rte cone Sain We er ae mate UyeAC GOS eee sees 51 56. 86
SDB) UT US RIA Gece Sagan ann A HEA POE oa IL A aS STS SuliyA OOO Ree as == 57 52. 63
BUYS. CO SET Kae re Septet an apm NIT EH I UC) sl BORA Sta July 25 WGOG ses gas tes 99 52.6
INCAS O Gey RWUR OiRe yarn seme are ae ea ape ete DR ace eee 2 a0) A eee O0S Seen ee: 29 51.75
NATCHEZ MISS Oe ht SoitaGn oie ai a esa sic nail vy Vue velar MND URL ARN WENO CEZSe G0 Re ce ee eae 82 51.3
RATA WACINE OTN Os Cs LO RNP iy ole NT Me Anse JOY ZA 1GOSe ee es 29 48.27
aller tswallex exces sae re Oy Ae ak RANA NG te ait 2 Palys MOOT ieee 19 47.3
1 DFE VES MRE sess coe es eta en arc LO eae ne Aug. 10, YO bees ae eens 193 47.15
PATTY COTM RC Nesey Attlee DEH L ieay ey ute to eerie win Winat. He, stay Aug. 6, 1O0S0 eis a 260 46. 92
Victoria, Rodis revel EU Nee) MA Miata ret Kk -...1 July 29, OOS Ys Neuere 140 45.71
Foster, TU SNR Bc OR Ie NN agi BN REC eoM ot ua Un RG a Sept. 7, 1907........ 22 45.47
Tallulah, NTA ape aie Mica 5 Sia SOONG ISI SANTIS asd RLU OR ada AMEN Dec. 20, 19092 os 4. 85 44.7
LeNoptl OTe eo) DENS e-des Ace eat ter tra nie IED ER AE Sr ce aS Aug. 29, 1907..-.... 41 43.9
[SU GURE RY CSSy OLS) CUS A he Re OE SE eSNG De es an Ve ut a Oct. 28, 1908........ 37 37. 84
NET OG RANT: Keio eatery 8 hens inl ra MA auc Sus YF SENSU UMA Sept. 16, 1908....... 69 33.33
INGE COMA 2S Si se ie cs Le el es Sa eae ree ag NR gare EN Sept. 4, 1908.......- 63 31.74
HOMIce PA een nan on eRe TR EN RIM GU | Tully Que t003 ne 284 26.05
VECO MPIC Neary etry Dec mee aE en hin senna Misi ee echare ei gea ed SR Sept. 20, O06 es 109 23.8
ITER ON TO es erase a piel dacs NE ea OT EN sag. sca eae sacs Aug. 31, 1906......- 347 21.3
Tallulah, spe rs ene ar ala a Bea AA ea Ge eR Sept., TOGO Hse 495 15. 35
IN FALLEN BOLLS.
\WiGO BEY, UNep.ec ins aaeala = Ae ie emu nL Neem nual sel a We ear ee yee June 17-29, 1908 97 14. 4
JNU BREWING Fa ep es ea Fee AR re It RIV Ae eat aS July 29, 1908.......... 22 9.09
MAS eral eys al Mem Sg eB Ie SRSAROE a St Ya 5 eta ips a ena SU Aue 24 1907). 22 oie 12 8.3
OTST CATT AM OOM cero Cie aay via Baa Ss a sae Nal Sept. 18, 1906.........- 34 5.9
Victoria, Lace CRIMES Aa ON Nich aun. Gn UEE Jan., AGO epee 87 5.74
IN HANGING BOLLS.

UD Ye ai esos wy fl De aS a Par es A a a ROR OES RCE AME DT 90M ES kee 11 36.3
WarlvieT Geyrl excep ire easig ray niweily Ie ats eee nyc (Ns Meat mIRC ahi aN INORG OR RIO ete meee ae 20 30.0
(ENG) Bay GL AB Ns A RS il we ares ONS nS July Be ISO ses eee cine 38 26. 31
1S) Peart eee a RU eH ANGI ese Mol NAL URLs Sees as) ne AAT LHI lS CLC) ears AR 38 26.31
DALPAM LONI OMe Re xen pease are ere case Wem SCN aa aT oe July oi, 1GOS i ieee 35 25.71
Victoria, Mes ea SE a spoke aN cape NOS oa VE WU oy oe Rage TY NE Aug. 10, LOO TRB eee 110 22.7
IN GCCHC Zp Masse Serene apes sen clio ee coe Mein ei cice iany eap em Jan., 1909 ews et 125 14.4
Marshall, ANB ek VE eS UeE een LT AU RGR a Yet a NO Peon TEL AN SAN Aug. 22, 1906.......... 52 13.5
Trinity, [Resetae iy eae THe AMS MMO Ratu fey etait K ‘Aug. QO NOO Gee 142 12.0
AVVELC Opal @ sccuia eens inn cp MIE AR cpa Uae MN i pe Aue SelAcn a MR Sept. 20, 1NGOGi ane tee 303 11.8

between the above figures.

IN FALLEN SQUARES.

Highest records of total insect control of the boll weevil.

Locality. Date.
INH SIONS NED eos 5 UA a RA 18 Na Ct ce oe a aR a ae AL UNy phe baal bo (07 ea
Hallettsville, HABE a se ae a lee Phen Se RN Na INGO Real IGN ee eee
Overton, SNe. ae 020 May Re aige eM RRBRT iG Sie Yeiade A Aug. ile 906i Aaee eee
Beeville, ANE. 6.5 aoe erate tae amg aan So EUS ee RAND FEN eey eae | c Osea ab er are
INALCLOTIO’ Sexier ne Aeren iin arise Ley sich tela Mea Py ent N, MuUliy29; AIOSe eee seen
Va xerenyall Sy 4 Neb; cess > Sys a en a Ue aN Re SUT phe ee Mec tea Sees a a aN OE Sept. il LOOG Sanaa Ne
Cuero, Vere Oei i RAIN) SO op rite, OKO E O8 De REA S g a June 20, 1908.........-
CON ya Sp eC ge RTE ete Eat Ny es ed tN RU roel es PP ANTE 2S OOS ees seme
Vidalia, TR yee eis eas oh 5 eee Oh een as ATE I ors pe Aa a A Sept... 15; 1908s. 2222.2

Number

of stages.

These records give merely the highest percentages for each State
in each year. There are many other records which might be included

Percent-
age of
insect

control.

96. 11
92.00
85. 20
78. 80
78. 38
77.40
73. 60
66. 69
61.20
49.71

24 INSECT ENEMIES OF THE BOLL WEEVIL.

Highest records of total insect control of the boll weevil—Continued.

IN HANGING SQUARES.

oe pecan
Vit umber| ageo
Recaity Date. of stages.| — insect
control.
TSR OVE ated Neo ae ee ee ie apres a nL OD ALN IOS Sate Oye ae er Se Aug. 1, ine Bt See 75 84. 00
Arlington, fl Weg ME Sy ES a Bey Ga ay ie ance eu I July, 19 Ns args See Cee 55 75. 44
ATLAS SAR Ge eta ae ee ae et repo ame eey ee RN ee Ne eye oe os, So pee CG Kayes ars or wedlnae se Se 57 69. 99
Victoria, cl Dye: leper aye Yet > Sacer Mis Mi Ree penal | TUEby 29S 908 aee eee 87 58. 54
Waco, Aor ae OURO talet GN I taping EN re da ON bee OL | July 1, lode ee 99 56. 56
Th EEN RRSS STEN (0 I a SS es io Sad yee Ne Sena (ea arm a Aa Sept. ie LQG ES eka sae 244 50. 90
AVKGU GY ast ad Rep eas Se rennet FR Ue es SO i NP ae ee ena Neg a [Aries SDOOT See Node 253 50. 00
IN HANGING BOLLS.
ae Yeioy aed (6 Ul CP A ia Neer ee ae ee ee | Sept. 29, 1906.......... | 145 58.30
BWV COS RG Rees Reco oe St oo Cece ena see eee ee ale Ree eRe ds 1906 eran: 421 42.04
OVervong Tews. cy oe. Ser eae Raa ae miso eNO ETAL Ae rch tO Oa ee ee ee ees 89 40. 50
iManshiel dilaa so2 ae so ic See ee ome ae wae oe Rao | Se ‘Oi, NGO! sins ee 479 33.00

THE CORRECT BASIS FOR COMPARISON OF MORTALITY STATISTICS.

As has been explained, the examinations have been made from

various sources. It is therefore necessary to arrive at some true

basis for the comparison of these data before an exact knowledge of
the conditions existing can be obtained. The first mortality of the
weevil is that due to proliferation. Dr. Hinds, in Bulletin 59 of
this bureau, has shown that the average mortality of weevil stages
in squares from proliferation is 13.5 per cent and that the average
mortality in bolls is 6.3 per cent. In the absence of further data
these two percentages are used as a basis for obtaining the weighted
average mortality.

As nearly as the proportion can be estimated throughout the entire
season, 15 per cent of the weevil stages are to be found in bolls and 5
per cent in hanging forms. Whether these arbitrary estimates be
true or not, this is the only manner in which it will be possible to
compare the mortality by the different factors in the various years.
On this basis, therefore, a series of hypothetical tables has been
erected.

In order to show how the hypothetical average differs from the
average obtained from the total examinations, two tables are given
for each year, the first being a table giving the actual conditions in
the four classes of infested material and the second table being a
hypothetical table based upon 10,000 weevil stages on the arbitrary
basis of 5 per cent of the stages in hanging forms and 15 per cent of
the stages in bolls. The process continues by first subtracting the
mortality by proliferation and then computing the mortality from
climate, predators, and parasites from the remainder. The percent-
ages of mortality given in the total line are based upon the total of
10,000 stages.

SHARE OF INSECT CONTROL IN WEEVIL MORTALITY. 25

1906.—The data on the mortality of the weevil in 1906 may there-
fore be condensed and tabulated as follows:

TaBLE VII.—Boll-weevil mortality in 1906.

Percentage of stages killed by—

Number of Penenuage Total Der
il OUSUA LCS Ba ae ern penne MCOMUAC CLO
Class of forms. Weve re :

stages. alive. Climate. | Predators. | Parasites. | M™tality.
an g@ine WOllses ss. enee eo 5, 452 66. 59 10. 98 16. 52 5. 90 33. 41
Elan eine SCUMaAreSe sei sae 4,118 46. 20 20. 30 20. 50 13.00 53.80
HatlenvbOllS ees ae cise ee 4,969 64. 53 13.34 19. 01 . 90 35. 47
Hallen(SQuUaresi\secia sacs oo 25,534 34. 80 31.20 30. 70 3.30 65. 20
Totals and averages. ..-- 40,073 44,19 25.15 26. 31 4,31 55.81

TasLe VIII.—The hypothetical or weighted average mortality of the boll weevil, 1906.1

© 1906—Mortality from—
S
E Prolif
gq 1 ea Climate Predators. | Parasites. Total.
84| ¢
Class of forms. S a a te, : we ' a F ain eos :
© OF Dif | a oo = oo =| oO| G&G o =
So Slee ee eel eS eee ale
oO » By q pes By Yes oO Pon oO a oO
I Heya =the tne) C= tea NEL) (seep Me TSEES | Povey URS UESE || opin tS Ns es
SSeS e ey eae oo a ee
B 5 16 5S at hel eis ct iy asp = bite) alls} = oO 5
a Ane Net Ve Pa) es A) (fee Ale |fet Zz |S AZ
Hanging bolls....-... 0.75 75) 6.30 4.7 70. 3)10. 98 7. 7/16. 52 11.6) 5.90) 4.1) 37.4 28.1
Hanging squares......| 4.25 425\138. 50 57.41 367.6)20. 30 74. 6/20. 50 75. 413.00} 47.8] 60.0) 255.2
Total hanging..| 5.00) 500)... 62M 43709 |ee ao SON ale, ST Ole BlsO| ee 283.3
Hallenwbolls--2 2 3-250 14, 25) 1,425) 6.30 90. 0/1, 335. 0\13.34) 178.1)19.01| 253.8) .90) 12.0) 37.4) 533.9
Fallen squares........ 80. 75] 8, 075)13. 50/1, 090. 1/6, 984. 9/31. 20/2, 179. 3/30. 70)2, 144. 4) 3. 30/230. 5} 69.8)/5,644.3
Total fallen... .. 95.00} 9,500;...-- 1,180. 1/8, 319. 9)..... 2 d0le Also cee 2,398. 2). ime DADO lees 6,178.2
Totals and aver- f x

Ages eeescnee 100. 00}10, 000/12. 42)1,242.1).....-- 24. 39/2, 439. 7/24. 85/2, 485. 2) 2. 94/294. 4/64. 61/6, 461.5
J

1 Given 10,000 weevil stages.

1907 —The mortality during 1907 was 54.27 per cent when figured
from the total number of stages and total mortality, thus showing a
decrease of 1.54 per cent from the mortality of 1906 figured in the
same manner. The parasitism showed an increase of 4.01 per cent.

TaBLE I1X.—Boll-weevil mortality in 1907.

Percentage of stages killed by—

: Number of | Percentage Total per-
Class of forms. weevil Of Stakes maranatha ee [AT |) COMape On
stages. alive. Climate. | Predators. | Parasites. mortality.

Te fr rr | a ner

an cinepollse ase ees see 431 76. 80 8.58 3. 02 11.60 23.20
Hanging squares............-. 2,612 51. 40 14. 50 7.50 26. 60 48.60
Hallentbollsgieee sence se seen: 342 49, 42 31. 28 14. 91 1. 46 50. 58
Hallenisquarese a= s-seee--2- 2. 10,020 42.90 33. 50 19. 90 3.70 57.10

26 INSECT ENEMIES OF THE BOLL WEEVIL.

Following the plan adopted for the 1906 records these figures
may be weighted for comparison with the earlier records.

TABLE X.—The hypothetical or weighted average mortality of the boll weevil in 1907.4

g 1907—Mortality from —
g
5
Prolifer- : : cola Mate
ale: soni Climate. | Predators. | Parasites. Total.
om) n
Class of forms. “3 Sp = ° a 2 fee ae ee aie .
sa| 2/5 | 3 Slee VS Sees
Px) “ Cae hee i o8 = o§ = o§ rn ole es
= ies Breas (ies Sed) Soo lea te 2 ee beacons
S| Bee] Bol Slee) see lee 2 a Sel epee
2 2 3 Q 2 2 ; 2
5) or = Og Os Og a or
= = = = g Po q So | Seat es) |S S
oO = a) 3 © ® H = oH 5 > & = 5) 5
Be A | Z eS 8 Z a Z |= | ZA |S a
Hanging bolls. ....--- 0.75) 75} 6.30) 4.7) 70.3 8.58 6.0} 3.02 she 60} 8. 129. 70; 20.9
Hanging squares......| 4.25) 425/13. 50 57.4, 367.6)14.50) 53.3) 7. 50| 27. 7\26. 60) 97.8)/55.50) 236. 2
Total hanging..} 5. 00} 500)... | 62.1| 437.9|....- 59. Esso Jk | DOSS Rees 105.9)....- 257.1
Fallen bolls.........--| 14.25] 1 425] 6. 30) 90. 01,335. 0 31. 28) 417.6) 14.91 91; 199.0] 1.46) 19.5)50.90) 726.1
Fallen squares..-...--- 80. 75| 8,075) 1s. 50/1 , 090. 1/6, 984. 933. 50/2, 339. 6) 19. 90/1, 389.8] 3. 70!258. 4'62. 80/5, 077.9
Total fallen.....| 95.00, 9, Ale 180. 1/8, 319. 9} .. . ..|2, 757. I fee 1, 588.8).....|277. 91-....|5,804.0
Totals and aver- ; | | |
OLCS Sarees ae 100. 00 10. vile 43)1, 242. i By Roam 28. 16/2, 816. 5|16. 18/1, 618. 6} 3. 83/383. alae 61/6, 061. 1

1 Given 10,000 weevil stages.

This table shows a weighted increase of 0.89 per cent for parasites
and a weighted decrease of 4 per cent for all agencies due to the
falling off in control by predators.

1908.—The mortality during 1908 was 44.34 per cent when figured
from the total number of stages, the total mortality thus showing
a decrease of 9.93 per cent from 1907. The parasitism showed an
increase of 1.68 per cent.

TaBLE XI.—B6oll-weevil mortality in 1908.
pee

Percentage of stages killed by—
Number of | Percentage

Class of forms. weevil of stages
stages. alive. Climate Predators. | Parasites pe abebeue
AN SIN CUM OlISe mess se eee 1,839 49. 40 38. 48 3. 97 8.15 50. 60
Hanging squares. ..--...------ 5, 922 49. 01 19. 24 9. 80 21.70 50. 99
Wallen Poliss sess ees eee 941 70. 25 20. 08 5.95 Beal 29.75
Malienisquaresess sess aeas- 20, 844 57. 28 20. 30 15.18 7.15 42.72
Totals and averages.....| 29,546 55. 66 | 21.21 | 13.12 10. 00 44.34

¥ 4

Ne

SHARE OF INSECT CONTROL IN WEEVIL MORTALITY.

27

Following the plan adopted for the 1906 and 1907 records these
figures may be weighted for comparison with the earlier records.

TasLE XII.—The hypothetical or weighted average mortality of the boll weevil in 1908.)

~J
o
fo}
5
= Prolifer-
a. ation.
nN
mo) H
; : Bota Ines
Class of forms. <= E = <
tO hers =
Sp Oh eh ee
8 Spates onl es
a | 2 |g8| 2
iS) g S) E
5 Ses
ow Va eae 17,
Hanging bolls.......- 0.75 75| 6.30 4,7
Hanging squares.....-} 4.25) 425)13.50) 57.4
Total hanging..| 5.00) 500/..-.. 62. 1
Fallen bolls......-.---| 14.25} 1,425) 6.30) 90.0
Fallen squares...--.--- 80. 75} 8, 075/13. 50}1, 090. 1
Total fallen.....| 95.00| 9,500 .....|1,180.1
Totalsandaver-| - :
ALCS sense 100. 00/10, 000/12. 42/1, 242. 2

1908—Mortality from —

Climate. Predators. | Parasites. Total.

rs) od [so og |o oJ |S o

. HI} w| w| S =

I o = o 4 o =| Do. a

Sh tale ae ae tae eae

g | o eas o SN Dil esis 3)

a |3q . 86 : 88/2/88) 2

oO oO

8 Be Se Sas aie e FE E :

fa Ay Z Ay A Oy Za |e 4
70. 3/38. 48) 27.1] 3.97 1.4} 8.15} 5.7/51.80) 38.9
367. 6|19. 24) 70.7) 9.80) 36. 0/21. 70} 79. 8/57. 30} 243.9
437 Oe OF 28ers SMa ee Sou aes 282.8
1, 335. 0/20. 08] 268.1) 5.95) 79. 4.3.71) 49. 5 34. 10} 487.0
6, 984. 9]20. 30/1, 417. 9/15. 18/1, 060. 8) 7. 15/499. 4/50. 30/4, 067. 7
8,319. 9)..... 11, 686. (0) ese OM (ae ayae ARE O |e nen 4, 554. 7
Wes sities 17. 83/1, 783. 8/11. 77|1, 177. 1) 6. 34/634. 4/48. 37/4, 837. 5

Number of| Percentage

1 Given 10,000 weevil stages.

TasBLeE XIII.—Boll-weevil mortality in 1509.

This table shows a weighted increase of 2.51 per cent for parasites
and a weighted decrease of 12.24 per cent for all agencies, due to
the falling off in control by both climate and predators.
| 1909.—The mortality during 1909 was 41.73 per cent when figured
from the total number of stages, the total mortality thus showing
a decrease of 2.61 per cent from 1908.
also a decrease amounting to 4.68 per cent.

The parasitism showed

Percentage of stages killed by—

Class of forms. weevil of stages
stages. alive. Glamater
Hanging Wolisue encase oes 1,534 53. 33 37. 94
langine squares. 5.2222 4.22252 1,959 61. 16 12. 96
Mallen polls ee eee ree see ese 573 54. 82 27. 74
MAUEMISQUALCSS ae seme eo 7,587 58. 79 26. 58
Totals and averages_.... 11,653 58. 27 25. 84

Predators.

Parasites.

All agen-
cies.

a | | | | |

28 INSECT ENEMIES OF THE BOLL WEEVIL.

Following the plan adopted for the three preceding years these
figures may be weighted for comparison with the earlier records:

TaBLE XIV.—The hypothetical or weighted average mortality of the boll weevil in 1509.4

g 1909— Mortality from—
5
A Prolifer- |
ae saa Climate Predators. | Parasites. Total.
oo]
Class of forms. Pod sane : = ied) es Serta Slee :
Sa Ss ° z om z Sl. 8S OS PO mS o
2 a) = oo) Ss K 2 s = ) S ha © iS = Ub eee =
@ | 5 |#3| 5 | 3 l$8l 5 |28| = lf8l 5 lf] =
g 2 |98| 3 = |88| 2 188) 8 )eS) S eels
© § |S EI s |S8) @ tse tq jes as |
5 5 5 =} 3) Be 3 Be 5 BS | et) aes 5
a A |e Z ae a A a Z o ZA |S Pal
Hanging bolls. ----..- 0.75 79| 6.30 4.7; 70.3)/37.94| 26.7) 5.21 3.7) 3.52) 2. sle0 13) 87.6
Hanging squares......| 4.25) 425|13.50| 57.4] | 367.6|12. 96] 47.6) 6.38] 23. 4/19. 49] 71. 6/47. 29) 200.0
Total hanging..| 5.00} 500)...-- padde CSVee)eeee= (EBGi\eooee 27. ; ASES= | Ves liscase | 237.6
Fallen bolls...........| 14.25] 1,425] 6.30| 90. 0/1, 335. 0|27. 74| 370. 3/15. 00 200. 2| 2. 44] 32. 6/48. 63) 693. 1
Fallen squares.....--- 80. 75) 8, 075/13. 50/1, 090. 16; 984. 9] 26. 58/1, 856. 6/12. 39) 865. 4) 2. 24/156. 5)49. 14:3, 968. 6
Total fallen.....} 95.00] 9,500]... .|1,180. 1/8, 319. 9]... [2,226.9]... .|1, 065. 6....-|189.1|.....|4,661.7

Totals and aver- | | | | | | |
BLCS ses ass - 100. 00)10, 000)12. 42)1, 242. 2)....... 23. 01/2, 301. 2)10.992)1, 092. 7| 2. fe 2)48.
\

| | Pree |

1 Given 10,000 weevil stages.

This table shows a weighted decrease of 3.71 per cent for parasites
and. a weighted increase of 0.62 per cent for all agencies due to an
increase in climatic control. 3

In the following table is given a comparison of the weighted aver-
age control by all agencies for the four years.

TABLE XV.—Weighted average mortality of the boll weevil, 1906-1909.

Weighted average mortality due to—

Years. ;
ee Climate. | Predation. Parasites. AA Seen:
GO G ee ee ae sett eR et ee eas 12. 42 24.39 24.85 2.94 64. 61
OO peo ee ee eee ec rey eee ees 12. 42 28.16 16.18 3. 83 60. 61
OOS ees et ai a eS I kg eS 12. 42 17. 83 THT H/ 6. 34 48. 37
TE YS) Se ce a a aA el ita ee Ran ie 12. 42 23.01 10. 92 2.63 48.99

In view of the fact that certain cotton varieties retain the infested
squares more than others, it is interesting to make another hypothesis
on the basis that 50 per cent of the infested forms are hanging. The
year 1908 is chosen to illustrate this phase of the subject.

SHARE OF INSECT CONTROL IN WEEVIL MORTALITY. 29

TasLE XVI.—A hypothetical average mortality of the boll weevil in square-retaining

varveties .
3 1908. Mortality from—
=|
FI
asi Prolifera- : Preda- :
2 3 : ont Climate. cores Parasites. Total.
Classofforms. |_2| 2% |= — — — =
on wn lo) i] “ Onn i OM eH ou oe] Oo i)
Divx al ro) o bi ow rob) oe | oO ow Cob) ) o
BO alee oCiccall eects ree, Moores ll) doves si oo siemens reise | So eealiores
= Dein (Sires © q baats| ® |4g p=n| o |S i)
ql 2 |8s| 8s 3 f3a| 85 |8elS8Sl8a) 45 |85]| 8s
2° 5 ye gh the = EE om 2 e mrs £8 5H |o*| 5H
ov {<b}
at Z | Z fae} | fatbal | lr fates Maal) (septal (ial ew Z
Hanging bolls.......- 7.50) 750) 6.3) 47.2! 702.8/38.48) 270.4] 3.97] 27.9] 8.15) 57.3) 53.7) 402.8
Hanging squares....- 42.50) 4,250} 13.5} 573. 7/3, 676. 3)19. 24} 707.3] 9. 80/360. 3/21. 70) 797.8) 57. 3/2, 489.1
Total hanging. .| 50.00] 5,000}... .- 620. 9/4, 379.1]... .. OTT esac 38852 kame Sho qllleeeee 2, 841.9
Hallenbolishs sass: 7.50] 750| 6.3) ° 47.2) 702.8/20.08) 141.1) 5.95] 41.8] 3.71] 26.1) 34.1) 256.2
Fallen squares.......- 42.50) 4,250) 13.5) 573. 7/3, 676. 3/20. 30] 746. 3/15.18|558.1) 7.15) 262.8] 50.3/2, 140.9
Total fallen. ...-. 50. 00) 5,000}... --. 620. 9/4, 379.1)..... 887.4]. .... 599. 9]... -- 288: .9]eee ee 2,397.1
Totals and aver-
Ages ates ha 100. 00/10, 000)... - - 1 Z4ATN S| ose 18. 65}1, 865.1] 9. 88/988. 1]11. 44/1, 144. 0/52. 39/5, 239.0

1 Given 10,000 weevil stages.

This series of tables, wherein the mortality of the weevil is given
an accurate basis for comparison, brings to light some very important
points. This is especially the case in Table XVI, which is based
upon the hypothesis that 50 per cent of the infested forms are hang-
ing. By comparing this hypothesis for the year 1908 with the
table of the same year in which it is considered that only 5 per cent
of the forms are hanging, it will be noticed that under the condition
of the greatest proportion of hanging squares the total control of the
weevil would be 52.39 per cent and the number of parasites to 10,000
weevil stages would be 1,154; whereas, with the smaller proportion
of hanging forms, the total control of the weevil would be 48.37 per
cent and the total number of parasites 634 to 10,000 weevil stages.
Now this shows a gain of 4 per cent in the actual control of the
weevil and almost double the number of parasites to 10,000 weevil
stages. Naturally, under such conditions it would follow that the
parasitic control would be even higher than that which has been used
as a basis for the estimate and would increase in rapid proportion.
In view of this showing of the fact that the larger the proportion of
hanging squares to the entire amount of infested forms, the larger
the insect control becomes, we recommend that those who are inter-
ested in the breeding of cotton varieties attempt to secure varieties
of cotton which will combine the necessary qualities of productive-
ness, length of lint, and early maturing with the square-retaining
tendency. It may be pointed out that the varieties known as Rublee
and Cook’s Improved are not only conspicuous for the square-retain-
ing qualities but also for their desirability under boll-weevil condi-

380 INSECT ENEMIES OF THE BOLL WEEVIL.

tions. Several other varieties have been noticed to have this
same tendency, but they have not the other characteristics to recom-
mend them. In this connection we refer the reader to section 4
(p. 21),m which it has been shown that at least two States have had
a higher average control of the boll weevil in hanging squares than in
fallen squares when all of the records available are considered. It
will also be noticed in section 5, under Table XI, giving the actual
control of the boll weevil in 1908, that hanging squares and hanging
bolis were decidedly in the lead in the total control over either fallen
squares or fallen bolls. While this has not been the case in the other
years under consideration, we nevertheless consider that the pres-
ence of a nursery for the parasites in the field is most desirable.
Undoubtedly these hanging squares constitute such a nursery.

6. A STUDY OF HOW AGRICULTURE MODIFIES INSECT CONTROL.

From studies made during 1907 the following comparisons may
be made to show the number of factors that it is actually necessary
to consider in order that differences in parasitism may be understood.

At Arlington, Tex., records were kept on a field in the red loam
post-oak country or “‘cross timbers,’ another in the Trinity River
bottoms, and a third on the black waxy prairie. The first was
planted March 12, the second April 1, the third April 5. On August
28 the weevil infestation of squares in the timbers was 80.5 per cent,
in the bottoms 94.3 per cent, and on the prairie 21.4 per cent. At
the same time the parasitism in fallen squares on the timbers was 3.12
per cent, in the bottoms 1.9 per cent, and on the prairie 2.56 per cent.
In the timbers the parasitism of hanging squares was 39 per cent and
in the bottoms 24.78 per cent. The variable factors are soil, flora,
time of planting, variety of cotton, and weevil abundance. Hang-
ing squares were found in 1906 to be more highly parasitized in
timber land than on the prairie, and fallen squares inversely. There
appears to be an indication of the value of early planting. This
first field was the earliest field known in the vicinity and it showed
a high parasitism in hanging forms throughout the season.

At Calvert, Tex., were two fields on the prairie, one planted March

11 and 12, the other April i. On June 21 the weevil infestation of the
first was 18 per cent and of the second 21 per cent. On July 5 the
parasitism in the first was 2 per cent and in the second nothing.

At Denison, Tex., were two fields, one in the red clay, the other on
sandy loam, neither surrounded by timber. On the first the stalks
were burned February 28, on the second March 15. Both were
planted March 30. On August 27 the weevil infestation on the
first was 88.3 per cent, on the second 87.6 per cent; the parasitism in
fallen squares on the first was 6.31 per cent, on the second 2.85 per

5 (ete reer —r <a oo = ee eh ee te ee
i

HOW AGRICULTURE MODIFIES INSECT CONTROL. on

cent; the parasitism in hanging squares on the first was 5.79 per
cent, on the second 11.53 per cent. Here the only variable condi-
tions were soil, possibly weeds, and time of plant destruction. The
parasitism in the two classes of forms was diametrically reversed.

At Terrell, Tex., were two fields on the sandy prairie, both planted
in March, but having different weeds present. The weevil infesta-
tion August 26 on one was 65.2 per cent, on the other 97.5 per cent,
while the parasitism in hanging squares on the first’ was 29.5 per cent
and on the second 25.6 per cent. The variables were field surround-
ings and weevil abundance.

The unknown influence which entered most of these examples was
very probably the relative abundance of the different species of para-
sites. This may best be illustrated by the hanging squares from the
timbers and bottoms at Arlington, which are quoted above. In the
timbers the determinable parasites proved to be 16 Hurytoma tyloder-
matis, 10 Microbracon mellitor, 6 Cerambycobius cyaniceps, 5 Micro-
dontomerus anthonomi, and 3 Catolaccus spp. In the bottoms there
were 17 Cerambycobius cyaniceps, 13 Microdontomerus anthonomi, 10
Eurytoma tylodermatis, 8 Catolaccus spp., and 7 Microbracon mellitor.
The rank of the species was almost entirely reversed.

Probably the most important point in the entire set of examples:

is that the earliest crop had the most parasites. To show this in
another way we may refer to the conditions on the experimental farm
at Dallas. The first part of the field to put on squares was the first
part to show parasites. On July 8 infested squares were to be found
in six plats, but only on this earliest plat was there any parasitism—
5.7 per cent. On July 19 it and the adjacent plat were still consid-
erably in the lead.

That the earliest field should show the highest parasitism was
expected by the writers in view of the early spring observations.
The parasites in hibernation, whether on the boll weevil or on winter
cohosts, all reached maturity in the latter half of March at Dallas.
It was reasoned that cotton, squaring and attacked by April 15,
would get the hibernated parasites in any part of the State; that
cotton squaring and attacked by May 15 would get the first genera-
tion of parasites from the cohosts, and so on. It is reasonable to
expect that cotton with squares infested in season to attract hiber-
nated parasites or a new brood from cohosts will fare better than
cotton that commences squaring when all the parasites are concen-
trated upon neighboring cohosts. This cotton must wait until the
period of the favored cohosts begins to wane before the parasites will
begin to seek new scenes of activity. Although it was so reasoned,
it was hardly expected that there would be sufficient proof to warrant
voicing the proposition.

By INSECT ENEMIES OF THE BOLL WEEVIL.

A series of examinations was made in the vicinity of Victoria,
Tex., in 1907 and 1908. On October 9, 1907, Mr. Cushman noted
that fall destruction of the cotton was being carried on quite exten-
sively, but in different manners. On the east side of the river, south
and east of town, was an area in which practically all of the cotton
had been defoliated by the cotton leaf-worm. This area was sepa-
rated by the river and by a wide strip of huisache timber from
other cotton areas. In other directions were located fields stripped
by grazing, some that were plowed under, and one field only was
found which had received no treatment.

On June 17, 18, and 19, 1908, fallen squares from several of these
fields were examined, with the following results:

TaBLE XVII.—Boll-weevil mortality in various cotton fields, Victoria, Tex., 1908.

Percentage of mortality, 1908.
Treatment, 1907. ee Total.

; : Preda- :
Climate. eee Parasites.
Destroyedistalks; Septemlberweea= =e tee eee eee 314 | 18.18 5. 43 4.14 9. 23
POW EOMOCLO DCR eee a8 eee ate arn ened 296 | 13.80 ti 20) 3.00 3. 00
Plowed: December <s..5 2 ase = 58. sao eee eee Seen 354 | 60.70 14. 40 41.20 5. 08
Grazed October naatee So ae ae ec Ae eS 144 | 44.40 24. 30 15. 97 4.16
Os Senate A ari teers anne Mee neat a AE Sawa ots 290 | 37.50 25. 50 7.20 4.80
HINES Ue enti CAL ates a Ue eae dS tno Sor SE re aS 480 | 29.30 20. 60 2.50 6. 20
eee ae are Ee Se ete Ad RRL Mis HS 513. | 52.80 27.00 11.10 14. 40
TID) es ese foci oan Sees ee eS ates eae oe 375 | 23.70 16. 50 3.10 4.50

These striking differences in the percentage of control can not be -
attributed to the differences of treatment in 1907, although that
may have had a bearing. The different fields had different weeds
and plants surrounding them, they received different treatment in
the spring of 1908, and there are many other reasons why no one
basis of comparison can be chosen. The table is offered to illustrate
how wide a difference in natural control can be found in fields only a
few niiles apart and proves conclusively the value of individual effort
in the fight against the weevil.

Numerous other instances are contained in the notes that are quite
as striking as the one to which reference has been made. ‘There is
every reason why each planter should follow out as complete a pro-
gram against the weevil as he can, because each effort reduces the
total infestation of his neighborhood.

7. CLIMATIC CONSIDERATIONS.

The climate of the hibernating season of 1906-7 was very unusual,
so much so that the boll weevil hardly became quiescent, and the
emergence was largely during March, whereas normally it is in
April. The boll-weevil parasites mature simultaneously with the

CLIMATIC CONSIDERATIONS. oo

great wave of boll-weevil emergence. A glance at the accompanying
diagrams (figs. 4, 5) will-show that in Louisiana the monthly mean
temperature was from 3° Fahrenheit (November) to 10° Wanuary)
higher than the normal, and in Texas it varied from normal
(November) to 10° above normal (March) during the entire winter.
On the other hand, the accumulated moisture from November 1,
1906, to March 1, 1907, in Louisiana was 5 inches below normal and
in Texas t inch below normal.

Cotton was planted in March and April (1907) and normally would
have squared in May and June, but it was retarded a month by the
low temperature in April and May, during which months the monthly
mean temperature was 2° to 3° below normal in Louisiana and
3° to 6° below normal in Texas. In addition to the cold of the
spring, the precipitation in Louisiana from March 1 to July 1
was 7 inches above the normal and in Texas 2 inches above. This
cold and the presence of volunteer cotton tided the boll weevil over
until the planted cotton was up. The parasites were obliged to seek
cohosts from March 15 until late in MayorinJune. The cold, damp
weather undoubtedly retarded their development so that the first
generation was ready to attack such boll weevils as were breeding
late in May and early in June. As only a few fields held this advan-
tage to the parasites, these fields naturally became much better
stocked with parasites, as has been pointed out in another paragraph.

The summer and early fall months showed a slight deficiency in
rainfall and a slightly higher mean temperature—to such an extent,
however, that the season was considered dry, for the cotton did not
put on a very luxuriant foliage, and thus gave the sun plenty of play
on the fallen squares. The result is evidenced by the high percentage
of mortality from heat shown in the mortality tables. The increase
in parasitism may be ascribed to the same cause. |

The mean temperature of October, 1907, was normal in Texas, but
10° above normal in Louisiana. This warm season was followed by
a very sudden drop in temperature on November 11, the ‘‘norther’’
lasting until the 15th. This caused the November mean in both
States to be 3° below normal (Texas 53° F., Louisiana 56° F.). In
both States during this one month the precipitation was 3 inches
above the normal. In northern Texas about 30 per cent of the adult
weevils were killed by cold. The temperature at Dallas! reached
14° on November 13, which was 11° colder than was experienced in
1906 and 21° lower than at any time in November, 1905. The boll
weevils were not prepared for this cold, as they were still in great

1 The record was made both by the minimum thermometer and the self-registering thermograph at the

laboratory in East Dallas, and is a few degrees lower than the official record at Oak Cliff, about 5 miles
to the west and across the Trinity River.

16844°—Bull. 100—12——3

34 INSECT ENEMIES OF THE BOLL WEEVIL.

numbers on the plants and many immature stages were developing
in green squares and bolls.

Table XVIII gives the results of the examinations made immedi-
ately after the freeze.

TaBLE XVIII — Mortality of the boil weevil in Texas, November, 1907.

Mortality due to—

Condi-

Place. Date. Form. Location.) 4:5, | Stages.
1907
IDallaseeeeee eee Nov. 14] Squares...| Fallen. .| Green... 93
IDSs tene caae BE | SR (a eee (aS (0 Freee IDTyAe Sa 151
Brownwood....- Nov. 15 |-..do-. Plant. Green
Navasota...-.-.- Nov. 19 |...do. (a) dose (1)
Calvert... 252-5: Nov. 20 |...do.... do doze: (1)
Dallass. Sas ee Nov. 14} Bolis...... Fallen IDWAe ee 13
Ofna eee lose do.. ..do Plantes OM ze: 8
Brownwood....- Nov. 15 doses. eae domes Green a
INavasotas.2s--_- INOWe LON 2 dor fee sine doses. Mixed 56
Calwerte.nesso-o = Nov. 20 |...do doles douse 21
WACO Sas S eee Nov. 21 |...do doses dou 16
ISHOLTO meson eoee dons: do do.... HOknss-4- 27
Totals and Squares 3542 scare mela tacos ee 246
averages. \ Fa eh aod {pails CEE Ae SS Ee Se ates ielere 148
Totals and
AVCTAGCS Se cob ees LS eae a a are lal Le ne ie are eidesjotise Ate 394
1 Several.

2 Most of the death from “‘other causes”’ in bolls was due to proliferation, which seems to be stimulated
by frosts.

From this small number of stages no general statement can be made.
Of the 394 stages 55.3 per cent were killed by cold. Of the stages
in green squares or bolls, 98 per cent were killed by the cold.

The most interesting point is that although 98 out of 100 weevil
stages in green forms were killed, a parasite larva was found to have
just hatched from its egg on a weevil larva killed by cold. Three
other similar cases were found in dry forms. Seventeen cases of
parasitism were found on the 394 stages. Among these were two
living eggs of which one was an entirely new type and also two pup»
which proved to be Habrocytus piercet.

The remainder of the winter of 1907-8—that is, from December to
March 1—had a mean temperature a few degrees above the normal,
but with several severe cold spells. During the four winter months
the precipitation in both States was above the normal. The short
cold spells with warmer intervening weather and heavier rainfall
were disastrous to the boll weevil. The February examination to
ascertain the mortality of the weevil indicated about 98 per cent
mortality. Asaresult of the extreme scarcity of weevils in the spring
and summer in most parts of Texas, there was a great reduction in
the number of parasites. In fact, m the northern portion of the
Texas black prairie the parasites were forced to seek other hosts. A
killing freeze in November, 1908, again killed many boll weevils.

HOW INSECT CONTROL FOLLOWS WEEVIL. 35

Following the cold of November, 1908, the winter was unusually
warm, being at least 5° F. above the normal in both Louisiana and
Texas. From March 15 to July 15, in both States, the temperature
was almost normal. However, by this time there was an accumulated
deficiency of precipitation in each State of several inches. The
months of July and August in Texas were extremely warm and many
places recorded the maximum temperatures for their entire period
of records. While the heat was less excessive in Louisiana, it never-
theless reached very high pomts. This extreme weather quae these
two months had a tremendous effect upon the boll weevil and upon
its parasites, although records taken after some of the hottest days
showed that the mortality of the boll weevil from the heat was con-
siderably higher than the mortality of the parasites of the boll weevil.
After the middle of August a period of renewed growth of the cotton
plant gave the boll weevil an opportunity for increased development
and consequently permitted a large number of weevils to mature
before the hibernation season. Incidentally with this fall brood of
weevils, we find that there was a very great increase in the parasites,
especially in Louisiana. The following two diagrams (figs. 4, 5)
illustrate the temperature of the years under consideration.

8. HOW. INSECT CONTROL FOLLOWS THE DISPERSION OF THE BOLL
WEEVIL.

From an economic standpoint it is very important to know what
kind of natural control of the boll weevil can be expected in newly
invaded country. Since 1904 it has been noticed that maximum
infestation is generally reached by August 1, and that simultaneously
an extensive dispersion of the boll weevil takes place. At this period
the boll weevils fly to fields many miles beyond the parasites. The
climatic conditions during the dispersion period are such as will not
seriously interfere with prolific breeding of the weevils in the newly
infested territory. The extent of the dispersion is limited only by
the number of weevils flying and the amount of food supply available.
In the fall of 1909 the sparse production of cotton in southern Missis-
sippi brought about a dispersion of 120 miles into new territory. |

Our knowledge of the insects which attack the boll weevil shows
that most of them are derived from the parasites of similar weevils
that are native to the region infested. Therefore, if parasites and
predators are present in the invaded region, it is reasonable to expect
that they will immediately begin attacking the boll weevil. This
assumption has been proven in many definite cases. At Mimden, La.,
in 1906, a parasite larva was found in a green square infested by the
first generation. At Roxie, Miss., where the weevils had been present
only a few weeks in September, 1908, ant work and parasite work

10n

lated infestati

“T ‘BO6/-LOG! NOMLL MIF ea

Later in the season of 1908, an iso

>

INSECT ENEMIES OF THE BOLL WEEVIL.
was found at Roadside, in Yazoo County, Miss., about 40 miles beyond

were easily found.

36

>
K
:
:

wuees i
Roe NY

WOMLPLIA/ PLEA \OFLP

iN

ZA

VINAIOAT,

A,

but it was noticeable that the weevil

was parasitized in this particular field.

y]

the regular line of infestation

3

STATUS OF WEEVIL AND CONTROL BY INSECTS.

9. THE STATUS OF THE BOLL WEEVIL AND ITS CONTROL BY INSECTS.

During the seasons of 1908 and 1909 the examinations of the boll

weevil to determine its status

ated that there had been a

demonstr

XN

eae

§ S7TPOLIO'|\YIGWRULATS| LSNO/NS

7
fe

(
i i f
| {
(REALTOR joo
i
YY L
, 7 .
j
ti !

| Azar_|

oa al ae
: t = N ~

(‘TeUISTIO) “606 PUL ‘SO6T ‘L061 Ul SUOTZVIIVA O1YCUNT[O CU

eISINO'T SULISISHI[] WeIseIG—'S “DLT

ce

~~ Ny
SRR

SNS

a

O07 = +=
7=
TMIOIY TRVLIE FHS PNOISTIIO7 = :
III” TEWIYON FOHATAY UNPTSHIOI = >

ea ey

AYLTINGTS| AYLTINET | YPEWTIOTT | YTEWINON

a

ot N P, . ‘|

a ae \ i

-" t \ im | i}

. A {
. }.

fa) oO
Ks) w
| ST ee ier ee
AATHNISHES SFTFAOIA FAMECYSAWIL

R

ie
©

tremendous falling off of the weevil in all western and northern Texas.

In August, 1909, there was less than 10 per cent infestation in half of

38 INSECT ENEMIES OF THE BOLL WEEVIL.

Texas and in all of Oklahoma. At the same time a maximum infesta- ©

tion was found in all of that part of Louisiana lying south of the Red

River and in Mississippi for about 20 miles east of Natchez. Ananaly- —

sis of the parasite records for this same season shows that the parasite
control of the weevil in these sparsely infested regions of Texas was
very light, whereas the control in the heavily infested regions of south-
ern Louisiana and Mississippi was correspondingly very high. The
inference drawn from this observation is either that the boll weevil
had ceased to be the predominating weevil species for parasitic
attack in the lightly infested region, or that the parasites had been
destroyed by the heat. That the parasites were not all destroyed
by the heat is demonstrated by many records of the same parasites
on other species of weevils during the fall and winter of 1909.

10. A BRIEF STATEMENT OF THE VARIOUS CLASSES OF CONTROL EXER-
CISED UPON THE BOLL WEEVIL.

Before passing from this part of the report, which deals with the
general conditions obtaining, it is necessary to say a few words con-
cerning the classes of control which are of importance in repressing the

boll weevil. The first agency which is responsible for mortality of the © \

weevils is the resistance of the cotton plant to attack, evidenced either
by the toughness of the plant tissues which must be punctured, or by
the proliferation of the tissues, which destroys the weevil eggs and

larvee by crushing. When the infested form falls to the ground or.

withers on the plant it becomes immediately a subject for numerous
other factors of control. Intense heat kills many stages. A large
number of parasite species seek out infested squares for their prog-
eny; myriads of ants, beetles, and mites find nourishing food by
merely cutting their way into the infested forms and devouring the
weevil stages. In addition to these, sudden cold freezes countless
numbers of developing weevils. Neither are adults free from adverse
conditions. Many are killed by heat, or cold, or drowning; many are
picked up by birds and lizards or preyed upon by other insects; and
finally multitudes are starved on account of the ravages of other
insects upon their food supply. In this report we are able to deal
only with the three factors which are determinable in the control of
immature weevils, namely, climate, parasites, and predators.

11. PRACTICAL CONCLUSIONS DERIVED FROM STATISTICAL STUDIES.

The following conclusions of economic importance have been
reached from a study of this large series of statistics:

I. The month of August is the most important month for the con-

trol of the weevil by insect enemies. As this month is also the most

f

BIOLOGICAL COMPLEX. 39

important in the control affected by climate, it should be considered
as one of the most critical times of the year for controlling the boll
weevil. When a sudden drop in the temperature below freezing
occurs in the month of November before a large proportion of the
weevils has entered hibernation, and while many are still immature,
an excellent control of the species can be obtained. As, however,
this is only an occasional occurrence, it can not be relied upon and
every measure possible should have been carried out to prevent the
weevils from going into hibernation at all.

II. Hanging squares are the most important infested parts for the
work of parasites, and fallen squares in a similar degree for the work
of the predatory enemies. It has been demonstrated also that in
certain years the total control by all agencies is greater in hanging
squares than in fallen squares, and furthermore that in the more
humid States this condition is the prevalent one.

III. It has been shown by examples that the total mortality of
the weevil can be increased in proportion as the number of hanging
squares in a given area is increased and likewise that the pro-
portion of parasites to weevils is increased. It is therefore recom-
mended that plant breeders attempt to develop varieties of cotton
which will retain the squares, but will also have the other desirable
varietal characteristics Heceseary) for the production of an early
cotton crop.

IV. The insect control of the boll weevil is dependent in a large
measure upon the operations of the farm and for this reason all those
field practices which have been included in the system of cultural
control of the boll weevil are further recommended as tending to
increase the insect control.

PART II. BIOLOGICAL COMPLEX.

In Part I of this bulletin one set of facts, composed of statistics,
was dealt with, and it was merely hinted that the causes of these con-
ditions were very complex. In this part is presented another series
of facts, even more significant than the first, but much more difficult
to present in a tangible manner. The study of these biological
factors received its first impetus when at Clarendon, Tex., in 1905,
Mr. C. R. Jones and the senior author were fortunate enough to learn
the biologies of three species of weevils and to find that all these were
parasitized more or less abundantly by the same parasites as is the
boll weevil. It was already known that some of the parasites of the
boll weevil attacked other weevils, but the significance of this fact
had not been realized.

40 INSECT ENEMIES OF THE BOLL WEEVIL. |

With this simple beginning the search for other hosts of the boll-
weevil parasites was started and we have now built up the knowledge ;
of the following complex:
_ Owing to the complicated nature of the data to be presented in

this part, these have also been arranged in the following sections:
1. A list of the insect enemies of the boll weevil.
. The hosts of boll-weevil parasites.
. Mites which attack the boll weevil.
. Fhes which parasitize the boll weevil.
. The hymenopterous parasites of the boll weevil.
. Biological notes upon the parasites of the boll wee
. The development of the parasites.
. The distribution of the parasites.

9. The parasite seasons.

10. Adjustment to new hosts.

11. Beetles which prey upon the boll weevil.

12. Lepidopterous larve which are incidentally predatory upon
the boll weevil.

13. Ants which prey upon the boll weevil.

14, Biology of the cohosts of the boll-weevil parasites.

15. A list of the host plants of the cohost weevils.

16. A summary of the most important biological facts.

CON GO Or SP W LO

1. A LIST OF THE INSECT ENEMIES OF THE COTTON BOLL WEEVIL.

The boll weevil is known to be attacked by 29 species of parasites,
while 20 species of predators attack the immature stages and 6.
species of predators attack the adults. These species are listed as
follows:

Arachnida.
Acarina. Sarcoptoidea.
Tarsonemidee. Pediculoidine.
Pediculoides ventricosus Newport (parasite on larva), Mexico.
Pediculoides sp. (parasite on larva), Louisiana, Texas.
Tyroglyphide.
Tyroglyphus breviceps Banks (parasite on larva), Texas.
Insecta.
Orthoptera. Mantoidea.
Mantide.
Stagmomantis limbata Hahn (predator on adult), Texas.
Hemiptera-Heteroptera.
Reduviide. |
Apiomerus spissipes Say (predator on adult), Texas.
Coleoptera. Adephaga.
Carabide.
Evarthrus sodalis le Conte (predator on adult), Louisiana, Texas.
Evarthrus sp. (predator on adult), Louisiana.

A LIST OF THE INSECT ENEMIES.

Insecta—Continued.
Coleoptera. Polyphaga. Diversicornia.
Cantharide.
Chauliognathus spp. (predators on larva), Louisiana, Mississippi.
Cleride.
Hydnocera pallipennis Say (predator on larva), Texas.
Hydnocera pubescens Le Conte (predator on larva), Texas.
Cucujide.
Cathartus cassiz Reiche (predator on larva), Texas.
Lepidoptera. Bombycoidea.
Noctuidee.
Alabama argillacea Hiibner (defoliator, cuts off food supply).
Hymenoptera. Formicoidea.!
Dorylide. :
Eciton (Acamatus) commutatum Emery (predator on larva), Texas.
Poneride. .
Ectatomma tuberculatum Olivier (predator on adult) Guatemala.
Myrmicide. Cremastogasterine.

Cremastogaster lineolata (Say) var. clara Mayr (predator on larva) Texas.

Myrmicidee. Solenopsidine.

Solenopsis geminata (Fabricius) var. diabola Wheeler (predator on larva),

Louisiana, Mississippi, Texas.

Solenopsis molesta Say (=debilis Mayr) (predator on larva), Oklahoma.

Solenopsis texana Emery (predator on larva), Louisiana, Texas.
Myrmicide. Myrmicine.

Monomorium minimum Buckley (predator on larva), Louisiana, Mississippi,

Texas.

Monomorium pharaons Linneeus (predator on larva), Arkansas, Louisiana,

Oklahoma, Texas.
Pheidole sp. near flavens (predator on larva), Texas.
Pheidole crassicornis Emery (predator on larva), Texas.
Dolichoderide.
Forelius maccooki Forel (predator on larva), Texas.
Dorymyrmex pyramicus Roger (predator), Cuba.

Dorymyrmex pyramicus (Roger) var. flavus McCook (predator on larva), Texas.

Iridomyrmex analis André (predator on larva), Texas.
Formicide.

Formica fusca subpolita perpilosa Wheeler (predator on adult), Mexico.

Formica pallidi-fulva Latreille (predator on larva), Arkansas.
Prenolepis rmparis Say (predator on larva), Arkansas.
Hymenoptera. Chalcidoidea.
Chalcididee. Chalcidins. Smicrini.
Spilochalcis sp. (parasite), Texas.
Torymidze. Monodontomerine.
Microdontomerus anthonomi Crawford (parasite), Louisiana, Texas.
Eurytomide.

Eurytoma tylodermatis Ashmead (parasite), Arkansas, Louisiana, Mexico, Okla-

homa, Texas.
Bruchophagus herrere Ashmead (parasite), Mexico.
Eurytoma sp. (parasite), Texas.

1 All of these ants have been determined by Prof. William Morton Wheeler.

49 INSECT ENEMIES OF THE BOLL WEEVIL.

TInsecta—Continued.
Hymenoptera. Chalcidoidea—Continued.
Perilampide.
Perilampus sp.‘ (parasite), Louisiana.
Encyrtide. Eupelmine.
Cerambycobius cyaniceps Ashmead (parasite), Arkansas, Louisiana, Mississippi,
Oklahoma, Texas.
Cerambycobius cushmani Crawford (parasite), Texas.
Cerambycobius sp. (parasite), Mississippi.
Pteromalide. Pteromaline.
Catolaccus incertus Ashmead (parasite), United States.
Catolaccus hunteri Crawford (parasite), Louisiana, Mississippi, Mexico, Texas.
Habrocytus piercei Crawford, Louisiana, Texas.
Lariophagus texanus Genprard (parasite), Texas.
Eulophide. Tetrastichine..
Tetrastichus hunteri Crawford (parasite), Louisiana, Mississippi, Texas.
Hymenoptera. Ichneumonoidea.
Ichneumonide. Pimpline. Pimplini.
Pimpla sp. (parasite), Texas.
Braconide. Sigalphine.
Sigalphus curculionis Fitch (parasite), Louisiana, Mississippi, Texas.
Urosigalphus anthonomi Crawford (parasite), Texas.
Urosigalphus schwarzi Crawford (parasite), Guatemala.
Urosigalphus sp. (parasite), Texas.
Braconide. Braconine. Braconini.
Microbracon mellitor Say (parasite), Mexico, United States.
Braconide.
Unknown species (parasite), Texas.
Diptera. Cyclorrhapha.
Phoride.
Aphiochzxta nigriceps Loew (parasite), Texas.
Aphiochxta fasciata Fallen (parasite), Texas.
Aphiocheta pygmexa Zetterstedt (parasite), Texas.
Tachinide.
Myiophasia xnea Wiedemann (determined by Coquillett) (parasite), Texas.
Ennyomma globosa Townsend (parasite), Louisiana, Texas.

HYPERPARASITES. ”
Diptera.

Plastophora (Pseudacteon) crawford: Coquillett on Solenopsis geminata Fabricius.
2. THE HOSTS OF BOLL-WEEVIL PARASITES.

As has just been stated, the boll weevil has 55 species of insects,
which are known to attack it. Among the parasites are to be found
7 which are occasionally accidentally hyperparasitic. At least 1 par-
asite is known to attack one of the predators. The accidental preda-
tor (Alabama argillacea) is attacked by 12 parasites, 46 predators,

1 This species may be a parasite of a Chrysopa larva or of some lepidopteron which had entered a weevil
cell.

2 The enemies of Alabama argillacea Hiibner afford some interesting sidelights on the complexity of the
biological relations of cotton insects.

MITES WHICH ATTACK THE WEEVIL. ei 43

and 1 hyperparasite. Among these 46 predators are 6 which also
prey upon the boll weevil. At least 1 very common predatory
insect is known to prey upon many of the boll-weevil predators.
Fifty-five species of weevils are known to be attacked as cohosts of
26 species of parasites and of the 19 species of predators which attack
the boll weevil. These 55 species of weevils are known to breed
upon 91 species of plants, most of which are to be found in the vicinity
of the cotton fields. Three of these weevils sometimes breed upon the
cotton plant. Among the great number of parasites which attack
the 55 cohost weevils, 44 species are definitely known to science and
at least 6 species of hymenopterous parasites are known to attack
these 44 species of parasites. This complexity could be carried still
further, but probably enough has been stated to show how the many
influences of nature are dependent upon one another. The state-
ments are illustrated graphically in the accompanying diagram (fig. 6).
_ The principal point of importance in all of these facts is that the

boll weevil has been deriving its parasites from these 51 species of
weevils and from other weevils which are not known to us, and
there is every reason to believe that some of these other 44 species
of parasites, or still additional ones to be discovered, may be drawn
over to the boll weevil as parasites in the future. The weevils serv-
ing as cohosts and the parasites are listed in the accompanying table
(fig. 7) in such manner as to show the nature of the interrelationships.

It will be noticed from this table that 6 weevils, namely, Laria
sallzr, Larva exrgua, Smicraulaz tuberculatus, Anthonomus albopilosus,
Tyloderma foveolatum, and Trichobaris tecana each have 4 of the
boll-weevil parasites; 4 weevils are attacked by 3 of the parasites,
15 of the weevils by 2 parasites each, and the remaining 37 by only 1
parasite each.

Of the parasites, Cerambycobvus cyaniceps attacks 18 hosts, Hury-
toma tylodermatis attacks 16 hosts, Catolaccus wncertus 14, Catolaccus
hunter. 13, and Microbracon mellitor 12. These 5 parasites are also
regarded as the most important parasites attacking the boll weevil
itself. Perhaps this importance is due to the fact that they have
a larger number of native hosts and are hence in greater abundance
around the cotton fields than the parasites having fewer native
hosts. .

3. MITES WHICH ATTACK THE BOLL WEEVIL.
ACARINA. TARSONEMIDA.

_ The mites of the genus Pediculoides are assuming an important
réle among insect parasites, two species being accredited to the
boll weevil.

2 INSECT ENEMIES OF THE BOLL WEEVIL.

Pediculoides ventricosus Newport (fig. 8). This mite has been
somewhat prominent in the study of the boll weevil since its first
notice in 1901 (Rangel, 1901) under the name of Pediculoides ven-

THE BOLL WEEVIL COMFLEX.
THE COTTON PLANT

ENEMIES

/
HYVPERPARASITE

/ PREDATOR

| THESE NUMEROUS ENEMIES ALS
ARE KNOWN 70 ATTACK AT LEAST.

=a fee
See (OTHER
Bee WEEVILS

HVPER—
Yo FARAS/ITES
a N/T eee

OTHER (meee
PARASITES i

Fig. 6.—Diagram illustrating the boll-weevil complex. (Original.)

triculosus. Mr. Banks has stated that it may possibly be different
from the European species, but as it is known throughout this
country under the above name it is so quoted here. Mr. Rangel

PARASITES OF WEEVIL AND OTHER HOSTS.

45

e

S2LIS Ved 70 HEWON 74102 RINSNRSNINS SASS SSSR SSRIS SS SISSNE SA SSS ARES SN
Ae Kade Neo A Ba Ee Ba mS

nV HEH rN

IZA UV MNHIS SIIHATVHSO4f/) lst els ete | Sala
IWONOHLIVY STIHASTAPH/SOAf) Bie See seeeseeean BS
SINONTIIANS Sihitad TWAS SRE SESS seu Peeee aS
VAFPLNO STHIDILSVALTL Lanse See eee seeks aS
SNINGXIL SIOVHASAYA 7 foc. een seeseees is aS
ees i eect ie AL a Sd IRE IIRs PoE Eee pols Jes IES Ese ES BIS
ESS) te SOEBRE BS SBS INE eS himaialmsiaaies ESELEEE)
gad OC eee fay
ee ee a Ee BEEBE Se SSE Ber Ohe Eee
sleep a a mS

Apo A a K 6 CIN
suv/REOENE CELE be ee ie eel leis
LL ka | LN a EE ss 2 a a PN ss a Sa Ee TIA a CL
Pee ee ee ES ee Gs eee ae eee eer rN
hee a a 0 a 0 mG)
oe ooo) SE eee bee e ese e SSL ase Raeee sees [oul

Paere lish cece a a a mS
Sst Bl SE a a
semonony amoonde CEs
oe oie DERE eePSSGe SASS SSeS hSrshe beece ae RE SRS SsReenases
oe See SCE BeOS SU seas e ee Se seeS see hoe Gs fe i a et Ol FI
ep ee SS Beene a CESS neeC eeereseelesceoecoshasaeoes
CFs al eed ead ad oes ac a ah Vets fe a i oi [| ee a

NSS = x i)
% | Sq oR S K
wR S TSSHkG S Org OFS oS 2 9
% TTR CRASH. bk Saeed = = os RN

Qos SNe SS SITSTogSosess Seg CBko Be KF eha e
BS VY FY, SQoy TVS VSTHISGERSSSS RUSH SV sys 2 Sek SSu &
QeS gcEUSSTeye SPR ETSI SETS S S So tUSS IIR GMSHle SELSH sg

; AG) ~o S <
NET CERTLAEIS LNT IT RIDES EPEOPPDETE EEE RT LETTS: Redeg sees
GSLISSTYLHS LSS Hgusyksau SINAN UNS LSC PS TS OS

RAINING IPCI REINER SESS GSR SOC 2 SOS we
GE GFPSS»AsayHyellsE 2

SISSIES SS gS SIE aE ESSEC CO IEEE SSS SaESoSIESIE CSU GS aa gE SIRES ESI ESS ESS
SS SS SSSI T OTK TET HH GCR SFE SOVEVIGAS CS QyY
W LTTE T TESTS SSS TTS TT LTS RRR REESE RS OSS SS TSG TRS SRS LSE RS RK
CE XX SHSCTTTHE LISS SSS FSS SSS ssesgs VEN TR STERSSS 8S Sp

RSS SSS SSE SS SS RSS SE SSS SSS TSS Eg SS TTS 8 FS SS SGN SS TFET TTI S

(Original.)

Fie. 7.—Diagram giving the parasites of the boll weevil and their other hosts.

46 - INSECT ENEMIES OF THE BOLL WEEVIL.

pointed out that these mites reproduce viviparously and that their
offspring are mature and fertile at birth. He found that when they
attach themselves to a host the abdomen commences to inflate
until it becomes many times larger than the thorax. The time
required for engorgement varies from 2 to 5 days. When the abdo-
men commences to grow, the young commence to leave the parent.
The males fertilize the females before leaving the parent’s body and
shortly afterwards die. An average of 100 female offspring to an
individual was recorded. Rangel found that in 48 hours, 21 stages
out of 40 in squares were attacked by the mites. In larger series of
tests, after four days 50 out of 153 weevil stages in squares were
attacked, or 32.6 per cent.

Fic. 8.—Pediculoides ventricosus: a, Adult female before inflation of abdomen with eggs and young;
b, adult female after inflation of abdomen with eggs and young. Greatly enlarged. (Redrawn from
Brucker.)

Pediculoides n. sp. This mite was discovered in the laboratory at
Dallas, Tex., June 13, 1907, by the senior author. Careful observa-
tions on the length of generations were made with the following
results: A gravid female was isolated June 13 and on June 17 there
were 31 gravid mites. All but 5 were removed. On June 19 there
were many offspring, one mite being well grown. -On June 21 the
fourth generation began to appear. In other words, between June
13 and June 21, that is, in Jess than 8 days, there were two com-
plete generations. Another genealogy was as follows: Parent isolated
June 13, second generation began to appear June 14, were mature

me le =e — eu f

FLIES WHICH PARASITIZE THE WEEVIL. 47

June 17, and reproducing June 19. On June 24 the third generation
was reproducing. In this case there were 11 days covering two com-
plete generations.

The mites appeared willing to feed on any insect food available,
as they were first found feeding on stages of Trichobaris compacta,
then on boll-weevil stages, and finally on a Baris, on _ boll-weevil
parasites isolated in rearing tubes, and on Hydnocera pubescens.
They were reared readily on larve of Chlorion cyaneum and Polistes
rubiginosus. Mr. John B. Railsback, of Forbing, La., found that
they attacked the larve of the bollworm and other smooth cater-
pillars very readily.

TYROGLYPHIDA.

Tyroglyphus breviceps Banks was described as a weevil enemy |
from Victoria, Tex. This, or a similar mite, was found to be very
abundant at Calvert, Tex., in 1906. |

4, FLIES WHICH PARASITIZE THE BOLL WEEVIL.

Very few Diptera are known to be primarily parasitic upon boll
weevils, but the genera Myiophasia and Ennyomma in the Tachinide
seem to be confined to hosts of this nature. The genus Aphiocheta,
of the Phoride, contains at least 3 species which have been reared
under circumstances pointing to primary parasitism. The larve of
the tachinids work singly and those of Aphiocheeta several to a host,
but in both cases as endoparasites. When the former become full
erown they completely fill the skins of the weevil larve and fre-
quently the appendages of Myiophasia penetrate to the exterior.
The weevil skin partakes of the character of parchment and becomes
a cocoon within which the fly larva pupates and from which the
adult emerges. On the contrary the Aphiocheta larve leave the
host when they have reduced it to a shell and pupate in the weevil
cell. }

The flies evidently prefer to attack weevil stages in moist, shaded
spots in preference to sunny locations. By this habit they become
very valuable in fields located in bottom lands where the dry condi-
tions conducive to parasites like the hymenopterous parasites are
absent. The puparia of Myiophasia and Ennyomma are so near
like that of the chalcidoid internal parasite Tetrastichus hunter that
they can be differentiated only by the larger size of the dipterous
puparia.

PHORIDA.
a

Aphiocheta ngriceps Loew (determined by D. W. Coquillett).
Aphiochexta fasciata Fallen (determined by D. W. Coquillett).
Aphiochxta pygmexa Zetterstedt (determined by D. W. Coquillett).

48 INSECT ENEMIES OF THE BOLL WEEVIL.

On September 12, 1906, in dry hanging bolls collected at Dallas,
Tex., a weevil larva was found parasitized and isolated in a separate
tube, with the following record: ‘‘ Very small parasite larva on small
weevil larva.’’ On September 26 a single specimen of Aphiochexta
mgriceps Loew was reared and the following note made by the
senior author: ‘‘ Found dipterous puparium, skin of hairy para-
site larva (may be the dipteron or a hymenopteron); also remains
of weevil larva.’ On October 6, 1906, in hanging bolls collected
at Dallas a weevil larva was isolated with the note, ‘‘Weevil larva
full of dipterous larve.’’ Eleven larve left this host larva and
. pupated. On October 29, 7 Aphiochxta (%) fasciata Fallen and two
A. pygmea Zetterstedt were reared. Atleast the latter case seems to
be very strong evidence of primary parasitism. These flies are reared
frequently from bolls and many are perhaps scavengers. Two other
species, A. epeire Brues and A. scalaris Loew, have also been reared
from cotton forms at Calvert, Tex. Records made in 1911, at Tal-
lulah, La., by Mr. Harry Pinkus, point conclusively to primary
parasitism.

TACHINIDA.

Myiophasia xnea Wiedemann is recorded from a number of very
important weevils. Of these, Balaninus nasicus Say (the acorn
weevil), Conotrachelus juglandis LeConte (the walnut weevil), Ampe-
loglypter sesostris LeConte (the grapevine gall-maker), Conotrachelus
afinis Boheman (the hickory-nut weevil), and Conotrachelus elegans
Say (the pecan-gall weevil) are all weevils which enter the ground for
pupation, carrying their parasites with them, and consequently it
becomes necessary for the flies to emerge from the weevil cell through
several inches of earth before attaining freedom. The only weevils
which this fly attacks and which do not enter the ground for pupation
are the boll weevil and Trichobaris compacta Casey (the Jamestown-
weed pod weevil). Very few records have been made to ascertain
its developmental period, but the three records at hand indicate from
22 to 29 days as the period from collection of the infested material
to the maturity of the fly. This period would cover largely the under-
ground period only.

Ennyomma (Loewia) globosa Townsend. Several specimens of
this fly were reared during 1907 by Mr. C. R. Jones at Alexandria,
La., as primary parasites of the boll weevil. It is a very common
parasite of Chalcodermus enews Boheman (the cowpea-pod weevil)
in the Southern States.

}o. THE HYMENOPTEROUS PARASITES OF THE BOLL WEEVIL.

So much information has been gained concerning the hymenopter-
ous parasites of the boll weevil that it will be necessary to omit
many of the technical facts learned about them. The present sec-

HYMENOPTEROUS PARASITES OF THE WEEVIL. 49

tion is concerned with the sources of the parasites and with important
records of their occurrence, while the other interesting facts to be
presented are included in the five following sections:

CHALCIDOIDEA. CHALCIDIDH. CHALCIDINZ. SMICRINI.

Spilochalcis sp. A single male of this species was found dead in a
weevil cell with the remains of the weevil and its own exuvium in a
hanging square collected August 10, 1907, at Victoria, Tex.

TORYMIDH. MONODONTOMERINA.

Microdontomerus anthonom Crawford (fig. 9). Brachytarsus alter-
natus Say was formerly the only weevil recorded as a host of this
species. A male and female of this parasite were reared on Septem-

Fic. 9.— Microdontomerus anthonomi: Adult. Much enlarged. (Original.)

ber 12, 1907, from pods of the flowering shrub Amorpha fruticosa, at
Dallas, Tex., which were highly infested by (Bruchus) Laria exigua
Horn. In 1906, in which year this parasite was first discovered, it
ranked as seventh species in importance as a boll-weevil enemy.
In 1907 it advanced to fourth place and was very important in the
central black-prairie region of Texas. In 1909 the easternmost
limit of our records was Tallulah, La., which did not become infested
by the boll weevil until 1908.

EURYTOMID®. EURYTOMINI.

Eurytoma tylodermatis Ashmead (Bruchophagus herrere Ashmead).
This is without doubt one of the most important species under
consideration, having a range of distribution practically coexten-

16844°—Bull. 100—12——4

50 INSECT ENEMIES OF THE BOLL WEEVIL.

sive with that of its host, the boll weevil. In previous publi-
cations it has been recorded as a parasite of Lizus musculus Say, L.
scrobicollis Boheman, Apion segnipes Say, Anthonomus heterothece.
Pierce, Anthonomus squamosus LeConte, Tyloderma foveolatum Say,
and Orthoris crotchit LeConte. To this list may be added (Bruchus)
Laria exigua Horn in Amorpha pods at Dallas, Tex.; (Bruchus)
Laria salle. Sharp in pods of huisache ( Vachellia farnesiana) col-
lected at Victoria, Tex.; Spermophagus robiniz Schénherr in pods of
the honey and water locusts (Gleditsia triacanthos and G. aquatica)
in Louisiana; Macrorhoptus spheralcie Pierce in pods of Spheralcia
at Del Rio, Tex.; Smicraulax tuberculatus Pierce in mistletoe stems
at Dallas, Tex.; Trichobaris texana LeConte in stems of Solanum
rostratum at Cisco and Victoria, Tex.; and also Trichobaris trinotata
Say; and finally it was reared in September, 1908, from Baris sp* in
roots of ambrosia at Camden, Ark., by C. E. Hood.

Eurytoma sp. A female primary parasite and a male accidentally
secondary on Microbracon mellitor Say were reared from hanging
squares collected August 10, 1907, at Victoria, Tex.

PERILAMPIDA.

Perilampus sp. A single individual was reared from the boll weevil
in an isolated weevil cell by C. E. Hood from squares collected Sep-
tember 7, 1907, at Shreveport, La. Another specimen of Perilampus
was reared from squares collected at Granbury, Tex., August 8, 1907,
but its source could not be proved. If it were not for the definite
record made by Mr. Hood these species could hardly be placed in
this list. This record may possibly be based upon the parasite of an
intruder in the weevil cell instead of upon the weevil itself. It is
not impossible that after several years this parasite may be found
normally as a boll-weevil parasite, as was the case with Sigalphus
curculionis.

ENCYRTIDH. EUPELMINZ.

Cerambycobius cyaniceps Ashmead. The list of hosts of this spe-
cles as previously recorded included Anthonomus albopilosus Dietz,
(Bruchus) Laria obtecta Say, L. exigua Horn, Lizus musculus Say,
Trichobaris texana LeConte, and Tyloderma foveolatum Say. To this
list may be added Laria bisignata Horn in pods of Acuan allinoensis;
L. ochracea Schaeffer in pods of Vicia sp.; L. sallei Sharp in pods of
huisache ( Vachellia farnesiana); Spermophagus robvme Schénherr in
pods of Gleditsia triacanthos at Alexandria, La.; Lixus secrobicollis
Boheman in stems of Ambrosia trifida and A. psilostachya; Apion
rostrum Say in pods of Baptisia tinctoria at Washington, D. C.;
Tachypterellus quadrigibbus Say in fruit of Crategus mollis at Vic-
toria, Tex.; Smicraulaz tuberculatus Pierce in stems of mistletoe

HYMENOPTEROUS PARASITES OF THE WEEVIL. 51

(Phoradendron flavescens); Tychius sordidus LeConte in Baptisia
pods at College Station, Tex.; Trichobaris compacta Casey in pods of
Datura stramonum at Paris, Tex.; and, furthermore, on Languria
sp. in stems of Gaura sp. at Ballinger, Tex.

Cerambycobwus cushmant Crawford. This parasite was reared in
small numbers as early as 1906 in southern Texas from the boll
weevil. It has been reared from Laria ochracea Schaeffer in pods of
Vicia sp.; from L. sallzei Sharp in pods of Vachellia farnesiana; from
Arecerus fasciculatus DeGeer in fruit of chinaberries ( Melia azeda-
rach) at Victoria, Tex.; and from Trichobaris texana in stems of
Solanum rostratum.

Cerambycobwus sp. On February 23, 1909, a male of a green species
of Cerambycobius was reared from the weevil in squares collected at
Natchez, Miss., January 19.

PTEROMALIDH. PTEROMALINA.

Catolaccus hunter Crawford. This is the species which in all pre-
vious articles on the boll weevil has been known as Catolaccus incertus.
Its hosts as now known are Laria compressicornis Schaeffer in pods of
Acuan illinoensis; Tachypterellus quadrigibbus Say in fruit of Crategus
spp.; Smicraulaz tuberculatus Pierce in stems of Phoradendron flaves-
cens; Anthonomus xeneolus Dietz in buds of Solanum spp.; Anthono-
mus albopilosus Dietz in seeds of Croton sp.; Anthonomus eugenir
Cano in fruit of pepper (Capsicum spp.); Anthonomus heterothece
Pierce in heads of Heterotheca subaaillaris; Anthonomus nebulosus
Le Conte in buds of Cratzgus spp.; Anthonomus signatus Say in buds
of dewberry (Rubus spp.); Anthonomus squamosus Le Conte in heads
of Grindelia squarrosa nuda; (Acalles) Gersteckeria nobilis Le Conte in
joints of Opuntia spp.; Zygobaris xanthoxyl Pierce in berries of Xan-
thorylum clava-herculis. 7

Catolaccus incertus Ashmead. This parasite is also very common
and is known from a considerable number of hosts, among which are
Laria exigua Le Conte in pods of Amorpha fruticosa; Apion decolora-
tum Smith in pods of Meibomia paniculata; Apion grisewm Smith in
pods of Phaseolus spp.; Apion nigrum Smith in buds of Robinia pseu-
dacacia; Anthonomus albopilosus Dietz in seeds of Croton spp.; Antho-
nomus aphanostephi Pierce in heads of Aphanostephus skwrrobasis;
Anthonomus fulvus Le Conte in buds of Callirrhoe involucrata; Antho-
nomus nigrinus Boheman in buds of Solanum carolinense; Anthonomus
signatus Say in buds of strawberry (fragaria virguniana); Auleutes
tenuipes Dietz in buds of Galpinsia hartweqi; Ceutorhynchus n. sp. in
crown of Selenia aurea; Baris cuneipennis Casey in roots of Helenvwm
tenuifolium and Calandra oryza Linneeus in corn.

52 _ INSECT ENEMIES OF THE BOLL WEEVIL.

Habrocytus pierce. Crawford. This is a brilliant green parasite
resembling Catolaccus anthonomi Ashmead. It is reared from the boll
weevil mainly in the fall and from hibernated individuals in the spring.
It has been reared from Laria compressicornis Schaeffer in pods of
Acuan illinoensis. (See fig. 10.)

Lariophagus texcanus Crawford. There is every evidence that this
species is a true parasite of the boll weevil, although it has not been
positively reared by isolation from the boll weevil. On August 17
and 27, 1907, two specimens were reared from material collected at
Hallettevills, Tex., August 13; on August 19 and 23 three specimens
were reared from cotton squares collected at Victoria, Tex.; on August
29 three specimens were reared from squares which were collected at
Eagle Lake, Tex., on August 14. The Victoria lot was peculiar in that
it furnished the first records of Cerambycobius cushmani Crawford,
Spilochaleis sp., and Eurytoma n. sp. This species is described as a

parasite of (Bruchus) Laria prosopis Le Conte. It

undoubtedly also attacks L. sallext Sharp, which
also breeds in the pods of huisache; furthermore,
the species was reared from stem galls of Leucosyris
spinosus containing Anthonomus ligatus Dietz.
Tetrastichus hunteri Crawford. This interesting
new parasite of the boll weevil was first reared in

of the boll weevil collected at Natchez, Miss., by
H.S. Smith. It is internal in weevil larve and
pup and has even been reared from immature
: adults. A parasitized individual can easily be told
Fig.10-—Habrocytus by its brownish color and smoothening of the vari-

piercei: Pupa. Much oys segmental wrinkles. In more advanced stages

eee a 22 of Ee parasite’s development, the parasitized in-
dividual becomes a mere brown skin of parchment. This skin serves
as puparium for the parasite. The developmental period is of con-
siderable length in the fall. Specimens isolated in November do
not mature until April or May. In 1908 it was found only at Natchez,
Miss., and Monroe, La., but in 1909 it was reared at a number of places

in Louisiana and also at Arlington, Tex. This species gives an ex-

cellent example of the adjustment of native parasites to the boll

weevil.
ICHNEUMONOIDEA. ICHNEUMONID2. PIMPLINZA.

Pimpla sp. On January 27, 1909, a larva of this species was

isolated from a weevil larva in squares collected at Nacogdoches, Tex.
This became a mature female on February 23.

the fall of 1908 from isolated parasitized individuals *

:
h
ee
z

————

HYMENOPTEROUS PARASITES OF THE WEEVIL.

BRACONID. SIGALPHINA.

Sigalphus curculionis Fitch. Previous to the summer of 1908 the
first record of rearing this species (see fig. 11) from the boll weevil was
considered doubtful, but beginning in August it was reared repeatedly
in material from Ruston and Monroe, La., and Natchez, Miss. Its
other hosts are Conotrachelus afinis Boheman in hickory nuts; Cono-
trachelus elegans Boheman in petioles of hickory at Dallas, Tex., and
in galls of Phyllozera devastatriz on pecan (Hicoria pecan) at Dallas
and Victoria, Tex.; Conotrachelus juglandis Le Conte in walnuts (Jug-
lans nigra); Oonotrachelus nenuphar Herbst in fruit of plum, peach,
etc.; Tyloderma foveolatum Say in stems of Onagra bienms at Wash-
ington, D.C.; Trichobaris texana Le Conte in stems of Solanum rostra-
tum; Trichobaris trinotata
Say in stems of potato (Sol-
anum tuberosum); and Zygo-
baris zanthosyli Pierce in
seed of Xanthoxylum clava-
herculis.

Urosigalphus anthonome
Crawford has never been
reared since the original
records which were made \
at Brownsville, Tex. Fig. 11.—Sigalphus. curculionis: a, Male; b, female; c, an-

Urosigalphus ee Wena tenna. Allenlarged. (After Riley.)
Crawford. This Guatemalan boll weevil parasite has never been
reared in the United States.

Urosigalphus n. sp. At Arlington, Tex., in 1909, a single specimen
was reared from an isolated cocoon.

BRACONINA.

Microbracon mellitor Say.1 This parasite (see fig. 12) still holds
the lead as the most important boll-weevil parasite. Its other
host relations are only partially discovered. The following hosts
have been ascertained: Desmoris scapalis Le Conte in heads
of Sideranthus rubiginosus; Smicraulax tuberculatus Pierce in
stems of Phoradendron flavescens; Anthonomus albopilosus Dietz in
seed of Croton spp.; Anthonomus eugenit Cano in fruit of pepper

1 Bracon mellitor Say is recorded by Girault (1907) as a parasite of the lesser peach borer (Synanthedon pic-

tipes Grote and Robinson) and of the peach borer (Sanninoidea exitiosa Say). The gregarious habit of these
parasites appears to prove that the determination wasincorrect. Mr. F.E. Brooks, of West Virginia, has fur-

- nished the record of this species from Sanninoidea exitiosa and also from Craponius inequalis Say at French

Creek, W. Va. The determinations were made in the Bureau of Entomology. Dr. F. H. Chittenden states
that he reared this species from the strawberry leaf-roller, Ancylis comptana Froelich (fragariz Walsh and
Riley), at Cabin John, Md., July 9, 1899. It is probable that all parasites of Lepidoptera determined as
Bracon mellitor belong to some other species. The lepidopterous and coleopterous parasites are not dis-
tinguishable by structural characters, but are so different in habits that it is considered advisable to
call the lepidopterous parasite Microbracon dorsaior Say and the coleopterous parasite M. mellitor Say.

54 INSECT ENEMIES OF THE BOLL WEEVIL.

(Capsicum spp.); Anthonomus fulvus Le Conte in buds of Callirrhoe
involucrata; Anthonomus squamosus Le Conte in heads of Grindeha
squarrosa nuda; Conotrachelus nenuphar Herbst in peaches; T'yloderma
foveolatum Say in stems of Onagra biennis; Craponius inequalis Say in
fruit of grape (Vitis spp.), and Baris sp. in roots of Ambrosia sp.

6. BIOLOGICAL NOTES UPON THE PARASITES OF THE WEEVIL.

A number of very interesting facts, which deserve mention in an
economic bulletin, have been learned about the biology of the parasites.

ABUNDANCE OF PARASITES.

It is unusual for the parasites of the boll weevil to be found flying
in numbers. Their work is in a general way quietly and unostenta-
tiously done, but occasionally it is the privilege of the observers to
see swarms of parasites hovering around the food plant of their
favorite host. At Clar-
endon, Tex., in August
and September, 1905, Mr.
C. R. Jones and the sen-
ior author witnessed large
numbers of Microbracon
nuperus and Tetrastichus
sp. hovering around the
highly infested pods of
Mentzelia nuda. Thepar-
asitism in pods gathered
at this time was so high
that much superparasit-
ism by Tetrastichus upon
theMicrobracon occurred.

At Ruston, La., in Octo-

Fig. 12.— Microbracon mellitor; Adult. Muchenlarged. (F +
ged. (From ber, 1907, the sentor
Hunter and Hinds.)

author saw Catolaccus
huntert flying in all directions and resting on the flowers and
leaves of Heterotheca subaxillaris and found a very high parasitism
of Anthonomus heterothece by this species. In November, 1908,
in this same field at Ruston, a very high percentage of parasitism
of the boll weevil by Catolaccus was found. In September, 1908,
Mr. Hood saw many species of parasites around the flower heads
of Vernonia at Camden, Ark. During the same month Mr. H.S.
Smith found Catolaccus huntert swarming on Croton capitatus contain-
ing Anthonomus albopilosus. Such observations are very important
because they suggest excellent sources for parasites to be used in
introduction experiments or suggest forcing of the parasites to the
boll weevil by the elimination of the host.

me i

BIOLOGICAL NOTES ON THE PARASITES. 55
FREQUENTATION OF NECTARIES.

The feeding habits of adult hymenopterous parasites have long
escaped observation, but within recent years the intensive study of
parasites has proved that very little can be accomplished in the prop-
agation of parasites unless they can be fed. In the case of the para-
sites of the boll weevil it is impossible for the adults to obtain nour-
ishment from the host in which they are ovipositing, as has been
proven in the case of parasites of externally feeding insects. The
host plant of the boll weevil, however, furnishes the desired food.
The nectaries of cotton are about as plentiful as those of any other
plant. The majority of varieties of cotton have three large nectaries
on the leaves and also have them on the outside and inside of the
involucre, as well as on the inside of the flower. Frequent observa-
tions of cotton plants which were producing considerable nectar have
enabled us to observe practically all of the parasites of the boll
weevil, as well as all of the ant enemies and many other insects.
Some of these insects which visit the nectaries are injurious to the
cotton plant, but the majority seem to be beneficial.

The quantity of nectar secreted by various varieties of cotton is

quite variable. The variety which seems to secrete more than any
other which has been observed is the Egyptian Mit Afifi. This variety
is frequently surrounded by large numbers of beneficial hymenop-
terous insects, although at the same time it appears to be very sus-
ceptible to boll weevil attack.

HELIOTROPISM.

The majority of the hymenopterous insects which have been under
observation in this investigation appear to be positively heliotropic.
In general this tendency can be utilized in rearing-cage technique to
induce the parasites to go into small tubes placed in the rearing boxes,
from which they can be easily removed. It has been noticed in
rearing cages in which there were growing plants with plenty of food,
air, and heat, that the parasites sought the lightest portion of the
cage rather than the plants which could give them some shade from
the hot sun.

The activity of the parasites is greatest when the sunlight is most
intense. . Observations at the nectaries of the Egyptian cotton con-
firmed this. When the sun was shining the parasites were very
active at the nectaries and flying around the plants, but when a
cloud passed over they seemed to disappear entirely. On cloudy
days none of the Hymenoptera, except the most industrious bees and
wasps, was to be found at the nectar. Trelease (1879) states that
“the extrafloral nectar of the cotton plant is far more abundant
during night and in the early morning than at any other time, and

Se

56 INSECT ENEMIES OF THE BOLL WEEVIL.

this is true whether we consider the involucral or foliar glands.’”’ The
parasites probably frequent the nectaries during the morning sunlight
hours and then are equipped to go about their other duties during the
hottest part of the day.

In addition to these actual observations as to the preference of
parasites there are other very strong proofs of heliotropism. It has
been found that there is a decided increase in the parasitism of weevil
stages in hanging forms exposed to the sun over those in fallen forms
which are more or less shaded. It is also apparent that the fallen
forms most exposed to the sun receive the greater amount of para-
sitism. Among the hymenopterous parasites there is only one at
present which seems to prefer a moist shady place for its work.
This is Tetrastichus hunteri Crawford, which is an internal parasite.

SEXES.

A numerical study of the records of rearing of parasites from the
boll weevil shows that in the majority of the species the males are
relatively fewer than the females. The following table will show the
percentage of each sex and also the number of parasites upon which
these percentages are based.

TaBLeE XIX.—Relative percentages of the sexes of boll-weevil parasites.

Percentage of sexes. .

: Total in-
Species. dividuals.
Female. | Male.
Per cent. | Per cent.
SMA CrOUTACOR AN ELUILOT see sa ee on eee ee ae en ee eee 980 60. 78 39. 22
COLGLACCUS TATU CT a ee ee OI NS Ss = aS a ee ct ee os 809 78. 37 21. 63
COLOLOCCUSANCENI US ee Seen eerie 2a eee eee oars ee ae ee a eee eee 429 81.12 18. 88
ST GULOCUTUS UENCE = ee Rae = ON ee ee Sh ed A ee 30 100, 00-1 Sete
CET ATROYCOUEUS EC USO TEI oo oe ee ae ee ee es Sr eS 64 | 71. 88 28. 12
Cen@mby Cours CYANICEDS 2. SS Sse As Se see ee eae + ees 509 70. 34 29. 66
COMING Y COPEUSTS ee ae ae ae me ee ae oN ee a ee ee 1 0 100. 00
UNYOM IMG QUOOOSG ee ek Se ars See ee ee Sera ee ee wee ee 8. = 37550 62. 50
LE AET SLO NLC YEOGENTIELES Ato oe se ENS oo oT an a ee ee 433 64. 90 35.10
PETA LOTING SP ace a ee aay eos AE So tO SS Oe ee Ses es 2 50. 00 50. 00
EATIOPRAGUSACLONAUS Banners eS ee Oe ee Ee eee 2 50. 00 50. 00
MicrodOnsOM ChUS QTURONOTN ao Oa PA eo See ae ae ee ee ee 223 84. 76 15. 24
MiayiODNOSUO PTE Sey ee eee ree See ie ree ee ree ee ee 2 50. 00 50. 00
SAGA DRUS CUTCU OTIS eee A Ns Gee Sete ee ge pee aera ee = 13 61. 54 38. 46.
VLE USTICI US MPU TUE CT Ei oe oe Ie ee Rea ee ene IIE ee 41 100.0041: 25 es

OVIPOSITION.

Tt has been found by numerical study of the large number of para-
sites collected during the last five years that whenever the parasitism
in a field reaches between 50 and 70 per cent there is a strong likeli-
hood of reduplication, with resulting superparasitism. The exact
records of superparasitism obtained in this investigation have been
published in another article (Pierce, 1910). Parasites have no power
of discerning the presence of another egg on the prospective hosts,
and hence there occurs at times a tremendous duplication of energies.

Bul

100, Bureau of Entomology, U. S. Dept. of Agriculture PLATE Il.

ROC LE
Se VUE YAZ IZ

LVS NAQS
ae, WNENGNS ERS >
YhIT 11S SR
LA Le NNN ONS
EL J 1SNSSSN SSS
PLIST ILS SENN SS

POP RI ead STC RCD peor

IDO Oe a OOO EPS

Ea@s OF BOLL-WEEVIL PARASITES.

Fig. 1.—Type Il. Microdontomerus anthonomi; Calvert, Tex., August 23, 1907; color white;

size 0.38 by 0.11 mm. Fig. 2.—Type VI. Unidentified egg; Dallas, Tex., November 14,
1907, color white; size 0.85 by 0.19mm. Fig. 3.—Type I. Cerambycobius cyaniceps; 3a, View
from side; 3b, view from end; color white; size about 0.8 mm. Fig. 4.—Type Ill. Eury-
toma tylodermatis; Dallas, Tex., August 22, 1907; color gray; size 0.68 by 0.21 mm.; 4a, side
view of another egg. Fig.5.—Type lV. Catolaccus hunteri; Dallas, Tex., August 22, 1907;
color white; size 0.62 by 0.22 mm. Fig. 6.—Type V. Unidentified egg; Glenmora, La.,
August 23, 1907; color gray; size 0.44 by 0.11 mm. (Original. )

eR ROmD 5

Bul. 100, Bureau of Entomology, U. S. Dept. of Agriculture. PLATE III.

we

ro

PARASITES OF WEEVILS.

Fig. 1.—Eurytoma tylodermatis, pupa. Fig. 2.—Catolaccus incertus, pupa. Fig. 3.—Ceram-
bycobius cyaniceps, pupa. Fig. 4.—Microdontomerus an thonomi, pupa. Fig. 5.—Larva of
microbracon. Fig. 6.—Microbracon mellitor, pupa. Fig.7.—Larva of chalcidoid. Much
enlarged. (From Pierce.)

einen are DAR,” BY

DEVELOPMENT OF THE PARASITES. 57

It may therefore be possible that a parasite will visit the s same square
several times and oviposit. In general it may be said that as the
primary parasitism of the boll weevil increases the superparasitism
also increases, with the result that sometimes the parasitism might
be considerably increased if every egg reached a single host. The
following instances will illustrate this. At Calvert, Tex., 41 stages
were attacked by 44 parasites, although only 36.5 per cent of the
weevils were parasitized. If every parasite egg had reached a host,
there would have been 107.3 per cent parasitism. At Dallas, Tex.,
out of 309 weevil stages, 44.6 per cent were attacked by 216 parasites.
The possible parasitism was 69.9 per cent. Many other instances of
this kind could be given, but these two cases illustrate the condition
perfectly as it exists in many places during the fall of each year.

The time for oviposition apparently differs for the various species.
Microbracon mellitor, as a rule, oviposits before the boll-weevil larva
has constructed a cell, that is, several days before the flared square falls
or dries. Eurytoma tylodermatis appears to oviposit in squares on the
plant after the normal time of falling and hence is more important in
hanging dry squares. Catolaccus spp. and Microdontomerus antho-
nomi favor fallen forms for oviposition. The chalcids generally
oviposit after the weevil larva has formed its cell. Tetrastichus
huntert is most frequently found in fallen squares.

7. THE DEVELOPMENT OF THE PARASITES.
THE EGGS.

The eggs of the boll-weevil parasites are all oblong-elliptic and
either smooth or sculptured. The eggs of several species have at
one or both ends a small tube which is tied into a knot. Six types of
eges of the boll-weevil parasites have been closely observed and
designated by number in the records of rearing. These are illus-
trated on Plate II. The eggs of all the boll-weevil parasites are
placed in the weevil cell or on the larva or pupa and usually without
injuring the latter.

Type I—Type I is the egg of Cerambycobius cyaniceps. It was
determined for the species by the use of a mica plant cage in which
the parasite was isolated with newly infested squares. This egg is
about 0.8 mm. long, pure white, cylindrical, unsculptured, and with
a narrow neck, which is reid into a knot, probably by the Ovi-
positor after ane latter has released it. (Plate i fie, 35)

Type [I.—This is the egg of Microdontomerus Dona as shown
by the rearing of an isolated specimen. The color is ie the egg
being distinguished by the slightly papillose sculpture and by the
nipple at one end. It measures 0.38 mm. in length and 0.11 mm. in

breadth. (Plate IT, fig. 1.)

58 INSECT ENEMIES OF THE BOLL WEEVIL.

Type [11.—This is the egg of Hurytoma tylodermatis, found by the
isolation of a larva seen in the act of hatching, collected at Dallas,
Tex., August 22. The egg is dark-gray and thickly covered with
spines. It measures 0.68 mm. in length and 0.21 mm. in breadth.
The process at one end is frequently twisted. (Plate II, fig. 4.)

Type [V.—This egg was practically identified as that of a Cato-
laccus by isolation of specimens collected August 22 at Dallas, Tex.
The color is white and the egg is covered with very small tubercles or
papille. It is 0.62 mm. long and 0.22 mm. broad. (Plate II, fig. 5.)

Type V.—This egg was taken only at Glenmora, La., August 23,
on two weevil stages, and has not been identified. It is dark-gray
and very spiny, but the spines are larger, longer, and sparser than in
Type Ill. The length is 0.44 mm. and the breadth 0.19 mm. (Plate
II, fig. 6.)

Type VI.—This new type was discovered November 14 at Dallas,
Tex., and has not yet been identified. There is no sculpturing what-
ever. It is pure white. The length is 0.85 mm. and the breadth
0.19 mm. (Plate IT, fig. 2.)

THE LARVA.

The larvee of the boll-weevil parasites live as readily on dead food
as on fresh food. The hosts generally die within a very short time
after the larve begin attack. The larve have been found pretty
well grown with dry weevil larve as food. They have been found
on weevil larvae and pups indiscriminately and several times under
the elytra of teneral or unemerged adults. Just before transforming
from the larva to pupa there is considerable meconial discharge.
The majority of the boll-weevil parasites are external feeders, but
the larve of Myiophasia xnea, Ennyomma globosa, and Tetrastichus
hunter, are internal feeders. These larve kill the host in a short
time, its skin becoming shriveled and forming a perfect puparium
for the parasite. Pupation takes place within this skin. (PI. III,
figs. 5, 7.)

Pupation.—All the chalcidoid parasites have naked pupz. The
braconids usually form silken cocoons of characteristic size, shape,
mesh, or color. The cocoons of Microbracon mellitor are very vari-
able in size, color, and consistency, so that they appear almost to
belong to different insects. The cocoons of Sigalphus curculionis are
generally of a rather bright yellow and with very fine silk. The
pupal exuvium of the various species of chalcids and braconids is
sufficiently characteristic to enable a skilled observer to determine
the species after the parasite has left. (Pl. ITI, figs. 1-4, 6.)

Rapidity of development.—It is rather difficult to make an accurate
study of the developmental period of parasites, especially when every
adult parasite that matures under observation must be saved, if

DEVELOPMENT OF THE PARASITES. 59

possible, for further experimentation or for determination. It is
inadvisable to isolate many of the parasites until the larva is partially
developed, as the isolation seems to dry out both food and larva.
In the study of the parasites all those in the same stage were placed
on the same tray. When they passed to the next stage in develop-
ment they were transferred to another tray. In this manner an accu-
rate record was kept of the development. In order to determine the
total length of the breeding period it seems best to take the total
period from the collection of the material to the maturity of the last
specimen and add a plus mark (+) to this figure. . The total period
can hardly be more than 2 or 3 days longer than the longest period
thus obtained, as the egg period is very seldom more than 3 days.
To obtain the exact length of the pupal period, the maximum period
is taken to be the longest time from the observation of a fresh or
newly-formed pupa to maturity, and the minimum time is taken to be
the shortest period from the observation of the grown larva to ma-
turity. Having thus accurately defined the pupal stage, the relative
limits of the egg and larval stages are obtained by subtracting
the pupal stage from the total developmental period. Table XX,
which follows, presents all of the available data as they have been
reduced in this manner to show the length of development of the
various stages. It will be seen that most of the species pass their
entire developmental period in from 20 to 30 days between June and
October 15, but that after the middle of October the developing
stages are caught by the cold weather and the development is sus-
pended until spring. Thus, it is noticeable that parasites becoming
larvee in early October and November have a short larval period of
probably less than 20 days, becoming pupe before the cold wave
and passing a pupal period of about 150 days. Parasite larve
which hatch a little later are caught in the larval stage and hibernate
thus for from 120 to 150 days, then becoming pupe and maturing
in from 15 to 40 days. It will be noticed that Microbracon mellitor,
Kurytoma tylodermatis, and the two species of Catolaccus have short
developmental periods during the summer, while the species of
Cerambycobius have a little longer period. It will be noticed that
Habrocytus piercer has only appeared in the fall of the year. This
species has been recorded four years in succession and never before
October. On the other hand, Microdontomerus anthonomi seems to
be almost exclusively a summer parasite, having never been recorded
after September. Of course the species of which we have records
throughout the breeding season are the ones most important. This
statement is borne out by the figures on the relative numbers and
importance of the different species.

oa

60 INSECT ENEMIES OF THE BOLL WEEVIL.

TABLE XX.—Lengths of developmental periods of the boll-weevil parasites.

Sep- |October}| October | Novem- | Decem-

Species and period. June. | July. |August. tember.| 1-15. 15-30. ber. ber.
Microbracon meliitor: Days. | Days. | Days. | Days. | Days. Days. Days. Days.
CONE IARVASS <4) eee 13-19 6-8 10-15 6-10 6-20 G-EE oe ee cosarecs 118+
ERD A Se Sent gece eee 2-8 5-7 3-8 +8 10-26 BATAVAG hs See 14+
POC Re ono ent eee 21+ 13+ 18+ 14+ 30+ 150+ 156+ 132+
Habrocytus piercei:
Nerang larvae os o5 5s| Sache salsa <2 ase ee ee ee PO si he apse cea 21+ 70+
PDA ew ccones cece cated tere one faecal fog nee Sea eee eee 1) a te eae 13+ 15+
PR Oba Jase ose se aCe Zot ee se ee ee SAC ss esct oen cee 34-138-+ 85+
Catolaccus hunteri:
Egg and larva........-:----| 9144+] 14-17+) 9-12+) 13-15 $-25-F | ge. sees ee ee 70+
PASS sanoee sae Sh Oe ease 6-11 47 5-7 6-8 eT Mt ee te et ee 7-12 15+ -
ROTA S ase sere ce ane 20+ 21-+- 16-17 | 21+ 16-33-E [Sse 2 16+ 85+
Catolaccus incertus:
Egg and larva...:...------- 9-10-+-) 9-12-++| 5-11+| 9-12+) 20-40 |...-......|..2...2.2-|.....2..
aan ee et eee ee 7-8 6-9 6-8 6-9 1S" OS SSS ot [or bee oe rae ee
Rota ee see ee eee 17+ 18+ 13-15 | 18+ Bota eeeees aaee 67-E ee es
Cerambycobius sp.:
pe ANG WAVES 2225052 --2 cls 2 st secs et epee ea hae esas oe cen ack poe ae ae hal eee 65+
Fe aes ee a eee ne) Penne SPS PRG en ee | Epes rae | Deere yee | ne See ell Pre eek 18+
otal S54 So SSE esas ew Sea foed eakoe bee tee a ete es see c ec aera aa fae ae eeeaees Pe en eemee 83+
Cerambycobius cushmani:
I SANG ArV das so 9s eee ae 13+ 16+ (io) eee ee eee el eee eee eee
re et a eke ae RS |e ee oe ES 3+ -12 (abi on on ee ee eee
ROA SeS cae see ae pe ee 16+ 28-+- BSaE | hocee or [5 cakes oes ee ee eee
Cerambycobius cyaniceps: :
Nem and larvas. 252.2 =e 10+ 6-9 W5=19" 5) 36220 SSIS Set eke ss) peeks eee 84-113
120] 30 eee ee ee een 9+ 10-13 7-11 8-12 date 3. Peete ase eee 19-37
SPO Gabe eee es ee 19+ 19+ 26+ 28+ 25-4 129+ 139+ 121-150 -
Ennyomma globosa:
Mer and@tarvacs:. 3235 = teos Shs che HI-16-F ps oon af so bees seo). 5 ese |S eee
d 233.0 eet a SG A aaa ae | Pe AG A alse eho eS a ee eee
POER LG aes pe ees 202 ¢ las ccc sls ote aoa fice oes Seam rate ee sere ee eee
Eurytoma tylodermatis:
Mean Gd arva a. se oe eee 11-12+] 410+] 12-25+| 89 1i+ 1 2 eae! ba eee el or 110+
IPE ats eho scoe et ae ee 5-H 6-12 5-9 7-8 15-23 20-2 ieee Sss 17-25
Retalsrs ose see eee es 17+ 16+ 21-30 | 16+ 26+ tS eee 135+
EQriOpnagus teTanus 92922 = ele 16- l A le e Se ee
Microdontorterus anthonomi:
NANG Marva eee. 32 2 epee N4=45) |] UI=365 POG Eo els ae eee
UPA soot caso ae wee ote eee 6-7 5-6 G-4l fi s..2es| hee: eee eee
A ia A oer 20+ 21+ Vi-4l | 82-87 Gecko es see SS pee
Myiophasia 2nea:
WeSland larvae. es. Ae a | oe epee ee SsE0 has sac ecpot 2s Ses 3 ee ae es eee
1 0 2 2 ee ee eR ue ek A Se ee SH): os. ase os ae ee oe
LO 21 es oe eee Oe) Se aed Per ete 46--Y oid eerie ee Se sae ee See
Pimopla sp.:
1 DES (G el eS ay eee gee ee eel Sere amet PE ea ey eee reg) Le Cale el cere Mere Mo ob a 62+
PUP asaoce sacs ees eee Soe a ee oe eee epee es ee eee one 15-
Wotab Saco sac 2 Sse eee ele SS ae ake eK Rs ase oe fee ee eee 87+
Sigalphus curculionis:
Figte (2316 W) i a fe a pe | egee al (a a aee see oS ey oe ee E720. oo oS se eee
Pda sasosces Sebel Pee eeres paca eee, epee see IB=16 ficesc cass eee ae ee eee
TROLA ROS Sr 5h) Seeaseneey PIG Pe Se bate Pee ae RR ges fey SR oe S32b 1 cc acaet aes eee eee
Tetrastichus hunteri: ;
DPE IG MRSA ae ne eee eae eee SaTT-EIS. ces 2 o:| Senet mee [oeeeeee 443-716-F). - SoS apne aeeeee :
IRipacd.c cee eo kee eae HGSIGE ip ste as a aa aieeeeeen SE-B SS re eee
POCA eae ees Coe Sa eee PH aie | eee) ener ee) ee L74-216-2 |S Se eee
Urosigalphus anthonomi:
SRANCMARV AHS: 0os2 ooo sae ates ee Sp ee eee 626) Le eih ae S422 SS eee
PUP = ae ie oe sre so ec foes Roe poe es fae tee 926) io oe 2525 | Soe a oe ees eee ee
Total: Stee we Fe ee ee ee a eed $5) po. Sb ee ee eee
Urosigalphus sp.:
geand larvae. 2 estes asaps cece ae es reer (ener meee oer Pere ner | art ei RO ke
(PDA cot Se ae oe ee ee eee eee Sa ee 2 Sa Eo eee eee ene nee eeneee eee abe ee ae

DISTRIBUTION OF THE PARASITES. 61

8. THE DISTRIBUTION OF THE PARASITES,

Parasites of the boll weevil have been recorded from every part
of the territory so far invaded. The records are so numerous that
we are able to show statistically which are the most important para-
sites of the weevil. The following list gives the species in their
numerical rank for the entire period from January 1, 1906, to Jan-
uary 1, 1910, giving only the number which were accurately deter-
mined foreach species. The first seven species are the mostimportant,
as has been shown in almost every section of this report. The last
nine species may be considered as more or less adventitious or acci-
dental. These species may possibly never be recorded again, or, on
the other hand, they may become in the near future among the more
important parasites. This very event has happened in the case of
three or four of the other more important species. Up to 1906 only
four of the first five in this list had been recorded from the boll
weevil. The other species have been added since and some of them
will become very important as the weevils enter the moister wooded
regions of the Hast.

TABLE XXI.— Numerical rank of the parasites for the entire period, 1906 to 1910.

Number Number
Species. of Species. of

records. records.
MMACTOURACOTUUNELULON oc. lee feo 2.147) | ennyomma\ glovosa- 22: ee ee 35
COLOlaccuUs WU Nene os ie ee oe ee 1,094 || Lariophagus texanus............5....---- 8
OGLOINCCUSIINCETILUS teen oe foe eee SVS Mytophastaneneaees |e ares 4
Eurytoma tylodermatis.........-.-------- DOS | ELCLTY LOTUS Dae ere oer enn ee 1
Cerambycobius cyaniceps.....---.-------- OMAR Cena mbycoviUs: Sp seem see) soca Sen 1
Microdontomerus anthonomi........----- B02) |S MiloChalci ssp tens so sso seen ee 1
REENESTICIUUS LNs e ee n eae ee 168 || Urosigalphus anthonomi.............---- 1
Cerambycobius cushmani.....-.-.-...-.-- MOP MUMOStGMLDIALS SPs cerns see seers sar 1
SIG CL DNUSICUNCUUONISH ns smear aee se eee Sill RETULCMNPUS SP eee ee eee eee 1
ETC OTOCULIUS) DICT CO e eee erate ee 36 || Pimpla sp....--- Das As No tap IRE a i Au nD 1

A study of the value of these parasites by years has shown that
the majority of the species had not occupied the same rank in two
successive years. The accompanying diagram (fig. 13), giving the
yearly rank of the boll-weevil parasites from 1906 through 1909,
shows that in each year new parasites were recorded and that in some
cases these parasites continued to attack the weevil. Muicrobracon
mellitor appears to vary but little in importance in different seasons,
while Catolaccus hunter. shows increasing importance year by year.
Some of the other parasites of considerable importance appear extremely
variable in their relative rank. It will be noticed that Habrocytus
piercei has occupied the ninth place three years in succession and
is now in eighth place. This parasite occurs in small numbers, but
may at any time become a leading parasite in Louisiana and Missis-
sippi. In addition to giving the yearly rank of the species this
diagram also shows the proportion of the sexes observed each year.

62 INSECT ENEMIES OF THE BOLL WEEVIL.

In order to show the regions in which the various species are of
greatest importance, the accompanying map (fig. 14) is presented.
This shows that while Mecrobracon mellitor has yielded more individuals
than the other species, it is the predominating parasite in by far the
larger proportion of the infested territory. It can also be seen that
much more can be expected from the other parasites as the weevil
moves eastward into their territory. Microdontomerus anthonomi is
quite important throughout the central black-prairie region of Texas.
Eurytoma tylodermatis is more important in north-central Texas and
also in the coast region of Texas. Cerambycobius cushmani is charac-

1906 1907 1908 1909
BRACON SELLITOR ERACON MIELLITOR BRACON MELLITOR CATOLACCUS HUNTERS
Lg [> <email RE EM I el
CATOLACCUS INCERTUS CATOLACCUS HUNTER! CATOLACCUS HUNTER! ERACON PIELLITOR.
<a —ieemalie> < <a ea
CERAIBYCOBIUS CYAIVICEPS LURYTOMA TYLODERIZATIS CATOLACCUS INCERTUS CERAMBYCOBIUS CYANICEPS

<< nv Mon << cle << el

CITOLACCUS HUNTER! \ /ICRODONTOMERUS ANTHONON CERATIBYCOBIVS CYANICEPS TETRASTIGHUS HUNTER!
<< ey << EO ea (Seal
EURYTOITA TYLODERMAIT/S TETRASTICHUS HUNTERI CATOLACCUS INCERTUS
|3|> << eee > << ( e Be

CERAMBYCOBIUS CUSHMAN)

<tieP whale

MICRODONTOMERUS ANTHONOL4!

LURYTOMA TYLODERITATIS /IICRODONTOMERUS ANTHOMO/4i
ry
i
EURYTOMA TYLODERMATIS

<— aa ee Ree ea
VIVIOPHASIA AENEA LARIOPHAGUS TEXANUS HABROCYTUS PIERCE/
<TR <A ick aroma >
HABROCYTUS FIERCE! HABROCYTUS PIERCE/ S/IGALPHUS CURCULIONIS
> , V £ ie
VROSIGALPHUS ANTHONO/1 EURYTOMA SP \) YCRODONTOMERUS ANTHONOMY PIMPLA SP _
<BOMECEIE (a>
kt eee ENNYOMMA GLOBOSA CERAMBYCOBIUS CUSHMANS UBOSIGALPHUS SP
PERILAMPUS SFP MYIOPHASIA_AENEA CERAMBYCOBIUS SP
<4 ea
SP/LOCHALCIS SP. LARIOPHAGUS TEXANUS
<a
Fie. 13.—Diagram illustrating yearly rank of the boll weevil parasites, 1906, 1907, 1908, and 1909.
(Original.)

teristic of the counties grouped around Victoria County, Tex., but a
few specimens have been reared from the boll weevil at Alexandria,
La., by Messrs. Cushman and Jones.

9. THE PARASITE SEASONS.

For the convenience of this work on parasites of the boll weevil,
the year has been divided into definite parasite seasons correspond-
ing with certain groups of conditions. The year opens with the
hibernation period well underway. In so far as the parasites are
concerned those which hibernate as immature insects mature gen-
erally about the middle of March. This marks the end of the hiber-

THE PARASITE SEASONS. 63

nation period or winter season and the opening of the spring season.
From March until the middle of June or sometimes July there are no
cotton squares for the weevils to breed in. Consequently the para-
sites are obliged to seek other hosts. The swmmer season is defined
as beginning with the production of squares in which the weevils and
their parasites may breed. Thus this season continues until squar-
ing ceases—that is, until late in the fall when cotton is killed by frost
and is succeeded by the winter season. However, we frequently dis-
tinguished a fall or postmagration season, which begins with the first

==

Cc ik 7a a ag

Smee 7

OF MEXICO

l= CATOLACCUS INCERTUS...\|
2=JETRASTICHUS HUNTER. |

Fic. 14.—Map showing the distribution of the more important parasites of the boll weevil. (Original.)

attack of weevils upon the bolls in August and ends with the heavy
frosts in October or November. The fall season is also character-
ized by a renewed growth of squares.

I. THE HIBERNATION OR WINTER SEASON.

| The most important parasites which winter as immature stages
upon the boll weevil are Microbracon mellitor, Catolaccus hunter, Ceram-
_ bycobwus cyaniceps, Eurytoma tylodermatis, Tetrastichus hunteri, and
_ Habrocytus pierce. The last two species are characteristic of
_ winter examinations in Louisiana and Mississippi. The predatory

64 INSECT ENEMIES OF THE BOLL WEEVIL.

coleopterous larve Hydnocera pubescens LeConte and H. pallipennis
Say are very frequently found hibernating as larve in the boll-
weevil cells or in the cocoons of Microbracon mellitor. The stage in
which these various parasites pass the winter is given very concisely
in the table of the developmental periods (Table XX) in section 7.
During January, 1910, Mr. Hood repeatedly found LEurytoma tylo-
dermatis and Catolaccus huntert hibernating in dry cotton squares
and bolls and especially in hanging moss at Mansura, La.

II. THE SPRING SEASON.

It has been demonstrated that there is a definite period between
the hibernation season and the first infestation of squares, extending
from the middle of March to the middle of June. What happens to
the parasites during this period is of considerable importance and a
great amount of work has been done in the search for intermediate
hosts.

In the case of Catolaccus hunteri the question was very satisfac-
torily answered. At Richmond, Tex., a large number of dewberry
buds infested by Anthonomus signatus was gathered March 21, 1907,
and this species of parasite was reared continuously between March
28 and April1. At Victoria, Tex., Mr. J. D. Mitchell collected, on April
23, 1907, a lot of haws (Cratzgus mollis), infested by Tachypterellus
quadrigibbus, and on May 7 he reared this species of parasite. In-
vestigations as to the distribution of these weevils added to the
formerly known records of Anthonomus signatus in dewberry buds:
Natchitoches and Shreveport, La.; Texarkana, Ark.; Muskogee and
Ardmore, Okla.; and Trinity, Richmond, Waco, Dallas, and Mar-
shall, Tex. Tachypterellus quadrigibbus was found breeding at
Shreveport and Natchitoches, La., and Victoria, Tex.

At Dallas, Tex., the buds of Galpinsia hartwegi were found to be
infested by Auleutes tenuipes as early as April 24. This species is a
host of several species of Catolaccus. The buds of Callirrhoe involu-
crata were found at Dallas to be infested by Anthonomus fulvus as
early as April 1, and on the same date Anthonomus exneolus was first
observed to be breeding in the buds of Solanum torreyr. Solanum
eleagnifolium, with Anthonomus xneolus both in its buds and in the
fungus leaf-galls, and Solanum rostratum with this weevil in the buds,
appeared early in April. All of these plants continued susceptible to
weevil work up to the end of the spring period, or until cotton began
to square. Numerous specimens of Catolaccus were reared from the
Solanum-infesting species of Anthonomus.

Myiophasia znea was reared April 11, 1907, from Conotrachelus
elegans in galls of Phylloxera devastatrix on the petioles of Hicoria

THE PARASITE SEASONS. 65

pecan, collected April 2, 1907, at Victoria, Tex., and was reared June
5, 1907, from material collected May 4 at Dallas.

Sigalphus curculionis was reared in considerable numbers between
April 28 and May 7, 1907, from Conotrachelus nenuphar in plums
gathered at Texarkana, Ark., March 26; and between April 29 and
May 17, 1907, from Conotrachelus elegans in galls of Phylloxera devas-
tatriz on pecan, collected at Victoria, Tex., April 2; also between June
5 and 14, 1907, from the same species in material collected at Dallas,
Tex., May 4.

Cerambycobius cyaniceps was studied very carefully at Victoria,
Tex., by Mr. J. D. Mitchell during the winter of 1909-10 as an
enemy of Z7’richobaris texana in stems of Solanum rostratum, and of
Lizus scrobicollis in stems of Ambrosia trifida. Mr. T. T. Holloway
conducted experiments in longevity by feeding sugared water to the
parasites. Emergence began, in the lots of Trichobaris, on February
1 and continued until April 8. The last parasite lived until May 31.
The total period of activity was 119 days and the average period
lasted from March 11 to April 1. The longest record of longevity
was 71 days and the average 21 days. Emergence began from the
lots of Lixus on March 2 and continued until March 24. The last
parasite lived until May 11. The total period of activity was 70
days and the average period was between March 13 and April 4.
The longest record of longevity was 67 days and the average 22.

Eurytoma tylodermatis was reared from the same lots and treated
in the same manner. Emergence began from the lots of Trichobaris
on February 3 and continued until March 21. The last parasite
hved until April 30. The total period of activity was 86 days and
the average period was between March 10 and March 30. The longest
record of longevity was 42 days, and the average 20 days. Hmergence
began from the lots of Lixus on February 22 and lasted until April 17.
The last parasite lived until June 1. The total period of activity
was 99 days and the average period lasted from March 16 to April 11.
The longest record of longevity was 79 days and the average 26 days.

Ill. THE SUMMER SEASON.

The first boll-weevil parasites of the year are reared late in May or
early in June in southern Texas, but in a very short time squares are
forming all over the entire cotton belt and parasites may be found
everywhere in small numbers as the summer progresses. The per-
centage of parasitism increases rapidly and generally becomes very
high after August 1. Most of the important parasites may also be
found on their normal summer hosts.

16844°—Bull. 100—12——5

66 INSECT ENEMIES OF THE BOLL WEEVIL.

About the middle of August squares commence to fail, and few
squares are to be found by September 1. This condition may be
said to begin the fall season, when the parasites are largely obliged
to seek other hosts or to attack the boll weevil in bolls.

IV. THE FALL OR DISPERSION SEASON.

Coincident with the decline in square production is the beginning
of the boll-weevil dispersion which extends into new territory around
the entire periphery of the infested region. In the fall there is a new
growth of squares which furnishes food for the weevils before entering
hibernation and also furnishes an opportunity for very high parasitism
just preceding hibernation. It is during this season that parasite
swarms are recorded and hence this is a very critical time for obtaining
and transferring desirable parasites to new regions. During this early
fall season there are several very important ways of propagating the
parasites already present in the vicinity, as will be shown later. The
fall season of the year closes abruptly with the first killing frost, for
this crisis precipitates the hibernation period.

10. ADJUSTMENT TO NEW HOSTS.

It is a very striking fact that the continuously breeding boll
weevil is attacked by parasites which In many instances attack nor-
mally weevils having but a single generation annually. Some of these
parasites attack one host after another throughout the entire breeding
season and may be found in activity at all periods except during
hibernation. This condition is well illustrated by the accompanying
diagram (fig. 15) giving the seasonal rotation of Catolaccus hunteri
and Cerambycobius cyaniceps. Whether these parasites were origi-
nally single-generation species like their hosts is a question we can not
now decide, but we now know that they have become adapted to
many species. This fact can be most easily proven by reference to
the list of hosts of the boll-weevil parasites given in the second section
of this part (p. 42). It appears possible that the constantly changing
factors of nature cause the various species to be continually adjusting
their habits to new environments and new hosts. In other words, the
groups of parasites from which the most available enemies of a new or
introduced species may be obtained are those groups in which the
parasitic habits are the most variable. A parasitic species that is as
readily at home on a stem weevil as on a bud or seed weevil is probably
able to attack many different species.

The most striking example of the adjustment of new parasites was
furnished in 1907. A lot of hanging squares collected by Mr. J. D.
Mitchell on August 5, 1907, at Victoria, Tex., on a field known as the
Haskell field gave a percentage of 61.5. There was something so

ADJUSTMENT TO NEW HOSTS. 67

striking about the nature of the record that Mr. Cushman was sent
immediately to Victoria to study the surroundings of this field and
report upon the possible reasons for the high percentage of parasitism.
Mr. Cushman reported after considerable study that there were
only two factors which, it seemed to him, might have an influence
upon the parasites of the boll weevil. The first factor was the com-
plete lack of fruit upon the huisache trees (Vachellia farnesiana)
which is the normal food of Laria sallzi. The second factor noticed
was the absence of flowers on the Callirrhoe wnvolucrata, the host of
Anthonomus fulvus. Mr. Cushman reasoned that the point would be
proven if we should rear from the boll weevil some of the character-
istic parasites of either this or the other species. As a result of
rearings from the material collected in this field, the principal par-
asite was Microbracon mellitor, the typical boll weevil parasite, but a

SEASONAL ROTATION ¢ OF HOSTS. BY CATOLACCUS HONTER CRAWFORD.
[FEBRUARY MARCH | APRIL | MAY "| JUNE | JULY | AUGUST |SEPTEMBER|OCTOBER WOVEMBER

ANTHONOMUS SIGNATUS. oe
| OMUS AENEOLUS. Bee Rey
TACHVATERELLUS_QUADRIGIBE eae
D 4 ae GRANDIS| BOHEMA

: ee ISL BOIL Po

WZTWZZITA DECEMBE)

SEASONAL ROTATION OF HOSTS BY manele: Co Bee

TACHYPTERELLUS QUADRIGIBBUS SHY
VCHYUS we LECOWTE
ee cas co} ie
ys BoE ann or
oo ; — IS Ber)
TRICHOBAR/S TEXAWA LECONTE __. eke GARIS TEXANA ae ONTE.
Db

p Siichavua BERCULATUS. ae
oe eae ne)

Fic. 15.—Diagram illustrating the seasonal rotation of hosts of Catolaccus hunteri and Cerambycobius
cyaniceps. (Original.)

species which is also a typical parasite of Anthonomus fulvus. It
is probable that the latter species furnished some of the Micro-
bracons for this infestation. The next most important species was
Cerambycobius cushmam, a typical parasite of Laria salle. and of
Arecerus fasciculatus which breeds in the fruit of the chinaberry tree
(Melia azederach). In addition to this species, this same field yielded
3 other new parasites of the boll weevil, 2 of which are known to
be parasites of the Laria. These were Eurytoma sp., Spilochalcis sp.,
and Lariophagus texanus.

To illustrate the divergence of habits among parasites the host
relations of Catolaccus incertus may be cited. This parasite attacks
several species of Laria (Bruchus) which are internal seed eaters and
pupate in their feeding cells; such weevils as Zygobaris zanthoxylt and
Auleutes tenuipes, which are seed or bud feeders and pupate in the

68 INSECT ENEMIES OF THE BOLL WEEVIL.

ground; and Anthonomines, which dwell in buds (Anthonomus gran-
dis), in flowers (A. aphanostephi), and in hard seed (A. albopilosus).
But it draws the line apparently at stem dwellers and is replaced by
Neocatolaccus tyloderme on Lixus, Tyloderma, and Ampeloglypter.
Cerambycobius cyaniceps 1s as much at home in a stem as ina bud,
and so also are Eurytoma tylodermatis and Microdontomerus anthonomi.
The Braconide appear to be more particular as to food but the most
noted of all,- Microbracon mellitor, has no preferences between stem
dwellers and bud dwellers.

Thirteen miles southeast of Yazoo City, Miss., on November 1, 1909,
the senior author found an isolated artificial focus of infestation by
the boll weevil over 30 miles from any infestation of the same age
and 20 miles beyond the regularly infested region. Out of 8 squares
picked, containing 5 stages, 1 parastized stage was found.

11. BEETLES WHICH PREY UPON THE BOLL WEEVIL.

The attack of the insects predatory on the adult boll weevil is purely
accidental. They may be very numerous, but the only ones recorded
and verified are Evarthrus sodalis Le Conte and another species of the
same genus. There are, however, several insects which have an actual
value through their established habit of either breeding in the square
upon the boll-weevil stages or of entering the square and consuming
the weevil. We shall refer to four of them.

Hydnocera pallipennis Say. A single beetle of this species was
reared April 6, 1907, after 183 days in its cocoon, and over 214 days
isolation in the rearing tube. It was collected in a boll-weevil cell at
Waco, Tex., August 28, 1906. The cocoon is very finely threaded,
loosely woven, and only single layered. The stage of the beetle can
easily be observed at any time.

Hydnocera pubescens Le Conte. This clerid is a very common
breeder in the weevil cells. Its larve have been found not only
feeding upon the various weevil stages but have been taken frequently
from Microbracon cocoons which they have entered at a much
younger stage.

Cathartus gemellatus Duval. This cucujid beetle is both a predator
and a scavenger, its larve being frequently found, however, feeding
upon boll-weevil stages which they must have killed.

Chauliognathus spp. The larve of these lampyrid beetles are very
common in the squares and bolls of cotton in Louisiana and Missis-
sippl. In one instance undoubted proof of the attack of such a larva
upon one boll-weevil larva was recorded. Many other very sus-
picious observations were made but no definite proofs found.

ANTS WHICH PREY UPON THE WEEVIL. 69

12. LEPIDOPTEROUS LARV@ WHICH ARE INCIDENTALLY PREDATORY
UPON THE BOLL WEEVIL.

Alabama argillacea Hiibner. The cotton leaf caterpillar is distinctly
an enemy of the boll weevil and of considerable importance. When
it defoliates a cotton field a month or more before the frosts it often
destroys immature weevils in the cotton squares and cuts off the entire
food supply of the adult weevils remaining. These weevils may be
able to suspend their activities and begin hibernation but it is well
known that weevils entering hibernation early in the fall can seldom
survive a long hard winter, or live until cotton is up in the spring.
Those that can not hibernate either die of starvation or rise in flight
to seek cotton elsewhere and may perish in the effort. Itis presumed
that a very high percentage of flying weevils fails to find cotton.

The leaf worm is attacked by 18 predatory bugs, 16 predatory
beetles, 6 predatory wasps, and the following ants: Dorymyrmex
pyramicus flavus McCook, Forelius maccookt Emery, Solenopsis
geminata Fabricius (these three ants are enemies of the boll weevil)
and Monomorvum carbonarvum Smith. Ten hymenopterous parasites
and one hyperparasite are known, and in addition the leaf worm is
attacked by a predatory fly and by two parasitic flies.

13. ANTS WHICH PREY UPON THE BOLL WEEVIL.

HYMENOPTERA. DORYLIDA.

Kciton (Acamatus) commutatum Emery. This ant was taken by
Mr. C. R. Jones at Beeville, Tex., attacking the boll-weevil larvee in
squares. Dr. W.M. Wheeler states that it is commonly parasitized
by a round worm of the genus Mermis.

PONERIDA.

Ectatomma tuberculatum Olivier. The ‘“‘kelep,”’ or so-called Guate-
malan ant, is a native of Mexico and Central America. Like all other
ponerids it is slow in action. The winters have proven too severe
for any of the imported colonies. The rate of development is so
slow and the movements of the adults are so sluggish that little
could be hoped for from this species even if it could become accli-
mated in this country.

MYRMICIDA.

Cremastogaster lineolata (Say) var. clara Mayr. This ant is also an
enemy of the boll weevil, having been recorded attacking immature
stages at Dallas, Tex., by Dr. W. E. Hinds. It has frequently been
seen in the rearmg cage carrying off insect prey. The species lives

70 INSECT ENEMIES OF THE BOLL WEEVIL.

in hollow stems, sticks, and galls and is commonly seen at the necta-
ries of cotton or attending aphides, membracids, etc. Prof. F. E.
Brooks has recorded this ant as an enemy of Heliothis obsoleta, the
cotton bollworm.

Solenopsis geminata Fabricius. The ‘fire ant” (fig. 16) is very
common in Texas cotton fields, where it is always an enemy of the
boll weevil, as well as of the cotton bollworm (Helvothis obsoleta)
and the cotton leaf worm (Alabama argillacea). In Louisiana,
Arkansas, and Mississippi it is very seldom seen in cotton fields,
except in southern Louisiana, where unfortunately it is in danger of
extermination by the Argentine ant, Jridomyrmex humilis Mayr.
This species divides credit for the greater part of the ant control of
the boll weevil with the other species of Solenopsis, two species of
Monomorium, and with the various spe-
cies of Pheidole. Its nests are placed in
the cotton fields, generally near the base
of the plants, and from these the foragers
go out in all directions in search of food.
The workers have learned to detect the
presence of the boll weevil in the squares
and in a short time can effect an entrance
into the weevil cell from which they either
draw the weevil bodily or convey it in
parts to their nests. This ant is some-
times found on the plant, but most com-
monly it does its work on the ground.
The species is parasitized by (Pseudac-
Fic, 16.—The “fire ant” (Solenopsis eon) Plastophora crawford: Coquillett at

geminata), an enemy of the boll Dallas, Tex.
hale Ree ae . Shannen Solenopsis molesta Say (debilis Mayr).
This minute ant was taken in the act of
attacking a boll-weevil larva by Mr. Cushman at McAlester, Okla.
This species and the next are so similar in appearance that they
may be easily confused. Prof. F. E. Brooks has recorded it as an
enemy of Craponius inxqualis.

Solenopsis texana Emery. This minute ant is a common enemy
of the boll weevil in Texas, Louisiana, and Mississippi. The entrance
holes are very minute, but sometimes the ants enter the squares in
great numbers. On October 31, 1907, at Thornton, Tex., Mr.
Cushman found 85 individuals attacking a weevil larva in a single
square. It is mentioned in the investigation records as attacking
the weevil at Alexandria and Monroe, La., and Cuero, Lampasas, and
Llano, Tex. It is also recorded as an enemy of Heliothis obsoleta.

Monomorium minimum Buckley. This common house ant (fig. 17)
is a very valuable enemy of the boll weevil and is common in cotton

ANTS WHICH PREY UPON THE WEEVIL. PA.

fields. It is recorded in the Dallas collection as attacking the boll
weevil at Llano, Lampasas, Albany, Henrietta, Arlington, and Dallas,
Tex., Ruston, La., and Roxie and Port Gibson, Miss. The species
has been taken attacking the immature stages of T’richobaris com-
nacta, Anthonomus albopilosus, and Anthonomus fulvus. It generally
attacks these weevils as well as the boll weevil on the plant, entering
the infested bud or square in search of its food.

Monomorium pharaonis L. This cosmopolitan house ant (fig. 18)
is another of the most important boll-weevil enemies, being very

Fig. 17.—The little black ant (Monomorium minimum), an enemy of the boll weevil: a, Fe-
male; 6, same with wings; c, male; d, workers; e, pupa; f, larva; g, egg of worker.
Enlarged. (From Marlatt.)

abundant in the cotton fields of certain sections. It is represented
in the Dallas collection as attacking the boll weevil at Victoria, Tex.;
Fosters,. Ruston, and Monroe, La., and Camden, Ark. It also
attacks the weevil on the plant. In southern Louisiana it is being
exterminated by the Argentine ant (/ridomyrmex humilis).

Pheidole sp., near flavens. At Arlington, Tex., August 31, 1908,
Mr. Cushman found abundant evidence of the control of the boll
weevil by this species. It attacks the weevil larve both on the
plant and on the ground.

72 INSECT ENEMIES OF THE BOLL WEEVIL.

Pheidole crassicornis Emery. At Lampasas, Tex., September 23,
1908, Mr. Cushman found this ant a very abundant enemy of the
boll weevil.

DOLICHODERID.

Foreluus maccooki Forel. At Beeville, Tex., August 138, 1906, Mr.
C. R. Jones found a high mortality of the boll weevil due to this
species. Dr. Wheeler has recorded the fact that this ant prefers
bare, dry ground for-its nests. The species also attacks Alabama
argillacea and Heliothis obsoleta. On September 7, 1908, at Dallas,
Tex., Mr. F. C. Bishopp took specimens in the act of attack, and
September 21, 1908, Mr. Cushman took others at Llano, Tex., attack-
ing the weevil.

Dorymyrmexz pyramicus Roger, the ‘‘lion ant,’ protects solitary

tree cotton from the boll weevil in Cuba (Schwarz, 1905).

Fia. 18.—The little red ant (Monomorium pharaonis),an enemy of the boll weevil:
a, Female; 6, worker. Enlarged. (From Riley.)

Dorymyrmex pyramicus (Roger) var. flavus McCook. This com-
mon ant of the cotton fields has only once been taken as an enemy
of the boll weevil, namely at Texarkana, Tex., by Mr. R. C. Howell,
but its abundance would make it a very important species if it
should develop a fondness for weevil larve. It is an enemy of
Alabama argillacea and Heliothis obsoleta.

Indomyrmexz analis André. Specimens of this ant were found
attacking the boll weevil by Dr. W. E. Hinds. This species is nor-
mally a honey ant, but occasionally takes insect food. It is very
common in cotton fields, especially in Louisiana.

Iridomyrmex humilis Mayr. The much-feared Argentine ant has
been taken attacking the boll weevil. It is, however, a friend to
the weevil because it exterminates Solenopsis geminata, Monomorwum
pharaons, and Iridomyrmex analis (Foster, 1908).

tsi oh aie Piatt)”

WN SS aaeemlmaEa2——ey—y———E—E—E—— — — —AOn—O—— EE EE —

BIOLOGY OF THE COHOSTS. vas)

FORMICIDA.

Formica fusca (Linneeus) subpolita (Mayr) perpilosa Wheeler. This
species of ant is normally a honey feeder, but it is recorded by Rangel
(Rangel, 1901c) as a predator on adult boll weevils in Mexico.

Formica pallidi-fulva Latreille. A single instance of this species
cutting its way into a square infested by a boll weevil was observed
by Mr. Hood at Ashdown, Ark., September 2, 1908.

Prenolepis imparis Say. A single instance of this species cutting
its way into a square infested by a boll weevil was observed by Mr.
Hood at Ashdown, Ark., September 2, 1908.

14. BIOLOGY OF THE COHOSTS OF THE BOLL-WEEVIL PARASITES.

The biologies of the parasites concerned in the boll-weevil complex
have already been discussed. It now remains to consider the native
weevils which have already or may later enter into the complex of
cohosts of the boll-weevil parasites. Many of these weevils are
native to the territory already occupied by the weevil, while others
will become important as new territory is added. Other families of
Coleoptera and even other orders of insects may later be found to be
of more or less importance as cohosts of boll-weevil parasites. The
late Dr. William H. Ashmead stated that Microbracon mellitor had
been reared from many Coleoptera, while Cerambycobwus cyaniceps
bred in cerambycids and other beetles. It is important also to note
the record of Cerambycobius cyanceps irom Languria. Our own
observations have been confined to the Coleoptera of the families
Lariide, Anthribide, and Curculionide.

PHYTOPHAGA. LARIIDA.

(Bruchus)' Laria salle, Sharp. This bruchid is characteristic of
the Gulf Coast prairie of Texas. It breeds in the pods of huisache
(Vachellia farnesiana) , is a continuous breeder, and is generally highly
parasitized by Urosigalphus bruchi, Cerambycobius bruchivorus, CER-
AMBYCOBIUS CYANICEPS?; CERAMBYCOBIUS CUSHMANI, LARIOPHAGUS
TEXANUS, HURYTOMA TYLODERMATIS, [orismenus sp., and several
other undetermined parasites. :

Larva exigua Horn. This bruchid is apparently Austroriparian and
Carolinian. Its principal food plant is Amorpha fruticosa, in the seed

1 The generic name Bruchus was first used by Geoffroy in 1762. Only one species is admissible in our
code of nomenclature and this is CerambyxfurLinneus, which is also the type of Ptinus Linnzeus 1767,
The genus Laria was described by Scopoli in 1763 and the type thereof has been designated as salicis
Scopoli, a synonym of Dermestes pisorum (pisi) Linneseus.

Linnzeus’s conception of Bruchus dates from 1767 and the type thereof was designated by Latreille
(1810) as Dermestes pisorum Linnzeus. Hence we see that Bruchus Linnzeus (1758) is preoccupied by
Geoffroy (1752) and an isogenotypic synonym of Laria Scopoli (1763).

Although the genus has been subdivided into several genera, our American species have not been
studied with regard to such subdivision and it is hence best to consider all as inthe genus Laria,
sensu latiore.

2 The names of boll-weevil parasites are printed in small capitals; others in italics.

74 INSECT ENEMIES OF THE BOLL WEEVIL.

pods of which it breeds prolifically. It is a continuous breeder and
is highly parasitized by Cerambycobius brevicaudus, CERAMBYCOBIUS
CYANICEPS, Horismenus sp., Heterospilus prosopidis, Eurytoma sp.,
MICRODONTOMERUS ANTHONOMI, CATOLACCUS INCERTUS, and several
other species.

Larva obtecta Say. The common bean weevil is known to be para-
sitized by CERAMBYCOBIUS CYANICEPS and Bruchobius laticollis.

Larva compressicornis Schaeffer. This bruchid, which breeds in
the pods of Acuan wlinoensis, is parasitized by CERAMBYCOBIUS CY-
ANICEPS and Heterospilus prosopidis.

Larva ochracea Schaeffer. This bruchid, which breeds in the pods
of Vicia sp., is parasitized by CERAMBYCOBIUS CYANICEPS, C. CUSH-
MANI, Hurytoma sp., and Heterospilus prosomdis.

Spermophagus robinie Schaeffer. This bruchid is very common in
the pods of the honey locust (Gleditsia triacanthos), and the water
locust (Gleditsia aquatica), both of which are trees belonging to the
humid Austral zones. It is parasitized by Heterospilus bruchi, CER-
AMBYCOBIUS CYANICEPS, HURYTOMA TYLODERMATIS, and Urosigalphus

bruchi.
RHYNCHOPHORA. ANTHRIBIDZ.

Brachytarsus alternatus Say. This beetle probably breeds under
many different circumstances. The only records are from a fungus
gall on Ipomea pandurata, and from the stems of Elymus virguncus
and Sideranthus rubiginosus. It apparently belongs to the humid
Austral zones. It is parasitized by MicRODONTOMERUS ANTHONOMI
and a Bracon.

Arecerus fasciculatus DeGeer. This very widely distributed
Lower Austral insect (see fig. 19), known commonly as the coffee-
bean weevil,
breeds in stored
vegetable prod-
ucts, in the seed
of Theobroma
cacao, in the
berry of the coffee
tree (Coffea ara-
bica), in diseased
cotton bolls, in
seed pods of
Cassia occident-

Fig. 19.—The coffee-bean weevil (Arzcerus fasciculatus), a cohost of boll- j
weevil parasites: a, Larva; b, adult; c, pupa. Enlarged. (From Chit- alis and @. ob-

sn eee
enden.) tusifolia, in seeds

of Indigofera tinctoria, in green and decaying fruit of Mela azedarach,
in green and dry cornstalks, and in dry acarian galls on [pomea

See I gS SOR,

BIOLOGY OF THE COHOSTS. 15

lacunosa. In the Melia betries it is parasitized by CrERAMBYCOBIUS
CUSHMANI, EURYTOMA TYLODERMATIS, and PEDICULOIDES sp.

CURCULIONID®. APIONINA.

Apion segnipes Say in Cracca virginiana is parasitized by Eury-
TOMA TYLODERMATIS.

Apion decoloratum Smith. Dr. Chittenden records this weevil as
breeding in Meibomia paniculata and parasitized by CaTOLAccus
INCERTUS.

Apion griseum Smith. Dr. Chittenden records this weevil as
breeding in Phaseolus retusus, P. wright, P. polystachyus, and
Strophostyles pauciflora, and parasitized by CATOLACCUS INCERTUS.

Apion mgrum Smith. Breeds in buds of Robima pseudacacia and
is parasitized by CATOLACCUS INCERTUS.

Apion rostrum Say in pods of Baptisia is parasitized by CERAMBY-
COBIUS CYANICEPS.

CLEONIN A.

Lizus musculus Say. This weevil is known both from the Lower
Sonoran and Austroriparian zones. It breeds in the stems of Poly-
| gonum pennsylvamcum, P. portoricense, and P.
punctatum, making an oblong gall or swelling.
It is parasitized by EuRYTOMA TYLODERMATIS,
CERAMBYCOBIUS CYANICEPS, Neocatolaccus
tylodermx, Glyptomorpha rugator, G. novitus,
and Horismenus lixiworus.

Iizus scrobicollis Boheman. This weevil (fig.
20) is probably confined mainly to the moist
Austral zones. It breeds abundantly in the
stems of Ambrosia trifida, A. artemisizxfolia,
A. psilostachya, and Hehanthus spp. It is

quite highly parasitized by Piinobiusmagnijicus,

Bee tee ove tyicoiay 4 EURYTOMA TYLODERMATIS, CERAMBYCOBIUS

cohost of boll-weevil para- CYANICEPS, Glyptomorpha rugator, G.mavaritus,

pee edna) i Ga tet, Vino belfrage, Microdus similis,

and Horismenus liaworus. Mr. Townsend has

described Lixophaga parva from a specimen reared from this weevil
at Dallas, Tex., August 15, 1907.

ERIRRHININA.

Smicronyz tychoides LeConte. This weevil breeds in stem galls of
various species of Cuscuta. It is parasitized by MicRoBRACON MEL-
LITOR and Hutrichosoma albipes.

76 INSECT ENEMIES OF THE BOLL WEEVIL.

Desmoris scapalis LeConte. This weevil (fig. 21) occurs mainly
on the black prairie in Texas and breeds in the heads of Stderanthus
rubiginosus. It is parasitized by MicROBRACON MELLITOR.

ANTHONOMINA.

Macrorhoptus sphxralcix Pierce. This weevil was found breeding
in stems of Spheralcea angustifoha. It is the host of Huryroma
TYLODERMATIS. |

Tachypterellus quadrigibbus Say. This fruit weevil breeds in the
seed of apple, pear, Cratzgus oxyacantha, and Cratzgus mollis. It is
known to us to be parasitized by CERAMBYCOBIUS CYANICEPS and
CATOLACCUS HUNTERI.

Smicraulax tuberculatus Pierce. This species breeds in the stems
of mistletoe (Phoradendron flavescens) throughout Texas, and evi-
dence of its work has been observed in Louisiana and Mississippi. It
is parasitized by HuRYTOMA TYLO-
DERMATIS, CERAMBYCOBIUS CYANI-
CEPS, CATOLACCUS HUNTERI, and
MiCROBRACON MELLITOR.

Anthonomus fulvus eConte.
This weevil breeds in the larger
buds of Callirrhoe involucrata and
CO. digitata. It is a characteristic
woodland and meadow insect in
Oklahomaand Texas. Theknown
parasites are CATOLACCUS INCER-

Fig. 21.—The ironweed weevil (Desmoris sca. TUS and MicRoBRACON MELLITOR.
palis), a cohost of boll-weevil parasites. En- °
larged. (From Hunter and Hinds.) Anth ONOMUS Sg natus Say. The

strawberry weevil is mainly char-
acteristic of the humid Austral zones, and it breeds in the buds of
strawberry, blackberry, dewberry, raspberry, Rubus villosus, Poten-
tilla canadensis, and Cercis canadensis. It is parasitized by Micro-
bracon anthonomi, Calyptus tubtator, CATOLACCUS HUNTERI, C. INCER-
Tus, and C. anthonomi. The two latter species were described from
this weevil.

Anthonomus albopilosus Dietz. This little Texas weevil breeds in
the capsules of Croton capitatus, C. engelmanni, and CO. texense. It
is known to us to be parasitized by MicroBrRacon MELLITOR, CaTo-
LACCUS HUNTERI, C. INCERTUS, and CERAMBYCOBIUS CYANICEPS.

Anthonomus mgrinus Boheman. ‘This species is eastern in habitat
and breeds in the buds of Solanum carolinense, and the potato (S.
tuberosum). It is the host of Entedon lithocolletidis, Eriglyptus
robustus, CATOLACCUS INCERTUS, and (. anthonomi.

- this weevil.

BIOLOGY OF THE COHOSTS. UTE

Anthonomus xneolus Dietz. This Texas weevil breeds commonly
in fungus galls on the leaves and in the buds of Solanum eleagmfolrum
and S. torreyi and also in the buds of S. rostratum. It is parasitized
by CaToLacoUs HUNTERI and a Kurytoma.

Anthonomus eugenw Cano (xneotinctus Champion). The pepper
weevil (fig. 22) breeds in most of the cultivated and wild peppers
and may be considered a serious pest. It is parasitized by CaToLac-
CUS HUNTERI, MICROBRACON MELLITOR, and PEDICULOIDES VENTRI-
COSUS.

Anthonomus squamosus LeConte. This is a weevil typical of the
gypsum prairie of the Lower Sonoran Zone, although occurring less
abundantly in the western edge of the moist Austral zones. It
breeds in the flower heads of Grindeha squarrosa nuda, G. wnuloides,

and perhaps also on other Grindelias and Helianthi. It is known -
‘to us to be parasitized by MicroBracon
-MELLITOR, CATOLACCUS HUNTERI, and Eury-

TOMA TYLODERMATIS.
Anthonomus nebulosus LeConte. This wee-
vil breeds in the buds of Cratzegus in Louisi-
ana and Arkansas. It is parasitized by Cato-
LACCUS HUNTERI and Sigalphus sp.
Anthonomus heterothece Pierce. This small
weevil breeds in the flower heads of Hetero-
theca subaxillaris and probably other asteroid
flowers. It is parasitized by CaToLaccus
HUNTERI and EuRYTOMA TYLODERMATIS. Fic. 22.—The pepper weevil
Previous records by the senior author on ‘Anthonomus eugenti), a cohost

of boll-weevil parasites. En-

Anthonomus disyunctus LeConte all refer to larged. (From Hunter and
Hinds.)

Anthonomus aphanostephi Pierce. This weevil breeds in the heads
of Aphanostephus skwrrobasis, and is parasitized by CaToLaccus

INCERTUS.
TYCHIINA.

Tychvus sordidus LeConte. This Austroriparian weevil breeds in
the pods of Baptisia bracteata and B. leucantha. It is parasitized
by CERAMBYCOBIUS CYANICEPS.

CRYPTORHYNCHINA.

Chalcodermus zneus Boheman, the common cowpea weevil (fig. 23),
is abundantly parasitized by EnnyomMa GLOBosa, and is likewise a
host of Ennyomma clistoides and SIGALPHUS CURCULIONIS.

Conotrachelus affinis Boheman. This weevil breeds in hickory nuts
and is parasitized by My1opHasiA ZNEA and SIGALPHUS CURCULIONIS.

aS te nes

78 INSECT ENEMIES OF THE BOLL WEEVIL.

Contrachelus juglandis LeConte. This is the walnut weevil, which
is also parasitized by MyropHasia #NEA, Cholomyia inequipes, Meta-
dexia basalis, and SIGALPHUS CURCULIONIS.

Conotrachelus elegans Say. This weevil breeds abundantly in
the petioles of hickory, the galls of Phylloxera devastatriz on pecan,
in pecan nuts, in leaf
rolls on hickory, and
finally in the roots
of Amaranthus retro-
flecus-~ It “is = ire-
quently parasitized
by MYIoPHASIA ZNEA
and SIGALPHUS CUR-
CULIONIS, and occa-
sionally i Cholomyia
inequipes.

Conotrachelus nen-
uphar Herbst. The
common plum curculio (fig. 24) breeds in the pulp of drupes and
pomes. The larve are parasitized by Cholomyia wmexquipes, SIGAL-
PHUS CURCULIONIS, MICROBRACON MELLITOR, and Porizon conotracheli,
and the eggs by Anaphes conotracheli.

Conotrachelus naso Le-
Conte. The common
acorn weevil is para-
sitized by SIGALPHUS
CURCULIONIS.

Tyloderma foveolatum
Say. Thiscommon wee-
vil breeds prolifically in
the stems of Onagra bien-
ms and Epilobium. It
is highly parasitized by |
Neocatolaccus tyloder Me, Fie. 24.—The plum curculio (Conotrachelus nenuphar), a cohost
CER AMBYCOBIUS CYAN- of boll-weevil parasites: a, Larva; b, adult; c, pupa. Much

enlarged. (From Chittenden.)
ICEPS, EURYTOMA TY-
LODERMATIS, MicROBRACON MELLITOR, SIGALPHUS CURCULIONIS, and
Urosigalphus sp. nov.

Gersteckerva nobilis LeConte (Acalles). The common prickly-pear
weevil is parasitized by CATOLACCUS HUNTERI and by several other
species.

Fie. 23.—The cowpea weevil (Chalcodermus xneus), a cohost of
boll-weeyil parasites. Enlarged. (From Chittenden.)

BLOLOGY OF THE COHOSTS. 79
CEUTORHYNCHINZ.

Auleutes tenuipes LeConte. This weevil breeds in the anthers of
buds of Galpinsia hartwegi on the Texas black prairie at least. It 1s
attacked by CaroLtaccus INcERTUS, Jicrobracon sp., Hutrichosoma
albipes, and possibly by Catolaccus nigroxnea.

Craponius inxqualis Say. This weevil breeds in the fruit of the
grape. It is parasitized by MicrRoBRACON MELLITOR and Stiboscopus
brooksv.

Rhinoncus pyrrhopus Boheman. This weevil breeds in the stems
of Polygonum and is parasitized by CERAMBYCOBIUS CYANICEPS.

Ceutorhynchus n. sp. This weevil breeds in the crown of Selema
aurea and is parasitized by CATOLACCUS INCERTUS.

BARIN &#.

Baris cunevpenms Casey. This weevil breeds in the roots of
Helenium tenuifolium and is parasitized by CATOLACCUS INCERTUS.

Orthoris crotcha LeConte. This
Lower Sonoran weevil breeds in the
seed pods of Menizeha nuda. It is
parasitized very highly by Microbracon
nuperus, KURYTOMA TYLODERMATIS,
and a species of Tetrastichus.

Trichobaris texana LeConte. This
species breeds very abundantly in
stems of Solanum rostratum, and is
hence more or less a Lower Austral
insect. Its parasites are CERAMBYCO-
BIUS CYANICEPS, HURYTOMA TYLODER-
MATIS, MIcROBRACON sp., and SIGAL-

PHUS CURCULIONIS. Fig. 25.—The potato-stalk weevil ( Tricho-

Trichobanis irinotata Say. The po- baris trinotata), a cohost of boll-weevil
Cat t le aT (fi 9 5) b d i parasites: a, Beetle; 6, larva from side;
ato sta weevil gS: reeds 1m c, pupa; d, section of potato stalk opened
the stems of many Solanacez, includ- _ to show larva and pupa in situ. ct b, ¢,
: . Five times natural size; d, natural size.
ing Solanum carolinense, S. melongena (grom Chittenden.)
(ege plant), S. rostratum, S. tuberosum
(potato), Datura stramomum, D. tatula, Physalis longifolia, P. philadel-

phica, P. lanceolata, P. heterophylla, and P. virginiana ambigua. It is

known to be parasitized by SIGALPHUS CURCULIONIS and EKuRYTOoMA-

TYLODERMATIS.

Trichobaris compacta Casey. This weevil breeds in the pods of
Datura stramonium and is also recorded as breeding in Datura mete-
loides. It is parasitized by CERAMBYCOBIUS CYANICEPS, MYIOPHASIA
NEA, and PEDICULOIDES VENTRICOSUS.

80 . INSECT ENEMIES OF THE BOLL WEEVIL.

Ampeloglypter sesostris LeConte. The grapevine gall weevil is para-
sitized by MytopHasia £NEA, Neocatolaccus tyloderme, and Calyptus
tibvator.

Zygobaris xanthoxyli Pierce. This weevil is abundant in the berries
of Xanthorylum clava-herculis. It is parasitized by CaToLaccus
HUNTERI and SIGALPHUS CURCULIONIS.

BALANININA.

Balaninus nasicus Say. This weevil breeds in acorns. It is para-
sitized by My1opHasiA £NEA and possibly by Trichacis rufipes.

CALANDRINZ.

Calandra oryza Linneus. The cosmopolitan rice and corn
weevil (fig. 26) breeds in acorns of several species of oak, in galls of
Phylloxera devastatrix on Hicoria pecan, in
old cotton bolls, and in all kinds of stand-
ing and stored grain. It is parasitized by
Meraporus calandre, M. vandinei, M. uti-
bilis, M. requisitus, and CATOLACCUS IN-
CERTUS. Other parasites have been re-

. ported abroad.

FGhED) d chat bate 2D 0M LIST O8 THE. HOSTS PraNTms eigen:

parasites. Enlarged. (From COHOST WEEVLLS.

ae In order to show more plainly the num-
ber and variety of plants whose presence around the cotton field,
if infested by their typical weevils, would influence the parasite
control of the boll weevil, the following list is presented, using the
classification of Britton (1901):

Plant. Infested by—
|
Privcum,: SauUvUuMm wheal) = 2f 2 jee s sees | Calandra oryza I:.
LUA OUP OU TULCHS eee geen te yt | Anthribus alternatus Say.
a : {Arecerus fasciculatus DeG.
Zea mais (Com)...-..---+2-+--++++-----+--- ‘(Calandra oryza L.
Orem SGU (FICC) oe i. as eee Calandra oryza.
uglns Tera (walnut). <os2 se oe oe ee | Conotrachelus juglandis Lec.
ELCOTESP> (MICKORY,) —22 en ea ae ee Conotrachelus affinis Boh.
=Hacoria athe (hiekory): S28 ose ae. see Conotrachelus elegans Say.
HLICOTIG PCCON: (PECAN) a8 ese ahs = Conotrachelus elegans. : |
Balaninus spp.
QuLenciisssp pie te bet Seay i ee eed ses Conotrachelus naso Lec.
Calandra oryza L.
NGTAMENATOM:flOVCSECNS ae 2 ne eee a Smicraulax tuberculatus Pierce.
Polygonum pennsylvanicum.....-..--------- Taxus musculus Say.
OLSGOUI POLLOTLECTUS Cars ta Pere es Tizus musculus.
= Taxus musculus.
EOI TE EIT CRATES SS Boe eae pyrrhopus Boh.

ATROTONAIVUS TELTOSLCEUS - 2 2s en eee Conotrachelus elegans Say. ‘

HOST PLANTS OF COHOST WEEVILS.

81

Plant. Infested by—
SICLERIORCUREO age). arash See ae ee ist Ceutorhynchus sp.
wows vellosus: (black berry,)-ss- 5222 2k = Anthonomus signatus Say.
ows triivalis" (Gewiberry)s ssa see ee a5 - Anthonomus signatus.
Rubus occidentalis (raspberry). .--.--------- Anthonomus signatus. = -
Fragaria virgimana (strawberry).-.--------- Anthonomus signatus.
ROLEIUUNLGRCONAGCNSIS sett Ses beet a fe Anthonomus signatus.
EEUSECOMINUNTS (DEAE) aoe st gat eee yao as Tachy pterellus quadrigibbus Say.
Maluswmnalus (apple) teiee = Uae 2. Hie atta gener
; . Tachypterellus quadrigibbus.
Crategus mollis (haw)..--.--..-----..--..- ennai nebulosus Lec.
Crataegus OTYACanthase a. 2 28 oS ene to: Tachypterellus quadrigibbus Say.
UCAS) (((OUUITIO)|21) 5 See Retest ee Reto Nee oe Conotrachelus nenuphar Hbst.
TAMYGEGLtS! Per sica. (PEACH) 4s. 02 en eee. yee Conotrachelus nenuphar.
Amygdalis persica (mectarine)s 4. 2.2852... -. Conotrachelus nenuphar.
Amygdalis armeniaca (apricot).......--.----- Conotrachelus nenuphar.
Vachellia farnesiana (huisache) .....--.---- (Bruchus) Laria sallxt Sharp.
ACUATUMULUMOCNS Seen ae ae aes | Lama busignata rorm.
Strombocarpus (screw-bean)...-...-.---..- Laria prosopis Lec.
Prosopis glandulosa (mesquite). ....---.--- Laria prosopis.
Corciscamadensis( Ged bud) Aiea setae 22 Anthonomus signatus Say.
COSSIGTODLUSTOUG Ie Sacto. Sse sae ae Arecerus fasciculatus DeG.
COSSTGROCCI ENT Als Sas Sy han Bee Arexcerus fasciculatus.
Gleditsia aquatica (water locust).....-...... Spermophagus robin Schon.
Gleditsia triacanthos (locust)...........----- Spermophagus robime.
Vignaaunguiculata (cowpea). (2.2 222222 2. - Chalcodermus xneus Boh.
IDO PUSTORONACEHLG: ==, SE ORR ge. Ss Tychius sordidus Lec.
BO PLUSUMNCUCANING wae em ist Ton Se S75 F. Tychius sordidus.
XO PLUSUO UUILCEORUG ai anaette ee SR yer ata eS & (ches Apion rostrum Say.
PATILOT PD) Oa ULLUCOSC = ola nena eee SS poe Laria exigua Horn.
HAUG ON ChAMtIMELONUG Nam ae 52 as 2 ps Sees: Arecerus fasciculatus DeG.
COCCHRUITG UIA bree TRAE oS Apion segnipes Say.
NROUUIIUO PSCUGACHCIG He pee eh eee, Apion mgrum Sm.
LOGS) 1 55 Me IU ia eRe Ee Laria ochracea Schaef.
METOONUIGY PANICULATA Es eet aN seen sek 2 Apion decoloratum Sm.
UZ ASCOLUSI POLY SLAC US ime Ries tees Sls Apion griseum Sm.
Hg ASCOUUSHRCUUUSIUS 2 eta Var eis oe alee eeu Apion griseum.
ULOSCOULSLLURUGINULU Sets ie a Pes tocol = Apion griseum.
Phaseolus vulgaris, (bean) 22: 2302: ssc32 =. : Chalcodermus xneus Boh.
AOSCOMUSOUUGUINS a. San ES ee eee ee Laria obtecta Say.
SO PhOStYLUSIPOWCL{LOT Ors =) ps2 8.) 82) Apion griseum Sm.
Xanthorylum clava-herculis........:-------- -| Zygobaris xanthoxyli Pierce.
Mena azedarach: (China tree) 2.233252 - 2522. Arecerus fasciculatus Dietz.
Croton capitatus.......---- BE ay hase ane - Anthonomus albopilosus Dietz.
CrOLOMCNOCLINON Toe ee saan aan es Anthonomus albopilosus.
COLONMCECTIS ase: S55 ELA SaaS SNE Anthonomus albopti cea
“8 Ampeloglypter sesostris Lec.
HOS SIDIB. (GIDE) 2 ceo Sate eb ca a ace ake (ence Deeaualis Say.
Callizrhocdigitatasteot: eer iat Aah eee 8 Anthonomus fulvus Lec.
Calliinoemvolucnata ee yan smh ee cel Anthonomus fulvus.
ISP EROUCCONAMGUSTNIOUG . ido 9 nears Vk ae Macrorhoptus sphxralcix Pierce.

Anthonomus grandis Boh.
Arecerus fasciculatus DeG.
Chalcodermus eneus Boh.

| Calandra oryza L.

WNC TEE ACI CORI eats ina Bese nie eae ee Orthoris crotch Lec.
Opintian(prickhy.peat) a. se see oe Gerstxckeria nobilis Lec.
Opuntiarcigelmannt 24.22 20.25 53 ses eee Gerstxckeria nobilis.

EB DULOULULTIUES Dae eieto yah tye ee eo Tyloderma foveolatum Say.
(COIS) UAB Se See am eRe ee sini eye Na Languria sp.

ONOGTO OVENS SE 2508 IRE EISEN On Tyloderma foveolatum Say.

16844°—Bull. 100—12——6

oe ee

82 INSECT ENEMIES OF

Plant.

Galpinsia hartwegi os: = 22 =i a
Tnomed lacunosa=. SS asc55 5 Sse eee ee
Ipomea pandurata
Physalis heterophylla. fe sis 015s oh ae
Physaus lanceolatas ses pee ee ae
Physaistongyoud <a ek eee Nee
Physalis philadelphica
Physalis virginiana ambigua

Olam: CAarolinense. 205 aie i ae

Solanum eleagnifolvum
Solanum heterodox. 2.2 420. Jee ee
Solanum melongena (egg plant)......--.--.--
Solanum rostratum
Solanum rostratum

Solanum torreyr

Solanum tuberosum (potato)

Capsicum annuum (pepper) --.--------------
DOEULOSSTOMORUEMN = ee ee hae ee ee

GLUT UO ee ae een eh ae Sg
Ambrosia artemisizefolia
Ambrosia psilostachya
VAT OTOSUEETUIUAE a 3 oes hp VS eee ee As
Grendetia vnitloudes: (2-4 ae Sanam sepa
Grindelia squarrosa nuda
Heterotheca subarillaris2 ek Syae a ssemen ek
CASTER ISMIUCLIIOUUISH ame nh een ae ae kee Neer.
ISLACTONERUS TUDUGULOSUS= as sae Sara So =
Sideranthus rubiginosus.....-.----
Aphanostephus skirrobasts-~.2. 425-2222 2- 2.
Helianthus spp. (sunflower)
Helenium tenurfolium

THE BOLL WEEVIL.

Infested by—

Auleutes tenuipes Lec.
Arexcerus fasciculatus DeG.
Brachytarsus alternatus Say.
Trichobaris trinotata Say.
Trichobaris trinotata.
Trichobaris trinotata.
Trichobaris trinotata.
Trichobaris trinotata.
Trichobaris trinotata.

: penne nigrinus Boh.

Anthonomus eneolus Dietz.
Trichobaris texana Lec.
Trichobaris trinotata Say.
Trichobaris texana Lec.
Anthonomus xneolus Dietz.
ees eneolus.
| Trichobaris texana Lec.
Cio nigrinus Boh.
Trichobaris trinotata Say.
Anthonomus eugenu Cano.
epee trinotata Say.
Trichobaris compacta Casey.
Trichobaris trinotata Say.
Iizxus scrobicollis Boh.
Lizxus scrobicollis.
Iizxus scrobicollis.
Anthonomus squamosus Lec.
Anthonomus squamosus.
Anthonomus heterothece Pierce.
Anthonomus aphanostephi Pierce.
Desmoris scapalis Lec.
Brachytarsus alternatus Say.

Anthonomus aphanostephi Pierce.

Lizus scrobicollis Boh.

Baris cuneipennis Casey.

16.-A SUMMARY OF THE MORE IMPORTANT BIOLOGICAL FACTS.

1. The boll weevil has 54 enemies, including parasites and predators,

2. These enemies are native to other insects which are to be found
in the vicinity of cotton fields.

3. The interrelationships of the boll weevil and its parasites with
surrounding insects are very complicated.

4. The parasites are sometimes found in great numbers.

5. The cotton plant, by its production of nectar, furnishes a very
powerful attraction for parasites and predatory insects.

6. The development of the boll-weevil parasites is as rapid as that
of the boll weevil.

7. Most of the parasite species are well distributed.

8. The parasite species attack other hosts in the spring and have
a generation before the boll weevil is ready for them.

9. New species of parasites are becoming adapted to the weevil
each year,

ECONOMIC PRINCIPLES INVOLVED. 83

10. Other cotton insects, by their ravages upon the food of the
weevil, sometimes reduce the numbers of the boll weevil itself.

11. Much valuable work is done by the ants, which are present in
many fields.

PART III. THE ECONOMIC APPLICATION.

The economic application of parasitic control to the boll-weevil
problem is dependent upon accurate knowledge of a multitude of
conditions. The preceding two parts of this bulletin have been
devoted to a statement of the many phases of the parasite situation.
It must be understood at the beginning of this part that we consider
the utilization of parasites and other insects inimical to the boll
weevil as intimately connected with good agriculture. The boll-
weevil problem, from a parasite standpoint, is entirely different
from any other parasite problem ever studied. In other cases such
means as introductions from foreign countries may be utilized. In
the present case the main problem is to devise such agricultural
practices as will increase the effectiveness of the parasites already
present. | ;

In order to facilitate the treatment of the economic methods to be
suggested, this part is also divided into sections, which are as follows:

1. The economic principles involved.

. Interpretation of parasite statistics.

. Interpretation of the biological complex.

. How to profit by existing conditions.

. How to plan for the greatest possible control.
. Propagation and artificial introductions.

. Objectionable practices.

. The economic significance of the investigation.

OIE mb ww

1. THE ECONOMIC PRINCIPLES INVOLVED.

The attempt at utilization of insect enemies in economic ento-
mology is now receiving so much attention that the authors will
set down the principles which appear to have been the foundation of
the work they have done.

1. Insects in a state of nature are more or less completely held in
check by natural agencies, in which other insects frequently figure
as of direct or indirect importance. Many insects are controlled
almost entirely by their insect enemies. No insect is without its
natural checks.

2. The relationships between an insect and its enemies can not
be expressed by a simple ratio, nor are they in any way invariable.
The agencies operating for and against the welfare of a given species
are so many and of such inconstant magnitude, due to the activities

eee

84 INSECT ENEMIES OF THE BOLL WEEVIL.

of other agencies, that the effects of one or two agencies of control
with known strength can not be estimated, because of the many
other agencies either unknown or of unknown strength.

3. When an injurious insect escapes from its natural surroundings
to a region where conditions are favorable for enormous reproduction,
it may become a pest, but it is never absolutely free of natural checks.
Anthonomus grandis has never been free of its checks although it
escaped those of its native home. These agencies of insect control
are inherent to all countries. An insect parasite is as likely to escape
its original surroundings as a phytophagous insect.

4. When an insect finds in its vicinity a variety of food closely
related to its native host, and that food is more succulent or more
abundant, there is a possibility that sooner or later the more inviting
food will become the normal food. This possibility becomes stronger
when the original food supply fails, if the species is to be continued.
Not only phytophagous but entomophagous insects have frequently
been proven to have thus changed their food habits—whether from
preference or necessity it is not known. Leptinotarsa decemlineata
(the Colorado potato beetle) is an excellent example of this change of
habit among phytophagous insects. All of the boll-weevil parasites
are examples of parasites which have adjusted their habits in the
presence of their original hosts.

5. A crisis in the history of a species occurs whenever the food
supply fails. The species may either disperse in search of food, as
the boll weevil does each autumn, or it may hibernate or estivate,
or it may select a new host, or the species may perish. All these
results occur in nature. All of these alternatives may be chosen by
different individuals of the same species. It is safe to assume that
when a species is found to have many hosts that it has undergone
many crises and that the resultant species is a highly developed and
adaptable form. A species most limited in food habit is most liable
to restriction or extinction and consequently of a lower type than
one able to meet any emergency.

6. When a desirable parasite species is known to be adaptable to
various hosts, a crisis may be artificially superinduced by the timely
elimination of the favorite hosts, thus forcing the species to attack
the most predominant near-by related host in the vicinity, or it may
be taken to an entirely remote or foreign locality and placed near a
field containing many insects closely related to its original host,
which it may learn to thrive upon.

7. The species most available for utilization are those most adapt-
able to changing environments, or those having the most hosts in
the given locality. These other hosts will serve as nurseries for
parasites. :

)
;

INTERPRETATION OF PARASITE STATISTICS. 85

8. Certain parasites with more or less established habits require
that their hosts be in certain habitual locations (for example, Neocato-
laccus requires stem-dwelling hosts), and in like manner there are
conditions which can be made more favorable for parasite attack
through cultivation or through plant selection. Furthermore, since
parasites require different conditions, it is desirable to alter the
existing conditions so as to make them favorable for as many species
as possible. In the case of insects extending their range over many

different climates it should be the aim to introduce parasites best

adapted to the prevalent conditions.
2. INTERPRETATION OF PARASITE STATISTICS.

From the great mass of parasite statistics given in Part I a number
of important facts need to be considered.

Parasitic and predatory attack is strongest from August until frost
time. Hence it may be presumed that whatever artificial propaga-
tion is to be done will be most profitable when conducted during this
period, provided it does not interfere with early fall destruction of
stalks, the fundamental cultural remedy against the boll weevil.

The greatest control of the boll weevil by insects and also by
all agencies is in hanging squares. As has been stated in Part I
(sec. 3), the hanging squares are a result of a diagonal absciss layer,
which causes the drying square to fail in separating itself completely
from the plant. These squares die on the plant and afford a very
favorable position for parasitic attacks upon the weevils within.
The statistics show that insect control in fallen squares is greatest in
the moist States of Louisiana and Mississippi. This is undoubtedly
due to some of the new parasites which are accustomed to attacking
woodland weevils and other insects characteristic of this humid
region. The insect control in hanging squares is the greatest in the
comparatively dry States of Texas and Oklahoma. These dry States

also have a higher combined natural control in the fallen squares ©

than in the hanging squares, largely because the climatic conditions
cause a higher mortality of weevil stages in squares lying on the
heated surface of the ground. On the contrary, the humid States
have a higher mortality from both climatic and insect agencies in
hanging squares than in fallen. Furthermore, it has been proven, in

Part I (sec. 4), that an increase in the amount of hanging squares.

will increase the total control. Having these facts in mind, the
obvious conclusion is that it will be desirable to have varieties of
cotton which have this tendency best developed. Among the varie-
ties which are now known to retain their squares are the cluster
varieties, including the Rublee.

86 INSECT ENEMIES OF THE BOLL WEEVIL.

The figures show that parasitism becomes very high under favor-
able conditions and also that agriculture modifies the insect control.
Obviously therefore those agricultural methods which will favor the
highest insect control must be sought. These methods, as now
known, will be dealt with more fully in a following section.

It was feared for a long time that the parasites of the weevil would
be held in control by the warm climatic conditions which affect the
boll weevil. Thisis notso. We have found abundant proofs of the
fact that a temperature which will kill the boll-weevil larva will not
kill the egg or small parasite larva in the same cell, and that the
parasites can develop to maturity on the dried remains of the weevil.

The temperature fatal to the boll weevil is 123° F., a temperature

frequently reached on a hot burning soil. We have found in several
years that a low temperature which will kill the boll-weevil larva, is
also not fatal to the parasites, for in November, 1907, when 97 per
cent of all the boil-weevil stages were frozen, no evidence whatever
could be found of mortality among the parasites. The minimum
fatal temperature of the boll weevil is 12° F.

3. INTERPRETATION OF THE BIOLOGICAL COMPLEX.

The complicated biological factors which have been noted in Part IT
have been summarized briefly in section 16 of that part (p. 82).
The interpretation of these facts has been suggested in a number of
places throughout the second part. Hence only a few words are
necessary at this point.

The fact that surrounding each cotton field there are numerous
plants harboring weevils and their parasites is of extreme importance
in this problem. These parasites are generally capable of attacking
the boll weevil under conditions of necessity or alternative choice.
The aim is therefore to find all the methods by which these parasites
may be forced to leave their native hosts and attack the boll weevil.
In fact, the entire second part has been devoted to giving these facts
in order to bring out this single point.

4. HOW TO PROFIT BY EXISTING CONDITIONS.
COLLECTION OF COTTON SQUARES IN SCREENED CAGES.

_ As has just been pointed out, there are conditions around the cotton
fields which are potential of a considerable increase in the parasitic
control of the boll weevil. Probably no other method will yield
better results than the gathering of the cotton squares which are
infested and placing them in wire-screen cages of 16 or 18 mesh to
the inch and placing these cages in selected parts of the cotton
fields. This method is not new in entomological practice. It has

‘ Paap ieee de atiigee o's ea hts

HOW TO PROFIT BY EXISTING CONDITIONS. 87

been used with great value in the freeing of apple orchards in Europe
from the apple-bud weevil (Anthonomus pomorum). The boll weevils
can not pass through a 16-mesh or 18-mesh wire screen, while the
parasites can do so, and therefore the release of these enemies will
be constantly increasing the proportion of parasites against the
weevils. Even if a 14-mesh wire screen is used, only a small pro-
portion of the weevils can escape through it and some gain is effected
by the release of the parasites. In order to demonstrate numerically
just how this would happen, three hypotheses are presented. In the
first case squares are collected and put in a 16-mesh wire cage, and
in the second case squares are collected and put in a 14-mesh wire
cage, and in the third case no squares are collected. The average per-
centage of control which follows as a result easily demonstrates the
advantage which will be almost immediately gained.

It has been contended and proven that many weevils escape
through the ordinary wire screen.

I.—Gwen 10,000 developing stages in a 1-acre field.

(A) Collect squares containing 50 per cent of the stages and place in 16-mesh

BCHE CIC CAC pera ate ass a cd ly oS ere a San oN Ur NaC lt UR 5, 000
Normalsparacitism isto per Cemtes:s2h5 6 30) PGs eeu aes 250
AmiissamdeheateKall:AQperCemt cat! kee ce ee Si ele 2, 000
WRorsteliitiy meres ee oe Po Neh iar pyr a Ne ON AC EM 2, 250

TBURSeCG tas Se ce ar ee et a REACTED peers MTEC anager RPE NI UE Ese 2, 750
There escape through 16-mesh screen—

NOSPeMCeMt wcewals essai ke Aly ieee yi Mee) re dia er aban: Saw 275
YOM eT Comb parasites. eames ee ek eh ee Les Gk eld aie 225
(B) There remain in the field squares containing 50 per cent...............-- 5, 000
Normal mortalibyasrabove,45 percent. 2.2%. .2205 2252.8 Pose 22 ee LO 2, 250

Breed senor sss eee spss OIG NL Sa es aa my ran a ce Ca 2, 700
SomMenweewls escaped (Ol CASES sees kesh otha ok sae Pa Ue: 275

ihotalaweevals‘at:- end of eeneration=: 26.) 225.2. Fr soe. ee 3, 025
LED BISH LSS) THEA | yes eh ea ei Nols os eis ata Min Ut a LS AOR a 250
Rarasilesiescaped arOmiCaves seo. SSI Se Sn se as 225

BATASIteshPLeseni aM MlCldsscnc smh eet s Me ec we 475

There is 1 parasite to every 6.3 weevils.

II.—Gwen conditions as above, squares in 14-mesh screened cage.

CS) perom collected squares, breed aan. ty heh bonds Wavoe i ee ual | aaa 2, 750
There escape through 14-mesh screen—

ACL CONG WEEVILS. cece ac52 ved hoe oe ees BSN gee ie es 1, 100

PMI ATASICES Mama ica cir tet era ak SN ae CN Ie ks a 250
(Giebreecdiimetel dec al iy etek eae ee Mee ei Ene Tea AOS A ao as 2, 750
Hisermecdeimommca meds cs: sk xB My Ni ey GG th ane Cue De Auge LE he 00

Movaleweevals atiend ofeeneration 32-8 ses 2/2. .22 ot) e Po ee oe 3, 850

88 INSECT ENEMIES OF THE BOLL WEEVIL.

Parasites reared <<< 1.8. ad< aan See eee ee ee eee eae ee 250
Parasites escapedic: 20 s..50 ia seen oa eae A eae aoe eee eee a 250
Parasitessinfieldec: oe. ett red ene te ee yk 500

yy

There is 1 parasite to every 7.7 weevils.

ITI.—Gwen 10,000 developing stages in a 1-acre field.

No‘squares are collected ..2 0-2 oe iw aee hh ars Shee ea en ee 10, 000

Normal parasitisms oper cents? :. 57. 2 0 oe. cas 2 en ee ee 500

Ants and heat, kill 40 percent: 223.02 ses Ses Se See ee es 4,000

Mortality in esto oc 5 2 se ce oo han? cee ine one ee Jee eee 4, 500
Breedint cee fede. See 2 Ee Gd ee re 5, 500

Parasites reared, 500.
There is 1 parasite to every 11 weevils.

SUMMARY.

Squares collected and

placed in—
Squares not
collected.
14-mesh 16-mesh
wire cage. | wire cage.
NiceuiisireniaIness sae ht ce Seer Pee Rey SRE a an 5, 500 3, 850 3, 025
IPATARTECH OMAN hea cocaine 4 wean c ects = Y omiteiey ee toe 500 500 475
Ratio of parasites to weevils--.2..--...2..<2...- qe: ileal erg 1:6.3

ELIMINATION OF COHOSTS.

Another practice of undoubted value in bringing about a higher
percentage of parasitism upon the boll weevil is the elimination of
the cohosts of the boll-weevil parasites at proper times. To show
what has been done in this line, the case of the Dallas farm in 1907
may be cited. On July 19 of that year a very large hedge of weeds,
Ambrosia trifida, infested by Lizus scrobicollis was cut. These weeds
were along the fence adjoining a part of the cotton field which had
been under close observation for parasites throughout 1906 and the
spring of 1907. In 1906 there was not found in any plat on this
farm a higher parasitism than 2 per cent by Hurytoma tylodermatis in
hanging squares. Eurytoma was very numerous in the Ambrosia
weeds next to this field, but did not appear to attack the weevil in
large numbers. Before the weeds were cut in 1907 the two plats
nearest these weeds averaged 26.76 per cent and 16.79 per cent
parasitism by Eurytoma. On August 17, about a month after these
weeds were cut, the two plats just mentioned had, respectively, 37.50
per cent and 26.66 per cent parasitism by Eurytoma. This striking
gain adjoining the weeds was not reflected by parts of the field farther
removed.

ee es,

HOW TO PROFIT BY EXISTING CONDITIONS. 89

EARLY DESTRUCTION OF THE COTTON STALKS.

There can be little doubt that the early destruction of the cotton
stalks, in addition to depriving the boll weevil of its food plant, will
also cause the parasites to seek a series of hosts which can carry them
through the winter period. In order to prove that fall destruction
does not have an injurious effect upon parasite control, we would cite
the discussion of the Victoria fields, in which various methods of fall
destruction were carried out, as discussed in Part I, section 6. As a
further proof the famous Olivia fall-destruction experiments may be
considered.

On October 1 to 10, 1906, all of the cotton plants in over 400 acres,
constituting the entire Olivia cotton community in Calhoun County,
Tex., were cut and burned under the direction of Mr. J. D. Mitchell.
According to the rearing records in our possession, the parasites
developing in this cotton would all be mature before November 10,
and if they hibernated, would have to do so as adults. Noother
cotton existed within 12 miles, as the community is completely iso-
lated by water and marshland.

Cotton was planted about March 15, 1907. On April 15 no boll-
weevil work could be found, but on May 7 a single weevil was found
after a careful examination of eight fields. On the same date at Six
Mile settlement, across the bay near Port Lavaca, there was consider-
able infestation. If the parasites hibernated as adults they would be
dead long before the middle of June. If they could have hibernated
as immature stages they would have matured by March 15, and under
normal conditions three generations would have passed by June 15.
The infestation was still very slight in July. It must be argued,
therefore, that any boll-weevil parasites must be breeding on some
other weevil, if they did not perish.

On August 22, 1907, Mr. Mitchell found parasites with weevil-
infested squares on a field in the opposite part of the community to
that in which he first found the weevil infestation. The obvious
inference is that a rotation of hosts occurred during the period of the
boll weevil’s absence.

Having planned the cropping system, it is also best to prepare the
fields early for cotton and plant as early as possible. Of course, most
of the reasons for early planting of cotton are well known and the
practice is very common, but in this connection it must be said that
such early planting has the actual advantage of enabling the para-
sites to start early.

Care must be given to the choice of the cotton variety which is to
be used. Frequent recommendations have been made of varieties
with light foliage, early maturing fruit, short nodes, and determinate
growth. All of these qualities are favorable to parasite control,

A - — =
SS SSS SSE ———

90 INSECT ENEMIES OF THE BOLL WEEVIL.

especially since such plants afford much more sunlight on the ground.
The ants and also the parasites prefer much more to attack the
squares which are dried out than moist squares. It seems that they
can more readily penetrate the linings of the square. In addition to
these qualities of the cotton variety, the use of a variety with at
least a moderate amount of nectar is also advised. The reason for
this has been explained in preceding paragraphs. Finally, the tend-
ency of plants to retain the squares must be again mentioned. If a
variety can furnish the desired qualities of early producing, produc-
tiveness, and quality of lint, as well as a diagonal absciss layer on the
square, that variety should be chosen above others.

If at all possible, it is advisable to plant the rows far apart or on
the check-row system, in order to give the necessary amount of sun-
light. The cultivations to follow this should be with the purpose
of obtaining a dust mulch, for with such a mulch the surface of the
soil may be heated to a much higher degree than by deep and lumpy
cultivation, and the control of the boll weevil will thus be greatly
increased, through the drying effect upon fallen squares.

=

o. HOW TO PLAN FOR THE GREATEST POSSIBLE CONTROL.

As it has been proven that many agricultural processes are favor-
able to the development and attack of parasites and enemies, there
can be no question but that it is desirable to plan to obtain the great-
est amount of this beneficial aid.

There are a few plants which have no objectionable qualities in
themselves which might with good reason be planted adjacent to the
cotton fields in order to induce the attack of weevils which act as
hosts of the boll-weevil parasites. For instance, the presence of a
hedge of blackberries or dewberries along the fence means the pre-
sence of Anthonomus signatus, the blackberry bud weevil, with its
numerous parasites, all of which attack the boll weevil. The para-
sites would be able to carry on a generation in the spring before the
boll weevils were breeding and would mature in plenty of time to
attack the first developing stages of the boll weevils.

It would seem advisable to plant a hedge of the flowering shrub
Amorpha fruticosa, which is the host plant of Laria exigua. This
little wevil is very abundantly parasitized.

In planning the cropping system there can be no possible harm
in arranging to have a forage or hay crop adjacent to the cotton
field. In case a forage crop is used, cowpeas with the ever-present
cowpea pod weevil would undoubtedly bring about the presence
of several important parasites. The early removal of the cowpeas
for fodder would force the parasites to attack the boll weevil. In
the case of a hay field, the process of haying and subsequent curing

‘
Pp eke ete) eee ei sae Bet

do iematdielinic A. 3 See 1g, Op ES

PROPAGATION AND ARTIFICIAL INTRODUCTIONS. 91

would enable the parasites present in the various weeds to escape
and attack the most abundant host, namely, the boll weevil.

If, with all these precautions, the boll weevils are very numerous
in the field, and the expense is not too great, much can be gained by
picking the squares and placing them in cages, as has been described
in a previous section.

Finally, at the proper season for haying, the actual methods of
cutting and preparing the hay will without doubt furnish still greater
control to the weevil. Some time in September, if not before, whether
haying is carried on or not, there should be a thorough cutting of all
weeds around the cotton field in order to force the parasites to the
boll weevil and also to get rid of favorable hibernation quarters for
the boll weevil.

6. PROPAGATION AND ARTIFICIAL INTRODUCTIONS.

The propagation of parasites under artificial conditions and their
introduction are attended with a great amount of labor and expense
and have many technical difficulties. The simplest form of propaga-
tion is the collection of infested squares at one locality and the ship-
ment of them to another locality, where they are placed in the field
to await results. There are good reasons for attempting thus to
introduce parasites. It has been found by very close observations
that the parasites are not evenly distributed, but that each species
has a more or less definitely defined geographical region. This is
no doubt due to the distribution of the normal host weevils. The
purposes of introduction are to take these parasites from their native
localities and place them in geographical regions in which they do
not at present exist. Definite proofs that results can be obtained
in this manner were to be had at Dallas on the experimental farm in
1906 and also in 1907. The 1906 experiment has been fully described
in the first report on the parasites of the boll weevil (Pierce, 1908a).
In 1907 a similar experiment was tried by the release of large numbers
of adult parasites. These parasites were carried to a field in small
screen cages containing foliage, so that the parasites might not become
overheated. The cages were opened in the shade, and the parasites
allowed to fly out in any direction which they pleased. While many
species of parasites were released in this manner, they did not all
show the results that were expected, but the release of Catolaccus
wmeertus in a given part of the field accomplished an increase in the
control in hanging squares by this species. In two other parts of the
field Microbracon mellitor was released, and it also showed good gains.
As Microbracon was released on this farm both in 1906 and 1907, it may
be useful to compare the percentages of parasitism at various periods.
In August, 1906, this species furnished 8.5 per cent parasitism in hang-

92 INSECT ENEMIES OF THE BOLL WEEVIL.

ing squares; in September, 1906, this had risen to 10.2 per cent; in
July, 1907, the parasitism by this species was 35.2 per cent, and in
August, 1907, it had risen to 39.8 per cent.

At Shreveport, La., in 1908, many specimens of Catolaccus incertus
and Microbracon mellitor were released. Table XXII gives an idea of
the results and shows the expected increase by Catolaccus in both
hanging and fallen squares and by Microbracon in hanging squares.

TaBLE XXII.—Experiment in artificial introduction of Catoloccus incertus and
Microbracon mellitor, Shreveport, La., 1908.

Percentage of mortality. | Gain in mortality.
Class of forms. Plat. Date. | Total
Total. | Pata- | Cato- | Micro- 7 ae Cato- | Micro-

sites. | laccus. | bracon. laccus. | bracon.

sites.

Per cent.|Per cent.| Per cent.

Fallen squares... .- Release...| Oct. 5 | 36.44 5.93 4,23 170! jax osake ok Sees | ee ee
DOP ee eae S80 (hee Oct. 28 | 37.96| 15.74} 10.80 2.16 165 150 33
1D Yo ee Sere Check. - -- - Oct. 5 |} 40.00] 10.50 5. 55 3.100 |b as oe eee ed ee
DORR ses ue pee GOreae ee WOGta 28a) 42-35 216500 8.00} 5.00 52 44 35

Hanging squares...| Release...) Oct. 5 | 47.15 9.90 4.39 3.298). 2 ee es ee ee ee
1D eae faa Oneae er | Oct. 29 64.86 | 37.84 | 10.81 :

RELEASE CAGES.

In order to obtain satisfactory results from the release of infested
material, it is necessary to place the material in cages from which
the injurious weevils can not escape but which will still allow the
parasites egress. This principle has been explained in other sections.
There is also another important consideration in the construction of
the cages. When a large amount of material such as this is collected
in a small space it furnishes great inducements to attack by colonies
of ants. The only way that the material can be protected from total
destruction by ants is the isolation of the cage on legs by the use of
‘inverted cups” containing oil, or by greasing the legs in some
manner. .

TRANSFER OF ANT COLONIES.

Since the work of ants is always very favorable to control, means
should be devised of increasing their numbers in the cotton field.
The dust-mulch method of cultivation is very favorable to the ants
in that it does not disturb their colonies after they have commenced
breeding. This is a very important matter to consider. The late
Mr. F. C. Pratt, in working with the horn fly (Lyperosta irritans
L.) discovered that fresh manure containing numerous fly larve
is very attractive to Solenopsis, and that these ants seem. to trans-
fer their whole colony at times to the manure. Mr. Wilmon Newell,
in connection with the Argentine ant investigations, at a later
date, found that he could trap immense colonies of the Argentine

‘Sai eo Soe et Sate

Saves i ine met

Sapp kiarin a eens ner asia

OBJECTIONABLE PRACTICES. 93

ant (Iridomyrmex humilis) by means of boxes containing manure.
These observations are very suggestive, for they point out the possi-
bility that colonies of ants can be obtained by placing fresh manure
in boxes near ant colonies. When sufficient numbers have entered,
they may be boxed up for removal to a place desired. In this manner
great colonies could be transferred bodily for considerable distances.

7. OBJECTIONABLE PRACTICES.

There are several practices which are quite objectionable from the
standpoint of encouraging the parasites and most of which have
also been found objectionable from purely. cultural standpoints.
When the cotton is planted closely on moist soil its growth is mainly
vegetative and consequently immense stalks may have very little
fruit. Agriculturists have always pointed out that large cotton
plants need plenty of room in order to produce fruit. Field examina-
tions to determine the mortality of the boll weevil from various causes
have always shown that the parasitism is greatest in the portions of a
field where the foliage is lightest. A notable example was found at
Natchez, Miss., where in a single field the growth was very irregular.
One spot seems to have been used for feeding cattle and was very
fertile. On this spot the cotton grew 6 or 8 feet tall and the ground
was densely shaded. Here the mortality of the weevil was very low,
and there was scarcely any control by insects. One hundred feet
from this was a thin piece of ground where the cotton plants were
barely 2 feet high, but they were loaded with bolls and showed a
much higher percentage of mortality, especially by insect enemies.
An actual count of the number of bolls in the two parts of the field was
greatly in favor of the smaller plants.

Late planting has been proven objectionable from almost every
standpoint from which it has been viewed. Under existing circum-
stances there are no valid arguments for late planting. From the
standpoint of control by parasites late planting simply delays the
attack of parasitic enemies and reduces the amount of control in the
fall at a critical time.

It is believed that the use of varieties which always tend to drop
their squares is objectionable if varieties with the opposite tendency
can be found with the same qualities of production.

The practice of picking squares and then burning them can not be
condemned too strongly. The planters are by this practice almost
nullifying the good work that they do by picking the squares. They
are doing nothing more or less than destroying their best friends
when they burn these squares. This may be proven by an hypothesis
similar to those presented (p. 87) in demonstrating the value of collect-
ing the infested squares.

94 INSECT ENEMIES OF THE BOLL WEEVIL.

Given 10,000 developing stages of the boll weevil in a 1-acre field.

(A) Collect and burn squares containing 50 per cent of the stages............. 5, 000

This destroys all parasites as well as weevils.
(B) There remain in the field squares containing 50 per cent of the stages

PLESeM ba. 55 = 2. Sos e 2 cok He Sete | arte Sree a rete ies eee Pn a ere eee 5d, 000
Normal "parasitism oper cents. 622s eens se ee see en ee 250
Anvisrand heat kull40' percents. s2:52-k oe ee ea eee se ere 2, 000
Mortality: 25355 Societe ale dee aos a elena eee re 2, 250

Weewvills tovbreedes.2 dsc ead 304 dee a ee eee 2, 750

Parasites present in field, 250.
There is 1 parasite to every 11 weevils.

This method undoubtedly greatly reduces the total number of
weevils in the field, but, as can be readuy seen, the proportion of
parasites to weevils is the same as if nothing had been done—namely,
1 to 11. It was shown that by placing the squares in 16-mesh wire
cages the proportion would be 1 to 6.3. Hence it appears that by the
caging method the planter leaves in his field a more active agency
than he had before to attack the many weevil stages he has undoubt-
edly missed, whereas by the burning method he has not in the least
improved hie field conditions.

8. THE ECONOMIC SIGNIFICANCE OF THE INVESTIGATION.

In final summary the following points are emphasized:

I. The control of the boll weevil by wnsect enemies is sufficiently great
to gwe it a high rank in the struggle against the pest. A considerable
portion of the insect control would not be accomplished by any other
factor; hence rt 1s by no means to be neglected.

The number of species of insects attacking the developing stages
is 49.

The control in any given place consists of the combined work of
several different species.

Places having the largest number of controlling insects have the
highest percentage of control.

In many places insect control is considerably greater than climatic
control or than any other class of factors.

The average insect control is 20 per cent of all immature stages or
two-fifths of the entire natural control.

The cotton leaf-worm is a valuable enemy of the boll weevil when
it defoliates the cotton after September 1, a date beyond which new
squares can not be expected to mature. It kills many weevils by
starvation, kills many others while consuming the squares, and
finally forces a premature hibernation which is generally fatal.

ett

ee ee

ose taal spears ¥en et nc tt

peat tee

a ail ik

ee 4

ECONOMIC SIGNIFICANCE OF INVESTIGATION. 95

Il. The amount of control due to the various factors at work in any
given place should be increased if possible. Parasites can be introduced
into new fields.

- In order to prevent serious injury to cotton, the mortality of the
weevil should be above 90 per cent. It has averaged over 57 per cent
for four years and has reached almost 100 per cent several times.

While climatic influences occasionally bring the control above 90
per cent, they can not be regulated or in any way directly utilized.

Although the insect enemies present at a given place may be
accomplishing the greatest amount of work possible, the fact remains
that if other species of enemies are introduced and become estab-
lished the control can be increased.

Since certain-parasites attack the weevil preferably in dry locations
and others in wet locations, and since some prefer to attack the
infested forms on the plant, while others seek them on the ground,
it is possible to select the insect agencies so as to obtain the least
amount of lost or duplicated energy.

III. The parasites and predators which attack the boll weevil are
native wmsects, already present in a gwen territory before the weevil
arrwes.

The predators, especially ants, will attack almost any kind’ of
insect food. The parasites in nearly all cases are capable of attack-
ing almost any kind of weevil breeding in herbs above ground or in
fruits.

The parasites have proven their ability to adjust themselves to
the boll weevil by attacking it in its first generation in newly infested
territory in instances where the actual sources of the parasites were
demonstrable.

The weeds surrounding the cotton fields contain many weevils
which are harboring multitudes of available parasites. These para-

sites may be induced to attack the boll weevil by the timely elimina-~

tion of their native hosts. .

This leads to the recommendation that planters cut the weeds adjoin-
ung the cotton fields, along the roadsides, turn rows, and fences about the
tume of the maturing of the crop.

It also leads to the recommendation that a field adjoining the cotton
be used as a pasture or hay field, and that this field be mowed early in
the fall. The usual haying will also bring about the same result—
namely, the elimination of other plants harboring weevils which
attract the parasites needed in the cotton patch.

It is advisable to have in the vicinity of the cotton field such plants
as the dewberry, Croton, Amorpha, cowpeas, etc., which contain other
hosts of the boll-weevil parasites.

cS

96 INSECT ENEMIES OF THE BOLL WEEVIL.

IV. The cultural methods of controlling the cotton boll weevil are the
most favorable methods of cotton culture from the parasitic standpoint.

The tendency of the principal boll-weevil parasites to prefer light
and heat leads immediately to a long series of recommendations.

Heat on the ground is essential for the climatic control of the
weevil and also for parasitic control. Although the two factors
overlap, each accomplishes considerable independent control.

A heated condition of the ground may be accomplished by planting
the rows far apart, by check-row planting, by flat cultivation, or by
planting varieties which have short limbs, or shed ther foliage early, or
have small leaves.

Fall destruction of the cotton plants cuts off the parasites from
breeding on the weevil but sends them to other hosts upon which
they can breed a winter generation, or pass hibernation, thus gaining
not only a generation on the weevil but providing for themselves
during the winter.

The early planting of cotton in the spring makes it possible for the
earliest parasites to attack the weevil.

V. The tendency to retain infested fruit, which is displayed by
certain varieties, is worth consideration.

The fact that many more parasites are reared in hanging squares than
in fallen squares makes it desirable in humid regions to have many of
the hanging squares in a field in order to serve as a nursery of parasites
for the weevils in the fallen squares.

Varieties which show an elongate diagonal connection between
the square and stem will tend to have an imperfect absciss layer
formed. Prominent in this class are the cluster boll varieties of
cotton. Jt 1s recommended that search be made for such a variety as
will retain its infested forms and at the same tume fulfill the other require-
ments in the making of a good crop.

VI. Any step which will diminish the number of weevils and not
diminish the number of parasites in a field will of course increase the
percentage of parasites present.

The most important step of this kind 1s the collection of infested squares
and placing them in cages with a screen through which the weevils can not
escape but the parasites can.

Ant colonies may be introduced into the fields in boxes of fresh
manure.

Tf squares are collected, they should not be burned, but should be placed
im screened cages.

INSECT ENEMIES OF THE BOLL WEEVIL. 97

BIBLIOGRAPHY.

ALDRICH, JOHN MERTON.
1905. Catalogue of North American Diptera. <(Smithsonian Misc. Coll., vol. 46,
. no. 1444, pp. 426, 427.
ASHMEAD, WILLIAM Harris.
1896. Descriptions of new parasitic Hymenoptera, 1]. <Trans. Amer. Ent.
Soc., vol. 23, pp. 218-219.
1902a. A new Bruchophagus from Mexico. <Boletin de la Comision de Para-
sitologia, vol. 1, no. 9, pp. 404. (For Spanish translation see Herrera.)
Banks, NATHAN.
1904. A treatise on the Acarina or mites. <Proc. U. 8. Nat. Mus., vol. 28, no.
1382, pp. 74-76.
1906. A revision of the Tyroglyphide of the United States. <U.S. Dept. Agr.,
Bur. Ent., Tech. Ser. 13, p. 17, Pl. IV.
BARREDA, L. DE LA.
1904. El picudo delalgodén. Salvacion de la riqueza de la frontera. <(Comision
de Parasitologia Agricola, Cir. 6, pp. 14-27.
CHITTENDEN, FRANK HURLBUT.
1893a. The strawberry weevil. <U.S. Dept. Agr., Div. Ent., Ins. Life, vol. 5,
pp. 181-182.
18936. Observations on some hymenopterous parasites of Coleoptera. <U. S.
Dept. Agr., Div. Ent., Ins. Life, vol. 5, p. 250. :
1895. The potato-bud weevil (Anthonomus nigrinus Boh.). <(U.S. Dept. Agr.,
Div. Ent., Ins. Life, vol. 7, pp. 351-352.
1897. The strawberry weevil (Anthonomus signatus Say). <U. 8S. Dept. Agr.,
Div. Ent., Cir. 21, p. 4.
1899. Insects injurious to beans and peas. <U. 8S. Dept. Agr., Yearbook, 1898,
pp. 242, 245.
1902. The potato stalk weevil. <U.S. Dept. Agr., Div. Ent., Bul. 33, n.s., p.17.
1908. An injurious North American species of Apion, with notes on related
forms. <U.S. Dept. Agr., Bur. Ent., Bul. 64, Pt. IV, pp. 30-32.
Cook, Orator FULLER.
1904a. Anantenemy of thecotton boll weevil. <U.S. Dept. Agr., report 78, pp. 7,
May 27.
19046. Report on the habits of the kelep, or Guatemalan cotton-boll weevil ant.
<U.S. Dept. Agr., Bur. Ent., Bul. 49, 15 pp., August.
1905. The social organization and breeding habits of the cotton-protecting kelep
of Guatemala. <U.8. Dept. Agr., Bur. Ent., Tech. Ser. 10, pp. 55.
CoQuILLETT, DANIEL WILLIAM.
1897. Revision of the Tachinidse of America north of Mexico, a family of para-
sitic two-winged insects. <(U.S. Dept. Agr., Div. Ent., Tech. Ser. 7,
pp. 50, dl.
CRAWFORD, JAMES CHAMBERLAIN.
1907a.. New hymenopterous parasites of Anthonomus grandis Boh. <Can. Ent.,
vol. 39, pp. 133, 134, April.
19076. New North American Hymenoptera. <Journ. N. Y. Ent. Soc., vol. 15,
no. 4, p. 179, December.
1908. Some new Chalcidoidea. <<Proc. Ent. Soc. Wash., vol. 9, pp. 157-160,
April 6.
1909a. New Chalcidoidea. <Proc. Ent. Soc. Wash., vol. 11, p. 52.
19095. Two new species of the genus Tetrastichus. <Proc. Ent. Soc. Wash., vol.
11, p. 150, October 5.

16844°—Bull. 100—12——7

sys es

aI

REE RT LEE

CS I ENTE

98 INSECT ENEMIES OF THE BOLL WEEVIL.

Foster, F. ©
1908. The introduction of Indomyrmex humilis (Mayr) into New Orleans.
<Journ. Econ. Ent., vol. 1, no. 5, pp. 289-293, October 5.

GIRAULT, A. ARSENE.
1907. The lesser peach borer. <U.S. Dept. Agr., Bur. Ent., Bul. 68, Pt. IV,
p. 40.

Contains an undoubtedly erroneous reference to (Bracon) Microbracon mellitor Say.

HERRERA, ALFONSO L.
1902a. Sexto informe acerca del picudo del Algodon (Insanthonomus grandis
I. C. Cu.): Pequefia avispa paradsita del picudo. <Boletin de la
Comisién de Parasitologia Agricola, vol. 1, no. 9, pp. 403-404.
This is a translation of Ashmead 1902b.
19026. Hormiga destructora del picudo. < Loc. cit., pp. 404-406.
This contains a translation of Wheeler, 1902.

Hinps, WARREN ELMER.
1907a. An ant enemy of the cotton boll weevil. <U.S. Dept. Agr., Bur. Ent.,
Bul. 63, Pt. III, pp. 45-48, February 5.
19076. Some factors in the natural control of the Mexican cotton boll weevil.
<U.S. Dept. Agr., Bur. Ent., Bul. 74, December 14.

Hoop, CLARENCE ELLSWORTH.
1909. Types of cages found useful in parasite work. <Journ. Econ. Ent., vol. 2,
no. 2, pp. 121-124, April 15.

Howarp, LELAND OSSIAN.
1897. The Mexican cotton boli weevil (Anthonomus grandis Boheman). <U.S.
Dept. Agr., Div. Ent., Cir. 18.

HuNTER, WALTER DAVID.
1907. Some recent studies of the Mexican cotton boll weevil. <( Yearbook
U.S. Dept. Agr., 1906, pp. 318-322.

Hunter, WatterR Davin, and WARREN ELMER HINDs.
1904. The Mexican cotton boll weevil. <(U.S. Dept. Agr., Div. Ent., Bul. 45,
pp. 104-110.

List of parasites and predators; adds as new records Sigalphus curculionis Fitch, Cato-
laccus incertus Ashmead, Urosigalphus (robusius Ashmead), Bracon (dorsator Say), and
Eurytoma tylodermatis Ashmead, as well as an entomophagous fungus, Aspergillus sp.

1905. The Mexican cotton boll weevil. <(U.S. Dept. Agr., Bur. Ent., Bul. 51,
pp. 143-150.

HuntTER, WALTER Davin, WitMoN NEWELL, and WiLuIAM Dwicut PIERCE.
1907. The insect enemies of the boll weevil. <(Louisiana Crop Pest Comm.,
Cir. 20, December.

Matty, FREDERICK WILLIAM.
1902. Report on the boll weevil. <(Public Printer, Austin, Tex., pp. 23-24.

Records (Bracon) Microbracon mellitor Say, Cerambycobius cyaniceps Ashmead, and
Eurytoma sp. from the boll weevil.

MorGaAN, ALFRED COOKMAN.
1907. A predatory bug reported as an enemy of the cotton boll weevil. <U.S.
Dept. Agr., Bur. Ent., Bul. 63, Pt. 1V, pp. 49-54, February 8.

NEWELL, Wi~Mon, and R. C. TREHEARNE.
1908. A new predaceous enemy of the cotton boll weevil. <Journ. Econ. Ent.,
vol. 1, no. 4, p. 244, August 15.

+ ED ie Rtas

‘
$

BIBLIOGRAPHY. 99

Pierce, Witu1AmM Dwicut.
1907a. Notes on the biology of certain weevils related to the cotton boll weevil.
<U.S. Dept. Agr., Bur. Ent., Bul. 63, Pt. II, pp. 37-44, February 5.
19076. On the biologies of the Rhynchophora of North America. <Ann. Rept.
Nebraska State Board Agr., 1906-7, pp. 261, 263, 268, 269, 270, 271,
272, 274, 276, 278, 279, 282, 283, 285, 295, September.
1907c. Contributions to the knowledge of Rhynchophora. <Ent. News, vol. 18,
pp. 356-363 (October), 379-385 (November).
1908a. Studies of parasites of the cotton boll weevil. <U.S. Dept. Agr., Bur.
Ent., Bul. 73, p: 63, January 21.
19086. The economic bearing of recent studies of the parasites of the cotton boll
weevil. <Journ. Econ. Ent., vol. 1, no. 2, pp. 117-122, April15. |
1908c. Factors controlling parasitism with special reference to the cotton boll
weevil. <Journ. Econ. Ent., vol. 1, no. 5, pp. 315-323, October 15.
1908d. A list of parasites known to attack American Rhynchophora. <Journ.
Econ. Ent., vol. 1, no. 6, pp. 380-396, December 15.
1910. On some phases of parasitism displayed by insect enemies of weevils.
<Journ. Econ. Ent., vol. 3, pp. 451-488, December 15. :
Pratt, FREDERICK CHARLES.
1907. Notes on the pepper weevil. <U.S. Dept. Agr., Bur. Ent., Bul. 63, Pt. V,
p. 56, February 9.
RANGEL, A. F.
1901a. Segundo informe acerca del picudo del algodon (Insanthonomus grandis
I. C. Cu.). <Boletin de la Comision de Parasitologia Agricola, vol. 1,
NO.0,, pp. L7l—N io;
19016. Tercer informe acerca del picudo del algodon. <Boletin de la Comisién
de Parasitologia Agricola, vol. 1, no. 6, pp. 197-206.
1901c. Cuarto informe acerca del picudo del algodon (Insanthonomus grandis
I. C. Cu.). <Boletin de la Comisidén de Parasitologia Agricola, vol. 1,
no. 7, pp. 240-261.
RILEY, CHARLES VALENTINE, and LELAND Ossian Howarp.
1890. Some of the bred parasitic Hymenoptera in the National Museum collec-
tion. <U.S8. Dept. Agr., Div. Ent., Ins. Life, vol. 2, pp. 348-350,
S533 vol. 4) pp.l22. 123.
ScHWARZ, EUGENE AMANDUS.
1905. Note on Dorymyrmex pyramicus Roger. <(Proc. Ent, Soc. Wash., vol. 7, p.4.
TOWNSEND, CHARLES HENRY TYLER.
1895. Reports on the Mexican cotton boll weevil in Texas (Anthonomus grandis
Boh.). <U.S. Dept. Agr., Div. Ent., Ins. Life, vol. 7, pp. 295-309.
1908. The taxonomy of the muscoidean flies, including descriptions of new genera
andspecies. <(Smithsonian Mise. Coll., vol. 51, no. 1803, pp. 56-58, 86.
WHEELER, Witit1AM Morton.
1902. Formica fusca Linnzus, subsp. subpolita Mayr., var. perpilosa n. var.
<Boletin de la Comisién de Parasitologia Agricola, vol. 1, no. 9, pp.
406-407. (For Spanish translation see Herrera.)

O

VASE ORAL ae ES

SP FESS SLT EE