202 MR GEORGE BROOK ON THE 



with acid alcohol extracts all the stain from the yolk, so that the protoplasm 

 remains deeply stained with carmine, while the yolk retains its original pale 

 yellow tint, which is, however, too weak to show in thin sections. Unless the 

 acid alcohol is allowed to act for some hours the yolk is not entirely robbed 

 of its stain. 



Fig. 3 (PI. XIII.) represents a section through the middle of an egg, prepared 

 as I have described. The germinal protoplasm is seen to be finely granular under 

 a high power, but the granulation is so delicate and regular that, with a low 

 power, the protoplasm appears quite homogeneous. The large yolk spheres 

 occupy the greater portion of the egg, and are comparatively evenly distributed. 

 There is not so great a variation in the size of the yolk spheres at this as in 

 later stages. Towards the margin, however, there is a diminution in size, 

 leading up to the small yolk granules which lie immediately beneath the egg 

 envelope. Kupffer says nothing as to the origin of these yolk granules, but 

 Hoffmann has indicated their possible mode of origin. According to this 

 observer, the unripe ovum contains little protoplasm, and a large number of the 

 large yolk spheres about '035 mm. in diameter. It is to be presumed that at 

 this stage there are no yolk granules, although this is not distinctly stated. 

 Later, in the ripe unfertilised ovum, the large yolk spheres are fewer in number, 

 and there is more protoplasm. While most of the yolk spheres are compara- 

 tively homogeneous in structure, a few are seen to be filled with smaller spheres 

 only '002 mm. in diameter. These are often seen lying in clusters outside the 

 larger spheres, and would appear to be set free by the rupture of the walls of 

 the latter. Hoffmann does not, however, offer any theory on the subject, nor 

 does he give any figures. 



There is undoubtedly a greater proportion of protoplasm to food yolk in the 

 ripe ovum of the herring than is to be found at an earlier stage. The small 

 refractive yolk granules are also not present in the unripe ovum so far as my 

 observations go, but I have not been able to trace their origin. I have never 

 observed a collection of small spherical "granules" inside one of the yolk 

 spherules, nor any collection of them which would lead one to infer the origin 

 assigned to them by Hoffmann. Judging from the subsequent behaviour of 

 the food yolk, I am not inclined to accept Hoffmann's view without further 

 evidence. The yolk of Teleostean fish ova, and particularly of the herring, is 

 simply a collection of food material which has no cellular value whatever, and 

 does not undergo a segmentation comparable with that of amphibian ova. If 

 Hoffmann's views are accepted, there must either be a segmentation in the 

 yolk spheres themselves, or the small yolk granules are the result of a 

 mechanical subdivision of the larger masses. The former proposition appears 

 entirely unwarranted by the future behaviour of the yolk. A mass of yolk, 

 which can form around itself a thin wall, subdivide its contained food material 



