216 CEYLON PEARL OYSTER REPORT. 



SCHULZE (72) in Cacospongia scalaris. POLEJAEFF (74) has already called attention 

 to variability in the development of the cameral canaliculi in this species, so that we 

 can hardly consider the presence of long exhalant canaliculi as a character of specific 

 value. 



RN. 56 (Gulf of Manaar). 



Hippospongia dura, LENDENFELD (66). 



The single specimen consists of an irregular massive body tapering gradually 

 upwards (?) or on one side (?) into a sub-cylindrical fistular process, the conical 

 extremity of which is perforated by numerous small, round apertures leading into the 

 internal vestibular space. Similar apertures are scattered more sparsely on other 

 parts of the sponge. Internally the sponge is cavernous, with wide, sub-cylindrical, 

 vestibular spaces, more or less sub-divided by trabeculse of smaller diameter than the 

 vestibules. The vestibular spaces are covered in at the surface partly by a somewhat 

 parchment-like dermal membrane and partly by superficial extensions of the trabeculse 

 themselves. The surface is rather uneven, but not distinctly conulose ; a consider- 

 able amount of calcareous and other foreign matter is attached to it. The texture, 

 in spirit, is hard and only slightly compressible, and the colour rather dark brown 

 throughout. The specimen measures about 110 millims. in greatest length (height?) 

 and 60 millims. in greatest breadth at right angles to the length. 



The skeleton is an unusually close and fairly uniform network of rather stout, 

 amber-coloured horny fibre. There is no distinction between primary and secondary 

 fibres, except at wide intervals, where the network becomes somewhat closer and 

 forms stout columns radiating towards the surface. In these columns many primary 

 fibres run side by side and nearly parallel with one another, connected at frequent 

 intervals by short, transverse secondaries to form a very stout, but ill-defined, com- 

 pound, trellis-like fibre. Elsewhere the stouter fibres frequently run parallel with 

 the surface of the sponge and are connected together by more slender fibres which 

 run transverse to the stouter ones. In other places again the network is quite 

 irregular. The stouter fibres, forming the bulk of the skeleton, are about 0'04 millim. 

 or 0'05 millim. in diameter; the more slender ones are very variable. Usually the 

 fibres are quite free from foreign bodies, but broken spicules may be found occasion- 

 ally in fibres of the trellis-like groups. There is no special dermal skeleton, but the 

 ordinary reticulation of fibres comes close to the surface, which is covered by a thin 

 cuticle like that of Hippospongia intestinalis and H. anomala. A remarkable feature 

 of the skeleton fibre is the brilliant yellow colour which it assumes in sections stained 

 with picro-carmine. The superficial cuticle stains in the same way, and sometimes 

 appears to be continuous with the more superficial skeleton fibres, but this is a point 

 which requires further investigation. 



The flagellate chambers are nearly spherical, only about 0'024 millim. in diameter, 

 and the ground-substance between them is finely granular. There is a rather thin, 



