Decembeb 21, 1906.] 



SCIENCE. 



817 



when the young larva of the Hessian fly en- 

 gorges itself with sap. Now, the production 

 of the rapid changes bringing about osmotic 

 pressure constitutes precisely the procedure 

 employed to bring about the separation of the 

 blastomeres and their evolution into several 

 distinct individuals, as has been shown by the 

 experiments already mentioned of Loeb and 

 Bataillon. 



Polyembryony is connected with the ques- 

 tion of the fixation of sex, and offers from 

 this point of view an especial interest. 



Bugnion observed already in the course of 

 his studies upon Encyrtus (1891) that all of 

 the individuals coming from a single cater- 

 pillar most often belonged to a single sex. 

 A total of twenty-one observations carefully 

 controlled gave the following result : five times 

 of males exclusively; nine times of females 

 exclusively; three times a great majority of 

 males; once a great majority of females; 

 three times males and females in nearly equal 

 numbers. 



Marchal has stated similarly that with 

 Polygnotus, those coming from a single larva 

 of the Hessian fly almost always belong to the 

 same sex. 



These facts, which Bugnion thought should 

 be attributed to an occasional parthenogenesis 

 (the caterpillars giving birth exclusively to 

 males having been, according to his supposi- 

 tion, those which had been pierced by a non- 

 fertilized Encyrtus), are now to be explained 

 in a much more rational manner. 



With man, true twins enclosed in the same 

 chorion probably come from a single egg. 

 While different hypotheses have been suggest- 

 ed, especially lately (Rosner, 1901), it is nat- 

 ural to suppose that twins develop by the sepa- 

 ration of the egg into two parts, (spontaneous 

 blastotomy). Then it is established that true 

 twins are always of the same sex. Exceptions 

 to this rule are explained by the fact that cer- 

 tain unusual twins are formed by the joining 

 of two eggs. 



Another case presents itself with the mam- 

 mals, which seems much more comparable to 

 those of Encyrtus and Polygnotus, namely, 

 that of the armadillo (Dasypu^ or Tatusia). 

 These animals give birth, according to the 



species, to a litter of from four to eleven 

 young which are all and always of the same 

 sex. It has been noted by Ihering (1886) 

 that all of the foetuses are enveloped in a com- 

 mon chorion and belong, therefore, to the type 

 of true twins. Eosner (1901) explains this 

 fact by the habitual presence of several ovules 

 in a single graafian follicle, and has even con- 

 cluded that all of the cases of monochorial 

 multiple birth can be explained in the same 

 way. But Cuenot (1903), reviewing the ques- 

 tion, has found that with the species studied 

 by Rosner (Tatusia novemcincta Linnaeus) 

 the monovular follicles are twenty times more 

 numerous than the pluriovular follicles. It 

 is then impossible to admit that the latter 

 only furnish the fertilizable eggs, and the 

 author concludes that, according to all proba- 

 bility, the multiple births of armadillos come 

 from a single egg. 



The discovery of Marchal, therefore, comes 

 extremely apropos to throw new light upon 

 this interesting and greatly discussed ques- 

 tion. In the cases where Encyrtus and Polyg- 

 notus issuing from the same larva are almost 

 all males or all females, it must be admitted 

 that this is a natural consequence of poly- 

 embryony, and that one would expect the 

 sexes to be separated in this way wherever the 

 embryos come from the division of a single 

 egg. 



The fundamental fact coming from this 

 study is that every caterpillar or larva which 

 contains a single chain of embryos gives birth 

 to imagos of the parasite belonging to a single 

 sex, but as the same caterpillar frequently 

 contains two or three chains it will not be 

 astonishing to find males and females given 

 out in quite equal number. The cases in 

 which we find individuals of both sexes, but in 

 unequal numbers, are to be explained by the 

 partial aborting of one of the chains and the 

 survival of only a few individuals, while the 

 other chain develops normally. 



It is seen, _theref ore, that the discovery of 

 polyembryony confirms a fact already sus- 

 pected but until now incompletely demon- 

 strated, and that is that the determination of 

 the sex in the fecundated egg is definitely 

 hrought ahout before the first segmentation 



