PLATE XC. 327 



lines of the skeleton, when they reach the mner 

 surface of the dermis, instead of terminating there, or 

 of passing through that organ, are spread out beneath 

 it to some distance in long slender lines, so that when 

 viewed with a low power through the dermal mem- 

 brane, they might be readily mistaken for the fibres of 

 a very slender Desmacidon. This peculiarity in the 

 distribution of its structures, at the first sight, com- 

 plicates their characters to a very considerable extent ; 

 lout when once recognised it Ijecomes a valuable dis- 

 criminative character. 



From tlie extremely irregular construction of the 

 skeleton it is rather difficult to say whether there are 

 any tension spicula on the interstitial membranes ; 

 and although I searched diligently for them I could 

 not find any of the anchorate retentive spicula on 

 them. 



The only species with which J. clitbia is liable to be 

 confounded is I. gracilis ; their skeleton structures are 

 very similar, but in the latter the spicula of the primary 

 lines are much stouter than those of I. duhia, and in 

 that species the external form is massive and sessile ; 

 while in I. gracilis it is ramose, and the deinnal mem- 

 brane is aspiculous. 



Desmactdon rotalis, Bowerhanl: 



Plate XC. 



Sponge parasitical on zoophytes or fuci. Surface 

 even and smooth. Oscula simple, dispersed, minute. 

 Pores inconspicuous. Dermis retiform, rete arranged 

 in hexradiate confluent rotulatc areas ; radii compact 

 and multispiculous ; retentive spicula bihamate, simple 

 and contort, few in number, and rarely bidentate, 

 iiiequi-anchorate, and dentato-palmate, inequi-an- 

 chorate, very minute. Skeleton compact, fibres 

 frequently dividing dichotomously, terminatin g abruptly 

 on the inner surface of the dermis and dividing to form 



