A study in insect parasitism : a consideration of the parasites of the white-marked tussock moth, with an account of the...


Material Information

A study in insect parasitism : a consideration of the parasites of the white-marked tussock moth, with an account of their habits and interrelations, and with descriptions of new species
Series Title:
Technical series - U.S. Dept. of Agriculture, Division of Entomology ;
Alternate title:
Insect parasitism
Physical Description:
57 pages : illustrations ; 23 cm.
Howard, L. O ( Leland Ossian ), 1857-1950
Govt. Print. Off.
Place of Publication:
Publication Date:


Subjects / Keywords:
White-marked tussock moth   ( lcsh )
Parasitism   ( lcsh )
Orgyia   ( lcsh )
Orgyia   ( fast )
Parasitism   ( fast )
White-marked tussock moth   ( fast )
abstract or summary   ( marcgt )
bibliography   ( marcgt )
federal government publication   ( marcgt )
non-fiction   ( marcgt )


Includes bibliographical references.
General Note:
I�n Insects. Philadelphia, 1850-1897, (v. 7).
Statement of Responsibility:
by L.O. Howard.

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
oclc - 04272601
lccn - agr09003097
lcc - QL462 .I57
System ID:

Full Text

Technical Series No. 9.



T~~ I T- N







Washington, D. C., February 1, 1897.
SIRu: I have the honor to submit for publication, under the title "A Study in Insect Parasitism," an account of the principal factors which brought about the almost total disappearance of tussock-moth caterpillars on the shade trees in the city of Washington during the summer of 1896, after an extremely injurious outbreak of this insect which took place during the summer of 1895. The details of this study may be consid(ered in a degree typical of many sudden interruptions of the rapid multiplication of injurious species of insects. On account of the fact that the detailed consideration of the problem involves the close record of many details of little pl)opular interest, as well as the description of a certain number of new species, it seems best that the paper should be published iii the Technical Series of bulletins of this division.
Secretary qf Agriculturc.


Introduction. ...------------------------------------------------------------
List of primary parasits. --- ..-.-..-.-.-.--------..........-.....--....--.....
A. Hymenoptera .--------.-.--..--..-.-----.....-----.-......--....---.. 7
B. Diptera ....---------------------------------------------------------- 7
List of hyperparasites. .. ...------------------------------------------------ 7
List of probable parasites .. ..---------------------------------- -------- 8
List of scavengers ...-------------------------------------------------------- S
Pihlda inquisitor .. ...--------------------------------------------------- 8
Pimpla annulipes ----------------------------------------------------------- 18
Pimpla conquisitor ---------------------------------------------------------- 18
Amorphota orgyi:e. ----------------------------------------------------------- 20
Meteorus commit ..--------------------------------------------------------- 21
Meteorus hyphantrie .-------------------------------------------------------- 22
Limneria valida ------------------------------------------------------------ 2"
Limneriasp----------------------------------------------------------------- 23
Theronia fulvescens -------------------- ------------------------------------ 21
Apanteles delicatus --------------------------------------------------------- 24
Apanteles hyphantria-------------------------------------------------------- 21
Chalcis ovata ..----------------------------------------------------------- 25
Pteronalus cuproideus ..--------------------------------------------------- 28
Cratotechus orgyi.--------------------------------------------------------- 28
Option bilineatuin. ...------------------------------------------------------ 30
Ichnieumni sabeyaneus------------------------------------------------------- 30
Ichneumon e~erul-us .----------------------------------------------------- 30
Telenomus egg parasites reared from other species of Orgyia. ------------------ 30
Teleiinius californicus -------------------------------------------------- 30
Telenomus orgyite ------------------------------------------------------ 30
HeIiiteles townseudi --------------------------------- ---------------------- 31
Bathythrix meteori.---------------------------------------------------------- 31
Bathythrix pimpLw -..------------------------------------------------------- 31
Adiastola americana -------------------------------------------------------- 32
Otacustes periliti ------------------------------------------------------------ 32
Allocota thyridopterigis .... ........ ... ..------------------------------- 32
Pernomachus insolitus ---------------------------------------------------- 33
Spilochalets debilis ------ ----------------------- -------------------.. 3
Vabrocytus thyridopterigis ..----------------------------------------------- 34
Dibrachys boucheanus ..--------------------------------------------------- 35
Syntomosphyruim esurus ------------.. ------------------------------------- 38
Eupelmus linnerie ..------------------------------------------------------ 39
Elachistus cacmea-e ---------------------------------------------------------.... 39
Elasinus atratus ...--------------------------------------------------------- 40
Asecodes albitarsis ...------------------------------------------------------ 40
Frontina aletia -.---.. .. ..---------------------- 41
Froutina frenchii. ..--------------------------------------------------- 12
Tachina ella -------------------------- -------------------------------- 42


Euphorocera clqripeniis ..------------------------------------------------- 43
11intheniia q ua(ripustulata ....---------------------------------------------- 44
Exorista griseomicans ...--------------------------------------------------- 44
The scavenger flies ....------------------------------------------------------ 44
The other insects ------------------------------------------------------- 46
General considerations ---------------------------------- .................... 48
Table showing host relations of the hymenopterous parasites................. 52
Appendix-I)escriptions of the new species:
Amorphota orgyie ....................................................... 53
Bathythrix neteori-- .------------------------------------ ................ 53
Bathythrix piipl; .- ----------------------------------------------------- 51
Adiastola americana ...------------------------------------------------- 54
Pezomnachus insolitus .... ............................................... 54
Apanteles delicatus. ----------------------------------------------------- 55
Pteroinalus cuproideus ----------------------------------------------... .55
Eupelnms imneri-e. -------------------------------------------------- 56
Elasmus utratus ......------------------------------------------------- 56
Appended note.................. .......... .................................. 57


Fig. 1. Different stages of Pimpla inquisitor -------------------------------- 9
2. Eggs, larvae and COCOOns of Pimpla inquisitor --...-.....-....--.... 10
3. Adult of Pinpla inquisitor fron side .------------------------------- 11
4. Cocoons of PiImpla inquisitor .......-................................ 14
5. Stages of Pimpla conquisitor........................................ 18
63. Cooon of Pimpla conquisitor --------------------------------------- 19
7. Amorphota orgyihe, with ncoon-..................................... 20
8. Meteors cominuis, with cocoon -- ............--------------------- 21
9. Meteors hyphantriae, with cocoon -------------------------------... 2,
10. Liumeria valida, with cocoon ....................................... 23
11. Apanteles delicatus, with cocoon ------------------------------------ 24
12. Chalcis ovata, with lupa and pupal exuviuim ------------------------ 26
13. Pupvs of Cratotechus orgyi.----------------------------------------- 29
14. Bathythrix pimpl. .-------------------------------------------------- 32
15. Otacustes periliti --------------------------------------------------- 32
16. Spilochalcis debilis, with parasitized cocoons of Meteoruis and Amorphota ....-------------------------------------------------------- 33
17. Habrocytus thyridopterigis .--------------------------------------- 35
18. Stages of Dibrachys boucheanus ..---------------------------------- 35
19. Syntomosphyrnun esurrs ..----------------------------------------- 38
20. Enpelmus liinueri -------------------------------------------------- 39
21. Elasnins atratus, with pupa .--------------------------------------- 40
22. Asecodes albitarsis., with parasitized pupa of Dibrachys ------------- 41
23. Euphorocera claripennis, with puparium ---------------------------- 43
24. Stages of Gaurax ancbora .-..........-..........-...... ......-.... 45

The white marked tussock moth (Orgyia lencostigma Smith & Abbott) is ani insect which prior to the early seventies was known mainly from the occasional defoliation of the leaves of orchard trees by its larve. Harris referred to it as an enemy of apple trees and rose bushes and sometimes of other trees and shrubs." Fitch mentioned it as an enemy of apple trees and rose bushes; while Riley, in his First Missouri Report (1869), referred to it as an apple insect, but stated that it feeds upon different kinds of trees, such as elmn, maple, horse-chestnut, and oak, seeming, however, to prefer the apple, plun, rose, and pear.
Since a somewhat indefinite date, which we can fix no closer than the early seventies, this insect has become more or less prominent as an enemy to shade trees in the cities of the northeastern United States, and correspondingly has become less prominent as an enemy to orchards. The beginning of its rapid increase in our cities is nearly coincident with the beginning of the remarkable multiplication of the English sparrows after their introduction into this country, and there seems little doubt that this coincidence is really a matter of cause and effect. One of the early results of the introduction of the English sparrow was the practical extermination by this bird of the cankerworms, which at that time were the principal insect enemies of our city shade trees. The removal of the cankerworms afforded room for the multiplication of the tussock moth, which, from the fact that its larva is hairy, was not eaten by the sparrows, and consequently multiplied with rapidity.' Furthermore, the tussock moth must be considered as one of those species which are becoming attached to cities-which are slowly altering their habits and accommodating themselves to city environment.
For many years the shade trees in more northern cities, and notably in Boston, New York, Brooklyn, and Philadelphia, have suffered severely from the attacks of this hairy caterpillar. Until recently, however, the species has not been excessively abundant in the city of

'This supplanting of the one species by the other was also undoubtedly due in part to the driving away by the sparrows of the native birds which previously had fed upon the tussock-moth caterpillars. Le Conte has shown (Proc. Am. Assoc. Adv. Sci., vol. 23, p. 44, 1874) that the larva of Ennonmos subsignaria, a measuring worm which had been very injurious to the shade trees of Philadelphia, was replaced by the tussock-moth caterpillar, through the sparrows eating the former and avoiding the latter.


Washington. Its place in that city seems to have been taken largely by the fall webworm (larva of fyphautria cunea Drury) and the bagwormn (larva of Thyridopteryx ephl)ememrf)rormis Haworth). The tussockmoth larvae, however, have been moderately common year after year, but not until 1895) did they increase so as to attract general attention. In 1895, however, Washingtoni suffered from an extraordinary outbreak of this insect, as the writer has already pointed out in his article on The shade-tree insect problem in the eastern United States," published in the Yearbook of the Del)artment of Agriculture for 1895, pages 361-384. In this season the trees along many streets were comp)letely defoliated, and in the autumn they were plastered in the most unsightly way with their cocoons.
In August, 1895, the writer commenced a careful investigation of the life history of the insect with the feeling that, although much had already been written upon the species, there was still more to be learned aboutit. As a matter of fact, new and important points in the life history have been determined in the course of this study, but none of them are of so much interest as the facts relating to the extraordinary phenomena of parasitism, induced by the unprecented multiplication of the species.
Many observations are recorded which show the prominent part which parasites take in the reduction of an insect which under favorable conditions has exceeded its normal bounds in respect to numbers. Such constant struggles between species in the apparent effort to preserve a jiist balance are met with by every observer in the course of nearly every season's observations. The writer has, however, always desired an opportunity to study the exact details of such a struggle upon a large scale, and this opportunity was alibrded in this instance.
In Boston, New York, Brooklyn, and Philadelphia the tussock moth has not seemed to fluctuate seriously in point of numbers. It has been more or less injurious every season for a number of years. In Washington, however, in 1895, there was a rapid and enormous increase, and great fears of its continuance year after year were expressed by city authorities, by the newspapers, and by citizens interested in the beauti0
fhl shade trees of the city. By the time the third generation had made its appearance in September, 1895,1 it was noticed that parasitic aInd predaceous insects were present in enormous numbers. The cocoons of the second generation were the rallying point of hordes of Ichneumon and Chalcis flies, while several species of predatory Heteroptera were present in great numbers. The trunks of nearly all of the large trees in the Government parks were sanguinary battlefields where the resulting mortality was excessive.
Prior to the beginning of these observations seven species of probable primary parasites of the tussock moth had been recorded. Fitch
SObservations of 1896 have shown without doubt that there are three annual generations in Washington instead of two, as previously stated. (See Bulletin No. 10, Division of Entomology, p. 33.)


had described two as Triclio!rammut o(kryuf and 7' *b'ateirua. Tile first of these. however, seems to be a Cratotechu-Ls which has not since been reared. while the second is a hyperp-(rasite oft the geus llolcopelte.
There is not the slig'itest evidence, however. tb:at this sec-ud parasite ever attacks the tussock moth, since Fitch simply found it wvalkiiig upon rose leaves in September, wherere" lie says, it was very probably .4earcihig for these same caterpillars iii which to deposit its eggs."
Riley rearedl Piimph! iinquisitoi-, an undetermined Tachinlid fly, a true egg p)arasite of the genuis Telenonius. and two specie.-- of the grenus Pteroniahis, while he aloobtained coco)ons,- of a Mficrogaster from tussock-moth claterlpillars, without. however, rearing the adult. The adult of what is probably this, species was reared later by Prof. F. ..Webster and was described by Mr. Ashinead a.s AIpaueles orqjYIw (B ull. Ohio Exp.
Sta., Techi. Ser. 1, p. 159). Dr. Weed records (Psyche, V, 51) ipla conquisitor S-iNorinal, 1.. July 14, 188:11. Dr. Lituer is also said to 'have reared a species of rfetrasticbuls from this larva-. This last species is undoubtedly hyperparasitic. and one if not bo)th of tile two species recorded as belonging to the genus FPteron)ialus -ire also probably
hyperpa ra sites.
The clo)se observations which were made upon abundant material
beginning early iii August, 1895, and1 carried through until the autumn o)f 1896, have resulted iiot only in vastly icreasing this list, but iii giving us some (definite ideas as to the habits of the species involved and
as to their more or less intricate iterrelatious.
The species reared are as follows:

A. 11/p ei trao.
Pimpla inquisitor Say. ,Thlev'onia fiilvescens BrullA".
Pimpla conquisitor Say. Apanteles delicatus n. sp).
* Pimpla annulipes Say Apateles liypbaiitriiw Riley.
Aiorphota. (Jrgv-N ii Chalis ovata Say.
M.Neteoruis corumunis Cresson. Pteronalus ctiproidleus n. sp.
Meteorus livphantriw Riley. Cratotecliis orgyiwe Fitch.
Limineria sp. Telenimis or,,zviak- Fiteh.
Limiieria valida ('ressoij.
Frontina aletiai, Riley. Euphorocera elaripenunis macq.
Frontina frenchii Will. E'xorista jgriseomica ns \-. (I. W\.
lachina inella *\Valk. Winthernia 4-pnst ulata Vl)b.


Hemiteles townsendi Ashniea Spi loclialci s debilis S,"ay).
Bathvthrix meteori n. sp. Eupelinuis liinneri~i ii. SP.
Bathytlirix pimplw ii. sp. Dibrachys lboucheantis ( Ratz.)'.
Adistola americana n. sp. Elachistus CaCwwociw H-ow.
Otacuistes periliti Ashi. Elasmus atratus n. sp.
Habrocytus thy ridlopterigis -\shuin. S vitomosphvruni estirt' is e)
Pezomachuis insolitus ii, sp. Asecodes albitarsis Ashrn.

Ichneumon subcyaneus Cresson. Ichneumon corruleus Cresson.
Allocota (Hemiteles) thyridopteri-is Riley.

(Reared from dead chrysalids or COcoon masses.) Diptera.
Helicobia helicis Towns. Limiosina sp.
Sarcophliaga spp. 2. Hoinaloiyia scalaris Fab.
PIlhra nigriceps Loew. Gaurax a nuhora Loew.
P hora incisuralis Loew. Neoglaphyroptera bivittata Say.
Phora fasciata Fall. Diplosis sp.
'ihora agaraci Lintner.
In addition to the insects listed above, a species of Chrysopa was reared from the cocoon mass, the larve of Anthrenwus raris were found feeding upon the d(lead pupae and empty egg masses, and a mite of the genus Tyroglyphus was found to have the same habit.
Other active agents iii the further reduction of the numbers of the tussock moth caterpillars were an undetermined disease upon which 11Mr. A. F. Woods, of the Division of Vegetabl)le Pathology, has been at work, and the I leteroptera I'odisus .spinosus, Euschistus servus, and Prion idls cristatus.
The total number of primary parasites reared in the course of the observations was 2,14 of which 185 were Diptera and 1,937 Hymenop)tera. The secondary parasites numbered about 1,000, but this is an estimate, as no exact account was kept of the many specimens of the minute Dibrachys and Asecodes. The writer deems it advisable to give in some detail the observations made upon each of these species in order before drawing general conclusions, since the basis for such conclusions must be the knowledge of the habits of the individual species involved
Pimpla inquisitor Say.

This important and widespread parasite of lepidopterous larve occurs in California, Texas, Iowa, Illinois, Missouri, Ohio, New York, and the District of Columbia, and is probably to be found in all parts of the United States, except possibly in boreal regions. It is parasitic

'All observed closely investigating recently formed cocoons of the tussock moth and apparently about to oviposit. The third species is parasitic upon Pimpla.
-The observations upon which the following statements are based have largely been made by Mr. Pergande, the assistant in charge of the insectary. Messrs. Pratt and Busck have also helped, and Mr. Coquillett was for-a short time in charge of the breeding experiments during Mr. Pergande's absence. The writer made certain field observations in the fall of 1895 and summer of 1896, and carefully dissected and examined very many Orgyia cocoons during December, 1896. The dipterous parasites have been named by Mr. Coquillett. and Mr. Ashmead has assisted in the naming of the Hymenoptera.


upon a large number of lepidopterous larv-e, feeding upon such Bonbycids as Clisiocampa and Orgyia, upon certain Phy(itids, Tortricids, and large-sized Tineids. It was by far the most abundant of the parasites reared, no less than 1,659 having issued in the course of our observations upon the parasites of Orgyia leucostigma.
Observations were first begun on August 31, 1895, at which time five eggs were noticed upon a larva which had spun its cocoon and hlAd died \\without transforming to pupa. Two similar eggs were found upon another caterpillar under like conditions.. A few days later many more were found. They were generally attached to the dorsal portion of the abdominal segments of the caterpillar and usually at the juncture of two segments and always set up on end, attached( by the small end. They were not firmly attached and were occasionally rubbed off by the contortions of the larva, being retained, however, by the silken threads of the cocoon.

FIG. 1. --'impla inquisitor: a, full-grown summer larva; b, hibernating larva; c, mouth-parts of larva;
d. adult female; e, abdomen of adult male from side-all enlarged, c greatly enlarged (original).

These eggs are much elongated, somewhat wider at one end than at the other, pure white in color, and perfectly smooth, appearing highly polished. The average length is 0.98 imm. and the greatest width 0.18 mm. The duration of the egg state must be very short and probably a matter of but a few hours. After hatching, the larvie were found to feed externally upon the body of the caterpillar, the mouth parts (closely applied to the skin, and, in fact, obviously sucking blood through a minute orifice. Their growth was rapid and there was no perceptible exuviation. The seven eggs found on August 31 had hatched the following day and by September 3 the larvae appeared to be nearly full grown. When full grown, the largest larvae are 9.5 mm. in length by 3.2 mm. in greatest diameter, fusiform in shape, and slightly curved in normal position. The color is yellowish white, some specimens taking on a pinkish tinge, and there are no markings. The


spiracles are very minute. There is no anal opening, but its position is indicated by a short curved transverse impression just ventrocephalad of the anal extremity. The mouth parts are so well shown at fig. cle as to ne'ed no description. The form and general appearance of the full-grown larva are also shown at fig. la.
On September 6 these larvae were found to be spinning their cocoons. These cocoons, at first white and afterwards turning gradually to a pale yellow-brown, becoming much the same color as the cocoon of the
()rgyia, are denser in structure and composed of a considerably finer quality of silk. They are long oval in shape, the longest reaching a length of 16 nmn. by 4 mm. in diameter. The cocoons spun by tilhe larva issuing from a single caterpillar are closely applied together and adhere so firmly that it takes some little force to separate them. They

e a b
FIa.2.-impla inquisitor: a, shrunken tussock nioth caterpillar bearing parasite eggs in situ; b. egg: c eggs in sie, dl larkne, full grown, fecedinz on spun-up caterpillar; e, cocoons of parasite within ()rgyin co on-a. d. e slightly enlarged, c still more enlarged, b greatly enlarged (author's illustration, from V-earbook, Dept. Agric.. 1895).

are applied side by side and so closely that their oval outline becomes more or loss anguhar at the points of application. The fastening together- is not exactly w-egular, since one or more cocoons will frequently extend farther out than the others. Twenty-fbur hours after the spinning of the cocoon the larne were still found unchanged, unt at forty-eight hours, or on September 8, they changed to pup,.
All of the adults issued together on September 14, leaving fifteen days as the duration of the life round, as follows: Egg, one day; larva before spnining cocoon, seven days; larva after spinning cocoon, one day; pupa, six days-total, fifteen days.
This rapidity of development is somewhat surprising. The writer has recorded a much greater rapidity witl certain Chalcidihds, but is not familiar with any record of a similarly rapid development with an


Ichneumonid. Ratzeburg (Die Ichneumnonen der Forstinsecten, p. 14, Band 1) states that the shortest life round of any of the parasitic Hymenoptera known to hiim is that of Pimpla *ftlripes, from July 15 or 16 to August 3-twenty or twenty-one days. In this instance the host was Lithosia quadra, a spun-up larva of which was found July 16 bearing eight small white Ichneumonid eggs. On the 17th the eggs had hatched: on the 19th the larva had become quite large: on the 20th they had begun to spin: on the 23d they had inclosed themselves in small white cocoons; on the 25th they had transformed to pupae: on August 3 the colors of the adult were noticeable, and on the 5th the first adult issued.
Adult females of Plimpla inquisitor were watched many times when searching for caterpillars of the proper condition for oviposition. On September 10 five individuals were observed to oviposit in tussockmoth caterpillars. The first one stung a larva which was just spinning up. The Pimpla ran up to the posterior end of the cocoon, stood for a moment with antenme vibrating on the cocoon as though to ascertain the exact condition of affairs, then suddenly ran a short distance forward bringing its ovipositor into a somewhat forwarddirected( position, and quickly inserted
it through the silk and into the body of the caterpillar, which at once conmenced to struggle violently. The wasp was not disturbed and remained ill )positioll for some minutes. The caterpillar was then examined and it was found that a Tachinid egg had already FiG.3.-Pimpjla inqui.itor: Adult female.
been deposited upon it between the from side-enlarged (orial)e
from side-enlarged originall).
second and third thoracic segments. It was placed aside to endeavor to rear the parasite, but the attempt was not successful and nothing was bred from it. A second and third one were observed to act in the same manner, the operation in each case lasting about two minutes. No adult parasites were bred in either case. A fourth oviposited in a webbed-up caterpillar, which when stung, already carried some eggs of the same species. A fifth, after running about for fifteen minutes examining a coon here and there, was seen to enter the opening at the anterior end of one of the cocoons and remained inside of it for several minutes. The cocoon was removed and examined and it was found that the caterpillar was dead and with it was one half-grown larva of this parasite, the offspring of some previous visitor. No adults were reared as the result of any one of these five observations. It was noticed on the following day that two and even three females of the Pimpla would occasionally engage at the same time in ovipositing in the same cocoon. In the


majority of the cocoons examined the caterpillars were contracted, but had not p)upated. ln one case, however, one had transformed to pupa, and near its posterior end were found a number of recently deposited eggs of the Pimpla.
Further observations upon oviposition were made on September 13. On one cocoon containing a recently spun-up larva two of the parasites were seen ovipositing at the same time. One of them, after running about for a few seconds, suddenly thrust its ovipositor into the body of the larva, which commenced to wriggle in pain, but the parasite was undisturbed and pushed its ovipositor in to its full extent, remaining motionless in this position for about five minutes, during which time the abdomen slightly moved in a peristaltic manner as if forcing the eggs out and down through the ovipositor. It then withdrew the ovipositor, turned around, took a step or two, and gave the caterpillar three thrusts in quick succession. It then went a little farther and gave several more thrusts, and continued the same operation four times more, when it again thrust its ovipositor in as far as the body, remain. ilg in this position again for several minutes. When the cocoon was examined, four eggs were found loose inside the cocoon. Here again, however, nothing was reared. Another one was observed ovipositing in a recently spun-up larva. It was seen to thrust the ovipositor in three times, lasting one minute, one and onehalf minutes, and one-half minute. It was also noticed to thrust the ovipositor in at another place in the cocoon five or six times in quick succession, as if in search of the larva. When the cocoon was examined four eggs were found, all in a bunch between the first and second segments, and in an upright position, but so loosely attached that they became detached after the cocoon had been slightly handled. It seems as though the preliminary stinging of the caterpillar were done to quiet it so that the eggs when laid would not be detached by its struggles. A parasite placed in a tube with a larva which had been stung was later found to have deposited five eggs, which were adhering to the glass and not to the larva.
An observation was mnade September 14, which indicated that all of the larvae do not feed externally, since from a discolored caterpillar collected September 4 and placed in a tube a larva of this Pimpla issued on September 14, and spun a cocoon, the adult issuing September 20.
It sometimes happens that a larva is stung when just on the point of transforming, and in such a case the transformation to pupa is occasionally accomplished. The recently formed pupa is also occasionally stung, as has just been shown. Contrary to the general rule holding when caterpillars are eaten out by the Pimpla larvae, they seem to feed within such pup&,, and eventually to spin their cocoons within the Orgyia pupa skin. Several times in December, 1896, apparently perfect female pupae of the Orgyia were found within their cocoons, with a closely applied layer of four or five Pimpla cocoons and a single break in -the


pupa skin from which the end of another Pimplacocoon slightly extruded. In all such cases the removal of the perfect pupa skin showed its contents to be a very closely packed mass of Pimpla cocoons. In one case of this kind the pupa skin was found to contain six large female cocoons of the Pimpla and one larva which, had died while engaged in spinning and had only succeeded in spinning a thin web, while attached to the outside of the pupa skin were four large perfect Pimnipla cocoons.
With this species, as with so many other parasitic iymenoptera, and indeed as with so many other insects in general, there was a marked priority in the issuing of the iales. The proportions o(if the sexes were carefully noted among those issuing in March. 1896, with the following rather striking results:
March 3..................... ----------------2 March 17-------------------....................-- 11 9
4...................---------------------... 1 18................------------------.... 10 3 9
------------------- 21 19 -------------------- 9 149
7 The-pair-i-sui-g-Mar 6- 20 --------- --------- 1 20 2
9 a------------------- 11 21 -------------------- 1 59
10 --------------------- 9 23 --------------------. 2 99
11 -------------------18 24 -------------------- 1 19
13 -------------------- 7 1 6 ------------------- 8 6
14 ---------------------- 8 I 27 ---------------------1 19
16 -------------------- 7
The pair issuing March 27 were placed in the same vial and watched. They at once copulated and repeated the act seveiteei times during the first ten minutes and six times during the next ten minutes. They then rested for a tiiie. During the next hour. however, they colpulated five times and during the succeeding two hours six times. On March 28 both were still alive, but on the 30thl both were (lead.
It is worthy of remark that these spring individuals from overwintering cocoons are smaller in size than those which issued the previous September.
It happened on several occasions that the adult Piinpla was observed to oviposit in tussock-moth caterpillars which were already infested with Tachinid larve. Several such caterpillars were isolated for observation, and in every case but one there was no development to maturity of either the dipterous or the hylvienopterous parasite. In one case, however, an adult of the Tachinid Euphoro(era claripennis issued fr6m such a caterpillar. The probabilities are that its larva was already well grown when its host was stung by the Pimpla and that the larvwe of the latter failed to find sufficient nourishment for development.
Such instances would seem to show that the maternal instinct is not so prescient as has been supposed, and that all the preliminary investigation of the host insect by the mother parasite and all the apparently anxious soundings and tappings with her antenna, while apl)pearing to satisfy her that everything is all right, do not always result in the depositing of the eggs under just the proper conditions. It is altogether likely that other parasitic Hymenoptera occasionally. and perhaps frequently, make similar mistakes, and that many parasites suffer


from this rivalry based upon erroneous instinct, as well as from the attacks of hyperparasites. Such mistakes are of course much more likely to occur during such times of extraordinary multiplication than when the species are normally abundant.
During the entire series of observations careful records were kept of all specimens of Pimpla reared. The following table indicates numbers and( dates:
Nm- Num.1" 11-1um
Date. r. Date. Nue. 1)ate. Num. Date. Num.
lwr. er. be, er.
1895. 1895. 1896. 1896.
Sept. 12 .......... 5 (Oct. 9. ......... 6 Mar. 19......... 23 July 9......... 2
13 .......... 15 11.... .... 5 20......... 21 10 ......... 23
14.......... 12 12 7 t.......... 6 11......... 22
If .......... 6 15 ......... 2 23 ......... 11 13......... 224
IT ....... 9 Nov. 6......... 2 24......... 2 14......... 50
18 .......... 12 9......... 3 26......... 14 15......... 82
19 .......... 22 14......... 2 27......... 2 16......... 100
20 .......... 19 .......... 30.... .... 7 17......... 60
21 ..........De96 1)c ..... 1 31 ......... 5 18......... 60
2 ......... 90 1896 Apr. 1......... I 20.. 100
24 .......... 76 Jan. 9 ......... 1 2......... 3 21......... 50
.......... 2 M ar. 3........ .. 2 3......... 2 22. 50
26.......... 20 4......... 1 6 ......... 3 23....... 20
27.......... 7 5......... 2 7......... 1 24.... .... 12
28.......... 4 7......... 6 9 ......... 1 235 ......... 10
0.......... 5 9......... 11 10......... 5 27......... 3
Oct. 1.......... 1 10 ......... 9 11 ......... 1 28......... 4
2 .......... 4 11 .......... 1 13 ......... 4 31......... 1
3 .......... 3 13 ......... 8 14 ......... 8 Aug. 4......... 1
4.......... 12 14......... 9 16 ........ 1 10......... 2
5.......... 7 16 ......... 7 8......... 5 26......... 1
7.......... 17.... .... 19 20......... 19
8 ......... 18......... 14 21 ......... 2 Total..... 1,659

It will be observed that the periods of greatest abundance were from September 12 to 26, 1895; from March 9 to 26, 1896; from April 10 to
20, 1896, and from July 10 to 25, 1896.
The percentage of parasitism by this species was indicated rather
plainly by four series of experiments. From 100 cocoons collected September 10, 1895, there issued 38 adults; from 9 about 500 cocoons collected August 23,
there issued 300 specimens; from about
300 cocoons collected August 21, there issued 172; from 604 cocoons collected Between June 30 and July 8, 1896,
c there issued 729.
We have not previously stated defiFI. 4.-Pimpla inqui.sitor: a, nass o(f mnale the of
cocoos: b, do., cross-sectioned;(: c,ross itely thenumber of specimens of the section of largest cocoon ,ass foun d. parasites nourished by a single Orgyia. mass of female cocouns-nat. size (origi- The number in summer varies from 1 to nal).
l). ~10, with perhaps an average of 3 or 4.
This number refers to the large well- nourished larve, mainly females. In the late autumn more are found. Ten to fifteen Pimpla cocoons in a single Orgyia cocoon are not at all unusual at this time of the year, while from a single Orgyia cocoon collected by ir. Busck in September, 1896, were reared no less than 23 Pimplas, all of which were males. This particular Pimpla cocoon mass was cross-sectioned, and is illustrated at fig. 4c.


The total number of specimens of the Pimpla reared appears to us to be very extraordinary, and in itself almost accounts for the nearly total annihilation of the first 1896 brood of tussock-mothl caterpillars.
The specimens of the Pimpla issuing ill March and April, 1896, and of these there were 252, undoubtedly passed the winter in the Orgyia cocoons and mainly in the larval stage. They were plainly smaller on the average than specimens reared in the following midsummer and the previous September. No observations were made during the winter of 1895-96 which would determine the actual condition in which the insect passes the winter, this statement depending upon observations made during the middle of December, 1896. On D)ecember 10, 1896, several cocoons of the Orgyia were opened and in some of them a number of specimens of this larva were discovered, in one instance as many as six. All were approximately of equal size and were apparently little more than half grown as compared with full-grown larvae taken in late August and early Septenber. They were, however, much contracted. Instead of being fusiform in shape. as are the full-grown summer larve, they were broad oval, the broadest portion a little behind the middle of the body. At the same time, however, their greatest diameter did not exceed that of the late summer fusiformi larva'. In spite of the smaller size, however, these larvae must be full grown, since each was inclosed in its characteristic cocoon. One of these larve is shown at fig. lb, and its size line should be compared with that of fig. la, which represents a September larva. The small size of the hibernating larva is remarkable and is unquestionably almost entirely due to contraction, as is evidenced by the shortness of the body segments and by the somewhat serrated margin and by the fact that both the head and anal end are well drawn in. In color these hibernating larve aresof a brighter and lighter yellow than those taken in late summer, and it can plainly be seen that there is a greater quantity of adipose tissue. It is worthy of remark that in the Orgyia cocoon containing six cocoons and hibernating larva of the Pimpla there was also a single adult male of the Pimpla. It is doubtful that this individual was the offspring of a prior oviposition, aind it was probably simply a rapidly developing individual from the same batch of eggs. Exact observations have not been made upon this poiit, but it is likely that, as is the casewith rapidly developing and externally feeding larva' of Euplectrus, as shown by Schwarz, the removal of an individual larva far in advance of the others results in the death of the remainder, probably through the admission of air to the wound and the consequent severe inflammation of the host.
Observations upon the method of hibernation otf Ichneumonids are sufficiently rare to render this observation on the hibernating larva' of some positive value.'
'Forstmneister Tischbein, ini Entomnologische Nuachrichten, 1871, pages l5-160, shows that with the genera Ichneumon, Amblyteles. D)icelotus, .Ethecerus, HIerpostomus, Heterischus, mand Phaogenes, the females (presumably fertilized) hibernate as a rule as adults, the males dying in autumn, lie did not determine by microscopic examination-whether these overwintering feinmales were fertilized.


We have just called attention to the fact that there were two spring periods of abundance in the emergence of the adults, viz, from March 9 to 26 and from April 10 to 20. The significance of this apparent periodicity is not that it means the development of a second generation, but that there occurred late in March and early in April, 1890, a col( spell which interrupted the issuing of hibernating individuals. As a matter of fact, there was no appropriate food for the development of an abundant generation of the parasite at this time of the year, since the tussock-moth eggs were only just about to hatch aiud other possible hosts hibernating as half or full grownl larve or as pup(ae were not abundant.
The exact length of time which the parasite remains in hibernating quarters is longer. perhaps, than would be anticipated, as is indicated by the fact that a number of the early March individuals, and, in fact, aill of those which issued up to and including the 9th of this month, came out of cocoons collected September 6, 1895, thus giving a resting period of practically six months. Those issuing later than March and in April emerged(l from cocoons taken from the trees in the spring of 189,;. but which were spun the previous September and October.
It will be noticed that a number of specimens issued during November. Whether these individuals successfully hibernated in the adult condition or whether the occurrence of a warm sunshiny day induced them to oviposit in hibernating pupe of the Orgyia (and we know from observations that, although the spun-up larva is their favorite condition of the host, they will often oviposit upon pupae) is a matter which we have not decided from exact observation. We know that certain Ichneumonids do successfully hibernate as adults and that there is this possibility with this species. It is safe to say, however, that the normal hibernating condition is that of the larva after its cocoon has been spun.
All the hibernating cocoons examined were readily distinguished by a closer, tougher, and more parchment-like consistency. There was less of the more loosely spun outer silk, although there was usually more or less of this looser silk surrounding apparently the whole mass of Pimpla cocoons in any given Orgyia cocoon.
The breeding season of the insect at Washington, then, extends from April to October. We have shown that a single generation may be produced in fifteen days, and with plenty of larval food in the proper condition there would be a possible development of about ten annual generations. As a matter of fact, if we were to consider the Orgyia as its sole food the development of the possible early generations at least would be sadly interfered with by the lack of hosts. The caterpillars of the first generation do not begin to spin up before July, and the confusion of generations of the host insect, owing to the retardation in development of some individuals and the acceleration in others, does not begin to take place in any marked degree until after this


month. There are other insects, however, in the early part of the season upon which the Pimplas do oviposit; yet it is doubtful whether under any circumstances there are more than six or seven generations annually.
As abundant and as hardy as this species seems to be, it does not escape the attacks of enemies of its own. A specimen of Euschistus serwus was observed to capture an adult female of this Pimpla when she was engaged in oviposition. At least two secondary parasites have been reared from its cocoons, viz, Dibrachys boucheanus and Allocota thyridopterigis, and it is altogether likely that one or more of the other hyperparasites mentioned will feed upon this species. It has a distinct friend, however, in Asecodes albitarsis, which is parasitic upon its p)rincipal enemy, the Dibrachys. Details concerning the three species will be given later.
The completeness of the destruction of the Pimpla by the Dibrachys and by other causes during the latter part of 1896 was surprising. About one cocoon in forty contained living Pimpla larvi. As illustrating a common condition of affairs in December, the result of a careful examination of a bunch of thirteen cocoons of Pimpla found within a single cocoon of Orgyia collected December 14 is here given: Cocoon No. 1 contained dry fragments of the Pimpla larva, four fuill-grown larve of Dibrachys boutchean us, three larve just transforming (having already voided the meconiumni) and one pupa of the same.
Cocoon No. 2 contained a dried-up larva of the Pimpla and its meconial grains. Cocoon No. 3 contained the rotten body of a Pimpla larva. Cocoon No. 4 had a small hole in its side and shriveled and mutilated remains of Pimpla larva, the meconium grains and pupal exuvia of a number of specimens of the Dibrachys.
Cocoon No. 5 contained shriveled larval skin of Pimpla and white meconial grains. Cocoon No. 6 contained rotten body of Pimpla larva. Cocoon No. 7 had a small hole and contained one pupa skin of Dibrachys, one sound pupa of same, five masses of Dibrachys meconial grains, a fragment of skin of a Pimpla larva, and two larve of a dipterous insect, probably the little scavenger fly, Gaurax anchora.
Cocoon No. 8 contained a developed but contracted and somewhat mutilated male of the Pimpla, and seven larvaw and four pupe of the Dibrachys. Cocoon No. 9 was very thin and not completed; contained shriveled body of a Pimpla larva which had died before finishing its cocoon. Cocoon No. 10 contained a bit of Pimpla skin and fragments of an unknown pupa, the adult of which had issued through a hole in the Pimpla cocoon. Cocoon No. 11, same contents as No. 10.
Cocoon No. 12 contained four larva and two pupe of the Dibrachys and a strip of Pimpla skin.
Cocoon No. 13 contained a shriveled and mutilated male pupa of the Pimpla and six larve and one pupa of the Dibrachys.
Very many more Pimpla cocoons were opened at this time, but the above is typical of nearly all and includes all of the characteristics except that at rare intervals a hibernating and healthy Pimpla larva would be found, or rather a cocoon mass all containing healthy larve, for, in general, the conditions were about the same in each mass.
11859-No. 5- 2


Pimpla annulipes Say.
This species, which is also an important and widespread parasite of lepidopterous larve, and which is particularly noted as perhaps the most efficient enemy of the codling moth in certain parts of the United States, is also a parasite of the Orgyia, although a very insignificant one, at least during 1895-96, when compared with the species which we have just considered. It occurs in California, Texas, Missouri, Michigan, New York, Maryland, Illinois, and the District of Columbia, and probably has about the same distribution as Pimpla inquisitor. It is normally a somewhat larger species and is readily distinguished by the remarkably distinct and complete black bands on the tarsi and tibia It has previously been reared not only from the codling moth,

Fiu. 5.-M'irpla cnquisitor: a. larva; b, head of same; c, pupa; d, adult female-all enlarged (d original, ~t, b, c redrawn fr om Fourth Report U. S. Entom. Com.).

as above stated, but also from several Crambids, Tortricids, and Phycitids, as well as from Papilio ajr and J)atana ministry. In March, 1890, it was bred from overwintering cocoons of Orgyia leucostigma received from M. H. Beckwith, Newark, D)el., and in the course of the extensive rearing experiments of 1895 six specimens issued from cocoons of the tussock moth between September 16 and 25. Its life history is probably very similar to that of 1. inquisitor.

Pimpla conquisitor Say.
This widespread and abundant Pimpla is one of the most important parasites of the cotton worm of the Southern States, Aletia argillacea. It is also parasitic upon the bagworm in New Jersey and the District of Columbia and is an important parasite of Phryganidea in California.


It has also been reared from the tent caterpillar of the orchard (Clisiocampa americana), from the melon worm of the South (Phakeliura hyalinata), and other Lepidoptera of economic importance. The species is found from California to New Jersey and south to the Gulf of Mexico. Cresson also records it from Canada.
With the cotton worm, the larva of which is slender, this large parasite issues only from the chrysalis, so far as we know. With the tussock moth, however, it is the larva which is parasitized. Fig. 6 shows admirably the appearance of a full-grown caterpillar from which the larva of this Pimpla has emerged and spun its cocoon. The cocoon is too big to be contained within the larval skin, which therefore splits longitudinally along the venter and serves I
simply as a dorsal cap. The only specimen at hand which has been reared from this particular host was collected in the condition shown in the figure by Mr. Coquillett, at Chestertown, Md., October 30, 1895. It was brought to Washington, and from it the adult issued from the orifice shown at fig. 6 on December 7. This would seem to indicate a hibernation in the imago stage, but it is quite possible that the emergence was premature, owing to the FIG.6.-Pimpla conquisitor: Two views of artificial temperature of the room in cocoon under skin of tussock moth caterwhich it was kept, and that un(ler pillar-enlarged (original).
normal conditions the hibernation would have been within the cocoon.
The figures of the early stages (fig. 5 a, b, c) have been enlarged from those published in the Fourth Report of the United States Entomological Commission, at page 112, and the larval mouth parts are probably incorrect. The adult, however, was drawn from the specimen issuing from the cocoon above referred to. The species may at a glance be distinguished from either of the other Pimplas above mentioned by the narrow yellow bands on the dorsum of the abdomen.,
The occurrence of the large cocoon in this instance is very interesting, since, when it infests the chrysalis of Aletia argillacea as well as the pupaw of other species, this parasite not only makes no cocoon, but spins no silk. A number of empty pupae of Aletia from which Pimpla conquisitor had emerged have recently been examined with this point in view, and no trace of silk could be found. Very many specimens were reared from the chrysalids of this host during the cotton-worm investigation. Most of them issued in the spring, but some made their appearance in the late autumn.


Amorphota orgyim n. sp.1
The interesting little Ichneumon flies allied to the genus Campoplex (Campoplegini) are general and widespread parasites of lepidopterous larve. They have been seldom reared in the United States, although something over twenty species of the old genus Campoplex are known to occur in this country. Those which have been reared have come from larvae of Noctuide or Bombycida, while in Europe they are known to attack the larvw of these two families as well as many species of Geometridae. Two species of
Orgyia in Europe are attacked
by them, viz, Orgyia gono/stigma by Campoplex carbonarius and O. antiqua by 0.
The species under consideration lays its eggs in the
half-grown caterpillars of 0.
leucostigma. Probably but a
single egg is laid in each caterpillar, since but a single larva
of the Amorphota is found to
issue. It was noticed that in
-August, 1895, a number of the half-grown larva had been
S parasitized by this species.
SThe development of the parasitic larva was undoubtedly
Fi. 7.-A morphott or i ': Adult female above, with its rapid, although no definite cocoon and shrivelled skin of the caterpillar host be- observations were made which low-enlargcd (original).
would determine its length.
IUpon reaching full growth the parasitic larva invariably leaves the Orgyia caterpillar, usually issuing from a large hole on the underside of the body behind the thoracic segments, and beginning immediately to spin at its posterior extremity a stoutish white cocoon marked near each end by a ring of large irregular brown blotches. The color of these blotches varies in intensity from a light yellow brown to a dark mahogany brown, and their size, and particularly their shape, is also variable. In addition to these bands there are smaller irregular blotches at either end of the cocoon. This cocoon is attached by its side to the leaf or bark upon which the caterpillar happens to have been crawling. This attachment is not especially firm, and the parasitic larva apparently makes no special effort to attach it, and it probably adheres simply through the viscidity of the silk when first spun. The cocoon is, however, frequently somewhat flattened on the side. In general it is subcylindrical in shape, with rounded ends, its average length being 6 mm. and its diameter 2.3 mm. SThis, with the other new species, will be described in an appendix. (See page 53.)


The discarded skin of the caterpillar host is slightly or not at all attached to the cocoon or to the leaf or bark by the parasitic larva. It is usually blown away by the wind in a short time, but sometimes remains attached by its prolegs, or perhaps by silk which it spun before its death, to the leaf or bark upon which it happens to be at the time, so that the appearance shown in fig. 7 is frequently observed.
Observations made in late August seem to indicate that the period from the completion of the cocoon to the issuing of the adult insect is only from seven to ten days. Four specimens of the adult were reared altogether. These issued two on August 29, one on September 17. and another on September 21. The remainder of the cocoons just referred to as having been more or less numerous gave out" secondary parasites, the most abundant being Spilochalcis debiliN, which will be treated more or less in detail in a later paragraph.
The Amorphota in issuing from its cocoon gnaws an irregular roundish hole at one extremity, of perhaps two-thirds the diameter of the cocoon itself, but if a cap is thus removed it does not remain, attached by a hinge and has not been observed. The orifice in none of the cases noticed was made at the9 center of the tip, but a trifle to one /' )
side, as shown in fig. 7, presenting "7 almost the appearance of having / / I /I
been pierced by a hyperparasite. r /-Meteorus communis Cresson. <7A
This common and widespread ,
species was a more or less important I
factor in the reduction of the num- I,
bers of the tussock-moth caterpillars FIG. 8.-]&,iA'r,,e lnmiui,: Adlt fe ale from in the early autum n of 1895, and side. ,iih ciity C(orogin a,,,.,)lg would doubtless have been a still more important factor in the early summer of 1896 were it not for the fact that in the autumn of 1895 it was attacked by secondary parasites to such an extent that in 1896 not a single specimen was found or reared! This insect, which belongs to the family Braconidfe (whereas the others previously considered belong to the Ichneumonide), is recorded by Cresson from Canada, ConnecLicut, and New Jersey, and occurs in the National Museum from Texas, Missouri, and the District of Columbia. It has been reared from Tineids, Botids, and Bonibycids, and with Orgyia leucostiqma attacks the half-grown larvae in much the same manner as does the Amorphota previously described. The parasitic larva issuing from the body of the half-grown caterpillar, which immediately becomes shriveled and distorted, spins for itself an oval cocoon of tough, parchment-like brown silk, over the surface of which are scattered threads of a coarser and lighter-colored silk. The toughness and


elasticity of these cocoons is extraordinary. It is almost impossible to open them without destroying their contents unless they are previously softened. The end from which the adult subsequently issues is the more pointed, and is usually lighter in color than the rest of the cocoon. The cocoon itself is attached to the leaf or bark by the coarser threads at the opposite end, and frequently this attachment is of such a nature that the cocoon hangs to a distance of several millimeters from its point of attachment. It issuing, the adult cuts off an even lid at the smaller end of the cocoon, the lid remaining attached by the coarser outer threads just mentioned as by a loose hinge. The dimensions of the cocoon are: Length, 4.8 mm.; greatest diameter, 2.2 mm. The period of development of this insect appears to be rapid, although we are able to give no definite figures. The larval development certainly takes place in less than ten days, while in. early autumn the pupa stage seems to vary from ten to twenty days. Owing to the extensive secondary parasitism, the method of hibernation was not ascertained.
Adults were reared September 19 to 23, but from
the cocoons kept later in
the season there issued
only secondary parasites.
Spilochlcis debilis, the byperparasite mentioned in
connection with the Amorphota, is also the most imI portant enemy of this

Meteorus hyphantrise Riley.

Asingle specimen of this
b a common parasite of IfyFiG. 9-Maetoru ]yh phautriw: a, adult feiiale frofl side; b, phantria cunea, described enipty cocoon showing cap-enlarged (from Bull. 10, Div. by Riley in Bulletin No. Entom. U. S. Dept. Agr.). }10 of this Division, was
reared September 23,1895, from Orgyia leucostigma. The cocoon of this species is somewhat larger and of a darker brown than that of MJfeteorus communis, and it hangs suspended by a long strand of silk, as is well shown in fig. 24, Bulletin 10, here reproduced. In 1386, the great web. worm year in the District of Columbia, this species was very abundant. Accordingto observations made that year, ten days elapsed between the completion of the cocoon and the emergence of the adult, and it was fouid that a majority of the adults had issued by November 1, making it probable that the insect hibernates in the adult condition. With this species in 1886, as with the M. communis in 1895, secondary parasitism was very marked. Out of 450 cocoons of the Meteorus placed in a glass jar the latter part of September only 70 adults issued, the remainder giving


out secondary parasites. Thus only 16 per cent of the cocoons produced the species which made them, while 84 per cent were parasitized.
Mr. Slingerland has recently reared this species in New York from Xylina laticiierea, and it is probably or may become a general parasite of free living lepidopterous larve.

Limneria valida Cresson.
This species, which is also a parasite of Hyphiatria cunea, was one of the unimportant primary parasites of the Orgyia. It was not reared in 1895, but made its appearance among the rearings of July, 1896. It issued from collected cocoons of the Orgyia, and but three adults were reared, all issuing July 17, 1896. This apparent scarceness of the species was only apl)parent, however, for many of its empty cocoons were found among the
cocoon masses of the Orgyia the following
December. The cocoon is rather long f
ellipsoidal, averaging
7.5 mm. inll length by
2.8 mm. in greatest
diameter. It is comnposed of two distinct
coverings, the outer
one of weak, closespun, crinkly, gray or
grayish-brown silk,
readily peeling off in
a sheet, and the inner one close, toug h, Fi. 10O.-Limneria ralida: Female from side, with empty cocoon at
parchent-like, dark right, and male abdomen above-enlarged (original). parchmenlt- like, dark 91
brown in color, with golden reflections, of the type common among the Ophionine. The adult issues from a round hole without a cap, gnawed almost precisely at one end, and, judging from the evidence at hand, hibernates as an imago. At all events, of the many cocoons examined December 14 to 17, 1896, all were empty. All of the cocoons found at this date were associated with spun-up larva of the Orgyia, and none with the pups of this insect. In one instance a single Orgyia cocoon contained three of Pil)pla inquisitor, all empty, and one cocoon of this Limneria, from which a hyperparasite had evidently emerged.

Limneria sp.
An allied and congeneric species issued in a single specimen on July 16, 1896. Its abdomen was unfortunately broken off, and the specimen was otherwise damaged to such an extent as to render it of uncertain specific position.


Theronia fulvescens Brull.
A single male of this species issued from a mass of Orgyia cocoons on July 24, 1896. It is a species which was originally described from Colorado, but which is more or less common in the Eastern States. It has not previously been associated with its host, so far as we know.

Apanteles delicatus n. sp.1
The small white cocoons of this species, occurring singly, attached to the bark of trees over which Orgyia larvae were crawling, were reasonably abundant in September, 1895, and iii the autumn of 1896. This species again was greatly hindered in its beneficial work by the abun. dance of secondary parasites. so much so that but four specimens of the adult have been reared. These specimens hibernated within their cocoons and issued in March and April, 1896, from cocoons
collected September 9 and 10, 1895. From other cocoons collected at the same time issued specimens of 1)ibrachys boucheanus and Spilochalcis debilis. A number of these cocoons were collected and opened D)ecember 14, 18906, for the purpose of ascertaining whether the parasite hibernates within its cocoon as .. larva or as pupa, but all, with a single exception, instead of the Fia. 1l.-Apan tls delicatus: Adult nlde, withempty original constructor of the cocomoon heclow-enlarged (original)..
coon, contained hyperparasites. Of these the most abundant was Elasmus atratus, which seems to be a very destructive enemy of this species. From two to four of its pupe were found in nearly every Apanteles cocoon examined. In one, however, were two larv of Dibrachys bouchcanus, and the single exception referred to contained a (lead adult of the Apanteles.
On September 10, 1895, the writer saw a nymph of Podisus spinosus trying to penetrate a cocoon of the Apanteles with its beak; but, although watched for some minutes, it was unsuccessful, and crawled away in search of some easier prey.

Apanteles hyphantrise Riley. This species, which is an extremely common and important parasite of Hyphantria cunea, was mentioned in Bulletin No. 10 of this division, pages 53, 54. It made its appearance early in the season of 1886, and attacked only half-grown caterpillars of the web worm. During the
Described in the appendix, page 55.


autumn its cocoons were found to coritain, in the main, secondary parasites. When affecting Hyphantria, the white silken cocoon is formed almost under the middle of the half-grown caterpillar, and is fastened securely to the object its host happens to rest upon and slightly to the host itself, which is, however, readily carried to the ground by wind and rain, and can tihercfore only be found in position in the more sheltered places, such as cracks in the bark of trees. But one Apanteles is found in a caterpillar, so that each white cocoon indicates the death of a victim. The cocoon of this species is readily distinguished from that of the foregoing species by its denser consistency, its apparently tougher silk, and greater opacity. The cocoon of Apanteles delicatus is, moreover, more firmly attached to the bark of the tree upon which the host insect has been feeding and by a greater number of attaching threads, as shown in fig. 11.
In the summer of 1895, as will be noted later, the larve of the fall webwormn were frequently found (usually in the half-grown condition) wandering about the trunks of trees affected by the ()rgyia, this peculiar condition of affairs resiilting mainly from the destru(.tion of the food supply by the very unusual numbers of tussock-moth caterpillars. Apanteles hyphaontriwc was very abundant, and a large proportion of these wandering, unprotected half-grown webworns were stung by this species. It happened in a number of cases that tussock-moth caterpillars of a corresponding stage of growth were also stung, and it can definitely be said that this Apaiiteles, while apparently preferring the webworm, is also a true primary parasite on occasion of the tussockmoth caterpillar. Its habits when attacking this latter host are similar to those just described. Its cocoons were abundant upon the tree trunks during the winter of 1896-97, and examination showed that they were almost without exception infested by hyperparasites. Among these were recognized Elasmos atratus (by far the most abundant), Dibrachys boudheanus, and Npilochalcis debilis. Asecodes albitars us was also in one case reared firom a cocoon which had been infested by Dibrachys boucheanus.
Chalcis ovata Say.
This widespread species, which is found all over the United States and Canada, and which also extends into the West Indies, is a very general parasite of Lepidoptera, issuing as a rule from their pupte. The writer in 1885 (Bulletin No. 5, Division of Entomology) recorded it as having been reared from pupe of Thyridopteryx ephemerafornlmis, Apatura clyton Aletia argyillacea, Desmia maculalis, Cacacia rosaceana, Gelechia galke-solidagin is, and Botis alnialis. It is rarely parasitic on the fall webworm (Hyphantria cunea), but is the most important of the primary parasites of the tussock-moth caterpillar next to Pimpla inquisitor.
It was first observed ovipositing in the tussock-moth cocoons on September 7, 1895, when the writer observed it in some numbers flying


about the trees defoliated by the caterpillars of this insect. A female of the Chalcis was watched while engaged in ovipositing in a fiesh chrysalis. She flew around the trunk of the tree, examining one cocoon after another, perhaps six in all, until she found a male cocoon containing a freshly transformed pupa. This cocoon she felt over from one end to the other, her antenna constantly vibrating. Finally she inserted her ovipositor, withdrawing it after a few seconds, then inserted it again and again a third time. At the fourth insertion she apparently penetrated to the right spot, for she forced her ovipositor through the silk, hugging the cocoon close and pushing her abdomen as close as possible to the pupa. In this position she remained two and a half minutes, finally withdrawing her ovipositor, walking to the upper end of the cocoon and remaining for a minute or two actively cleaning her legs and antennit, rubbing them against each other for sometime, after which she was captured. The pupa
which had been
stung was placed
in a vial, and from
it on September
14 there issued a
T- healthymalenoth.
We can not surmise the cause of
failure of this apparently successful oviposition on
the part of the parasite. On September 10 another
FIG. 12.-Chalcis orata: i, pupa; b, parasitized pupa of Orgyia; c, adult; specimen was obd, outline of same from side; e, pupal exuvium-enlarged (original). served ovipositing
in a pupa which was evidently several days old and which bore signs of having been previously parasitized. This second pupa was preserved, but did not give out an adult parasite. It died and dried up. Another specimen was observed ovipositing in a shrunken larva which bore a Tachinid egg on the dorsun of the sixth abdominal segment. The movements of a large dipterous larva could be seen through the skin of the caterpillar. This specimen also was placed in a vial, but eventually dried up and no parasite issued. The two adult specimens observed ovipositing on the 10th were placed in vials with cotton stoppers and on the 11th were still alive and active. They were removed to different tubes and provided with pupe of the tussock moth in an advanced state of development. One of them oviposited in the same pupa twice and the other once. On the following day (September 12) both died. On the 13th the moth issued from one of the pupo oviposited in on the


11th, the egg of the parasite having been laid too late to interfere with the development of the adult Orgyia, or perhaps the already greatly toughened skin of the pupa was not penetrated by the ovipositor of the Chalcis.
Owing to the failure to rear parasites from pupw or larvTe in which the adults were observed to oviposit, the length of the life round can only be surmised. It is unquestionably, however, very short. The tussock-moth caterpillars were transforming to pupae rather rapidly at the end of the first week in September, and, as just noticed, the Chalcis flies were abundant at this time, and were actively engaged in ovipositing The first adult of the parasite was reared on September 17 from a discolored pupa found September 13. The second one issued on the 18th .from a discolored pupa taken on the 17th, and the third on the 19th from a pupa taken on the 4th. The longest period under observation in the fall of 1895 occurred in the case of certain cocoons taken on August 23, from which the majority of the Chalcis flies issued from the 19th to the 25th September, and from which a yery few belated individuals issued as late as October 19. The probabilities are that at this time of the year the entire life round of the parasite occupies on an average from three to four weeks. A little later it is naturally lengthened. From the mass of cocoons taken September 10 the last Chalcis issued October 4.
In issuing from the Orgyia pupa this Chalcis invariably gnaws a hole through the dorsum of the first and second thoracic segments. This hole is usually a little to one side of the median line.
It is an easy matter to recognize the Orgyia pupie which have been parasitized by this species. -Not only is the large round hole near the head of the pupa characteristic of Clalcis ovata-that is to say, characteristic among the parasites of the tussock moth-but if the chrysalis be broken open the pupal exuviuni of the Chalcis will always be found, and this is in itself so characteristic as to be recognized at a glance. The sheaths of the large hind femnora and those of the corresponding tibiae and tarsi projecting straight out at nearly right angles from the femora distinguish this exuvium from that of any of the other parasites. It is well shown at fig. 12 e. Very many of these parasitized )up-' of the Orgyia were examined. Never more than a single pupal exuvium of the Chalcis was found. If more than a single specimen develops in a single Orgyia pupa, it has not been observed. This is rather strange, for it is seldomn that the entire contents of the body of the pupa are devoured by this parasite, and in many female pupe from which an adult Chalcis had emerged fully one half of the- eggs were undisturbed and sound.
No instances have been recorded in which this insect seemed to be hyperparasitic, but there was in the collection a swollen larval skin of an Orgyia caterpillar apparently parasitized by some other insect, from which had issued from the anterior ventral segment a specimen of this Chalcis, according to the label. The specimen was carefully dissected


and was found to contain a dipterous puparium about the size of that of Exorista griseomicans, and from a hole in its anterior end some insect had emerged.- No trace of the pupa skin of the Chalcis could be found, however, and I am of the opinion that the label was incorrect.
In all, 260 specimens of Chalcis ovata were reared, and dates of issuance will appear from the following table:
XNum- NuinDate. P'lte. Date Dll- 'll-ate. N m
i)ate. ber. be r. 1)ate, ber. ber.

1895. i895. 1895. 1896.
Sept. 17 ---------- 3 Sept.30 ......... 2 Oct. 17 .......... 1 July20 ........... 2
18 .......... 3 Oct. 1 ........ 1 19 ........... 1 21 .......... 7
19 .......... 14 2 .......... 3 26. --------- 1 23.......... 7
20 .......... 14 3 ......... 1 1896. 24..... .... 3
21 ....... ..... 22 4 :1 July11.......... 25 ..............7
23 ........... 39 7 ......... 1 13 .......... 1 27.......... 5
24 .......... 18 8 ....14....... 3 28........ 8
26 ......... 18 11 ......... 1 16 ........... 8 30........ 2
27 ....... .. .8 15 ......... 1 17 ..........
28 ........ 5 16 ........ 3 18 .......... 3 Total ..... 260

It will be observed from this table that no rearings were made in the spring of 1896 from cocoons taken the previous fall or from cocoons taken in the early spring. From this fact it seems reasonably certain that this insect hibernates in the adult condition.
We have no positive evidence that it possesses secondary parasites, but specimens of Dibrachys bouckeanus have been reared from pupe of the Orgyia, the first inhabitant of which we do not know, but it may very well have been this species.

Pteromalus cuproideus ii. sp.
On October 5, 1895, two specimens of this parasite were reared from the 100 Orgyia cocoons collected Septenber 10 to ascertain the percentage of parasitism. All of the species belonging to the section of the genus Pteromalus, to which this belongs, are primary parasites on lepidopterous larve, and we may therefore safely consider that this species is a primary parasite of the larva of the tussock moth.

Cratotechus orgyim (Fitch).
The adult of this insect has been reared only by Dr. Fitch. In his Second Report ovi the Noxious, Beneficial, and Other Insects of the State of New York (separate edition, p. 216, 1856) he gives the following account of this insect:
I once gathered two of these caterpillars, which I placed, with some leaves, in a box. Two days afterwards one of them was found to be dead, and the other, being lively and vigorous, was removed to another box. Next day what appeared to be the ends of little worms were seen protruding from the body of the dead caterpillar. Upon the following day these worms were found to be seventeen in number. They had all left the dead carcass of the caterpillar, and just above it on the side of the box 'Described in the appendix, page 55.

they had arranged themselves in a circular row and had changed to pupae of a milkwhite color, 0.12 [inch] long and half as broad, hanging by their tails, with their heads downward and their backs against the side of the box. This was upon the last day of July. Next day they had changed to a pale red color and had somewhat shriveled, each having discharged a little cluster of clay-yellow grains, which were adhering to the side of the box at the tip of their bodies. They subsequently altered to a black color, and on the 6th of August they hatched the winged insects, which were of a brilliant brassy-green color, with a blackish-purple abdomen and white legs, and about the same size as the pupe.
Both Dr. Fitch's account of the method of transformation and an examination of the fragmentary specimens still remaining in his collection, now in the possession of the United States National Museum, indicate that this parasite belongs to the genus Cratotechus.


FIG. 13.-Cratotechus orgyie: Pupa, in situ below, with greatly enlarged pupa above (original).

In December, 1896, a few specimens of the pupa of a Cratotechus or some closely allied Eulophine genus were found among the cocoons of an Orgyia, and we assume that they belong to this species. Efforts to rear the adult, however, have failed up to the time of writing since all, or nearly all, had been stung by a hyperparasite-Elachistius caceciw. A group of seven of these pupe attached to a leaf about the shriveled remains of a half-grown Hyphantria larva is shown at fig. 13, while one of the pupe enlarged is illustrated in the same figure.
These exposed pupa of Cratotechus are frequently in other instances parasitized by species of Tetrastichus, and it is possible that the species of this genus reared by Dr. Lintner has for its true host Cratotechus orgyi .


Ophion bilineatum Say.

This species is mentioned in this connection with, unfortunately, no certainty as to whether it is a parasite of the Orgyia or not. The reason for introducing it is simply that in December, 1896, a single specimen of its easily recognizable cocoon was found in a mass of the Orgyia cocoons and apparently inclosed in the outer envelope of one. The cocoon was opened, but the adult had issued at some previous time. On the whole, it is not likely that it was a parasite of the tussock-moth caterpillar. Ordinarily this species is reared from larger Bombycids, such as the Arctians and Saturnians, and it is quite likely that the larva making this cocoon had issued from a larva of one of these groups, had spun its cocoon upon the bark of the tree where it was subsequently found, and had afterwards been spun over by the Orgyia larva' in making their own cocoons.

Ichneumon subeyaneus Cresson.
This species was observed in September investigating full-grown larva, and recently spun cocoons of the Orgyia. The writer watched several specimens, and has little doubt that it may be put down as a parasite of this species, although no specimens were actually reared. It is a widespread species recorded from Canada and generally throughout the United States, and is said to have been reared from several Bombycids, as well as from nciatus ventralis.

Ichneumon ceruleus Cresson.
The same statement may be made regarding this species. It was observed in September, 1895, under exactly similar circumstances, and has practically the same distribution, but has been reared only from pupae of Bombycids.

Telenom us cablfornicus Ashmead.-There are four female specimens of this genus in the collection of the United States National Museum, which were reared by Mr. Ehrhorn from an egg mass of a species of Orgyia at Mountain View, Cal.
Telenomis orgyim Fitch.-This minute egg parasite issues from the egg mass of the closely allied Orgyia nova. With this insect the eggs are left naked and not covered with the white, frothy substance with which the female of 0. leucostigma covers and hides her eggs. It may result from this fact that this parasite will not be reared from the egg mass of the species of Orgyia under consideration. It has been reared from the eggs of a species of Orgyia at Ottawa, Canada, by Mr. W. H. Harrington, as well as by Dr. Fitch in New York. The species referred


to on page 34, second edition Bulletin No. 10 of this division, is probably T. orgyiw, but we are unable to substantiate the statement that it has been reared from eggs of 0. leucostigma.

Hemiteles townsendi Ashmead.
With the consideration of this species we begin the hyperparasites, leaving the primary dipterous parasites to be considered after all of the Hymenoptera have been treated. So far as known, all of the species of Hemiteles, as well as of at least most of the genera into which it was subdivided by Foerster, are hyperparasites. They are reared from all sorts of lepidopterous larvae and pupe, from dipterous puparia, from oak galls, and from spider cocoons, but in every case, so far as the published records go, where the exact host relationships have been determined, they have been found to be hyperparasitic. There is a species, however, at Washington which is very deceptive in this regard. It issues from the egg cocoons of certain spiders, and a close examination of the cocoons from which it issues fails to show even the fragmentary remains of a primary parasite.
In the case of the present species (Hem iteles towinsendi) no such difficulty exists. The type specimen was collected in Michigan by Mr. Townsend, and in December, 1889, another specimen was reared from the pupariim of a dipterous insect at Washington. There have been four specimens reared from the cocoons of the tussock moth. All four issued in early November, 1895, and it is practically safe to say that they came from contained puparia of one of the Tachinid parasites of this insect. Of the four specimens, three are females and one is male.

Bathythrix meteori n. sp.'
A number of specimens of a very distinct Hemiteline were reared during 1896. The first specimen issued March 5 from a cocoon of lleteorus communis, collected in December, 1895. All of the others, thirteen in number, issued between July 21 and August 5, 1896. from the mass of 624 cocoons of the Orgyia collected between June 30 and July 8, 1896. It is worthy of remark that not a single specimen of the Meteorus was reared from this lot of cocoons. Of the thirteen specimens, eight were females and five males.

Bathythrix pimple n. sp.2
Four specimens of this insect were reared August 29, September 18, 28, and 30, 1895, from masses of cocoons of the Orgyia under observation in the rearing cages. Its exact host relationships are not known, but it is probably a parasite of Pimpla inquisitor.
Described in the appendix, page 53.
Described in the appendix, page 54.


Adiastola americana n. sp.i Three specimens of this species, two females and one male, were reared November 6, 1805, from cocoon masses of Orgyia under observation. This species also is likely to be a parasite of Pimpla inquisitor.

Otacustes periliti Ashm. A single specimen of this species issued frt-om a cocoon of Meteorus communisMarch 5,1896,the
Meteors havingpreviouslyparasitized an Orgyia larva. This species was reared from the same host at Washington in 1882, but has not since been found.

Allocota thyridopterigis (Riley). A single specimen of this inFlo. 14.-Bathythrixpimplr: Female-enlarged (origi- sect, which is commonly reared
nalj. from the bags of the bagworm
and which was first described by Hiley as a parasite of this species, was seen crawling rapidly over the trunk of a poplar tree in September, 1895, investigating Orgyia cocoons. It was doubtless attracted by


FlaG. 15.-tacustes periliti: Female, dorsal view, at right; same, ovipositing in Meteorus cocoon, at
left-enlarged (original; the latter from field sketch by Pergande).

the abundance of Pimplas, and as Riley has shown that it is a hyperparasite of Pimpla con quisitor, we may safely put it down as a secondary parasite in the case of Orgyia leucostigma, the more especially as no 'Described in the appendix, page 54.


specimens of the bagworm were noticed in the vicinity. That this species should be attracted to the Pimpla only when contained within the bags of the bagworm would be somewhat of an anomaly.

Pezomachus insolitus n. sp.'
The species of the genus Pezomachus may or may not be hyperparasites. The genus somewhat resembles Remiteles, and there have been published statements to the effect that the two genera are identical, the one comprising the one sex and the other the other sex. These statements, however, are due to a faulty appreciation of the generic characters and to the fact that species of each genus are often reared from the same host, notably from the egg cocoons of spiders. So close an observer as Foerster, as a matter of fact, placed the two genera in

Fi. 16.-Spilochalcis debilis: a, adult: b, antenna of same; c, Meteorus cocoon, and d, Amorphota
cocoon from which this species has issued-enlarged (original).
different families, the one being the type of the Hemiteloidae and the other of the Pezomachoidm. Three male specimens of the species under consideration were reared November 9 and 19, 18951 from the cocoon mass of the Orygia.
Spilochalcis debilis (Say).
This beautiful little species, which is recorded by Cresson from Delaware, Indiana, and Illinois, and is also known to the writer to occur in North Carolina and Long Island, New York, has frequently been reared in the course of these investigations. There can be little doubt that some of the species of Spilochalcis are primary parasites of Lepidoptera. Others are with equal certainty hyperparasites. For example, Walsh's Smicra albifrons, which belongs to this genus, is with little doubt a tertiary parasite, since it issues from a Pezomachus which is parasitic upon a Microgaster which destroys the army worm (larva of Leucania unipuncta).
'Described in the appendix, page 54.
1-1859-No. 5- 3


The present species is a very important hyperparasite. It was extensively reared at this office in 1886 from the cocoons of lMeteorus hyphantrim and is the species referred to as Spilochalcis sp. on page 57 of the second edition of Bulletin No. 10 of this Division.
During the observations upon Orgyia parasites in 1895-96 it was reared from the cocoons of Amorphota orgyia, of Mleteorus communis and I1. liyplhantrit, and of the two species of Apanteles just considered. Moreover, empty cocoons of Limneria ralida have been found pierced by holes of about the size of the exit holes made by this species, so that it is very possibly a parasite of this species also. There is a further possibility that it may parasitize Pimpla inquisitor, although I have no definite proof.
The cocoons of the Meteorus from which this insect has issued may readily be distinguished from those cocoons from which the Meteorus itself has issued by the fact that the Spilochalcis makes a round hole before the tip and does not cut off the circular hinged cap which iA characteristic of the issuing hole of the Meteorus. This appearance is well indicated upon fig.16. With a lesser degree of certainty, the same distinction may be made with regard to the cocoons of the Amorphota. Here the issuing hole of the true maker of the cocoon is not only considerably larger than that of the Spilochalcis parasite, but it is usually made nearer to the exact end of the cocoon than is that of the parasite. The same may be said of the Apanteles cocoons. It is noticeable that the specimens which issue from the Aminorphota cocoons are somewhat larger than those from the Meteorus cocoons, while those from the Apanteles cocoons are still smaller. No observations have been made upon the early stages of this species. The total number of specimens reared is twenty. Fifteen of them issued between September 3 and October 5, 1895; one January 4, 1896, and the others April 20 and 22, 1896. Those issuing in the autumn consisted of both males and females, while those issuing in April were all males, were darker in color, and all came from the cocoons of the Apanteles. These facts suggest a dual hibernation, either as adults or as larve or pupa within the cocoons of the host insect, but it is worthy of remark that no specimens of this insect in any stage were found during the extensive examinations made the middle of December, 1896.

Habrocytus thyridopterigis Ashmead.
This Pteromalid is rather commonly reared from the bags of the bagworm, in which the writer has assumed it to be a tertiary parasite, having for its host Riley's Htemiteles thyridopterigis, which in its turn is in all probability parasitic upon Pimpla conquisitor, so often found in the bags.' Eight specimens only of this species have been reared in the SAn observation is recorded in Proc. Ent. Soc. Washington, I, 28, showing that a Chalcidid larva feeds in the bags on the larva of the Hemiteles. This chalcidid is with little doubt Habrocytus thyridopterigis.


course of the Orgyia observations, and all practically between April 16 and May 18, 18967 in the jars containing cocoons of the tussock moth collected in March and early April. Of the eight specimens. four were males and four females. The probabilities are this species overwinters within the cocoon of its
host. What its true host
may be we can only surmise. It is probablyoiie of
the Hemitelines parasitic
upon Pimpla inquisitor.
Dibrachys boucheanus
This widespread and
very abundant hyperparasite is common in Europe,
and was frequently reared Fii. 17.-Habrocytbs thiridopteri(qii-enlarged (original,. by Ratzeburg, Brischke,
and later observers, from different hosts. Ratzeburg, in the first volume of his Ichneumonen der Frostinsecten, pages 196-197, states that he reared a number of specimens as secondary parasites iii the year 1842, at the close of the feeding period of Liparis dispar (the gypsy moth). He took many Microgaster cocoons in July of that year from the larve of the gypsy moth found dead upon the trees. Hardly half of the MicroA6

FIG. 18.-Dibrachys boucheanus: a, larva: b, upa: c, adult female-greatly enlarged; d. head of larva; e, antenna of adult-still more enlarged (original). gasters issued from these cocoons, the remainder giving forth this hyperparasite, which issued from needle-like holes in the cocoons, the true makers of the cocoons issuing from holes made by cutting off circular caps. He found that in some cases the larvie of the Microgasters were wholly eaten out by the parasite and prevented from pupating. In other cases the larve retained sufficient strength and substance to


transform to pupe, and were killed in that state by the parasite. In one case he found a Microgaster pupa which had been destroyed by the pupa of a Pteromalid, from which a tertiary parasite, which he considers possibly to have been this species, had emerged. The development of the hyperparasite he found to occur rather late, for on the 18th of December he had living pupe and adults and even one living larva, although the gypsy-moth larve had been killed as far back as the month of July by the Microgasters. In his second volume, pages 189-190), he states that he bred this species in 1847 from Microgaster cocoons from Bombyx salicis, and at the same time from hanging cocoons such as Perilitus spins. From neither of these kinds of cocoons did he obtain the true maker, but only this parasite, together with two other secondary species. In his third volume, pages 231-232, he mentions an instance, on the authority of Tischbein, in which this species was probably a tertiary parasite, destroying Hemiteles socialis, which itself was parasitic in Microgaster cocoons.
Brischke records the rearing of this species from Microgaster glomeratus. f1L fulripes, and MeCteorus unicolor.
Mr. Ashmead. in the course of his careful studies of the Pteromalinm of the Uited States, has decided that Fitch's Cleonymus clisiocampCe, reared from thle chrysalis of Clisiocampa americana and stated by Fitch (erroneously, it Mr. As i head's determination be correct) to be a primary parasite of this species, is identical with this European species.
It has been reared as a hyperparasite from several insects at this office. It is recorded among the hyperparasites of Hyphantria cunea, under the name of Pteromalus sp., in the revised edition of Bulletin No. 10 of this Division as parasitic upon Apanteles hyphantrie, Meteorus hypianttriw, and Limneria pallipes ( -= ralida).
In the observations upon the parasites of the tussock moth this species was by far the most abundant of the Chalcidids reared. It is parasitic upon Pimpla inquisitor, upon both of the species of Apanteles, upon both species of Meteorus, and probably also upon Limneria ralida when affecting this host. It was first reared August 30, 1895, from one of the isolated cocoons of Pimpla inquisitor, and in the course of the ensuing year hundreds of specimens were reared. Nearly 400 were counted, but many more issued. The mere labor of counting the specimens on some (lays of abundant issuing from the mass of Orgyia cocoons placed in the breeding cages was too great to be undertaken. From August 30 to November 6 a number of specimens issued nearly every day. The most abundant periods of issuing were about the 1st of September, the last of October, and the 1st of November. None were reared during the winter of 1895-96, but they began issuing again from overwintered cocoons on March 31 and continued until the end of June, beginning again toward the end of July and issuing at intervals through the rest of the season. The number issuing from a single specimen of the host seems to depend entirely on the size of the host.


Although many host cocoons have been dissected, an instance has never been found where the host larva (and it is the larva apparently vhen full grown which is attacked) was not almost entirely destroyed, the only remaining fragment being the strip of shriveled skin. The only exception to this rule which the writer has noticed was in the case of an Apanteles cocoon which was examined December 31, 1896. In this cocoon was found a single pupa of the Dibrachys, while the perfectly developed Apanteles showed simply a large cavity at the base of the underside of the abdomen. The largest number observed to issue from a single cocoon of Pimplt anlipes has been twenty-three. In the instance in which this count was made the specimens emerging were counted the cocoon was cut open and the number of pupal exuvia was found to correspond. Successful transformation had been effected in every case and the adults had all issued through two holes, one near one end of the Pimpla cocoon and the other near the other end. Pimpla cocoons were found, however, during the winter of 1896 in which there were as many as twelve exit holes of this parasite, while in another instance twenty-two specimens of the Dibrachys issued from a single hole in a Pimpla cocoon. This time, however, the hole was near the middle of the cocoon.
A number of cocoons of Apanteles were cut open December 13, 1896, and in one case contained two larvc of this I)ibrachys, which transformed to pupe December 29. The pupa, were at first milk-white, and no darker color, even in the eyes, was observed. Thirty-six hours later the eyes had become red, and ten days from pupation (January 8, 1897) the adults emerged. From a single cocoon of ieaeorus cowmn nis the highest number of this species reared was six.
Judging from the large number of specimens which issued at the end of October and first of -November, and from the fact that the dissections of the Apanteles cocoons Just referred to in )ecember showed larvwl and pupae, this insect seems to hibernate both as adult and as larva. The midwinter condition of this species in the cocoons of Pinipla is described in the closing paragraphs of the section on Pinpla inquisitor. It occurred both as larvwe and pupae December 13 to 16, and adults issued in warm rooms December 27.
The larva and pupa, as well as the adult, are shown at fig. 1 The simple nature of the mouth parts of the larva is especially brought out. So far as can be ascertained, the maxille are not differentiated, nor is there any differentiation of the parts of the labiunin. The only parts observed are mandibles, labrum, and labium. The mandibles are very minute and are excessively sharp, looking like minute slightly curved spicules.
A singular and practically inexplicable instance was observed December 16, 1896, where in a cocoon of the Orgyia a dead female chrysalis was found in an advanced stage of development. The moth was apparently just ready to emerge at the time of death. All of the scales on


the body and legs were fully formed and the wings were also fully developed. On breaking the body across, in the interior of the abdomen were found two active living larv, which were entirely indistinguishable from the larva of this species. The specimen was put aside to await developments, and the writer has at this time no explanatory comments to make.'
This species has been observed to oviposit in freshly spun cocoons of Meteorus and of Amorphota. It has also been seen investigating cocoons of the Orgyia, presumably for the purpose of ascertaining whether they contained Pimpla cocoons. It oviposits customarily in parasite cocoons containing the as yet untransformed larva, although it may also lay its eggs in the parasitic larvvc which have just issued from the host and are about to begin spinning. We have no evidence that this species is often a tertiary parasite, but it is quite within the bounds of possibility that it may become so by virtue of ovipositing in cocoons of Meteorus,
Apanteles, or Amorphota which have already been stung by
Spilochalcis debilis.

Syntomosphyrum esurus

This interesting little hyperparasite was
first reared from the
cotton worm of the
South (Aletia argillacea) in the course of
the cotton-worm investigation by the U. S. Department of Agriculture in 1878. In Cornstock's report on cotton insects, 1879, and in Bulletin No. 3 of the U. S. Entomological Commission, it was mentioned under the name Cirrospilus esurus. In the Fourth Report of the U. S. Entomological Commission it is designated as Tetrastichus esurus. The chrysalids of Aletia formed during the latter part of the season were frequently found infested with this parasite, each chrysalis nourishing a number. The parasite was found to be generally distributed and was reared in Texas, Alabama, and Georgia. Adults issued all through the autumn, during mild winter weather, and in the spring. It was considered in the first two reports published to be a primary parasite. In the fourth
'The only recorded instance of an at all similar case known to the writer was given by Rev. T. A. Marshall in the Entomologists' Monthly Magazine for December, 1896, in which an ichneumonid larva is said to have been found alive in the body of an adult Acherontia atropos. The writer's record (Proc. Entom. Soc. Wash., I, 95) of the rearing of Ichneumnon instabili8 by Scudder from an adult of Wnei8 8ernidea was based on a wrong reading of a manuscript note of Scudder's.


report the statement is made that it is possibly not a primary parasite, but that it infests one of the larger Ichneumonid parasites of the cotton worm. "This supposition is, however, contradicted by the fact that Aletia puple parasitized by this Chalcid are always found packed to overflowing with the Tetrastichus, whereas were the latter simply parasitic upon Pimpla or Chalcis they would in all probability be found only in the abdomen of the Aletia pupa."
In 1894 Mr. Ashmead (Trans. Am. Ent. Soc., Vol. XXI. pp. 343-344) described a parasite reared June 10, 1891, by Dr. A. D). Hopkins, at Morgantown, W.
Va., from Orgyia
leucostigma on miapie, as Syntomosphyrum orfyia'.
This species proves __.
to be identical with
Riley's esurus
and introduces it
among the tussockmoth parasites. It
has not been reared
at Washington
from the tussock
FI. 20.-Eupelws lim ieii F cmale-enlarg ed (original).
moth, but during
the spring of 1896 a large series of specimens was reared from overwintering chrysalids of the fall webworm. A single parasitized chrysalis was examined after these parasites had emerged and no trace could be found of a primary parasite. The exact host relations of this species must therefore still remain in doubt, although from group habits it is reasonably safe to say that it is a hyperparasite. In this event it may be secondary (feeding upon Pimpla) or tertiary (feeding upon Bathythrix).
Eupelmus limneriae n. sp.'
But two specimens of this insect were reared in the course of the investigations upon the tussock minoth, both issuing from cocoons of Limneria ralida. It is the same species which was recorded in Bulletin No. 10 as being a hyperparasite upon Limneria and upon Meteorus.

Elachistus caceciae Howard.
This insect, which was originally described in Bulletin No. 5, first series, of the Division of Entomology, page 28, as probably a primary parasite upon Cactecia rosaceana and Hyphantria eunea, now appears, without doubt, to be a secondary parasite. A single male issued in a vial in which the pupae of what we take to be Cratotechus orgyiwce Fitch, shown at fig. 13, were being kept to rear the adult.
'Described in the appendix, page 56.


Elasmus atratus n. sp.i
The most abundant of the parasites of Apanteles hyphantrise was, during the winter of 1896-97, this curious little Chalcidid, which was mentioned under its manuscript name in the writer's article entitled "The habits of Elasmus," Insect Life7 Vol. IV, pages 253-254, as being a common parasite of Apanteles hyphantrim, in the District of Columbia. No specimen of this insect was reared in 1895 and none in 1896 until the end of December. December 13 a number of cocoons of the Apanteles were found in the cocoon mass of the Orgyia, and all were cut open, as described under the section on Apanteles, for the purpose of ascertaining the hibernating condition of the Apanteles. In nearly every cocoon thus examined were found from two to four small black pupte, one of which is shown at fig. 21. These pupae were at once seen to be


FIo. 21.-Elasmus atratus: Adult male at left, with hind leg and female antenna. and with pupa at right-i-enlarged (original).
different from the pupse of Dibrachys, and were placed in a vial to rear the adults. The adults issued December 29 to January 15, 1897, and proved to be Elasmus atratus. The date of issuing is doubtless abnormal, since the specimens were purposely kept in a warm room to force early issuing in order to determine the species as soon as possible. Normally, they probably would not have issued until spring.

Asecodes albitarsis Ashmead.
This insect plays in the economy of the Orgyia the apparently unusual rble of a tertiary parasite. Its immediate host is Dibrachys boucheaus. We have secured the pupa of the Dibrachys under circumstances which render its identity positive, and it has been carefully illustrated at fig. 18b. The identity of this pupa having been ascertained, the actual SDescribed in the appendix, page 56.


observation on the Asecodes issuing from the pupa shown at fig. 22, and through the round hole on the ventral side below the head at once establishes the tertiary parasitism. Very many specimens of this species were reared. The dates of rearing in numbers were September 10 to 12, November 6, 1895, January 20, April 29, and July 30, 1896. We have found
neither the pupa
nor the larva of
this species. A
strange variation
in habit was observed September 9, 1895 and
again in -December, 1896, when a
number of specimens of this species issued from
oneof the puparia of a little scaven- FI. 22.-Asecodes albitarsis: a, pupa of Dibrachys from which this parager fly, Garax site has emerged; b. adult female; c, male abdomen-greatly enlarged ger fly, Gaurax oigua.
With this species we have a possible though unproved instance of quarternary parasitism. In the closing paragraph of our consideration of the Dibrachys we have shown how it may become a tertiary parasite. In such an instance as this the Asecodes, should it be attracted to such a stung cocoon of Meteorus, for example, would become quarternary.

Frontina aletiae Riley.

This is the first of the primary dipterous parasites to be discussed.' It is one of the most important enemies of the cotton caterpillar of the South, from which fact it derives its specific name. It was found in the course of the investigations of the cotton worm that this fly issues from puparia formed by maggots issuing from the larva and occasionally the pupa of Aletia argillacea in Georgia, Alabama, Mississippi, and Texas. It has been estimated at different times that the proportion of caterpillars killed by this species has reached 40 per cent. Its further distribution is stated by Mr. Coquillett to be Massachusetts, Florida, California, and Venezuela. It has been reared at this office from several different host insects. In addition to Aletia and Orgyia it has issued from Cerura sp., Dasylopha angina, Halisidota maculata, H. tessellata, Heterocampa manteo, and Lagoa opercularis.
'The writer is fortunate in presenting these few notes in having the benefit of the assistance of Mr. D. W. Coquillett in the determination of the species and, further, the opportunity to consult the notes on habits and distribution which he has brought together for a monographic paper on the Tachinide of North America.


It was first met with in the course of the observations upon Orgyia on August 30, 1895. A larva of the Orgyia which had already spun up could plhiinly be seen to be parasitized. A large dipterous larva could actually be seen through its skin. No eggs, however, were found upon its body. It was placed in a vial to breed and on the following day two Tachinid larve left the empty and dry skin of the caterpillar and transformed to puparia, which, however, dried up and nothing was reared from them. On September 7 a tussock-moth caterpillar was found which bore Tachinid eggs upon its back, and from this larva an adult of this species issued on Sep)tember 18. On September 20 an adult emerged from a puparium the larva of which issued from an Orgyia larva found September 5, and which was collected for the reason that it had a black spot around one of the spiracles; the larva issued on September 9 and transformed to its puparium, making the duration of the pupariuim stage eleven days. The same day a specimen issued from a tussock moth caterpillar which was observed upon September 11 to have been oviposited in byPimipla inquisitor. On September 6 an Orgyia cocoon was opened which contained a chrysalis upon which a dipterous larva was apparently feeding externally. This maggot contracted into a puparium, and a fly of this species issued on September 23. On September 13 a full-grown tussock-moth caterpillar was found which had a sick appearance and a pale-colored head. On September 18 a full-grown Tachinid larva issued from it and contracted, and the fly issued September 30. The total number of specimens of this insect reared was forty-five. They issued quite regularly through September; two specimens appeared in October, one in January, two in February, and the remainder between the 14th and 27th of July, 1896. During the examination of Orgyia cocoons in the winter of 1896 no specimens of this insect were found in any stages. In all probability it hibernates as an imago.
Frontina frenchii Will.
This species is very widespread, occurring from Massachusetts to California, and has been reared from no less than twenty-two different species of Lepidoptera, including Bombycids, -Noctuids, and Rhopalocera (Papilio turnus and P~jrameis cardui).
In the Orgyia observations it was not an important species and was not reared at all in 1895. Seventeen specimens were reared, all in July, 1896. In the case of two of the specimens the flies issued July 6 from two puparia found June 29. There is nothing in the observations which indicates the duration of any of the different stages.
Tachina mella Walker.
This species is also widespread, occurring in Toronto, Canada; Franconia, N. H.; Massachusetts; New York; District of Columbia; West Virginia; Florida; Texas; New Mexico; California, and Oregon. It has


been reared from several different Bombycid larve at this office. It appeared both in 1895 and 1896 in the present series of observations. On September 16, 1895, a single adult issued from an Orgyia cocoon taken September 7. In the different series of rearings undertaken to ascertain percentage of parasitism eighteen specimens issued, five in September, 1895, and the remainder in July, 1896.

Euphorocera claripennis Mlacq.
Thiswas the most abundant of the I)ipterous parasites. Like the other species, it occurs practically all over the country. It has been reared at this office from twenty-seven distinct species of Lepidoptera, including among the species of greatest economic importance A grotis ypsilon, Aletia argillacea, Anarsia lin~eatella, Ceratomiia catalpwe, Datana ministra, Feltia herilis, Hydrtecia ininanns Iliqphmitria
cunea, Leucania. albilinea, and
Mamestra trifolii.
It was first reared on September 21, 1895, from a pupa
of Orgyia collected September 7, the Tachinid larva having left the chrysalis before
pupating. September 23 a specimen issued from a puparium found within a chrysalis
of Orgyia on September 6.
On September 24 one specimen issued from a larva found
September 7 with a single Fm. 23.-Eupho,cracr claripejni i: Adult Vwith enlarged Tachinid egg attached to it. auntenna and with empty pupariumn at right-enlarged (original).
Another issued on the same
day from a larva found September 4 with a Tachinid egg attached to it, and still another from a larva found September 11, also with a Tachinid egg upon its back. This last had pupated within the body of its host. On the 30th of the same month a specimen issued from a caterpillar found in its cocoon September 11, and in which Pimpla inquisitor had been observed to oviposit.
Ninety-seven specimens in all were reared; eighty-one of then issued between September 19 and October 15, 1895, a single specimen issued April 16, and the remainder appeared from July 14 to 28, 1896. The usual method of hibernation here must also be in the imago state, although inii the case of the fly which issued April 16 the puparium must have overwintered.


Winthemia quadripustulata Fabr.
This, again, is a species of wide distribution, occurring throughout the northern portion of the United States and also in Germany. It has been reared from twelve species of lepidopterous larva at this office, including certain cutworms, the army worm (Leucania unipuncta), the grass worm of the South (Laphygma frugiperda), the promethea caterpillar (Attacus promethea), and several sphingid larva. Four specimens were reared in the Orgyia observations, three issuing in late October, 1895, and the fourth April 14, 1896, this latter having evidently hibernated in the puparium.

Exorista griseomicans v. d. W.
This species, which we know only from the District of Columbia and Central America, and which has been reared from this host insect alone, made its appearance in our experiments only in late July, 1896, when four specimens were reared from the mass of Orgyia cocoons under observation.
Helicobia helicis Towns. Limosina sp.
Sarcophaga (two species). Homalomyia scalaris Fab.
Phora nigriceps Loew. Gaurax anbchora Loew.
Phora incisuralis Loew. Neoglaphyroptera bivittata Say.
Phora fasciata Fallen. Diplosis sp.
Phora agaraci Lintner.
It seems hardly worth while to give any detailed consideration to any of the above species. All were reared from Orgyia cocoons, and most of them from cocoon clusters which had overwintered and were collected in the spring of 1896 and placed in rearing jars for observation. It was especially noticed that nearly all of the scavenger flies, particularly the Sarcophagids, after issuing, remained among the litter at the bottom of the cage, whereas the adults of the parasitic Diptera, and Hymenoptera as well, immediately flew to the upper parts and sides of the cage, apparently trying to find a way out.
By far the most abundant of the eleven species was the little Oscinid, Gaurax anchora. This little fly, shown at fig. 24, was first reared on September 10, 1895. Its larve and puparia were found abundantly in old Orgyia cocoons the larve of which had been destroyed by Pimpla inquisitor. The larve of the Gaurax were found to feed upon the quite dry remains of the caterpillar.' Other larve were found in dead pups early in September. Several hundred were reared in all, of
1Loew, Centur.,VII, page 111, mentions the fact that Osten Sacken found the larvM of this species in New York feeding gregariously upon the chrysalis of the Cecropia moth inclosed within its cocoon. In Osten Sacken's catalogue the fact is recorded simply in the words "inquilinous in cocoons of Attacus cecropia." The same species was reared from the cocoons of the gypsy moth (Porthetria dispar) by Professor Fernald and his assistants.


which only five appeared in the fall of 1895, the remainder appearing in May, July, and August, 1896. Active larvae of this species were found feeding upon the remains in old cocoons in the middle of December, 1896, and remained rather active for several weeks in spite of cold weather.
Living puparia of one of the species of Sarcophaga were found in the old cocoon masses of the Orgyia in December, 1896, and nearly all of the specimens of both species issued in July and August, 1896. The other Sarcophagid, Helicobia helicis (Towns.), originally reared from a snail (Helix thyroides Say) by Mr. H. A. Surface, Warren County, Ohio, seems to be a general scavenger after all. Eighteen specimens were reared in October, 1895, from cocoon masses of the Orgyia, and three more July 27, 1896. It is a widespread species, and Mr. Coquillett

FiG. 24-Gajrax anchora: a, larva: b, anal segment of same; c, spiracles of same; d, puparium; e, adult;
!, antenna of same-a, 1, e, enlarged; b, c, f, still more enlarged (original).

has recorded it from Charlotte Harbor, Florida, Mississippi, Illinois, Los Angeles, Cal., Ohio, and the District of Columbia (Proc. Acad. Nat. Sci. Phil., 1893, p. 317).
The four species of Phora were not reared in large numbers. Of P. nigriceps about thirty were bred from dead larvT of Orgyia at the end of July, 1896. Of P. incisralis five specimens were reared September 1, 1896. Of P.fasciata eight were reared July 28 to 30, 1896, and of P. agarici one specimen was reared September 30, 1895, from discolored chrysalids of Orgyia collected September 13, while on October 28 of the same year about a dozen specimens were reared from another discolored chrysalis, also found on September 13. The facts gained by the rearing of these four species of this genus present no positive evidence on the question as to whether Phora may ever attack living insects.
Of the Limosina a single specimen was reared September 2, 1896,


while the only specimen of the Anthomyiid (Hfomalomyica scalaris)' which was reared was found in the breeding cage with cocoons of Orgyia on September _6, 1895. AeoglaphIyroptera birittata, a beautiful little black and yellow Mycetophilid, was reared in two instances, the first on May 28 and the second one April 30, 1896. They were found in the breeding cage with cocoons of the Orgyia collected October 15 of the previous year. The Diplosis was reared in April, 1896, and again in September, 1896, the-first time from cocoons collected October 15, 1895, and the second time from cocoons collected in July, 1896.

Among the other insects found feeding upon the tussock-moth caterpillars or reared from their cocoon masses none are worthy of especial mention with the exception, perhaps, of the larvae of A uthrenus various. These larv were found constantly in the old cocoon masses of the Orgyia and under the batches of eggs which had already hatched, where they were engaged in feeding upon the dry remains of pupe and eggs. This is quite in accordance with the known habits of this insect. Its frequent occurrence in the egg masses led us to surmise that it might be found to feed upon the living eggs, and thus become a true enemy of the tussock moth. Predatory habits begin in this way. This surmise was proved to be correct in February, 1897. Mr. Busck, who had been instructed to watch carefully all egg masses which lihe should observe out of doors, with this object in view, brought in, February 14, two egg masses which had evidently been eaten by some insect. Both egg masses were eaten on one side of the surface. One was eaten into from the under side, and in the hole was a cast skin of the larva of a Troloderna, presumably T. tarsale. These egg masses were carefully examined, since it was suggested that simply the frothy secretion might have been fed upon, but eggs were found which had been half eaten, and examination of the other eggs in the mass showed that they were unhatched and living. Later, on the same day, Mr. Busck was sent out for additional material, and succeeded in finding a spot where the egg masses were numerous and collected 50 or more larve of Anthrenus rarius and Trogodernia tarsale hidden among the egg masses. These were brought in and kept under observation in confinement. It was soon proved beyond all doubt that they eat sound eggs of the tussock moth with avidity, and certainly just as readily as they eat dead animal matter, which was given them at the same time. We have thus the beginning of what may ultimately prove to be a very beneficial predatory habit on the part of these Dermestid larvae. The only record known to us of a true predatory habit on the part of a Dermestid larva is published by A. Kuwert in the Stettiner Entomologische Zeitung for 1871, page 305, where it is stated that the larvae of 1This species is common to Europe and North America. In Europe its larva has been found in human excrement.


Megatoma undata were found inside the cocoons of Tenthredo lutea. Kuwert considered their work in killing pupe of this sawfly of considerable importance and ascribed the immunity of his garden from the work of the sawfly larvwe to this cause. It seems, however, that there may still be doubt about the accuracy of his conclusions.
As illustrating the necessity for the most careful examination of all the circumstances surrounding a case of supposed parasitism, and as bearing upon the subject of Anthrenus rariuts, the fact may be mentioned that on December 17 a living specimen of an adult Proctotrypid parasite was found in the center of an et ipty egg mass of the Orgyia. The species was new to the series of Orgyia parasites, and study showed it to be Laelius trogodermatis Ashmead, a parasite, as the name indicates, of Dermestid larve. Further examination showed a full-grown larva of the Anthrenus in the same egg mass. Mr. Ashmead informs the writer that he has previously reared this Proctotrypid from Anthrenus rarius, and the presence of this specimen on the Orgyia egg mass was thus explained.
A very constant feature in the examination of the Orgyia cocoon masses during December, 1896, was the finding of many of the characteristic cocoons of a Chrysopa. They were hidden away among the mass in all conceivable situations. Some were empty, some contained parasites, while others contained the hibernating liarva of the Chrysopa. During the preceding summer adult Chrysopas were reared in some numbers from Orgyia cocoon masses placed in the rearing cages for study. The frequent parasitism of these Chrysopa cocoons introduced another element of danger in drawing conclusions as to the true parasites or hyperparasites of Orgyia. For a time the writer considered one of the species of Ilemiteles (sens. lat.) reared from the Orgyia cocoons as one of the tussock-moth hyperparasites. While opening many Orgyia cocoons, however, in December, one was found which contained several old and empty cocoons of Pimpla inquisitor, and in one of these, back at its far extremity, was the puparium of a Sarcophagid fly, from which the adult had issued. More in a spirit of idle curiosity than in anticipation of any result, this empty puparium was cut open and here again in its far extremity was another object, the cocoon of this Chrysopa. The cocoon was intact, and upon cutting it open it was found to contain an adult of the Hemiteles, which up to that time had been considered a hyperparasite of Orgyia. The larva of the Sarcophagid having reached full growth after feeding on the remains of the Pimpla pupa skins, or perhaps dead pupae as well, had transformed within the Pimpla cocoon and the fly had issued. The parasitized full-grown Clirysopa larva, seeking to hide itself as effectually as possible, had crawled first into the Orgyia cocoon, second into the Pimpla cocoon, and third into the empty puparium, and in its far extremity, with its little remaining strength, had spun its own cocoon only to die immediately thereafter from the attacks of the larva


of a primary parasite already containing eggs of the hyperparasitic Hemiteles!
The abundance of these Chrysopa cocoons in these situations does not, it seems to us, prove that the larve which constructed them had fed upon the Orgyia. It is more likely that they were attracted by the abundance of dipterous larv, particularly of the larve of the little Gaurax, although this is supposition only.
The work of predatory Ieteroptera was not especially marked, and in fact for some reason these insects do not seem to prey as extensively upon tussock-moth caterpillars as upon the fall webworm and other caterpillars. We have already called attention to the exemption from bird attack which this species possesses, and in fact the conspicuous coloration of the caterpillar seems to mark it as an especially protected form.
Except for the remarkable number of species involved, there is nothing, perhaps, after all, so extremely unusual in the extensive caseof p)arasitism of which we have just given the details. Wherever aplantfeeding species from some cause or from some combination of causes transcends its normal abundance to any great extent, there is always a great multiplication of its natural enemies, and this multiplication is usually so great as to reduce the species to a point even below its normal. Exceptions to this rule are seen with especially protected species, which, through the possession of some distasteful or repugnant quality, have no predatory or parasitic enemies. Even in such cases, however, disease steps in and fills the want. I need only refer to the chinch bug as a familiar example of this class of injurious insects. It possesses no parasites, but when it increases beyond the bounds of what may be called nature's law, for want of a better term, bacterial and fungous diseases speedily carry it off.
With all very injurious lepidopterous larve, however, we constantly see a great fluctuation in numbers, their parasites rapidly increasing immediately after the increase of the host species, overtaking it numerically and reducing it to the bottom of another ascending period of development. The unusual number of parasitic forms in the present case, however, and the extreme thoroughness of their work, as well as the plain evidence of the important part which hyperparasites have played, renders this particular case of perhaps more than usual interest.
The culmination of the Orgyia attack may be placed at the end of August and early part of September, 1895. At that time almost every poplar, soft maple, box elder, elm, alder, birch, and willow in the city of Washington was completely defoliated, while other maples, sycamores, horse-chestnuts, ashes, and many other trees were badly damaged. The rapidly developing parasites (some of them, as we have shown, requiring only two weeks for the development of an entire generation) by this time had become so numerous that it was an exception


to find a healthy caterpillar which one of the parasites was iot engaged in stinging. The rearings which were undertaken -at this time show a parasitism of' perhaps 90 per cent of the caterpillars.
The species principally involved in this work were Pinipla inquisitor and Chalcis orata, both primary parasites. The remaining species were not abundant and very few hyperparasites were reared in the autumn of 1895. It is worthy of record, however, that many caterpillars were killed at this time by some disease the nature of which has not as yet been ascertained, although some work has been done upon it by Mr. A. F. Woods. About 10 per cent of the cocoons at the end of September gave forth adult moths, and of these, of course, only about onehalf were females. Ten per cent, then, of the possible females of the fall generation laid their eggs, and the majority of these eggs hibernated successfully, so that iii April and May, 1896, there was a inoderately abundant hatching of young Orgyia caterpillars.
In the meantime, however, the extremely abundant primary parasites had successfully hibernated alnd were in position to make their presence felt. It resulted that at the points of observation this first generation of tussock-inoth caterpillars was practically exterminated, alid in June it was a very difficult matter to in(l enough living individuals to carry on rearing-cage experiments at the very points where they had been present the previous year by thousands anid thousands. Rearing-cage experiments at this time showed the actual percentage of parasitism to reach above 98 per cent. The details of one of these experiments follows:
On June 30 and July 8, 1896, 624 cocoons of the tussock lmoth were collected without discrimination from the trunks of trees in the U. S. Department of Agriculture park. From these 624 cocoons issued the following parasites:
Pimpla inquisitor ------------------------------------------------------------ 729
Bathythrix pimpl ------------------------------------------------------------ 13
Limneria valida ------------------------------------------------------------- 1
Theronia fulvescens.....------------------------------------------------------ 1
Chalcis ovata --------------------------------------------------------------- 69
Dibraehys boucheanis ------------------------------------------------------- 150
Asecodes albitarsis ----------------------------------------------------------- 1
Frontina aletiu --------------------------------------------------------------- 7
Frontina frenchii ---------------.--------------------------------------------- 14
Tachina mell: ....-----------------------------------------I----------------- 12
Euphorocera claripennis ---------------------------------------------------- 15
Exorista griseomicans -------------------------------------------------------- 4

Total ------------------------------------------------------------------ 916
Thus from the 624 cocoons were reared 916 parasites, all but 64 ot which were primary. Of moths 12 were reared, and with the exception of these 12, all of the others which were not parasitized, and these I Estimated.
11859-No. 5-4


were only a few, died from disease. The exact mortality ratio was, therefore, 98.08 per cent.
In the later months of 1896 the abundant presence of hyperparasites was made manifest. Many hundreds of specimens of Dibrachy.s boucheanus were reared from the cocoons of the Pimpla, and while the next most important primary parasite, Chalci. orata, has not definitely been proved to have its own parasites, it is almost safe to state that it also is destroyed by this species or some other of the secondary parasites. The other primary parasites, notably Mecteorus commu nis, Limneria ralida and the Apanteles, were also, almost without exception, destroyed by hyperparasites. principally Spilochalcis debilis and Dibrachus bouchea(n us.
The effect of this hyperpl)arasitism began to be noticed by the time the third geieration of tussock-moth caterpillars became full grown. It was no longer so difficult to find specimens as it had been in June. They were still rare, however, and the number of egg masses laid in the late autumn at the points where these studies were carried on was very small.
The length of time which it will take the Orgyia to recover from this externinative parasitic attack canm not be surmised. The partial recuperation toward the close of 1896 was a matter of some surprise, and must be attributed almost entirely to the work of hyperparasites. Ordinarily recovery from a severe case of parasitism following an undue multiplication of lep)idopterous larv;e is rather slow, as is instance by the published records regarding the army worm (Leucania unipuncta).
We should naturally have expected a period of abundance of tertiary parasites to have followed that of the secondary parasites. This, however, was not the case. Tertiary parasitism seemed to be comparatively rare and was only definitely proven in the case of Asecodes albitarsis and Dibrachys bouchean us, the latter being usually a secondary parasite. The majority of the specimens of the Asecodes issued in the late fall and early winter of 1895-96. There must be a limit to this work of parasite upon parasite at sonime point, and it seems certain that tertiary parasitism is rare, and that (quaternary parasitism seldom occurs. An interesting fact was, however, noticed in the late fall and early winter of 1896, and that was that many of the parasitic larvae died apparently as the result of disease, although possibly from some other cause-as, for example, the puncture of a hyperparasite, the larvae of which did not-develop. A number of d(lead larve of Pimpla inquisitor and Dibrachys boucheanus were found in the Orgyia cocoons in December, 1896, more or less shriveled and slightly moldy, but apparently whole.
It must be stated that nearly all of the observations which we have recorded were made on or in the immediate vicinity of the grounds of the U. S. Department of Agriculture, and that observations made incidentally and by no means with tlhe same thoroughness in other parts of the city show that conditions vary in neighboring localities, and that


at different points, distant from each other less, perhaps, than a mile, there was by no meaiis the same relative abundance of species. It canl be seen at the time of this writing (January, 1897) that hibernating egg masses of the Orgyia are certainly four times as abundant at some places in the city as they are at others, a fact which points plainly to a greater scarcity of primary parasites at the first-named points. The tussock-motlh caterpillar is itself a slow traveler. Its p)rimiary parasites naturally congregate at the points of greatest caterpillar abundance. At points where the caterpillars are scarcer they are thus less exposed to the attacks of their parasitic enemies., and it results that there may actually be an increase of the species at one point simultaneously with a decrease at another. This, then, at once suggests that in a small way artificial transportation of the Pimpla iii particular may often be of some practical benefit.!
The part played by the dipterous parasites in this instance was not great. Only 187 specimens in all were reared. All were of rather well-knowi Tachina flies, which are general and widespread parasites. Even though the work of these species was not iml)ortant in this instance, as is well known, they frequently )play a most iml)ortanit part in the re(huctioli of the numbers of injurious larvae.2
It is an extraordinary thing that these flies are by no means so restricted in their host relations as are the parasites which belong to the order Hymenoptera. The 1)parallelism between structure and host relation, which is so striking among the Hymeioptera. seems to be practically absent with these parasitic Diptera, as Brauer's tables plainly indicate.
It has been shown, for example, that Pimpla inq ibsitor, the most numerous of the parasites mentioned, is a very general feeder on lepidopterous larvwe, and in such cases as this extensive parasitism of ()rgyia in Washington, if the right moment were seized, a surplus of the parasites could readily be sent to such points as Boston, for example, where the tussock moth appears to be abundant nearly every year, and where, perhaps, the species would he found to attack even the gypsy moth, although, according to the reports of the gypsy moth ('ommittee, this species has not as yet been reared from this host.
,-The writer has searched for hours in grass fields overrun with army worms without finding a single specimen of the worm which did not bear upon its back the eggs of Wintheinia 4-pustulata. It has long been known that many of these eggs fail to produce any result, through the molting of the caterpillar before the hatching maggots have an opportunity to work their way into its body, but the observations made by Professor Fernald and his assistants in their work upon tLe gypsy moth in Massachusetts have thrown a new light on the number of failures in Tachina parasitism. On page 385 of the 1896 report upon the gypsy moth it is stated that during the summer of 1893 Mr. Reid collected a uznber of caterpillars oni which the eggs of the parasites had been laid. Two hundred and thirty-five of these caterpillars, having from 1 to 33 eggs on each, were fed in cages until they changed to pupw, and from these 226 moths emerged, but only 4 dipterous parasites were secured from the entire number. The caterpillar which had 33 Tachina eggs on it molted before the eggs hatched, passed through all its transformations, and the moth emerged in good condition. In 1895. 50 larvw bearing eggs were isolated, and 43 moths resulted and no parasites. Later 252 caterpillars were taken in the field, each bearing dipterous eggs, and were fed and carried through their transformations without the appearance of a single adult Tachina fly.


The main function of the Diptera in the interesting struggle which we have described was that of scavengers. The scavenger flies were much inore abundanit thaii the parasitic flies, and the (lead bodies of the caterpillars, the chrysalis skins, and the remains of the different species of parasites were abundantly preyed upon by species of this class.
The table which follows shows the exact host relations of the Hymenoptera, reared in the course of the observations:

Table 8ow~ing ho.st rchationN ofthIe Ih~menoptermis Paratsitcs st) far es Ikici are kiiown Or81)' sirisi'd.

[h1ost. Orgy ja Ivucosigrua.]

P~rimaury parasite. endr asi..Tert iary pairaite. osieniatPinipla inquisitor ......... I)raclhvs 1,orrheanuri ....... AstcodeM i hitarsis...
A Ilocot thjyrid1o ptevigis----auou ...... Bat hYthri x pimpki......Dibrachv% houciteanns Adiastola anericana ........ Pi"Pla- alnuhipes........f
Pimula ronqui.-itor ........ Allocota titYridopterigis ...... Hahrocytus tlryr.
Ic hneitnion c(Pruleis ..
1111(.eu111m) subicvanleus...
A..orphota orgyvia'.....Spiilochaleis dtehilis ..... .....1)ihiachys. Iitchea- Asecodes allhit:,r'Ills
Di1hrachy-s boncheanus ....... Askecodcs albitarsis...
Nr(-teoruis corimmunis....... Spilochab-is dvIbiis ........
Otacustes periliti ..........
B'athythrix iucteori ......... LDihracehvs bolicheanus .....I Aleteornls hvplmntIi~ na...... SpI lorhalcis debilis ..........1Elpl cn I Is 1imneriar'
Otacil.-At's periliti ...........
1Dibr'achvs houcireanus ....... Liuuwcria sp .............
Linlieria \ alida ..........IDibrachivs houchranus ....... Tetrastwlcirs sp.?..
Eupeluimns Iiunerite....
ilastiurs tru..........
'rh.ronlia fI'lvescens. ....
AJpanteles by phantriia' plcili ii..........1)ihracbys ounch.?....A ecodes nlbijar.

IDibrachN.s boucheanns ....... Ase(-odes alhitarsis?..I
Elasnius atratus ............
dipntldeica t it,.....Spi lochairis dci'his .....
IDibraclivs bolichanus,3... ELa sinu1s atratus ............
('halcis, ovata ............

C'a toteech us orgyiai........Mlac h Iis t Is (-Ic c i a.t........

Telenornus orgyila-.........-rsihr ......

As far as we were able to ascertain, the dipterous parasites had no hyperparasites. It will be seen that there were seventeen species of primary Hymeuoptera and six of primary IDiptera. Of the secondary hymenopterous parasites there were thirteen species, several of them, however, affecting more than one species of primary parasite. There were but two species of tertiary parasite, so far as positive observations go, but three other species were strongly suspected to sustain this relation, while, as indicated, one of the species of which we have proof of tertiary parasitism may under certain conditions, in our opinion, prove to be a quaternary parasite. In the table those species followed by an interrogation point are problematically placed. The parasitism. indicated is strongly suspected, but bas not been definitely proved. The

placing of all of the other species is based upon exact observations.


Amorphota Foerster. Verh. d. Natiurhist.Ver. d. Preus. Rheinl. u.Westplh. 1i68, p. 151.
Amorphota orgyise new species.
Female.-Length 9.5 mm; expanse 11 mm. Color black; abdomen rufous, with darker petiole; disk of second segment and hind )order of same black; all legs rufous, except hind tibia and tarsi, which are black; trochanters, tarsi, and spurs of hind tibia light yellow, nearly white; antennal scape rufous; mouth parts and tegul;e whitish; head and thorax densely and finely punctate, with sparse whitish pubescence, more marked on metanotumin; metanotum with shallow central longitudinal groove with transverse elevations; abdominal segnmentts 3 to 6 subequal in length; 7 shorter; 2 longer: the latter two-thirds length of petiole.
Male.-Resembles female, except tli at abdomen is darker, dusky shade extending down its dorsum: femora are of darker rufous: scape black.
Described from two females, one male, reared from Orgyia leucostigma.
Habitat: District of Columbia.
U. S. National Museum type No. 3509.

Bathythrix Foerster. Verh. d. Naturhist. Ver. d. Preuis. Rheini. uii. Westph. 1868, P 176.
Bathythrix meteori new species.
Female.-Length 7 mm.; expanse 11 mm. General color black and rufous; head, mesoscutum, mesopleura and metathorax black; all legs, including coxae, abdomen, prothorax, mesoscutellum, tegulhe, rufous; abdomen darker toward tip; front and middle coxa and trochanters, all tibie and tarsi lighter; sheaths of the ovipositor black; antennae rufous with the joints accentuated with black; darker toward tip; mandibles dark rufous; palpi white; head and mesoscutmn finely shagreened; mesopleura nearly smooth, shining; metanotum with several distinct raised areolets; entire thorax with fine white pile, more abundant upon metanotum and at borders of clypeus; abdomen shining, faintly longitudinally aciculate-punctate; abdominal segments decreasing in length from 1 to 7.
Male.-Resembles female except that abdomen is black at tip; antennme fuscous; thorax shining and very faintly shagreened.


I)escribed from many male and female specimens reared from cocoons of Meteorus.
Habitat: .District of Columbia.
U. S. National Museum type No. 3511.

Bathythrix pimple new species.
Fcnwle.-Length 6.5 mm; expanse 11 mm. General color black; head and thorax opaque; abdomen shining; all legs, including coxp, ruftous, with the exception of black band at the tip of hind femora and at tip of hind tibi; hind tarsi blackish, except at base of first joint; mandibles dark rufeus: palpi nearly white; mesothorax very faintly and sparsely punctate; metascutum divided into six areolets; entire thorax with sparse fine whitish pile; lower tface with very close and dense whitish pile. particularly marked at edge of clypeus.
described from three female specimens, presumably parasitic upon Pimpla inquisitor.
Illabitat: District of Columbia.
U. S. National Museum type No. 3512.

.ldiustola Foerster. Verh. d. Naturhist. Ver. d. Preus. Wheinl. u. Westph. 1868, p.180.
Adiastola americana new species.
Fm((l.-Length 5 mm.; expanse 9 n1m. General color black; all coxw. and femora and joints 2 to 7 of abdomen rufous; hind tibie and all tarsi black; front and middle tibia, rufous; teguhle whitish; mandibles and palpi black; petiole black, finely longitudinally aciculate; ovipositor sheaths nearly black; head and mesonotum very finely and very closely granulate, subopaque; mesoscutellum shining; areolets of metascutum marked, the hinder median one very concave, showing on profile view two sharp prominences; pleura shining, but very delicately shagreened.
Male.-Resembles female except that abdomen is entirely black and that second abdominal segment, as well as petiole, is closely longitudinally aciculate, though not so coarsely as petiole.
Described from two females and one male reared from cocoon masses of Orgyia le-costigna and presumably parasitic upon Pimpla inquisitor.
Habitat: District of Columbia.
U. S. National Museum type No. 3510.

Pezomachts Gravenhorst. Ichneumon. Europwea, II, 1829, p. 867. Pezomachus insolitus new species.
Mattle.-Leugth 5 mm.; expanse 10 mm. General color black; antenna dusky; all legs, including coxie, dark rufous; under side of abdomen, especially at base, somewhat rufous; clypeus rufous; mandibles rufous, black at tip; palpi dusky, nearly black; tegula dark; fore wings yel-


lowish at base; head and mesonotum shining, very faintly shagreened; metanotum closely punctulate with areolets faintly indicated; abdo meln shining, very faintly shagreened; joints 4 and 7 subequal in length; 1 to 4 gradually decreasing in length; stigma pitchy black, yellowish at proximal extremity.
Described from three males reared from cocoon masses of Orgyia leucostigmina.
Habitat: District of Columbia.
U. S. National Museum type No. 3513.

Apanteles Foerster. Verh. d. Naturhist. Ver. d. Preus. Rheinl. ii. Westph. 1862, p. 245.
Apanteles delicatus new species.
1ale.-Length 2.8 nn.; expanse 6.2 mm. General color black; all legs, except coxe, light reddish-yellow, hind tibie somewhat darker at the tips; palpi light honey-yellow; face faintly punctate, with a slight incomplete median carina below insertion of antenna: clypeus regularly rounded; metascutum with faint median longitudinal carina; abdomen rather coarsely punctate on first three joints: remaining joints smooth; joints 1 and 2 with slight median marina which is absent in joint 3; dorsal plate of first segment with somewhat rounded sides; p)leura faintly shagreened.
Described from two males reared from Orgyia leucostigma.
Habitat: District of Columbia.
U. S. National Museum type No. 3514.

Pteromalus Swedenis. Vetensk. Acad. Handl.. 179, T. 16. Pteromalus cuproideus new species.
Female.-Length 3.5 im.; exl)anse 6 n. General color brownishgreen; antennal scape rufous; thfunicle dusky; all legs except cox;e light honey-yellow; tegule darker honey-yellow; abdomen and metathorax concolorous with remainder of thorax and head, except that the basal two-thirds of second abdominal segment is blue or purplish, the second segment occupying a little more than one-third of the dorsal extent of the abdomen; joint 3 as long as 4 and 5 united; 6 and 7 subequal, longer than 3, the terminal segment conical; head and thorax uniformly and rather coarsely punctate; metascutum with medial longitudinal carina and two curved lateral carina.
Male.-Differs from female in possessing yellowish antenna' with scape black at extreme base; color somewhat brassy.
Described from one male and one female reared from OrgyiM lcucostigmiia.
Habitat: District of Columbia.
V. S. National Auseum type No. 3515.
Comes closest to Ashmead's 1. J()cronucts.


Eupelmis Swederns. Yetensk. Acad. Handl., 1820, pp. 136,376. Eupelmus limneriae new species.
Female.-Length 3 mm.; expanse 4.2 mm. General color dark metallic green; all coxm metallic green; front femora honey-yellow, with (lark. somewhat metallic stripe on outer side: middle femora honey-yellow, somewhat darker above; hind femora metallic; front and middle tibiae honey-yellow; hind tibie with rather more than basal half fuscons: tibia' yellowish white; all tarsi yellowish, black at tip; antenna black, somewhat metallic; body moderately stout; abdomen about as long as thorax, widening gradually to joint 5; joint 6 rather abruptly acuminate; head about as wide as thorax; mesoscutum sparsely punctate; miesoscutellum and mesop)leura closely and finely shagreened.
Male.-l)itfers in having all femora honey-yellow; hind tibie dusky; middle tibiae dusky toward tips; punctation of mesoscutellum similar to that of mesoscutum.
Described from one male and one female reared from cocoons of Limneria ralida.
Habitat: District of Columbia.
1. S. National Museum type No. 3516.

Elasmus Westwood. Lond. Edinb. Phil. Alag., III, 1833, p. 43. Elasmus atratus new species.
Female.-Length 1.6 nmm.; expanse 3.2 mm.; greatest width of fore wing 4.6 rmum. Face and vertex with moderately sparse large punctures; pron tun and mesoscutum regularly scaly, with sparse hairs; meso. scutellum very finely shagreened, shining; abdomen smooth, rather longer than head and thorax together; pleura and hind coxam shining, the latter finely aciculate; antennam very short; funicle joints subequal in length and about as long as wide; club slightly flattened; hind tibie above with three closed cells of spines placed end to end and a portion of another cell at each extremity; general color black, shining; thorax with faint metallic cluster; meso-postscutellum not differentiated in color; front and middle tibie dusky, rather lighter at bases.
Male.-Antennal branches dusky, reaching to base of club; in other respects resembles female.
Described from many male and female specimens reared from Apanteles hyphantri Apanteles delicatu.s, and Limneria ralida.
Habitat: District of Columnbia.
U. S. National Museum type No. 3517.


Prof. V. Deprez has given, under the title "Une Invasion de Dasychira pudibunda," in the Annales de la Societe Entomologique de Belgique for 1895, pages 333-335, some account of an outbreak of a European species closely related to our white-marked tussock moth in the vicinity of Carlsboarg, in the years 1892 to 1895.
Vigorous efforts were made by the authorities to put a stop to the work of the insect. The forestry administration employed a special oil, made in Germany, to cover the eggs and thus to asphyxiate them, and also made use of different methods for preventing the climbing of the trees by the caterpillars. Moreover, the inhabitants of the surrounding villages were requisitioned in May to collect the perfect insects. ", Nevertheless," says Professor Deprez, whatt can human means accomplish against such a prodigious quantity of enemies?" "Happily," he says, "nature, always foresighted, has placed a remedy beside the evil, and the researches of the most eminent naturalists have established the law that when an injurious insect develops in abnormal numbers its parasites are but little behind it in becoming proportionally multiplied. Thus, with the species of which we are speaking we have noticed an increasing multiplication of its natural enemies-the Ichneumon flies-which charge themselves with the duty of reducing considerably the number of the caterpillars, and often of reestablishing the equilibrium which had been broken for several years."
Experiments were instituted to determine the increase of the parasites. In the winter of 1892-93, among 200 chrysalids, 30 were found to have been parasitized. A year later, from the same number, 53 were parasitized. A year later, in the winter of 1894-95, from the same number, 95 were found to be parasitized.
How far this instance falls, in the completeness of the parasitism, below the instance which we have described in the foregoing pages will at once be evident. His concluding paragraph, freely translated, reads: "Will this proportion of the parasites increase still further (luring following years, as has been noticed in other countries, so as to bring about the complete stoppage of this unusual outbreak of the caterpillars, or will the parasites themselves be destroyed by other parasites, which, limiting their destructive action, will thus prolong the caterpillar invasion ? Future observations will determine."
11859-No. 5-5