REPORT ON THE TETRACTINELLIDA. 
159 
cortex approach and unite, forming a single thin layer, which is continued forwards along 
with the investing epithelium on to the cloacal tube, furnishing the tissue of the skeletal 
wall. From the lower fibrous layer of the cortex a thin layer of tissue is continued 
inwards over the base of the tube and another over the adjacent choanosome of the 
sponge-body, thus furnishing the walls of the lacunar spaces. These layers of tissue are 
more coUenchymatous in character than that from which they are derived. The interior of 
the base of the cloacal tube is occupied by a mass of coUenchyma channelled by numerous 
excurrent canals. 
Development of the Skeleton . — It has already been stated that the cladomes of the 
prthodieenes become more and more closely crowded together as they are traced towards 
the base of the cloacal tube. This approximation is continued as the tube enters the 
sponge, and is accompanied by a decrease in size, till only very small and early forms 
are reached. These small spicules lie within the coUenchyma of the base of the tube 
arranged in vertical radial planes, one behind the other, the youngest situated most 
externally, and successively older forms occurring towards the interior. This is well 
shown in transverse sections made at the proper level through the sponge (PI. XVII. 
fig. 14). The spicules cross the wall of the tube from within outwards obliquely, the 
cladomes cropping out at its surface one above another, the lowest lying being the 
youngest. Looked at from the outside the cladomes and cladal ends of the rhabdomes 
form a structure over the base of the tube like finely woven basket-work (PI. XVII. 
fig. 11). Evidently these spicular series are modified radial fibres, adapted to the sides 
of the cloacal tubes. 
The chief interest in these spicules is centred in their mode of development. Weltner 
has enquired whether they are to be regarded as primitive forms or reduced trieenes, 
and the foUowing observations furnish a conclusive answer to that question. The earliest 
form of spicule, as shown both by its characters and position, is a minute oxea, 0’07l mm. 
long by less than 0'002 mm. in width ; the distal end appears to be less sharply pointed 
than the other. In the next stage observed we meet with an oxytylote, already 
developing minute cladi ; the rhabdome, finely pointed at the oxeate end, measures 0T42 
by 0‘002 mm.; at the tylote end it is thickened into a small bulb 0‘004 mm. long by 
O'OOS mm. in width, from which at the border of the summit minute conical spines 
extend outwards and forwards. The number of spines could not be ascertained, owing 
to the position of the spicule. In the next stage, one more frequently observed, the 
rhabdome has attained a length of 0‘275 mm. by 0'0025 mm. in diameter. It tapers to 
a finely pointed oxeate end, beyond which the axial fibre protrudes, bare of its siliceous 
investment, apparently into the surrounding coUenchyma, but in reality more probably 
into the scleroblast. The tylote thickening is about 0'004 mm. in diameter, and three 
small cladi, two of which are about 0'0118 mm. in length, project from it. A spicule 
in this stage is represented in the figure (PI. XVII. fig. 12). A stage later and the tylus 
