HYALASCUS SAGAMIENSIS. 93 



directly adjoining dermalia are, as usual, apposed side by side 

 throughout their entire length. 



The ffasiralia (PL VII., üg. 4; PI. VIII., fig. 2) are all 

 hexactins in which the free proximal ray is much longer than 

 any of the other rays. Length of paratangentials, 90-110 n. 

 Distal ray as long as paratangentials or somewhat shorter. 

 Length of proximal ray, 185-275 /^. Breadth at base of rays, 

 10-14 /J. All the rays taper very gradually towards the coni- 

 cally pointed ends. (PI. VIL, fig. 4, does not faithfully repre- 

 sent this point. The general shape of the gastralia is better 

 shown in PI. VIII. , fig. 2). Except at the base of rays and on 

 the central node, Ijotli of which parts are generally smooth, the 

 surface is beset with numerous microspines similar to those on the 

 dermalia. The microspines on the free proximal ray may l^e 

 slightly more strongly developed and more distinctly outwardly 

 directed than those on the other rays. In my preliminary de- 

 scription ('96) of the species I have said that the gastralia, on 

 account of their specially developed proximal ray, might be 

 called hexactin-pinules. That statement I beg now to withdraw 

 for fear that it may lead to an over-estimation of the degree of 

 differentiation shown by the proximal ray. The gastral hexactins, 

 it may be said, are no more specially characterized than are 

 those with prolonged proximal rays in certain other Possellids 

 (f. i., SiaurocalyptuB glaber, lihabdocalyplm ungaiculatus, etc.) 



Oxyhexasters (PL VIL, figs, 7-10), re])resented by heuiihex- 

 actinose and somewhat less frequently l)y stiictly hexactinose 

 forms, are abundantly present in the choanosome as well as in 

 the gastral layer. Normally developed oxyhexasters, in which 



