ORIGIN OP MBTAMBEIO SEGMENTATION. 69 



PART II. 



Application of the above Hypotheses to the Verte- 

 BRATA, Annelida, Arthropoda, Mollusca, and certain 



SMALLER GROUPS. 



Fig. 7 represents a diagram of the ideal ancestor of all the 

 above-mentioned Triploblastica. It closely resembles the 

 common ancestor of the Coelenterata but may be supposed a 

 little more advanced in specialisation. For instance, the peri- 

 pheral excretory pores (o) have a regular arrangement. This 

 animal is supposed to have a bilateral symmetry shown in the 

 gut pouches and in the excretory pores. It is supposed to 

 have an elongated mouth partly differentiated into two parts, 

 and the nervous system is generally diffused over the oral 

 surface (which will henceforth be called the neural surface) 

 with a tendency to specialisation into a narrow tract. 



This ideal ancestor soon gave rise to two stocks, the first 

 differences between which may be supposed to depend on the 

 shape of the body. 



In the one stock the mouth and anus (which soon became 

 separated) remain on the neural surface, a prseoral lobe was 

 developed on the abneural surface of the body (fig. 12) ; this 

 prseoral lobe being carried first in movement became specially 

 sensitive and the nervous system largely developed. 



This stock is the Invertebrate stock. The prseoral part of 

 the nervous ring in consequence of the shape the body has 

 taken becomes enlarged and sense orgaas largely developed in 

 connection with it. The hinder praeanal parts of the nervous 

 ring have more or less approximated to each other, and are con- 

 nected by commissures and become swollen at intervals where 

 many nerves pass out to the locomotive organs (appendages). 

 The postanal part of the ring becomes weak and often dis- 

 appears, never having more than a commissural function 

 (absence of nerve-cells in postanal connection of lateral nerve 

 trunks of Peripatus, vide Balfour on Peripatus ca- 

 pensis). 



