832 DR. G. J. HINDE ON RECEPTACULITIDZ. 
whereas in hexactinellid sponges of the genera Hyalostelia, Zitt., 
and Holasterella, Cart., the spicules reach a maximum length of 9 
millim., and a thickness of ‘5 to *9 millim. But, on the other hand, 
in the genus Jschadites, the spicules are not at all larger, and many 
of them are in fact much smaller than the surface-spicules of Palzo- 
zoic and Jurassic hexactinellid sponges. 
The only other structural feature which remains for comparison is 
the inner or upper surface-layer which, as we have seen, is only de- 
veloped in the genus feceptaculites. As already mentioned, the 
structure of this plate in the typical example of the genus, R. Neptuni, 
is not definitely known, but in the allied species, R. occidentalis, it 
consists of a continuous layer or plate, perforated vertically by canals 
giving access to the interior of the specimen, and also with short 
horizontal canals crossing each other at right angles, penetrating its 
substance, and communicating at their points of intersection with 
the canals in the vertical rays of the spicules. In its complete form 
this inner plate appears to me to be analogous to that which forms 
the inner or upper wal! of the Jurassic hexactinellid genus Porocy- 
pellia, Pomel*. 
The circulation of water through the organism in the Receptacu- 
litidee appears to have followed a similar course to that occurring in 
sponges generally. It seems to have found admission through the 
narrow linear apertures beween the margins of the spicular plates of 
the outer surface, and after passing through the interspaces between 
the spicules, which in the living state were lined by the sarcode, to 
have found its exit through the aperture in the summit of IJscha- 
cites, or through the perforated inner layer in Receptaculites, or freely 
into the open cups of Acanthochonia and Spherospongia. 
The resemblances referred to above appear to me to justify the 
conclusion that the Receptaculitide constitute a family of siliceous 
Hexactinellid Sponges. The body-walls are composed of spicules of 
the hexactinellid type, but modified by the development of regular 
rhomboidal or hexagonal plates in place of the head-ray of the nor- 
mal spicule. ‘The spicules are arranged in definite order, so that 
the summit-plates form oblique, curved, or spiral rows, whilst the 
four horizontal rays mark radial and concentric lines. In one 
genus the body-wall is bounded by a perforated inner layer. 
The character and position of the spicules ally the family to the 
Paleozoic genera Protospongia, Dictyophyton, and Plectoderma. As, 
however, the component spicules of the skeleton do not appear to 
have been connected together, otherwise than by the sarcode (save 
in the genus Receptaculites), the family would be ranked in Zittel’s 
suborder Lyssakina. 
Ifin some structural features the Receptaculitide stand apart 
from fossil sponges of a more recent geological horizon, other un- 
doubted Paleozoic sponges also exhibit abnormal characters. Thus it 
is only in the genera Protospongia, Salt., and Dictyophyton, Hall, that 
there exists a surface-structure of definite larger and smaller quad- 
rate areas; in Astylospongia, F. Rom., the canal and spicular struc- 
* See Zittel, Neues Jahrb. 1877, p. 364, t. 5. f. 1a. 
