Yolk of Teleostean Ova. 87 



statement : — " Sur quelques oeufs, j'ai vu une communication 

 s'etablir entre le germe et la gouttelette du globe vitellin, 

 comnie si le germe allait puiser a cette source de nutrition ; " 

 and in the figure vvliicli lie gives (pi. i. iig. 14) it is difficult 

 to explain the existence of tlie column of protoplasm con- 

 necting the globule and the germ, except as indicating a 

 tropliic function, unless tlie ovum were abnormal, which it 

 most probably was. Unfortunately the ovum in question was 

 not fertilized, and the subsequent fate of the globule was not 

 ascertained. In studying the complete development of the 

 ling, gurnard, catfish, and other forms at St. Andrews 

 unusually favourable opportunities were afforded for tracing 

 tire destination of the globular bodies, and the evidence 

 gained strongly militates against Van Bambeke's theory that 

 the germinal disc is nourished by them. Not only do they 

 show no decrease in size and persist in the pendulous yolk 

 for some time after the embryo is liberated, but the actual 

 position of the globules in the early ovum is unfavourable to 

 Van Bambeke's view. 



As already stated, the normal position of the globules is con- 

 stant, viz. in the upper segment of the ovum, at the vegetal 

 pole, and they therefore occupy the region most distant from 

 the germinal disc. In a small number of eggs, however^ this 

 is not the case, and a remarkable example described by 

 Agassiz and Whitman * is the ovum of Cottas grceiilandicus^ 

 in which from ten to forty globules are more or less evenly 

 scattered over the surface of the yolk. This ovum, strange 

 to say, is ]jelagic ; whereas all other Cottoids, so far as known, 

 have demersal eggs, and all alike are abundantly supplied 

 with large oil-globules. A similar condition occurs in the 

 ovum of Trachinus vipera — in which the oil-globules according 

 to G. Brook are " scattered over the upper hemisphere of the 

 yolk, and lie between it and the vitelline membrane " or cap- 

 sule f. The large globule in Lota vulgaris is central, but 

 this position is very unusual, and it is perhaps permissible to 

 suggest that Van Bambeke may have mistaken the apparent 

 for the real position of this body. The globule always rises 

 to the upper side of the egg, and when the latter is on the 

 stage of the microscope, unless by very careful manipulation, 

 the actual position of the vesicle cannot be made out. Viewed 

 under the usual microscopic conditions, the oil-globule in 

 Motella, Trigia, Molva, &c. appears to be central, when in 

 reality it is not so. The oil-globule in truth occupies diiferent 



* ' Studies from Newport Mar. Lab. — XVI. Develop. Ods. Fishe?/ p 7. 

 t G. Brook, Journ. Linn. Sue, ZooL^gy, vol. xviii. p. 274. 



