84. BRITISH FOSSIL SPONGES. 
quadrate framework, and the interspaces are filled by numerous irregularly 
arranged cruciform or five-rayed spicules, the whole forming a smooth, connected, 
surface membrane. In the enveloping dermal layer of Cystispongia, Roem., and 
Camerospongia (D’Orb.), the siliceous membrane is very fine and delicate, and no 
traces of spicules or of their axial canals can be detected in it. 
Octactinellide.—In this small group, only represented by the genus Astreo- 
spongia, the skeletal spicules do not appear to have been attached in any way to 
each other, but they are distributed without apparent arrangement beyond that 
the stellate dise is generally parallel to the upper surface of the Sponge. The 
spicular rays rest upon and cross over each other, but without leaving any definite 
interspaces or canals. There are no indications of a special disposition of the 
spicules on the outer surface of the Sponge to form a dermal layer. 
Heteractinellide.—The skeletal-spicules in this group appear to have been dis- 
tributed quite irregularly in the body of the Sponge, their rays interlacing with 
each other loosely, and the smaller spicules filling in the spaces between the rays of 
the larger forms. In some instances, fragments of the skeleton are met with in 
which the spicular rays are lightly cemented together ; but this probably results 
from a secondary deposit of silica during fossilization, like that which occurs in 
the tetractinellid Pachastrella from the Chalk of Flamborough. 
In the dermal layer of some of the Sponges of this group the spicules are 
disposed so that their disc-rays interlace and fit into notches in adjoining rays 
(Pl. VII, fig. 1 6), whilst the radial ray projects into the interior of the sponge- 
wall (Pl. VII, figs. 1, 2). The rays also are, in some instances, partially fused 
together, in others they are compressed and completely united with those of 
adjoining spicules, so that there are only small oval apertures in the dermal layer 
(Pl. VII, fig. 3). 
Calcisponges.—In the large majority of fossil caleisponges the skeletal-spicules 
are arranged side by side in close contact with each other, so as to form cylin- 
drical fibres, which anastomose together. The spicules do not appear to be organi- 
cally attached or fused together in any way, and it is somewhat surprising that the 
fibres thus composed should have in so many instances resisted disintegration. In 
some cases there is a relatively large axial spicule in the centre of the fibre, which 
is enveloped by smaller filiform spicules; whilst in others the spicules are through- 
out equal or subequal. In one genus (Protosycon) the spicules appear to have 
the same arrangement as in the existing Sycones, that is to say, they form a series 
of horizontal tubes. 
The four-rayed spicules of the dermal layer in some instances are irregularly 
disposed over the outer surface of the Sponge, and the apical ray is directed into 
the sponge-wall ; in other cases the spicules are irregularly and closely mingled 
together, so as to produce an apparently compact, smooth, or wrinkled membrane. 
