274 



'ENDEAVOUE" SCIENTIFIC EESULTS. 



towards the outer surface, they are supplemented by the inter- 

 polation of others which arise from their connecting fibres. 

 Their connecting fibres either may be simple, forming direct 

 transverse connections between them — in which case the re- 

 ticulation is more or less rectangular — or, less frequently, may 

 branch and interunite, thus forming between them an interre- 

 ticulation, and rendering the pattern of the skeleton irregular. 

 The fibres are here and there, at very wide intervals, echinated 

 by spicules (similar to those within the fibres) which, as a rule, 

 project but slightly beyond the spongin. 1 he meshes of the 

 reticulation are tympanized by very thin membranes (visible, 

 perhaps, only in washed-out specimens) in which moderately 

 abundant slender spicules lie scattered. 



There is no indication, in the present condition of the speci- 

 mens, of a dermal membrane or of a dermal skeleton. The 

 main fibres terminate abruptly at the surface whh their end- 

 spicules projecting slightly beyond the spongin. 



Spicules. ■ — Microscleres are absent; the megascleres are, for 

 the purpose of description, separable into three groups : — 

 (i.) The. spicules of the fibres, straight styli, cylindrical 

 to within a short distance of their pointed extremity, 

 ranging in length from 80 to about 12011, and rarely 

 slightly more than 4 p in diameter ; occasionally an 

 oxea is found amongst them. 



(ii.) Straight or but slightly flexed interstitial spicules, 

 identical in form w ith the preceding, but mostly much 

 longer, attaining to a length of 220 /t, and not more 

 than 3.5 p in diameter. Oxeote modifications of these 

 occur more frequently than amongst the fibral spi- 

 cules but are nevertheless comparatively rare. 



(iii.) Variously curved and flexuous, very slender, intersti- 

 tial spicules, about equal in abundance to those of 

 the second group, frequently exhibiting oxeote modi- 

 fications, seldom exceeding 1.5 /i in diameter, and of 

 very variable length ; the longest observed measured 

 264 fi. 



The identity in form of the spicules of the first two groups 

 and the complete transition between them in the matter of size, 

 permit of no other conclusion than that they are but partially 

 differentiated derivatives of an originally single spicule-form. 

 I regard them as homologous with the priucipal megascleres 

 of other species. Whether the spicules of the third group are 

 merely variants of the same, or whether they are represent- 

 ative of auxiliary megascleres, I am unable to decide. 



