4 PERCY SLADEN TRUST EXPEDITION 
The canal system, except for the obliteration of the vents, due probably to temporary 
contraction, agrees closely with that of cxx. 7, and here again there appear to be remnants 
of an endogastric network. 
The spiculation (Plate 3, fig. 2) differs from that of cxx. 7 only in the absence of 
any distinction between deep and dermal triradiates, all bemg of about the same size as 
the deep triradiates in the specimen above described. 
The histology exhibits great peculiarities, which I am inclined to associate with 
the contracted condition of the sponge and with the presence of the numerous large 
ova. The mesoglea is much thickened, especially in the neighbourhood of the ova, 
which lie separately in spherical cavities not lined by any special epithelium™. Frequently 
the presence of an ovum (Plate 3, fig. 3, ov.) causes a conspicuous inward bulging of the 
wall of the ascon-tube in which it lies. Most remarkable is the presence in the mesogloea 
of enormous numbers of small spherical bodies of a yellowish colour and varying from 
about 0°004 to 0°006 mm. in diameter. These bodies are abundant throughout the 
mesogloea but especially so immediately beneath the gastral surfaces of the tubes and 
around the large ova (Plate 3, fig. 8, yb). They lie in the meshes of a network formed 
of delicate strands of mesogloea with here and there small nucleated amcebocytes or 
stellate connective tissue cells (collencytes). They occasionally appear to adhere to the 
surface of an ovum or even to become embedded in the peripheral cytoplasm of the latter. 
Sometimes very similar bodies are actually found imbedded in the nucleus of the ovum, 
but I do not think they can really be the same. These yellow bodies are so remarkable 
that I append a detailed account and discussion of the subject to the description of this 
species. 
There are apparently no collared cells at all, the inner surfaces of the ascon-tubes 
(easily recognizable both by their position and by the projecting apical rays of the 
quadriradiates) being lined, at any rate to a large extent, by a flattened pavement 
epithelium, similar to and quite as well developed as that which lines the outer surfaces. 
This absence of collared cells from the gastral surfaces is highly remarkable, but is to 
some extent in harmony with the observations of Minchin [1900] on contracted specimens 
of Clathrina, and with those of Weltner [1907] on the disappearance of the collared 
cells in Spongilla (Hphydatia) during the winter months. Whether or not, in the 
present case, the collared cells are represented by the yellow bodies in the mesogloea will 
be discussed presently. 
The ova are very large, attaining a diameter of about 0°086 mm., although they have 
evidently shrunk to some extent, so that they no longer fill the cavities in which they lie. 
They nearly all seem to be in about the same stage of growth (Plate 3, fig. 3). The 
cytoplasm is coarsely granular, or flocculent-looking ; there is no vitelline membrane and 
no sharp outline to the cytoplasm. The nucleus is large and centrally placed, it usually 
has an irregular outline and no nuclear membrane. It is differentiated from the cyto- 
plasm by its denser and much less coarsely granular appearance and frequently contains 
‘a small number (? from 1 to 3) of conspicuous spherical bodies (nucleoli ?), closely 
* Such an epithelium when present is probably derived from amceboid wandering cells becoming flattened. 
A few such may sometimes be detected around the ovum. 
