212 
THE VOYAGE OF H.M.S. CHALLENGER. 
which vary from about 0‘03 to 0’05 mm. in thickness; at the margins of the chones small 
fusiform cells are accumulated, but do not appear to form a sphinctral ring, nor is any 
special sphincter to be observed at their inner ends, which open freely into the incurrent 
canals of the choanosome. 
On approaching the oscules the microstrongylose layer is thickened to form the 
marginal lip (PI. XXVITI. fig. 15); and the cortex, reduced in thickness and otherwise 
modified, as by the gradual disappearance of the sterrasters, is continued inwards as the 
cloacal wall. In a recess of this, just below the oscular margin, the wall presents the 
structure shown in fig. 18, PI. XXVIII. It consists of a layer of apparently homogeneous 
matrix, 0'07 mm. thick, staining with hsematoxylin, and containing various apparently 
cellular structures ; most exterior are slender threads about 0’02 mm. long, directed at right 
angles to the outer surface, ending at one extremity against it, and at the other enlarging 
into a rounded or fusiform body, from which are continued one or two slender thread-like 
processes deeper into the interior ; in some cases these can be traced into connection with 
somewhat similar but stellate corpuscles which lie in about the middle of the layer ; 
nearer the inner face of this, where it adjoins the sarcenchyma of the choanosome, are 
s im ilar fusiform bodies lying tangentially. I have spoken of the matrix as only 
apparently homogeneous, because in some sections it appears to be divided up into oval 
or rounded polygonal masses about 0*02 mm. in diameter, each enclosing one of the more 
deeply stained bodies alluded to as a corpuscle ; and bounded by pale linear interspaces. 
It would thus appear that the smaller more darkly stained body is not the whole ceU, 
but only a differentiated part of it. I do not venture to offer an opinion on the 
significance of these appearances, which appear, however, worth recording. 
Choanosome . — The mesoderm is a well-marked sarcenchyma, and, except about the 
openings of the chones, seldom becomes collenchymatous ; the incurrent and excurrent 
canals are not provided with collenchymatous walls ; a sarcenchymatous wall of no great 
thickness, the sarcencytes of which are usually elongated or fusiform in shape, is, however, 
sometimes present. Occasionally by a modification of the ectosarc these cells pass into 
fusiform collencytes. As a consequence of the absence of thick walls, the canal system 
presents a strikingly open appearance, and velar diaphragms are rare. Scattered through 
the sarcenchyma are pigment-cells (PI. XXVIII. fig. 22), which become more numerous 
in the neighbourhood of the larger canals. They have the same characters as those of 
the cortex, and are traceable into cells composed of spherical colourless granules which 
take a deep stain with hsematoxylin ; sometimes one meets with a cell half composed 
of these protoplasmic granules and half of the pigment-granules into which they become 
converted. The cells composed of protoplasmic granules are probably derived from deeply 
staining amoeboid cells, possessing a large nucleus and nucleolus, that are found here 
and there within cavities of the sarcenchyma (PI. XXVIII. fig. 21). In addition to 
pigment-cells numerous oval darkly stained bodies of a problematical character are 
