MATERIALS FOR A MONOGRAPH OF THE ASCONS. 311 
formed in the interior of the sponge wall, projecting into the 
eastral cavity (figs. 46, 57, 68), in which case they arise from 
the dermal epithelium lining the gastral wall of the oscular 
rim in exactly the same way as they are formed on the outer 
surface of the sponge. The fact that the inturned dermal 
epithelium, which is for the most part destined to give rise 
to porocytes, can also secrete monaxons, is an important 
point in considering the origin of the gastral rays of the 
quadriradiates. 
The slender monaxons of L. complicata are remarkable 
in that the lance-head is either rudimentary (fig. 7) or absent 
(fig. 8). The stages in the formation of these spicules are 
shown in surface views in figs. 1—8, and in sections in 
figs. 42—47 and 57. 
The most noteworthy points in this development are, first, 
the great distance between the two formative cells before 
the spicule appears; and secondly, the clear space, which 
has been described above, which precedes the formation of 
the spicule (figs. 1—5). In places where the body-wall is 
thickened, for instance, at or near the point of attachment of 
the sponge, the spicules may be formed very far in from the 
surface (fig. 45). 
(b) The stout monaxons are fairly abundant, but their 
earliest stages are, for some reason, difficult to find; perhaps 
because they grow with great rapidity. Asa rule they have 
a very distinct lance-head, and figs. 10—15 show, without 
need of further explanation, how the lance-head is formed 
by the thickener and then gradually abandoned by it. Occa- 
sionally, however, the lance-head is completely absent (fig. 
16). The great size of these spicules makes it very difficult, 
in fact, practically impossible, to obtain perfect stages of 
their growth in sections. Some are shown in figs. 48, 71 a, 
(0: 
When I commenced my investigations on these spicules I 
made use of the method, which I now condemn, of brushing 
off the gastral collared epithelium. In such preparations I 
found many young spicules, similar to figs. 10 and 11, but 
