Mr. W. S. Kent on two new Siliceous Sponges. 219 
dency to divide superiorly, where they become lost in the layer 
which composes the roof of the vaulted n enclosed be- 
enser than the 
outwards, while the opposite attenuate extremity passes through 
the superior layer of sarcode into the fascicles of large acerate 
spicula forming the upright supporting pillars. A few of 
these last acerate spicula are also occasionally met with pro- 
jecting among the spinulate ones. The third and last type of 
ipei, ^ palmato-inequianchorate," occur, but not abun- 
antly, in the sarcode investing the supporting pillars; and 
with these may also occasionally be found a few stray fascicles 
of the smaller acerate form. 3 
No especial oscular system appears to belong to this sponge, 
and it would seem probable that all matters of nutrition are 
received through temporary apertures occurring in the thin 
cortical layer of the sarcode, rejectamenta being got rid of 
through the same channels. : 
Rhaphidotheca Marshall-Hallii are the generic and specific 
The second sponge I have to notice is, like the bare 
found associated with a coral, the single y oae obtai 
ifera, var. antho- 
provided with a distinet cortical skeleton; but the nature of 
this skeleton throughout makes it necessary to refer it to the 
