STUDIES IN THE EETINA. 43 



We thus have not only new elements (cones) forcing their 

 way down between the rods, but the cone nuclei forcing their 

 way to and through the memb. limitans ext., to become rod 

 nuclei. The question then arises, how is the supply kept up ? 

 For in growing retinas it is obvious that a great multiplica- 

 >v tion of the Hliclei of the outer nuclear layer must be provided 

 for. I have never seen any karyokinetic divisions in any 

 retina in parts Avhere rods and cones have begun to form. 

 They can be seen all over the retina just within the memb. 

 limitans ext. in embryos before any sacs have been protruded ; 

 but after the centre of the eye has become functional, and 

 rods have appeared, the karyokinetic divisions are confined 

 to the rim of the retina where the embryonic simplicity 

 persists. The immense number of fresh nuclei required by 

 the outer nuclear layer is not, therefore, supplied by the 

 division of those already present, most, if not all of which 

 would be actively functioning, and thus quite incapable of 

 mitotic division. The answer to the question of their origin 

 is, however, easily solved ; a survey of retinas of any extent 

 reveals a steady migration of nuclei outwards from layer to 

 layer, but especially from the middle nuclear to the outer 

 nuclear layers. As I propose to give the evidence for this 

 in exteuso in a separate paper, I will confine myself here 

 to saying that the nuclei which have been already frequently 

 seen in the outer reticular layer (cf. fig. 5, e and/), and have 

 been invariably regarded as belonging to that layer, are, in 

 reality, nuclei travelling through it from the middle nuclear 

 layer outward. I have been entirely unable to confirm the 

 existence of Krause's " membrana fenestrata," nor the later 

 view of Schieferdecker^ that this outer reticular layer is at 

 least primarily sustentacular, with a specialised nucleated 

 tissue. I have studied these nuclei with great care and for 

 many months, and while it is not always possible to say that 

 they are migrating, this is so frequently the case that I am 

 disposed to assume it to be so universally. The evidence for 

 this conclusion is so overwhelming that it has frequently 

 ' 'Archiv f. mikro. Anat.,' xxviii, 1886, p. 305. 



