170 GENERAL CONSIDERATIONS. 



mature condition, and the coelomoducts, having the form of canals of about 

 equal calibre throughout, lead in the usual fashion to the exterior. Their lining 

 epithelium is composed of low, usually cubical cells, non-vacuolated, without 

 any traces of concrements or crystals, and indications of any glandular activity 

 whatsoever are totally absent. I am therefore strongly of the belief that the 

 coelomoducts in the Neomeniina are solely concerned in the reproductive process. 



Granted that the section of the alimentary canal, including the cirrose 

 area and the buccal sensory ridges, is the homologue of the Chiton snout, or at 

 all events a comparatively late formation, the remaining portions are decidedly 

 Chiton-like. In most species there are both dorsal and ventral salivary glands 

 which show a surprising amount of variation, ranging from scarcely distinguish- 

 able bodies to others of great size and a high degree of complexity. As in the 

 case of the Chitons the dorsal set typically opens through the dorsal buccal wall 

 while the outlets of the ventral pair are in the neighborhood of the radula. 



The radula, in a number of species is lacking, and in several others it is in a 

 degenerate state, being reduced to a peg-like body (Chaetoderma) or to a very 

 few teeth which are reported to lack a basement membrane. On the other 

 hand there are species, such as Kruppomenia (Nierstrasz, '05) and Limifossor 

 (Heath, '05) which have typical radulae, as regards location and component parts. 

 Odontoblasts, cells which form the basement membrane, and enamel cells are 

 all present, and the resemblance to the radulae of other molluscs, as figured by 

 Rossler for example, is surprisingly complete. 



The fate of the subradular organ appears to depend closely upon that of 

 the radula. In every case that has come under my observation it is lacking 

 or is reduced to the merest rudiment when the radula has disappeared, and in 

 some species it is in a highly degenerate state when the radula is in a similar 

 condition. It is to be noted that the nerve supply to this organ may be a much 

 more conservative set of organs, persisting in Strophlomenia scandens, for example, 

 after the organ has ceased to exist as a well-defined structure. 



The limits of the posterior end of the pharynx and consequently of the 

 anterior end of the oesophagus are not sharply defined histologically and in 

 the absence of embryological evidence they remain problematical. In fact the 

 oesophagus is sometimes disregarded as a definite section of the gut or is included 

 in the description of the pharynx. In some species it appears to be bounded 

 anteriorly by a circular fold and posteriorly it is probably terminated at the 

 commencement of the stomach-intestine. 



The intestine is differentiated into a well-defined stomach and intestine, 



