BY H. LEIGHTON KESTEVEN. 139 



shape, and aoain become solid; during this time, none of them 

 were surrounded by a clear area of plasm staining purple. 



The old nuclear membrane is apparently still present, in some 

 cases, until a much later period; but, in others, it seems to have 

 broken down at the time when the new nucleus first develops its 

 membrane. 



Up to this time, the ultimate oogonium has lain free in the 

 lumen of the gland or oviduct; it now becomes attached to the 

 wall of the latter, either to one cell by a foot, as in Fig. 9, or, in 

 the absence of the foot, to several cells. 



Growth in size appears now to take place verj^ rapidly, and 

 there is a great increase in the quantity of yolk-spherules. No 

 changes are observable in the new nucleus. 



Concurrently with the increase in size of the cell and quantity 

 of yolk, the karyosomes left free in the old nuclear area appear 

 to be dissolved without showing any further signs of activity. 

 When the cell, now to be regarded as a primary oocyte, has 

 reached its full size, it is once more set free into the lumen of 

 the oviduct. 



There is present in the mature primary oocyte only one set of 

 spherules, variable in size, certainly, but differing in no othei' 

 way from one another. All are yolk-spherules or none are yolk- 

 spherules. The spherules which were formed by increase in size 

 of the chromatin-granules are, therefore, similar to, and may be 

 regarded as being the same as, those formed later. 



After these observations had been made, and the conclusions 

 given below had been deduced from them, I sought, in the pub- 

 lications of previous workers, for comparable observations and 

 deductions. I cannot better give the results of my examination 

 of literature than by the following quotation from a paper by 

 Caroline McGill(5), who, on p. 21 9 of the paper quoted, expresses 

 the opinion that "it seems probable that chromatin may have 

 something to do with yolk-formation." 



Will (7) thinks that the larger nucleoli of the amphibian 

 germinal vesicle pass out into the cytoplasm, and there become 

 yolk-nuclei. MacCallum(4) concludes that, in the ova of Am- 



