, , �
, " , I ( t . ,f ,f " "',X1) \� I ,,1 ":; �.', A/
, ,� \ '\. I'")
,-.'\ " 1 '''{ I \
\ ,
t· J ::r' I'< /' '!� f 5(;' ", \ ).----
"I
... , "--':-/. , /;;' \';.;_:�;�,:� ;';
This volume is the property of the University of Saskatchewan, and the literary rights of the author and of the University must he respected. If the reader oh tains any assistance from this volume, he must give proper credit in his own work.
This Thesis by .. EIJ.,IS.McMILLA.N : . has heen used hy the following persons, whose signatures attest their accept ance of the ahove restrictions.
Narne and Address Date
,.� A THESIS
ent itled
A PR3LIMINARY STUDY OF THE POLYEMBRYOUIC CUTWORM
PARASITE BERECYNTUS BAKERI VAR. GEMMA GIRAULT
BY
Ellis McMillan
presented April, 1930. to the Faculty of Arts and
Science of the University of SasKatchewan, in part ful filment of the requirements for the degree of Master of
Science. FOR E W 0 R D TOT H � THE SIS
The field and laboratory worK involved in this
study has been performed by the writer during the
period 1927-1930 as a part of the investigational worK
pertaining to the duties of Assistant Entomologist,
Canada Department of Agriculture, and is used here by
permission of the Dominion Entomologist, Mr. Arthur
Gibson.
The development of the full plan of the study,
the worK in its entirety (except for certain routine
phases), the assembling of data and the conclusions
brought out, should all be credited to the writer. Ac
Knowledgments in full are given of any assistance which
has been received.
In the following presentation, both on account of
convenience and because of the large amount of data in
volved, the worK has been divided into four principal
sections. The first deals with the biology of the red
baCKed cutworm (Euxoa ochrogaster), the chief host of
the paraSite being studied, and is merely a summarized
statement of previous worK done by K. :M. King and
N. J. AtKinson. The second section deals with the
systematic study of the paraSite consisting mainly of
a treatment of larval morphology. The biology of the
parasite in all its aspects is fully treated in the
third section. while the fourth part outlines the pre- -�-
liminar,y studies on the polyembryonic development of the parasite.
Finally, credit for the original conception of the problem goes entirely to Mr. King, as also does some of the subsequent , supervision. ------...,_....".,...... ",..."...... ,,=---...=.��-�---
ACKITOWLEDGMEUTS
An investigation of the nature of the present study, in parts so detailed and so intricate, requires the assistance of others who are fully aware of the dangers and pitfalls to be met with in polyembryony.
It is with great pleasure, therefore, that the writer ac�nowledges his indebtednessto the following, from whom material help has been received.
Chief among. those to whom the author wishes to ex press his gratitude in the'pre�ent �ork is Mr. K. M. King,
Federal Entomologist for Sas�atchewan; to him belongs the initial conception of the problem, his permission and re commendation were necessary for the carrying on of part of the wor� in Branch hours and for the use of some mat erial presented herewith. Of equally great assistance was his continued interest in the problem, manifested by kindly criticism and many useful suggestions.
To Dr. L. G. Saunders, under whose supervision the study has been conducted, I wish to express my apprecia tion of his help and consideration. It has been the
�indness of Dr. W. P. Thompson and Dr. Saunders which has made possible the presentation of this work as a thesis.
From H. G. Crawford, Entomologist in Charge of Field
Crops and Garden Insects, Canada Department of Agriculture permission was obtained, upon Mr. King's recommendation, ---- _. -_. -----=��==""""=��=-==-��--
to use in the present thesis data and material obtained
by 1fr. King and the author as officers of t:qe Entomological
Branch, and to him the author wishes to express his sincere
t hanks ,
The author has been fortunate in obtaining valuable
advice from specialists in insect polyembryony. To
Dr. H. L. Parker, European Parasite Laboratory, United
States Bureau of Entomology, my most sincere thanKs are
given for the advice received concerning larval morphology
among the Chalcididae. From time to time,during my re
search, considerable help in interpreting the polyembryonw
development of the parasite,has been received from
Dr. R. W. Leiby, North Carolina Department of Agriyulture,
and I am exceedingly grateful for his sympathetic interest
in my difficulties. To Dr. Patterson, University of Texas,
grateful thanKs are due; this specialist in polyembryony
brought to my attention relevant literature which proved
invaluable.
For laboratory facilities and supplies I would expre�
my indebtedness to the Entomological Branch and to the
Department of Biology of the University of Sask.atchev,-an.
Messrs. A. Arnason, R. Glen and L. Paul, fellow
officers at the Sask.atoon Laboratory, have materially
helped at various times in the routine phases of the field
and laboratory work.. Their friendly interest in the pro
blem is also sincerely appreciated.
-�- �- �- - --...;;:.,,::;,- ;.....__ __;..._��===--_�=_=_O'"...... _=__------T _\.B2::,E 0 Fe\) UT �UT S
LIST OF FIS-TIRES
I IUTRODUCT Io:r 1
II BIOIOr'jy OF THE CHIEF EOST. :::t1!O_\. OCHROGASTER
( SUI��L\';ty ) 3
II I SYSTEMAT IC 11
IV IITORPHOI.OGY 19
1. The egg 19
o ..... The La rva
(a) Larval types in the chalcid oidea. 21
(b) External anatonzy 28
(c) Internal anatomy 29
3. Prepupa 32
4. Pupa 32
V BIOLOGY OF THE PARASITE
1. Emergence of the adults 35
2. Longevity of the adults, their habits and reactions. 39
3. potential progenitiveness ,45
4. Copulation 47
5. Oviposition 49
6. Parthenogenesis 55
7. Appearance of parasitized e�gs 60
8. Feeding of parasitized larvae and non-parasitized larvae. 61 9. Comnarison of the amount of time spent Lu the different stages of development by no rmal and para- sit ized larvae. 66
10. Hosts. 69
11. Super-parasitism of Euxoa ochrogaster. 72
1·::>..... Interrelationship of dise�se and parasitism by Berecyntus. 75
13. The general effects of parasitism on the hast and on the parasite. 77
14. The of food with reuro- relationship � ductive activity and longevity. 82
15. The relationship of abund�nce of the
parasite to weather conditions. . 86
16. Seasonal history. 107
17. Percentages of parasitism and economic importance. 116
VI PO 1YEliIB�YO BY
1. General considerations on polyernbryony132 ( a) thea ries 132 (b) polyembryony in animals 136
2. Historical review of literature on polyembryony, chiefly insect poly- embryony. 144
3. No t e s on the host and parasite 154
4. Development of the egg (fall period) 156
(a) preoviposition 156
(b) ovipoeitio� to cleavage 157
(c) egg after oviposi�ion 158
( d) d e ve Lo pine rrt in the host eg;s 162 1. maturation and fertiliz- ation. 163 Z. cleavage to overwinter- ing stage. 164 (e) EuffiLaryand discussion of the development' of the para.site egg to the overw int er ing st age. 168
5. Develo:pment of pazas Lt e body in host larva (spring and e arly euumer ) 169
(a) development of the parasite body after �tching of the host but previous to any indication of polyembryony. . 170
1 • po 1ygerma 1 rlJ9..SS 17�� 2. f 0 rrna t Lo n 0 f po1ygermal mass 173
(b)' first indications of polyembry- ony. 1. polyembryonal mass 175 2. the Lnt e r etub ryo na L matrix 176
(c) pseudogerms, pseudomoru1as� pseudo embryos and pseudolarvae 178
(d) .independant life of par�site in the lymph of the host 183 1. the larvae 183 2. prepupa1 stage and pupa- tion 156
6. Relationship betv.een specific poly embryony of parasitic hymenoptera and experimental polyembryony� 187
7. Rela.tionship be twe e n specific poly embryony of parasitic hymenoptera and other methods of agamic reproduction. 188
8. F:.wtors conditioning or aiding specific polyembryony in par:3.sitic hymenoptera. 194
9. Rearing da.t� on the sexes. 200
r!umber of adult s emerging from host. 203
10. The origin of mixed broods in poly- e mb ry o n i,c hymenopt era. 207
VII T:::CH�Trr�UE 215
-- -� ------�====------
1. Rearing methods. 215
2. Histological technique. 222
VIII 'llHE CO:::Et�:::::'ATIO�r OEl STRUCTU� AND HOST
�ATIOU -'i.HOnG TES EHCYRTHL\E 225
IX SUI!I:U.RY 229
X BIBLIOGRA.EHY 235 LIST OF lIT GtJRFS
Atter page l5-----PlateI: Antenna, head, mandibles and ovary of Berecyntus.
l8-----Plat e II : Adult, pupa and larva of 1!. baker! •
20-----Plate III: Eggs of B. baker!.
27-----Plate IV: Larval types in the Chalcidoidea.
4l-----Chari I: Length of life of adults of !.bakeri.
52-�---Chart II: Duration of oviposition and feeding � puncture.
63-----Chart III: Amount of tood consumed by healthy and parasitized
'. larvae of Euxoa ochrogaster.
66-----Chart IV: Length ot ttme spent in the different stages of
development by nor.mal Eu�oa ochrogaster larvae and
by larvae parasitiaed by Berecyntus.
76-----chart V: Mortallty of Euxoa ochrogaster from insect parasitism
and "diseases" of larvae and pupae at Saskatoon
during 1925.
, . 82-----Plate V: :'Diagramatic representations of three shapes aaaimed
by larvae of Euxoa ochrogaster parasitized by
Berecyntus.
se----Chart VI: Ntunber of Berecyntus adults emerging from Euxoa
ochrogaster larvae, the eggs fram wbidt these
larvae hatched having previously bee� kept at
different temperatures.
.. -_ _._- --"""-"--':;;..;._-...... __....;..------__...... 90----Chart VII: Nmnber of adults of Berecyntus emerging fram Euxoa
tristicula larvae, these larvae having previously
been reared together under laborator.y conditions
from. the egg to the fourth moult and then subjected
to different temperatures tor a period of 30 days.
97-----ChartVIII: Number of adults of Berecyntus emerging from Euxoa
tristicu1a larvae subdected to alternate thawing
and freezing conditions.
102----Chart IX: Correlation of' the a.IJX)unt of' mnshine wi th the
percentage of parasitism of Berecyntus. l24----Chart X: Estimated parasitism of Euxoa ochrogaster at
Saskatoon by Berecyntus 1925-1929.
l73---Plate VI: Egg of Berecyntus after insertion in host egg;
section Showing conditions of' hibernation; section
showing condition saowing after hibernation.
l77----Plate VII: Po1yembryonic chaige; section through a polygeDn.
183----P1ateVIII: Free larvae; asexual larvae. - 1 -
I N T ROD U C T ION
As early as 1923, the attention of Mr. K. M. King
Was drawn to the occurrence in SasKatchewan of a hymenop
terous insect parasite of the family Encyrtidae,
Berecyntus bak.eri How. which was parasitizing noctuid
larvae. Becauee of the great number of adults developing
in a single host larva, it was thought that this paraSite
developed polyembryonically.
A study of this insect was undertaKen by the writer
in 1927 at the suggestion of Mr. King, Since it was one
of the major paraSites of the red-bacKed cutworm (the
most important cutworm occurring in SasKatchewan). This
study was cant inued for three years, during which time
much information concerning the parasite was secured.
The data here presented indicate that Berecyntus
,. baKeri develops in a polyembryonic fashion, the stage in
which parasiti�m ini�ially occurs being the host egg.
The various stages in its development are somewhat simi
lar to those described by Leiby for Copidosoma Gelechiae
How., but there are striKing differences also exhibited,
not only with the former species but with other poly
embryonic species. Such are to be expected since the
development of only a few polyembryonic species have been
described and since each of these differs essentially
from any other •
- ..
•.�'-- .. �- - - - -'-'----�___:.e:;_, ...... "-"==-__...... ;;;;.-�__ _ - 2 -
If parasitism demands a nice physiological adjust ment, it is exhibited to a high degree by egg parasites, since they affect the host organism at an earlier and less highly differentiated stage of ontogeny. The later syn- chronism of host and parasite also reveals a marvello�sly balanced association.
The present work: is termed "A Preliminary Study" I and mak:es no claim to being anything more. The external
I morphology is completed as far as seems necessary until adequate comparative studies with other species can be conducted; considerably more work: is needed with respect to internal anatomy. The biological studies of this egg parasite are more complete than any other section of the study. The term "preliminary" fully indicates the status of the investigations on the polyembryonic development of the parasite. Indeed there is still suffiCient work: necessary in this phase to complete our understanding of the parasitic development that a major research problem could be initiated. The lack: of suitable equipment and the writer's inexperience in such a detailed study made
it impossible to complete the work: .in a Master's theSiS.
--_ .. -.--.::: BIOLOGY OF THE CHIEF HOST - EU!OA OCHROGASTER ( 8UMMARY) - 3 -
NOT�S ON THE BIOLOGY# OF THE RED-BACKED CUTWORM
(Euxoa .Q.£!gogast er Gn.), TEE C flIEF HOST
The biology of Euxoa ochrogaster in the Prairie'
Provinces has been described in considerable detail by
King( '26). The species has but a single generation
per year and hibernates in the egg stage. The larvae
are present from the first warm spell in spring until
about the end of June during which the normal larvae
pass through six instars. Pupation takes place in the
soil. The moths are common for several weeks beginning
the latter part of July but oviposition does not begin until early in August. The eggs are laid in loose soil.
The larvae spend most 0 f the ir 1 ives there also 'but usually come to the surface at night to feed. The red
backed cutworm is a native species now associated al most exclusively with cultivated fields.
Rather widespread outbreaks of greater or less
severity have occurred ('26) in this Province at inter
vals of from four to six years. The most recent reached
its climax in 1925. At Saskatoon the species was re
latively scarce in 1923 though probably more plentiful
than during the preceding year. In 1924 it was nearly
••••••••••••••••••••••••••••••••••••••••••••••••••••••
Footnote# - The information contained in this section was obtained chiefly from the paper published by King
( '26) and t'he thesis of At kinson ( '26) •
- - 4 -
twice as abundant as in 1923. The damage also waS
greatly increased and the larval mortality decreased
by a phenomenally dry summer. As a result of the
latter, combined with exceptionally favourable weather
in the fall, the larvae were fully ten times as plent i
ful in the spring of 1925 as in the previous year.
However, a very high death rate in June 1925, combined with factors affecting the moths, reduced the numbers
of the species in 1926 to about one-fifth of those of
the previous season. In 1927 the species was present
in about the same numbers as in 1923. In 1928 apparent
ly this species was again on the upward trend as larvae were consi�erably more abundant than in 1927. Condi
tions in the fall of 1928 were very favourable for a
large increase in the larval population in the spring
of 1929 but did not materialize as much as was expected.
Nevertheless the red-bacted population waS larger than
in the previous year. A moderate outbreaK of this pest
is forecast for 1930, the severity depending upon the
spring weather oonditions.
AtKinson in his Master's thesis '�6 (unpublished)
admirably des.cribes the various stages of the red-baolr.ed
outworm and also presents some interesting data whioh is
relevant to the present problem. - 5 -
The egg of Euxoa ochro�aster is spheroidal, flattened
dorso-ventrally, glistening milk-white when newly laid
but losing this glistening color and changing to a dull
gray within three Or four days at normal temperature.
The chorion is thin and semi-transparent but tough.
The outer coat of the egg has a reticulate appearance,
but the chorion of the basal fifth of the egg is more
delicate and glistening, without any definite reticula
tions. The following dimensions are ta�en from twenty
five eggs, .five from each of five moths:-
I s i z e Diameter I Height
Minimum 0.671 mm. 0.530 mm.
Average 0.691 mm. 0.545 mn,
Maximum 0.718 mm. 0.577 mm.
Oviposition ta�es place exclusively in the soil.
We have never obtained eggs on vegatation either in the
field or in captivity. The eggs are laid at a depth of
i to � inch. Dry loose soil is preferred to wet or
ca�ed soil. Most of the eggs are laid in masses of from
fifty to one hundred eggs. This is particularly true of •
the first few nights' ovipos it ion. A. few solitary eggs
are cemented together with a clear colourless substance
which is slightly soluble in water. The extent to which - 6 - the eggs are attached to one another varies considerably; sometimes they fall apart if touched but more often are closely attached and may only be separated from one an other by being moistened. Even this only removes a small amount of the agglutinating material.
The following measurements of larvae of Euxoa ochrogaster were based upon ten larvae of each instar:-
BODY GTHB INSTAR mm.
1 0.324 1.9 3.0
2 0.483 3.2 4.3
3 0.752 4.6 8.0
4 1.250 9.1 15.7
5 1.915 16.3 24.1
6 2.754 26.3 38.0
Body length A is the length of the larva at the beginning of the instar prior to any feeding in that in star. Body length B is the length at the end of the in star when feeding for that instar has ceased.
Data concerning the life history of the moths
indicate t under condit Lena that there is an laboratory .... average period of twelve days between emergence and the first �ting, with a variation of from seven to twenty days; a period of one or two days between mating and
-----.--�------�---==------7 - oviposition, and an average oviposition period of nine- teen days with an average·total life of 32.3 days. All of the periods might be lengthened under field conditions, since at the time when the moths are flying the tempera ture is normally fairly high� The important point from the viewpoint of parasitism is the fact that there is a preoviposition period averaging fourteen days with ex tremes of from eight to twenty-one days.
The oviposition records of seven moths kept in captivity showed the following totals per moth:- 1304,
1154, 1303, 726, 1314, 917, 1507 and 1175. The fact that each moth is capable of laying more than 1000 eggs accounts for the rapid increase of this species under favourable conditions. Moths are present in important numbers for about seven weekS, beginning late in July, wh�le the aver age period of oviposition is, in nature, over three weeks.
The fertility of eggs laid in nature is above ninety per cent, more probably between 95 and 100 per cent.
The development of the eggs takes place subsequent to the emergence of the adults and the number of eggs laid is influenced very greatly by the size of the fat-body. The
development of the fat-body of the adult is largely in
fluenced by the larval environment, especially the food
supply. - 8 -
DEVELO?1mUT AND HATCHING OF � OCHROGASTER EGGS
• In the following disaussion two terms will constantly oaaur, whiah are used to designate two states.
These terms are morphologiaal maturity and physiologiaal maturity. The first is applied to larvae whiah are ap parently fully deve�oped morphologiaally, judged by the faat that they are aative when disseated from the eggs and appear normal morphologiaally in every respeat; the seaond is applied to larvae that will feed within a per iod of a few hours following disseation from the eggs.
With respeat to the' time taKen by larvae to reaah morpho logiaal maturi�y, the variation in individual eggs was not great at high temperatures, but at ,lOW temperatures, around 1200., there was a variation of as much as five days in eggs laid by the same moth on the same day.
Subsequent to the arrival at morphological maturity there is normally a prolonged period of dormanay prior to the attainment of physiologiaal maturity. This is the state in whia� the winter is passed. Numerous stimuli have been to the eggs with the of ' applied object breaKing . . this dormancy but, with the exaeption of the subjeation of eggs containing morphologically mature larvae to high or low temperatures for a considerable time, no conclusive results have been obtained. The very uneven hatahing in such cases is probably due partly to abnormal conditions, but Since the morpholog�aal development taKes place at - 9 - varying rates it is reasonable to assume that the physiological would probably do so in nature also. In nature the period during which dormancy must be main tained, i.e. from the time of attainment of morphological maturity to the cessation of warm weather in the autumn, may be as long as three months; it appears possible that eggs laid in July might hatch in the fall if August was very warm. The probability of fall hatching is remote
Since a sufficiently warm spell seldom, if evert follows a cold spell of sufficient length to brea� the dormancy.
THE EFFECT OF SOIL MOISTURE on LARVAE
This species has no well-defined moisture require ments though the optimum probably lies between twelve and twenty-seven per cent of capacity. More important, how ever, is that it can withstand great extremes of moisture.
NATURAL C ONT RO L
At the present time we have no indication of any appreciable mortality-of the eggs under natural condi tions. The data are purely qualitative but it is note worthy that at least two very serious outbreaks have occurred following weather which one would expect to be detrimental to the eggs. If any appreciable larval mortality is caused directly by phySical factors it is probably only during the early instars.
,.,
-
-.------10 -
LIFE - CYCLE
The average life-cycle of Euxoa ochrogaster may be summarized as follows:-
Egg - August 15th to April 15th.
Larva - April 15th to June 26th.
Pupa - June 26th to July 23rd.
Adult - July 23rd to september 5th. )
SYSTEMATIC AIm MORPHOLOGICAL
• - 11 -
SYSTEM,,\.T IC
Berecyntus bakeri var. gemma Girault belongs to the hymenopterous superfamily Chalcidoidea, family Encyrtidae and subfamily Encyrtinae.
Among the many thousands of minute hymenopterous insects existing in the world and to which have been given the popular name Chalcid-flies, there is probably no single family that is of more interest or 0 f greater economic im portance than the family Encyrtidae. The species in this family, like the vast majority of other Chalcid-flies live parasitically in the eggs, larvae, or pupae of other insects, particularly the Hemiptera-Homoptera and Lepidoptera, but hardly a single order of hexapodous insects is wholly free from their attacks. But in this family, and more especially in the subfamily Encyrtinae, the species are of more than ordinary interest and importance, since so many of them are found attacking and destroying the scale and bark-lice (Coc cidae and Aleyroididae) and the plant-lice (Aphididae and
Psyllidae), containing some of the most destructive and troublesome pests with which fruitgrowers, agriculturists, and florists have to contend. Certain genera are definite ly restricted with reference to their selection. Thus
Aphycus is an eCto- or endoparaSite of Coccidae, particular ly of Coccus L.; Blastothrix almost exclusively parasitizes '�------..�--���--�....--��------
- 12 -
Coccus and Pulvenaria while Ageniaspis is mainly confined
to ce zt a Ln lepidopterous genera. On the other hand, Euphemus
affects a wide range of species, having been reared from the
eggs of Saturnidae and other of the large:r: ,Lepidoptera, from
the puparium of Glossina and from the Cecidomyidae, Coccidae
and various Coleoptera. The family is of great interest from
the fact that certain species of Encyrtus, Litomastix, Q£E
idosoma and Berecyntus, which parasitize Lepidoptera, are
known to exhibit polyembryony.
The family Encyrtidae is remarkable for its incompar
a.bly rich and diverse display o,f generic characters, in sharp
contrast with certain other Chalcidoid families, such as the
Pteromalidae, which are more uniform in both appearance and
structure. The number of genera of the Encyrtidae, a.lthough
now mounting to over 300, will be gradually increased by
future work, and the immensity of the group can be appreciated
only by those who have had occasion to arrange or study col
lections of small parasitic Hymenoptera, especially from trop
ical regions, or by those who have delved deeply into chalcid
taxonomic literature. Only a small amount of work has yet
been done on these insects from the Tropics, however, except
Girault so by in Queensland, Australia, that doubtless many
and interesting genera species await description. The family
is also no means in by poorly represented the temperate reg
ions of the a few globe and species have been recorded even
from the Arctic Zone. - 13 -
The family Encyrtidae is readily distinguished from all others in the Chalcidoidea by the large. non-impreseed mesopleura. the large triangular mesepisternum. which does not extend to the front coxae. and by the large saltatorial spur of the middle tibiae. which is most freguently long a� stout, or dilated at base, and usually armed with a double row of black teeth or bristles. No other family possesses this large saltatorial middle tibial spur, and only a few species, in one or two of the other families, possess the non-impressed mesopleura.
According to Ashmead (1900), Berecyntus bakeri is placed in the tribe Mirini, characterized by the hypopygium not prominent; hind tibiae with only one spur; body most freguently, but not always, metallic; antennae variable.
To the tribe Mirini belong the vast majority of the known
Encyrtidae. It is distinguished from the other tribes prin cipally by the mandibles, which are somewhat differently shaped, and always t rident at e at apex_. In most 0 f the gen era, these have three small, equal, or very nearly equal, teeth, while in others the outer tooth is the longest and most acute. One of two genera, however, have�e outer two
teeth- l�ng�r .. th�n the inner. The marginal cell in the hind wings is usually long and narrow, nearly obsolete, but nev er broad, as in the Encyrtini; while the h'tnd tibiae have only � apical spur. - 14 -
Dr. L. O. Howard's description of the adult (1898)
follows:-
"Berecyntus, new genus.
"Female - Comes nearest to Prionomitus Mayr., from which, however, it differs in its lengthened face. Head
seen from above semi-circular, vertex broad. Occelli at
angles of an obtuse-angled triangle; eyes sparsely hairy,
broadly oval; cheeks longer than eyes; scrobes elongate;
epistoma with a large, rounded, longitudinal carina; man
dibles stout, long, 3-dentate, with a long, sharp apical
tooth. Antennae inserted slightly above mouth; scape long,
slender, subcylindrical, slightly swollen near middle and
reaching nearly to top of head; pedicel subcylindrical,
nearly four times as long as broad; funicle joints snort.
increasing slightly in width from one to six and each about
as long as broad; club about two-thirds as long as funicle,
broader at base than sixth funicle joint, obliquely truncate
from tip to near base. Mesonotum flat, scutellum slightly
elevated, rounded at tip; scapulae narrow, not quite meeting
at tips. Abdomen flat, broadly oval, po Lnt e d at tip. Mar
ginal vein of forewing very short, broader than long; post
marginal much shorter than the rather stout stigmal vein;
costal cell of hind wings very narrow but extending nearly
n to ho 0 k.let s •
Berecyntus bakeri, Howard, new species.
- Female length, 1.4 mm.; expanse, 3.4 mm.; greatest -� � ..
- 15 -
width of forewings, 0.63 mm. Head and mesonotum densely and
shallowly punctate, with irregular, frequently hexagonal
punctations, those of head rather finer than the others; mes
oscutellum with close, shallow, very elongate punctat ion; mes
opleura faintly longitudinally aciculate. Head, thorax, and
abdomen lustrous. General colour highly metallic green; mes
oscutellum bronzy; flagellum of antennae black.; all legs black.
except femorotibial k.nees and tarSi, which are brown. Wings
hyaline, veins dark brown.
One femal� collected in Colorado by Mr. C. F. BaKer.
Type - Catalogue No. 5030, U. S. N. M.
Berecyntus bakeri Howard, var. gemma Girault.
Female - Length 1.15 mm., metallix blue-green, the
middle knees and all tarSi except the distal joint, reddish
brown, the venation black.ish, the forewings lightly infuscated
from the head of the submarginal' vein distad to the apex.
Funicle jOints increasing gradually in width d,istad, I quad
rate not half the length of the pedicel, subequal to 2, 6
somewhat wider than long. Pedicel dist inctly longer than wide
at apex, nearly haif the length of the club which is a little
shorter than the body of the scape. Hind wings with about
sixteen lines of uniform discal cilia where broadest. Fore
wings with only about two short lines of coarse cilia proximad
of hairless line. Whole body scaly reticUlate. A very short,
delicate carina between the axillae. Scutellum more finely
scutum. OVipOSitor not prominent. .,.------'
- I
' I ,
--�
._- ,� .--- 2
I ", I I
Ir---OVP
�---..:;:===--""",I�
?lateI:(I)Antennae of B.bakeri (2) Head of famale in frontal view (3) Mandible of female,left.2_frontal view; right_ ventral view. (4) sketch of ovaries ( ov)showing that they are directed caudad parellel with the ovipositor(ovp) but that their tips are inclined dorsocephalad. (muchn enlarged) - 16 -
Male similar to female but the scutellum is scaly like the rest of the thorax, the funicle joints all somewhat longer, subquadrate, the club shorter; the third tooth of the mandible
is longer, the other shorter. Wings hyaline.
Differs from the typical form in having the fore-wings narrower and the venation black; both forms bear a rectangular
fuscous patch along the stigmal vein. Type locality, Ottawa.
Types - Catalogue No. 19318 U. S. N. M. Described from a male and a large number of females labelled IlFrom an Euxoa
larva, Q,ueensboro, Ontario, A. Gibson;" also "from larva ·of
Radena devastatrix ottawa, July 12, 1914, A. Gibson".
Girault also distinguishes two other varieties of the species bakeri, (1) euxoae (2) arizonensis. However, Dr. A. B.
Gahan, another specialist in the Chalcidoidea (in a personal communication April 17th, 1929) states that from a casual ex amination of the types of these three so-called varieties, he
is of the opinion that they are hardly worthy of recognition.
Berecyntus bakeri Howard var. euxoae Girault.
Male - Similar to female melanocera Ashmead - except:
The scape is somewhat shorter, the club distinctly shorter
(not a third longer than the pedicel); the funicle 6 is quad
rate, 2 longest, colored as bakeri. The pedicel may be short er, the wings more or.less dusky. Reared from larval-Euxoa auxiliaris, State Colleg@, N. M. (D. E. Merrill). ., - 17 -
Types. Catalogue No. 20097.
The female similar. Later, both sexes from the same
host, Lethbridge, Alberta, July 1918 (E. H. Strickland).
Described by Girault in Descriptiones Stel1arum Novarum
p •. 18, May 1, 1917.
Berecyntus bakeri Howard yare arizonensis.
Female differs from bakeri bakeri in having the fore
wings distinctly more dusky, infuscated throughout from the
bend of the submarginal vein.
Described from 4 females from Phoenix, Arizona, March 20
1915. From Chorizagrotis species.
Types - Catalogue No. 19319 U. S. N. M.
DISTRIBUTION
The species Berecyntus bakeri is spread over practically
the entire North American oontinent. Although there is some
restriction to_ the distribution of the three varieties, there
is considerable variation in the localit ies for which these
varieties are recorded. Probably the distribution of the
.varieties is somewhat dependent upon the distribution of the
hosts. The species is parasitic chiefly on the following gen-
era:- Euxoa, Chorizagrotis, Agrotis, Feltia, Rhizagrotis and
Sidemia. The variety euxoae has been recorded as parasitizing
chiefly the army cutworm, Euxoa (Chorizagrot is) auxi1iaris.
No host has been noted for the Variety arizonensis • ..'.",.
- - 18--
The remaining host genera are attacked by the variety gemna,
with the Euxoa group being particularly susceptible. It has
not been possible to obtain a� information regarding the
that if the specificity of any of the variet Lea , is, variety •
euxoae �s able to parasitize eggs nor�lly attacked by the
variety gemma.
The varieties of this species have been recorded from
the following localities:-
var. gemma
Queensboro, OntariO; Ottawa, Ontario; Winnipeg, Manitoba.
, r var. suxoae
Lethbridge, Alberta; State College, New Mexico.
var. a�izonensis
Phoenix, Arizona (on Chorizagrotis)
J'
Berecyntu8 bakeri
Oregon, Utah, Montana, Vernon (parasit Lc on Peridroma saucia) .�.
2 4
Plate II: (I) adult of Bereeyntus baker! Yare gemma Girault.x40 (!)pupa,ventral view %50. (3)mature larva, greatly enlarged, (actual length2.5 mm) (4) head capsule of mature larva •
' .. " .. ''''
- 19 -
MORPHOLOGICAL
Study of the Egg
The newly deposited egg is short and broad, typically
elliptical in shape, occasionally elongate-ovoid, but never
distinctly flas�-shaped or with a small pedicel, as described
by Silvestri for Litomastix. The egg is slightly refractive
and almost colourless. The shape and colour vary for individ
ual eggs. The individual egg varies considerably in size, but
the size and shape of the eg� undergo no appreciable change upon oviposition. The following measurements were made of ova
dissected from the ovaries of female parasites. The measure ments were made at the longest and widest part of the ova.
The average measurement of many eggs indicates a length of 38
to 50 microns (about 0.04 mm.) and a width 0 f 25 to 35 microns
(about 0.025 mm.).
The protoplasm of the egg appears to be very finely
granular and uniform; ,no vacuoles appear to be present. The
presence of a wea�ly defined membrane becomes eVident in eggs
that shrin� considerably following some fixations. However,
this membrane is qUite tough. It is possible that a very
delicate vitelline membrance is present.
Although the small size of the egg made it very diffic
ult to section accurately, some success was obtained. Each
egg contains a nucleus situated in the arrt e r-Lcr region (narrow
er part); in some cases another small body is noted, probably
'_ -
- 20 -
being a nucleolus. However, this is not certain, in that no
such body is given by Leiby' for Copidosoma gelechiae, a very
closely related species. If the egg has been fertilized the
elongated sperm is found in any part of the egg, but usually
in the median or posterior part. It is usually found in a
curved position, probably indicating constant activity.
Never more than one sperm has been found in any egg and it
always stains a deep blac� in iron-hematoxylin.
As has been mentioned above, the eggs of Berecyntus
have no pedicel, as is the case with practically all of the
is the other not .j; ,; family Encyrt idae; Copidosoma only genus
on is .. I: having this attachment the egg. Usually this pedicel
'I well differentiated from the egg and is the means by which
the pazas Lt e egg is attached to the external wall of the host
egg. Par�er states that it is the intermediary by which com
munication is established to the outside in order to obtain
air, and that it plays exactly the same role as does the ,
.. "
tracheal sheaths and integument in the tachinid larvae.
Schedius tuvanae How., an egg paraSite of the gypsy moth, has
the long stal� attached to the host egg at the point where
the egg of the gypsy moth was punctured by the female parasite.
In regard to the later life of Sohedius within the host egg,
the is taKen verbatim from Bureau of . t following Entomology I "
- bullet in [ p , 180), by Howard and Fis ke : '1J
1 2
Plate III: {I)and (2) eggs of Berecyntus bakeri (3) egg of Encyrtus � (after Parker)showing pedecil (4) section of parthenogenetic egg five minutes after oviposition showing the nucleus(top large) and the genn cell deter.minant(bottam).
- - 21 -
"When the egg hatches, the larva do es not ent irely leave the shell, but remains with its anal end thrust into it, and the stalk, which is hollow, becomes functional and acts, like a lifeline attached to a submarine diver in supplying a connection with the outer air. It would appear that this stalk is actually that of the integument of the first-stage larva. "
Anatomical study of the Larvae.
(1) General Considerat ions on the Larval Form.
Larval types in the Chalcidoidea.
During recent years workers 'on the biology of parasitic
Hymenoptera have made several attempts to classify the larvae of these insects in groups based on certain striking features of the external anatomy. Richardson ('13) and Wheeler ('23}, previous to 1924, had distinguished 9 or 10 types of hymen opterous larvae distributed among the several parasitic fam ilies. Previous to Parker's and Thompson's work no effort had been made to define such larval types within the superfamilies oft he paras it tc Hyrne no pt era, altho ugh in many 0 f the gro ups several types of larvae had been described. However, the data gathered by Parker and Thompson have demonstrated the exist ence within the superfamily Chalcidoidea of a number of well defined larval types. These types are readily recognizable in the primary larvae but later the morphological differences
• !l11,! , ....)
- 22 -
beeome less apparent. In all, seven larval groups are dis- III t inguis he d.•
Group I:
The first includes the larvae of the , • primary ,I group
genus Aphelinus (Eulophidae or Aphelininae). The larvae are
almost spherical, composed of a head and thirteen segments;
the head ,possesses a pair of short antennae and some sensory
The is without hairs or ' , organs. integument bare, any sensory
' ...... - �-. spines. There are 8 pairs of open spiracles, one on segment
, : II, the others on segments IV to X inclusive.
Group II:
This group includes the great majority of the Chalcid r e , larvae. The ectophagous larvae of the pteromalidae and cer :.
tain Torymidae, Eulophidae, Eurytomidae, Elasmidae, and Pet
romalidae, present a head and thirteen well-developed segments,
both head and body being feebly chitinized and pigmented; the
head capsule bears a number of relatively short sensorial hairs
or papillae and one pair of short conical truncate antennae;
the mandibles are gently curved; the body segments under low
magnification, appear glabrous, though a careful examination
with a high power objective will reveal the presence on,e�ch
• • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • I· ! I
Footnote 1 - The Comparative Anatomy of Chalcid Larvae is a review of R.L.Parker's Doctor's thesis (published in French) and the R.t.Parker on " paper by ('24)'and W.R.Thompson ('25) , Chalcid Larvae.
_*"1
- - 23 -
... ,.\ segment of an anterior band of minute spines together with
two or three pairs of very small ee neo rLa.L hairs; in this
stage the tracteal system presents four pairs of open spir
acles situated on segments II, IV, V and VI.
Group III:
The sexual larvae of the polyembryonic Chalcids, a.ll
of which belong to the family Encyrtidae (Copidosoma, 14!�
mastix, Encyrtus, Berecyntus) constitute this well-charact
erized group. The external anatomy here is extremely
simple; no spines or sensoria are present; the mandibles
are very feeble, poorly developed and practically invisible;
the tracheal system is closed.
Group IV:
Another group of very curious and interesting first
stage larvae are also found in the Encyrtidae and repres
the ius ented by genera Microterys, Blastothrix, Sched ,
Phaenodiscus, and certain species of the genus Encyrtus;
are those which are attached to the outer wall of the host
(or egg shell in the case of Schedius, parasite in the
, of Porthetria means of the shell of the .J' dispar) by egg par
a.site itself. The pedicel of the chalcid egg transverses
' , the body wall of the host, and in some cases it is"hollow
and appears to act as a respiratory tube, comparable to the
integumental and tracheal sheaths of the dipterous para-
sites of insects. The bodies of these larvae are composed _)
-
- 24 -
a ' and ten thirteen , head of head J of a segments instead and
usually found in chalcid larvae. This difference is due
the the last four !", � in all probability to fusion of segments
. the but in the I later ,I. during embryonic development stages
these segments are clearly differentiated. :Most of the
ind ividuals of this group are metapneustic, lik.e first-
stage tachinid larvae, possessing only a single pair of
I l : I open spiracles situated on the posterior extremity., Sil
vestri has described many larvae of this type.
Group V:
.. " " There are a certain number of forms . �. endophagous
which float freely in the body cavity of the host 'egg, J ,., ',: �,:
larva or pupa. The larvae of the genera Chalcis, Aspid
I j ,\'. iotiphagus, Cerapterocerus, Tetrastichus are of this type.
They are often designated by the name of "ichneumoniforme •
larvae", in that they are very similar in form to the
primary larvae of certain Ichneumonids. These larvae in
,j "c , IlL ..�...... �.... most cases are strikingly different from even their clos�
est relations, systematicallY speaking, in that they poss
ess a more or less elongate caudal appendate or tail. The
various members of the group of caudate larvae have much
" , in common anatomically, often being almost exactly alik.e
',' in construction although widely separated systematically.
Larvae which have already been studied can usually be
identified specifically when subsequently encountered, but
, .1 - 25 -
larva of this cannot at be an unknown group present
referred with certainty even to its family on morpho
logical characters alone. Such a larva might be a tetra
stichine, a chalcid, an encyrtid or an aphilinid.
AS mentioned previously the larvae are character-
" ' ized principally by the presence of a caudal appendage,
usually of very simple form, although in some cases it is
bifurcate; the head is longer than wide, somewhat pOinted
anteriorly, and bears two strongly uncinnate mandibles;
the cephalic sensoria are mostly inconspicuous and few in
number. The body is distinctly segmented and generally
armed with cuticular spines, true sensorial hairs, how
ever, appearing to be usually lacking or at least very
poorly developed. All of the larvae of this group are apneustic in the first instar.
The asexual larvae of the polyembryonic Encyrtidae
belong to this group also; but they are easily distin
guished from other larvae, by the absence of a heart,
respiratory system and reproductive system.
Group VI:
This group includes the larvae of Torymus, Eurytoma,
Euphemus, Cerambycobius, Leucospis and Ditropinotus. They
. possess a well head followed by thirteen seg , developed
ments; the body is feebly chitinized and unpigmented, but
the head is brown in colour and strongly chitinized; the
- - 26 -
mandibles are more powerful than in some of the other
groups and are more strongly curved, being approximately
· " comma-shaped; each body segment presents two or three
pairs of long sensory hairs; in addition, each segment
bears a median or submedian girdle of large and conspicu
ous cuticular spines; their larvae bear four pairs of
open spiracles.
'-. ,t Group VII:
Still another group presenting well-defined morpho
logical characters is constituted by the primary larvae I , � I
k.nown as "pLan Ld La" in which the dorsal surface 0 f the ,'I " head and body segments is heavily pigmented and strongly
chitinized, so that the larvae resemble miniature arma
dillos. All of the larvae of this group which have been
identified beiong to the genera, Orasema, Perilampus,
PSilogaster, Schizaspida and Stibula, included in the fam
ilies Perilampidae and Eucharidae. In the perilampid lar
vae the body is composed of twelve segments in addition to
the head, while in the eucharids the head is fOllowed by
eight segments in the genera Schizaspidia and Stibula,
with eleven in Psilogaster. All these larvae, with the
exception of Schizaspidia, bear numerous spines, varying
in dispOSition, and which probably are ambulatory in func
tion. They likewise bear a limited number of sensory setae
They are very acti�e, though not more so than are those of - 27 - the Torymidae, Eupelmidae, Eurytomida6, and Leucospidae
in their more res't ricted environments. These larvae are extremely interesting on account of their varying para sitic habits and modes of existence.
The following table, taken from ParKer ('24), shows the larval types of the various families in the su1'erfamiJy
Ohalc ido idea.
Larval Larval Gro up. Family. Family.
I Eulophidae (Aphelininae) VII Perilampidae
II Torymidae (Megastigmini; Eucharidae Pedagrionini; Ormyrini) VI Leuco idae s1' . Eurytomidae (Ar'chirileyia) V Ohaleididae Petromalidae VII Perilampidae Eulophidae
Elasmidae II,VI Torymidae
III Eneyrtidae (Copidosoma, II,VI Eurytomidae Litomastix; sexual larvae)
IV Eneyrtidae VI Eupelmidae
V Eulo1'hidae (Asdiotiphagus; III,IV,V Eneyrtidae Tetrastichus)
II Eneyrtidae (Cera1'teroeerus) Miseogasteridae .
Chaleididae (Chaleis) II pteromal idae
VI Torymidae (Torymus; Ditro II Elasmidae p Lno t rue )
Eurytomidae I,II,IV Eulophidae
Leueospidae Not Known Triehogrammidae
Euphelmidae •
. '- .. _ ------
1 2 3
4 6
5
'Plate IV: Larval types in the Chaleidoidea (I) primary larva of Aphelinus longielava.representing group I. (2) first stage larva of Dibrachys boucheanus,representing groupII.(3) first stage larva of Gopidosoma thompsonii,ramoved fram its envelope in the developing chain, representing group III. (4)first stage larva of Eneyrtus ferrugineus representing group IV. (5)first stage larva of Chaleis fonscolambei, representing group V. (6)first stage larva of TO;ymus sp.,representing , group VI. (7) first stage larva of PerilEmpus hyalinus ,representing group VII. (after Parker and Thompson). ,
\�- - 28 -
2. EITEIULtL ANAT01lY OF BE:;tECYNTUS BAKERI
The larva.
The very young larva shows thirteen distinct seg ments in addition to the head; the sk.in of the larva is of a transparent white color while the contents of the
body are usually of a dirty yellowish-white, the Same color as the contents of the host egg. The head is hemispherical with a mouth clearly visible, but there are no mandibles.
No sensory organs are visible either on the head capsule or on the body. In each segment there is a branching tracheal system but no open spiracles are found. The fully devel
oped larva measures about 2.5 rom. in length and 0.6 mm. in
its greatest width; however, it becomes shorter and more
compact just prior to pupation.
At the end of the larval development, the sexual larva is cylindrical and stretched out, slightly bent vert
ically, not much larger at the centre than at the ends, am plainly divided into a head, and thirteen segments. It is
of whit ish color and the, integument is bare. The head is white, of an almost spherical form, a little wider than it
is long, the opening 0 f the mouth is clearly define d, the
lips being well chitinized. The mandibles are narrow,
stretched out and sharp, of a testaceous color and with .a
broad base. Despite the fact that the tracheal system is
closed in the first-stage larva, the mature form has 9
pairs of open spiracles. -
- 29 -
3. INTERIU.L AUJ.TOMY OF BERECYUTUS BAKER!
The brief description of the internal anatomy that
is presented belongs exclusively to the adult or mature
larva. Due to the a na.l.L larvae being examined, it was
ve ry diff icult to dissect out the varfous syst ems clearly.
Respiratory System
This system is made up essentially of two large
lateral trachea which extend from the first to the tenth
or eleventh segment of the ·body: The trachea resemble
those of other insects: round tubes slightly chitinized.
In the primary stages, the larvae are apneustic, that is
the sp Lrac Ie s are closed and do not function. It is
believed that such internal parasitic larvae as these
derive their oxygen from air. that is contained in the
blood of their hosts, and that this is done by osmosis
through the cuticula of the larva, the SKin of the larva
being furnished with a vne two r-k of fine tracheae. what
ever number of spiracles may be present in the first-stage
larvae, they do not undergo any change until the final
stage when they show- nine pairs of open spiracles.
�he Muscular System.
Great difficulty was experienced in obtaining any
informat ion concerning the mua ou.Lar- system, due to the
fact that the muscular fibers of an insect are not en
closed in a sheath and are therefore made up of many - 30 -
distinct fibers of a minute size and very difficult to
see or stain. The only strands found were those belong
ing to an oblique muscle which was found traversing the
first three segments of the body.
The Circulatory System.
The circulatory s.ystem,of the adult larva is very
simple. The main part (and the only part observed) is
made up of a heart of bulbous shape composed of several
chambers which continues anteriorly as a long simple tube
(aorta). This heart is situated dorsally in segments X
and XI. The tube extends through the thorax into the
head (as in all insects) where it probably opens near t.he
brain. The blood, as in all insects, is nearly colorless,
although the body contents of a larva appear yellowish.
The Nervous System.
The nervous system of all chalcid larvae is similar
to that of other insects; however, due to the very minute
ness of this system, no observations were made.
�he Adipose Tissue.
On opening up the body of a larva, one of the most
conspicuous things to be seen is fatty tissue, in large
masses. The adipose body is in the form of a bed of cells
very rich in small droplets of fat; these cells together
form a networ� situated on each side of the body, in the
\ - - 31 - space between the stomach and the side wall. As a rule, the median dorsal and ventral dorsal regions have no adi pose cells; however, in adult larvae, adipose cells are observed everywhere. In the adult larva, also, the adi pose bodies are easily detached from other organs, chief ly the alimentary canal. Apparently the chief function of 'the adipose t issue is the storage of nut riment.
The Reproduct ive Slstem..
The reproductive system is represented in the male
by a pair 0 f small oval bodies in segment XII; in the female, by a pair of elongated bodies, in segmentX. No effort waS made to study these organs in a detailed manner.
The Digestive System.
The digestive system is, in general, qUite simple.
It begins with a pharynx in the shape of a bent tube. The
oesophagus is very short and is probably constricted at the point where it empties into the stomach. The stomach is merely a slender sac, probably held in place by muscular
tissue. At the last larval stage it is closed posteriorly and does not communicate with the posterior part of the
body. In larvae of certain kinds of insects the stomach
is a closed sac, the passage being closed between the mid
and hind intestine. In these instances the food is always
of a fluid nature and there is but little solid residium.
However, with Berecyntus, the food of the larvae is not ..
- 32,-
entirely of a.' fluid nature and t he re is co ns Lde rabj a
residue. This residue is not deposited until the pre-
pupal stage is reached, at which time, connection between
the stomach and hin�-intestine is re-establish�d. Thus,�.
is another instance of the co-ordination existing . perfect '.
between the host and its parasite, because if this re
sidium were to be freed into the body cavity of t�e host,
serious changes would doubtless be caused.
PREPUPA
Just before entering the pupal stage, the la�vae of
Berecyntus discharge an excessive amount of excrem�nt,
usually to be' found in the form of many irregul,ar pellets
at the anal end of the pupa. This is followed by a
lengthening of the anterior body segments and the shaping
' . of the pupa within the pupal cell. The prepupal form
requires 2 to 3 days unless retarded by unfavourable con
ditions.
:PUPA
The parasite larvae eventually transform internal1y
into naked, more or less coarctate pupae. The larvae,
inha�iting the host insect in great numbers, when about
to pupate, cause a marked inflation in the host larva by
the formation of oval or rectangular cells around the
parasite. This gives it the appearance of being honey
combed; as far as can be ascertained there is no definite - 33 -
method of arrangement of these cells. In Copidosoma
J gilechiae the cells are arranged parallel to the long
itudinal axis of the host. Riley has aptly called
.Q..:. gilechiae "The inflating-chaleis-fly".
An interesting point concerns the na tu re of the
cocoon-liKe cell surrounding each pupa. At one time
Leiley considered this e t ruct Lo n to be an advent it ious
tissue of the host. He discredited the pos�ibilities of
it being a silken cocoon, or a membrane secreted in the
surface of the chalcid or the loosened last .' general body
larval skin of the parasite. However, the writer has
observed the following changes under the microscope: When
the parasite larva have finished feeding, they invariably
shrink in size. When shrinking they shed the outer layer
of the cuticula, and this cuticula hardens around the
larva, forming a pupal chamber. In literature, newly
published, Leiley ('28) tak.es this viewpoint also, stating
that the pupal cell is formed by molting of the mature
larva, just as in the Diptera. He modifies his statement
somewhat by mentioning the posSibility of the cuticula
being ha.rdened by secretiOns f�om the larva. No glands
of any sort have ever been found in Bereeyntus larvae.
There are two main functions for this structure:
(1), it separates the pupae from one another, thus prevent
ing any possible injury to the delicate pupae; the possib-
of to ility injury late maturing pupae from early emerging
\ - 34 -
adults is also prevented (Z) the cells act as a support
for the host, otherwise the body wall would entirely
collaps�.
The pupa, when first formed, is creamy white in
color and closely resembles the adult in shape. Soon
after formation the eyes and ocelli turn pinkish in color.
The appendages are closely folded to the body. The head
and tip of the abdomen are bent slightly forward. The
wing pads, legs and antennae, although folded close to the
body, are visible through the thin pupal skin. AS it gravis
older it gradually turns d�rker in color, until 3 to 4 day.3
before emergence it is of the same black color as .the adult.
The measurements of the pupa are almost the same as for
the adult. Pupation takes approximately 20 to 25 days to
o c cu r ,
\ BIOLOGY OF THE PARASITE
I - 35 - EMERG�NCE OF THE ADULT
The adults shed the pupal ak Ln within their chambers and remain therein after doing do for one or two days. After this time they emerge by gnawing an exit hole through the wall of their chambers and the body wall of the host carcass, thus escaping into the open. The adult stage is reached about 15 to 25 days after pupation. In the field the adults have been ob served from the early part of May until the end of
August, the maximum emergence occurring during July.
As the eggs of � ochrogaster are not deposited until the latter part of August, it is probable that the females live for moderately long periods under field condit ions, the length of life depending upon the weather condit ions.
The host carcass after emergence is covered on all Sides with small holes eaten by the adults in order to escape. In most cases the holes have very ragged
. edges. Some are large, looking as though numerous adults I had emerged through them; others are very tiny with just sufficient room for qne adult to emerge. However, from the number of holes observed in a host carcass, it is
adult in makes an very improbable that each emerging opening of its own. In all likelihood the adults on the outer side of the host. larva (that is, the adults emerg-
and later adults come out ing first) maKe openings through the same apertures. J - 3.6.-
It has been noted fre�uently in the rearing of
',,' Berecyntus bakeri hosts in tin cans and in vials that
the adults take a considerably longer time to emerge
when soil conditions are dry than they do when moisture
conditions of the soil are moderate. The adults may
.... 1 appear to be almost ready to emerge, that is the host
may appear to be black in color (due to the presence of
j', , adults) but none emerge until a little mOisture is added
to the dry soil in the tin, or to the cork of the vial. I II.j
The moisture factor may be needed to soften the .' • help j
, , tissue of the host larva, the adults then finding it . , ,\
easier to eat a moist tissue than a , through through dry ; , ,I ,
brittle one. !U.; I .',
After emergence, in each little chamber, there J., ','
j if found a number of small round pellets, these being
larval exuviation and the pupal skin. When E. ochro
gast!! is the host, only a small number of these pellets
are while with E. tristicula there is found (7-10), - .
invariably a larger number, usually between 12 and 25.
No explanation of this is suggested, although it would
appear that different varieties of the same species
might be attacking the two hosts mentioned. In each
host larva there is always found a small number of dead
do or diseased pupae and adults which not emerge.
Adults from the same host were usually of the same size
and emerged at the same time. on a check of some field - 37 -
material, it vae found that emergence in the same brood
continues two a.nd even three days after the first day of
emergence. i7hile not nany records of the length of eme r-
t n no ted that t he numbers were some- gence were ake , it was
" times qUite noticeably greater on the second and third days.
Wolcott· in work.ing viith the chaLc i.d Trichogramma
minutum Rile y found this parasite to show an emergence
trouble with response to light. H� had, it being strong
ly positively phototropic, so placed his cages in the
dark. When the cages were removed from the dartt,he found
the the great er number c ontaine d no adult s but that in
next hour of exposure to light a large emergence took
normal time place. He came to the conclusion that "the
hours sunrise" and of emergence is approximately two after
minutum that in his cages "--six times as many adults of !.!.
hour of emerged in the first hour after dark per previous
some such con daylight in the same day". With B. bake rt
in the dark dition has been noted. Host carcasses placed
of adults in normal time. usually allowed the emergence
I ; the either no However, from host carcasses placed in light,
occurred or the dat e of was emergence whatever emergence
and resulted in a much higher delayed materially usually
understood when mortality than normally. This is easily
normal of is in one ea nsiders that the place emergence · '�------
- 38--
the soil. There is also the possibility that the de
layed emergence when exposed to light was partially due
to other factors, such as temperature and moisture. But
this oondition has occurred so often under laboratory
conditions that more weight must be given to the lif:ht
been response than to other factors. No information has
obtained as to the time of the day when normal emergence
reaches its maximum, nor is there any indication that
such a condition exists.
Some interesting observations were recorded con
cerning the emergence of adults from a host whence a
mixed brood issues. In practically all the cases coming
both from one host to my attention where sexes emerge
larva, the emergence of adults taKes place in a very
" � I "
short time (about 'one hour). As the adults usually lin-
, ) .
for some time after ger near the host carcass emergence, i' _ •• the possibility of copulation taking place is considerably
the first enhanced. In a few instances, the majority of
adults emerging were males; this also increased the poss
seen that ibility of fertilization occurring. Thus it is
and although parthenogenetic development is possible
these two factors concerning occurs quite frequently,
the of adult emergence noted above, decreases poss.iliility
result if eventual extinction of the species which would
tOOK only parthenogenetical development place. - 39 -
LOlIGEVITY OF THE ADULTS, THEIR HABITS AND REACTIOl;S
In order to secure data on the duration of the adult stage, 80 adults (58 females and 22 males) were distributed
into three groups and subjected to different conditions.
Group 1 included 20 females and 6 males, placed without nourishment in small vials plugged with cotton and exposed to the air in the laboratory in a dry condition. Group 2
conSisted of 20 females and 18 males in small vials plugged with cotton, the plugged ends of which were inserted in a plaster blocK Kept constantly saturated with water. The
cotton absorbed sufficient moisture from the plaster blocK
to supply water for ingestion. Group 3 contained 18 fe
males and 6 males kept in a saturated atmosphere in the
same manner as group 2, but were provided with sugar solu
tion for nourishment. The results are summarized in
table I. It will be noted that the greatest length of life was 20 days; that in each group the average length of the
female exceeded that of the male; and that the parasites
left in a saturated atmosphere with sugar solution for
nourishment had the longest average length of life, which
amounted to 11.2 days. Throughout the experiment, the
temperature of the laboratory in which it was conducted
varied from 450 F. to 780 F. with an average temperature o of 61 F.
At a later date an experiment somewhat similar to
the one described above was run, with the object of - 40-- learning the importance of moisture on the adults. Two groups of parasites were used, 20 in each (10 males and
10 females). Group 1 was placed in a dry atmosphere and with sugar solution for nourishment. Group 2 was placed in a saturated atmosphere and with sugar solution for nourishment. The results are summarized below in table II.
TABLE - LENGTH OF LIFE OF BERECYNTUS BAKERI ADU!.TS
(Group 1 was kept in a dry atmosphere with no nourishment available; group 2 in a saturated atmosphere with water a9cessible for ingestion; group 3 in a saturated atmo sphere with sugar solution for nourishment.)
Male Female No. Longevitl No. Longevitl Total Group. Used MaXimum Average Used Maximum Ave. Average LO!lQ:evit:v- .-
I 6 5 days 3 20 6 days 5 4.5days I I 1 2 10 12 6 20 h6 9 days , days I S.Odaysl t 3 I 6 days 9 IS days 12 I 115 I 120f 1 J llo2days!
TABLE II -
(Group 1 was Kept in a dry atmosphere, Group 2 in a sat urated atmosphere, and both were supplied with sugar solution for nourishment.) . ------P------� ----.. ,
- 41
TABLE II -
(Group 1 was kept in a dry atmosphere, Group 2 in a saturated atmosphere, and both were supplied with sugar solution for nourishment.)
. Male Female I l. , I ITotal No. Longevitl No· , LO£gevi tl IAverage
Grou'D Used Maximum'Avera�e Used :rvfu.x imum Ave. Longevit.Y • ! , 1 10 8 6 I 10 10 7 6.5 , I I , I daysl 2 10 14 I 9 10 19 12 11.0 days I , , t I ,
From the above results it is seen that the average
longevity of adults is increased only by two days when
they are kept in the same dry atmosphere, and one group
supplied with sugar solution. It would appear that dry
conditions seriously influence their activities and that
the addition of sugar solution is not sufficient to en
able them to live for a long period. This evidenc·e also
would interlink with the previous work on tropisms.
be Since the atmospheric moisture would usually less in
would amount during hot sunny days, this factor cause
the adults to be partly negatively phototropic in re-
• action under natu�al conditions.
Extremely low or high temperatures undoubtedly
life. Several have some effect on the length of adults, o of 24 F. for however, were subjected to a temperature
noticeably ill effects. a period of 5 hours without any
- � -
�1 � � __-_-_-::_- ..�.-� ·J ..
�. 'i
GroupL. IT. ill, lA.
. Chart I: length of life of adults of Berecyntus bakeri. (see tables land II) ,
- 42 -
Adults subjected to heat expired within one minute at
temperatures from 1050 F., to 1200 F.
Adults have been Kept alive for 15 days in large
outdoor cheesecloth cages measuring 2 x 2 x 1 feet.
Here they were observed to feed upon cotton soaked in
sweetened water and upon droplets on honey scattered
here and there on sweet clover leaves. Other parasites
that were not fed died when placed in similar outdoor
cages within one weeK after emergence.
When in confinement the adults crawled rapidly
but very seldom flew. Although equipped with wings, the
adults of both sexes appear to laCK the power of sustain
ed flight. Locomotion seems to be accomplished largely
by means of jumping or running. Adults that were ob
served in the field running on the surface disappeared
with startling rapidity when an attempt was made to
collect them, their movements, when approached, resembl
ing those of flea-beetles. This ability to jump made
the species very difficult to handle in the laboratory
the of cages and many adults were lost during process
feeding or when they were being transferred from one
vial to another. It was found that in all cases the
»' males were much more active than the females. Although
the small size of the parasite makes it very difficult
to observe under field conditions, it was noted that the
the wind. adults were spread short distances by
. ....
-."-- ...... -�-.__...... __ .:;:-_.--�-:.::.._----_-__=.== .:-:;,;:;.4 ,
- 43--
In confinement, the adults generally moved away
trom the light. This was found to be more true with
adults emerging from � ochrogaster than for those emerg
ing from � tristicula. However, with both this negative
ly phototropic condition was noted to some extent, although
there must be of necessity a great variation in the degree
of this tropism. When suddenly disturbed they feigned
death by drawing up the legs and antennae close to the body
and remaining in this attitude for a few seconds. In
actual death the antennae and legs are found stretched a�ay
from the body. When at rest the body is usually held in
crouched position, but with legs and antennae not drawn as
close to the as when death. I :�' body feigning
For nourishment, the adults readily ta�e up sugar
or honey solution. In laboratory experiments a mixture of
about 40 percent honey and 60 percent water has been found
the best food tried. When allowed to go a few days without
water, they became very thirsty, as was shown by the quic�
ness with Which they accepted it.
In the field, the adults have been found walking
over the surface of the soil in moderate numbers. A few
have been noted on alfalfa and clover plants. lIet sweep
ings have revealed their presence on these plants in great
er numbers than could be seen with the naked eye. There
the is no doubt but' that these two forage plants supply
... - ,�--- - - 44 - parasitic adults with easily accessible food, and their presence on these plants was for this sole purpose. It has been reported that adults have also been found on the heads of tall sunflowers. Since there is a'very plentiful supply of nectar present in sunflower heads, it is quite possible that Berecyntus might be 'found there, but I doubt if it is able to fly such a height. Of course, it could easily crawl up the stalk. The adults will lap up water and the juices of a wheat leaf which bas been bruised or brOKen, and will nip at the short pubescence on the wheat leaf. In a large field cage in which wheat was growing, some nocturnal habits were observed with a flashlight.
Many of the adults were found at twilight sheltered on the groove of the wheat plant. Here they apparently spent the night, bunched together in a compact group with their antenna� drawn in against their body and their legs bunched. They were not disturbed even when a strong flash light was directed on them from a short distance. Evident ly they were fast asleep_
- - --J,�' __-_�.-_'_"_"'�=_��'_'�.�_-_- __�4 , - 45 -
PO�ElnIAL PROGENITIVENESS
In order to ascertian the full egg-laying capacity
of Berecyntus ba�eri adults, the ova contained in the
, ovaries of 10 individuals were counted. The abdomens of
living females were crushed under a cover glass in gly
cerine and the ovaries forced 0 ut under pressure. ( In
this species the ova found in the adult stage are fairly
well developed in size at the time of eclosion). In count-
ing these eggs a very high degree of accuracy was made
possible by spreading the eggs from one ovary at a time
over an ocular micrometer disc ruled into 1 mm. squares.
The results showed a maximum number of ova per individual
·of 300, a minimum of 145 and an average of 200. More eggs
were sometimes found in one ovary than in the other of the
same individual. Assuming that a female oviposited to her
full capacity, and that on an average 1000 adults are pro
duced from a single egg, the progeny from an average tamale
would come to 200,000 adults. Deducting from.this figure
the 35% which will develop into males, we have left a total
of 130,000 females with the capacity of producin� 130,000,000
adults for the following generatioh. It can be thus easily
understood how easy it would be for'this species to rapid-
its ly cause the .disappearance of its hosts and eventually
own destruction if it were not for the natural restrictions
placed on its reproduction.
.. �,---- -_-_-_-__._-'··�==_-_"���=� ���..AI ,.
- 46 -
..: TJ..B-::: I
lruI.:BE� OF E'iG3 COUNTED IU OV.�R�S
lIo. of egq:s ITo. of eg:g:s
Adult no. 1. 200 Adult No. 6� 145
no. 2. 300 Uo. 7. 175
No. 3. 150 No. 8. 200
No. 4. 220 No. 9. 210
No. 5. 240 No .10. 200
The greatest total number of o v Lpo e Lt Lo ns made
under the writer's observation by an individual Berecyntus
female in confinement amounted to about 80 and they were
all depasited in the caurse .of 48 hours; and there is no
reason to believe that if sufficient host materDl had been
available there would have been many mare laid. Of course,
under natural c.onditions .oviposit ian is limited by the
difficulty of finding host material.
� ���� �===_· ·--_·_---- A. .. "'------.._ - ,
- 47 -
COPULATION
The union of sexes and fertilization of the
female takes place in the manner common to most chalcids
soon after .the adults emerge and lasts but a few seconds.
The males seemed more active than the females. On con- I
tact with one another, instantaneous recognition of sex
waS seen. If both were males, they kept on moving. On
meeting one of the opposite sex the female was very quiet
(if not previously fertilized) while the male grasped her
wing tips with his front pair of legs, hooked his head
over the tips, swung up and held to the wings with the
other legs. Thus hanging. in an underslung position copul
ation took place, usually lasting from 5 to 10 seconds.
During copulation the males kept their wings in constant
motion, probably for balance. Some females kept moving
about while was carried on, as though attempting copulation I
to shake the male off;, others remained quiet.
After copulation the male separated from the female
with a sudden jerk. The female remained quiet and very
carefully preened its abdomen with its legs. The male
wandered off, cleaned its body but not so carefully as the
female. The male would perhaps go immediately to some
other passing female. If a female already had been fert
ilized, and is met by a male, it jumps in the air away
from the male (noted often), thus suggesting that fertili-
zation occurs only once.
'- .. �., _ .... - ...... __. ,
"!' 48. -
The presence of females is apparently recognized
by the males without establishing contact with the ant-
ennae, since upon their introduction into a cage contain
ing males, the males become suddenly and increasingly
act iva. ,�------,
- 49 - OVIPOSITIon
The process of oviposition was often observed under
laboratory conditions. Oviposition often begins, during
the first day of adult life and extends over the whole per
iod of adult longevity. In one instance a female of
Berecyntus bakeri was introduced into a tin containing
Euxoa ochrogaster eggs. This female had been contined in
a vial co nt a tntng many adults of both sexes and had prob
ably been fertilized. It had not previously been exposed
to any eggs of its host. Immediately upon being introduced
into the vial, the female began vibrating her antennae
rapidly and swinging her head slightly from side to side
in order apparently to cover more area with her antennae.
The finding of a host egg by the female is determined by
what is apparently only accidental contact with the distal
segment of the antennae for otherwise the parasite may
walk beside or even over a host egg and not recognize it.
It is known that the sense of smell in chalcid adults is
very acute and is usually located in the antennae and in
this specific instance the difficulty of finding small
host eggs in the soil is so great that the antennary sense
is very highly developed. As soon as she identifies an
egg she halts and apparently concentrates on the spot since
an increased rapidity of motion of her antennae is noted.
Sometimes it may happen that she seems to lose track of the ,
_. 50 -
egg momentarily but when this occurs she usually turns in
small circles until it is found again. When the egg is
located she strokes it quickly for a moment with both
antennae and if satisfied with this examination, quickly
draws up her abdomen so that the tip of the ovipositor
touches the surface of the egg.
Through the binocular microscope it was seen that
the female was poised over the selected egg, the legs
braced against the Sides of the adjoining eggs, and most
of the pressure being exerted through the posterior pair
of legs. During this period the tip of abdomen is at
first raised slightly, the ovipositor is withdrawn from
its sheath and directed downward and baCkward, while the
antennae are bent in front of the head. The ovipositor
is directed from one end of the egg to the other as if
to make sure of its position and character. If all is
well she brings the ovipositor to a halt at about the
middle of the egg and commences to oviposit. Apparently
to the adult the chorion of· the host egg is of the same
toughness in all places, although the basal end is
supposedly more delicate. In no case was it noticed
that more than a part of the length of the ovipositor
was inserted. In a few seconds the ovipositor is brought
in contact with the host egg, which is finally punctured
by a slow drilling movement, after which the egg is de-
d
__ P_OS_it�...... __W_h_e_n_t_h_u_s_e_ng_O'_a_g"",,e_d_t_t_h_e_f_e_m_a_l_e_i_s_n_o_t_e_a_S_i_l_y -IIIIIJ_ '
--- -
r
_- 51 -
disturbed and the function is normally performed even in
small capsules 0 r vials.
After the opening in the egg seemed to be completed,
where a small drop of brownish liquid appeareo at the _point
the ovipositor was inserted in the egg. It is impossible
to state whether this liquid exuded from the female or came
to be from the egg. In some cases, the female appeared
sipping the liquid around the opening in the eggshell, or
possibly feeding on the contents of the egg. After the
egg was deposited the female turned around and manipulated
her mandibles and antennae in the small drop of liquid
which remained over the opening in the egg. This liquid
soon hardened and formed a transparent, light brown, waxy
substance which apparently sealed the o vf.po sLt Lo n puncture.
This puncture spot is characteristic of parasitized eggs.
After completing oviposition in this egg, the female re
sumed her examination of the remaining eggs in the cluster.
The following records were made at different times
of the day on the duration of oviposit ion. The last two
counts were taken on females that were almost spent.
DURATION OF OVIPOSITION
Female No. Duration Length of feeding at puncture
1 50 sec. 1 min. 30 sec.
2 58 sec. 5 min. 5 eec.
----�- ...... ------'------I �,-----.------�� -�- ,
- '52.-
Female Uo. Duration Length of feedi� at Euncture
3 60 sec. 2 min. 40 sec.
4 60 sec. 3 min. 35 sec.
5 50 sec. - - -
6 70 sec. 2 min .• 20 sec.
7 70 sec. 4 min. 40 sec.
8 65 sec. 90 sec.
9 90 sec. 2 min. 10 sec.
10 95 sec. 2 min. 10 sec. I
. It has also been observ&d that the longer a female
oviposits in succession, the more tired it becomes and
hence takes longer to complete its oviposition.
Inside the laboratory the females oviposited any,
time of the day but when t ake n into bright sunlight, they
refused. Under field conditions it is most likely that
oviposition occurs throughout the day, although the great
er part would probably occur during the cooler portion of
the day. The parasites seemingly pay some attention to
the age of the host eggs. A group of newly laid eggs
when to exposed parasites invariably showed a higher per
centage of parasitism than did eggs that were much older.
fact is This borne out by an examination of eggs of
various ages for traces of the egg punctures. The follow
counts ing of egg punctures on 50 eggs of various ages
throws some light on the subject. The same number of · I
) I I I I c.
___ oVIposit/on _____ .f,,("Jillj a.t F';Jnc-tu.p.�
-_
"/ / / / / �
......
-- -... __
>
- __
100
at Chart II: Duration of ovi�osition and feeding �uncture. - (see table I)
I "1 ".�� Jl __���=- �� � ,
_'!:O ,53.-
females (5) was used for the same length of time (48 hours)
in each group of eggs.
NUMBER OF EG� PUNCTURES IN 50 EUIOA OCBROGASTER EGGS
1. Newly laid eggs - 28.
2. Eggs, 3 days old - 20.
3. Eggs, 1 weeK. old - 12.
4. Eggs, I month old - 7.
From the above results it would appear that female
parasites find it easier to o v Lpo s Lt in eggs not older than
3 days. A probable explanation of this fact may be that
the chorion of the host egg remains in a semi-plastic con
dition for a few days after oviposition and is thus more
easily pierced than after the hardening process has taken
There is no doubt that the adults place. do try to ovi I I in old but posit eg�s apparently they find it very diffi , cult to pierce the chorion. In cases where they do succeed
the chorion has not become as hardened as it normally would
be (due to a variety of causes) and in these Cases it bas
been noted that the oviposition puncture is always found
in the basal part of the egg, where the chorion is usually
more delicate,. Even in these instances, it takes longer t to be for oviposition accomplished. However, with ne�ly
or a few deposited eggs eggs only days old, there is no
definite correlati�n between the punctures and the basal
of the host It part egg. has been noticed that adults in - 54 .:.. ovipositing in older host eggs run their' ovipositor over the surface of the host egg, and apparently the process of oviposition is accomplished only by means of a trial and error method.
The parasite seldom deposits two eggs successively in the Same host egg but after crawling away in search of another egg, it may accidently return to the same egg and oviposit again in it. Sectioning of host eggs has re veal�d no instance where two parasite eggs are deposited at the same time.
Laboratory studies also show that the female para sites are not reluctant to oviposit in a host egg that has been previously attacked by one or more other females.
Indeed it is not difficult to have as many as 7-10 para site eggs in a single host egg. This has been done a number ot times under experimental conditions so that the I early development of the eggs could be studied. Such ·1 frequent oviposit1ons by various females in host eggs do I not probably occur in nature, but the females of
Berecyntus has a decided tendency to fail to recognize the fact that a host egg has been previously oviposited i in another female. This is not the case with by generally ,I I I other species of polyembryonic hymenoptera. 1 I
,. -4 ,
- 55 -
PA�THEUOGENESIS
Experimental rearings have shown definitely that
parthenogenesis·does occur in Berecyntus baKeri. In such
a case the resulting offsprings are invariably males.
During the winter months of 1928, a number of experiments
were conducted in the laboratory to determine whether
parthenogenesis does occur. The date from these experi
ments are given in the section on "Rearing data on the
sexes". However, a short summary of the results is of
interest here.
Only virgin females of � baKeri were placed in
lamp chimnies containing � ochrogaster eggs as hosts;
they oviposited about as freely as those in other cages
where there were both sexes and the act of oviposition
seemed normal. Males emerged from all the larvae origin
ating from these females, cle�rly showing that the species I is parthenogenetic and indicating that unfertilized females
give rise to a generation of males. However, the average
number of the progeny is much lower (between one hundred
to two hundred and fifty less) than the average number
,'," ' ." obtained in similar experiments conducted with fertilized
females.
It has been seen that the unfertilized egg of
� baKeri produces only a polyembryonic brood of males.
must be characterized the r The 'germ cells of such males by
haploid numbers of chromosomes. The presence of the half - - •. 56. numbers is due to the fact that in the parthenogenetic development of the matured egg, there is no compensating process for restoring the full number. As a consequence of the reduced number of chromosomes in these males, the maturation divisions probably would be modified in such a way that two, instead of four, spermatozoa are produced from each spermatocyte.
Due to the fact that under natural conditions the adults must find it very difficult in the soil to find the opposite sex, it would almost appear that partheno genetic development was the most frequent method of re production (some copulation, of course, occurs at the time of emergence). If such were the case, agamic generation would then be usually accidental and not self-perpetuating; and its occurrence in this species would be either the beginning of an alternate generation or, according to the theory upheld by Adler in the Cynipidae, a surviving race of an early complete and total parthenogenesis •. However, if only parthenogenetic development occurred, eventually it would lead to the extinction of the species, through the development of allrmale broods.
Anyone having practical experience in breeding in sects �nows that of all the physiological processes con- cerned in reproduction, that leading to pairing is most liable to be affected by the abnormal conditions usually found in c3.ptiVity. Many insects that, when fertile, will - 57--
oviposit freely and may easily be reared in captivity, will
not pair except under special conditions of space, heat,
food, moisture, etc. But as well as in captivity, abnormal
conditions would be encountered in nature by the spread of
a species into new localities or, less frequently, by a
change in the climatic 'conditions in its old home. In the
former case, the occasional accidental spread to a locality
isolated from the main habitat is more likely to present
new conditions than the gradual spread on the borders of
the present range.
If the species is one which can produce young par
thenogenet ically, this interference with the process of
pairing will have a direct and cumulative effect on the
ratio of the sexes. The change in sex ratio can be treated
if certain assumptions are in order to mathmetically made. I I � reduce the problem to a very simple case. These are (1) I
that the size of the colony remains constant. (2) that
there is no differential mortality between the sexes or
I between the fertilized and unfertilized eggs. (3) that , i
the fertilized eggs produce equal numbers of each sex and
lastly (4) that the species is momogamous.
First, there is considered the condition of the
sex ratiO in a normal locality where all the individuals
pair as far as possible. Whatever be the result of the
parthenogenet ic eggs there will be a permanent position
of stability at 50% of each sex, and if some of the "
. .1 ��-- .. ----....,._.-=..------__"".-- ,
- 58·-
temales were aocidentally destroyed in one generation,
there would be a return to the equality in the next gen eration. If, however, the number of males was reduced
below in 5�� a monogamous species, not all the females
will be able to pair. The result would now depend on the
sex produced the by untertilized egg. With � baKeri, in whioh parthenogeneSis invariably produces males, there
would be an increase in the percentage ot males in the
next generation followed by a return in following gener ations to equalit�.
We have next to consider the effect of failure to
pair on the normal 50% equilibrium. TaKen for complete
ness that there will be three different strengths of
disability, namely 10�, 30% and 100%, that is, that in
the first case ten out of 100 of the sex . ,�,. every , specified
do not pair, in the second case 30, and in the last case,
none pair. The following table shows the results to be
expected from such co ndit ions •
I ' , •
T 1 RT.H! T -- J" 10% 30% 100% Females Males Females Males Fema.les Males PI 100 100 100 100 100 100 Fl 90 110 50 150 0 200 Fa 100 100 100 100 0 0 F3 90 110 50 150 - - F4 100 100 100 100 - -
In this case we have an alternation between the
equality of sexes and an increased number of males except
in the extreme form, where the males increase to 100% in
: �
j �-- -
,
- 59�
the first generation and, as a result, the colony dies
out.
It has just been shown tha.t in parthenogenet ic
development producing males, if pairing is prevented the
colony dies out owing to the production of a generation
consisting entirely of males. An extreme case of in
ability to pair is in the foundation of a colony in a
new locality by the accidental production of a single un
paired female or a few isolated individuals. Such an
occurrence is taking place frequently under natural con
ditions. In a parthenogenetic species producing males,
.' . not only must both sexes be introduced but they must be
there at the same time and close enough together to en
able them .to find each other. �hus it is seen that
� bakeri would be under a conSiderable handicap in ex
te�ding its range of distribution.
I� 1, -= =- _ ,
- 60 -
APPEARA.nCE OF P.tRASITIZED EGGS
Internal appearance.
For a long time after oviposition, the internal
appearance of a parasitized egg does not differ from the
normal, the contents consisting of a dullish-colored
liquid. The only change apparent is a slight thic�ening
of the liquid contents.
External appearance.
Superficially, the external appearance of a para
sitized egg does not differ from the normal, there being
no change in its color or shape. Microscopic examination,
however, reveals the presence of the sealed oviposition
puncture, and this is the only external indication by
which a parasitized egg may be recognized •. Normal Euxoa
ochrogaster larvae have frequently hatched from eggs
having oviposition punctures of the parasite. This is
probably due to the fact that these particular eggs were
not successfully parasitized.
I �
"'" � AI . _ j - 61 -
FEEDING OF PAR�SITIZED LARVAE AND NON-P_L�SITIZED LARVAE
It has been proved by various wor[ers, chiefly
D. G. Tower with the army worm (Cirphis unipuncta) and
S. S. Crossman with the gipsy moth (Porthetria dispar)that
larvae parasitized by internal parasites eat less food than
normal larvae. With the armyworm, parasitized by Apanteles
militaris, a braconid endoparasite, approximately one-half
the amount of food waS eaten, while healthy gipsy moth larvae
eat two to three times as much as larvae parasitized by
Apanteles melanosceles.
In nine larvae out of ten of the red-baCKed cutworm
parasitized by Berecyritus, there is an additional instar.
This seventh instar or !:. ochrogaster has an average head
width of 3.12 mm. in comparison with a normal head-width
in healthy larvae of 2.75 mm. The body length of the para
sitized larva is also considerably greater than in normal
larvae. It would be reasonable to believe then that the
larger size of the parasitized larvae would result in a
greater amount of food being consumed. Moreover, due to
the fact that there is such a large number of paraSitic lar- I
vae to be nourished, this supposition would be strengthened.
Several feeJing records were kept of � ochrogaster
larvae which were known to be free of paraSites as checkS
against similar feeding records of larvae in which Berecyntus
was present. While the method of estimating the amount of I � �------==�==------,------_.".. - 62 - consumed Vias not it v.aa con- -.' food entirely satisfactory,
"". sidered to be accurate enough to show the trend of the feed-
ing. The amount of food consumed v:as measured· in square
inches, geranium leaves in squares being used to simplify
calculat ions. Tvyo series of experiments were conducted, one
in which the amount of food consumed during the whole larval
life was measured, and in the other the amount consumed dur-
ing the last three instars, the period of greatest normal
s hov. the . The tables obtained from " feeding. following results
the feeding experiments.
AMounT OF FOOD CONSU11ED DURING THE PERIODS
Lr WHICH THE GREATEST FEEDIUG OCCURS IN EUXOA OCHROG.t\SER
Amt. consumed by Amt. consumed by No. of adults Larva no rma l, larvae. parasitized larvae emerging from No. in 4,5 & 6th inst ars in 5,6 & 7th instars parasitized 1.
" ,
1 17.5 sq. ins. 19 sq. ins. 425
2 20. 15.5 300
3 14. 22. 525
4 18.5 22. 675
5 22.5 27. 700
6 16.5 30.5 950
7 23.5 21.5 500
8 20.5 29.5 850
9 21. 28.5 850
10 21. 30. 1000
_tVE.=tJ.. GE 19.5 sq. ins. 24.5 sq. ins. 677.5
I.: _.I - ·63 -
TAB1E II - AMOUIlT OF FOOD COUSUlllED DURING THE
COMPLETE LIFE-CYC LE OF EUXOA OCHROGASTER
Amt • consumed Amount consumed Number of adults larva by by emerging from No. no rmal larva paras it ized larva parasitized larva.
1 18. sq. in. 26.5 sq. in. 650
2 19.5 34. 970
3 24. 26. 610
4 26. 31.5 900
5 24.5 22.5 510
6 27.5 28. 950
7 28. 28. 800
8 24. 32.5 915
9 18.5 28. 850
10 19.5 35. 900
�VERAGE 22.95 sq. in. 29.2 ·sq. in. 805.5
Percentage increase - 27.5%
The above experimental studies were conducted in the
Laboratory where the temperature varied from 450F. to 78oF.
with an average' temperature of about 60oF. The eggs in both
'experiments were all obtained from the Same moth and laid
within a period of three days. In the first experiment, both
normal and parasitized larvae were given an abundant supply
----��=------I 5"1ua.r-e .. '5tf. i n " ...... _ <, ",.,'!' -- ...... � .. _P'! . , 1- _ .••. - •.• 'Chart III: �ount of food consumed by healthy larvae of Euxoa ochrogaster and by larvae parasitized by Berecyntus bakeri.(see tables land II) _, 64 - of food before the period at which a record was �ept of the amount eaten. The tins containing the larvae were examined , each day and fresh squares of geranium leaf introduced-when the previous squares lost their freshness. The data thus obtained definitely indicated that para sitized larvae eat more than normal larvae. In the two ex- periments, the parasitized larvae in one consumed approx imately 25% more food than unparasitized larvae and in the second 27% more. This is in line with what one would expect on account-of the size of the parasitized larvae. From these results it seems that the abnormal size of the parasitized larvae is due to the extra amount of food consumed. However, the writer is of the opinion that this is only of secondary importance and that the inflated size of the larvae is primarily due to the large number of internal parasitic larvae present. No doubt the extra amount of food consumed is chiefly for nourishment of these parasite larvae. It was also seen from the feeding data that there is considerably more feeding in parasitized larvae in the earlier instars than in healthy larvae. Also feeding in parasitized larvae stops at least five days previous to the prepupal stage than it does in the normal larvae. During this five days, appar e�tly, the parasitic larvae are nourished from the body con tents of the host; only a few Shreds of muscle, the alimen- tary canal and tracheae are left undevoured. I - 66 - COl!PARISON OF T� AlJOUNT OF T n.s SPEHT IN THE DIFFER!WT ST�G�S OF DEVELOP11IENT BY NORIiLi..L AUD PARASITIZED LARVAE In connection with the feeding experiments conduct ed to obtain the amount of food consumed during the comp lete life cycle, some data were obtained as to the length I� ,. i ot time spent in each instar by the parasitized larvae. / I.� , . These data are shown in the following table, contrasted with the data obtained by Atkinson as to the length of time spent by normal ���oa ochrogaster in each instar. The experiment was carried on in the laboratory with the temperature ave,rag- o 0 ing 62 F. or 17 C. TABLE I - TIME SPENT IN EACH IUSTAR BY EUXO_� OCHROGASTER LARV...tE Normal Parasit ized Increa.se Increase stage la.rvae larvae in days in percent 1st instar 2.2 days 2.3 days 0.1 days 4.5% 2nd Lnat a r 3.8 days 4.0 days 0.2 days 5.2'% 3rd instar 5.6 days 6.5 days 0.9 days 16.1% 4th Lnat a r 6.4 days 9.5 days 3.1 days 48.4}6 5th ins tar 8.6 days 10.2 days 1.6 days 18.6% 6th instar 16.3 days 12.5 days 11.2 days 68.7% 7th instar 15.0 days Prepupal ( 6.3) days (2.5)days -9.0 days -34.6% Pupal 26.0 days 17.0 days TOTAL 68.9 days 77.0 days 8.1 days 18.1% Ave. increase � _._- .. _ I I. - 'I J. , "\1 ..___ Pc\'y-u�;;iz.r'. - 1 ___ ""'0"".', Vetficull;n�s. roc rea» .. IOJrt'1«; I . r lIe ,"�I . . "' ...... " � -, . � ....,...... lo. , . "i :'_. J •• J ., "j Cllart IV: of time spent in the different of , I Length stages development by nomal Eu:x:oa ochrogaster larvae and by larvae parasiti zed by Berecyntus baker1.(see table I) ' , ... - 67 - From the above data it is seen that the development of the parasitized larva is retarded and that a longer per iod of time is spent in each instar than by larvae free of the parasite. Despite the fact that feeding in parasitized larvae stops at least 5 days earlier relative to the pre pupal stage than it does in the normal larvae, there is a much longer period of feeding in such larvae. From the above table it is seen that 42.9 days are approximately de voted to feeding by normal larvae, while 60 days is the per iod in which feeding occurs in p�rasitized larvae. This longer period of feeding probably accounts for the increase in amount of food consumed by such larvae. A noteworthy feature of the above table is the fact that it clearly in dicates that although there is relatively more feeding in parasitized larvae in the earlier instars than in healthy larvae, the amount of retardation experienced in the earlier ins1ars of parasitized larvae is considerably less than in later instars a.s measured by the length of time spent in I . each instar. This is merely another example of the perfect- ed association existing between the parasite and its host, Since the earlier stages of the host (and also.the most del icate and most easily injured) are not influenced to any ex tent t�t the host will be injured and eventually react against the parasite. As mentioned above, the experiment waS conducted with an average temperature of 62 degrees F. Under the varying �,._ - 68 - temperature conditions prevailing in the spring and early summer, the length of time spent in each instar would vary to some extent from the period obtained experimentally due to the temperature conditions. Table II gives the mean daily minimum, average and maximum temperature for April to July at saekat oon, base d on the records of a ten-year pe r Lo d • T .Li3-=.E II - l'lE_�H DA.I�Y TEMPER_l..TURES IN DEGRE�S FAHRElmEIT . AT ,sASKATOOU, BASED on TEI1-YEAR AVERAGE • Average Maximum Uinimum April 42 48.1 27.2 53 63.2 37.7 M3.y , June 59 71.7 47.2 July 64.2 77.9 51.8 In view of the fact that the eggs are generally with in the first half of the soil and the larvae and pupae v.ith- in the first two inches, the above figures have considerable ./' . significance. It is noted that the average temperatures of the first three months are all below the t empe rat ur e at which the experiment was run (62o). The July temperature is very close to that used in the laboratory. As a result, an ad justment must be made for temperature. This would bring about a considerable retardation of the early instars o�ing to t he low t€mperatures os A.pril, llay and early June, follov' ad by a slight aoceleration of the later stages during late �bout the of by higher temperatures j - 69 - these months. It is a rational ae sumpt Lo n to state that the avera�e len�th of the life cycle of normal Euxoa ochro�aster under natural conditions would be considerably greater than .. · 68.9 days as recorded experimentally, and it is believed that 80-89 days would be a more approximate est imat e. It follov,'s from this that there would be a corresponding increase in the average life-cycle of a parasitized host, from 77.0 days as recorded experimentally to 90-95 days. AS a result of this, the amount of food consumed by parasitized �vae would be relatively greater than recorded in the feeding experi ments. HOSTS " Parasitism by Berecyntns bateri is not specific. At it has been reared from Euxoa ochrogaster Gn., ' .. Saskatoon, 11 (J is Euxoa detersa "ilk.. , Euxoa tristicula Morr., Choriza9'rot A .�:< " • is Sm. ia thanatoloQ.'ia Sm., Rhizagrot flav'icollis , Felt _!!E : erabilie WIt., and Feltia ducene WIk. The variety euxo ae t., ' t., .. ) " has also been recorded by other workers as parasitizing the army cutworm, Chorlzagrotis auxiliaris Grote, and the'glassy cutworm, Sidemia devastator Brace. The parasite appears to found exhibit a marked preference for species whose larvae are in cultivated land. < , Induced pars it ism by Berecyntus bakeri has been ob tained in the laboratory with the following spe cies- Euxoa vert ical is Grote, Emwa tessellata Harr., 'Poliarenigera Steph. �--- ___'''''"' J - 70 - Euxo3. dargo Stkr., Eu.._�oa divergens WIlt., and Emma. .£e.!!!- pestris Grt. There may be some doubt concerning the last species, since there was some confusion in determining the adults of � campestris and � verticalis. It would almost appear that Berecyntus bakeri is able to parasitize any. eggs readily avai.lable. ' Under field conditions it is more or less limited to t bo ee species. ovi positing in cultivated fields. The reason for this lies in the fact that the parasitic adults can more easily work. their way into the soil in loosely cultivated fields'than they can in the firmly packed soil conditions of native prairie. The cutworm species in which parsH ism was in duced are all forms (with one exception) ovipositing in nat ive p r'a Lr Le and are thus more or less free from natural i. co ns ro I In cutworm common to , by,this parasite. species both cultivated and natural conditions, such as R. flavi collis, one would ,expect that the parasitism by Berecyntus would be much higher in the cultivated land and this is " exactly the condition which exists. With relation to Euxoa tessellata, parasitism was obtained in the laboratory, although it has never been act ually recorded from the field. However, due to the diffi culty in distinguishing tessellata, �ogaster and verti calis larvae, it is probable that parasitism does occur under field conditions. But in no case would the degree of parasitism be as high in tessellata as in t�e other two - 71 - '" species. This is due to the method in which the eggs of tessell�ta are deposited. The eggs, being laid in�sses and covered over by a fine outer coating, are somewhat pro- tected from· the adult parasites and it would only be in a few instances where parasit ism might occur. 'However. v he n the eggs of tessellata are removed from their· protective covering, they are readily parasitized. To obtain conclusive e�idence that the limiting fac- tor in parasitism of eggs by Berecyntus baKeri is the soil "', conditions, some experimental work. was carried on in the laboratory. Eggs of Euxoa ochrogaster, lhe chief.host of this parasite, were placed in two distinct soil conditions, (1) in soil, kept quite loose. (2) in soil firmly packed. In both cases, eggs were of the same age, and soil conditions, " '/ except for'the firmness,. were the same. The ·eggs were placed in the soil at the normal, oviposition depth (linch). �he percentage of parasitism of the eggs. in the loose soil was about 40 percent, while practically no parasitism was obtain ed in the firmly-pack.ed soil. These results indicate with out any doubt that the�in factor governing the degree of parasitism was the condition of the soil. The same experiment was carried on with respect to those species not recorded as being parasitized under nat- urll conditions. Eggs of E. verticalis, E. divergens and - - � dargo were used. In every case parasitism of the three species wer e obtained in loose soil, but practically none .I'�---- - 72 - under soil conditions similar to native prad r Le , Experimental work. W3.S unde rt ake n to ascertain if the adults of Berecyntus bak.eri would parasitize eggs deposited on vegetation. As has been already stated, the vast majority of the eggs parasitized are laid in the soil. Experimenting with Euxoa ochrog-aster, eggs of this species deposited on , . 1'1", green vegetation at a height of 4 to 6 inches above the soil surface were practically immune from parasitism. Only one t" egg·out of twenty-five placed there were a t t ache d , However with eggs of Feltia ducens and pQlia renigera, which are laid on or an it , '. normally vegetation rubbiSh, average paras . ism of was obtained. No will be made , ' .. twenty percent attempt to explain this situation at the present time. I , ,.,ol I OF , SuP�R- PARAS IT ISM EUXO.l OCHROGASTER : , .' , No indications of hyper-parasitism have been found. Super-parasitism occurs to quite a consid�rable extent, as must always occur where severai species of parasites attack the same host. As defined by Fiske "Super-parasitism is that form of symbiosis resulting when any individual host is attacked by two or more species of primary para'si� or by one species more than once." A better descript ion of the situation is obtained with respect to � ochrogaster when the term "mixed super-parasitism" is used .. The defini tion of super-pa.rasitism by Smith f its the present case more - closely "Super-parasit ism is that form of sy�biosiS Occur- - 73 - '" ring where there is a super abundance of parasites attack ing an individual host. No dissections have been made, the only observed cases involving the external Paniscus sp., (ichl'leum or1id) in combinat ion with either Meteorus vulgaris Cress. (Braconid) or Berecyntus bakeri How. On account of its life-history the fourth instar parasite, Sagaritis at�insoni Vier. is probably not involved in super-parasitism except with Berecyntus. Thus we find Berecyntus parasitizing .', . eggs of the red-backed cutworm primarily, the larvae develop- .' . , ing being secondarily attacked by other parasites. In t�e ' ';, case involving the external paraSite Paniscus and Berecyntus, d I. \ both parasites probably complete their life cycles, with the virility of both being leesened. The number of Berecyntus adults emerging in this combination is considerably below the average, indicating that there was not sufficient nour ishment to feed the normal number of larvae. In the case where Sagaritis atkinsoni is linked with Berecyntus in para- I .-' a Lt Lsm, the results of this combination is not known. From the life-history of the fourth instar paraSite, it is certain " , that the host will perish in the fourth stage and from this 'J'" it is evident that the Berecyntus parasitism would be totally i" lost. Superparasitism of Euxoa ochrogaster by hymenopterous ,I paraSites appears to fall short of the scheme, i.e. one para site lives, and·the other dies or else is greatly impaired; however, in any case the survivor brings about the destruc- tion of the host. In the field super�parasitism is distinctly I J - 74 - v , -J, more prevalent than rearing data would tend to show, owing to the fact that only the survivor is recorded, the other , "'II before noted. I parasite �sually dying being I ." One important problem connected with parasitism is • ',I, the question of competition between the parasites •. More than one species may attacK the same individual larva, and if the less efficient species survives, its attack serves to reduce the numbers of the more efficient species. For • J,) instance, J!.:. bak.eri cvi.po s i.t e in the egg of Euxoa ochro '. gaster and unless the egg stage of Berecyntus 'is' very long ,.:; it should have the advantage of an early start over the other host; but the duration �f the combfned egg and larval , , , " stages of Berecyntus is comparatively great and it is poss ible that a rapidly growing parasite, even though hatched later, might survive and destroy the Berecyntus larvae. Again, certain species of parasites attacK the later in stars of the larvae and may sometimes attack larvae which are already infested with a rapidly developing species of parasite, well-advanced 'in development. This may cause the death of the host within a few hours or days after these species oviposit, thus preventing their development. Cert- a i n species, such as Sagaritis can o vfpo e Lt in the larvae already parasitized by Berecyntus and complete their devel opment before the egg parasite can. Larvae infested �ith Sa�aritis are retarded in development and never become as - 75 - large as those infested with � baKeri. It therefore seerrs probable that this species is liKely to survive at the ex pense of Berecyntus if their eggs are deposited in larvae already infested with it. It see�possible that a large number of sp�cies of . ,. '. �. parasites attacking the same host may result in less, rather • 1,1 than more, effective control of the host. But to reach more definite conclusions concerning the conflict between the " " pa ras Lt e e of Eux�a ochrogaster, it will be necessar.v to carry on further experiments and to study the life history of the I I, various parasi tes in detail by dissecting the infested larvae • INTER-RELATIONSHIP OF DISEASE AND PARASITISM BY BERECYNTUS BAKERI. The data herewith presented is t ake n from the study of this phase of the red-backed cutworm work carried on by , ,I K. M. King and N. J. AtKinson. This data is taKen ent�rely from the records obtained at Sas katoon during the summer of 1925. During that summer a high degree of mortality amongst red-backed cutworm larvae was cause d by a bacterial disease. The following table taken in part from King and Atkinson's paper is of interest:- � .---- - 76 - ", . 110RTJ.LITY OF E. OCHROGj.STER FROM INSECT PJ..RASITISU AND 'l'DIS�1SEsr' OF LA-RVE & PUPAE DURING 1925. '" .AT S.ASL�TOON Berecyntus Mortality from Date of e o I'Le c t ion bake ri disease - �. 1Tay 15 16.6 16.6 :May 20 2.0 5.9 May 26 7.5 30.0 June 9 4.0 54.0 June 10 5.0 60.0 June 25 12.0 66.0 25 June , 6.0 80.0 The differential death rate did not affect all species of paraSites equally, depending largely on the I time of emergence of the paraSite from the host. It was t evident both from its rate of emergence in 1925 and from the increase in the rate of parasitism by it in 1926, 'j, that Berecyntus baKeri received an advantage from the prevalence of the cutworm disease. The reason for this is �t however, since does not "I" clear, Berecyntu� kill its host until the latter has made its pupal cell and indeed such host larvae have a longer average period of feeding than normal larvae� The only suggestion that can be made is that this observed advantage may be connected with the apparently increased metabolic activity of the host until just prior to its destruction by the parasite, a condition evi'denced by its size and voracity, larger usually by an a. ' .. {'o __ Pn..ra.s,t,."". __ Ol',ea..s<,. 5 / / / / 3 ( '/ � / "'0} ;).0 q j Ma._} Ju.ne. .Chart V: Mortal! of ty � ochrogaster from insect parasitiam and "diseases" of larvae and pupae at Saskatoon during 1925.(After Xing and Atkinson) I j _.... - 77 - THE G�llSR1� �FnCTS OF P.l..1USITISI,l ON" THE HOST AND ON THE P ...�ASIT�, W ITH SP�C IFIC REFE�nC� TO BERECY:L;TUS BAl�RI In spite of the fact th�t the life of a parasite is commonly regarded as an easy one, it is full of dangers I and leads eventally to extinction. Since the object of this form of life is the getting of food, a matter com�on 1 to all living things, it is after all only a specialized mode of living. If parasitism has been accomplished by 1 following up an advantage, then in mar-w respects it is an achievement and if it is a result of necessity it is no less so. " The term "dagene rat e as applied to the parasitic mode of living is hardly appropriate unless the term may be applied to the reduction of certain senses which be- 1 corne more or less to the such as certain useless organism, I I sense organs and organs of locomotion. This form of de- 'I generacy, so far as the paraSites at least are concerned, might better be deSignated as a form of "hopeless special- izat ion", according to Wheeler, since, as he puts it, it leads eventually to e�tinction. Wheeler remarts that "The individual paraSite buys its rare successes very dear- ly, for it must often run the gauntlet of great reSistance i -I and animosity on the part of a too healthy host, and must I • at the same time carefully avoid seriously injuring that I I host and thus' bringing about its own destruct ion - - - - ,.:'"'.P"""'"\ ---- - 78 - Parasitism so far as the race is concerned is anything '. ' but a promising or profitable business". But then we are reminded that parasites form but a small part of'the species of animals and plants which as Root has put it "Have perished from the earth without leaving descendants because their specialization for particular environmental conditions was so beautifully exact that i,t left them hopeless, with no choice but extinction, when those en- vironmental conditions were replaced by others. n 1 Parasites are adapted to their mode of life in two general respects, �Siological in that they require � a certain species or group of species as food reacting I' accordingly, and morphological in that their structure is modified to this end. A third adaptation might be termed "rhythmical" in that the life history 'of the host must be j somewhat suitable for successful parasitism. These adap tat ions become the more intense and exclus ive as t his mode, of life progresses, i.e. the further one goes bact in the history of a parasitic species, the more nearly liKe its free living relations it becomes. The interrelation, both as to behaviour and struc ture, between the parasite and the host becomes more per- r fect as the symbiosis grows more intimate. The true par asite and its host represent a type of machine with all its parts functioning co-ordinately, hence it is diffi- cult to discuss the behaviour of the former without also - 79 - dwelling on the behaviour of the la�r. The stage in this relat Lo nsh Lp when the former cannot exist w itho ut the latter is certainly reached in many instances, but one may indeed wonder whether the reverse condition is ever actually achieved through the agency of parasitism. If one were to tabulate the several qualifications which a species must possess in order to become a success ful paraSite, it ��uld soon become apparent that all suc cessful species, whether free-living or paraSitic, are endowed with these same fundamental qualities. Parasit- ism does not operate under laws of its own but under laws fundamental to the life of all animals. The larvae of Berecyntus bakeri, even before they begin to feed, exert a very marked effect upon its ho st, although not always very noticeable. These hundreds of paraSite larvae, while living internally within the host larva do not seem to cause it the slightest inconvenience and it is impossiue to distinguish between a parasitized and a healthy larva up to the sixth instar of the host larva. In this perfected state the relation between the host and paraSite is a marvellously balanced association. Timberlake (112) suggests that some paraSites rray be so t r similar to their hosts in effluvia or physical being that their presence is not felt or resented. It may be that parasi tea acquire this immunity wit hin their hosts. How- the ever, despite lack of outward signs of eff�cta, there , � .:._--- ... - 80 - must be occurring internal changes which are not apparent until the host larva is almost ready to pupate. There is one definite difference between l3rvae p�r�sitized and t ose not - in dissecting parasitized lar v�e there are pathological changes in the body tissues, in particular the flaky fat bodies are often converted into a number of comparatively gigantic globular cells which are sometimes unattached and float freely about in the blood of the insect. Parasitized larvae in the later instars may often appear to be sluggish and not as active as healthy larvae. Then suddenly its to»por increases and through the semi-transparent skin are seen hundreds of small white par- aa Lt Le larvae. In tViO days at the mo e t the host is dead. In some cases tLe host la�va has partly completed its pupal cell. It has also been noted in many cases that· parasitized la.r1.8a do not go beneath the surface to make its Pllpal cell as do normal larvae. In the case of prepupae, a sort of earthen cell is nude to »upat e in and normal pupae are near- ly always covered with soil. But parasitized larvae usually are not c ornp Le t e Ly covered over w Lt h soil, ·probably due to the great size of the larvae, which prevents it from entire ly remaining within the earthen cell made for a normal larva and probably due to the fact that the parasitic condition - 81 - stops the wor� of the larva before completion of the cell. The host larva at this stage of its life is merely an I empty sKin full of parasites. Tvvo other definite effects the host larvae have I upon been observed. The average size of fUll-grown larvae of E. ochroQ,'9.ster parasitized by fu. ba.keri is considerably l � above that of t hose not paras it ized. The enlargement in � , size involves not only the body but also the head and is o:ften associated VI ith an abnormal seventh larval instar which has been observed under no other condit ions. In this event the sixth instar tends to be subnormal in size. A great many measurements of head widths of parasitized lar V9.e were rrade and in ninety pe rcent of E. ochrogaste r lar vae, the abnorma� seventh instar was present. The average head width for t. his seventh instar was 3.12 mm, with the largest n�asured being 3.90 mm. (This larva also had an abnormally large body). �f.lhis is a. considerable' increase in size in comparison with the average size of full-grov;n normal larvae - 2.75 mm., viith the largest being 2.94 mm. The same condition was found in parasitized larvae of Felt ia ducens and Rhizagrot is flavicollis. normal larvae, of the last species have a seventh instar with an average J�---- .� , I - 81b - head �idth of 2.96 mm. while the parasitized larvae in the seventh instar average 3.12 mm. in width. There are two possible v,rays in which an increase in size of the host headwidth can take place. ( 1) By a g-radual increase for each instar. ( 2) By a decided increase at some stage of the host' 8 larval development. It has been found that· the latter method is the one which occurs. There is a minimum of increase in the first three instars( due to the fine adaptat ion of the paraSite to its host) with a decided increase in head width in the fourth or fifth instar, depend ing on which instar the para.site larva are freed into the general body c�vity of the host. The acceleration of development in the later sta:!es of the paras it ized hosts is of course, uro'ught about by the presence of the paraSitic larvae but �hether directly or indirectly has not been ascertained. It may be due to glandular secretion but nothing resembling glands have been found in the paz-as it e larvae. t r I .��,---- �- , t - 82 - The second definite effect of the larval parasites upon the host is that, at the time when the host is about t e r aa Lt Lc La rvae to pupa , the pa inhabiting the host· in g reat numbers, cause a ma r ke d inflation in the host larva by the ttmation of oval cells around e.arly pa ra s it Lc larva. It is from this fact that �iley called these po Iyembryo n i,c chalcids "The inflating chalcis-fly". The �nfl3.tion varies in amount, depending upon the number of p.3.rasite larvae present in the host. THS ::i��_-\'TIOJJSHIP OF FOOD 1.l:ITH REPRODUCT I'VE _\.C T IVITY AIID �O�JG�VITY. "Hymenopterous pa rae L tee oviposit, then die", is a brief generalization found in various forms, usually a. little amplified in literature relating to parasitism. This e t a t e me nt is very general. If taken not too liter- ally, it implies a period of maturity and oviposition re- latively short in which life is but little prolonged by t food and water. If we were to rea.son about this matter, it is only rational to aSSume that the cell contents of the bodies of mature egg:"laying hymenoptera should be su bject to just the same Laws 0 f exhaust ion and renewal as govern the cells of animals in general. This would mturallY imply a need of food and water in the life- � I , '"" ---- _-__.. late V: Diagramatic representation of three shapes assumed by larvae of Euxoa oChrogaster parasitized by Berecyntus bakeri. - 83 - activities of these creatures. In order to be most highly efficient a female parasitic hymenopteron should be able to survive through long periods while she is engaged in seeking a suitable host in which to oviposit. So it would seem that were the eggs to mature but slo�ly in the ovaries, not all at one time, and could the crea ture be sustained meanwhile by suitable food, she wo u'l d be a parasite more efficient than one whose death follow ed promptly after her first oviposition. It would also seem that in general, for high efficiency, the male might well be of relatively long life, just in order that each male might survive 'periods in which he failed to find the female of his species. Of course, such reasoning in ad vance of obtaining a large body of facts implies one thing - that food with these adults shall have its usual functions of restoring exhausted tiesues and of prolong ing the 'life of the organism. In answer to the above, with direct reference to the polyembryonic chalcid under study, the great numbers of the parasite is nature's method of getting around the im- perfections in the adult's reproductive mechanism. Due to the fact that its hosts are eggs laid in the soil, the chances of oviposition are decreased greatly for the fe male. To overcome this, nature provides for a greater num ber of adults so that the species will not become extinct. and Both males females of the parasite under natural con- - 84 - ditions do not live for long periods, hence nature does not make the sp ecies specifically attacking one host, but allows for a large number of hosts. With Berecyntus the ovaries are mature on emergence of the adult and oviposi is ' tion can take almost This beneficial :. place immediately. to the species in that in its short life there is no waste . , time for maturing of the eggs. Throughout this study of the biology of the para sitic adults, a mass of informat ion has 'been collected from various experiments along many lines� I wi�l merely summarize in short form some of the most interesting ob servations relative to food supply and longevity and re pro duct ion. 1. If unfed for more than 3 to 5 days at an average temperature of 70 degrees Fah., adults will not survive. 2. Females will feed on fluids exuding or sucked from oviposition punctures, even if they are well-fed on honey water. 3. Females fed honey-solution but given no chance to oviposit may live longer than females fed with honey water and given opportunities to oviposit. 4. Oviposition is plainly an automatic reaction to an olfactory stimulus. The antennae probably receive this stimulus, though the mouth-parts may have a part in its reception. - 85 - 5. The acts of oviposition and of subsequent feed- ing at punctures requires the expenditure of much muscular energy, for which expenditure ill-fed females are not pre pared. Unless. fairly well-fed and strong, they will make no serious attempts at oviposition, but are apt to starve in the immediate presence of a possible food supply v:hich they lack sufficient strength to obtain. 6. Females live longer than males, no matter whether fe d or no t , From the above observations, included with other • ,1. notes, various inferences can be made. 1. In nature the adults find a varying and irregu lar supply of f::>od which in its effect on sexual activity and longevity is equivalent to honey water, and which is of paramount importance. 2. That in the laboratory the life of the adults can be prolonged to some considerable degree by art ifi- cial feeding. ra of a ras Lt Lam '" For a 'working bas knowledge of pa , there is needed many exact stUdies of the effects of those various factors which work together in what we term cli- mate, and of the effects of various Kinds of food, on the duration of the period of reproduction in paraSitic hymen- optera. ,"---- 86 THE RELATION OF-T� ABmmAUCE OF THE PARASITE TO WE_\.THER CONDITIONS It is often considered that parasites are controlled by different and probably more intricate laws than are other organisms. However, this is not true of insects parasitic on other insects. As to causes and laws concerning abundance, little appears to be �nown, except in a general way. With sufficient numbers of individuals of the one or several host species, Shelford states that the abundance of any paraSite is controlled in the main by the' following:- 1. Survival during adverse periods or periods of inact ivity. 2. Rate and duration of development of new generation. 3. A.ctivities controlling fertilizat.ion, egg-laying, migration from host to host. 1. SURVIVAL There is always more or less mmtality during periods of inactivity. The abundance of individual paraSites presem after a period of hihernation, aestivation or cessation of activity is markedly influenced by weather and climate. The survival of the paraSite is of interest in relation to the survival of the host. In this regard, Shelford definitely states that there is no definite rule which can be laid down as to conditions suitable for the survival of paraSite and I j - 87 - host during rest periods. IVith Berecyntus bakeri the "rest " r d of hIbe r.na.t or pe Lo is the period i.ng , overwintering. This is accomplished in two ways, in the egg, as an egg, and in partly-grown larvae in early polyembryonic stages of development. Atkinson in his study of the red-backed cut worm state-s that there is no indication of any appreciable mortality of the �ggs under natural conditions. His evi dence is wholly qualitative but is convincing, in that t wo very serious outbreaks of the red_'back.ed cutworm have occur- red following weather which one would �xpect to be detri .mental to the eggs. It must ,be remembered that the parasite� egg is within the host egg and is subject to- the same envir ',. onmental factors. Apparently the natural enVironmental con ditions as well as not injuring the host do not cause any . the e To check this mortality to p aras Lt , up qual itative hyp- of E. ochro�aster known to be parasitized by othesis, eggs - Berecyntus were placed in cold storage, with temperatures o 0 0 . registering 32 F., 0 F. and -10 g. In none of these temp- eratures were the parasites or the host eggs seriously in- jured or impaired. Since the eggs of the host are laid within the first half inch of soil, unless protected by a covering of snow, they are subjected to all the Vicissitudes of the weather. But thermograph records at Saskatoon show t ha t the soil temperature will very seldom ever go lower o than 0 F. So it is seen that the abundance of the parasite is not correla.ted with winter temperature. i J - 88 - ',,------. T.1BLE - NU11B� OF P.lR...\.SITE ADULTS E1j3GIUG FROM EtrIOA OCHROGAST�R L..i.RVJE, THE EGGS FROM VJHICH THESE' LARVAE HATCHED P�ZVIOUSLY IUVIUG BEEN KEPT AT DIFFERElJT TEI.rPER\.TURES. Temperatures. o Egg No. 630F. (lab. Temp.) 28-320F. -10 F. , " 1 400 adults 530 790 -- =# 2 325 adults 800 670 500 .3 500 adults 720 320 470 4 650 adult s 750 420 430 5 650 adults 530 560 560 6 800 adults 870 320 710 7 500 ad ruts 620 400 480 8 400 adults 560 620 390 9 850 adults 470 550 -- # . " 10 700 adults 325 440 520 AVERAGE 577.5 617.5 534.0 (436) 506� , ", ,___------...._-----_....-----'-----1 # - No larvae emerged from these eggs; average is taken from basis of 8 larvae. In the above experiment 10 eggs of � ochrogaster I , known to be parasitized by a fertilized female were placed in the temperatures recorded for a period of 30 days. While some variation is s hown, it has been the cront inual experience of the writer that the number of adults emerging is an exceed- ingly variable quantity. As the temperature to which the .. / / //� " q / __ ,3·F I __ 3a-, M / ______06 I ------IO·_ / I I I ; / / ( ) f ) ,/ '" " / / " ['" I I I I I "so Chart of VI: Number baker! , adults . Berecyntus emerging trom Eux(i)a ochrogaster larvae ,the eggs trom which these larvae hatched previously having beeu kept at different temperatures. (see table I) - 89 - eggs are subjecte� is lowered, there is also a slight de crease in the number of adults emerging. It is not known if the reason for the two non-hatching eggs in the group o subjected to a temperature of -10 F. can be attributed to the temperature or not. The parasi te also overwinters in partly-grown larvae, such as Feltia ducene and Euxoa lristicula. The stage of the parasite's life cycle during the winter is dependent upon the stage of development attained by the host larva. With both the host species mentioned, the host larvae over winter usually in the 5th or 6th instar. At this stage the pa rae tt Lc larvae have been freed from the polyembryonic chains and are free-living larvae. It is known that the over�intering larvae of ducens and �risticula are not subject to serious mortality due to low temperatures. Parasitized larvae of Euxoa trist LeuLa in the 5th and 6th instars were 0 0 placed in cold storage at temperatures of 32 F., o F. and 0 -10 F. No wide differences were found for any of the temp- eratures. This effect was measured by the average number of adult parasites emerging !ventually from ten hosts, in comparison with another group of ten parasitized larvae of the same host epe c Ie s Kept in hboratory conditions. It is thus apparent that the winter temperatures have no direct influence in the parasites' abundance. � �\�---- - 90 - T.lBIE II - Nm.rn!R OF P_�RASITE ADULTS E1ISRGING FR01I E. l'RI8- -- TICUL.i. L1RVXE, THESE L�RVJ.E PREVIOUSLY H.\.Vnm BEEN RE_LlED TOGETHER UNDER LABORATORY CONDITIONS FaOM THE EGG TO THE FOURTH MOULT, AND THEN SUB JEC TED TO THE FO LLOVnUG TElJPERAT URES FOR .A. PER- IOD OF 30 DAYS. Temperatures Larva No. -lOoF. 1 220 160 200 170 2 250 120 130 160 3 200 200 170 140 4 120 240 150 200 5 175 170 210 130 6 200 210 140 175 7 200 190 170 195 8 250 220 185 140 9 120 220 235 170 10 190 180 2QO 180 AVERAGE - - - 204.5 196.0 179.0 166.0 In the above the larvae 'were reared from eggs of the . same moth and were known to be parasit ized by a fertilized female. They were reared in the laboratory at °630F. until the fifth instar. Then the larvae were all placed (except o I 10 Kept under laboratory conditions) at 32 F.; after 7 days, I --� -----���--� - � 1\.-� I -- 63°F. -- 30<>� ------0·." ...... - -10. Chart VII: Humber of adults of B. bakeri emerging tram Euxoa tristicula . larvae,these larvae previously having been rearea together under laboratory conditions trom the egg to the fourth moult and then subjected to different temperatures for a period of thirty days.(see table II) - , - 91 - 20 of these larvae were placed at OOF. for 24 hours, and then 10 of these larvae were subjected to a temperature o of -10 F. These precaut ions were tak.en in order to not plunge any group of larvae into a lower temperature sud denly, this sudden change being more severe for some groups than for others. MOreover, under natural conditions, there is usually a gradual change in temperature and no very abrupt drop. Although the results do not show a very pro nounced_difference between the two extremes of temperature (an average of 38 adults), there is again noticeable the fact that as the temperature is decreased, there is 1ik.e- wise a Similar decrease in the adult emergence. Of course under natural conditions, the low extreme of temperature o (-10 F.) would be very rarely reached, and the effects of the other temperatures is not so great but that it can be explained as caused by other factors. However, parasitized larva of Euxoa tristicula when subjected to temperatures of 320F., OOF. and -10°F. in the first', second and third instars for a period of 30 days � show a very serious effect caused by the low temperatures. and second instar larvae of E. tristicula when sub- First - . jected to anw of the above temperatures were always so ser iously affected that none ever survived. The same was true for third instar larvae when subjected to temperatures of OOF. and _lOoF. At 320F. the third instar larvae usually --- �------�-�------..------... - 92 - Survived but the parasite abundance was seriously affected. The following t able shows the result S sUIIlIll3.rized from this experiment. TABLE III NUMBER OF PA.RA.SITE ADULTS E1ffiRGIUG FROM E. TRIS- -- . , TIe ULA. I&.RVAE, WHEN THE FIRST, SEeO lID, . AIID THIRD INSTAR LlRVA3 ARE SUBJECTED TO THE FOLLOWING TEMPERATURES. •• 'I 1st instar larvae �nd instar larvae brd instar Larva No. to:- to:- larvae exposed eXEosed eX80sed - 32 F. OOF. -100F. 32 F. OOF. -100F. to: 32,0, -10] 1 0 0 0 0 0 0 20 0 0 2 0 0 0 0 0 o· 100 0 0 3 0 0 0 0 0 0 60 0 0 4 0 0 0 0 0 0 80 0 0 5 0 0 0 0 0 0 90 0 0 6 0 0 0 0 -0 0 110 0 0 7 O· 0 0 0 0 0 40 0 0 .8 0 0 0 0 0 0 100 0 0 9 0 0 0 .0 0 0 70 0 0 10 0 0 0 0 0 0 80 o· 0 AVERA.GE 80 From the above results, the only stage from which parasite adults were obtained was that of the 3rd instar sub jected to the warmest temperature used (32°F.). But even here the average emergence is almost decreased by 60 percent. - 93 - In this experiment all the precautions used in previous experiments were followed, such. as Knowing that the eggs were parasitized only by fertilized females. The only explanation of the decreased number of adults emerging from o 3rd instar larvae at 32 F. is that the polyembryonic chain or polyembryonal maaa in which the embryos are formed is very seriously influenced by low temperatures. This may be either a direct or indirect action. If this is true in nature also, any paras it ized larvae overwint ering in the earlier instars will be seriously influenced by natural con diti"ons, provided the temperature reaches at least OOF. (�his temperature is quite often found, especially where there is little snow covering.) But it is indeed rarely the case that any cutworm species overwinters in the early in stars, unless weather conditions of the previous fall cause such a condit ion to exist. Another i.n;teresting s Lt ua't Lo n is found in connection with the effect of alternate thawing and freezing on the host and on the parasite. Unless weather conditions are extremely favourable in the fall for egg-hatching, Euxoa ochroga.ster never hibernates in the larval stage. So occa sionally does this occur that it can be neglected. However, ducens and E. tristicula (the chief both F. - overwintering - alternate hosts of Berecyntus in the larval stages) do over- winter in the 5th or 6th instar. They are therefore subject .J_. _ ..._,. ------_.""""'"',------_...... - ... '\\ ---, - 94 - to any thawing and freezing in the spring. It has been our experience that unless this condition of alternate thawing and freezing occurs very often, tha� no serious mortality is experienced by normal larvae. It is also true that the greatest mortality will occur when the ex tremes of temperature are the greatest, that is, with con t inued thawing and repeated temperatures Lowe r than the freezing point, the mortality'would be the greatest. The effect of repeated thawing and freezing on the parasitic larvae was determined more or less eXperimentally. The following experiment was carried 0 ut : (1) 10 parasitized larvae and 10 normal healthy larvae of Emwa trist icnla (averaging a late fifth instar) after dev eloping to that stage under laboratory conditions, (62?F.) in a of 29 to suff were placed temperature about 320F., just icient to cause freezing. After 4 days in this temperature, they were brought into the laboratory and Kept as c ro se to 45-5001. as waS possible (the average waS closer to 500F). Then after 4 days in the laboratory, the group was trans ferred again to freezing conditions in the cold storage plant 4 �ere under (29 to 32oF.). Again, after days they placed the laboratory conditions (50oF.). No mortality as yet had occurred in the normal larvae. But after 4 days' exposure to to laboratory conditions, replacement .freezing conditions, out in the some and being thawed again laboratory. mortality in the normal was noted larvae. All of the larvae began ----_._-_. .- 95 - to show unwillingness to feed and three of the healthy lar vae eventually died. I cannot state definitely, however, that mo.rtality was caused by the repeated thawing and freez but ing certainly the temperature factor must have been one of the most important factors involved. The average number of adult paraSites obtained from the ten parasitized larvae was only slightly smaller than the average of the ten lar- the vae of Same species �ept under constant laboratory con ditions. The difference, however, was so small (about 50) and the variation in individual larvae from the average Was so small that nothing of importance can be deduced from the results. TABLE IV - NUMBER OF PAR.-\.SITE ADULTS EMERGING FROM E1OC0A TRISTICU�� LARV.\E SUBJECTED TO ALTERNATE FREEZ rso AND TIrlWING CONDITIONS. (320F. ANn 500F.). Unparasitized Parasitiz(;}d Parasitized larvae Kept LARVA. NO. larvae larvae entirely in Lab. as eh� cks , 1 O.K. 170 adults 220 2 dead 150 250 3 O.K. 170 200 4 O.K. 140 170 5 O.K. 140 175 6 dead 150 200 7 O.K. 140 200 8· dead 170 250 9 O.K. , 120 170 10 O.K. 160 190 �VERAGE 151.0 204.5 , ..... , - 97 - and freezing, that they did not feed sufficiently to nourish the parasitic larvae, which t�erefore found it necessary to feed prematurely on their host; an hypothesis following dir ectly from this is that some of the parasite larvae did not obtain sufficient nourishment to complete their development and thus' there was found.a Smaller number of adults emerging from the host. However, it must be remembered that although the ex perimental conditions used above are sometimes found in nature, they do not materialize often enough to seriously influence the abundance of Berecyntus. Even when they do occur it will not be so pronounced. - TA.BLE V NU"MBE3 OF J?A.R�SITE ...-tDULTS E11�RGING FROM EUXOA TRISTICUU LAl&AE SUBJECTED. TO GREAT .EXTREIvIES OF THAWING _iliD FREEZInG CONDITIons (OaF. AnD 500F.) Parasitized larvae k.ept Larva Unparasitized Parasitized ent ire1y in laboratory No. Larvae Larvae as check.s. 1 dead dead Z20 2 small size 80 250 3 dead dead 200 4 O.K. 120 170 5 dead dead 175 6 small Size 70 200 7 dead 90 BOO 8 never pup' a dead 250 9 small size dead 170 10 small Size dead 190 �ver. 90 (4) 204.5 __....----- " 63-F. 3a""IJ{So�F. 0° 1V11150·F. ) I ( ) , , I , , ( , , I I , I I I I , I , I I VIII: Nwnber Phart of adults of !. bakeri emerging from EUxoa tristicula larvae subjected to alternate thaWing and freezing conditions. (see tables IV and V) , - 98 - The only other period of the parasite's life cycle which can be intluenced by the climatic ronditions is the adult stage. The adults are present in natural conditions from May to September with the maximum period being in July. The climatic factors which might influence the long evity of the adults are temperature, mOisture and sunshine. Since the adults are not long-lived, anw factor influencing the adults is not of much importance in relation to its abundance. Of course, it is realized that the availability of suitable food is a very vital factor in the longevity of the adult. Therefore, any factor or combination of factors which affect the food supply is of importance. Usually the food supply of Berecyntus is affected primarily by a combin ation of precipitation and temperature with the moisture fadPr being of chief importance. Naturally the temperature will influence the vegetation indirectly. But insofar as the weather data of the past seven years are concerned, there is no positive correlation between moisture, tempera ture and parasite abundance. In 1923, the estimated � cyntus parasitism was 0-10%, and in the growing season of the year, 17.32 inche� of rain fell from April to September, with a very heavy period in July of 5.28 inches. Compared with this is the estimated percentage of 35% in 1929 �hen the total rainfall for the same period was only 2 inches greater than for the month of July in 1923 itself. If there was any correlation, it would be expected in 1929, since the ._...... _-- , 99 - drought would affect the food supply to some extent. But the p�rasitism remained l�rge, therefore it is quite app arent th�t these climatic factors never influence the food supply of Berecyntus sufficiently to disturb the parasite popUlation. The direct effect of the physical factors of mois ture and temperature on the adults is not so well known. There is no doubt but that temperature has practically no limit ing effect on the adult itself. The temperature never goes to )80 point low enough to cause mortal�ty and as far as is known about soil temperature, a maximum temperature is never reached that might have a lethal effect. Of course, it is quite possible that such factors might condition the habits and reactions of the adults, such as oviposition. There is one other possible way. in which moisture might affect the adults. Since the greater part of the life-his tory is spent in the soil the effects of a heavy precipita tion may be �f some importance. If the precipitation was of sufficient amount to thoroughly supersaturate the soil, it is within the limits of possibility that the adults might not survive. But I do not feel that such a condition exists often enough to seriously influence the parasite populat Lo n , The only other climatic factor to be taken into conside�ation is the sunshine. Parman (128) in his paper on the Scelonid, Phanurus Emersoni Girault, a tabanid egg -_.. --- . \� .... , - 100 - parasite, states that that parasite does its most effective work.' during seasons with a high percentage of sunshine. Observations have shown that the ad�lts of Berecyntus baKeri are somewhat negatively phototrophic; from this fact it is logical to deduce that the adults would spread over wide areas (by appearing above the surface) and thus cause great er parasitism in years of keast sunshine. Such a correla- tion between sunshine and parasitism has been found to be true to a greater or less degree. )::e such is true, the amount of sunshine in one season affect Lng dire ctly the adults, and hence oviposition, would influence the amount of parasiti� obtained the followin� year. A reference to the table showing the sunshine records at Saskatoon and the per centage of parasitism by Berecyntus during the period 1922 to 1929 will aid in this discussio�. In 1922 the amount of sunshine was the second highest recorded, while little parasitism was recorded the following season. In 1923, 1696 hours of sunshine waS registered (the highest in the 8 year pe riod) while the paras it ism was only 12%. In 19�4, the amount of sunshine was 1601 hours, the third highest amount, and the parasitism in 1924, waS prac tically the same as in the previous year. In 1935, 1582 hours of sunshine was registered, the fourth highest, and the following year the parasitism jumped to 20 per cent. In 1926, 1479 hours were registered, with a para�itism the following year of 23 per cent. In 1927, only 1525 hours of -,101 - sunshine was recorded (the period lowest in the 8 year per iod, and the parasitism the following season is the greatest recorded, 40 per cent. In 19�8, the smallest amount .o f sun shine was registered (1469 hours) and the parasitism in 1929 was about 35 per cent. Thus it is seen tAat a certain am ount of.correlation exists and th�ugh discrepandes are pre sent when correlating the amount of sunshine and parasitism, these are necessary as found in nature; for example, if there were a very strong positive correlation between sun shine and parasit ism; it would be expected that the 1525 hours in 1927 would be. correlated with less parasitism in 1928 than the 1469 hours in 1928 would be with the parasitism in 1929. But such a strong correlation does not exist; more- over, the difference b-etween the years 1927 and 1928 in am ount of sunshine is very small (56 hours, spread over 6 months) and the difference in parasitism in the following years is very little (40 per cent and 35 per cent). It is thus quite possible that even with the noted discrepancies in the correlation between parasitism and amount of sunshine the previous year, that such a correlation is present to some extent. When loo�ing at this correlation from the viewpoint of the amount of sunshine recorded. in July, the month of maximum adult abundance, it is found that it holds again to more or less the same extent. In 1922, 386 hours was re- corded (the highest in the period 1922-1928) and the _ "_,,_------, �-" '. -".":;;;c-.,,.. - 102 - , ! parasitism r.as the lowest. In 1927 and 1928, 320 and 311 hours respect ively were recorded (the lowest and third lowest amount ot sunshine in the eight year period), �hile the parasitism estimated in the tollowing years was 40 per cent. and 35 per cent� respectively (the highest and second highest percentages in the Same period ot years). In the intervening years, the correlation is present also. Thus it is men that sunshine is a tactor to be con- sidered in connection with parasitism by Berecyntus. While other factors may be interlin�ed with the facts of sunshine, there is ,undoubtedly a limiting intluence on the abundance o f the pa rae Lt e. SUNSHINE RECORDS. Year Total Amt. of sunshine Total ParaSitism registered in hours in Amount by the 6 ms , period April in July. Berecyntus to Se-ptember. 1922 1637 386 not known, 1923 1696 317 0 - 10 1924 1601 326 12 1925 1582 335 12 1926 1479 317 20 1927' 1525 320 ,23 1928 1469 311 40 1929 1524 383 35 PRECIPITATION RECORDS. Year Total amount in inches Total Parasitism in 6 mos. period - - Amount April to September. in July. 1923 17.32 5.28 0 - 10 1924 3.45 .35 12. 1925 13.03 1.23 12. __"'-"------"'''' �.._-----_".�------, // / / / / ( \ \ \ ) 1'1'1" I. Po.ro.siti./Ill. ---. Slinsbine-bn1Ds. -_'__ -. 'jU"5hin�.-J"I:J. , S"1IA Chart IX: Correlation of the mnount of sunshime with the percentage of parasitism of Berecyntus bakeri.(see table) - 103 - PRZC IP r"-'AT IOU RECO.aDS. Year Total amount in inches Total Paras it is m in 6 mos. period - - Amount _ltpril to September. in July. 1926 11.67 2.50 20. 1927 13.17 5.44 23. 1928 11.48 5.14 40. 1929 7.61 1.31 35. TE1I.PEIU.TUl{E . RECORDS. Year Average mean temperatures Mean temperature for the 6 mo s .. period for July. Parasitism April to September.-- 0 1983 54.6 F. 66. o -.10 1924 53.0 66. 12. 1925 55.1 64. 12. I 1926 53.7 66.8 20. 1927 53.9 63.8 23. 1928 53.4 64.4 40. 1929 53.0 65.5 35 • . RATE A.ND DUR.i.TION OF DEVELOP�UT OF NEW GENERATION. According to Shelford the number of paraSites sur- viving an adverse periodis the basis upon which all incre- ments are built. Abundance of parasites in a given season granting an average number of survivors from a preceding seaSon is governed by:- 1. Rate of development to maturity, or the length of time between generations. of life. 2. Duration reproductive 3. Rate of reproduction during reproductive life. the These rates and deviations are influenc�d by interest to weather. Here once more it is of especial - 104 - compare the numbers of parasites with the abundance of the host and the relative effect of weather conditions upon such a variable factor. In considering the effects of the v;eather factors, direct application of each factor alone is impossible. The chief rel'iance must be placed upon the reaction of the organisms in question under con ditions which may be regarded as standard. Temperature and moistur� must necessarily be conSidered together be- cause they vary continuously and reciprocally. Usually temperature is most important in controlling the ratio of development but air humidity and soil. moisture are al ways an accompanying factor. Organisms are not inured to constant conditions and their progress is different under constant and variable temperatures, usually being more rapid under variable ones. The effect of soil moisture and temperature on the development of the host larvae and the paraSite larvae was not ab l.e. to be well determined, due' to the lack. of However suitable equipment for controlling such factors. � some �neral observations have been made and are of inter- est. Atk.inson ('26) showed that the optimum soil tempera- o ture for larvae of � ochrogaster waS somewhere near 27 C. He also stated that the probable optimum soil moisture was between 12 and 27% of capacity. Between temperatures o of 50-75 F. he found that soil moisture between limits of 12 and 40 per cent of capacity had little effect on -----,--.�------.. - 105 - the rate of metabolism. At 55% of capacity he found a well-defined retarding effecf, due possibly to the in crease of carbon-dioxide in-the soil owing to the pre vention of much aeration and the more rapid decay of the organic material in the soil. Since the nte of development of the parasite larvae depends to a large extent on the rate of development of the host, the effect of temperature on the parasitic lar Vae will be Similar to that on the host larvae. But I do not feel that the moisture relation is of such similarity. Despite the fact that the host larvae can withstand great extremes of moisture and any extremes of dryness liKely to be found in nature, the parasitic larvae are somewhat in fluenced by extremes of moisture and dryness. The effect on moisture on the parasite larvae is measured, not as the rate of development but by the number of adults Eventually emerging. Of co ur a e, the observations are very general and it is also realized that other factors may enter. Most of the observations were made in the course of rearing parasitized Brvae in the laboratory. It was in variably found that parasitized larvae exposed to lo� ex tremes of moisture were more seriously affected than v.hen exposed to high extremes. When parasitized larvae were reared in tins, the so il of which v:as k.ept very dry, the host larva waS able to develop. but the number of parasite adults emerging was invariably smaller than the .normal -._------�------�..� -- . t - 106 - number. Particularly is this true when the early stages of the host larvae are exposed to extreme dryness. Since the early instars of the host larvae are more susceptible to adverse conditions than the later instars and since the early instars contain the most delicate stages of the paraSite's life history, the cause of the decrease in paraSite adults is probably the result of a combination of effects, both on the host and on the parasite. It is li�ely that the smaller water content of the larval body influences the paraSitic body extensively. Since the paraSites do not attack the host directly until late in its development (so as to maintain its perfect state of parasitism) it is probable that many of the earlier stages of the parasitic development perish. It has also been noted that parasitized larvae subjected to high extrem�s of moisture produce fewer adults than normally. The effect, however, is not so definite as in the case of extreme low moisture. No explanation is advanced for this mortality. III - COnTROLLInG ACTIVITIES There are other factors, probably parts of those already discussed, which directly influence many of the activities of the parasite. Although Berecyntus bakeri has been found on blades of wheat and on sunrf.owe r heads, it is undoubtedly certain that at that time no direct bright surishine was present. It is possible that such J __ , , 1\ - 107 - I, act ivit Le s of the ad ui vs wouJ1 tak.e p'l i ce on dull days, and therefore, it is seen that light is a controlling factor in egg-laying. In cool, cloudy �eather the adults remain iru.ctive while the activities of the adult of its chief ho�t, is also decreased. IV - OBSERVATIONS The final test of abundance and its causes is abundance itself as correlated with weather conditions. As this would mean continuous observation of parasites and hosts by means of week.ly or daily quantitative coll ections throughout many years, this has not been possible. But there is no doubt but that such data would enlighten most of the unSOlved problems of abundance. SEASONAL HISTORY. Before proceeding to discuss the seasonal history of Berecyntus bak.eri, it is v:orthwhile to again summar ize the relevant pOints concerning the biology of the adult parasite. (1) Adults have been observed in the field from the early part of May until the end of August the maximum abundance occurring during July. (2) Adult parasites emerge about three weeks to one month after they are first visible, thus being only 5 - 8 days later than the corresponding moths would have been. (3) The - 108 - adult have parasites practically no preoviposition period and can oviposit soon after emergence and cop ulat ion. In discussing the seasonal history of this "many " hoe t e d egg parasite, it is necessary to remember that discussion of any its life-cycle will naturally be hypo thet ical, si nce it is ent irely impossible to follow out its life cycle under natural conditions. This is due to the fact th�t the parasite attacks the eggs of many and as species, such, maKes possible a large number of Eowe ve r life-cycles. , from the information available concerning the biology of the ad ult par as Lt e and the re lative p�rasitism and oviposition habits of its various hosts, a f�ly accurate picture of its seasonal history can be obtained. The species passes the winter either w Lt h Ln the host egg, as in Euxoa ochrogaster, its chief host, or within a hibernating host larvae, such as Feltia ducens. I will first discuss the hypothet ical life-cycle where only Euxoa ochroQ.'9.ster is available as a host. Adults 0 f the paraSite emerge in maximum numbers a little later than the moths of Euxoa ochro�aster. But the moths of � ochro Kister. have a long pre-oviposition period (averaging 14 days), and thus about a we e k to ten days elapses before eggs of its chief host are available for pa£asitism. The fact thJ.t the peak of abundance in the fields of Berecyntus dIIJiIIiI"' ...= . .� _r� �..__...... - 109 - barl..eri is ecmewhat in a dvanc e of the max irnum Occurrence of the stage of ochro�aster which is suitable for· attack, results in a moderate loss of effectiveness of these parasites. This is so Since the majority of the adult parasites cannot and do not live even for a period much longer than two we e ks under natural condit ions. In any case, it is very improbable that the adult parasites emerging from red-backed cutworm larvae in July live long or in enough sufficient numbers to account for a heavy par�- sitism of eggs laid by 1h ochrogaster early in August. Hov.: ever, there is no doubt but that some pa.rasitism does occur in this way, chiefly by adults emerging from late-develop- ing r e d=b ac ke d cutworm larvae. The a no unt of p a rae Lt ism accounted for in this way wo ul.d depend to a great extent on waather factors but it is very improbable that it would ever be of economic importance. Thus it is seen that the gen ral rate of parasitism by Berecyntus of its most abundant host is influenced by an imperfect rhymtical adaptation to that host# and that if E. ochrogaster was the only host present, the parasite I/: Footncm - this hypothesis is expressed by King and Atkinson ('27) in their discussion of this parae Lt e , Mr. K. M. t:in� has also in persoral discussion ment ioned other out in the above outline. noints brouxht- ... J. _ ------,------�-- - 110 - would never become of any importance economically. From this fact it is very evident that there must be a certain amount of seasonal selection on account of the varying life histories of the different hosts. It is kno�n that the parasites from the seasonal extremes of one host, such as the late larvae of Feltia ducens, may attack the opp osite seasonal extreme of another species, such as the early eggs of Euxoa ochrogaster. It is also known that the rate of development of this parasite is invariably more rapid than that of most of its great number of host species, at least at Saskatoon. Therefore, in order to obtain the high degree of parasitism indicated from coll ections, it is essential that there must be seasonal sel ection on the part of the parasites. Under co nd it ions where ]h ochrogaster is out stand in�by the most abundant host, .there is a loss of effect iveness of the parasites, relative to th� parasitism of that host. The majority of the parasites must depend for survival upon find ing eggs of less abundant early hosts, such as Feltia ducene, Euxoa tristicula, and §idemia � astator, all of which overwinter as partly gr-own larvae. Under the same conditions, the rate of � ochrogaster is low, since the greater majority of its eggs are exposed to attack only by a small b roo d 0 f parasites, for t he most part made up of late individuals such as those emerging from late-developing hosts, e.g. weltia venerabilis or . .--...__..... _ ...... iIIIIC'iI__iiiiiiii ... � - III - �uxoa ochrogaster. This hypothesis is strengthened not only by field observations on the abund�nce of the para site adults but more strongly by the much higher rate of · ! parasitism by Berecyntus of larvae of Feltia ducene than those of Euxoa oohrogaster occurring in the same field. This co nd it ion has been definitely established in c o Ll,« ections taken in different years. Similarly, the relative rate of parasitism in different fields is partially correlated with the extent and variety of t:re cutworm infestation in each during the previous year. If there are present, besides the chief host, � ochrogaster, moderate infestations of such species as Feltia ducens, Feltia venerabilis, Euxoa tristicula and Rhiza�rotis flavicolliS, the chances of parasitism of eggs of the chief host will be increased directly as the abun dance of the other hosts (other factors·the same). This has been established· in the collections from our standard cutworm plot, - when the pa ras Lt Lem of fu ochrogaster is high, it is usually correlated with a high parasitism of other host species in the same field. In considering the seasonal histories of the para site in relation to the Lnd ividual host species, some interesting facts are brought out. If Berecyntus was a specific parasite and could only attack. one species, three of its host would never reach a high degree of paraSitism, J . .-=--- - 112 - due to t he long per iod intervening betws en the emergence of the paraa Lt e e and the egg laying period; tv�'o of its hosts would always receive moderate parasitism, �hile two would be in a very advantageous situation. Euxoa ochrogaster, Sidemia devastator and �agrotis flavicollis all ovipoSit from three weets to one month after the emer gence of the paraSites from similar cutworm hosts; it is well known that neither S. devastator or R. flavicollis are - - parasitized heavi�y, due to this reason. F. ducens probably ---- and F. venerabilis are always to a moderate ex- -, parasitized tent, since only two weeks intervenes, thus making it poss- ible for paraSites from parasitized host larvae to attack eggs of the same species later. Euxoa tristicula and 9..!. thanatologia oviposit at prac'tically the Same time as parasites are emerging from the host larvae. From the above facts, it is seen that the red-backed cutworm would never attain a high paras it ism in this manne-r While in those where it could be d uc e na species achieved (tristicula, , venerabilis) the factors of limited abundance and the pres ence of other paraSites would eventuallY decrease the species to such an extent that the paraSitism would also be lessened. In the following discussion on the hYpothetical life-cycles of the paraSite, only the mOre important hosts are included, these being ochrogaster, pucens, tristicula, , - dIIlJt#"" .. _ l� - 113 - devastator, venerabilis, thanatologia and flavicollis. (see chart at end of section) Eggs of � ochro�a8ter laid late in August, are more liKely to be parasitized by adults emerging from F. venerabilis and C. thanatol o�ia than from other sources. Some will occur - parasitism from late-emerging adults of flavicollis, ducene and � �ogaster •. It is very unlikely that parasites emerging from tristicula and devastator will parasitize ochrogaster eggs. As has been previously stated, the parasitism of ochro�aster is influenced by the numbers of· larvae of other host species present; now it is seen that the chief species influencing parasitism would be venerabilis and thanatolog:ia. But since it is also Known that neither of , . those specie� have ever been very abundant, at least in such numbers as to di rectly influence the pa.rasitism of ochrogaster, the writer is convinced that the species dir- ectly influencing ochrogaster parasitism are the three tristicula devastator. Larvae of duc- species ducene, and -- � are usually numerous enough and vary enough in length of seasonal history as to influence the parasitism of ochrogaster to a moderate extent. Tristicula, when abund- �nt, certainly influences the parasitism conSiderably, but this species is only abundant in occasional years. The larvae of devastator have as yet never been very abundant. Eggs of � ducene, laid late in July and early in August, are open to attack from paraSites of all the other J - 114 - host species and al so from pa ras Lt ea emerging from ducens hosts. This is probably the reaSon for the higher paras it ism of this spec ies in comparison with that of other species in the same field; Eggs of tristicul� are open to attack by p�rasites from all species except venerabilis, and it is known that this species, when abundant, is very highly p�rasitized. But since it is also a hibernating species' for many other parasites, it is so kept in check that very seldom does it ever appear abundant in consecu t ive years. Eggs of devastator and thanatologia, like the pre ceding species, are laid at such a time that they can be attacked by paraSites from every host species. The eggs of venerabilis and flavicollis v:hich are laid during the last two weeks of August have the same chances as ochro gaster in the probability of being parasitized. As neither of these species is very abundant, it would therefore acc ount for the small degree of parasitism exhibited by those species. In summarizing, it �ould appear to be a safe state ment to state that the rate of parasitism of � ochrogaster by Berecyntus bakeri is influenced by the total parasitism of other host species by t'his parasite. The species attri buting the most effective influence are Feltia ducens, Euxoa tristicula and Sidemia devastator, which themselves �------J..-...--,. -r� � �__ - 11:: - there 1s J. -r:er./ u,,!I.ll,liu rt e d ue tr o .. cl, of c a ue e and e ff e c t fJ.ct)l"� pr e ce ...rt in the p.i rae Lt Le n of the red-bJ.(;t;,.ed cut- _."",. - � -- r SEASONAL HISTOBIE3 L1 WINTER SPECIES PERIOD MAY JUNE JULy AUGUST SEPTl!UBER I otAdults B. in egg tew moderate maximum slackening ott tew Bakeri. or larvae abundance Euxoa parasites eggs laid last , ochrogaster egg emerge early two weeks in in July August. Feltia parasites egg@ laid last wk. in & 1st 10 ens duo larvae July emergJutarlyin l' days in August. Euxoa larvae parasites e- eggs laid tr1sticula late in in July. merg,une m a e n eggs laid tirst Sidemia larvae Rarasitt8rge rn devastator June, early in 3 wks. in Aug. July. parasites eggs laid late s larvae and ChOriZa�t1thanato ogla emerge in July early July. August. paraSites e- Feltia late eggs laid last venerabil1s egg mer�ein & 2 wks. in Aug. earlyuIIn Aug. e- Rhizagl'Otis egg parasite eggs laid late tlavicollis merge in ear- in August. 11' July. --) , - 116 - P�RCE!l.L\.GE OF P_\.RASITISM J.ND :SCO:mMIC :c.:I'ORTANCE The data presented herewith has been accumulated by the officers of the Dominion Entomological Laboratory dur ing the period 1923 to 1927. While the detailS relate in the na.Ln to conditions at SasKatoon, the principles brougbli out apply throughout the province of Saskatchewan. That part of the data dealing with the period from 1923 to 1926 inclusive has been worked over and published by King and �tkinson ('27). Since that time the writer has been in active charge of the cutworm studies of the laboratory and many- of the collections have been ma de by him. Other off icers of the laboratory have contributed much to the cut- worm studies, in making collections of larvae in various localities, in laboratory rearings and, generally adding mi�cellaneous information. The information given herewith concerning the years 1927 to 1929 inclusive has been worked over by the writer, VJith frequent collaborations with Mr. K. M. King. Rather frequent outbreaKs of Euxoa ochrogaster of greater or less intensity have occurred in this province at intervals of from four to six years. The most recent reached its climax in 1925. At Sas�toon the species was relatively scarce in 1923 though probably more abundant than in the previous year. In 1924 it �as nearly twice was minimized by as plentiful as in 1923. The damage also the exceedingly dry summer. As a result of the latter, J___ - 117 - '_', combined with exceptionally tayourable conditions in the fall, the hrvae were abo ut ten times as numerous in the spring of 1925 as in the previous year. A very high mort ality in June of 1925, in combination �ith unfavourable conditions in that fall, reduced the numbers of the speci� in 1926 to about one-fifth of those of the previous seaSon. In 1927 the species was in about the same numbers as in 1923. In 1928 the species was again on its upward trend and larvae were becoming more numerous. A slight outbreaK was torecast for 1929 but weather factors of 1929 were unfavourable for such a condition. Nevertheless the species was more trouble- some than since 1926. Everything pOints to a moderate out break in the spring of 1930, the severity depending upon weather conditions. The to11owing remarKS as to parasitism by Berecyntus for the past seven year period is concerned chiefly with the red-back:ed cutworm and allies. During thE} summer of 192;3 not a single larva of any species was found parasitized by Berecyhtus. In 19�4, the parasitism for the SaSkatoon district rose to 10 per cent, ranging from zero at Plunkett to 16.7 per cent in one collection at Saskatoon. In 1925 it was reared from eleven out of fifteen collect ions �f lar vae represent ing t he province and waa pro bably in the others, the hosts being destroyed by disease before the parasites could emerge. Indications were, on the basis ot the number of eggs parasitised, that there waS no marKed increase or 1924 to 1925. decreaze in th� rate of parasitism from J_ l' - 118 - In 1926, the rate of parasitism by this Species increased considerably, the estimated parasitism being 20%. This was an increase of 8 per cent over the previous year. This percentage was obtained from collections taken chiefly at Saskatoon and the results were obscured by disease. In these collections the parasitism ranged from 6.0 per cent in one collection to 29.0 per cent in the last collection of the season taken on June 24th. One collection was taken at Scotsgu�rd, in southern SaSkatchewan on June 21" which gave approximately 40 per cent parasitism by � bakeri. In 1927, the amount of parasit ism by this species was about the same, probably a trifle higher (23%). Only a few coll- ections of � ochrogaster larvae were�made, this species of cutworm being relatively scarce and while the parasitism in these waS somewhat higher, the increase cannot be entirely accepted, because of the small numbers of larvae on which the percentage is based. Three collections that season showed percentages of about 20.0, 26.0 and 18.0. However, the percentages of parasitism by Berecyntus in the above collections waS somewhat obscured by disease in laboratory rearing. In 1928, 13 collections of � ochrogaster larvae from widely distributed pOints indicated clearly that para sitism by � bakeri waS higher than ever before. In the collections taken from the standard cutworm plot in the / Field Husbandry EXperimental Farm, 5 collections spread - 119 - over a. period from IJay 21st to June 28th, showed the follow ing percentages 26.0, 20.0, 36.0, 36.0, and 47.0. Of course, it is realized that these percentages are high, due to the local conditions existing in the standard location. This, will be discussed later on. Two collections taken at Lethbridge, Alberta, showed a parasitism of 19.0/0 and 29.0�. One at Drumheller showed high parasitism, 39.0� whereas two collections at Stirling, Alberta, displayed exceedingly small percentages, 4.0% and 3.0%. All in all, the estimated percentage of parasitism by Berecyntus in 1928, is very conservat ive when pla.ced at 40}&. But this is almost double the previous high mark established the year before. In 1929, larvae of � ochrogaster were more plenti- ful in various local it ies than for some years but unfo rtun ately not many collections were made. In those made a high mortality in laboratory rearing made it very difficult to establish an accurate pazae Lt ism by Be re cynt ue , Three coll ect ions taken in the mual standard location gave ·percentages of 26.0, 14.0 and 33.0. One collection tak.en at Clares:tolme, Alberta, showed no parasitism from Berecyntus but any probable paras it ism would have been overshadowed by the heavy mortality from disease, as in this collection 93 out of 102 larvae died from disease� However, it seems pOssible that parasitism by Berecyntus was probably about the same as in 1928, but in any event not more and likely a little 10wer.(35%) t' .. .-JIIi' _,.. •. � "'-",. _ - 120 - In the above estimated percentage� of parasitism of �. ochrogaster, by � baKeri at SasKatoon, the majority of the collections were taKen in one heavily cutworm infested plot (104) in the Field Husbandry Experimental Farm. The resultant estimates may therefore be higher in that this standard plot has been infested with cutworms for several years and that there was usually a mixed infestation, in- , eluding both late and early species, making condition more favourable for parasitism. All of the above discussion relates only to parasit ism of the red-baCKed cutworm. Although there is not as much data available other some interesti�- concerning hosts, . points are mentionable. In 19�4, Feltia ducene was parasi tized 25� in cultivated fields and only 8�� in native prairie. - In 1927, the larvae of � dueens were more plentiful in the n "standard :plot than in previous years but we re not not ice ably abund�nt in prairie pastures. The parasitism was very high, 35% being recorded. This high rate of parasitism, as compared with that of E. ochrogaster taken at the same t i.me in the same plot (about 20;G) is Similar to the results men- tioned for 19�4. ;7hi1e the eggs of � ducens have an incub ation period of only about 10 to 14 days, the time of ovi- position corresponds more nearly with the time of emergence of adults of � bakeri from larvae of � oehrogaster and this seems the probable explanation� the observed differ- ences. In 1928, !!. dueens was t ake n in few numbers in the _j __ - 121 - " "st and ar d plot, five collections, including this cutworm species giving a parasitism of 8.J, 11.1, 0, 0, and 8j.3�. This v.-ould appear to be considerably Lowe r than that record- ed for 0 chro g"j,st er ill t he same p l.o t • .\. collect ion of F.' � - ducens on Uay 10th in the e t and az-d plot revealed a parasit ism of 17.4 v.h Ll.e d. collection taken on lSay 23 on pru Lr i e pasture �hov.-ed no parasitism whatsoever. In 1929, � ducens larvae in the standard plot �ere not so highly paIasitized as were the E. ochrogaster �arvae, 3 collections showing parasitism of 10.0%, 8.3% and 40.0%. No parasitism data of value concerning Rhizagrotis flavicollis Vt3.S oi:iained until 1927. Up until 1926, this species ��s taken in few numbers from cultivated fields. In 1926, it v.aS moderately plentiful in the standard plot. in 1927, the larvae �ere from 2 to 3 times as plentiful in cultivated fields as in the previous year. The larval pop ulation in 1928 and 1929 were somewhat similar to 1927 with the moth catch in the light trap indicating a comparatively smaller number in 1928 and about the same numbers in 1929 as for 1927. In 19�7, & flavicollis was parasitized in cultivated fields by � Bakeri approximately 33%, while native conditions revealed a 6% parasitism. In 1928 all the collections in the standard plot showed � flavicollis moderately heavily parasitized four collections giving per centages of 12.0, 33.0, 50.0 and 0. One collection else �here in this area gav� 21.0 per cent. A collection at l - 1.22 - DrUL1Leller +ac J.l�J hea.vil:r hit, 16.6'; of t he R. fL1.\"'icolli� Li rv.re be Luz ]1J.rJ.�itized by Berecyn���. -\. e o L'Le c t i o n on the p ri Lr Le �l:.ov.ed the cp i c Le s to ha ve a p.i rac Lt Le m of 2[i.G'�: vh i Le one t9.�:6.1 ill d. :;;J.r�eH J.t PJ.ldbtonc �l:"ov'ed onl�- 8.1'�. In 1��9, three collections in the e t anda rd plot ehov e d & �icollis to be p.:l.rJ.8itized apfJro{in:3.iely 4.fl;�, ;;,0.0.': .;;l.w] frOI11 r e co ru e be Lo ng Lng to the Dominion El1tomologic.:1.1 L::ib- b a ke r l tJ.1.ml.)..ted is a c t ua L re co ru e d t:.J�l1tUC the pa.r3.sitisli , v h Ll e the pe r c e nt age given in the br ac ke t e is the e e t Luat e d Lnc e Lt Lem i'1to the pt ed p a.rae , Hut is, tak.il1g consiuerJ.tioll fi c t t h it the Li rvae dying of d Lae ac e ve re pa r'ae Lt Lz e d pro- po r t i.a.iat e Ly t , the ac t ua'l percentage. p.L::n�n'I$=1 nr 192[) OJ �lrx:O_1. OCRROG�\.ST�R ]To. of IJ.rv3.e Uo • of B. Dise.3..se r3.rasi- b3.L.eri t ism. ---- !Ja.y 27, Ha.nley 29 0 15 0 r.1cl.y 29, _i.be rnethy 20 3 3 15.6(17.7) uy 30, _tbernet by 30 0 18 0 r,uJT 30, Bt rase burg 40 1 31 2.4(11.1) ,:;> :> June 6, CUdViO rth 47 5 ...... 10.7(20.0) June 18 ,l:eot a 40 9 17 2�.5(39.1) t n in ...�verage par].s it ism for 5 collect ions at �3..sb.toon ake - Unive 1'C it y farm fields 6 • 7)/liJ - 123 - P A:::Li.S IT IS1,1 0 F � OCHROGASTE!t rn PLOT 104FHD (1926) No. o t larvae Ho. of B. Disease Parasitism ba ke r L, Uay Z� 50 8 14 16.0(2�.2) I�y 31 50 7 5 14.0{15.5) Jun. 8 50 8 12 16.0{21.0) Jun. 16 50 3 10 6.0(7.5) Jun. 16 50 7 30 14.0(35.0) Jun. 24 55 16 17 29.0(42.1) PA.:;U.SITISM OF � OCHROGASTE� In PLOT 104 - FED {1927} No. o f larvae no. of B. Disease Paras it ism bakeri. June 1 50 10 16 GO.0(29.4) June 9 I 50 11 13 26.0(30.0) I. June 18 50 9 12 ,I 18.0{2..5.6) PARAS IT ISM OF' L..tRV.-lE IN PIDT 104 - FED (19Z8) E. 0 chro g'ast er No. of No. 0 f B·., Disease Parasitism F. R. larvae bak.eri7"" d uCens Flav. Gl 50 ID .2 M3.y ID 6 • 0 ( .65 .1 J 8.�(20) 0 Ni3.Y 30 55 11 6 20.0(22.4) 11.1(12.5) 12 June 9 49 18 11 36.0(48.3) 0 33 Jun.19 49 18 14 36.0(51.4) 0 50 Jun.�8 68 32 14 47.0(59.2) 83.3(83.3} '0 PA1U.SIT ISM OF LARV.ill IN FHD PLOTS, EXCEPT 104 .( 1928) E. o c hro zas t e z- Plot no. of No. of B. Disease Percentage F. R. larvae bakeri d"(l9n�a.v • 1J.aY 10 105 17.4{17.7) June 1 1406 11 5 2 45.5(55.5) 0 0 June 2 106 41 16 8 39.0(48.5) O. 0 June 4 206 71 20 I 12 Z8.l( 3.3.9 ),50( 66.6) 21C 25) . J 1.11 [I ' r __------�------�-�,���.��-�--� P�L.1:t_\.SITISM IN OTHE� CO:;:'IECT IOUS - 1928. E. o chro gaster F. ducens f1av. - &. No. of larvae May �3, Sas k.a t a on, native prairie 0 25.6( 32.2) Xw13.y 31, Lethbridge, Alta. 53 10 10 18.8(2�.3)0 0 June 1, Stirling, Alta. 51 2 15 4.0(5.6 0 0 June 4, Drumheller, ..tlta. 56 22 16 39.2 ( 55) 0 16.6(25.0) June 11 Lethbridge, Alta. 55 16 8 29.1 ( 34 ) 0 0 June 14 Stirling, Alta. 67 2 27 3.0(5) 0 0 June 16 Naidstone, Sas k.. o 8.1 PJ.RASITIS�11 OF :SJ.RVJ.E IN PlOT 104 FED (1929) lJo. of No. of B. Disease Paras it ism F. R. ducaris f1av. �E� -bak.eri,.!.. -- May 16 10(16.6) 4.5( 6.6) May 29 54 14 12 26(43.7) 8.3(11.1 30. ( 33.3) Jun.14 57 19 21 33(52.8) 40.(66.6 8.3( 14.3) ESTI1t-\.TED P�RASITISll OF EUXO� OCHROGAST�R ..!.T SASKATOON BY BERECYl�US BAKERI. . 1924 1925 1927 1929 � - - � - � 0-10 12 12 20 23 40 35 r From the data available on paraSitism it is quite eVident that Berecyntus bakeri must be considered as one of the major parasitic species attaCKing the red-backed cut worm and allies. It has been very noticeable that this para site has been increasing very rapidly in numbers since the 1923 when spring of the first data concerning it wae obtained. This increase has been especially marKed in areas of heavy J .�=---=--== ___ I -- "a.b - - 1�3" ' -- ,9iHI ••.••••••• I'I:J.. '1 " ,. ,------" / / 10 JUIJ't. Chart X:Est�ated parasitism of � oChrogaster at Saskatoon by Berecyntus bakeri.(see tables} - 125 - mixed infestation. The data has also indicated that the degree of parasitism is greater in cultivated fields than in native prairie. There is not sufficient available data to justify arw remarks concerning the relative amountof para� Sitism in other cutworm species. It is well known that Feltia ducens, Rhizagrotis flavicollis, and Euxoa tristicula (as well as E. ochrogaster) are the three most important species in which Berecyntus is found. The comparative percentage of parasitism of these four species varies considerably each year, depending to a great extent on the abundance of the species. There is each year some mortality of the cutworm spec ies parasitized by Berecyntus whether insect parasites are present or not. This mortality varies from year to year, de pending upon the conditions which influence the contributory factors, but the average percentage of mortality (barring insect parasites) for any period of years is the same as for any other similar period of years, if the period includes a. sufficient number of years to make the average a fair one. This average mortality is not sufficient to prevent an increase in the species (other conditions the Same each year) nor is the parasitism by � bakeri great enough to prevent an increase of the exact of any species • Although percentages parasitism for various species by this parasite cannot be stated, it is the evident that it has a very important plan as a part of • l' - 126 - sequence of parasites which in conjunction with the other natural agencies retards the increase of the different cut worm species. The total parasitism by all species attacking Euxoa ochrogaste r in any one year is never higher than 80'% and even with this high rate, the direct effect of the para sitism in decreasing the cutworm population is not noticed, unless other factors such as disease and the weather enter. It is thus very apparent that Berecyntus, although one of the most important parasites, really has only a small part in limiting cutworm abundance. There are a number of reasons why this paraSite will probably never increase its rate of parasitism to the state where it might have a very decided effect upon cutworm a.bun dance. With the great abundance of paraSites achieved by polyembryony from one egg, it would be possine, if no other limiting factors were concerned, for it to reach the stage where total parasitism of its host took place or to the point where the effects would be the same •. But if such a point of were to be reached, it would mean the eventual extinction the paraSite itself. So probably it is o�ly a natural con .ne ve r and sequence that such a possibility shOuld occur there has been evolved a number of factors limiting the de gree of parasitism by this parasite. AS has been seen from the percentage data, the high- est parasitism is obtained in those fields where there is a J_ • l' - 127 - mixed infestation. The greater the abund�nce of the alt ernate cutworm hosts, the greater is the possibility of par aSitism by Berecyntus. But uSually infestations are not as mixed as was the case in the standard plot v:here the parasitism is so high. Usually two or three species might be present but would probably not include a great enough range to help Berecyntus; as �as pointed out before, Euxoa tristicula, Feltia dueens, and Sidemia devastator are the three chief overwintering alternate hosts of the parasite. � dueens and � devastator are never abundant enough to increase the rate of parasitism in EQxoa oehrogaster. Occasionally EQ� tristicula is very abundant, and at such times the parasitism of Euxoa ochro�aster increases mater- ially. But the total pa ras Lt Lam of tristicula is invar- iably so heavy that it reduces the abundance of this species very rapidiy. So it is seen that paras it ism will never be abetted to any degree by means of mixed infestations. It has also been learned that parasitism is greater in newly deposited eggs, although oviposition does occur in mature eg?s to some extent. It is easily understood th�t �he chances of the paraSite finding newly deposited is limited. Added to this e��s_c of its various hosts very is the natural difficulties of the adult in finding'eg?s in the aoi.L, _·Hl in all, these two f actors limit the amount of parasitism by the parasite to a ma.rk.ed degree. The eff- of on the B. bakeri also is ect diseases parasitism by --- l' - 128 - detrimental to the achievement of a high rate. In years out of severe disease, the effects of parasitism is wiped hosts. considerably by heavy larval mortality of the various Another interest ing point in connection with the relative-amount of parasitism in the different host species Studies of is the manner in which the host eggs are laid: the percentage of parasitism of egg clusters of various sized laid in soil have shown that the percentage of para- sitism increases as the size of the cluster increases. From of the laboratory studies it was found that most of the eggs red-backed cutworm are laid in masses of from 50 to 600 eggs. This is particularly so of the first fe� nights' oviposition. often fa ur..d. A few solitary eggs or smaller groups are also and the extent to The eggs are cemented together slightly varies which they are attached to one another considerably. understood more vo ul.d be It c a n be easily why parasitism soil. found in large clusters of eggs laid in the First, there is great difficulty in locating such a small object and no doubt a cluster is eas- as an egg in the soil large n the ier to lcx:ate than a Single e;;g. Again, whe parasite available for ovi- has located the cluster, more eggs are there is again the position, v.hile with one 'eg?;, presented in this ochro- d Lf f LcuI ty of finding another egg. So viay, � in clusters. gaster favours p':l.rasitisrn by ovipositing their tristicula and R. flavicolli� invariably lay eggs core than two or three. singly or in clusters of not l' - 129 - Q..:, thanatoloQ.'ia and b au....>r:iliaris are more inclined to lay their e�gs in clusters, although many of them are laid singly. c. thanatologia usually has'clusters not larger than ten or twelve eggs. Dhen laid on vegetation the eg�s of � ducens and � venerabilis may be grouped together but, they can hardly be called clusters as they do not touch each other. In the soil these eggs are almost al�ays laid singly, In the cae e of Euxoa tessellata, the e:;gs are lald in clusters but there is a protective covering over them which -prohibits extensive p:.1rasitism. Viewed from these f�cts. p�rasitism by this egg-parasite is favoured by the' following species laying eg�s in clusters - E. ochrogaster, Fel!J:2: ducens and Feltia ve ne ra b lLi.e , By o v Lpo s Lt Lng sing';' R. fl9.vicol1is and C. thanatologia make E. - tristicula, - ly - therefore parasitism more difficult for the adults. It is are a no the not a coincident that ochrogaster and duce� ng species mOdt heav�ly parasitized. Due to the fact that there is very little dispersion of the adult parasite by such factors as wind, the spread the of the parasite �ould be very gradual and slow. �hile effects in such localities where the parasite ras establish chance of out ed might be noticeable, there would be little seen that this Side infestations being affected. It is thus of "Lo caL" im r-ae t e i7hile a ma r one, is also only pa I , jo portance. l' - - 130 One oo Lrrt in reS'...I.rd to the immediate effect o f the para site all the host is interesting: from the e co no mt c v Levpo Lnt , It ha e been found t h at La r va e paras it Lz e d Berecyntus illvar i9..bly eat mOTe fOJd than healthy larvae. This �ould then be the cause of heavier crop damage in those areas where the a bund aric e of BerecYl1tus "It-,-as g re at ee t . There vo u.l d be b s Lanc e d a.ga Lns t one anothe:c the reduction of larval abund- ance by the parasite and the increased larval feeding c�used by the pdrasite. In uhich direction the scales would s�ili� in time, I wo ul d not ve nt ur e a guess. Except for the remarKable number of host species in- valved, there is nothing, a f t e r all, so extremely unus ua L La the extensive c ae e of paras it iSll of vrhich we· have just giver the details. �herever a plant-feeding spEcies from some e aus e or f+,om s orae combination of causes transcends its normal abundance to any gr e at extent, there is alrays a gre�t multiplication of its natural enemies and this mul- tiplication is usually so great as to reduce the species even be Low its normal. Exceptions to the rule are seen v.ith specially protected species �hich, through the possession of some distasteful or·�epugn3.nt luality, ha.ve no predatory ve r disease or parasitic enemies. Even in such cases, hov.e , steps in and f ills the v:ant. The chinch bug is 9.. famil iar example of this class of injurious insects; it poesEsses l - 1.)1 - no p rrae Lt e e but r.hen it Lnc re aee e bey and t he bounds of l:tJ.t may be called nature's Law, for want of :3. better term, bac t e r Lal a nd fungous d Lae ae e a speetlily c8.rr;/ it off. '.7ith a.ll very injurious lepidopterous larvae, hov- ever, we co ne t arrt Ly see a g re at f Luc t u.rt ion in numbers, their parasites rapidly Lnc r e ae Lng immediJ.tely a.fter the Inc r e as e of the host species, o ve rt akm= it nume r Lca.Tly and reducing it to the bottom of another ascending period of development. POL Y E M B R YON Y J�, l' - - 13Z GE:S1U: conSID��J.Tlorrs on POLYElif3RYOITY PolyembrJonic development consists of the dissocia tion of the egg, after fertilization and before the first indication of embryonic layers, into numbers more or less large. 1hrchal has given this type of reproduction the name of germinogony Jr specific polyembryony. The Lne t.mcee of polyembryonic development in the animal �ingdom are very rare. The only one which appears to present an analogy with that found in the parasitic hymen optera, i� found among the Cyclostomos, in the genus Lichen opora, whe r e Harmer (1895) has named it embryonic fission. Later, I will discuss the relationships which can be estab lished bet�een specific polyembryony of parasites, experimen- tal polyembryony, accidental polyembryony, and the various methods of agamic reproduction. At the present time, it is sufficient to set down two facts: first, that specific poly embryony of hymenoptera, in ta�ing the fertilized egg as the starting point, is the most precocious case of dissociation of the body, which has been described in the course of onto genesis; secondly, that polyembryony is the type which most closely approaches the experimental polyembryony resulting from the artificial sep�ration of blastomeres or blastotomy. THEORIES OF POLYE1lBRYOHY been in A great many different viev.s have expressed these are de ve Lc nt , Most of expLanat Lo n of polyembryonic pme 1 - 133 - pure conjectures �nd as such, have no place in any serious a t t e mpt at a scientific treatment of the subject. T1":e theories discussed below are those �hich seem to hold a grain of truth and which have gained a certain number of adherents. (2) Theory of polyovular follicles. An attempt has been made to account for polyembryony on the basiS of poly ovular tollicles. Since it was Known that the mammalian oV3.ry occasionally possesses such tollicles, it was ll9.tural to suppose that this.fact might furnish a clue to the pro blem of polyembryony. A great number o f citations ot cases indicates how widespread the occurrence of polyovular foll icles is among mammals; but, although they show a distribu tion among widely separated forms, their occurrence in a given species seems to be rare. In the li�ht of this fact it is impossible to asS ociate the occurrence ot polyovular follicles with the causa.t ion of polyembryonic developments in mammals. The fact that a single' armadillo was found with such follicles can have no greater importance than have the simil3.r spor adic cases in other manwals. The signif�cant fact is that in the ovaries of fifty individuals belonging to a specie� . which reproduces by specific polyembryony alone, not a single case of polyovular follicles waS found. It is well to emphasize the fact that the polyovular follicles do not lie at the basis ot polyembryony, tor to l' - 134 - accept this theory would be equivalent to a denial alto- I gether of the phenomenon of polyJmbryonic development in the mamma l La , A multiple gestation from ova wh Lch have ac�identally become associateu together during a part of the o var Ian history, through the fusion of adjacent foll- icles, may have no more significance than a similar gesta- tion resulting from ova from uniovular follicles but sim- ultaneously ovulated, as often occurs in many mammals. It is only to t hose cases in v:hich the several embryos of a multiple pregnancy have taKen their origin from a single fertilized egg that we must loot for facts to clear up this pro blem. This theory has been- discussed somewhat at length. in order to emphasize the necessity of keeping it entirely dis tinct from the subject of polyembryony. It is essential in studying this problem that all of the individuals should have the same germinal constitution; but this condition would never be fulfilled in multiple gestations resulting from ova from a polyovular follicle, even if this could be proved beyond question and even though all the ova came from a single mother cell; because in that event, each egg must be fertilized by a different spermatozoon. (b) Theory of blastotomy. According to this theory, each embryo is looked upon as the lineal descendant of one of the carly blastomeres. In the case of two embryos aris it p�s been ing from a single egg (identic�l twins) supposed � - - -f-� rfII/II"" .. k _ .....--- iiiiiiiiiiiiii.....tiliiiiilliiiiiiiiii..iiiiiiiiiiiiiiiiiiiiii...... _....iiiiiiiiii _ l - 135 - that each individual is t he product of one of the blasto meres of the two called stage, while in the case of four it embryos has been aris�s from assumed that .' each . embryo one of the blastomeres of the four-celled stage. _�nd so lot has been a.rgued� for those cases in which even a great- er numbe r of embryos COme from one egg. Patterson (113) presented evidence that polyembryonic development in the armadillo could not be explained on the of a basis spontaneous blastotomy, in the sense that e�ch is the embryo 1 ineal descendant 0 f a single blastomere of the four-celled stage. This rather simple mechanical or semi-me chanical explanation might ho Id in the spo rad ic cas ee of pOlyembryony, Li.ke those of duplicate twins and double monsters, but there is no evidence that blastotomy operates in the case of e pe c Lf Lc polyembryony in higher forms. On the contrary,. the evidence po Lnt s unmistaKably to a d iffer ent explanation, namely, that a type of budding lies at the basis of polyembryony. (c) Theory of budqing. The process qf budding is a very com�on method of reproduction among organisms. In plants it is practically universal,. and in animals, it is frequently met with, especially among the lower forms. In late in many cases asexual reproduction by budding occurs •. In the life cycle, as for example among coelenterates to such for�s as the common Hydra it is customary regard the the organism as an adult when budding begins. But - I �, . l __ �� ��..� ��� _ - 136 - appearance of budding is by no means confined to adult organisms, or even to late st�ges of development, for it may appear very early in the life cycle. Budding has be�n described in the earthworm, in cyclostomatous Bryozoa and in parasitic hymenoptera. It is best to regard polyembryony as a precocious type of biding, and this perhaps only in the sellEe that it occurs early in the embryonic life, and V\' ithout the impli cation that it has been pushed forv,ard to the life-cy'cle or superseded a budding which in the ancestral forms occurr- ed at a late period of development. This would seem to be the case at least in the Polyzoa, in which the embryonic , budding is followed in the sessile larval stage by the typical budding to produce the colony. Patterson ('13) brought out the important point that polyembryonic development in the armadillo can be inter preted as a type of budding and while to show that poly embryony is a budding process does not solve the question as to the determining cause of the division, yet it is a distinct step toward the solution of that problem. PO LYEMB�YO NY IN AU I:wLtLS 1. Introduction. The fact that an egg may, under certain conditions, produce two or more embryos, is perhaps no more remarkable l' .. ... 137 t han t h.rt an egg v;i�l form a single individual. The de.v- e Lo pme nt a.L factors involved ill the production of the in dividual embryo must be the same in both caseS. In zoo- 10gic3.l literature tte term pclyeobryony has been applied to cases in v:hich t ro or more individuals develop from a single egg during the course of its early development. The term V'-3.S first used by botanists and among plant em bryologists is applied to all cases in �hich multiple em bryos are formed in the embryo sac, irrespective of the origin of the embryos. Thus in plants, multiple em b ryo a (polyembryony) may arise from tv;o e�3-s, or from the split ting of one eg�, or from anyone of the follo�il� sources: l1uce�lus integument, synergids, antipodal cells, endospt.-rm 1:,- cells. Animal embryologists use the term in the restricted sense only, that is, it is applied by them to cases in which the several embryos develop from one egg.' Some objections have been offered to the'term poly embryony, especially in appLy Lng it to cases of tw Lnn Lng in animals. But so long as the term is used in a purely des criptive sense, and without implying any particular mode of development, there can be no serious objection to its' unive rsal appl Leat' Lo n , Tv.'inning it se If must be regarded as the simplest type of polyembryony. That this is true has been demonstrated in parasitic hymenoptera. Three types of polyembryony may be recognized. _L l - 138 - These are (1) experimental polyembryony, or the production of multiple embryos by a�tificial mea�s. (2) accidental or sporadic polyembryony, or the occasional production of multiple embryos in a species �n which development is typ ically monembryonic; (3) specific polyembryony, or the habitual production of multiple embryos in a give� species. Among the first to produce experimentally two or more embryos from the egg wa s Hae c l..el ('69). He cut into pieces the blastulae of Crystallodes and obtained from the larger pieces normal nrvae. Since then there h�ve been many successful experiments of a similar nature. The re- suIts of these and many similar experiments have brought out some yery significant facts with reference to the behaviour of isolated oastomeres. Apparently these differ- . , ences are dependant upon the degree of organization of tLe egg or of the blastomeres at the time of their separation. For example, the undivided Ascidian egg is highly organized and hence if the blastomeres are isolated, even at a very early sta=!e, they are incapable of producing complete in- dividuals but only parts of individuals. On the other hand, in forms like Amnhioxus the uncleaved egg is not so highly organized and consequently the isolated blastomeres of the 2 and 4 celled stages are able to produce whole embryos. By the time the 8-celled stage is reached, the organization becomes more or less established and each of � - " b -,. -" ....iiiiiiiiiiiiiiiiii=iiiiii;;iiliiiiiiiiiiiiilliiiil_... liiiiiiiii._ l - 139 - the several blastomeres �s attained a definite v�lu€ as an organ-forming region �nd is no longer entirely able to do so. The result of such experiments have been used by various writers to explain polyembryony. It is probable that certain cases, especially sporadic polyembryony, may arise. in nature by the accidental separation of the early cle�vage cells, but specific polyembryony cannot be ex plained in such a simple �ay. Sporad Lc polyembryony occurs among bot h the inver tebrates and the vertebrates. It usually appears in the form of twins or in the closely related form of double monsters. Cases have been reported among cestodes, coel enterates, echinoderms, anne�ids and arthropods, but most of the cases cited in literature are found among the ver- tebrates. Twins or double monsters have been reported in every class of the vertebrate, from the lowest to the highest. Even in man it is found in the familiar case of· If "ide nt ical twins • Specific polyembryony also occurs among invertebrates and vertebrates. It has arisen independently in several distinct groups of organisms. The four· outstanding cases stom are the following: (1) Embryonic fission of the cyclo atous Bryozoa; (2) Twinning in the earthworm. (3) Poly embryony in the parasitic hymenoptera; (4) P�embryony in the armadillo. l - 140 - POLY�lfaRYONY IH J:HE P�1.RASITIC RY1IEUOl'TERA.. In the following table an attempt is made to incluoo all of t he undoubted poly-embryonic spe c Le e , The first column gives the name of the polyembryonic species, the second the name of the host insect, the third the average number of individuals in a brood, and the fourth the observer:- TilLE 1. (over) The species listed. in the table belong to three families of the parasitic hymenoptera, namely Encyrtidae, Platygastridae and Dryinidae. Further investigations will undoubtedly bring to light many additional polyembryonic species. There are many pOints of similarity in the meth od of development as found in the several species so far investigated but there are also striking differences. Many of the accounts so far given for the development of polyembryonic insects are incomplete and because of this it is difficult to prepare .J. su:nnnry covering such im portant questions as the origin and evolution of the type of development in the Hymenoptera. However, the recent observations of Leiby and Hill ('2j) indicate that poly embryony in insects begins as a twinning process in which the egg, at an early stage, divides into two parts, each of which ultimately forms a complete indiVidual. As the - - - J = .-- \ -; J Parasite Host Brood Observer Platygaster hiemalfs Phytophaga destructor 2 I.e iby and Hill '23 !-·1 Platygaster vernalis n n 8 Leiby and Hill '24, Pl�tygaster felti Vlalshomyia texana 11 Patterson '21 Polynotus minutus (Phytophaga destructor ' (Phytophaga avenae 11 1hrchal 04 Encyrtus mayri Oecophyllembius neglectus Silv. 11 Sil vester '15 Ageniaspis testaceipes Lithocol1etis cramerella 13 :Marchal 104 Ageniaspis fuscicollis ' Pray�Q olechus 14 Sil vest ri 08 subspecies maysincola Platygaster felti Rhopalomyia sabinae 18 Patterson '21 Encyrtus variicornis Anarsia lineatella 28 Sarra '15 Aphelopus theliae Thelia bimaculata 50 Kornhauser '19 Copidosoma buyssoni Coleophora steffani 58 Silvestri '14 Ageniaspis fuscicollis Hyponomentus malinel1us 100 Marchal '04 Copidosoma ap. Olethreutea variegana 148 Sarra '18 Copidosoma gelechiae Gnorimoschema gallaesolidaginis 163 Leiby 122 Copidoaoma gelechiae Gnorimoachema salinaris 191 Patterson '15 Paracopidosomopsis floridanus brassicae i161 Patterson '17 Autographa . Berecyntus baKeri Euxoa auxiliaris 1289 Snow 125 Litomastix truncatel1us Plusia gamma 1481 Silvestri '06 Copodisoma tortricis Tortrix comariana ? V.'aterson '22 l' - 141 - number of individuals �rising from the egg (of different species) increases, the process becomes more and more co l ica.t e d in a rnp , finally culminating highly specialized mode of development. • I DISCUSSIon A general survey of the subject of polyembryony ma�es it clear that this type of development has arisen independently in several different groups of animals and in some of those groups has undergone a distinct evolu tion. Since polyembryony has thus arisen and evolved it is not to be expected that its exact mode of expression would be the same in each of the several groups in which it is found. For this reason, any attempt to apply a general theory as to the cause of polyembryony is certain to meet with difficulties. Nevertheless, it is possible that in the final analysis it will be found that the causal factors underlying the production� multiple em the same caee s of bryos are in'all , irrespective the.ex act mode of origin or the number of embryos arising from the egg. Various theories have been advanced to explain the occurrence of polyembryony, such as "the blastotomy theory", "the fission theory", "t he budding theory" and "the physiological isolation theory". Whole some of these theories may have merit, yet I thin� that no one of l' 'I - 142 - them is ad e quat e to account for the origin and develop ment of polyembryony among the different animals groups. The var Lous theories have been discussed in a p re vLous se ct ion. In discussio� on the subject of polyembryony it has not been customary to emphasize the obvious fact that specific polyembryo-ny is a developmental characterist ic which is definitely inherited and as such it must arise and become incorporated in the hereditary mechanism in a manner similar to that of any other inheritable char- acter. That is to say, it must arise as a variation having survival value and hence affective in adaptation. Specific'polyembryony can be interpreted as a form of adaptation for the reason that it results in an increase in the number of offspring from the egg. This might have the effect of increasing the chances of survival of any species having this type of development. There is considerable evidence to show that 'poly- embryonic development has undergone a very distinct evol ution within certain groups, e.g.t the parasitiC hymen- optera.· In the hymenopter.3. it is possible to arrange a • series in which polyembryony began as a simple type of twilli1ing and gradually increases incompletely until in some species more than a thousand individuals are pro-' duced from one e�g. Moreover it is possible to trace l - 143 - throughJut this series the several steps by which this complexity has taken place. The s�me is probably true for the armadillos, al though the eviden�e is less complete. The facts on the development of armadillos are still fragmentary but as Patterson ('13) clearly indicated, the armadillOS v e re formerly multiparous and then gradually evolved to a condition ofl.C.bltparity. In the living species, the uterus is of the simple type, somewh9.t like that of the human. Nevertheless, some of the non-polyembryonic species sti]l show a strong tendency to ovulate and gestate two eggs. Some few species are said to be entirely uniparous. It is from this condition that polyembryony has evolved; first, by the production of identical twins (probably), the e the then by formation of quad rup.Le t , and finally by production of from eight to twelve embryos from the egg. Anothe r po int of significance is t he fact that before specific polyembryony is established in a given group of organisms the monembryonic type of development in the group often has become highly specialized, and it ,is probably true that without such specialization, polyembryony would not become established as the exclusive mode of development. This discussion h�s been devoted almost entirely it to cases 0 f spec ific polyembryon-y but in c onc Lus ion l' - 144 - seems desirable to make a suggest ion concerning what has been termed sporadic polyembryony. �uch cases occur throughout the entire animal kingdom, although their appearance in any given species may be very rare. In some species sporadic polyembryony gives every evidence of being hereditary. Thus in the case of human ident- ical twins, several writers have pOinted out that such twinning must be inherited, be cause 0 f the fact that it appears with very great frequency in certain families (Davenport '20). However, most cases of sporadic poly embryony give no evidence th�t heredity is involved. From reading the reports on such cases one gains the convict ion that the ir occurrence is the result o'f "envtr- onmental accident s" such as the separat ion 0 f the early blastomeres. I would suggest that the results obtained in experimental polyembryony offer an explanation of the cause of such sporadic cases rather than of specific polyembryony. LITERATURE on P01YEUBRYOHY no. attempt can be made to completely review or even'to list here the wide field of published literature the bearing an important relationship to study of poly embryony. This phenomenon waS investigated in other gro ups of the an Lua.L kingdom before it was discovered to l' - 145 - exist in insects. The studies of many workers on thiS of type development in the invertebrate group were pub lished under a diversity of names; and also, since most of these were European investigators; it was Lt.po ae t.b'Le to obtain many of them. However, a few of the most im portant p�pers will be mentioned and some indiation given as to their connection with insect polyembryony. It must be "remembered that the phenomenon of polyembry- o ny as it ex Ls t s in the ins ect kingdo m Ie very rare in the animal world. ,j 1§12 - For OU' know'l.e dge of t he early development of twins in e art hwo rrne v:e are dependent almost wholl;)T upon the account of Kleinenberg on t he embryology of , 1�, of J, , Lumbricus !!'apezoides. A single egg developes into tvo j, .,j, embryos; here multiplication is accomplished by intern�l budding in the gastrula stage. 1893, 1895-1896 Harmer - Among the cyclostomatous found a t ak in the po Lyaoa , Ha.rmer budding Lng place egg a.t the beginning of d evel opnent , The egg segmented in such a TIay as to give a polymorula ma.ss or premature em bryo wh i.ch sent out digitiform extensions separat ing secondary embryos at their extremities by budding. In the genus Lichenopo�, this budding is replaced from the beginning by the dissociation of the prirllary embryo it self into a large number of secondary embryos. There is here a phenomenon ,{uite comparable to that found in -m 146 - ras Lt ic pa hyme no pt e r a , 1900 - Bataillon ehowe d t h.rt the production of sudden ch ange e in the e?g of the Lam-prey induced by osmotic pressure constituted one of the best ue t ho de to use in order to obtain 'separatio:rl of blastomeres and their evolution into distinct individuals (experlmentJ.l po Lyernb ryo ny ) • 1901 - S3.int-Remy, in studying the development of the Cestode wo rms , found that there was formed arourid tte embryo a primal7 precocious formation in the sh3.p� of an external envelope. This envelope is nearly homo logous to t he amnion found in po Lyembxyo ni.c -d evelopment in insects. 1913 - Patterson described the polyembryonic dev- elopment occurring in the Arnndillo, Tatusia no vemc Lnc t a , He discusses the theories of pOlyer"1bryony and their value in relation to this phenomenon in insects. �ince polyerJbryonic development in insects hJ.s only been stud Le d for not more tha.n '3. qus rt e r of a cen of tury, t h e r e is no t a very w I d e and a iverse field published literature to re v te v , Indeed in this time not e d mc r e than e e+e n or ei,g-ht workers have Lnve at Lg at ttis method of development. It is satisfyins-, boy-ever, to knew t11"1t t he TIorll. done v:a.s carried out very extensively and much info rra.rt i o n is already known; But there still r ema Lne m:3.l1y unsolved p ro bIerae. which '\";i11 only be cleared l' - 147 - up thrJugh further re�e�rch. A very complete biblio- .;rapho) of Ltt e r a t u re relating to polyembr;:loL./ is to be found in Patterson's article (1927) "PolyeLlbryony in Anima.ls ", A short historical review of the wo r k on insect polyembryony v Ll.L be given. 1§21 Bup-nion � TIhile Bugnion found many embryos in �ncyrtus fuscicollis, he considered that each arose from a se p rr-at e e gg , �1 How�rd merely stated facts that he ha.d learn ed but did not attempt to explain hov polyembryony was brought about. 1904 - I.Tarchal. To Doctor 1!.a.rchal be Lo nge the distinction of fir8t describing the phenomenon of poly- embryony in insects, although his w�rK in part vas fore- s had owe d the ions of Eowa r d and the wo rk of by suggest. . Bugnion upon the parasite �ncyrtus fuscicol1is. He clear- ly demonstrated that the flexuous tube of embryos founq. by Bugn Lo n in Eilcyrtus originated from a single parasite e�g, as a result of a new and peculiar type of cleavage. The parasite egg was deposited in the host egI and each parasite eg? thus gave rise to as many as 150 adult para- sites. M3.rchal called this' phenomenon germinogony. In the same study he showed that there were.present 12 to 15 embryos of AQ'enia.snis testaceiues in a simila.r flexuous t u oe found in some larvae of Lithocolleti! and he Lnt Lm rt ed - :filii' r., I _ _- -�'...... iiiiiiiilliiiiiiillil__iiiiiiiiiiiiiiiiiiiiiiiiiiiiiiil_.. ... llliiiiiiii__• I -, .-;. 148 ;.. tlut this parasite also developed by polyeubryony. rn ue r ac s of the Hes�ian Polynotus nut , a platyg3.strid pa Lt fly (Phytonh=iQ'9. destructor Say.) was another spe o Le s de monstrat ed by I.1archg_1 as having the same type 0 f develop ment. 4�bout ten individuals of this parasite developed from one host larva. 1906 - Silvest-ri. A men.o rab Le work on insect ·poly e�bryony w�s published in 1906 by Doctor Silvestri, in wh ich he showed t hat an average of 1800 or more adul ts of the parasite Litomastix truncellatus Dalm. developed from a single egg deposited by this parasite in the egg of its host, Plusia gamma L. Silvestri made known in this w ork " the discovery of the so-called "asexual Lar vae as well as t he no rmal sexual larvae. 1906 - Howard. A critical review of all the studies carried on by previous workers was published by Doctor Howard. Further, he added a great many observa tions of his own, and as a result, made a substantial contribution to the tnov.ledge of polyembryony. AS well as the review a f previous w ork, Dr. Howard went into some detail co nc e rn.Lng the fixing of sex. 1908 Silvestri. The maturation and early cleav as age stages of A�eniaspis fU3Cicollis (the same species tJat published uvo n by Ma.rchal in 1904, but not completely described by him), and also reviewed the early c Ie a va=e oft t e e e stages h30 p c ie • Furthe r, this study included a - tW .1- r· ...... -- oiiIiiIlIiioolliiiiiiiiiiliiiiiiiiiiiiiiiiiiiiil=====:;; ...... oiiiIiiIiilliiliiiiiiiiiiiiiii.... l' - 149 - brief co mpa r Lso n of the species fuscicollis 1:ith the sub- species pra/sinco1a, in which an average of t er; individuals develop from one parasite egg deposited in that of the host. 1910 Silvestri. A preliminary study of the poly embryonic deve lopment of Cop idosoma buyssoni V:d,S publ ished ' by Silvestri, comparing, its method of development with that of Lit omast Lx , 1910 Wheeler. Dr. Wheeler questioned the presence of the so-called "asexual larvae" and suggested that they might represent another species of parasite • 1914 Ma.rtin published some studie e on A8:eniaspis . fuscicollis, in which he gave most attention to the gror.th of theoocyctes, their maturation and fertilization and the early cleavage'stages., 1914 Silvestri described the matur�tion and fert- ilization of CopidoSOffi9. bu.yssoni Mayr., a pa ras Lt e of the lepidopteron Coleophora steffani Joannis, which he had referred'to as developing polyembryonically in a previous note ('10), and in which he stated th�t from 41 to 119 Lnd Lv Ld ua Le of this species emerged fro.m. one host larva, these being very probably the product of one pa rae i.t e egg. 1915 Silvestri demonstrated the early stages of Encyrtus mayri Uasi., a parasite of the lepidopteron OecophYl�embius neglectus Silva which he had previously l 150 considered to develop by means ot polyembryony. However, in this paper, Silvestri states that he thints this para site develops monoembryonica11y. Leiby, in writing about this paper, states that trom Silvestri's demonstration the development o r this parasite is clearly achieved by poly embryony. Approximately eleven adult parasites of this ape c Le s emerge from one host Lar va , 1915 Patterson. Some observations on the later development of Copidosoma ge1echiae How. �ere published by Fatt erson, in which he dealt chiefly with the biology of the parasite and the sex relationship of, the adults'pro duced by polyembryony. 1915 Sarra. In discussing the biology of Encyrtus variicornie Nees. a parasite of Arnasia lineatella,- Sarra considers that the parasite develops by polyembryony, 28 adults emerging fro In one host larva. 1917 (a) Patterson published his observations of Paracopidosomopsg floridanus, a closely a:llied species to Litomastix truncell�tus. Again his observations were whiefly of'a biological nature. 1211_(b) Patterson published his observations con cerning the so-called �asexual larvae� with full details and also observations concerning the determination' of sex in polyembryonic species. 1918 - Gatenby made a splendid review of the l - 151 - previous studies on polyembryony. 1918 Sarra: A species of Copidosoma reared from the larv� of Olethreutes v�rie��na Hb, of which as many • as 68 females and 80 males were bred from a single host larva, was thought by Sarra to have the polyembryonic type of �evelopment. 1919 Kornhau2er - A dryinid.Aphe1opus the1iae, a p�rasite of the membracid Thelia bim�cu1ata, develops presumably by the process of pOlyembryony. The egg of the paras�te can be deposited in an inmature Thelia of any of the five nymphal tnat a.re , It gives rise to about 50 larvae which reach. their maximum. development either in the fifth nymphal instar of the host or in its adult stage. The parasitic La rvae escaping from their host, thereby kill ing it, drop to the ground, burrow in, pupat e and become mature the following summer. The parasitized membracids show many alterations, especially in their sexual charact- erist ics. 1919, 1921 Patterson studied the polyembryonic dev elopment of PlatYQ:ast,er felti Fouts, but confined his vo r s chiefly to the sex-ratios in the species. 19';>';>�� Le1Lby. In 1922 published his splendid studies on the polyembryonic development of Copidoso�gelechi�e How, a parasite of the lepidopteron Gnorimoschena gallae solid��inis Riley. Leiby described the development from - 152 - pre-ovipoeition to the fully dev�loped larvae in a more detailed v;ay than did either Marchal or Silvestri in their i�vestigations. Some observations were also made concerning the biology of the parasite. This piece of work. has been constantly referred to by the writer in the study of Berecyntus,. since this species and the one studied by Leiby, Co-oidosoma gelechiae are very similar. 1923 Leiby and Hill. In this piece of work, the authors described t he twinning and mono-e�bryonic develop ment of Platygaster hiemalis, a parasite of the Hessian fly� This was done in a carefu� way, for the cleavage Qt�ges of development. This parasite develops both mono embryonically and poly-embryonically but by the second method only a ema L'l number of indi viduals ar� 1,1rod�ced from a egg (61). one-third of the' ,J. ,single A1?proximatel� eggs deposited do not devel�p beyond the cleavage nucleus stage, pro bably because they fail to be come invested by host tissue. The twinnin� development in insects here described for the first time is a simple type of poly embryony. On the other hand, the mono-embryonic develop ment of this parasite is very highly spe�ialized. Since Pl�tygaster hiema�is exhibits both types of developmeftt, in in it ,furnished a clue to the origin of polyembryony sects. The authors believed that the mono-embryonic dev of the elopment of some of the eggs, and the fact that some l' 153 - of e:sgs a g rcup are insemina.ted wh LLe o t her e of 9. group are not, v:ill account for the origin o r mixed .uro o d e of the p.rras tt e e and the occurrence of single :individuals of a sex different from that of the others of the brood. 19�± Leiby and Hill. A slightly more complex fonn of polyembryony is described here for Platygaster vernalis, 'I. another par as Lt e of the Hessian fly, in which an average of 8 Lnd Lv Ld ua La are developed from a single egg of the. parasite. This paraSite develops only in the mid-intestine of the host larva. The development of a closely-related species, POlynotue minutus Lindemann, wh Lc h is also co n- fined to the 'mid-intestine of the Hessian fly larva in Fr3.uce, has been described previously byl.�archal ('04). Mirchal's'papcr diu not tre3.t of the precleava�e or cleava�e stages of development in a sufficiently detailed manner to demonstrate polyembryony, put i� did describe quite fully the organogeny of the e�bryos. There are also some indications that � minutus and � vernalis differ slightly in the details of their development, although it is difficult to determine·this definitely, because the development of 11archal's species is not illustrated enough with sufficient histological preparations. In the _.-' present paper, emphasis haS been placed upon a study of microtomic sections a.nd the paper has been illustrated from this viewpoint rather than upon gross examinations. - 154 - The insect species ment ioned in t his review �epresent p�ra�itie hymenoptera belonging to the families Eneyrtidae, Platygastridae, and Dryinid�e, v.hieh have been definitely proved or sug -e e t e d as having po Lye mb ry o ni,c d eve Lo pme nt , It is not improbable tha.t w ith more research in these or related groups that other species will be found to be polyembryonic in their development and that they will demonstrate some striking variations in their methods of arriving at the same phenomenon - th�t of producing more'than one individual from a single egg. :mTES ON THE HOSTS J.N]) THE PA:..q.'lSITE. Berecyntus bakeri o v tpo e its during the summer a nd early fall in t he egg of its host, the time of oviposition depending upon the host species. With Euxoa ochrog�ster, its occurs late in t principal host, oviposition .uigue , Vi.hile for other hosts, oviposition ranges from early July to early September. In Euxoa ochrogaster, by the arrival of v. inter the developing egg or pa rae Lt e body is found as a polynuclear mass "ithin the completely formed host embryo which has developed simultaneously. The red-backed cutv:orm emerges from its egg early in May, depend ing upon weath,- r' cond it ions and commences to feed. By the middle of June, the larvae are nearly full grown and soon after the prepupae begin to make their pupal cells. l' - - 155 :.:etl1 w..Ll,e the o.a ras Lt e body in the ne v Ly hatched host 13.rva develops into a polygermal rms� while lodeed in the f�t-body. �s the host larva grows, the polygernal mass becomes a La r ge group of embryos which are later freed into the body ca vLt y of the host larva 3.S pa rae Lt Lc larvae. These larvae feed upon the body content of the host devouring the blood, fatty tissue, muscles and every thing except the chitonous parts. AS a result, the para sitized larva dies just at the time it would normally have pupated (late June). In the meantime the parasitic larva become full �rown, scatter themselves �ithin the host larva, each individual paraSite frming a separate but contiguous chamber in which it pupates� A parasitized larva then appears as a mummified carcass containing as many as one thousand 0 r mo re paras it ic pupae. The pa ra « sites rem3.iu thus for about three weeks, vhereupo� the individu3.1s in one carcass all emerge at about the same time. The females whether fertilized or n0t are ready to ovi�osit, but if males are present, copulation occurs first. The paraSite, like its chief host, has but one gene rat ion each year. The t Lme 0 f appearance of t he pupal 3.nd adult stages of the paraSite has been found to occur a few days later than the similar 'stages of the host. This condition holds des�ite the varying life cycles of the different host species. l' - - 156 � total of at least fifteen parasitic species haa been reared from �Q�� ochro�a2ter at Sastatoon. The heavy pa ras Lt Lsm b1 these species, the majority of vh Lch p ro bab Ly initia.te their developll1ent in the very youn;! larva and complete if before the host Larva pupa t e s or shortly after pupation, causes sc ue mortality of larvae vh Lch c o nt a Ln e t a ce e of Be recynt ne , ConsE-<1uent ly, ovi po s Lt ion in the host egg by Berec.:yntus is by no uea ne an aes urance that a brood of this species will nature. This condition is fre;auently met in natu re and is more fully discussed in the section on super-parasitism. With certain of the host species the method of pa rac it ic development varies somewhat. Pith such spec ies as Feltia ducens and Euxo� tristicula, the host egg is parasitized early in the summer and hatching. of the host larv� occurs at a later date. Both of these hosts over winter usually in the fifth or sixth Lne t a,r , By this t Lue tbe parasitic larvae have been freed into the body c�vity and it is in this stage of their development is resumed �ith the development of the host l�rva and is cOQpleted similar to that in Emwa o ch ro zas t e r , D�V�LOPI.:SET OF TH� �G. (FAL:' P:SRIOD) 1. Preoviposition - The eggs found in the OVaries are very elon:S'3.�ed Deasuring about 0.14 mn. x 0.02 mu. l � 157 - ° ':.'hey ..ire e:.. o LLe n ..it t he t v o ext r enn,°t i.e e , To see the egg's c Le irLy it ,ie ne(;�s.sJ.ry to fix slightly Lr. o em Lo a.cid and color for a corisiderable perio� (48 hours) in picrocaroine. The broad or posterior end of the egg co r-z-e e po nd e to the ve t a t Lve of ge pole other eggs., ,It is from this region thlt the tr�e eru�ryonic cells arE formed by the procees of c l.e ava-re , The pO..jterior pole presents a slight t h Lcke n Lng of the chorion. According to Bugnion, a micropyle should be found at this pole, but no such structure v.�s clearly defined in the eggs studied. The. anterior end of the e,gg gradually narro�s down and finally terminates in a short fin .. ce e a v.h Lch ger-lil�e pro , is quite cha.r sc t e r Le t Lc of the eggs of many pa.rasitic Hymenoptera. AS development pro- gresses, the content of this process is gradually taken i�to the egg proper, so that by the time the egg is de- posited in an host egg it is �o longer seen. z. Ovinosition to cleava�e. As has been stated' in the life hi$tory of Berecyntus, the eggs are deposited in the host eggs late in .L·mgust,·the time d e pend t n= to some extent on weather conditions. The par ae i.t e egs": d eve Lops within the developing host embryo until by the be::"inning of cold weat he r the pa r ae Lt e body is a. La rge syneyt Lum of emb ryo.: ie nuele i surrounded by a nueleat�� membranee or protoplasmic l�yer. This parasitic network is eQbedded in the host a fully developed munute embryo st ill v: ithin egg. 1 - - 158 Both par ae Lt e and ho e t P3.SS th_e w Lnt e r in this condition. UsuJ.11y only one pa rae Lt e egg is inserted i.nt o a host e:::, but if more ar e inserted, it is ge ne r a'l Ly the case tha.t the others will become aborted. E�g �fter ovinosition., After oviposition it is extremely difficult to find. by simple dissection the egg of the parasite. The egE; is very SIIl3.11 and it is placed in a relatively enormOus mass, m�de up of embryonic ele- ments of the hoet egg. Because of the fact that the chorion forms a h�rd layer it was �ery difficult to ob- tain series of the host eg�. This was very desirable in order to locate definitely the relation of the parasitic egg to host tissues" Nevertheless, sufficient sections were o b t a i ne d to study t his relationship. The ue s t 8ec- tions were o ba i ne d by using Gilson's liquid as 3. fiXer. To understand ,fully the dis'triuution of the pro ducts of the po1ygerm, it is necessary to call attention to' the relationship existing between the paras it ic egg and the host tissues. T;!is question is of very great in- terest for it can be showrt by means of a very simple ex- periment t hat the development of the pa ras Lt Lc egg is dependent upon the development of the host egg. This is true for the init ial stages in the development as well as for the later stages. lLrcbal ('04), Silvestri ('06) and Martin ('14) have all stated certain pOints concenling this relationship. - .-:IfIIIt' I - t 159-- In the case of _'I.'�,·eni3.snis, l.:J.rc.:hal e t at es that in order for the egg to develop com-rletely'it is essential that it be, placed in the general cavity of the- host embryo, that is, v,'ithin the embryonic region of the deaoping cater pillar (Hyponomentus). In all of his sections sho�ing the pa rae tt Lc egg, he always found it to be in the general cavity of the embryo, where it was normally developing-. However, Marchal did not ea ne Ld e r his studies-numerous and detailed enough to determine conclusively �hether or not some of the eggs are lost or die if placed in an unfavour able position, such as the intestine or yolk.. He implies, however, that some are lost because of this, since it does not seem probable ·that the adult parasite can seek. w It h the extremity of its ovipositor the embryonic region and to' choose .it in preference to the ether parts of the egg. Further � he is rather afraid to admit that with the adult being placed on the middle of the free portion of a host egg and having inserted its ovipositor vertically that the length of the ovipositor is so finely calculated that the egg is placed in the embryo. He does not even admit that the paraSite egg must be placed in the general cavity, and it is vf;;;ry probable that, simply deposited 'in some manner be twe e n the different embryonic elements, it must be mJved by the pressure of the side Tlhere the elements show between themselves the least cohesion, that is, from - 160 - can the side of the general cav tty , The objection be raised, it is true, t�t the distance sep�rating the host embryo froID the free surface of the egg v�ries, to a certain degree, f o LLow Lng the stages of development and that this embryo can be more 0 r less impressed in the vitelline mass. M:3.rchal did not believe that the methods of ovi- po�iting used by Encyrtus were infallible, principally be cause.of the small numbers of the host eggs parasitized, • despite the gre�t fury with which the eggs are attacked. From that it seems natural to conclude that a part of the germs of the parasite must have aborted because they have been placed in the egg of the host, finding some stages unfavourable to the ir development. In later stages, l.rarchal d Le co'ver e d that the epi- thelial layer which at first forms a cyst about the dev- eloping egg and then gives rise to the elongated tube of the chain of embryos, is the product of the host tissue. In the case of polynotus minutus, he discovered the Lnt - erestin� fact th3.t the eg� is lodged in the gastric pouch or stomach of the host (Cecidomyiid) and there curiously e nough undergoes its development. According to Silvestri's observations, the egg of Litomast Lx may be La Ld in any part of the host embryo (Plnaia.). The eg? is destroyed if la.id either in the in- test Lne or Lk . In yo late stages, the germ maSS may be J - 161 - found i11. any par t of the head. It is most f r e que nt Ly f o und in the thorax, either above or below the oesophagus. He a Leo found the pol,/germ in the nerve ganglia, especia.lly the brain g�nglia. In AJ�enia.spis, Itllrt in believes that the freq_uent occurrence 01 the egg in the thoracic ganglia of the cater- pil13.r is to be correlated with the laying time of the parasite. He finds that the egg clings to the ganglion in such a way th.3.t the typical e hape o f the ganglion is pre served. In the late stages of development, he could no find the rrn c o nne c t e d h t.o n v h i.ch longer po Lyge vLt a g angl , fact leads him to conclude that on account of its grovth the polygerm is forced out of the gangLLo n , My ovm observat ions on Berecyntus very closely para.llel those of Silvestri on Litomastix. The egg n�y be deposited in any part of the host egg, but diSintegrates if it happe ne to. be placed in the yolk. or intestine. In the nevlly hatched host larva, the egg may be found in any part of the body cavity or embedded in the tissues ad j3.cent thereto. There are two k Lnd e of t issues in which it is frequently found, namely adipose and nervous, chiefly the fo rmer. The polygerill is often surrounded or embedded in fat tissue d e ve Lo pe d from the host cells. The fat t issue pro bably starts to develop from mesenchyme cells �nd not only serves as =l source of rnt r Lne nt for the grovLng polygerm, - 162 - but it al.eo holds the embryonic mae se e together, thus delaying their dispersal. The rela.tion of the host tissues to the pa r ae Lt ic egg is very Lmpo rt ant in the development of the latter. It can be e ho wn t h at the de ve lopme nt 0 f the paras it ic egg is dependent upon the g rovt h of the host embryo. This has been s n hcw experimentally in the follcoving simple i3.y • .'l group of eggs laid in the laboratory by an unfertilized red-b3.cted cut�orm moth exposed for 24 hours to a l�rge g ro u j of f em aLe p a ras Lt e e , _is a. result,. parasitic eS'?s �ere usu3.11y deposited in each host eg�. � fe� daYb l�ter these v.er e ee c neu , The t e eg?s fixed and t Lo uo h ggs , of course, did no t develop (not fert ile) and in e xarn Lnat ion of the sections, revealed the fa.ct t h.i t 110t a single pJ.r:3.- site e g> developed. Under e LrniLrr conditions, but v It h fertilizeu uo t h eg-;:s, all of the p s.r ae Lt Le eSt2;s would have been lell '�dvanced in the cleav3.�e stages. �uch small eg;s as these paraSite egiS are devoid of yo Lk as a fundamental tissue building IJl').terial for 3. developing embryo; there is substituted a Jesser amount 'of eg� cytoplasm, �hich is abl� to function as 3. trophamni�n Develonment in the host e�g D,\Le to the mtnut e ne se of the Lost egg and of the pJ.rasitic egg, and hence of the components mating up these bodies, it V.:1S found impossible to o bt a t n such i11- formation cc ncc rn Lng the development of the pJ.rasite I me _.me- - 163 - in the ho s t 7he of mat ur a.t Lc n e:::. processes , fertilizJ.. " t i o n J.11d of c Le a v a je 3.13 occurring in B&recyr:tm:: ar e t he r e « fore not discussed e but i.r; to out to any xt e nt , order fill this a rt of the orrl c d eve Lo nt n p po tyembry pne , tnro rrnat Lo is given �8 t�keD from !ei�y's detailed studies on Copodiso@3. R'elechiae. 113.turation .3.nd fertilization. The rrL:I.turation·and fertilizatio � of eggs of the po Lyernb ryo nt.c Rym€noptera have been studied by Silvestri ('06) in Litomastix; by cilvestri ('08) and 1,Ia.rtin ('14) in �i3.2Vis; by Silvestri ('14) in Copodisoma bu.yssq_.illi; by Patterson ('.17) in P.3.ra-. copod isomo��; and by Leiby (' 22) in Copodosom3. g-elechb,e. In all po Lye mb ryo n Lc species it has been e hovn that there are eventually produced three polar bodies in both fert ilized and part�enogenetic eggs. The process of m�turation is identical in fertilized and unfertilized e g rs . The first stage of maturation OCClUS vLt h i,n a fev! minutes after ovi position Lnvo Lv l ng o nl.y the c hr-o ma t Ln , The first ma t urat Lon division results in reducing the number of chromosomes from s Lxt e en to eight. The second ma t u rat ion occurs almost imm ediately after the first is comnleted, or w Lt h Iri 45 to 75 minutes aft,- r o v i.po s it ion. Eight d ist inct chromosomes re ma i n ,J.ftt::r the second maturation. One po Lrr bo dy is pro duced in the first division wh Ll.e tViO more a r e seen in the second maturation, mak Lng three po Lar bodies in all; t hes.e 1 - 164 - are not t hrov.n off from the e':8. liter maturation is completed, there is a pe r Lo d of time (from the second to the eighth hour) in wLich the fe-male pronucleus changes its shape and a ppea.rarice , From the eighth Lou r to the tenth hour after oviposition� the nucleus assumes a stage of repose; the first.cleavage occurs at the latter time. Fertilization. The sperm rmintains its original for a short time after the '" shape only entering eg�. Ile ve r more than one sperm has been observed in an eg�t and there are indications t h a t a sperm is not alv.-ays f o und have under �. in e�gs deposited by females.th�t copulated ns to a rrn e xps r Lrne nt a'l eo nd Lt Lo ; however, failure locate ape does not necessarily indicate its absence. The entire sper- matozoon enters but the tail disappears and only the head is transformed into· the male pronucleus. The fusion of the male and female pronuclei results' in the production of the first segmentation or cleavage nucleus. Cleavaqe to overv.-interinq stage. Two g e ne ra.I r eg to ra of of the mature eg� can be distinguished by the position the nuclei before cleavage. The polar bodies, or the polar the nucleus if they have been already united, is found in anterior part of the egg. The posterior part of the egg contains the cleavage nucleus and the germ cell dettrmilmnt. not At a vdry early stage the two regions are separated by a membran�e nor does the ooplasm in each stain differently. - 165 - They are zirilpl:ir defined by po e i.t t on of their nuclei and tht; totally difft;rent part each region plays in later developrnellt of the egg. In later stages, the t wo regions are d Ls t inguished by the vary i.ng dens ity 0 f the ir o.o p l aern �nd by the presence of a membrane sep�rating them. In respect to cleav�ge, the eg� of pJlyembryonic insects differ from ttat of the typical inSect eg?, in th.lt the cleava.ge nuclei are initially accompanied by cyto- plasmic segmentation. Anoth�r point of interest is the fact th.l.t the course of development is in no way modified by feftilization. The history· of the cleavage, as well a.s that of the polygerm, is the same in the fertilized and unfertilized egg. E3.rly c le avaae , 5 d Lv Le Lo ns tak.e place in the early cle�vage stages. In discussing this stage of the parasite egg development, a comparison will be made of the early cleavage stages as found by Deiby for Copodisoma �ith those found by other wo rke re on other. spec ies of Hymenoptera. I.13.rchal ('04) found in the early development of Ageniaspis a prominent amoeboid nucleus in the region of the egg v-h Lch surrounded, but d i.s t inct from, the embryonic region. Trac- ing its development, he s bowe d tha.t this region wh Lch vaa just under the adventitious cyst (host tissues) was the sam� as the amnion of other insect embryos. He gave the name paranucleus to the prominent amoiboid body and the - 166 - region surrounding the p�ranucleus he termed the troph I" amnion. Silvestri, in his studies on Litom�stix, :first differentiated the pol�r region from the embryonic region and sho�ed that the first region surrounded the second. The pol�r region also contained an increasing number of nuclei originating in the fused polar bodies. He proved that in the species studied by l1archal, the paranucleus o rtg Lnat e d from the' same fused l?olar bodies (polar nuc Le uc ) and th:it the trophamnion v:as,a differentiated part of the egg which in Litomastix, he termed the pol�r ooplasm. 1Brtin ('14) merely confirmed the previous observ� t t o ns or Marchal and Silvestri in this connection. Leiby found a simillr uondition existing for Coptdosoma, the differentiation into polar and embryonic region occurring at a stage of development mid-v;ay between those recorded for Ao:eniaspis and Litomastix. The polar nuclei divide more frequeDtly than for Ageniaspis, which is expected'con sidering the great v.ar iat Lon in the resultant numbers of larvae. The function and d Is po e i.t Lo n of trhe polar ooplasm and its nuclei in Copidosoma are similar to those describea for _tR'eniasp is by 1hrchal, and thus are calle d the t ro phamnion and paranucleus respectively� Later Cleava�e Stages. The previously danribed dev elopment of the egg taKes place during the first 36 hours - 167 - after oviposition. The cell walls are no l0nt-er distinct " , I structures �nd the centre region reambles a veritable syncyt Lurn, The emb ryo n Lc region is co mp'Le t e Iy surrounded by the trophamnion. By the forty-eighth hour, the polar region appears. to be still furthEr differentiated in the form of a swelling, while the embryonic region ahov e no changes except that the number of nuclei has increased. Thert: is evidence that the polar region increases in size by its abeo rpti o n 0 f host material, for such eggs which are accidentally placed in the adult parasite v:here they a re not -Ln centact with the general cavity, i.e. just under the chorion of the host egg, do not develop such a not iceable increase in size. During t he period between the forty-eighth hour and the fifth day, the greatest change occurs in the polar region. Between the fifth and tenth days, the development of the parasite egg proceeds irregruarly since it is taKing place during the fall of the year; consequently, Similar stages .i re to be found Ln different hJ.::t e�.;a, .J.lthough it is Known that such para- site e ggs are not of the Same age. By the twentieth day the ent ire parasitic body is usually found completely surrounded by tissues of the host embryo so that the whole resembles a cyst. By the thirtieth day the development of the par ae Lt e and host has reached the stage that an normal host embryo is fully developed and nny be readily ... Lis: _e.....c_t_e_d_f_r_o..;,;,rn.oiiiiiiii:t=h=eiiiii8ieiiiE:glWi�iiiii,·iiiiiiiii;;;;A.='b=0=U=t=4�5-.d_a_y_S_,;;a;.,.f_t_e_r_o_v_i_P_0_S_�_.t_�_.o_niiiiiiiiiiiii'iiiiiiiiiiiiiiiiiil__ l - 168 - the p az-as Lt e body is found in the condition in vh Lch it passes the winter v. ithin the host. In the fiml winter- ing sta!e'the parasite �ody consists m�rely of the em bryonic region containing m�ny cleavage nuclei between , , which are groups of dense,granular particles absorbed from the paranuclear masses in the trophamnion. The egg, or ).s it should now be called, the parasite body is a veritable synyctum surrounded by a trophamni�n containing many small nuclei. Summary and discussion of development of the ,para site egg to the overwinteri�� stage. Studying the dev elopment of the parasite eg! from the time it was deposit- ed to the overwint ering period, there is revealed the fact that it does not increase greatly in size or volume, but that a great number of transformations .a r e carried out in preparing for a rapid growth in the spring when the host eg� h�tches and the' host larva commences to feed. In a 45 day period of development in the fall, it w as seen that the parasite body increased only twice its size in length and width. This would indicate a very remarkable and perfect biological adaptation, because the growth of the parasite body is indeed dependent upon the a�similation of host products, �nd if the parasite body developed to arw great er extent, the death of the host would result, because certain tissues of the host which are necessary for its 1 - 169 - own development in the spring would be Lmpa Lr'e d or destroyed. The fall development of the eg? is there fore restricted to cleavage of the embryonic region forming a considerable number of nuclei, and to pro tecting and nourishing this embryonic region by the diff erentiation of p�rt of the egg into a polar region •. DEVELO:??1E:TT OF P.\.3.ASITE BODY IN HOST L_'L.1={VA. (. SPRIUG & EAR""_Y Sm.!1.�T{.) In order to have available 'the earliest spring stages of the parasite's development, I found it necess I I' � ary to rear paras it ized Lar vae fro m eg.�s rh Lch ha d been pa raa Lt Laed the previous fall, .'since such host La rvae cannot be easily found in the field in the spring be- cause of their sm�ll size. P�rasi�ed eg�s were carried through the winter in cold storage and hat ched out in the spring. This was done primarily so that no o t he r factor would enter that might influence in any way the natural development. The same results were obtained, however, from the larvae obtained tr... rough induced para s Lt Lsm during the winter experimental wo r k , The pre caution to keep in mind in this method WJ,.S that after oviposition.of the pa.r as Lt e egg had taken place it "as essential t ha t the host egg be kept for a period of at least three weeks under laboratory conditions. This vas l - - 170 to allow development of the h�st ege and parasite egg to 9ccur• Under field conditions in the fall, a minimum of six: we e ks is necessary for such development to t ake place. From these f�cts it is highly possible th�t the pazas Lt e bodies mi�ht 1'3.SS the vLnt e r in an essentially similar condition, even though oviposition may h).v� been expt.rimentally brought about at a later tiLle th9.n normal. Such p�rasite eg;s h-sten in development if the hoct eggs are Kept indoors. Under normal conditi�ns, it is quite pro ba b l e t ha t t he development 0 f the paras it e egg is accel�rated if deposited in a host egg which at the time of p�rasitism contains an advanced embryo, and th3.t it is ret�rded if placed in a host egg whose embryo is still in the initial stage. Under normal field conditions, oviposition of the p3.n8ite probably occurs'at the time the blastula sta�e of the host embryo is reached. Development of the plrasitic body after hatchin� of the ho.:;t but previous to any indic'3.tion of uolyembryony. With the early activity of the host larva in the spring, chiefly its feeding, the parasite body rapidly increases in size. In opening a host larva of the first or second instar under the microscope, there can be fo'und in some of them, a small body resembling somewhat the genital gl3.nds of a larva. These rounded bodies float bet�een tte viscera, in the blood of the larva and are in reality -�--��� - 171 - parasite e�gs. By means of a membrane (troph::imnion) these p�rasitic bodi�s ma�e contact with neighboring host organs, and the position they occupy is a1v.ays very variable. Sometimes they are found invested to the posterior end of the digestive s"stem and more often, on the inside of adi pose tissue, r.hose continuity in such a case is interrupted by the parasitic body. It is quite rem�rKable that in the same egg of the host, one finds very often the parasite egg occupying a very simil�r position in relation to the organs of the. larva. It would appear t�at the parasitic body occupies a situation in the host larva determined by its rel�tion to the embryonic structure in the egg. Fur ther, it is quite possible that the position of the para site egg depends to some extent on the st�ge which the host is found at the time of oviposition of the parasite egg. The par ae Lt Lc body\";as examined in the host Ia r va by the following me t ho d: - the host larva VI as placed in a 10 per cent. osmic acid solution for 12 to 24 hours. After dissociation, it waS fixed in osmic acid and colored �ith picrocarmine. In its interior there can be seen by t e nce the e The numb e r of ranepa.r , paras it body. embryos seen were not very grea.t, due to the La e k of high micro , e in the same manne r scopic po wer On a preparation mad , but examined in Y"ater·v,rithout coloring, the nuclei in the embryo were v�ry �pparent since they are quite refringent. �------�--�------====..._------... .,.•__------�L - 172 - The entire pa r a., it e bo dy r:ay be nov- be tt er knovn ' as the rdygermal IUSS. Sometines, it is cy Lf.nd r Lca in eh e but on ap , often account of its size, it is do u Iil.e d upo n itself. Usu al Iy two of these f1i3.sses are :!bund in one ho s t La r va o ne on the side rax , dorsal in the pro t ho and the other beside the intestine in the first to third ab- do m i.na'l segments. Both are found in host adipose tissue, r.h Lch seems necessary for the normal development of the rin , s o me host hore ve r po Ly ge In larvae, , po Ly ge rma'L r.lasses have been dissecteu but which were not imbedded in adipo�t IT tissues. rne re such "nat.e d po Lyge r me are found, the-ir no rmal is r e t ar d e d in development considerably , resulting a d ec r e aae d nuubc r of adults appe::t.ring. These _9Jl;Yfel'Ill3.1 masses C3.n usually be seen under the b Lno cul a r by the means of reflected li:S'ht, and ::it the latest s t age of t:teir development, using the method described by StricklJ..nd for ro b c i.. o rm for ae i.t ee the III3.SSeS p inj ut lJ.rvae par , polygerEl3.1 are found, e ape c La l Ly the naked l)oly�erms. Thus it 1:3.08 be c n found thJ.t the result Lng development f ro u s uch poly- o their normal d e vej o nt but germs is not nl, delayed ill prue thJ.t the numbers d e ve Lo p Lng is much decreased. Thi::: is al,v a.ys o bv Lo u; vhe r e t r.o pol;igermal mae see .i r e present in the s ane host larva, 'with one err ro und e d by ad i.po s e tissue and the o t he r lying in contact y-ith muc c Le e or, as is nore f re que nt Iy the case, the alimentary canal. In such - 176 - , o a c ar,e , t h fo ::Lle r 'erln 'i. ill 1·0 due; e d e ve'l d pol�r __ .... ful1;r pe l:trv,;l. gress, w l LL be p3.rt b.llJ" de e t r o ye d by the r o re adv.3.l1ced l..lrv3.€ d e ve Lo p in8' fro m t h e prot e c t e d pol:17g�rm_. Leiby st�te8 th.3.t in Co�idosom� no larvae viII be produced by na l.. e d -po Lyge rraa.L LI..lSSeS but thi;:;; i c not ·absolul.tdy true in B�recY::ltu_§:. '-1. host larva V;3.S p ro be d t h at revealed two such unrrro t e c t e d pc Lygc r ma'l L'laSSeS, arid about ty:o hundred t rae Lt e e Eove ve r I t t he adul pa emerged. , feel t ha chances are very fevc, of such a cJ.se happening very often under nat ural c o nd it ions. Fornct.t ion of �e(;ondary poly:::ern13.1 nl.lsses - dis junc- tion or norulae and embryos. DespU'e the fact that a great many sections WE:re m..lde of host larva.e at a stage of their development vhe r e the po Ly ge rmaI Glasses are bre.J.t. ing up, forming polYL1orular masses and embryos, no 'definite information was obtained. The method used vas precisely the same as trut used by Leiby in Conidosoma, but it is felt that the microscopic ma_;nificat ion obtained vas not euf f i.c Lent to clearly Lnd Lcat e the different features in the sections. Therefore, a. short resume of this phase of the deve10�ment will be given as obtained by Leiby in his study on Conidoso.@§:. Due to the concentration of the trophamrnion between the germs (each of which is similar to the original polygexm) the entire poly�erm is divided i11tO two or three nae se e , These further sub- l secondary ,,� 1 3 ... 2 .. 1 Elate VI: (I)eggof Berecyntus bakeri(small)after insertion in an host egg of .. � ochrogaster This figure represents a section cut through the host egg .. The host egg contains an embryo surrounded by the amnion end. the vitellus .. In this is embryo found the parasite egg .. (2)section through parasmte body 44 days old showing approximate condition during hibernationin the completely developed host embryo.Note that some of the cells have became elongated and spindle -shaped. (3}section showing the condition found in the spring after the first development. - 174 - continues until divide into tcl��ry nusses. This process v'hich the late morula or early embryo sta�e is reache�, by from t Lue e ach no ru la or embryo is entirely separated aI1..y of them are enclosed into one po2.y otl:er. Eowe vc r , all tost tissue. morul3. or polyembryol�l fi�SS by the adipose these nar ke d c s are evi �H a sli�htly Lat e r stage, hange co ne Lde rab in size; dent: (1) the germs h.3.v€ increased Iy and the (2) a greater numb�r of nuclei are present; (j) become cell waLl.s in which the nuclei are present pro nounced for the first time since the early cleavage sta�es. de ve Loome nt of the An importance difference in the re�ched at this embrJos in Conidosoma and Berecyntus is of e a ch time. In Copidosoma a distinct segmentation germ takes There into two smaller germs of equal size place. is is no doubt but that in Btrecyntus this segmentation Silvestri for �itomasti�. more simil�r to that described by a continuation Instead of only one segmentation occurring, to ien or twelve germs of the g�rms (one germ giving rise de ve Lo to the before division ceases and before the germs p is to e Lc h an embryo) morula s t ave , acn of wh w�ll produce about 1200 individ be expected, Since Berecyn� nroduces 2000 v.he r e-, uals from one eg!, and Litouast ix about adults, 180 adults and as .Ureni.1suis fuscicollis produces only It is bab Ie CopidJsom:3. .Q'elechi� about 166 adults. y_uite prc all in no diviSion of the !erms occur at polyeF.,bryonic .. L t:l:__ - 175 - forms v he r e r 0111.; a fev; adults a.re p o d uc e d from one tgg. This e e rrt t gme a Lo n of the g- rme must be rega.rded .a s a highly epe c La'l Ls e d adapt at t o n for Lnc re ae i ng the numbc r of which are embryos to be developed from one egg. ��fter this segmentation has be e n completed, each division is known as a morula, since each is a mass of nuclei, and w Ll I give t o an embryo eve nt ua'l l y t rane ro rmtng into a 1a.rva. (The term embryo is used htre in the sense that in 'this stage of the emb ryo n Lc development, the err.bry onic layers will be differentiated; strictly speakil�, the morula is an early stage of the embryo.), The polyembryonal mass. The stages prevL:usly dis- cussed will rejuLr e (for BerE cyntus) until tht;' early fourth ' Lne t ar the vo r m of red-b::l.cked c ut e de ve Lopmer , Shortly after (3 to 4 days) advar... ced embryos have been observed, and the entire rllSS is KnOv.n as the polyembryolnl mass. Usually the l)O'yembryor.a.l mass v.ae observed to be very ir- , regular a.nd indefinite in outline. In some host larvae the L1testine appeared to almoet divide the entire mas s into sections. Sometimes it has been observed t ha t the mae s is parallel to the iongitudinal area of the host larva, and w ... t e e t , very c10se,if not in contact It h , the ii Lne In the fifth instar the mas e is mad e up of many embryos, we L'l ad- va.nced in develo]?ment, and the embryos are ready to escape from the polyeffi0ryonal masS into the body cavity of the host l.3.rva. -.•.�------��----�------�- 1 - 176 - From this point 011, the eraqryo.::; ne e d not d rav upon the host for nourishment, since the o rg ine of the embryo are differentiated. _tfter this differentiation has occurred the new'Ly formed larv3. is far enough ad vanc e d to escape from the po Lye mb.ryo na.L mas s and to take its nourishment directly in the form of blood pl�sm. Thus, the trophamEion and par anuc Le ua are d Lacar de d as far as their m.t rit Lve funct-ion is concerned, no f urt he r use in this line being ne ce ss iry. Hov:ever, the;iT are invar iably re t a Lne d as a so rt t of protecting envelope. Up to this e t a>e of t r,e pa rae Lt e s development, these two s t ruct ur e e have been of the utmost r t anc e in the me ana of nourishment Lmpo , being transferring from the host to.the developing germs, morulae and embryos. The interembr.lQ,nal matrix�_ In the p re ce d i.ng dis cussion, frey_ut::nt reference hs been made to the trophammion and the pa ranuc l eue and the relation they maintain with the developin� embryos. Leiby showed that the parasite egg in its early cleava�e stages is initally surrounded by mesenchyma.tous elements and that the oo Lar region is d iff erent ia ted from the host embryo, so that in it, paranuclear masses a.r e :produced v:hich are transferred d Lr e c t Iy to the emb ryo n Lc cyt o p Laem by the t zo phammi.o n, In the early stages this process allowed for a limited but sufficient amount of nourishment for the embryonic region, but in the spring, the rapid grov,th of the parasite body demands more nJurishment and thus the germs are formed. � .. iIit"'_.I!_�--z,,_... .,. ·L'-"""""" .....iiiiiiiiiiiiO�.....iiiiiiiiiiiiiiii;;;;;;;:;;=;;;;;;;;;;.;.;;;._-. ... IIIIIII _ , .h.t. 2 1 lilt Plate VII: part polye.mbryonic chain showing a large number ot po(Iiygenns.ot(a2)section a throUgh polygenn showing 6 germsjno segmentation to to� morulas has yet occurred. ht-host adipose tissue fll=ir�hamn1on. - 178 - fd.t-tissue be tv e e u the eu'uryos is eo mewhat d Ls vo rt e d ; Leiby states that in Copidosoma this interembryonal oatrix is composed of blood plas� fat cells and leucocytes, all of wh Lch appear to be in a condition whereby they may be best assimil�ted by the trophammion surrounding each em bryo. This matri� must be con�ered closely ar�lagous to the YOlK of monembryonic insects. It is quite probable tha t host t racheae would be abundantly present in such a ma.trix. Pseudog:erms, pseudomorulas, pseudoembryos and pseudo larv�. The process of formation of germs by the gouping of embryonic nuclei is app�rently by no means perfect. The result is that certain germs are imperfectly formed. Certain germs and embryos are therefore to be eeen in the v�riJUS stages, which cease development which the others develop normally. Finally, certain larvae fail to �eep pace·in their development with the normal larvae, and are therefore destined to pe r Lsh, Leiby calls these " forms "pee ud o Lar vae while Silvester uses the term "asexual I, la.rvae • These abortive larvde are commonly met with when ... p�r3.sitized host larvae ate dissected· at a time when the larva is host larvae appear inflated. If a parasitized be opened at this t Lme numerous parasit ic larvae will I""--��------ - 179 - fJund f1o�ting in the body fluid. � cl�se study of these forme ii, ill sh ow many po o rly d e ve Lo pe d larvae'. These asexual larvae are easily distinguished since they are usually smaller; the integument is bare; from general studies of sections nothing definite could be learned about the reproductive system, although Leiby reports it absent in Can�disoma, as �ell as the heart and malpighian tubules. According to Silvestri, to every 1900 normal larvae , produced in Litomasll! there are over 100 asexual ones. These larvae in this species had very str�ng mandibles and were probably useful in brea�ing down the tissues of the host for the consumption of their relatives, but they soon degenerate and never attain the adult stage. Silvestri calb this phe nome no n "precocious development of caste". 1',Tith Couodisoma. gelechiae, Leiby estimated that an ave rage of . ore e e a.ne 22� per cent • were peeudIdQ ar vae an v e re there f d t· d to perish and be devoured by the normal larVae. In Bere cyntus not more than 15 of these abo r't Lve forms have been f�und in anyone parasitized larva, and the average vLl L probably be smalle r than this number. Abortive embryos occur in the development of many different species of both invertebrates and vertebrates. It is difficult to assign any definite cauee to the de- generation of these embryos although it has probably some- -, . - 180 - thing to do with nutrition. In some cases it seems to be due to the f�ct that the division of the eg� r�s been carried too far. In other cases, degeneration is appar ently due to lack. of l"\roper n..t rit ro n, Uost of the poly- germs are early surrounded by a thick. layer of adipose tissue, upon which the ea.rly deve1op .. lent of the embryos depends. But other polygerms are almost if not entirely barren of adipose tissue and it is an observed fact that the mo rtaI Lt y of embryos in such cases is exceedingly high. Silvestri'.s statement that the asexual larvae of Litornastix aid in the brea k Lnc do vn of the host tissue is very difficult to understand, especially Since the sexual larvae are able by their o�n eadeavours to break. do�n and aae i mii.at e the t issue of the host. It seems Lnrpzo bab Le that a single species should have d e ve'l o pe d a.. peculi3.r sexless and moribund larva. for the particular purpose. In both Co"Oodisoma. and BerecY'ntus the abortive larvae a re not capable of that purpose, in that they have pract icall�' no mandibles. From experimental studies, it was found that asexual La rv re develop irrespective of whether or rut the parasite egg has, been fertilized. Leiby says that the first sign of these l.::l.rvae appears in young polygerms be t ve en 75 to 80 hours old. First, the asexual embryos is composed of a lar!e number of relatively small, closely packed cells, - -181 - whe r as e a primary mase is composed of a f ew La r ge cells, show no de f Ln i, te a rr an=erce nt e. nd the inner Lng Seco Ly , of the tv-o envelopes surrounding the embryo is decidedly t h t cke r than the corresponding ueurbr me of the primary mass. This fact is more evident in advanced sta?es. \lhile some p o Lyge rms produce aa e xua l emb ryo s at a Yery early stage, nevertheless the rnjority of such embryos dJ not aope ar until after the disassociation has t aue n p.lace. The p ro d uc t Lo n of e exual larvae ill a given polygerm is not confined to a Single pe r Lo d of development, but is a co nt Lnuous process, extending over a period of about two we e k a , The method of de ve Lopme nt in .une r Lcan species is very simil..ir to thai; of Litomastix. .icco rd tng to Leiby, the rr�in difference is that they fre�uelltly app- ear in groups, instead of arising singly from cert�in of the secondary masses. �ftet a ptriod of multinlication of es the of the contains secondary maes , body cayity host many gro�ps of asexual embryos, as well as groups of young asexuat larvae. There are also found free asexual .Ia rvae and some on the point of being set free from the capeu'l e , . The fre':i_uent appearance of asexual embryos or lar- vae in g ro i.pe suggests that, Li Le the s e xua.L erlbryos,. the ind i v i du al e of a group have a common origin, pro bab Ly aris- Lng thro u rh the d ivis iO:1 of a s Ln zl e se co ndary mase . Very careful dissection of a number of parasiti�ed - 182 -_ larvae reared in the laboratory gave the follo�in� d�ta 011 '::I..:;. exuat embryos arid larvae: -. ( the female para.site v7J.S a Llor.e cl to oviposit only once in each egg) 10 d3.Ys old - 4 free larvae, 3 embryos 12 d aya old - 2 free larvae, 1 embryo? 14 old - 3 no e days degenerat ing larvae, embryo • 15 old - 5 5 e r,l days free larl7ae, br�ro s • - .) 15 days old 4 free la.rvae, ... de ge nc rat Lng larvat; 17 days old - 1 degenerating larva. 18 da.ys old - no asexual la.rvae or embryos. This data Lnd Lert e e tha.t not more than 10 to 12 as exual La rvae are pro d uced by a s t n=Le egg. Tr.e:l also show the period in which the larvae are developed. The younge�t stage containing free larv3.e �as ten days old, the oldes" 17 days. The abortive -arvae have all degener- ated by the 18th day. These larva.e anparently do not live more than a very fev,- days a.fter being liberated in the body cavity, probably one to four days. Free larvae first appe a r on the tenth day and degenerating ones on the fourteenth. The last larvae escape from their e nve l cpee on the 15th day and' have d Laappe a r ed by the 18th da.;-" .. T'Tithout exception, al::' asexual Larvae degenerate. The be g Lnn tnj of this degel.ieration is mar ke d l.l;)1 a s l.o rt e n Lng then b Ll,e an'] t r.Le t Lng of the bo dy , The IJ.rvae become Lrnno - 183 J.l1U soon d Le Lnt e g rat e , Insofar 3.2 can be seeny they pe r ro ru no funct 10�1. There is auao Lut e Ly no evidence t h.rt e o they br a-, d r.n the Lost t issue in pr ep.i r at ion for Lt e .:.t..::.sil11il3.tion bJ the sexual Lar vae , They disJ.I'�eJ.l' at le.:.t.2t teu befJre the days sexual larvae are set free • .\.s s e d t at in the a uo ve discussion, the .i e e xual larv3.e develo) in adv3.nce of the sexua.l l�rv3.e. This is a no t he r vit..l.l f ac t o r in the d e ve Lo pme nt of the .large number of ad ults, because v.'ere the asexual larvae capable of completing their developldE;nt, they v o ul d destroy the host whe n it is only one-third or c ne=hal.f g rov n, ��s cl. re2ult' t he r e vro u'Ld become matured only 15 - 20 par ae Lt e s e in a brood c apabt of d e ve Lo p Lng nea.rly one t ho ue and , �Uso if t he e e a s exua'l Larvae could aucc e e d in t ali Lng fJ�ld . and tU11 it rro a r t ha such t;:; 0 uld be ill). ing , uld appe t a.dul about ten t Lr..e e the size of the e exual adults. . The Iarvae. In the early fifth instal' the p:1rasit Lz e d host 13.rva€ ..ire sli:htly sTIollen and larger tha.n unparasitized re Ld Lar vae , The embryos or ne-v:ly-forL1ed Lar-vae a he to gethE;r in one nnes by adipose tissue. Usually the inside of this hJ.s been partially consum�d by the developing enb ryo s , i�'hen the embryos are completely developed, 1,he;/ strai�hten out and are in a condition ready to be S8t free into the s'enEral body cavity of the ho s t larva. \ 3 Plate VIII: (I)1'ree larvae showing great variation in size Of larvae from a single host larva; all drawn to the same scale.First 4 or 5 of these would have been developed to maturity but such·larvae as those illustrated in F to H degenerate. Most cammon types of degenerating embryos are small oval-shaped masses(G,H) (2�larva at time of liberation into general bOdy cavity of host. larvae bottom (3)toptwo-asexual x8S; four-degenerating asexual larvae x8S. - 184 - J. . re fJUlld Lu the Jf the hcs t az' first boLly , they a.p pe in outline. The hove ve .. r i c shrivelled �€-�Lelltatiol.l., , distinct. lot t Lue of Li.Le r a t Lo u the l·;,l'vae me,j.- quite , sure 3.ppro�:iLi3.tely 0.40 rnLl:: 0.25 llll". d"'u:r the p:l.l�9.cite La r v ae have il1:"},e.sted their first f e e d on the blood of the host, the diet inct e e rtue nt at ion disap��e3.red 3.11d the Lirv 1.e a r e found mlgr3.t ing t ha ougho ut "t]-_e who Le bod;,r CJ. v- ity • • Lir is r Ld t a be 6Y.- The va.L grov;th vej:�r ap , and is t pected insomuch as the par ae Lt e e d eve j o pme nt is EO c l os c l�r associated to that of its host ('l:hich at this t Lie Is groviing ver�' rJ."f\idlY). �oon after their disTIersal Lnt o the body of the host, t l;e Lnt e e t t ne is found to be e o r rounded e qua l l.y on all sides bJ the La rvae , It Lc at this e t a.re th.J.t they co nrnence to f e e d upo n the f3.t-bod�' and the mus c Le s of the ho s t , In examining the sections r.ade of the 11rva.e, in the n Ld Lnt e e t Lne e an be found p1.rts of the f.3.t body auG muscles rihiuh 11.3..\'e been chere d into bits by the man d Lb Le e , In a short time t he miu- intestine becoQss very swollen, since it serves as a reservoir for the inJested fJod. In rome"larv�e, tLis org.111 Ius been so dilJ.ted thJ.'G its;, 3..118 v.exe almoct touchi�� the body �all of the p.1rasite. In the sixth inet 3.r 0 f t he hoe t. t l�e p� r 3.2 it e llrvae L,e aeure doli 0 ut ...L - 18::) - 1.5 :t: D.!) m:a. Tr_eir p r-e se nce o aue e s a decidedly ill- fLtted a):je3.rallc\:;' to t he host. P .l.rasit ized l..trvae are usuall�; three to eight mru , Lo nge r than no rrna I ones. The int e r Io r of the pa r ae Lt ic larva is nnde cpa que by a large number of f.lt cells; one C).l1 also o be e r ve by tr3.nspJ.rence the Lar ge stomach or mid-intest ine, vh Lch appears to be filled with a ye Ll ovLeh colored substance, h:11f liquid. If p...i.ra,sit ized sixth Lns t a r host Li r va e are dis sected, numerous parasite larvae are found to be floatillg in the body fluid. � close study reveals the fact thdt part of these are poorlJ developed, being the pseudolarv�e previousl.l mentioned. In the Lat e sixth and early seventh instar the parasitic larvae have increased in size to such v an extent t h a t they are almost in contact Lt h one anot1:er and e rcv.d the ho s t 1 ar va so t}:-I. tit is g re a tly dis t e nd eu • i.t first the Lar-vae sat isfy themselves by suck i.ng the lymph and the blood and their mandibles are not used ovn on the vita.l host o rg ane until such a time as the ir development is completed. Examinin� the host in the it is seen that most of the blood has been fifth Lnsta.r , devoured. In the early sixth instar the adipose tissue is consumed, since very· little of it can be found except . of the The muscles in a f ev: seattered isoli.ted Tl3.rts host. .lnll glands are then at t ac ke d and shortly the larval in- testine of the host coll.lpses. �t this st3.�e, the host ·�""""""""""""""""""""""""""""�....------....a -, - - 186 Lar va ar:r:'eJ.1.'c stupid '.:l.nd Lnac t i ve in e ont r as t v It h its previous acceler�ted �ctivity: The e e xua'l La rva e wh Lc h vLl.L produce a.dlllts· ma t ur e a tmu'Lt aneo ue Ly and irreguLirly arrange themselves vLt h Ln the host. This is e omewh rt in contrast to their disposal in Conidosoma �here they are arranged more or less parallel to the Lo ng Lt ud i.na L axis of the ho e t larvae. Pr'e vt o ue to tllis arrangement, the La rvae have entirely co nsume« the body contents except for some muscles, the a Lfme nt az-y canal and the host tracheae. They have also devoured all the pseudoembryos and'poorly �evelopeJ parasite l�rvae. The part of the body contents v:hich r'ema i n, together v;ith the outer loose layer of. cuticula of each parasite (in reality a moulted s�in), harden and form a chamber a.round each larva. In this chamber the larva remains quiet for about ten days. During this prepupal period, a rupture tales plCice in the uel's dividing the mid and hind intestine, thus a l l.ow Lng a co nt inuous pa aeagev ay in the alimentary canal whiuh heretofore alS been abstructed by these cells. This allows defecation in the form of small round pellets in the base 0 f the chamber. In t he meant ime, t he body vall of the host also hardens anu the larva a-ppears a mum mified carcass composed of many chambers each containil� a parasite larva • larvae do • »r the \s discussed ev toue Ly , parasitic not injure the host uct LL full larval size of host is - 1,-7 - 1 re�chea, ttu3 ftr�ishln: LLe :rc�tect �u��it� of fJoJ to t he d Lc t Luc t 3.uva.l.�ta:e of the Ia.L'asit.es. R�TArrIJlTs:rIP B:�TFI�:S:f �P�CIFIC PO::'YE:.J3RYO:TY O� P_�R.J:- 3ITld HYl.2:IO:PT��� VTD �!p�RII�nT_\'I. PO��MBRYO::Y. the e rrpe r Lme nt a I PJlyembr;;J1J.;t-r fou ud in the eg:s of differ- ent animals (!j<..:hiJ:.oderr.1s, Fishes, �mphibialls), for in isolating artificlaily the blastomeres of the eggs (blasto tomy), each b La s t o me r e or each group of isolated b Lae t orae i-e s . co 11St it ut es eI:1 s c e e e e an bryo it el f. In ]Jot:r... a , the g-s ehove a perfect Le o t r op Le m, t ha t is, e ae h of the parts co nt a Lne the he r e d Lt a ry factors co rap l.e t e Ly c rpab'Le of bringing abort the formation of an individual of the Same k Lnd , But in eXD�rimental blastotoI:1Y this reproductive faculty is often limited by insufficient supply of nutritive aube t ance e necessary to complete the process; the individuals thus pro- d uce d t , being in these conditions smaller han no rma L illdivid- ua Le it e r a l t o t o is , results that for xpe Lmerrt l b ae my t:tere Lmpo sed a nrl n i.nnuu 0 f si z e fo r the ae gment e re e ult in:! fro m the pa�cellin� of the egg, a minimum beyond �hich the process leading to the multiplication of the individual ceases to be realized or to be rea.lized in an imperfect manner. On the ot he'z- ha.nd in ill. tural found in , polyembryonY Eyrueno p- �L ,,� __ �..----�------� ------, - 188 - terous parasites, the question of mtnt uum size is not very import�nt �nd the p.rasite egg h�s at its disposal in the o r g.an Lc liquids of the host wher-e it is found, all the elements necessJ.ry to allo\ completion of the different parts result Lng from the d Lee o c La t Lo n of the egg; also this dissociation can be carried to' the extreme and leads to the form�tion af many perfectly formed individuals. ::tEL_t:i: 10 �� SH IF I3ET-r'EEU SP�C IF Ie .:.)0 LYE1JBYO lIT 0 F RYlv�HOPTS�U. _11m OT�R :METHODS OF .d.G.UIIC F-E PRODUC j_1 ION. Agam Lc reproduction Dr agamogenesis can be mani fested in insects in different stages of ontogeny. Speci,fic polyembryony of Hymenoptera is t he expression corresponding to the most precocious stage. Then follo�s pedogenesis (parthenog-enct ical pr-o ge ne e Ls ) wh i.e h ha s the qu�lity possessed by certain larvae of producing in the interior of their body young lar� (Cecidomyiid, belong in� to the genus Mias�or, and certain species of Chiron omids); final �y, one of the last e tage s is found in the agamogenesis of Pucerons in which rmture individuals pro duce in their ovaries neV7 individuals (regUlar cyclical parthenogenesis). ze ne e is The gap e e pi rat Lng germinogony from pe do is very large and is represent ed by a succession of all the J� _____ 1 - 189 - c Lae e ee o f embryonic de ve Lo pme nt and of Lar va l, evolution. If. aman� insects. we cannot find any intermediate made of reproduction uet�een these twa pracesses, there can be often faund in the an irnaI kingdom examples af agamic re productian intervening betTleen the stages mare or �ess precacious af embryonic development. Among the cy Lco e t o ma t o ue BryoJTaa, Ear'me r found a buddin� taking place in the eg� at the beginning of dev- e Lo pme nt , The egg segmented in such a v:ay as to give a po�ymarular mass or 0rimitive embryo w:t.ich sends out dig-. itiform extensions sep�rating secondary embryos at their ext remit Le s by bud ding; but, in the genus Licheno rora, this budding. is replaced .from the beginning by the dis sociation of the primary embryo itself into a l3.rge num ber of secondary embryos; there is here a phenc�enon quite cornpa ral.Le to th rt found in p ir as it ic hymenopt era. the vor ms the Among , Kleinenberg: described curious case of Lumbricus tra�ezoides, in T-hich the egg develops Lnt o two embryos; here multiplication is accomplished by internal bud d Lng in the g is t ru.Ia sta':3-'e. In the Tunicates, Dinlosoma offers a curious case of precocious budding of the embryo uhich presents the illusion of simultaneous formation of two embryos in the �same €g'g; hoy/ever, it is simply a case of internal budding of an embryo already d L tlferent La t e d , - · J.1I7-"''4w� '._ ,iio...... -.J _ 1 - - 1�1 A,g-aIno;2'enesis (after I.Iarcha1) �xternal Internal �mbryonic fission of Specific polyembryony of p3.ra- LichenoEora Bryo z o a , sit ic hymenoptera. Pedo genes is of larvae of certa in Cec idomyiids (Miastor) Budding in a1 t e rna toe P 3.rtheno g-enes is Pucerons)cyclical n gene rat iOI18 of Dol io- Cy ...ip Ld s )parthenogene- " um. Budding in 301 ter- Rot ifers) sis. - nat e ze ne r a t .o ne of "�nnelids • Budding producing in- Part he no gent s is of bees. and dividuals ent irely e- of Tenthredinids. v» 1 ved. In comp�ring the amnio� of Encyrtus �nd those of other p iras Lt ic hymenoptera developing mono- or polyembry onically with the plasmodium of the Orthonectides or wi.t h the sporocyst of the Trematodes, the question arises if this amnion is not re.;.Lly a true placenta. It is certain tha t if the 1 ife cycle 0 f an Orthonect id v; ith an unisexual plasmodium such as Rho1J..:.l.t� ophiocoma.e is compared vLt h that of Encyrim or Copidosom.§_, it wo ul d be Io'asier to cor relate the plasQodium of the first v.ith the trophamnion of the secor...d. Hov:ever, it is bel Le ve d t hat such an analogy is very super�icial; and if something Dore nearly homo- Lo go us to the a im Lo n is desired, it is found in the prim3.ry 1 - 192 - precocious for��tioa such as the envelopes (esp�cially ext e r ual ones) r.h Le h surrounds the I hex ac ant h embryo of Teni:3.s (Saint-�emy). In bo t h cases, the fo rm.at ion cons ist s of simp' e envelopes destined to protect the cg� and especially to place it in contact vLt h the nutritive centre. Also • these envelopes can be co ns Ld e re d as a very rudimentary temporary body, vhich will be e Li.m Ln.rt e d v:hen it ha s played its ne ce c aa.ry p a r t and when the primary germ cells vhich it contains willbe differentiated into a secondary body replacing the first and v;ith secondary germ cells. .uio the r que e t Lo n is solved in comparing the poly- embry-ony of Hymenoptera lnd other types of ae;anlogenesis; t hat is, if polyembryony in insects must be co ns Ld c r e d as hav ing preceded or follov;ed phylogem:tically the active types of a gamic reproduction, such as the pedogenesis of Cecidonzyiids or the cyclical partheno�enesis of the Cym p Lds , Harmer, in wo r k Lng v,·ith the Bryozoa, arrived at the conclusion that embryor:..ic fission mus t have been OTI.;_y a consequence of the b Ls.e t o ge ne t ic fa.culty of the adults. Perrier also admits this in a general �ay for all budding animais and he attributes the a-'pearance of these pre- cocious tdPes of agamogenesis to a continuous type of he r e d Lt ary action dett...rminin� the acceleration of the eli1oryonic phenomena and to which he gives the name 0 f t �L :..hliiily.g.e.n.e.s.�•.s_;_hlllle_c�o.nsbl�.i.d.e.r.S_.i.n_p.a.r.t.i.C.U.l.a._r.t.h_e_P.O.l_y.e_n.lb_r.Y_O_I_1_i_C _ - 19J - f rou do very e ne r ye t Lc ta.ch�T?elletic .cc t Lo n, tha t i:::: re e e nt pol�relJlbryJll�'" p Jnl�' dfOradlc3.11y arid r a t : (.L' ecc e Lo na.l a., co i re d to t r,e o t t pt Ly up Le r y pe e of ac:J.l .. o- tl:..e i uve r c c , opinion vl1icL co ne Lde r e po Lye . rJollJ ...I.S J.ll he no meno n , initial L:.:l.rcha.l does not t h Lnl; it '.�'orth v h il e to co ne Ld e r any other causes that t hcs e v.h Lch ha.1,};J€11 in e r xpe Ime nt a b.Lae t o t o my , Far PJlynotus. Jot 1e.3._ t, the act ua; c aue e e e e era sufficient to e xpl a Ln the p:_enor.le .io n. It Ls 3.1-::u 1)0,-2il)le tlut OtLC1' 0_,,(:(,S "111 be fOUllJ \ LE:.L'-:; t l,e p:l t pl.:i�1',_d b, t he e e ca r c eo vL 11 be 0111" of' ceconLl.:i"L';T Lupo rt.mce , But e uch CJ.Z('C vLl.L p ro ve uo t h Lu; a�...I.iu::..t fJ.cult./ of po Lye ub ryor l,c d e ve Lo pme nt of the e:::-g h",,8 h.rd primlt IT-el�T :br d e t c rm Lnarrt e t ho ; e rhi c h ,_..re in the ex terior, causes which are found in a stliKing manner aruon� the still uns t ud Le d it s_fk,cies , may be th3.t these causes have t hrc ne r. Lo ns d Le a-r ie ar e d r or. , ugh d.da!>tat , f other 8::€cie'::, but 1.h,.1. the f.1cl,.J.lt./ of pol:yell1br;"Ji1ic develop- ha s p e r e Ls t e d , This repla.cement of p r Lrn.t Lve c:terl1c.l.l the .::_�ne ',),y _,l;:; the c aue e e �',:-cimitivt:; , is indeed a f a e t .,' ...... �------'I!! - 194 - COr.1["011 to the e vo Lut ion of eve��" t1.in:' ani] canst Lt ut e s the principle of the 13.7 of heredlty. In order to arrive at SODe definit� concllsions on this subject, it wo u l d be hig-hlJ' desirable to be in to discuss it several or a a positlon concerning gre�t . number of C.3.268 of this of d e ve nt , Hovc ve r type Lo pme , it is felt t ha t t he re i� sufficient mat e r i.al a.t hand to furnish some clues for a ratiorul explanation of ce r t a in ph�ses of polyembryony. Tva primary conditions are e s sent ia.l to pe ruu.t tr.l.l. occurrence of specific po Lye nbr-yo ny , First, an a uo c t inexhau�tible su�ply of nutritive ffi4tcria' is necessary and tr.l.is must be in the close pro:drlity of the egf! dur i.cg it� d e ve Lopme rrt , Sec;ondly, there is ne cee s a ry an adar>ta- of the first to . it to be used in tion al�ow nronortion- .- to it 8 ne e d s • The first of these co nd Lt Lo ns is realized among all rao r e Hymenopterous parasites.• The second one is true to a or less degree amo ng a good number of Proctotrypids arid Chalcids. It is often man Lf e s t er. by a t h Ln.. .ne ee or even by the disa�p�arance of the chorion �hich sep�rates the t _ L p�lo.r...... n.t_e.g.!_f.r.o.m.".t.lli1.�e_.o.r.g.,a.11.i.c_c.e.I.lt.r_e_b.e.l.o.n.g!lli.n:.:....".t.O_.t_hl\!!!!e!!!!!l!!!l .. - 195 - pa r ae I t Lc Lost. It is rnan l f'e e t ef also bJl the pr e coc t.oue aripe ar-anc e of e pe c La'l f o r mat Lo ns (trophadmion) funct ion- in� �s 3. pl3.0enta 3.nd puttin� the p�rasite in contact �ithfue nutritive centre furnished by the host. It is also evident t h a t t he e e t wo conditions nec- ess::1.ry for the bringing about of po Lye rab r yo ny are not en ough: the first (a large suPply of nutritive material) is common to all paras it ic hymenopt €l'a, and the second, 3.S bas been already indicated in the p re ce d i ng discussion, is found amo ng a large number of paras it ic hymenoptera wh Lc h develop mo no embr'yorn.catIy , other cond it ions then mnst i:0vern the real izat ion of polyembryony. l.ur(;hal found in his e t udy of Polynot� minutus, a po2.yembryonic :Proctotrypid, certain recullir and e xee p t Lo nal conditions which he-conSidered freely accounted for the appeara.nce of polyembry,ol.l.Y be ca ue e these conditions v:ere the same as those which are fOUlld for ex- par Lme nt a '. blJ.stotomy. J.. mo ng �ll the hymeno p t e ro us paras ites, which have been reviewed in literature, the eg� is placed in the g,;;:neral cavitJ' or in a thick part of the host v Ie ce ra in such a. v,ay th;l.t the nutrit ive centre with which it remains in contact is the blood 0: the animal at whose expense it lives. In mt nutus on the Polynot� , contrary, I.'hrchal found the egi! 301\ ays placed in the gas t r Lc sac of the host and t he r-e larvae). result e the fact that. v hen .. L (�ECiaomYiid .....--�� ------.... - 196 - the Joung' larva. J: Lhe p az-as Lt ized O�cidomyiid cc rame no e s to nourish itself b;l fillilli_:; its gae t r Lc SdC l.it!J. sap fron whe a t sterns. the eg� finds itself abruptly plunged into .:l. centre hiVing osmotic rroperties different to t hos e in r. hd ch it origina.lly V'd.S found. I.loreover, the production of sudden changes induced by osmotic pressure co ns t Lt ut e s ont;;' .jf the best methods to use in order to obtain sep:l.ration of b Lac t o me r e e arid t he Lr evolution into e as been e c l be distinct Lnd Lv Ldua'l , has ahovn ape La Ly the e:Joeriences of Ba't a iLl o n (1900) all the eg_s of the :'amprey (bl:3.stotorny and e xpc rimental po Iyemb ryo ny } , If.a.l'ch:3.1 found such a condition present in Pol:ynotus e r mt!: .. tus and he c o ne Ld e d it sufficient 10 explain poly embryoey. He also found for the development of the egg of Polynotus ai.o t he r p a.rt Lcul.a r condition wh ich exactly corr ee po nu e to the second process by wh Lch experimental blasto be o bt d that is nourishment. Marchal found tomy can a Lne , living Cecidomyiid La rvae in v:hose stomachs we re found one or more e g>e of pOlynotus; the p.:l.rasite eggs are tossed a oo ut lil{_e a ball from one Side of the stomach to the other. 1.Ll.rchal c ona Ld e re this cond ition very anal a go ue to the mechanism by wh i ch Driesch o bt a i ne d d Le eo c t.a t Io no f blasto- meres in the e g rs of Ours ins, vhe n he nourished t l.e ee eg?s in sea vat e r ; and further w it h the phenomenon w hf.ch Ohun o c c a s Lo na.l Ly found in nat ure whe n the eggs of Ctenophous - - 197 - are nourished and aha kan n:/ v av e e at t he t Lrne of storms. It is thus r a t Lo na.L to a dm Lt that, if the variat ions of osmotic pressure in the e�g constitutt an aid to poly embrY:HlY for Po Yllot� minutus, the no u ri ahme nt of the egg in the gae t r Lo sac must at least be considered v.;;ry favourable for the production of the phe no me no n, If the l'ea.l Lz a t ion 0 f polyembryony e e eus to be eJ�plained without any difficulty in Polynotus minutus, l.1J.rchal found tha.t the same aiding factors a.re not pre sent for the polyembryonic chalcid, Encyrtus fuscicollis. - But even here, he found a very in.portant mo d Lf Lc at Lo n taking p l a ce in the p'La.ce where the egg w.ae deposited and polyembryony manifests exactly at the time the ch:3.nge is produced. In fact, the para.site egg is deposited in a very concentrated p I ac e formed by the host eg,� (Papilionidae) in the general cavity of the embryonic t arva , Then this little Lirva hat che e during the summer and passes the rest of the sunmer, fall and all winter without any nour- the i.clternal ishment or g rowth , Ma.rch:3.l considerea that fluids wh Lch had the same concentration at hatching as of those of the tgg, must undergo during the period that it hibernation some degree of daydration. He found filled was only in 'the spring whe n the little host larva . t hat its e t cmach V7ith S3.p from trees on vhich it feeds the p.i rae Lt e egg began to ahow its enormous develcpment, - - 198 - and po Lyemb'ryo ny d e c l vr ing it se If. It seems pass i ble to state that this l�st process is c�used by the passa?e of the hos t fluids from a. r e La't Lve Ly co ne errt r-a t e d s t.rt e to one rna re diluted. An objection could be raised th�t Encyrtus fusci is not collis the o nl.y p rrae Lt e that deposits its egg in a host eg� without causing the de�th of this eg� or stopping its development. lIarchal f o und. c t h= r Similar eX3.mples but also found that Encyrtus w�s the only hymen opterous paraSite he had studied, presenting a period of dormancy at the beginning of the d e ve Lo puerrt of the paras Lt e egg and it is very interesting to note that the stage of dormancy is very long. 1hrchal believes it only m.tural to think. tha.t the .3.rrest in the development is caused by a .slight degree of dehydration and that the hydration of the blood which follows in the spring is of assistance to polyembryony. From the a oo vs d Lec ue.e Lo n it wo ul d seem that spec ific polyembLyol1Y Of hymenopterous pirasites finds its chief ae s Le t Lng factors in the actual causes which haope n in blastotomy and experimental polyembryony. This inter pretation is more strit>.in·�· if it is rememhred that the ty;O types 0 f known polyembryonic development be Lo ng to two different fami}ies of hymenopterous paraSites (Chalci d idae a nd Proctot ryp idae) and t h;t mo s t 0 f the act i ve mem- of these fJ.milies develop monoembryonically. 1!ol'eove, ..L b�S ... - 199 - not only is polyembryony not a distinguishing family cha r ac t e r in one , but or two c aae s , it does l.l.Ot con stitute a gen�ric char3.cter. Marchal discovered that some of the species.belonging to the genus Polynotus developeJ in a monoembryonic manner. This fact alone, that in the same genus Po Lyno t us] wl.Lch is more often con�idered as a subgenus of Platygaster} there is found side 'uoth by side types �f development, v o ul d appear to make entirely inadmissable �very explanation that vould not attribute to the same actual causes v'hich are found in experimental blastotomy, a preponderant determining influence in the production of natural and specific poly embryony of Hymenoptera. It might then be. concluded that the ae s Ls t Lng fac tors a r e the same in the three types of Known polyembryony, that in - is, experimental po Lyemb ryo ny , in accidental poly and embryony (true twins) specific po lyembryolZY , and that the chief cause is that which Bataillon h�s demonstrated, particularly the.determining action ,in the first two �inds of polyembryony that is, the change in osmotic pres sure at the beginning of segmentation. It mus� be thoroughly Kept in mind in connection with the previous diSCUSsion that we are dealing with hereditary factors and that any prevailing factors merely co nd LtLo n the characters already present •. Concerning �L - - 200 - it is t h at the Be r-eorntue , found tv:o pr Lraazy conditions eas e nt ia1 to permit the occurrence of po Iy ernb ryo ny (food and ability to use it) are present. It is �o seen t14t the same f.ac t o r a aid ing the appe a ranc e of polyer,lbryony in are for that is the Polyn:Jtus_ present Berecyutus, t poly embryony d e c La re s itself at the period of the ye3.r v.he n the host Lar-va commences to feed, thus bringing about a change of the osmot ic pressure La the s t omach of the host. During the winter of 1928, 100 host eggs of ��xoa ochrog3.ster we r e pI ic e d in lamp-chirnnies in v:hich there were unfertilized fem9.1e p..irasites present. These virgin parasites were liberated one b� one fr�m severa, host car cae e e e of �u..."'{oa ochrog3.:::.ter, by opening the chamber in which each had p up a t e d (the o ut e rmoe t chambers of the host), in this �ay permittin� emergence several hours be fore all the p:!r3.sites of this e ame carcass v·ould normally emerge. Of course, it was necessary to liberate a grEat many adults in order to obtain a good supply of females. E3.ch of the female p a.r ae Lt ee V:9.S p Lace d in a small glass tube and a similJ.rly Lso Lat e d male was then introduced into e a ch vial. They v:ere left unt il co pul rt ion v.ae observ�d. In mother similar experiment the femJ.le p.1ra. Sites we re removed from the host ca.r c ae s e e but v:ere Kept • q._ - 201 - removed from 1113.1e8, thus l�no7.in3" definitely that they �ere unfertilized. Theze �ere allo�ed to also oviposit in 100 ochrogaster eggs. In the above t�o experiQents e�ch host egg T.�s subjected to oviposition only once. This v: �i2. a.;:;;certai:i.1ed by look-ing for the punctures in the eg obtain frohl uo t h experiments 100 eg�s tm.t were oviposit- ed in only once, more than this number 0 f eggs r ere pla.ced in at the start; the p�raeitized eggs were removed at in tervals u�til e�ch group had one hundred parasitized eggs. From the 100 eggs allo�ed to' be parasitized by fertilized fema.les, only 78 eggs·ha.tched �nd developed through to mat ur Lty . Of this number, 55 per cent of the wer e ma Ie 38 r cent v e r e maLe e ·..lnd broods pure fe , pe pure 7 per cent �ere mixed broods. From the 100 eggs oviposit- ed in the unfertilized ftmales, only 65 reached maturity and all produced pure male broods. The rearing of pure rna Le broods from unfertilized eggs and pure fema.le broods from fertilized eg�s is vhat we would' expect. Th�sedata therefore are interesting then, because of the pure male broods reared from ferti· lized eggs, �S w e L'l a s unfert Ll.I z ed e g�s, and because mixed broods �ere reared from fertilized eggs. These con- ditions are ao mewh a t similar to those ex Ie t Lng in Copo- I s o ma d z e Le ch Lae , Up to the time trot Leiby demonstrated - 202 - th.1t male brood.:: w e re re .... red f ro u fertilized €g�s of 9...:. gelE:'chiae, Patterson's r-e c o r d of the rearing of four broods of P.1racopid;)sornoDsis r.e r-e the only definite re cords of .rc t u.a'L rea.ring of alW of the polyembryonic E�niien optera. Previously Silvestri ha d hu d e the statement that fertilized p.a ra s Lt as produce fecn.les, '7hile unfertilized produce ma'l e s , It wo u Ld seem t h a t the more or less regularity v ith vrh Lch ma.le broods are reared from fertilized females under experimental conditions is sufficient evidence to ahov a doubt beyond that fertilized females may produce either a male or female brood. Of course, t l.e re is some possib t h-rt the ility fertilized female did not a Iways deposit an egg containing a sperm, in which instance Vie simply are the duplicating experiment of rearing m3.le broods from un e��s. C. C. fertilized -- Hill in demonstratin� the ('26) .- parthenogenetic development of PlatY'2:aster hiemalis, a Hessian fly p.1rasite, observed from sectioned material th.1t an ll"pregn'3.teJ female vt Ll deposit both fertilized and un fertilized eg.:;s at a sille-Ie oviposition. On the other h:irO leiby, r o r t.Lng vi th Conidosoma., ae ce rt a i ne d as 3. part of his experimental vo r k that sperms a r e r a t he r regularl�r fO'Uild in e::!e deposited by the S J.[l1t- fertilized f emal e e in o t r e r host e�"2':;;, wh Lc h v er e e e c t Lc ne d and studied, to le::1.ID if the;/ had been inseminated. leiby is t ho rou g'hLy convinced - 203 - t h a t 3. f(...r�ilized eg� 1.1.l.�7 ?ive rise to :1 ma'l e a� v e lj as a f�n�le brood. �rm.lB�R OF �U:S�PS �l.ISRGIUG FROll rOST. During' t he e urume r s 0 f 19':;8 and 19�9 collect ions of c ut v.o rm r La vae we r e nude ill the field for the purpose of obtaining larvae parasitized by �ecyntus. Thirty larvae (fifteen e3.ch year chosen at ra�dom) of the red- b ac . ... ed c utwo rm Euxoa , ochrof!3..2ter, parasitized by t h i.e c haLc Ld v.e r e placed in cages w it h c o nd it i o ne as ne ar nor- mal as could be obtained. _tfter the ras Lt e e had t e d pa pup.i , the a r aa t Lz e d v.e r e e p i hosts p Lac e d La sep9. .r at he a+y glass vLl.ls ..-:it 11 a cloth co ver Lng o ve r the t Ope I.�o i e t t.r e V"'3.2 added .1'::' r e qu Lr-e d t o this covering. _1..2 the ad ult s er,ler;:e d t they we r e left ill he vials unt LL e ue r g e uc e v ae co np.l e t e , -\-e1'e t h e n ch ro fo rrue d nurnbe r frJjll o�.J.e They Lo , the total u e e; c c rrn , hce t Ll.TVJ. co rt e d , and the nuiube r of each d t l ne d TLe fo Ll cv.Lng t ab l,e al.ov s t r,e r e e ul.t e o bt a Lue d frc ru t he s e CJ U :1. s in the sum.ie r e of 1928 and 1929. TAB"-_.E I - Iml.1B�� OF B��=CYlrTUS B.iL�RI I.DU:'TS EI,CRGIUG 19::>8 �\.!m 192� (numbe1'S 1-15 inclusive are from 1928; numbers IG-30 in- clude thE 1929 data. 1. G70 - M. 3. 1105 5. 959 looe - F. 6. 4. 835 - F. 1200 - .304 - .) 7. 10G5 - 15. 9SlJ 1.�...... OJ. 1.365 b. 3G5 IG. 914 - 1:. 24. 1101 9. 432 - o r. 1'" U. 17. 9<]7 - ",OJ. 82!) .L •• �� 10. 5')'"..... ;) 18. - 10G5 F. .... 0. 945 - 11. 673 I': • ''r 19. 788 - .37. 6�')0 ...... 1')..... 578 20. 8�J - 28. 6EO 1: • --, - 1.3. 7"-'U F. 21. 754 - I.!. 29. 7'20 0') ...... 14. 599 1002 - 30. 910 F • l.lillirLum - 385 Ave rage - 840 Ihximnn - 1285. The aver�ge emergence from the fifteen:larv�e collected ill 192.8 was 795; the ave r a.;e number in 1929 and �ere 1929, 5 pure fem�le broods, 9 �ere pure male b ro o d a .a nd 16 v.e r e mixed broods, t h at is, co nt a Ln l ng both m�les �nd females. Of course, it is r��lized tL:3.t a ouch be f c r e being ab l e to s t at e vLt h any degree of aCGUl'lC:») the rel�tive proportiou of the various linde of broods. TLe five pure fem;;:.le brooa� h3.ve 1006, 885, 755, 1085 :3.nd 910 individu�18 respectively, giving 3.n aver�?e of 905 a du l t e 'Per brood. This .3.';e1'3.,;e is a l.mo e t one hundreJ g re.rt e r t hari the .;I.vera.;::-e of t he t:r�irtJ' 'urJoos .... - 205 - The nine p�re n�le broJdz h�J G70, 675, S93, 914, 754, 825, 682 and �Z5 individuals respectively, ?iving an ave r a je of 755 adults per broods. This d,ver:.i::;e is a.bout 85 less than the total average tJr tLe 30 La rvae . It is the m Lze d b r o o de v.h i.ch are of special interest. The number of individuals in these broods varieB from 385 to 1285, �ith an aver�ge of 88& individuals per brood. The .percenta:se of r;'ales varies from 71.9'J� to 0.28�'�. In Table II are listed 16 mixeJ broods, inwhieh the sex of each in- d Lv Ld ua.l has been carefully determined. Thet;t: sr e a rrange; in the order of percent of males, from the highest to the lowest. T A.B I.E II MIXED B300DS OF BER�CYDTUS. Brood No. of Individuals Females I.1ales % of lJ:ales --- I 1105 310 795 71.9 .,) � 989 465 5�4 52.9 3 1206 808 398 33.0 4 1085 814 271 24.8 5 365 331 54 14.0 6 525 467 �8 11.0 7 578 532 46 7.9 8 599 563 36 6.0 9 977 928 49 5.0 '7.':> 10 788 756 D� 4.0 11 833 808 25 3.0 12 1092 1065 27 2.5 13 1265 1260 25. 2.0 14 1101 1085 16 1.4 15 945 935 10 1.05 16 720 718 2 0.28 L - 206 - The3e dat). r a Le e 3. number of interesting- points, only a few of w hi.ch can be c o nsi d e r e d here. One of the most stril,_ing features i8 the re Lat Lve aca.r c it y of pure broods. 53 cent ':Jf the e per broods xan Lne d r.e r e mixed bro o de , Of the pure broods, 30 per cent wer e pur e ma Le broods and 17 per cent were »ur e female broods. .rno t he r e qual.Ly strikin8" feature is the great preponderance of females in the mixed broods. In only t�o C9.ses out of the 16 mixed broods ex amined is the numbe r of ma l e e in e xce ae of the numbe t: of females (broods 1 and 2, Table II). Reference to Table II will S:r�OVi th'3.t in over 62 per cent of the mixed broods less than ten per cent of the in in Lve n brood are ma Lee v:hile dividuals any g , in 37.5 pe r cent of the cases there is less then three per cent of males. The most striKing case is th�t listed at the bottom of t he table, in which only two males were present. In v: ith cc ud Lt Lo n broods 'TN a nurube r eo nt rd.ct this , ith large of lrules and 3. few females have so far not been fo und , As stated above, in o nLy t wo cases out of siXteen are males in excess of females, and even in these the small�st number of females is almost as much as 28 per cent of the entire brood (brood I) The fact t�t fem:lles are so frequently in e�cess of males in the broods is Ofte of the most sig llificant points brought to light iJ.1 the study of the ee z- rJ.t io s • ------�------� - 207 - Of the t o t al number of individuals found in the e d v.e r e 30 broods xamj.ne , 64.2;� v:ere females and 35.8)� ma.les. TRS OR IG nr 0 F I.IIX�;) BROODS In :!?O:'TII.:::BRYO �TIC FYI.lEIJOPT:S::L\. The opinion prevails gen�r�lly amon� students of the polyembryonic Hymenoptera that an egg deposited by a fertilized f�m3.1e gives rise to a pure femal� brood of pJ.rasites, that one deposited by an unfertilized fE:;male develops a male brood, and.th�t a mixed brood (male and fem�le) is produceJ from eg�s deposited by a fertilized and unfert ilized female v:hen they happen to o vi.po e it in the same egg. .d..m.ong those who have advanced evidence in supnort of this gell(..ral theory are Bugnion ('92) and 1!.1rchal '04, in .Vteniaspis fuscicollis; Silvestri '06 in .=1t·;:,mastix trnncatellum; and Leiby and Hill '�3 and '24 in Pl3.tyg�ster hiem3.1is and R.!. verna.l is. Patterson '19 is inclined to believe t hat both males a nd femJ.les can and do develop froD a s Lng Le fert ilized e �g and that males only are p ro duc ed from unfert- ilized eggs. He is of the �pinion that mixed broods cannot be completely explained b;:l the tVio-egg hypothesis. �uch dizygote broods do exist but he claims there are tv-o ob- e t ae Le e v.h Lch come under the app Li.ca t Lo n of this hypothes:Ls:- 1. The individuals of a mixed brood emerg� simult�neously; t l.e r e f o r e it is ne e , ce e .ry to mal,e the ae aump t Lo n that - 208 - pa ras Lt e eg�s JIe a Lwaye laid at the same time. 2. In the gre�t majoritJ of mixed broods, fefuales far outnumber the males. It must be �ssumed then, that an unfertilized egg in the presence of' a fertilized e�g is inhibited from fully developing, because vhen laid alone in the host egg it produces a pure male brood, in most cases nearly as ld.rge as a pure female brood. Patterson ad vanc e d a h:.lpothesis e xpLa Ln Lng the development of mixed broods, basing it largely on the possibility of the abnormal behaviour of the chromosomes. He believes that the mned broods may be 3.ccounted for on a monozygotic basis by the Bridges method of somatic non disjunction of the sex chromosomes in an originally fert ilized egg; in which as the cleavage nucleus divides, cer tain of the bl�stomeres (cleava�e or embryonic nuclei) would receive on:y a single X chromosome, and each blasto mere so affected becomes thereupon the progenitor of one or 1L10re maLe embryos. The number of male embryos vo ul.d depend upon the time during cleavage that the abnormal be e e ha vio ur of the sex chromosomes to 0 k pL.J.ce. If, for xampl , t he be hav Lo ur v e.r e abnormal at, the first division, the resultant brood vould be composed of male and female adults in more or less equal numbers. If the abnormal behaviour of the sex chromosomes occurred relatively La t e in the diviSion period of the blastomeres, the brood of adults .... - 209 - developed therefrom vo u Ld be pre po nd e r.ant Ly female f.ith only a fe� m�le adults. _-\, simpler exp Lanat Lo n of the .rbo ve theory is as follows:- j. single fertilized egg ill..iY give rise to a few males as well 3.S many f ema.l ee , If Copidosom3. conforms to the g e ne r a l scheme for sex determination in insects, the fem3.les must have 2 X chromosomes and the ma'l e s the e LngLe X chromosome. Ordinarily during the precess of cleJ.va�e, all of the chromosomes Ln the fertilized e�g divide equally, so that all of thenuclei entering into the form ation of the will i. chromo ao me t embryos c8.rr;,: he n , l.us producing a brood of females. But if during the early development of the egg it should happen that the t vo X chromo somes in one or more cleavages should not diVide, but sep�r3.te, one going to each pole of the spind'e, each daughter nucleus r.o u.l d then rece ive a single X chromosome. If such nucleus l�ter divided in the typic.1l �ay aud gave rise to embryos, 'such 'Ii/ould be males. Patterson believes that this t he o rr will uc co unt for the development of most or many of the mix�d broods found in llJ.ture in various species of polyembryonic hymenoptera. In some later data ('17a) presented by Patterson on o ro od e of P'3.r,3,co-pidosorJpsiS, he finds that about 85 per cent of the bro o d a a r e mixed. He e t at es that this is only possible ..is a re eult of four different types of be- i o ux hav of the sex chromosomes in the c l e avage divisior�s. �.,� a."""""""""""""""BW""""" � .. - 210 - ' P3, t t e r s o n S sUI:es t i,o n is e uppo rt e d fro m an e xpe rille nt ell stand�oint by the re�ring of one mixed brood from a host egg p�rasitizeJ by a fertilized female. This brood ras composed of �,096 indiviauals. of vhich 1889 �e:e females and 207 v;ere males and Lnd Lcat e s that a fertilized egg can produce a mixed brood. ITo cytological evideu0e in sunport of the Patterson theory has as yet been presented. It appe�rs on the con trary, that there: ar e now sufficient data and observations available to substantiate the earlier belief th:lt mixed broods are of dizy�otic origin: that females are produced from fertilized eggs and that males are the product of unfertilized eggs. SID,'IT,1..iBY ..um DISCUSSIO�r JF SEX R..\.TIOS IU OTHER PO:SY�I,IBRYO:nC HYICEnOPT=R�!.. In Copi,d.osoma. gelechi.3.e and Platygaster vernalis. only a small percentage o'f the brood are represented by both sexes. Both species Lt kewLae d e pce Lt as a rule only of the ovi a single egg in the host egg at each insertion positor. Furthermore, it is rare that the egg will be at ta c ke d by tv.-a difft::rent females under m.tural conditions. The writer believes that the occurrence of mixed broods in these two species is thus readily accounted for by the tlo egg hypothesis. In Copidosoma truncell.3.tum mO�t of the broods are - 211 - mixed. '".7ith this ep e c Le s in approximately half of the ovipositions the female deposits tv,'O eg�s at a time. The sarond egg of a pair deposited by a. fecundated female is without a sperm in a.t Le as t 80 pe r cent of the instances. Ilo r e than one p rras Lt e will oviposit in a host egg under natural conditions. Reference to the d�ta of Patterson on the sex ratio of this species ahowe that in t-i.O CJ.SE:S out of 60 there �iS but one m�le reared in eJ.c� of tvo mixed broods thJ.t contained 912 and 1550 individuals res pectively. In a few other mixed broods the number of males is small in proportion to the females. In the opinion of the writer the o c cur renc e of a single male in a mixed brood is due to a mo no embryo n tc d evel o ome nt of the second of tv-a eggs deposited b� i fertilized female that did not contain a sperm. If this condition prevailS, �8 h�s "been proved f:)r P1'3.tYR':lster hiem3.1is, it is also not unlil:.ely t hat an ullfertilized egg m3.Y be inhibited in t he co mp l e t e c re avage of its embryonic nuclei to the end t h a t on1;7 a small num her of ma'Le s originate from such au egg, rhe reae a larger number v;ould be developed from the fertile egg even though it v e r e deposited by the same f'ema ·.e along v:ith the un fertilized egg. T:.That conditions govern the production of the proportionate number of m3.1es 3.nd females in a mixed brood of C. truncell3.tum is larg�ly a matter of conjecture. The great nWJber of individu3.ls (1000 to 2500) reared from - 212 - a si��le ho�t and the f��t th�t the species is the most ape c La'l Lz e d of the pol./emrJryonic h,/menoptEra., make .... <.:011- clusions on the origin of m Ixe d broods arrpe a.r perplexing. F.3.ctors wh Leh mu .... t be c ane Id e red and of vh.lch little is definitely kno�n are:- the origin of the so-called asexual lJ.rvae, the subdivision of the polygerms into ema Ll.e r e e c o nd a ry and tertiJ.ry masses tha.t became distributed ill the body of the host, and the failure of SOQe of the smaller groups of germs to mlture p�rasites bec3.use they do not become properly Lnve e t e c w it h host t issues so trut they can cont inue the ir development. On the bas is of what is be lieved to be the origin of mixed broods in other polyellibry onic hymenoptera, the opinion is ventureJ here th3.t mixed broods in C. truncell.3.tum are of dizygotic origin. Leiby and Hill arc quite positive th�t the Qixed broods of Pl.3.tl�� hiemalis and � vernalis are of dizygotic origin. The o c cur r e uo c of ma re e in mixed broods is due to the habit of a fertilized female in depositing a cluzter of three to seven eg�s, one or tV10 of which do not become inseminated during the act 'of oviposition, v hf.Le the remaLnd e r do be co me Lne em i.nat ed , There is some evi- d e nc e bJ.2ed on a count of chromosomes to Lnd icate also that re l a.r the insemiruted eggs develop polyer.1br;yollically gu ty , though o c c a s Lo naLl.y one e36 nay develop a s Lngl e female y,-hile its t\;in component becomes ,aLJorted. The eggs t hi t cu La a sperm hav e a tendency to develop mo noembryonl ca Ll.y , -�------�------� - 213 - �\. e t ud Jf the fJ.ctare re:ponsible f'o r the origin of mixed b ro o d s in the po Lj embr'yo n.Lc h,/mel1a�tEra has r-e ve aLe d uht: sJ"rikins- f_Lct tLd.t Lns eml.nat Lo n of an egg by a fErtilized fe ma Le may 9.pp:..rently be controlled. It i� certJ.in that not all egIS deposited by a. fertilized femJ.le invariably contain sperms thohgh they �ere depJsit ed by a fertilized female and ·thouBh they �ere deposited along �ith e�gs that do contain a �perm. The belief h�s been e�pressed by �eiby and Hill (12�) "th�t in the case of P. hiemd.lis t he fem:.l:e does not control Lne em Lna t Lo n �s the honeybee apparently does, but the short period of time required by the parasite to deposit 3. groi.p of eg:s indica.tes the probability of the eggs passing the ape r m atheca duct so rapidly durin: oviposition th:l.t all of t h., eg�'s do no t receive .1 epe rm", From the present s t ud te e on Bert-cyntus, t he vr Lt e r is led t ; c o nc Lud e tha.t vLt h this species the occurrence of mixed broods can only be e xpLa i.ne d by Patterso['�ls hypothesis - t ha t is, a fertilized e::-:.s can p ro duc e Lot only females but also m:3.1e�. In over ninety per cent of the ho e t e::3'8 that have been examined, only one pa rac ite egg h.e been Lne e rtel.1 a t J. t Liae , "":ten L1Jrt: t ha.i one et:'g has been de�")osited, it is -l.uite L)ossible th:it the t vo e;_;g hypothesis e xpj a Lns the origin of lllixed broods. But this o nl.y e xp l a Lns the situation in a. 12m3-II p e rc e ut age of the - 214 - rn Lre d Ul'JJUs. J�l .c co n.rt Jf t Lt s f ... wt t l e t· J eC;: :r...�T}.Jothes12 f a.Tl c cl:..ort of e xp'l i Ln i.n.; the prE':'C'1.CC Jf mi:ea �rocdc i� BereeJ�tu�. The �riter 1�:::: reared � m L:e.J Jro:J d f ro II J. ho ; t e :I ..... Ll T,'" LicJ� it r �s knc- 1 t ha � h d o J. fc.rtilized fl...l".1:11e i v l po e Lt e d only o nce . :;?'roEl c L: v tLi2 f a t , it e Ld e n; t h i t ...I. f c rt LtLz e o e r-: c a ... l 3.11U d o , do e e p ro uc e j,IL:e'j br o c c �LeTe h .. s ..ilEO 'been re_"l'ecJ f ro n host e:�2, in v h i c h a ferJ.ile [:.n07'1 "�O be fel�till:::E:u ' o e e d .J. e o o , h i.s vf oo Lt , p'lrt: nal br d It is .ir/�J1reni. froLl IJ e .' " t h i.c t ha t he Lne e rt e� __ 8 IDt l.t1.�erJbEdeci, .md re sulted in � pure D..i'e �rood. I� vier of the ..ibove J.nJ other 2tudle� it l� f�lt th:J.t Lns em i zi.it Lo n 01" La c l, o f it b,/ fl....ctiliz�J feLL.lcc is controlled in the po Lyemb ryo m.c hyne no pt c r a , ZinCe O.L1101 Ill.J.lt:S a r e pl"oduced f ro u u�J€rtilized eg�'c a nd since the s pe c L e C.111 o bvLous ty be co nt i.nue d 0:11J 'J';l the production e m Le e the of of f i , control LlsemLl1tiolll:ould be, und is, of advJ.l.Lt.J._!e to the adult pa r cs Lt e e of 3. br o c d , It should be remembe-red t h rt the a du I t 'P.1r.l..:-.ites 3.r'€ "I:ery minute; th�t �ith fe� exceptions e�ch s�eciee is limited in its d eve Lo pme nt to a definite hce t species; t h rt the chance ::>f a minute insect fil1Clillg its p ro pe r :t.:.;:::.t egg b,..' me me o f co.at ..I.ct r ith its arrt e nn ,e are e xce ed Lngl y remote; arid ·th).t the ·rodl-;.ctiol1 of mal e s a Lo n- Y ith fe m�les (mi�ed brJoda) �ould be adV3.Dta;eous to aD end, thJ.t , •.�MM""""""a. �� ""�� - 215 - fer.ules r c u Ld h .. ve :;re..l.tl...r c hanc e e of be co m ing fe:rtilized v he n L1J.le.::: a r e co-developed 'J.nd issue f ro rn t l,e e an.e hce t 3.1J1'1: vLt h the f erna.Le e , 8011t:l.':)� of insemim.tion u� t l;e fertilized f e rna l.e par ae i.t e TanIa t he r-e f o z-e be of a decided ).dv3.Dtage ia the nerpetu.3.tion of the species. r no Development bJ, po Lyemb yc ny , itself, h re t rt n> to do v:ith the ac t ual d e t e rrn Lna t Lo n of the e e x of the e gg , Polyenbryonic d e ve Lo pue nt nay obtain, no rutter vhether the e�:I is fertilized or uo t or r he t he r it is d e s t i.ne d to bring forth a progeny of unise-..:ual or one of b Le e xual individuals. Fa.ct$ of I)Jl.i'Hl1bryo ic d e ve Lo pnerrt a d d rat Le r corroboration to of t o - strong that cy Lo gy , thJ.t sex po t e nt La l.tt y of t he e3'g' is fi":ed at 3. "I':ery €-3..r1;/ s t a e of d e ve Lo rrt pme , doubtless in all ordill3.r.,' cases at the ti@e of fertilization. REJ.. RIHG ��TEOD,� One of the mo e t e s e e.rt La l points in the .stl1dy of this po Lyernbr'yo m,c p.i r aa i.t e ·r3.:.:o t he o b t a Ln Lng of a r1ent- iful supply of ptrasitized hosts for relriDg �nd for ex pc;rimel1ta�. study. Due to the minute size of the adult.::. a'nd the m Lx'ro e co p Lc e t ruc t u ree of the po Lye nb ryo u i,c dev e l.o rrt t he r e ';.).i::: a co ne Ld c r-a b'l e \,/:3,.- and 10s8 of pme , ta�e c e ar t o a )) .. rae Lt Lc rn.rt e r La L, This made it ne ce y obtain very 1.3.rg'e s upp Ly of ps.ras Lt ized ho s t e in the spring in - -_. __ .- -- .. -� , .....�------..------�-- - 216 - o r d e r to hive sufficient m.rt e rLa L t o .... tu.dy unt il the o e r i , C rm f l.LcvLn> p ng utwo La rvae v e r e or o ught in from the fiell18 Ls.t e in Ih.y urrt il the midl"le of June, depend ing upo n the species of c utr or m a.nd the we a t Le r condi- t i o ne , 1.UvJ.t:;' v:cre ha nd Le d in the f113.IDler described by I:in::- arid .�tkL:..son ('27). (It r"li;:-ht be mentioned here of th.3.t this phase the �ork ��s ntrt of a study an p�r� e Lt Lsrn of no c t u Ld luvae c a rr i.e d on.3.t the Z:.l.cll.toon ' Ent omo Lng Lcal Labo r at o ry a nd th.3.t the w r Lt e r v:a.s a b e to o ba Ln much i.nfo rmrt Lo n and a:::'l the p.:l.rasitized Berecyntus ho�ts from this �ork.) L.3.rvae pirasitized by Berecyntus become noti0e�ble in the laboratory about the ruiddle of June and are found' for a per Lo d a f about three ve e ks , Under labo r::i. t orJ co H ditions, if these larvae are left untouched, Lhe adults v;i11 eme rge in about 10 or 15 days. In order t o make it unnecessary to keep a continuous cycle of the paraSites r e a var o ue methods ve r e tried to ret .l.rd being red , t the eme r'ge nc e of the ad uLt e f ro m the host o.rr c ac ae e , The no._t successful method ��s the use of refrigcr�tion. If para- e Lt Lz e d hosts are p La c e d in mo de r a t e Ly dry soil and put in a temperature of not l:..igher than ·::0 to 450F., the 3.du.lts v:ill aot emerge. Ho�ever, should the temperature rise to o even 50 F. for as e ho rt a period as 3 to 4 days, eme r ge uce v:ill t a c e 1'11.:1.ce. The Love r limits of temperatur€- for .. - 217 - these p�rasltizea larvae �ere reaohe� at a�out the fret;3ing paint o r �b to ,�20F. Be Lev. t h Le v t e rat ur-e rrpe , the of degree mortality V.d.S fairly h i.ch and very little emer ge nc e took. place. The re3.S011 for tLis 1113.y be due to the fact th�t a large percentaze of the parasites in t he host v;ill be in the pupal stage a nd it is a veIl knor.n e nt ouo Lo g icaj fact that tilis stage is more suscept ible to injury t ha u other periods of insect life cycles. There is a'l so the possibility tint any Lurnat ur e adults wo present ul.d also be injured b�- freezing t eurpe r.rt ur-e e , This method of storing p",rJ...__it Lz e d hosts in lOT t r-a t ur e s is aa't Lc f ac t c a.Lt ho empe moderately ry , ugh Hot aLwaye , :!:ven ill temperatures whe r e no emergence has t a ke n place, other factors, such as rna H ture condit ions, may up set the -condit ions $0 th_�t co na Ld e r-a o Le mortality v'ill anc e e host a t the nos occur. In many inst , larvae placed t deSirable temperature, have failed to show any emergence. It is not e nt irely understood v:h�T this should occur v Lt h some host larvae rhile in others emergence viII ta�e pl�ce. _t wore r'e l La b Le method of k.eeping a sU.9Ply of Ber�cyntus material avail�ble i� that of storing the earl t uree , ier stages of p.i r.rs Lt lzed l.:irvae in frsezing erape rat into a t ra In this case, pa r ..tsit Lz e d l.,;j.rvae are put ernpe nt ture of 28 to 32°F. at a t Lue when the Id.rval de ve Lo pne of the pJ.r::isites is DOt complet6 and the low temper::iture merely cJ.uses a pe r i.o d of d o rrnancy in their development. #.. - ,,_ .. _ .. _ -- - .. - - - - 218 - La r vae arid p.i rae it Lc Lar v.ae ...-ill continue t he Lr d e ve Lcp- n.e r.t norlll3.11./. Thie meth.od t a ke s a. little more t Lr.e illit Lilly L1 t h a t a large numbe .: 0 f cu t wo rm t ar v are mus t be examined under the binoculars in order to d e t e r mf ue which are p ar ae Lt ized, but v.he�� these are f o uud a supply of pa rae Lt e ma t e r La'l v:ill be pe rmane at Ly ava tl abj e e Luc e uo e e r Lo ue mortality ill this ue t ho d ha s been recorded. It is only r e aeo nab Le th1.t this me t ho d should be success- ful, since this p a r ae i t e does o ve rv Lnt e r Lr, partly g rov n Lar vae of such c Lee as Feltia duce.re and spe , Euxo9. tri�tl �. In the case of Lar vae o f b trist ienla v.h Lch hiber- na t e ve :_-'y eha ll owl.y in t he so LL, t eurpe rat ur e s as Lev as 0 '.) 19 to .... 0 F. have .t10 lethal effe�t upon. pJ.rasite larv3.e ill the host. In the experimental v,-ork. carried on iLl co nne c t Lo n vlfth the biology of the adults and with Lid ue e d pa.r a s Lt Le.u., man.z difficulties v:ere encouu.tered in handling such em a'l L par9.sites. Trouble W.3.S also experienced ou accou�t of the large numbe rs ene rging from one host. Hce t La rvae v e r e usually p La c e d in heavy glass vials, 1 inch v.ide 3.11d 5 inches long. The open ends of the viale �ere covered vith cheesecloth tightly secured by a heavy rubber b aud , :Jo so il 0 r e and v 3.8 put in t he vials, e i:-J.c e fungc>us :?rov t r; ';'18 quite prev.3.lent in such c ae e s •.\. little mo i.e t ur e r.3.S usua.lly added to the ctee8ecloth, sinoe thic f�ctJr seellied - 219 - to ..icceler,:i:ue t r,e ti e of e ne re nce , r i.s t he direct r3.�rs of the e uu JTovetl fatal tJ the adult e in some cae ee aud s Lnc e the s adult v. ere uaual L, so a c t ive if :111ov ed t o reua Ln in the li,:-ht that they soon e xh aue t e d trJ.emsel vee, the vials cant aining them ve r e ke pt in the d a rk , r:'hen it ne c e e became. s a ry to ha nd Le tLese. tiny. p3.ra. sites or in Lr s a siTI.61J pa , �:ery difficult prob1ell1 arose. a ao d of the wc uLd - IhturJ.lly, g many adults escape on t rar;.. af e re nc e from one vL).l to a . ..other; it v.az. doleo llecessary to be able to ident ifJ the sexes w it hout injuril1� the ad ults. .t n is necessary to be able to d Ls t Lugu Le h t he sexes by the genital o z-gane , ) The use of ether ','. ..is tried, but its eff�ct Ta� too lethal on the adults. �fter a fev methods we re d Le c rd e d the used trieu and a , method by Philips ('27) in handling hymenopterous pa.r.3.sites of the �heat joint �orm �as used very successfully. The covered e vh ich the eme . is i.d of the vial, in adults have rge , d r d for a few mf.nnt e a v he r n t lie tUT11ed upside ovmwa , eupo insects collect in the upper end. The covering of cheese cloth is quictly removed, the mouth of the vial held over a small cJ.ke of ice and the end of the vial � given tvo or three vi8-'or�us taps wi.t h the finger. The- insects fallon ':Lhe ice is the c a s.e of ice and =1.11 are at O1.1ce benumbed. -, ne then. n'Lac e d in a sha lov; saucer in wh Lch the 'spt: c i.me -'- ad t he ce-:es s c r-a t e d , ruy be examined under a rn Lc ro e co pe a pa - 220 - The ice 81...J u l d be nea.r] �T f I.a � O!1 bot h the llPller aud 10'Ver f.UJfJ.Ces so tll.:l.t the Lne s c t s I"ill not slip off. ,::he11 t l.e c Lue epe na are reooved from t Le ice those of each sex should be placed in a vi:3.1 r ith strips of absorbent pa pe r arid in a ft;v; minutes they v,-ill ue co me �T 3..11d be a c t Lve and a.lert �s ever, a.PP3.rently none the �orse for the ex perience. In this v.3.y vials Dl:l.;y be et o e t.e d vLt h the re number of quired individu3.1s of each e e z , �-hen they be come dry they a re re3.dy to be t ra ns f e rred t o the breC'din,z ca.ges. In the e r Lue nt a.i v o r k v induced r ac i.er i xpe ith p.a it , it \':i.e found difficult to ha nd Le the a d ult e in ::i.ny ""ay to pr-e vent a co ne id er 3..bl e loss. r-hile the me tho d us e d vas not e fa c t o vere entirely at Ls ry , �ood results obtained. The e g,!s wh I ch v.e r e to be pir a., it ized ve re placed in three o unc e tin C3.11S, either directly in the soil a t 9. d e pt h of one-h3..1f a.n inch (the norI113..1 depth of oviposition bj Suxoa othroQ.'3.ste:c) or p Lace d on a v ax pape.r 011 top of t l;e soil. Best results a r e o b t a l ne d v.hen the e�;;s are Single and t e s , e t r c e t r;e in not in La r g e c Lue r In i he ae , soil the till i·-_U=: moistened slightly. The soil eho ul d be very fLlely sieved, quite loose, J..l.ld v Lt l, a little moisture. 4.1 ordinary s i ae d Larup c h Luney vL th the top c o ee r e d o ve r 7:ith cot LO::'l cloth 'I-3.S used to fit snugly il.l.to the tin C3.n. ,:(he a d ul,t e to be use oJ Lu t 11e Luduce d p3.rdc it iel.1 - 221 - "ere p La e e d Ll tLe 13Jl2 ch i.uney ..i�ld the cliir,ll1c;, t Le u e d r La c o ver t , I uf r...l.:: p l.e till c m 0J.13. su.a l l)iec€ p.rpe , on tOl) of tIle soil, J. little a bco r be .. .l.t c o t LJL1 ::J.tu1"J.teu rLt h l:..jl1e.l e o Lu t Lo n ·,'.:is P .. wed to 0.110\' fuou for 1.1 e id ul L e • \7...1.8 b./ pLl.cill: J. p i.c c e of v.h i.t e blot" Ll_:' p rpe r e.it ur.i t e d \ ith ho ney so lut Lo n in tte c a je , ihe use :)f colored b Lc t t Lng .p..iper re sut e d in the d e a.t l, of lluny d.dults. The e ag e w.rc then placed in the Ciark. arid e xam i.ne d every d ay , In thts \'d.y the adults v e r e t.e p t alive for some tine, .:It ae t ve e 0 d r e t e a b t Le for tvo ks , and b'O eul, Joineu. SJurues of .l.Ccident.:J.J_ �OSS. From t he v ex ; -)e6ill�lillg there we ie many Vd..,'S 1:1 whi.ch there """3.S 3. co n et ant Los e of va Luab l e and iuterest ing material; these sources of loss ve r e the fO�.. lov, i1'l€':- n 1·. El.ltanglement in the [io ney -u rc ps , �he c o ramo e nd of too 1d.r�'e a proportion of d.ll t r,e adults urul e r o be e rva t ion. 2. '£he dr�)i':.l.g of the ho ney -e o Lt-t Lo n .J.ilcl the d eat h of insects by 2tJ.rv�tion. 3. �,-:'o":J.�')es :f ad ult e " hen the ho ney-ceo Lut Lo n is ch.l.nged. 4. De at hs due to the mcm1dil1.=· of the ho ne.r-« J.tt::l'. Before or �fter tt€ �uult� h3.ve died, the D...I.ra- s Lt Lz e d e:!gs h.t the g rc u» CJ.::.1 be p Lc i.e d 'J1Jt und c r <.:l. - 222 are ttell e unce p Lac d in a em.3.l1 Oll€- J tin c ari , r ith a filter paper top for mo t e t ure r e t e at Lo n and placed in oven an to hat ch , J.'he be c t o pt Lmum t empe r at ur e for hatch ing is about 90 to 950F. (Ill the case of host egJs nev:- ly deposited, it is ne ce es iry to subject them to cold t e mpe r a t ur e e for some time in order to b("./l.Q.k the na tur al. perio d a f do rniancy , ) . In v. ork-ing vlith such small 0 bjecte a� c ut v.o rn arid the still smallerp·3,ra.citic eggs, ant] v Lt h e uch euall organs as found in the p ar ae Lt Lc larvae, it vas essential that the most au Lt ab Le ne t ho o e of technique be used. Na tur a.Ll.y t he r e v a s a conSiderable amount of "trial a nd error" ex-pended in find in::' the best method to be used in the various types of study involved. Therefore, in the following e t at emerrt of the technique emplo.jed in the pre sent problem, not only will the most successful method be me e outl Lne d , but vrill nt ion also the other me t ho d wh Lch gave moderately good results. Chitin. One of the grea.test difficulties encountered ��s the presence of chit in (chorion of the egg). It cannot be e ut Lr e Ly dissolved without destroying the und erIy Lng ., ....,�/oi...... IiiiiiiII I!!!!!!!!!!I!!I!! II!!!!!!!IIIJ!I_!!!!!I - 223 - tissue. -\. rae thad which r.a s s uc e ee s ful Ly used for gross e--ramil1ition af the par a s Lt Le egg v:ithin the hast egg was to Lr.rbe d the host eg3' in the usual vIay and then by use of a sharply.pJinted scapel� trim the paraffin bloc� in such a v;ay as t o remove the greater p a rt of the chitin. By this means the ze La t Lve position of the pa r ae Lt e eg:: wLt h i.n the host eg� W'3.S de t e rm tral , The use of a 5 percent caustic potash sol�tion heated to a temperature of 60 to o 70 F. for a period of 3 hours is of valuable aid in soften- ing the chitin so tha.t it can be easily picked off. Hoy- ever, this method waS not successful in examining the in- ternal structure of the p3.rasite egg. It \las found that piercing the chorion of the egg with a needle was necessary to permit the pe ne t rat Lo n of the usual fixing fluids. The most successful metboB em ployed to dissolve chitin was by using a solution of 70 per cent alcohol cantaining 5 per cent nitric acid far a period of 24 hours. �s the previau� process of fixation caue e s some s the removal of the some- hz'Lnkage , charion is �hat facilitated. The use of Gilson's fluid (nitric acid) for 15 ta 30 minutes softens the chitin and is also an excellent fixer. Sec t io n Lnz , Because o'f the fact t ha t the chorion forms a hard layer, it �tS very difficult to obtain series of the host ... , ... " ....__IIIIiIIII_�IiiIIMII_IIiiiII ... Ia!!!!!!!!I!!I!I!--II!!!!!I!!!!!!!!!!! --JZ4 - and Jf egg the p�razit� eg�. This r.as very dcslr�bl€ EJt to Lo c.rt e onl�l difll1itclJ the posit.:..on of.the pa ra- e it , but at of r egg sc t h; e Lat Lve development of the t v o eg�e. TLe ho e t e e e t ions r.e r e obtained by fi:ting the :for 2 L'linutes in Gilsoll'S liquid v a.rtue d to 70oFahrt::nheit �nd leaviag them ag�in for 12 ho�rs in the same li�uia at k.ept ::i temperature of 450F. '111e e2;''''s t hue tr eat e d ver e then passed up through successive alcohols (30, 50, 70, 85 and absolute), cleared xylol for 1 to 2 hours, i1.1fil- o t r a ted Yiit h pa ra ff Ln (.35 C.) fo r 2 hours, sect t o 11e d and stained v.-ith haemoto::glin. The use of Van Leeuwen's fluid (picric acid base) for 24 to 36 hours was also valuable, but a longer period was ne ce e s . .1ry to obtain results com- parable to those obtained with Gilson's fluid. In order to e carui.ne the iuw;:t.ture eggs in the ovar te e of the adults, it was necessary to fix slightly in a reat solution of osmic acid (1/10 of 1 per cent) for tvo hours and stain for 3. eo ns t de rab'l e period (48 hours ) in picro ca.rmine. It V:3.S often found more co s.ven Lent to exam i.ne the paras it ic body v!ithin the host La r vae "Ln Situ" rat Le r than by sectioning. The use of 10 per ce nt solution of osmic ac id for 12. to 2,;, hc ure "\,'i11 cause the macerat ion 0 f the host tissues to such an extent th:l.t the par as Lt Lo body v"7ithin call be �asily distinguished. �tfter using the osmic ,I" "1;11I,,liI'l,[,II.IIIIIII ...IiiII.-=.!Iii...... ------!!!!!!!!!!!!!!III!!I-----� - 225 - solution it wa e stained with pa rac a rrn Lne , In the int e r io r 0 f the no e t the r as it e IJ.rva. , p.i body e.m be seen by trJ.llSp;l.renct;' vLt h a p repar at Io n nude in the Same Tra.y, but exa.mined in wat e r v;ithout staining, the nuclei ill the embryos were very apparent since they are quite refringent. Osmic acid is by far the b eet dis sociator to use since the fat bodies of the host IJ.rv.ae ar e blackened by it and the presence of the p.aras Lt Lc body thus determined. THE COR�L.1.TIOlf OF 3T�UCTUR� jlTD H08T-:EU_·::S J..TIOH �-\.HONG rnm ElTCY�T IUJ.E . Even from a superficial study of par3.sitic Hymen opt era, 0 ne is much impresse d by the unifo rmity v:it h r.h ich paras�tes,of certain more or less restricted groups are parasitic upon insects of certain groups also of more or less limited extent. Very broad and ewee p Lng statements in this direction to v:hich the-re are, however, many ex- a.re ce Lo ns be nu , while the pt , may de Thus, Lepidoptera. parasitized by many representatives of all of the four principal families of parasitic Hymenoptera, those of the sub-family Ichneumonidae may in general be said to be paraSites of Lepidoptera. The species of the braconid subfamily Euphorinae aTe usually p�rasites of Coleopter�, those of the subfamily ::Microgasterinae are par ae Lt e a of _.:...... -..i__...iillliiiiiiiiiiiiiiiiillliiiiiiliiliil _ - 226 - Lepidoptera, those of the I.:octotrynid subfamily l'laty- g�sterinae �re p�rasites of Dipt�ra, mainly of Cecido myiidae, �nd those of the subfamily Dryininae of the Homoptera of the families �embracidae, Jassidae and Tettigoniidae. Inctancts of this Kind might be multi l, ied at the same t Ln.e s s p but, , groups in which much Le uniformity exists a r e also numerous. In the family Chalcididat:, to zh Lch the subfamily which I sh�ll p�rticularly discuss belongs, there is the s�me uniformity in some groups and the lacK of uniformity in 0 thers. Very few 0 f the subfamilies possess any great uniformity throughout their whole extent. The Tetrasti chinae, however, appear to be unifprmly p�rasitic on other hymenopterous p�rasites, v.hile the Elachistinae are pa.ra sites of Lepidoptera, and the Toryminae are parasites of gall insects, the pref�rence of the latter depending not so, much upon the structure of the host as upon its poss ession of the gall-maKing habit, sin(;€' they a t t ac k cyntpid, cecidomycid, trypetid, and even lepidopterous gall-maKers. In the majoritJ o� the subfamilies, ho�ever, there is a much greater subdivision of the correlation of structure and hab it. The Encyrtinae of Europe have been carefully mono> graphed by Dr. G. Mayr of Vienna. Out of the �3 genera n lice which are described by' 1.1ayr, 15 are p.s ra.e Lt Lc -upo bark. - 227 - exclusively. Other genLr� �re also specific p�r�2ites. There is one OJ.1ly genus, �llci�, whe re there is J.l0 absolute uni:foTmity in host relat Lo n To ithin g e n. ric bounds inso:far thJ.t the host insects of each p3.rticul:3.r genus J.Te closely rel�ted and o:f the Same g�neral type. Encyrtus is one of thoce unwieldy genera of varying limit, in v.-hich many species have been lumped, fr e que.rt Ly for insuf:ficient reasons and really :for want of a better 'place to put them. An examination of the American epe c Ic s in this .sub family shows the :follov.-ing results:- Among the species p�rasitic upon CoccidJ.e there- ).re three d ist Lnc t type_s; the spec Le s paras it t c upo n the ':rphid Ld ae form an independent type in the group distinguished we r ke d e by L'l=ma structural charao t re • .J.ID.ong those pLrasi tic upon Psyllid:3.e, an intere�ting condition is found. Those reared from gall-mak.ing l'syllidae belong to the eauie type vs that reared from a gall-mak.ing Cecidomyiid, v:hile those pa.r ae Lt ic upon free-living Psyllida.e belong to t r.o types, distinct from e ac h other and from the first .• The paraSites of the free-living dipterous larvae belong to a common type distinct from the others, while th�t reared from the dipterous gall-maKer agrees in n�in structural characters with those just mentioned from psyllid galls. The species reared from cynipid galls form another type and the most distinctly ma r ka d one of the who Le series. - 228 - The c e epe Le r e a re d from Le p idopt E r'o ue larvae be Lo nz to a common type, closely resemblin� the forms re�red from free-livillg dipterous larvae. Those reared from heter opterous eg�s and those from lepidopterous eggs belong to a common type and v:hile e e par abLe from each other by cer tain structural characters, these seem unimportant. The abo ve f ac t s merely. tend to shoy:.:.. (1) Another exemplicationof the axiom that struc ture is dependent upon habit. (2) Th1t wh LLe the true c l acs Lf Lcat Lo n depends en t ire�y upon s t ruc t urat 'detail, Vie may gal n ideas as to the relative value of char�cters by a �nor.ledge of vital habits; and (3)' That as soon as s uf f' Lc Le nt records accumulate, it will be important to examine the classificatory rear ings of the group-habits, p azt Lcu'l rr Ly of the host-relation, V7ith other groups of pcrasitic hymenoptera. In the above discussion only one side of this im- portant subject h�s been touched upon The other � ide is the structural differentiation of forms �hose host-relations of several and even fam are Ld e nt Lcal , ,Parasites genera ilies are prasitic upon the same h:Jst-type and even upon the same individual. - 2::!9 - s U :.! I: A R Y Berecynt� b3.keri Var. geuua Gira.ult belDngs to tl:..e hymenopterous SU1)er:fu.mil�l Ch3.1cidoidea., fa.mily �l1cyrt idae and subfaBily �ncyrtina.e. Bert:cyntus bak.eri is a primary polyembryonic pcl.ra site of :roctuid3.e, v;ith the chief host-species being t-he re d=backe d cutworm, �uxoa ochrogg,ster Gn. Parasite and host a�e single-brooded. llorphologic3.11y, the sexual larvae of this chalcid belongs to Group III, according to Dr. H. L. Parker's lar val classification. The asexual larvae are placed in Group V. The parasitic adults have been observed in the field from the early part of [aay unt il the end of Augu�t, the max imum abundance occurring in July. Under natural conditions, it is probable that female adults live for a period of about three wee ks , The potential progertltiveness of these fe'males averages about hiO hundred eggs. Copulation and fertilization take place soon after emer�ence last but a few seconds. o and. Parthenogenetic development 00CurS, producing a poly embryonic brood of wales. Parasitized host eggs can be recognized by the sealed oviposition puncture • ' - -- .. , . ��iiiiiiiiiliiiiiiii"'liiiiiIiIiiiiiI--- - 230 - In about ninety percent of pa rae Lt ized 13. rvae there is an additional t ar Lns , This reeults in a tv.enty-five percent increase in food consumption. Under laboratory conditions, parasitized larvae have a life-cycle eight days longer than normal larvae. Para sit ized larvae feed approximately sevent een days longer than healthy larvae which possibly accounts for the increase in the amount of food consumed. Parasitism by Berecyntus bakeri is not specific. At I Sasl\.atoon it has been reared from �uxoa ochrop:8.ster Gn., Euxo9. detersa \·ak., Eu.."'{oa tristicula Morr., Chorizap-rotis than9.tologia Sm., �a.'5ro�is flavicollis Sm., Feltia ven erabilis \ilk., Feltia ducens \,flk., and _sidemia devastator Brace. The paraSite exhibits a marked preference for species inhabiting cultivated land •. Super-parasitism of the red-backed cutvorm occurs, Berecyntus bakeri having been found in combination r.ith Paniscus �., an external parasite. � bakeri receives an a dvant a ge from the prev3.1ence of cut wo rm disease. the The morphological a�d biological adaptation of paraSite to its host is so perfect that the large number of internal paras Lt ic larvae cause no inconvenience whatever to the host larvae up to the sixth instar. There is also a rhythmical adaptation, which is not so synchronous. Parasitized La rvae are a'lwaye much enlarged (due to - .331 - internal a r as Lt Lc La r vae the e a r't he n ,many p ] , pupal cells of the host larvae are never co t ed up'le , and the l.:l.rvae are usually sluggish in acti0n. Internally the fat bodies are considerably changed. -In nature, the adults find a varying and irregular supply of food �hich in its effect on sexual activity and longevity is equivalent to honey �ater. Low temperatures have very little effect upon the paras it ic larvae, the only except ion being whe r e there. is a great diverge1.1Ce be twe e n alternate thaw Lng and freezing co nd it ions. A correlation between sunshine and parasitism exists to a more 'or less degree. no such c::.rrt:lation has been found in relation to soil moisture and soil temperature. The general rate of parasitism by Berecyntus bateri of its most abundant host is influenced by an imperfect . adaptation to that host. The rate is further influenced by the total parasitism of other host species by this para site. The species attributing the most effective influeDce are duc ens tristicula and Sidemia devastator. Feltia , Eux£§. There has been a gradual increase in parasitism of the red-backed cutwo rm by this par-as Lt e from 0 - 10 percent in 1923 to 40 percent in 1928 and 65 percent in 1929. s s in the of its hosts wh Lch The pa rae it e o vi.po Lt eggs it finds in the so il in summer or early autumn. --.�--- -� --_- -- .. - - _ -:' 232 - The Lt e par-as Viill deposit. its egg in the hce t e::;g at any time but sho�s some preferelice for newly deposited host eggs. It does not parasitized the young caterpill�r after hatching. In most cases it lays only one egg at a time. The newly-laid egg is short and broad, typically elliptical in shape. The broad end of .the egg is posterior and the narrower end is ant er Lor , The egg may be placed in any part of the host egg but does not develop uriless embedded in the tissues of the host embryo. In fertilization only a Single sperm enters. Poly- spermy never occurs. The maturation divisions are typical and result in reducing the number of chromosomes from sixteen to eight. The process is identical in fertilized and parthenogenetic eggs. The cleavage nuclei are from the first accompanied by cytoplasmic segmentation. Cleavage is confined to the posterior end of the egg and eventually results in a mor ula-like stage. a bud The po Lar region elaborates in the form of at the the expense of its host, forming the trophamnion and paranuclear masses. These bodies are protective and nutri- tive in function. - 233 - In the wintering stage the parasite body is a verit able synyctium surrounded by a trophamnion. A �emarkable biological adaptation of the p�rasite to its host is ,shown by the fact that in 45 days of develop ment "in the fall, the parasite body increased only twice its size in length and width. IThen the hos� larvae hatch in the spring, the p�ra- - site body is usually found lodged in the host fat-body. The embryonic nuclei group themselves to form germs. These increase in size and break. up, each giving rise to two morulas. The morula develops into a polygerm, Yihich consists of a number of pri�ry masses. Each primary mass consists of a group of embryonic cellS, surrounded by an inner mem- b rane , The primary masses multiply by constriction of the inner membrane. The products of these divisions are known as se condary rnasses, which in turn multiply by Similar -corr- strictions to form tertiary masses. The tertiary masses later divide to produce components. At acme time during the period of mult iplicat ion of or dis the masses, the polygerm undergoes fragmentation the sociation. The ma.sses become scattered throughout body to cavity of the host-larvae. Each morula gives rise an are dissociated embryo and when these are fully formed they as young larvae. - 8.34 - The process of fc rmat Lo n of germs is rather hap haza.rd. As a result, a few of these germs· lack. the pot entialities of developing into morulas and become pseudo germs. A similar condition exists for later stages, re sulting in pseudomorulaz, pseudoembryos and pseudol�rvae. Pseudo larvae do not live over three days in the body cavity of the host. Not over 12-15 such larvae arise in one host larva. The young larvae feed first upon the blood plasm of· their host and later the fat-body, muscles, etc. are attacked. By the time the larvae are full grown the entire body contents of the host larva has been devoured and the host is nothing but a carcass· containing many chambers in �hich the parasitic larvae pupate. The pupa.l stage lasts about 20-25 days following which the adult par as I te e eat their v!ay out of their cham bers and escape into the open to search for host eggs in which to oVi�oSit. An average of 840 adult paraSites is developed from a single ElL�oa ochrogaster egg as the host. Uith Euxoa tristicula as the host,·the number is conSiderably smaller, averaging about two h�1dred adults. From the small number of parasitized larvae examined be (30) over 50 percent of the parasitic adults emerging longed to mixed brood�. Of the total number of individuals found in the 30 broods e�aminedt 64 percent were females and \ ___ II ...... • __ - 235 - 36 percent were males. BIB LI 0 G�\'pHY A�H11EAD, I,':.r" H. - 1900 On the Genera of the Cha1cid - flies belonging to the subfamily Encyrtinae. 1?roc. of U.S. Nat. I.fus. Vol. XXII p , 32.3-412. ATKIuroiT, U.J. 1926 - A preliminary study of the biology and morphology of the red-baCKed cutworm. Master's thesis, U. of Sask.., unpublished. BATAI:SLOU, E. 1900 - Pression osmotique de l'oeuf et polyembryonie expe riment ale • C .R •. Ac. Sc. Paris.CrXX p. 1480-1482. BUG1HON. E. - 1891 - Recherches sur Ie developpement postembryonnaire L'anatomie et les moeurs de l'Encyrtus fuscicollis. �ecueil. Zool. Suisse. T. 5 p. 4.35-536. CAUL:S:::::.tY, ti. and llesnil, F. 1901 - Rech€rches sur les Orthonectides. Arch. al�t. micro. IV p. 381-470. cso SS:.u.lT, S. s. 1923 - Ap�nte1es me1anosce1es, an imported p�rasite of the Gipsy moth. Bul. 1028, U.S.D.�. DOTEU, S_i.t�u:i:L B. 1911 - Concerning the relation of food to reproductive activity·and iongevity in certain hymenopterous parasites. Technical bulletin 78, Agric. Exp. Sta., Univ. of nevada. G1TEUBY, J .B. 1918 - Polyembryony in parasitic hymenopter�. Quar. Jour. Micr. Scienee, vol. 63, pt.2 pp.175-196. 1916 - New Encyrtidae from U. A.merica. Psyche Vol. 2j, pp. 49-50. 1917 - Description of Berecyntus ba�eri Var. E��oae. Descriptiones St-ellarum novarum, -l1ay 1st, 1917, p.lS. HAEC�E-S, E. 1869 - Zur Entwictelungsgeschictte der Siphonophoren. Utrecht. .� _, - ... ,� �.... � ._- - - ._ -- � , - 237 - R.L�I.8�, S. F. 1893 - On the Occurrence of embryonic fission in Cyclo st crnat us Po lyzoa.. Q.uar. Jour. lacr. Sci. :::x::av. p , 199 -242. 1895 - Prelimim.r.f note on embryonic fission in Lichenopora. Proc. Roy. Soc. LVII p. 188. 1896 - On the development of Lichenopora verrucaria Fa.br. Quart. Journ. Lacr. Sci. xrxrx p. 71-144. HEFF�, It�RO LD H. 1928 - Differential effects of constant humidities on Pro rt r Lt e toplrce quinquemaculatus Hawo h and its pa aa , \Hnthe� quadripustulata Fabr. JI. Ec. Ent. vo L, 21, No.1., p. �13-221. HER:.rs, W.B. - 1926 Effects of par ae Lt Lsm on the host and on pa ras Lt e , Jl. Econ. Ent. vol. XIX, no. 2., p. 316-JZ5. HOY,T .I.:::m, L. 0 • 1891 - The methods of pupation among the chalcidid�e. Insect Life. Vol. 4, p. 193-196. 1891 - �he biology of the hymenopterous insects of the family Cha1cididae. l'roc. u.s. nat. I.ius. Vol. 14, p. 567-588. 1893 - The correlation of structure and host relation among the Encyrt inae. WildaY'- Quarter cent urz Boo k pp , 177-185. 1897 - A study in insect parasitisw. Bull. U.�.D.J.. Div. Ent. T. S 5. 1898 - On so me new parasit ic Lne e c t e of the e ubf am LLy :;:;nc�,r Vol. :II 2j7-.:! .... b. t Lnae , Proc. U.S. Hat. :rUSe p. -- . ., _. __ .. - -- - .I' �- - _ _ � • J,;-- - 2J8 - 1906 - Po Lye mbryo ny aad the fixing of sex. Science n.s. Vol. 2.,1" Iro. 625, p , 810. HO' '1::tD, L� 0; and Flm:�, ·r!. F.' 1911 - The importation into the U.S. of the parasites of the gipsy mo t h ar.. d the brown-t a tl uict h ; a report of pro�ress with·some cOl�ideration of recoveries and concurrent reports of this �ind. U.S.D.�. Bur. �nt. Eu1. 91; 312. 1926 - Some notes on technique of handling parasites. .n , Ec. Ent. Vol. 19 tre , 2, p. 311. KING, 1:. M. 1926 - The red-backed cut r.o rm and its control in the Praitie Provinces. Can. Dept. Agric. Pamph. 69, IT.S. KnIG, I.::. �,:. and ...\.Tt nTS):r, IT. J • 1927 - Quantitative methods of collecting and re::iring soil cutworms. Jl. Ec. Ent. Vol. 20, no . 6, Dec. 1927. 1928 - The biological control factors of the Lrnma t ur e staa-es of-�ux:oa ochroC!'aster Gn. in SJ.8k.atchev·an. Annals Ent. So c. A.Dler ica. Vol. TIl p � 167-188. K�InEI:ffi�G, If. 1879 - The development of the earthworm Lumbricus trare zoides. �u�rt. Jour. Uicr. Sci. -:XIX p , 206. KORlffi�US�R, s. I. - �39 - 1919 - The sexuBl ch8r�cteristics of the L.!elllbrf1cid Thelia bimaculflta Fabr. I. Extern!'l changes induced by E�;.Y��s. theliee Gahan. Jour. 1..0rph. Vol. ��, .1.10. 3 p. 531-636. - 1922 The polye ...bryonic develo.pllJ.ent of CO"l)ldosoma Gelechiee._, w 1th not e 6 0.i1 its bl ()logy. Jour. �orphJ 37; 195·265. - 19�,5 The origin of mix,ed broods in polye:..bryunic hY1t!enopter� • Annals Ent. Soc. Ame r c , Vol. 18, �io. 3. - s 1928 Polyembryony in Insect , 'I'ral1 S. 4th. Int. Cong�ess Entocology. P. 67�-�87. LI!..IBY, h .','[. and HILL, C. C. 19�3 - The tWir1l1ing t?!1d mOl"1embryonic development of P1�t'r."';!=lster h.t�.,rfl.�11s. EI parasite of the He s e i en fly. J our. of A6r. �eseerch. Vol. �5J p. 337-349! 1924 - The po Lvemb ryon i,c develo.:.,.f:ient of Pl�t:r3"'st_u vei.�{lalis. Jour. Agl"ic. l\esearc.u, Vol.:8{VIII, :Io. 8, p. 8�,9-S39. LAlloCHAl., P. 1904 - heci..lerches sur les biologie et le develo,;,.:pn!ent des hymenopteres parasites. 1. La polYGd).!'yuilie s�ecifique ou gc�uinogonie. Arch. Zool. Ex,.? et Gen., (4) T 2, p_tJ. £57-335. IJ J. .'l' Ill, . .l.,_ 1914 - Zur Enti,7icklungsgeschickte des polyel.�bryoll::11en Chalcidiers A-;;el1i�s,')is (1!:r1C2�� f!.l.scicollJ.� Zeit. f. ·;lis�. Zool. na, 110,8, p. 419-479. - s .•b onne 1924 ne cne rche s sur Le s forme ,Post-el rv ires des Ch81cidie�s. AI1nele s de 18 Societ e Erlt omologi que de Fr�'{lce, Vol. XCIII. -----, � .. - ._ __ � ..::._-_._,. --. .. -.,...... _- If .... - - �, v 1925 - Notes on the La rvae of the Ch..llciuoidea.. \.ll1'13.1s �l1t. �oc. .rne r , Vo 1. XVII I ne , 3, pp. 364 - 395. 1927 - Etudes sur la bio10;:ie des Lueec t es p..l.r_ates: 130 vie pJ.racitaire et 130 not iOH mo rpho Lo g Ique de 11 a da.pt at ion. AnnaLe s Soc. Ent. France, 701. ·(CVI l)P. 11':-l'1G. lS25 - Contritution a 1.1 biologie des Ch�l�idiel� entomophages. Alllul e s So c. :llt. Fr3.nce, 70 1.::� VI I, p. 4Jr.; _.1. Cf. • P _G: a:r. D. C. 1928 - E.:cperiLleiJ.tal d Ls aera l nat Ion of the ta.banid e;;;,; parasite, Phannrus �r.!ersol1i Girault and bio Lo gLca.L note:s on tr... e E.);Jecit:s. Circ. 18, U.S.D.t. Jan. 1928. 19L":, - Polyembryonic de ve Lo pne nt in T3.tu8iSl. !!:�� c Lno t a , Vol. 2 � no. , r :9-GG2. Jo ltC-.-lIo rph. _, 4, pp 1915 - Ob�Lrvations on the deve10p�ent of CoplJosJn� gelechiae, BioI. Bull. Vol. 29, lio. 6, Dp. 3:3-373. 1917 - Studies on the bio106;/ of P�l'auopid:)som);sis. I - �ata on the sexes. BioI. Bull. Vol. �2, �o. 5, pp. 291-:00. III - :.�tur3.tion and fertilization. BioI. Bull. Vol. �3, Ira. 2, :vp. 37-62. 1918 - IV - 7�e asexual l�rv3.e. BioI. Bull. Vol. �5, no� 6, pp. 3G2-377. 1919 - l'olyeLlbcyony and sex. Jour. Harid. Vol. la, lTo. 8, pp. 344-:J[j2. . 1921 - The development of Paracopidosompsis. - Jl. l':orph. Vol. 36, p:p. 1 69 • -_ -_._-. . ----. . --"- -'l:------�,-,,-i.··._ _ - Iii" "_.. • _,,", - '241 - 1927 - Polye[1bryony in .tnimals. Q,uart. Revie", of BioloV, Vol. 11, IrQ. 3, Sept. "2. 7. PEr:.r:?�, r.r. J. 1927 - Eurytoma parv3., and its biology d,S a parasite of the wh ea t jOint worm •. Jour. Agric. ResearL;h, Vol • .34, !To. 8. RICK�D�Olr, C. H. 1913 - Etudies on Splangia. - .. Jour !.TorI' h. Vo 1. 24, Ilo , 4, PI'. 513-557. RIT3Y, 'If. A. 1907 - Polyembryony and sex-determination. Science. Vol. 25, p. 106. 1901 - Contributions a l'etude au development des Cest odes. II. Le developpement embryo nna Lre du Taenia serrata. Arch. de parasitologie IV, pp. 143-156. 1915 - Osserv�zioni biolo�icbe sull' �1arsia line�tella Z. dannosa al frutto del mandorIO:Bol1. �a.b. Gen. ,lgr. Zool. Sup. Agr. Portici, Vol. 10, pp. 51 - 55. 1918 - La variegana ed i suoi parassiti. Vol. 12, : p p , 175 187. SR::'::'FOEill, V. E. _ ..... _ ., ... .-IA .. �- _ . � ��� - �.t2 - . Sm:SFO�! V. E. 1926 - The relation of abundance of parasites "to weather conditions. ,Jl. �con. Ent. Vol. IIV No.2, pp. 283 - 2b9. 1906 - Contribuzoni alIa conoscenza uio1o�ica degli -- Imeno\teri p�rassiti. I Bio10gia del Li tOI:13.st ix t runc e L'Iat u s (D'�.lm)! Annali della '1r.Scuola sup. d'a::'l'ic. di Portici, Vol. 6, pp. 3 - 51. 1910 - Notizie preiimim.ri sullo svilluppo die ConidosonH - buye aon l , - l�onitor anno , Vol. 2S8. Zool� Ital:. , 21, I'll). 2�6 SMITH! H. 1916 - An attempt to redefine the host re1�tiormhips exhibited by e nt orao phague insects. Jl. Eeon. Ent. II No.5, pp. 477 - 486. srrRIe I::L.UID, E. H. 1923 - Biological notes on parasites of pr�irie cut worms. Can� Dept. Agric. Bull. 26 N.S. SNOW, S. J. 1925 - Observat ions on the cutworm Euxoa auxf.Li.ar Ie Grote and its principal parasites.--- Jl. Ee. �nt. Vol. 18, pp. 602'- 609. TI�.rn��!d.. KE? P. H. 1912 - Experimental parasitism. U.S.D."i.. Bur. of Ent. T.S. 19, pp. 71 - 82. .. Z43 - TO::E3, D_\.IT I�::" G. 1916 - Comp3.rative study of the a no unt of food e at e n by parasitized and nJn-parasitized larvae of Cirphis unipuncta. Jour. j,gr. Research, "lol. VI, nc . 2. 1910 - The effects of parasitic and other Kinds of cast rat ion in insects. Jour. Exp. Zool. Vol. 8, pp. 377 - 438. �,'HEE TTR, E. I.'. - 1923 Braconid parasites of aphids • , .lnrals Ent. Soc. Amer Vol. XVI, lIo. I, p p , l-Z9. - sex 1927 Some problems of rat Io and pa rt he no.jene e Ie • J1. Genetics. Vol. VI, no. 4. r:o LO 0 T T, G • U. 1918 - An- emergence response of Trichogramr.n minutum Riley to light. Jour. Ec. Ent. Vol. 11, pp. 225 • ...... _- .,"'.�.. �..._--•...... ImIl...... IC:!�--