538 REVISION OF THE AXINELLIDiE, ii., 



gradually more and more divergingly disposed, as well as more 

 loosely compacted, and the fibres finally assume, in consequence, 

 a typically plumose aspect. The spongin cementing the spicules, 

 — though necessarily fairly considerable in amount owing to 

 their loose arrangement, — is, except in the stalk and oldest por- 

 tions of the skeleton, usually of scarcely more than the minimal 

 quantity required to hold them together, and seldom or never 

 forms a distinct sheath: where the spicules lie more widely 

 apart, it often becomes reduced to a mere film between them, 

 and here and there even leaves small open spaces or fenestrse. 

 Running upwards from the stalk, and continuing for some dis 

 tance into the body of the sponge, gradually dissolving as they 

 proceed, are a number of relatively stout strands of spicules, or 

 funes, evidently formed each by the fusion of several originally 

 separate fibres (PI. xxxvi., fig. 2). Connection between the main 

 fibres, apart from occasional anastomosis or direct union between 

 them by inosculation, is partly by means of relatively few, 

 obliquely-running multispicular fibres, similar in character to 

 the main fibres except in being usually of lesser stoutness, and 

 partly by means of connecting fibres proper. The latter are 

 mostly very slender, and usually contain few spicules or are 

 composed of spongin alone; they occur at irregular intervals, 

 sometimes singly, sometimes several together, and in the latter 

 case usually interunite also among themselves 



In sections of the sponge with the soft tissues intact, the 

 appearance of the skeleton is somewhat different. The presence 

 of spongin is scarcely apparent; the main fibres have a much 

 looser and more plumose aspect; and the connecting fibres are 

 seldom definitely recognisable as such, owing to the difficulty of 

 distinguishing between the megascleres actually constituting 

 them and others that are merely scattered between the fibres. 

 The more diftuse and plumose appearance of the main fibres is 

 probably due to the fact that some proportion of the more ex- 

 teriorly situated (and likewise more obliquely directed) spicules 

 entering into their formation are not attached by spongin, and 

 consequently are absent from the skeleton that remains after 

 maceration. In the more peripheral parts of the skeleton, the 



