378 Mr. H. J. Carter on Purisiphonia Clarkei. 
“fistulous,” used by Dr. Bowerbank (J. c.), since the latter 
implies a form which we are not justified in giving to the 
only fragment of this sponge brought before the public, even 
if, when entire, it could under the circumstances be called 
‘*fistulous.”” That the curved state of the specimen indicates 
a concavity, and that the large holes in its surface indicate 
growth in these directions respectively, with a wall averaging 
half an inch in thickness throughout, is all that can be said 
towards a description of the entire form, whatever this here- 
after may prove to be; but the forms of the same sponge are 
often so varied, that the next specimen may in this respect 
be totally different from the present one. In general structure 
(which, where the calcareous infiltration has been dissolved out 
by acid, is as beautiful and perfect almost as it ever could 
have been) Puristphonia Clarket is somewhat like the speci- 
men of Dactylocalyx pumiceus, Stutchbury, in the British 
Museum—that is, in the general thickness of the wall and the 
arrangement of the excretory canal-systems (this specimen is 
circular, flat-vase-shaped, 17 inches in diameter, and about 
an inch in thickness),—but in the minute structure totally 
different, as it is from the structure of all other hexactinellids 
with which I am acquainted, inasmuch as the bundles of 
linear spicules, cutting each other in all directions at different 
angles in the midst of the reticulated structure, are entirely 
absent in D. pumiceus; and although somewhat resembling 
those of Huplectella aspergillum, and similarly united by 
Jadder-like structure, they are comparatively short and ar- 
ranged in a totally different manner, never apparently cross- 
ing each other at right angles. Itis true that, where the 
spiculous structure bridging the oscules is rubbed off, the 
surface presents the lattice-like appearance of Huplectella ; but 
this is in an unnatural state, while the rest of the struc- 
ture of the wall, together with its thickness, resembles that of 
D. pumiceus rather than that of Euplectella. 
By chance I found, under 41-inch compound power, a 
microscopical fragment of the skeleton bearing a rosette, which 
I now possess, mounted in balsam; and this, as above stated, 
is globular, radiated, like that of Dactylocalyx subglobosus and 
also D. pumiceus (‘ Annals,’ 1873, vol. xi. pl. xii. fig. 6). It 
is the only instance in which I have found a rosette fossilized ; 
and but for the unusually perfect state of the decalcified skeleton, 
chance, and the use of a high microscopic power, this, in all 
probability, would never have occurred. Still it shows that 
the rosette may be preserved, although the combination of 
circumstances necessary for its detection may seldom happen. 
The presence of this globular radiated form of rosette, how- 
