248 CEYLON PEARL OYSTER REPORT. 



clear cell contents and rounded nuclei, and these cells are very distinctly ciliated. 

 Fig. 5 represents a section taken through the duct of the byssus cavity, shortly 

 above the point where the lips of the furrow have united to enclose a canal. It can 

 easily be seen that the duct consists of two portions, (a) a lower portion whose 

 epithelium is continuous with that of the open furrow and, like it, is non-ciliated and 

 provided with a cuticle ; the walls of this region are thrown into a number of folds ; 

 (b) an upper portion continuous with the crescentic demi-canal, and lined by the 

 same clear ciliated cubical or columnar cells. Fig. 6 represents a section through 

 the middle of the byssus cavity. At the upper (really the anterior) end of the 

 cavity is the crescentic demi-canal lined by the same clear ciliated cells as before. 

 The remainder of the cavity is broken up by septa, of which two thick folds on 

 either side of the demi-canal, a central partition springing from the lower (posterior) 

 end of the cavity, and two minor lateral folds may be particularly noticed. These 

 septa are covered by a ciliated epithelium, evidently of the same nature as that 

 lining the demi-canal, but the cells are very much elongated and enlarged at their 

 outer extremities. Those on the thick lateral folds are especially long, and diverge in 

 a fan-shaped manner from the band of connective tissue and muscle fibre which forms 

 the centre of the fold. This figure agrees in most respects with CARRIERES' drawing 

 of the byssus cavity of Cyprina islandica (5, fig. 12, B). In a section taken through 

 the deeper end of the byssus cavity, the characters of the epithelium are the same as 

 those of the preceding section, but the cavity has been divided into two by the forward 

 extension of the median septum, the crescentic demi-canal has disappeared, and the 

 lateral septa are smaller. These two anterior prolongations of the byssus cavity end 

 blindly close beneath the pedal ganglia. These three figures are drawn from 

 horizontal sections of the whole animal, and are therefore nearly transverse sections 

 of the foot. Fig. 7 is a highly magnified drawing (ZEiss' ^ immersion) of a 

 transverse section of the whole animal, which therefore cuts the foot and byssus 

 cavity obliquely. It corresponds to the top part of a section rather anterior to that 

 shown in fig. 6. The section was stained with safranin and licht-grun, and does not 

 show the cell contours very clearly, but the granules of byssogen, stained bright 

 scarlet, are very clearly seen. At by.gl. are seen the large polygonal glandular cells 

 occupying the central part of the foot. On either side these cells may be seen to 

 be breaking up and their granules are streaming outwards along definite lines to pass 

 either into the central tongue-shaped projection (which is a part of the here incom- 

 plete median septum) or into the lateral swellings projecting into the byssus cavity. 

 As they pass outwards, the granules form little pyriform or club-shaped masses, whose 

 swollen ends are directed towards the lumen of the byssus cavity, and it is evident 

 that they are travelling, probably by intercellular paths, to be discharged into the 

 lumen, and there converted into the material of the byssus. The granules themselves 

 are clearly not byssus substance, but " byssogen," as the lumen of the cavity is filled 

 with a granular material (not shown in the figure) which is not stained either by 



