R. W. H. ROW—REPORT ON THE SPONGES : NON-CALCARBA. ab 
remarkable genus, the two previous species being P. merguiensis, Carter (7), 
and P. cineriformis, Dendy (11). All three species come from the Indian 
Ocean or adjacent waters: P. cineriformis was obtained in Ceylon and 
P. merguiensis at various localities from Mergui Archipelago to Torres 
' Strait. 
These three species form a complete series illustrating the development of 
the cortical trizenes. In P. merguiens’s the spicule can be clearly seen to 
be developed from an ordinary plagiotrizne, since the rhabdome can always 
be distinguished from the ecladi in them. Further, young forms can be found 
in which the plagiotrizne form is not obscured. A considerable advance is 
seen in P. cineriformis, where the true triene form is entirely lost and the 
spicule is somewhat irregular. The trieenes are, in fact, very similar to those 
of P. eccentrica which possess all four rays. A further advance is seen in 
P. eccentrica, where, as described above, some of the trienes have become 
modified into ‘“ walking-sticks.’ Otherwise these two species are very 
similar, both in shape, colour, and skeleton. 
The genus is especially interesting as indicating a possible origin for the 
Lithistida. The appearance of two main types of trizenes in P. eccentrica 
also suggests that the Tetracrepid Lithistida and the Monocrepid Lithistida 
may have arisen from the same group, by selective specialism, from some 
such form as this, where spicules bearing a resemblance to desmas of both 
kinds occur. 
The evidence which can be brought forward in favour of a Tetillid ancestry 
for the Lithistida is, however, far too vague to enable one to do more than 
point out the possibilities suggested by the various species. 
Locality. The single specimen was obtained at Tella Tella Kebira. 
Distribution. Red Sea. 
CHROTELLA IBIS, n. sp. (PI. 85. fig. 2, Pl. 36. fig. 7; Text-fig. 8.) 
This species is represented in the collection by several specimens, all very 
similar in appearance. Lach consists of a subspherical mass, in the largest 
specimen 23 mm. in diameter. The basal portion of the sponge is wanting 
in all the specimens, probably being left on the support when the sponge was 
collected. The broken surface forms a large, more or less flat area, the 
diameter of which is very nearly as great as that of the sponge at its thickest 
point, so that if this area really represents the approximate area of attach- 
ment, the sponge will have the shape of a somewhat globose cushion. 
The surface is densely but minutely hispid, but the greater part of the 
projecting spicules are broken off close to the surface of the sponge. Imme- 
diately covering the surface is a single compact layer of sand-grains, 
entangled among the projecting spicules. 
There are several small oscula visible on each of the specimens, measuring 
