98 BULLETIN OF THE 



agine would have caused the author more concern than it seems to have 

 done. If his drawing accurately reproduces the conditions, there must 

 have been about four times as many villi as there were granulosa cells. 

 That in itself alone might not be of any significance to the author, espe- 

 cially as he disclaims any genetic connection between the granulosa and 

 the underlying capsular membrane, but it does seem as though it should 

 have received some explanation in view of the ultimate relation (Taf. I. 

 Fig. 10) which Hoffmann admits to exist between the radial fibres of the 

 capsular membrane and the cells of the granulosa. This is what he says 

 about the later (February) stage of the egg : The zona itself is seen 

 to consist of two layers, the inner much thicker than the outer. From 

 the latter there arise with small triangular bases long peculiar fibres 

 only 1 jx in thickness, which are stained in osmic acid precisely like the 

 outer layer of the zona from which they arise. The outer ends of these 

 fibres are thickened, even more than their inner ends, and form a con- 

 tinuous laver, between which and the zona radiata the fibres themselves 

 are stretched like so many columns. " Over the proximal 1 ends of these 

 fibres the granulosa cells are arranged in srich a way that a granulosa cell 

 fits into each thickened end (Taf. I. Fig. 10)." 



If after the [ovarian] eggs have lain in water a short time they are 

 transferred to osmic acid, it becomes very easy, says Hoffmann, to iso- 

 late both this layer (formed by the expanded ends of the fibres) and the 

 granulosa in the form of large shreds (" Lappen "). From such prepa- 

 rations one can readily convince himself, he says, that the expanded 

 ends of the fibres form a continuous sheet, and are not processes of the 

 granulosa cells, as one is at first inclined to assume, and that the fibres 

 are, as the examination of the early stages proves, only the greatly gr own- 

 out conical villi. 



That which seems to me to need explanation is, Why is the numeri- 

 cal relation between the villi and the granulosa cells so different at dif- 

 ferent stages in the growth of the egg, and why does this relation become 

 such an invariable one in the later stages of development 1 



It might be answered, in reply to the first question, that the granu- 

 losa cells undergo rapid multiplication, and that cell division occur- 

 ring twice for all granulosa cells between October and February would 

 explain the altered relations ; but is it not more reasonable to suppose 

 that, through some unexplainable accident, Hoffmann has been led to 

 attribute eggs to a perch which were taken in October from some other 



1 It is not clear to me in what sense tlie author can use the word "proximal " of 

 the ends of the fibres which are directed aica/j from the centre of the egg. 



