186 CEYLON PEARL OYSTER REPORT. 



from Hymeniacidon. The position of the genus amongst the Axinellidse is, of course, 

 open to question, but it seems to come at least as naturally here as anywhere else. 



Hymeniacidon petrosioides, n. sp. Plate XII., fig. 4. 



The type specimen (R.N. 151) is massive, cushion-shaped, strongly convex above 

 and irregularly concave below, having evidently been attached to the substratum at a 

 few points only. Upper surface fairly even but coarsely granular. Vents (probably) 

 few, small and scattered. Texture very hard and compact ; surface harsh to the 

 touch. Colour (in spirit) dull grey, with a purplish tinge here and there. Greatest 

 breadth about 47 millims. ; maximum thickness about 19 millims. There is another 

 much smaller specimen of irregularly massive form. 



The skeleton is a dense, close-meshed reticulation of short, stout styli, in which one 

 can readily distinguish stout, multispiculous main fibres running at right angles to 

 the surface at distances of about one spicule's length from one another, and connected 

 crosswise by isolated spicules and bundles of spicules, with other similar spicules 

 irregularly scattered in the soft tissues. 



Spicules. Short, stout, more or less curved or bent styli (Plate XII., fig. 4), 

 broadly rounded off at the base (occasionally slightly tylote) and gradually sharp 

 pointed at the apex ; fairly uniform in size, measuring about 0'39 millim. by 

 0'022 millim. (There are apparently no oxea. ) 



This species appears to be nearly related to TOPSENT'S Stylinos jullieni from the 

 Atlantic (48), but the spicules are much larger. In external appearance it bears a very 

 close resemblance to Thrinacophora durissima, with which it may easily be confounded 

 until microscopically examined, and, indeed, I am inclined to think that these species 

 of Hymeniacidon are really closely related to the massive species of Thrinacophwa. 

 We may also have here, so to speak, a point of contact between the Axinellidse and 

 Desmacidonidse, indicated both by the form and arrangement of the megascleres 

 and the presence of trichodragmata in the last-named genus. The spicular fibre in 

 Hymeniacidon is not plumose, or, at most, very feebly so, but it is impossible to draw 

 a hard and fast line between the plumose type of fibre characteristic of the Axinellidse 

 and the non-plumose type characteristic of the Desmacidonidse, &c. 



R.N. 151 ; 316 (Ceylon seas). 



Thrinacophora, RIDLEY and DENDY. 



Axinellidse with typically plumose skeleton and with microscleres in the form of 

 trichodragmata. 



Thrinacophora agariciformis, n. sp. Plate XII., fig. 6. 



Sponge consisting (usually, at any rate) of a short, thick stalk (which has evidently 

 been attached below), supporting a thick, rounded, cushion-shaped body which is 

 irregularly depressed above so as to form a more or less shallow, very thick-walled 



