1886. J STRUCTURE OF THE OVUM IN THE DIPNOI. 511 



the opinion that it may be derived from the germinal epithelium. 

 Owsiannikow suggests three possibilities ^ — either it originates from 

 cells which have made their way out of the blood-vessels (!), or from 

 cells of the subgerminal tissue (stroma?), or, finally, they may be 

 derived from the germinal layer. The latter alternative is adopted 

 by Owsiannikow on certain evidence, which he does not, however, 

 regard as conclusive. The outermost follicular layer of Lepidosiren 

 I have already (p. 509) shown without doubt to be derived from 

 the germinal epithelium ; I shall therefore adopt the name of 

 secondary follicular epithelium for this cipsnle, wliich indicates that 

 its origin is similar to that of the true follicular epithelium, which 

 may be briefly termed the follicular epithelium. 



The two layers that have just been described form a hollow sphere 

 enclosing a central cavity, which is partly occupied by a mass of 

 cells. It is very possible that in the fresh condition the central 

 mass of cells occu|)ies the whole of the space available, but this 

 is not the case in my preparation. A large portion of the central 

 cavity, particularly on the side turned towards the exterior of 

 the ovary, is quite empty, and no structures intervene between 

 the central mass of cells and the follicular layer. On the opposite 

 side, however, the central mass is in close contact for a con- 

 siderable area with the follicular cells, this area exactly corre- 

 sponding with the transitional area between the follicular and external 

 layers. These facts would suggest that the central cells are derived 

 from the proliferation of the follicular cells and ultimately of the 

 extra-follicular cells, as these two latter have been shown to be 

 perfectly continuous, the proliferation taking place in a certain 

 limited area only. In this case the apparent cavity which separates 

 the central cells from the follicular on one side will be an indication 

 (exaggerated by the action of the preservative reagent) that there is 

 here no real connection between the central and peripheral layers, 

 though they may be in actual contact in the fresh state. 



A number of the central cells are displayed in figs. 14-20 of Plate 

 LIII. ; they are more or less inegular in shape, rounded, and of 

 different sizes ; the staining-reagent has hardly affected the cell- 

 protoplasm, but has deeply stained the nucleus. The cell-protoplasm 

 is arranged in a reticulate fashion, and closely resembles that of the 

 follicular cells. Some of the cells contain two or more nuclei, which 

 seems to show that the cells themselves are in a condition of multi- 

 plication. The most remarkable fact about the nuclei of the central 

 plug of cells is their great inequality in size : some of the variations 

 are exhibited in those figures ; the variation is all the more remarkable 

 as it does not occur in the follicular layer, the nuclei of whose cells 

 are of quite a uniform size. There is almost every gradation in size 

 between the smallest and largest nuclei, a fact whicii perhaps indicates 

 that the smaller ones are the result of nuclear division. The largest 

 nuclei rather excel in size those of the follicular epithelium. There 

 is a similar difference of size in the peripheral layer of cells, 

 particularly obvious at those points where the peripheral layer is iu 

 1 Loc. cit. p. 30. 



