208 Observations on the Gestation of Indian Rays. [Nov. 26, 



we infer that, in early stages, the pavement layer forms a continuous 

 investment over the whole trophonema ; and that subsequently, by 

 the separation from one another at definite spots of the pavement 

 cells, stomata are formed, which, when the lumina of the glands are 

 established, become the mouths of the glands. Be this as it may, the 

 outer (pavement) layer is not traceable into the mouths of the glands 

 at any point in any of our sections. 



The trophonemata possess an exceedingly rich vascular supply. 

 All the sections show two large, but not very thick- walled, arteries, 

 one in each non-glandular margin ; minute arterioles here and there 

 between adjacent glands immediately beneath the epithelium ; and, 

 between the layers of glands of opposite faces, a plexus of sinuous, 

 cavities or capillaries. -i g"\ 



The capillary plexus in transverse sections of a trophonema is seen 

 to extend deeply between the glands right up to the superficial 

 arterioles ; and, in the part corresponding to the position of the great 

 axial vein of Pteroplatcea, it presents a solid or spongy circular 

 expansion, in the centre of which the future lumen of the vessel is 

 commencing to be formed. In transverse section of a group of glands 

 the capillary plexus is seen to form a polygonal network, in the, 

 meshes of which the glands lie. 



In comparison with Pteroplatcea micrura, the trophonemata of 

 Trygon walga are characterised by the possession of an epithelium 

 which is several cells thick instead of one cell thick, and by the 

 richer vascular supply of their glands, each of which is embedded 

 in a little capillary cup of its own, like a filbert in its husk. 



5. Conclusions. 



1. Comparison of the trophonemata in the two stages above 

 described, showing, on the one hand, at the onset of pregnancy, a 

 mucous membrane of large nucleated indifferent cells and of solid 

 unformed glands, and, on the other hand, when the term of pregnancy 

 is fulfilled, a surface layer of gland-casts of epithelium in various 

 stages of degeneration, appears to be conclusive proof that the glands 

 are developed for the special requirements of the pregnant state. 



2. As regards the function of these glands, the presence in every 

 case where a fresh pregnant uterus has been examined of a viscid 

 turbid or actually milky albuminous fluid, and further the finding in 

 the case of Myliobafis nieuhofii of one and the same secretion in the 

 uterus of the mother and in the intestine of the foetus, seem to fully 

 confirm our original conclusion that they are in all cases milk-glands 

 furnishing a secretion for the nourishment of the embryo. 



3. Regarding the channel through which the milk is carried into 

 the foetus, we think it to be in every species that we have examined, 



