160 Mr. H. J. Carter on Grantia ciliata, 
most of the shorter linear ones of Grantia spinispiculum, 
although even here there may be an admixture of both kinds. 
At the same time long, stout, terminally curved spicules of a 
similar form, but smooth throughout (that is, without spines 
and not so much bent at the ends), may be more or less plenti- 
ful in the conular “ bunches” of Grantia ciliata and G’. coro- 
nata; wherein the chief differences between these and those of 
the variety G. spinispiculum appear to me to consist. 
The capability of entirely closing the cloaca, which is essen- 
tially the “rectum” or termination of the excretory canal- 
systems in a// hollow sponges, whether calcareous or siliceous, 
by the extension of a sarcodic, sphinctral diaphragm across 
the mouth at the junction of the body with the collar-ring, 
first noticed by Bowerbank in Grantia ciliata (Trans. Micros. 
Soc. 1859, vol. vil. p. 83), although not so plainly seen as 
that provided for the opening of each radial chamber into the 
cloaca, together with the relationship of the ends of the cones, 
which, although closed in the dried state by a kind of spiral twist 
of the linear spicules (“ monoceles,’’ H.) around their apices, 
respectively, would appear to have an opening here in the 
living one, is a mechanism which I cannot understand, seeing 
that the inhalent channels which lead to the chambers are on 
their outer side, as distinctly indicated by the pores themselves 
in the triangular interspaces between the conuli on the surface 
(Pl. VIL. fig. 7). That the single tubular vent projected 
from the young Spongilla when grown from the statoblast 
does close up for a time after a surfeit of carmine, I have long 
since witnessed and described (‘ Annals,’ 1857, vol. xx. p. 30, 
pl. i. fig. 1); and if the sphinctral membrane at the base of 
the collar-ring acts in connexion with the conuli and sphinc- 
tral diaphragms of the radial chambers in Grantia ciliata 
and its varieties, they being all parts of the excretory canal- 
system, then the same kind of general closure may take 
place as in the young Spongilla under similar or other cir- 
cumstances. 
To attempt to describe all the varieties of form in the 
spicules of the sponge would be vain, from their great number, 
hence they can be only learned by a practical examination. 
To assign the use of the spined spicules too would appear to 
be wholly conjectural from their almost general distribution 
over the whole surface of the body, had we not a parallel case 
in the young spherical G'eodia &c., wherein the anchoring- 
spicules (“‘ anchors” and “ forks”) are developed over the 
whole of the surface, but for the most part only retained in a 
projected state where the Geodia is nearest to the object to 
which it may be attached, when they come into use for 
