in PHYLUM AND CLASS PORIFERA 105 



of the contained paragastric cavities, and in many other more 

 minute points ; in fact, we find as a result of the comparison that no 

 two specimens are exactly alike. These differences are so great 

 that some very distinct races or varieties of S. gelatinosuni have 

 been recognised, and some have received special names. HIT. 

 again, as in the case of the families and orders, the distinctions 

 are of an arbitrary character some writers on Sponges setting 

 down as several species what others regard merely as varieties of 

 one species. It is impossible, in fact, to draw a hard and fast line 

 of distinction between species and varieties. In the higher groups 

 of animals the attempt is made to establish a physiological dis- 

 tinction ; all the members of a species are regarded as being fertile 

 inter se, and capable of producing fertile offspring as a result of 

 their union ; but such a mode of distinguishing species is impos- 

 sible of application among lower forms such as the sponges. In 

 these lower groups, accordingly, a species can only be defined as 

 an assemblage of individuals which so closely resemble one an- 

 other that they might be supposed to be the offspring of a parent 

 form similar to themselves in all the most essential features. 

 And, according to the view taken of the relative importance of 

 different points of colour, shape, and internal structure, the con- 

 ceptions of the species and their varieties and mutual relationships 

 formed by different observers must often differ widely from one 

 another. 



3. GENERAL ORGANISATION. 



General Form and Mode of Growth. The simplest Sponges 

 are vase-shaped or cylindrical in form, either branched or un- 

 branched, and, if branched, with or without anastomosis or 

 coalescence between neighbouring branches. But the general 

 form of the less simple Sponges diverges widely from that of such 

 a branching cylinder as is presented by Sycon gelatinosum 

 (Fig. 69). 



From the point to which the embryonic sponge becomes 

 attached it may spread out horizontally, following the irregulari- 

 ties of the surface on which it grows, and forming a more or less 

 closely adherent encrustation like that of an encrusting lichm 

 (Fig. 75, A). The surface of such an encrustation may br smooth ; 

 more commonly it is raised up into elevations rounded bosses 

 cones, ridges or lamellae ; and the edges may be entire or lnl>r.|. 

 In other cases the sponge grows at first more actively in tin- 

 vertical than in the horizontal direction, and the result m;t\ ln> M 

 long, narrow structure, cylindrical or compressed, and more or Irs 

 branched (Fig. 75, B}. Sometimes vertical and horizontal growth 

 is almost equal, so that eventually there is formal ,-i thick 

 mass of a rounded or polyhedral shape (Fig. 75, C), with an even, 

 or lobed, or ridged surface. Very often, after active vertical growth 



