40 H. M. BERNARD. 



A still more beautiful confirmation of our conclusions is 

 afforded by the fact that, in young eyes, the oldest and 

 thickest rods and those with longest outer limbs occur in 

 the centre of the retina, and as we work away towards the 

 periphery the rods get thinner and thinner, until it requires 

 close attention to decide whether the element is in the cone 

 or in the rod stage. It seems to me possible to speak of 

 such stages, the rod stage being when the outer limb has 

 begun to grow towards the retina. In fig. 12, A and b are 

 from the centre of the retina, c nearer the periphery, but not 

 so far from the centre that the length of the elements has 

 perceptibly decreased. The rods in A and b were about 

 6 fi thick, in c the thickest was only 3 fx. Here the 

 earlier investigators could have found the transition stages 

 they were looking for in immense numbers had they looked 

 for them. 



It seems to me, then, that we can clearly regard the longer 

 and thicker rods found in the centres of young retinas as 

 the '' oldest,'^ for we know that all these elements are grow- 

 ing. The rod layer is almost crescentic in shape in very 

 young eyes, and at its thickest it is much thinner than in the 

 adult. In the advanced toad tadpole (with four legs, and 

 with the diameter of eye 1'12 mm., some of Avhose retinal 

 elements are shown in fig. 12) the thickness of the rod layer 

 in the centre of the retina was only 34 — 36 jjl. In the 

 adult it is at least 50 ix. Again, the thickest rod I could 

 find in this young eye was 6 jx, thinning away to 3 /i, while 

 I have measured rods in adult toads as much as 9 /z thick. 



I have noted also that rods may be thicker in the centre 

 of the eyes of adult frogs than towards the periphery. 

 Eods 8 n thick were common in the middle of a retina, 

 while 5 — 6 /x was the usual thickness at the sides. If the 

 thickest were the oldest, this would be intelligible. 



The transitional series between cones and rods based upon 

 the longitudinal measurements yielded by these young eyes 

 are very simple. As can be seen from fig. 15, which, as might 

 be gathered from the presence of the refractive globule. 



