254 PLATE LXXX. 



it in 1867 off Balta, Slietlaud. It coats the surface of 

 a small fragment of slate-coloured schistus not more 

 than four lines square. The thickness of the sponge 

 varies from not exceeding that of joaper to about half 

 a line. The surface is very uneven, rising in several 

 places into minute hillocks, and it is in many parts 

 covered with minute grains of sand which are too 

 firmly adherent to be readily washed off. When not 

 thus covered it is quite smooth. The oscula were not 

 readily to be detected in Canada balsam, but by dii'ect 

 light under a power of about 100 linear they were dis- 

 tinctly visible ; they were simjjle and minute. No 

 pores could be detected. 



The skeleton is loosely fasciculated and very irregu- 

 larly disposed ; many of the bundles are multisjDicu- 

 lons, while in other cases they contain not more than 

 from two to six or seven spicula ; and numerous single 

 spicula cross them in every possible direction ; and in 

 some of the thickest portions of the sponge they are so 

 numerous as to render the skeleton structures very 

 like those of a Hymeniacidon, but in the thinner por- 

 tions the distinction between the two modes of struc- 

 tui'e is very apparent in consequence of the spicula form- 

 ing a fasciculus, having their bases and apices always 

 coincident, a mode of arrangement never met with in 

 a Hymeniacidon. The spinulate spicula are large, 

 long, and rather slender, and intermixed with the 

 fully-developed skeleton ones are a few, exceedingly 

 slender and frequently flexuous. The fully-developed 

 spicula vaiy to a considerable extent in the degree of 

 the production of their globular bases, varjuug from 

 completely spherical down to almost purely acuate, but 

 the latter form is of rare occurrence. I could not 

 detect any other form of spiculum in the sponge than 

 those I have described above. 



