204 Mr. J, T. Cunningham on some 



says that I do not now hold the view that oil-globules occur 

 in the perivitelline space. It is true that in a paper published 

 in 1885, when I described the movement of the oil-globule of 

 the egg of Trigla gurnardus, I was led into the error in 

 supposing that the oil-globule moved between the naked 

 surface of the yolk and the vitelline membrane. At that 

 time I gave no further study to the ovum of the gurnard nor 

 had I studied any other ovum containing oil-globules ; 

 but before the paper by Professors M'Intosh and E. E. 

 Prince was published 1 had, in my paper on the development 

 of Teleosteans occurring in the neighbourhood of Plymouth 

 (4), already explained that the oil-globule in the ovum of the 

 mackerel and the gurnard is situated within the thin pellicle 

 of protoplasm which encloses the yolk. My words were : — 

 " Thus it is evident that the yolk is to be regarded as a liquid 

 enclosed within a layer of protoplasm continuous with the 

 blastoderm, and at the surface of this liquid, next the proto- 

 plasmic layer, moves the oil-globule." 



Of course these remarks of mine were intended to apply to 

 Scomber and Trigla only, not to the relations of the oil- 

 globules in all ova. But Prof. Prince, the pu])il and fellow- 

 worker of Prof. M'Intosh, in his paper on what he was pleased 

 to call " oleaginous spheres " in Teleostean ova (8) , stated 

 that in some eggs the oil-globules occur outside the yolk in 

 the inrivitelline space. His words are {loc. cit. p. 88) : — 

 " The oil-globule in truth occupies different situations in 

 different species, occurring within the yolk-mass or outside it 

 in the perivitelline space, or rather in a fossa or pocket 

 indenting the surface of the yolk. Examples of the latter 

 condition are afforded by the Gadoid ovum studied by Hseckel 

 and bv Motella mustela, Lopkius jytscatorius, MoJva vulgaris^ 

 and other forms." In tlie recent large memoir of Professors 

 M'Intosh and Prince (9) I find no reference to or contradic- 

 tion of this statement, and yet there can be little doubt that 

 it is as erroneous as my own earlier remark concerning the 

 oil-globule in Trigla — the truth being that in the cases men- 

 tioned by Prince the oil-globules are enveloped by a proto- 

 plasmic pellicle continuous with the protoplasmic layer which 

 envelops the yolk and are therefore immovable. In fact in 

 the case of Trigla and Scomber in the course of development 

 the oil-globule becomes enveloped by the j^rotoplasm of the 

 periblast, and the periblast is formed by the increase in thick- 

 ness of the original protoplasmic envelope of the yolk. 



In all non-pelagic ova, and in some pelagic, the yolk itself, 

 apart from the presence of oil-globules, is heterogeneous and 

 discontinuous. It usually in this case consists of a large 



