22 Mr. W. J. Sollas on a new Genus and 
From these we see that spicules which are usually pro¬ 
duced by the immediate radiation of four arms from a mother 
cell may pass into varieties in which the origin of some of 
the rays is postponed till the spicule-cell has assumed a linear 
growth and so made it impossible for fresh rays to appear 
except as secondary instead of primary radiations—a change 
analogous to that which sometimes occurs in the vegetable 
kingdom, where a whorl of leaves may exceptionally become 
elongated into a branch. Conversely a spicule in which 
bifurcation took place after the spicule-cell had produced an 
acerate form might easily pass into one in which the bifurca¬ 
tion took place at the commencement of growth, and thus rise 
to a truly triradiate form. 
If the acerate and triradiate spicules of Plectronella , widely 
different as they now are, did originally branch off from a 
common ancestral form, one is led to inquire whether the 
same may not be the case with the still more diversified spi- 
cular groups belonging to other kinds of sponges. We can¬ 
not attempt to enter into this question now; but it may be 
interesting to examine a few other sponges with a view to 
determining whether they present us with analogous varia¬ 
tions to those exhibited by Plectronella. A few hours’ exa¬ 
mination of two specimens of Dercitus Bucklandi has fur¬ 
nished us with the variations of its quadriradiate spicules 
figured on Plate VII. The commonest variety is that of 
fig. 43, where one of the four rays is bent, near its termination, 
about 30° from its normal course. Sometimes two or even 
three of the rays may be similarly modified, giving us, in 
the last case, a figure very like that of the Manx arms. This 
flexure is almost invariably the forerunner of a bifurcation of 
the ray, owing to the appearance of a second termination to 
the ray, as if to maintain its symmetry (fig. 40). The secon¬ 
dary rays so produced may themselves become flexed and 
eventually bifurcated (fig. 39). When three out of the four 
rays of a spicule are thus bifurcated, we are carried a long 
distance on towards the branched spicules of the more compli¬ 
cated Tetractinellidae. 
On the other hand, the fourth ray is sometimes completely 
suppressed; and the triradiate form (fig. 33) is then produced. 
Again, the third ray may not arise immediately from the 
young spicule-cell, but from one of the arms of a biradiate 
spicule (fig. 36) ; or the third ray may wholly disappear, when 
we have a biradiate form, bringing us back to a curved 
acerate spicule. Unfortunately the only instance I have 
observed of this was lost before I could sketch it; but there is 
no doubt that a more extended examination would reveal 
other examples. 
