86 PROXEOMENIA HAWAIIENSIS. 



cells the secretion is discharged. The cells (if the ^hmd are comparatively large, 

 with small nuclei and an almndance of a lightly staining secret ion, and are grouped 

 into several clusters separated from each other by a small amount of comiective 

 tissue. 



The paired venti'al salivary glands are long tubulai- unbianched structures 

 opening into the pharynx on each side of the front end of the radula. Their 

 position and general appearance are represented (Plate 14, figs. 7, D). Each 

 constituent cell is high and cohmmar in form, conijioscil of vacuolated cytoplasm 

 and possessed with nuclei ranging from spherical to slender spindle-slia])cd forms 

 correlated with different stages of glandular acti\ity. The secretion within the 

 main duct is finely granular and has only a slight affinity for logwood dyes. 



In this species the radula is relatively well developed and is located as in 

 other members of the genus. The teeth are formed by odontoblasts of the usual 

 high columnar type characteristic of the Chitons and prosobranchs and are of 

 the form represented (Plate 34, fig. 13). All the teeth are of essentially the same 

 form and number not less than from thirty-eight to forty-five in each transverse 

 row. There is no indication of a median tooth so far as the sections show but 

 each tooth adjacent to the mid line is somewhat smaller than its fellows (Plate 34, 

 fig. 13). In some species of Solenogastres the teeth are reported to be merely 

 thickenings of a continuous cuticular jilate, but in this species they are clearly 

 distinct, a well-defined suture nt)t only separating each tooth from the others 

 but from the basal plate as well. 



Immediately in front of the radula and somewhat covered by its forward 

 border are two areas of high columnar cells (Plate 34, fig. 2) that are more or 

 less sunken in a well-defined sheath. In another place (Heath "04) it has lieen 

 shown that these organs probably correspond to the subradular organ in the 

 Chitons and some of the prosobranchs. Their innervation is discussed in the 

 section on the nervous system. 



The usual relation of jiharynx and stomach-intestine are shown (Plate 6, 

 fig. 2). In another specimen the anterior dorsal coecum is considerably more 

 developed and there is also a small ventral one that extends forward between 

 the salivary glands. A short distance behind the pharynx the cells of the di- 

 gestive tract shade gradually into the relatively high pyriform hepatic cells of 

 the stomach-intestine. There are strong evidences that the distal jiart of these 

 cells loaded with secretory products separates from the remaining nucleated 

 portions and dissolves in the alimentary tract and that the process is repeated 

 indefinitely, the basal imcleated parts develojiing anew the glandular distal 

 portions. 



