58 Mr. W. J. Sollas on two new and 
especially to chamber C than to B, and that, accordingly, we 
must rather consider the fact that its spicular heads point 
towards the interior of C, than that they point away from the 
interior of B. An inspection of fig. 6 will at once furnish us 
with an answer to this argument, so far as it can be called 
argument; for there we find two diaphragms, the relation of 
which to their respective chambers is clear enough, a evidently 
belonging to chamber A, and 6 to B, while, at the same 
time, the position of the surface of spicular heads is reversed 
in each case, in a the points and in 6 the heads of the spicules 
being turned towards the interior of the respective chambers. 
In some cases, moreover, I have seen a diaphragm placed 
obliquely across an aperture where two chambers open into a 
third, and evidently so arranged as to determine a passage into 
one rather than into the other. 
It seems then, to my mind, that whatever the ultimate 
function of these bodies may be, their immediate morphologi- 
cal relation to the sponge is that of open or closed partitions 
between adjoining chambers; and the term “ diaphragm’? is 
therefore the most appropriate to them. 
The composition of the diaphragms may be best determined 
by placing one on a glass slide, adding a few drops of nitric 
acid, and boiling over a spirit-lamp till the acid has nearly all 
evaporated ; a few more drops must then be added, and the 
operation repeated as many times as may be necessary for the 
solution of the kerataceous cement which binds the spicules 
together. When this has been accomplished, the acid must be 
driven off completely by continued heating, and the spicules 
mounted on the same slide as has served for their preparation: 
no attempt must be made to wash them with distilled water, 
or to transfer them to another slide; either of these operations 
is sure to result in the loss of some or all of the small flesh- 
spicules with which the diaphragms are but sparingly sup- 
lied. 
By examining the edge of one of the diaphragms as an 
opaque object, under an objective magnifying about 100 
diameters, the arrangement of its spicules can readily be made 
out ; and no arrangement could be simpler (Pl. I. fig. 14). The 
more or less cylindrical spicules lie side by side, their mucro- 
nate extremities forming the inner and their globular heads 
the outer face of the diaphragm, so that the latter looks like a 
pavement of glass marbles (PI. I. fig. 15), all of the same size, 
and packed as closely as possible, and in consequence exhi- 
biting a quincuncial pattern; the inner face has very much 
the same appearance, with the single difference that from each 
marble of its pavement a small spike stands out erect. Across 
