STUDIES IN THE EETINA. 35 



We have so far confined our attention solely to the adult 

 frog, and the argument thus limited has been somewhat 

 complicated. This, however, is the order in which the 

 work was carried out and the results obtained. It may, 

 perhaps, be remembered that the relations of the rods and 

 cones was not the initial object of these researches, which 

 were undertaken to extend and elaborate the evidence I had 

 collected in confirmation of the theory of vision I published 

 in 1896.^ We shall now turn to other Amphibia, and espe- 

 cially to young forms in different stages of development, 

 where perhaps we ought to have begun, and show that they 

 fully confirm the conclusions above arrived at, and generally 

 in a much simpler and more decisive manner. 



The Toad. — I have not found any traces of sections or 

 nodes in the tips of the cones in the toad. Their continuations 

 have more the appearance of shrunken sacs (fig. 6). The 

 refractive globule so characteristic of the cones in the frog is 

 absent here, but is perhaps compensated for by the immense 

 size of the " ellipsoid," which in the fully developed rod may 

 attain a length of 10 ju with a breadth of 6 fx. It is very con- 

 spicuous in cone Cg (see fig. 7), but becomes somewhat in- 

 distinct in cones Cg and c^. In cones at this stage it some- 

 times appears as if there were two, somewhat ill-defined. 

 These finer details, and indeed the ellipsoid itself, require 

 special investigation, and we shall return to them again in the 

 second part of this paper. 



As in the case of the frog I suggest that cone c^ (fig. 7) 

 passes into the rod r^ by the filling up of the terminal sac, 

 and the squeezing outwards of the ellipsoid and other matter 

 in the basal part of cone c^. Again, also, the aspect of the 

 double cones, fig. 6, d, h, shows that the action of pressure 

 might well account for the long thin inner limb of rod 7*^ ; 

 while, lastly, fig. 6, i, shows how great the lateral pressure 

 within the proximal zone of the layer, brought about by the 

 intrusion of young cones, must be. Here we see the inner 



» '4nn. and Mag. Nat. Hist.,' 6, xvii, p. 162. 



