Patterson: Polyembryony and Sex 



351 



Since we had been able to show ex- 

 perimentally that if the host egg is ovi- 

 posited by a virgin female, the result- 

 ing brood of parasites is always male, 

 but if parasitized by a fecundated fe- 

 male a mixed brood develops, the ques- 

 tion was raised as to whether the fe- 

 male of Paracopidosomopsis deposits 

 two eggs at one oviposition, and if so, 

 whether the impregnated female (as 

 does the queen bee ) could lay unfer- 

 tilized as well as fertilized eggs. If 

 this were so, it might explain the ap- 

 pearance of numerous mixed broods in 

 this species. These questions have been 

 tested out by appropriate experiments, 

 and it is found that in two times out 

 of three, or in about 66% of the cases, 

 the female parasite lays two eggs in the 

 host &gg at a single oviposition. If it is 

 assumed that in every case one of the 

 two eggs is fertilized and the other 

 is not, there is still a discrepancy of 

 over 20%, because 87% of all broods 

 in this species are mixed. Further- 

 more, I have studied, by means of 

 smears, the two eggs laid at one oviposi- 

 tion by the fecundated female, and find 

 that in the majority of cases both eggs 

 are fertilized. In a few eggs it is diffi- 

 cult to demonstrate the presence of the 

 sperm, although a failure to do so does 

 not necessarily indicate that such eggs 

 are unfertilized ; because the position 

 of the Qgg on the slide may be such as 

 to obscure the sperm. 



In this connection the conditions in 

 Platygastcr are very interesting. The 

 host egg for this species is too small for 

 smear preparations, so that it is neces- 

 sary to resort to sections in order to 

 determine whether the parasite deposits 

 one or two eggs. The host eggs, which 

 are usually laid in masses, were ex- 

 po'^ed to a mixed brood of parasites, 

 pnd sections then made of these masses. 

 Some of the eggs will, of course, be 

 parasitized by more than one female, 

 but the position of the eggs in the volk 

 of the host egg will show whether there 

 has been more than one ovinosition. In 

 a group of 28 eggs the following data 



were obtained: Each of 14 eggs con- 

 tains one parasitic egg; each of 11 con- 

 tains two ; and each of 3 contains three. 

 In eight cases out of the eleven the two 

 eggs are far enough apart to indicate 

 two ovipositions, while in three cases 

 they are close together, suggesting that 

 the two eggs may have been laid at one 

 oviposition. In each of the three re- 

 maining eggs the three parasitic eggs 

 are not close together.^ In view of these 

 facts it is impossible to explain the ori- 

 gin of mixed broods of Platygaster, 

 tabulated in Tables IV and V, on the 

 two-egg theory. 



The suggestion has been made (Pat- 

 terson '15, '17a) that the appearance of 

 so many mixed broods in some of the 

 species of polyembryonic parasites may 

 be accounted for on a monozygotic ba- 

 sis, namely, that a fertilized egg gives 

 rise to males as well as females. This 

 result could be brought about through 

 the abnormal behavior of the two sex 

 chromosomes during the early cleavage 

 divisions of the egg. There are several 

 different ways in which this might oc- 

 cur, but the most prabable way is by 

 the means of somatic non-disjunction 

 (Bridges, '16). If this process should 

 occur, certain blastomeres receiving but 

 a single X chromosome would become 

 the progenitors of male embryos. The 

 sex ratios in the broods of Polygnotus 

 in particular are more easily accounted 

 for on this basis than by tlie two-egg 

 hypothesis. This is especially true with 

 reference to the large number of broods 

 which show but a single male. The 

 regularity with which certain sex ratios 

 appear in many broods {c. a., 9 females 

 and 1 male. Broods 88 to 99) suggests 

 that a single male is formed at some 

 very definite point in the earlv devel- 

 opment of the eggs. If somatic non- 

 disjunction does occur during the earlv 

 cleavage stages, it might be regarded 

 as a very primitive type of paedogenetic 

 development, in which some of the early 

 blastomeres, functioning as egg cells, 

 produce abnormal behavior of the sex 

 chromosomes in the cleavage division. 



;; T7r)r tl->p«e Hqtp T am indebted to Afrs. Lelia T. Porter, who is studying the development 

 and cytology of these insects. 



