438 
further resemblance is shown in the fact that in this recent genus also the linear and 
other spicules are partially united in an irregular manner by transverse bulks and 
siliceous estensious as in Purlsiplionia. Likewise in Ihe existing genera, Uertwlgia, 
0. Schmidt, and Jthabdo 2 }eciella, O. Schmidt, the skeletal spicules are firmly cemented 
together in the lower portion, whilst they are free in the upper portion of the sponge, 
thus exhibiting, as Oscar Schmidt remarks,* transitional characters between the 
Lyssakina and Dictyonina. These recent genera have been, however, ranked with the 
Lyssakina Group, and, considering the nature of the spicules in PurisijjJionia, it should, 
in my opinion, also be placed in the same group. In its massive wall of closely arranged 
spicular fibres, all firmly and irregularly cemented together, and penetrated by definitely 
bounded branching canals, this genus is very distinctly marked off from any other 
Ilexactinellid, whether fossil or recent. 
PuEisiPnoJfiA Clahkei, PowerbanJr, PI. 19. 
Purisiphonia Clarkci, Bowerbank, Proo. Zool. Soc., 1869, p. 312, PI. x.'iv., fig.s. 6 anti 7. 
„ „ Moore, Quart. Journ. Geol. Soo., 1870, xxvi., pp. 235, 210, PI. xvii., fig. 1. 
„ „ Sollas, Quart. Journ. Geol. Soc., 1877, xxxiii., p. 802. 
,, ,, Carter, Ann. and Mag. Nat. Hist., 1878, i., p. 376. 
„ ,, Zittel, Handb. Pal., 1878, i., p. 179. 
,, „ Eth. fil., Cat. Austr. Eoss., 1878, p. 104. 
,, ,, Hinde, Cat. Foss. Sponges Brit. Mus., 1883, p. 124. 
„ „ Vosmaer, Bronn’s Klassen nnd Ordnuiigen des Thicr-Reichs, 1885, 2 er. Bd. (Porilera), 
p. 206. 
“ Sp. Char. Ihe largest fragment of this species which has yet been discovered is 
apparently the main tubular stem of an individual with the bases of two or three 
branches. Tlie fragment is about one hundred mm. in length and breadth, and the 
walls are about fifteen mm. in thickness. 
“ The oscules are numerous, disposed without regular arrangement, from one to 
two mm. apart, and varying from one to three mm. in diameter. The canals very 
slightly diminish in their course in the wall until they bifurcate. 
“ The spicular fascicles of the skeleton are about twelve mm. iu length and from 
fifteen to twenty-five mm. in thickness, the linear spicules composing them, about nine 
or ten in number, are in close contact find so united by silica that the compound 
character of the bundle can in some cases only be clearly seen by the axial canals of the 
spicules (PI. 19, fig. 6). The linear spicules, not in fascicles, are, as a rule, much stouter 
than those forming the fascicles ; the connecting transverse processes are numerous, and 
so bridge over the spaces between the spicules that hut little more than oval or circular 
fcnestrie exist between them. Though, as a rule, these connecting bulks are stouter 
than the spicules, there are no canals present iu them, whereas they can be traced in 
most of the linear spicules. The groups of minute normal hexactinellid spicules appear 
to he very irregular in their distribution ; in some instances they are massed near 
the terminations of the canals. Their rays appear to be sub-equal, cylindrical, or 
tapering but very slightly and with rounded ends. The rays vary from '1 to T3 mm. 
in length, and ’02 mm. in thickness. These spicules, as shown in the figure (PI. 19, 
fig. 10), are very indiscriminately mingled together, and their rays are fused wherever 
they are iu contact with others. 
Obs. “This species was first described and figured by Dr. Bowerbank, who received 
the specimens from Mr. Charles Moore. He constituted for them the genus Picrisi- 
phom’a, a rather inappropriate name, since it might bo thought to indicate a connection 
with the Lithistid genus Siphonia, with which this form has nothing in common. 
Hie Sjjongien des Meerbusen von Mexico, 2te. Abth., pi). 42 and 62. 
