494 THE STOLOXIFERA AND ALCYONACEA 



The anatomy of Tuhipora has already been fully described (12 and 8) but it 

 seemed desirable, since Dr Ashworth (1) has shown that there is a marked divergence 

 from the general anatomical features of the Alcyonarian stomodaeum and mesenterial 

 filaments in the genus Xenia, to reinvestigate the structure of these organs in 

 Tuhipora. 



The stomodaeum possesses an ectodermal epithelium of clear columnar cells as 

 in Alcyonium and other forms, and there is a well-marked siphonogh-ph armed with 

 long cilia. No gland cells are found in this epithelium. There are six short ventral 

 mesenterial filaments bearing gland cells. These filaments begin on the free edge of 

 the mesenteries in contact with the inner opening of the stomodaeum. In the con- 

 tracted spirit specimens they may be seen to extend about 5 mm. down the free 

 edges of the mesenteries. It is impossible to estimate what their exact length may 

 be in the living expanded polyps. The two dorsal mesenterial filaments are very 

 much convoluted and extend from the stomodaeum as far down as the first tabula. 

 There are no gland cells in their epithelium. 



Judging from these anatomical facts then, the digestive processes of Tuhipora 

 must be carried on as in Alcyonium and Kophohelemnon, the digestive fluid being 

 secreted by the six ventral mesenteries and not in any degree by the epithelium of 

 the stomodaeum nor by the epithelium of the dorsal mesenterial filaments. 



The specimen was a male. The young sperm sacs in their follicles were found 

 on the six ventral mesenteries only, as described and figured by von Koch\ 



2. Glavularia viridis Q and G. 



A small piece of a colony of this species was sent to us in a partially retracted 

 condition. It was obtained at Welle Island (Sanaroa), D'Entrecasteaux group, in 1895. 



The specimen differs from those obtained in Celebes (9) in being decidedly 

 stouter in build. The polyps are not very long (about 35 mm.), springing from a 

 creeping stolon attached to a piece of dead coral. The terminal extremities of the 

 polyps have an inflated appearance due, undoubtedly, to the manner in which they 

 were killed. They are from 40 — 50 mm. in length and 8 mm. in diameter at the 

 free end. Only one transverse tubular connection between the polj'ps above the level 

 of the stolon was seen. 



The spicules vary considerably in length but the average is about 2'5 mm. 

 The tubercles are rather less prominent than those of the spicule figured by one of 

 us (9), but similar to those of other spicules in the collection from Celebes. 



In the larger spicules of Dr Willey's specimen there may be frequently seen a 

 curious bifurcation of one end. The same has been noticed in the Celebes specimens. 

 The lower, and therefore older, parts of the body wall are strengthened by the forma- 

 tion of horny fibres in the mesogloea, making the texture hard and brittle. 



' On referring to my preparations made in 1884 I find that 'I was in error in the statement I made 

 then that the gonads occur on the dorsal mesenteries in the specimen from Zanzibar. 



It is extremely probable that Dana (2) was mistaken in stating that the specimen he examined was 

 hermaphrodite, six of the mesenteries bearing sperms and two ova. Hermaphroditism is of very rare occurrence 

 in Alcyonarians and the presence of gonads on the dorsal mesenteries has not yet been satisfactorily proved 

 in any genus. S. J. H. 



