282 Dr. I). Rosa on some 



level of the outer ventral setre, but I could not determine 

 which segment bears the first of them. 



Septa 5-6 to 8-9 inclusive are thin, the following 9-10, 

 10-11, 11-12, and also, but to a less degree, 12-13, are 

 thickened. 



A gizzard is not recognizable, but septum 6-7 is more 

 deeply infundibulate than its neighbours, and we may 

 connect this with the earlier existence of a gizzard in the 

 6th segment. There are no calciferous glands; the intestine 

 begins behind the 18th segment, perhaps in 16 or 17. 



The hearts occupy segments 11, 12, 13, the last being the 

 largest. 



The two pairs of large spermathecse belong to segments 

 8 and 9 ; each spermatheca consists of a pyriform pouch not 

 distinctly marked off from its duct, and of a narrow tubular 

 diverticule which is longer than the main pouch. 



The broad grape-like ovaries are readily seen in seg- 

 ment lb*, as well as the testes in segments 10 and 11, all 

 these gonads being attached to the anterior septum on each 

 side of the neurochord. In front of the gonads, that is, on 

 the anterior face of septa 10-11, 11-12, and 13-14, the 

 funnels of both pairs of vasa deferentia and of the oviduct 

 are plainly visible. 



The small sperm-sacs in segments 11 and 12 have a 

 botryoidal appearance. 



The prostates occupy segments 16, 17, and 18, and recall 

 very nearly those of P. insularis (Rosa). Their glandular 

 portion has the appearance of a large sausage-like body, 

 formed by the apposition of the several parts of a slightly-coiled 

 lesser duct. The muscular duct which arises from the front 

 end of the glandular tube is moderately bent, with the con- 

 vexity inwards, and gradually increases in diameter as it 

 proceeds backwards, reaching at last the external openings on 

 the 18th segment. 



I could not see exactly where the vas deferens joins the 

 prostate, but I have little doubt that the connexion be- 

 tween both structures will be found to be the same as that 

 which has been described by Akira Jizuka for P. matsu- 

 shimensis. 



Our species seems to be closely allied to P. insularis (Rosa), 

 which I first described from specimens obtained in the Am 

 Islands *, and which has been more recently found also at 

 Ceylon (Michaelsen) f. Still a marked difference between the 

 two species exists, as in the spermatheca of P. insularis both 



* Ann. d. naturk. Hufinuseuiu, Wien, Ed. vi. 189."). 



t Mitth. aufi d. naturmstor. Museum, xiv. (Hamburg, 1897). 



