1886.] STRUCTURE OF THE OVUM IN THE DIPNOI. 523 



oue of the cells, which compose the central mass of the follicle, 

 acquiring a predominance in size over its neighbours or being differ- 

 entiated in any other way. 



The fact of there being two kinds of ova with a different mode of 

 development is not new to the Vertebrata. In my former paper I 

 have compared the follicle and its contents in Protopterus to the 

 " egg-nest " of Elasmobranchs, the points of difference being perhaps 

 on the whole greater than the points of resemblance. But, since 

 the formation of " egg-nests " is so general among the Vertebrata, 

 it seems to me that there is probably some genetic connection 

 between these structures and the "egg-nest" of Protopterus. It 

 has been shown that in Mammals, Elasmobranchs, and Reptiles 

 the permanent ova are formed in two ways: — either (1) by the 

 direct development of one of the primitive ova, which surrounds itself 

 with a follicular layer derived from the ordinary undifferentiated 

 germinal cells; (2) a number of primitive ova coalesce together to form 

 a nest ; their nuclei multiply, and some atrophy, serving as pabulum 

 for a limited number which subsequently separate off, accompanied 

 by some of the undifferentiated germinal cells, to form as many ova. 

 These two modes of development are not regarded by Balfour as 

 morphologically very different ; the latter mode of development has 

 been brought about to secure the adequate nourishment of a certain 

 number of cells which form the permanent ova. 



Balfour's ^ observations certainly do not show any morphological 

 difference between the ova produced in these two different ways ; in 

 both cases the ovum is the equivalent of a single cell ; but the 

 physiological difference is considerable. 



I have referred above (p. 512) to the presence of two kinds of cells 

 among the central cells, distinguishable by the characters of their 

 nuclei ; in some the nucleus was rather larger and more oval in form 

 than in others, where it was smaller and rounder, and, in fact, exactly 

 like the nucleus of a follicular cell. It is possible that the cells with 

 larger nuclei correspond to primitive ova and the smaller to the 

 ordinary germinal cells ; in this case the resemblance of the central 

 mass of cells to the egg-nest of the Elasmobranch will be dimin- 

 ished ; against this supposition is the fact that there are nuclei of 

 intermediate size, but these may have been produced by a recent 

 division of some of the other nuclei. 



Judging from analogy, however, it is probable that some of the 

 cells of the germinal epithelium are specialized into primitive ova from 

 undifferentiated germinal cells, although in patches of germinal epi- 

 thelium covering the ovary I failed to detect any such specialization 

 in the nuclei. 



The important facts to be borne in mind in comparing the egg- 

 nest ofthe Elasmobranch with that of the Dipnoi'' appear to me to be 

 the early formation of the complicated follicular layers in the latter and 

 the early commencement of yolk-secretion. Tlie germinal cells being 



^ Balfour, loc. cit. 



^ It is iraportaut to remember tliat tlie formation of "uests" is uot confined 

 in Elasmobranchs to the embryonic period (see Balfour, loc. cit. p. 415). 



