DEVELOPMENT OF THE CAPE SPECIES OF PERIPATUS. 527 



the cases, in which segmentation gives rise to a solid embryo 

 are quite as numerous, if not more so, than those in which the 

 reverse holds. 



I would even go further than this, and maintain that 

 Metschnikoff^s view, that the ectoderm is primitive and the 

 endoderm secondary — arising from the former by inwandering 

 — is not more in accordance with the facts of embryology than 

 the opposite view, viz. that the endoderm is primary, giving 

 rise to the ectoderm by budding-off cells — outwandering as it 

 may be called. In almost all Invertebrate groups there are 

 instances of the latter process, in which the ovum, before or 

 after division into two or more large cells, buds out a number 

 of small cells which form the ectoderm ; either itself, or the 

 large cells produced by it, persisting, and eventually, after the 

 appearance of a central cavity, forming the endoderm. 



I do not mean to say that the facts are more in favour of 

 this view than in that of Metschnikoff, but I think they 

 favour the one as much as the other. 



But granting the hollow blastula as a possible animal, I 

 agree with Biitschli in thinking that there are great physio- 

 logical difficulties in the way of accepting the process by which, 

 according to Metschnikoff, it may have become transformed 

 into a solid form. Surely a hollow sphere is in a much more 

 advantageous position with regard to nutrition than a solid 

 one; and yet Metschnikoff supposes that the transformation 

 into the solid form was due to the migration of surface cells 

 into the interior for nutritive purposes. The central part of 

 the animal being empty would require no nutriment, and even 

 if it did, what more convenient arrangement could there be 

 than a layer of cells to pass prepared nutritive substances into 

 it ? The ectoderm of Hydra or of Ccelenterates in general is 

 not fed by the migration of cells from the endoderm, and in 

 short we know of no instance in the animal kingdom of food 

 being carried from one part of an organism to another in 

 actively migrating cells. It seems to me much more likely, 

 if the ancestral Protozoon was a hollow blastula, that the first 

 diflfcrentiation would have been into locomotive sentient cells 



