646 HAROLD HEATH, 



that at a time when the detailed history of the development of the 

 Annelid trochophore is known in not more than half a dozen forms 

 broad generalizations are impossible or hazardous at best, but with 

 the hope that the following hypothesis may contain some grains of 

 truth and may serve to show some of the present limits of the problem 

 it is herewith presented. 



An attempt has been made in the preceding pages to show that 

 there are the strongest reasons for holding that the Annelid and Chiton 

 trocliophores are very closely related organisms, and that the various 

 quartettes of cells develop along much the same lines and are homo- 

 logous. Consequently the first somatoblast in the Annelid is the 

 homologue of that in the Chiton and is also I believe to be looked 

 upon as a further development of such a type. And not only so but 

 there are some reasons for the belief that metameric segmentation 

 may have arisen from an ectodermal segmentation similarly located 

 to that in the Polyplacophora. 



It is not claimed that metamerism in Annelids has been directly 

 derived from the incipient segmentation found in the dorsal ectoderm 

 of the Chitons. Such a claim would be involved in insuperable dif- 

 ficulties for the present at least. Nevertheless the two forms have 

 some points in common which indicate that segmentation in each may 

 have originated in much the same way. In both it aifects the first 

 somatoblast and is consequently confined to the trunk region; the 

 groove between the head and first trunk segment occupies the same 

 position as the line separating the first and second valves of the shell 

 in IscJmochiton ; the growth zone is located in the same position in 

 both; and in its later development as will be shown presently, 

 metameric segmentation gives other evidences of having been derived 

 from a type much hke that existing in the Polyplacophora. 



It is not known how the segmented shell arose in the Chitons. 

 It appears probable that a dorsal cuticular covering arose in the trunk 

 region of the trochophore which subsequently became charged with 

 lime salts. This structure became variously modified, being coiled in 

 the Gastropod, divided along the median dorsal line into two segments 

 in the Lamellibranch , and transversely into eight in the Polyplaco- 

 phora. Whether this cuticular covering was present in the trochophore 

 before the Annelid and Mollusc diverged cannot at present be deter- 

 mined, nor whether Annelid segmentation arose in connection with it 

 as in the Chitons, but if it be assumed that metamerism, whatever 

 may have been its nature, arose in the ectoderm on the dorsal surface 



