590 ARTIHIR WITiriRY. 



dcfiiicd as the trophohlast by Ilubrcclit (5), since its prin- 

 cipal function is to provide for the nutrition of the embryo. 



More recently (6) Ilubrecht has published a very remark- 

 able and what I venture to predict will become a classical 

 tiieory of the mammalian tropiioblast, in whicii he seeks to 

 demonstrate a phyletic relationship between the latter and the 

 superficial ectoderm ("embryonic epidermis," Balfour; "Dcck- 

 schicht," (joette) of the embryos of Amphibia. 



Ilubrecht points out that the vertebrate amnion is a de- 

 rivative of the trophoblast, which is a structure sui jijeneris. 



If the vertebrate amnion, in its capacity of derivative of 

 the trophoblast, has a |)rof()und phylogenetie significance, one 

 would be inclined to su})[)ose that an analogous significance 

 would attach to the embryonic membranes of insects. 



Hitlicrto the prevailing impression seems to have been that 

 the latter were of merely eenogenctic importance, and all 

 serious attempts to account for them morphologically have 

 been more or less coloured by this assumption. 



The embryos of a species of Peripatus which I found in 

 New Hritain last year (1897), of which I have recently pub- 

 lished a full description (14), seem to me to point the way out 

 of this somewhat barren and unsatisfactory position, and to 

 endow the embryonic membranes of insects with a singular 

 interest. Because, if it could be shown that a common prin- 

 ciple governs the theories applied to the explanation of the 

 embryonic membranes of insects and of those of the higher 

 vertebrates, the one theory would constitute an important 

 complement of the other. 



On this occasion I am only dealing with the embryonic 

 membranes of insects, because they are the best known, and 

 my own observations only bear upon these structures. There 

 can be little doubt that the embryonic membranes of scorpions 

 are capable of being explained in an analogous manner, only 

 the data are not suflicient in this case. 



My own idea is that three theories are necessary for the ex- 

 planation of the embryonic membranes of scorpions, insects, 

 aud vertebrates, and that a common principle underlies the 



