POLYEMBRYONY AND THE DETERMINATION OF SEX. 815 



are grouped in such a maniier that the}' form a single ovoid mass, 

 rocked In' the contractions of tlie stomach walls." 



The author, it is true, has not as yet observed Polygnotus in the act 

 of oviposition, but having- found the freshly deposited eggs in the 

 gastric cavity he succeeded in following the multiplication of the 

 nuclei, afterwards the grouping of the cells into a number of indi- 

 viduals just as distincth' as in Encyrtus. Polyembryony is there- 

 fore well established for this species. The only dili'erences from 

 E. fuscicollis are : First, that the morula stage follows a genuine 

 blastula with central cavity, preceding the formation of the em- 

 bryo; second, the proliferation of the germ being considerably less 

 active, the number of individuals proceeding from one egg appears 

 not to exceed a dozen in number. (Marchal, 1903 and lOOlr/.) 



Other examples borrowed from the whole animal kingdom may be 

 associated with the polyembryony of insects. 



In the Cyclostomids (Bryozoa) one encounters a budding which 

 takes place in the egg, at the beginning of development. In the genus 

 Lichenopora this budding is replaced by the breaking up of the first 

 embryo into a great number of secondary embr3'os. Thus we have 

 here a phenomenon comparable to that which we have found in the 

 Hymenopterous parasites. It should nevertheless be noted that the 

 secondary embryos thus formed already present an indication of the 

 embryonal lamellae (planula), whilst the morules of Encyrtus or the 

 blastules of Polygnotus do not present any apparent ditferentiation. 

 In the other Bryozoa {Lophopus, CristateUa) one likewise observes a 

 budding in the egg, but it occurs at a much later period; it is only 

 when the embryo, having already clearh' diiferentiated two ecto- 

 dermic and mesodermic lamella?, is going to transform into a free 

 larva that it buds forth many polypoids at its aboral pole. 



In the worms, Kleinenberg in 1879 made known the curious case 

 of Lmnbr'icus trapesoides, in which the egg develops into two 

 embryos; here the multiplication is caused by a sort of internal bud- 

 ding, interposed at the gastrula stage, before the differentiation of 

 the lamella has yet taken place. 



In the Tunicates the Di-plosomas offer a curious case of precocious 

 budding, which gives the illusion of the simultaneous formation of 

 two embryos in the same egg\ but in realty it is a case of the internal 

 budding of an already differentiated embryo (Salensky, Caullery, 

 Pizon, Perrier). In Pyrosorna the budding likewise takes place in 



a In obedience with the contractions of the stomach walls the polygerminal 

 mass of Polygnotus is cariied by rythmic movements, which transport it by 

 turns from in front backward and fi'om behind forward. This movement, of 

 which the effect can be compared to that of the shaking of eggs, would have, 

 according to Marchal (1904), an influence toward the division of the germ. 



