MATERIALS FOR A MONOGRAPH OF THE ASCONS. 309 
to migrate along the shaft in a proximal direction, they are, 
in reality, practically stationary, while the spicule moves 
outwards. Sections show that the thickener retains its 
connection with the dermal epithelium, and forms a con- 
stituent cell of this layer (figs. 42, 48, 45—48, 57). Very 
noteworthy is the frequent presence in the founder-cell of a 
distinct, clear space in the protoplasmic body, continuing the 
shaft of the spicule, and appearing like a mould in which the 
secretion of the growing shaft is laid down (figs. 5—7, 13, 
15, 79—-83, 85, 86). When the preparation has been very 
slightly corroded by the mounting medium or some other 
cause, the proximal end of the shaft often appears hollow, 
containing a cavity or canal continuous with the clear space 
in the protoplasm of the founder-cell (figs. 83—86). This 
appearance, perhaps, indicates that the first secretion pro- 
duced is of organic material which becomes subsequently 
impregnated with calcite to form the spicules, an interpreta- 
tion further borne out by the observation that in some cases, 
in preparations in which there was not the slightest evidence 
of corrosion, the very young spicule appears first as a clear 
space within the protoplasm of the two formative cells, and 
that the calcite, when it makes its appearance, does not at 
once completely fill this space (figs. 3—5). In this connec- 
tion I may draw attention to the very interesting results of 
Maas (1904 [1], [2], 1906), who has found that sponges 
reared in water deprived of CaCO, produce organic sclerites, 
not containing calcite. 
When the spicule is nearly fully formed the founder ceases 
its activity. As the thickener, however, continues to secrete 
- and to pass towards the proximal end of the shaft, the two 
formative cells are brought close together (figs. 7, 8, 17, 18, 
44, 57, 71a, 82, 83, 85—88). When. the spicule is fully 
formed, both cells leave it altogether. According to’ Wood- 
land (1905, p. 242) the founder leaves it first (compare my 
fig. 8) and the thickener later; a fact which proves clearly 
the complete distinctness of the two formative cells in the 
later, if not in the earlier, stages of spicule-formation. ‘The 
