ORIGIN OF MBTAMBEIO SEGMENTATION. 81 



of a worm-like animal with well-difiFerentiated mesodermal 

 tissues (a view which on physiological grounds is hard to 

 accept) but are rather to be traced back to simple ectodermal 

 pits in the two- layered ancestor developed for purposes probably 

 of aeration and represented at the present day in the Coelen- 

 terata by the subgenital pits of the Scypho-medusae, in the 

 embryos of Arthropoda by the pits into the cephalic ganglia, 

 and in the Vertebrata by the canal of the central nervous 

 system.^ 



The essence of all these propositions lies in the 

 fact that the segmented animals are traced back 

 not to a triploblastic unsegmented ancestor but 

 to a two-layered Coelenterate-like animal with a 

 pouched gut, the pouching having arisen as a result 

 of the necessity for an increase in the extent of the 

 vegetative surfaces in a rapidly enlarging animal 

 (for circulation and nutrition). 



The hypotheses are based upon the embryonic development 

 of the respective organs in the Triploblastica and the structure 

 of the living Coelenterata ; in other words, upon facts precisely 

 of the same nature as those which have been used in tracing 

 the evolution of the nervous and muscular tissues. 



Before proceeding to discuss the facts upon which the hypo- 

 theses rest, I may be permitted again to point out that it is no 

 part of my view to derive segmented animals direct from the 

 Coelenterata, but to derive both Coelenterata and segmented 

 animals from a common Coelenterate-like ancestor, whose 

 structure can only be elucidated by studying the anatomy and 

 the development of the living Ccelenterates, and of the higher 

 segmented animals. 



1 Sedgwick, "On the Original Function of the Canal of the Central 

 Nervous System of the Vertebrata," ' Proc. of Cambridge Phil. Soc. ' vol. iv. 



