MUSEUM OF COMPAKATIVE ZOOLOGY. 85 



cover nothing, he asserts that he saw in all cases with the greatest 

 distinctness that the zona radiata first appeared, and that when it 

 had attained a certain thickness then for the first time the villi and 

 the zonoid layer made their appearance, almost simultaneously. Un- 

 fortunately, Brock has given no details concerning the proof of this 

 assertion, either in figures or description. I am tolerably certain, not- 

 withstanding the very positive way in which he maintains his conclu- 

 sion, that Kolliker was right, and that he is wrong, for I cannot believe 

 that in so fundamental a matter there is such a difference between 

 fishes as would be implied by admitting both views to be correct. 



That the differences of opinion which the egg membranes have given 

 rise to are not exclusively due to the study of different fishes, is clearly 

 seen from the results reached by Kupffer, and soon after by Hoffmann, 

 in the study of the herring's egg. Kupffer ('78 a , pp. 177, 178) found 

 the yolk to be closely invested by an egg capsule 6 to 8 /x in thick- 

 ness, and the latter to be covered by a layer of viscid semifluid sub- 

 stance, which was found to be of nearly uniform thickness if the eggs 

 were dropped into alcohol without contact with water. In water it soon 

 becomes solid. The capsule consists of two firmly united layers, the 

 inner one being finely striate radially, and alone equivalent to the po- 

 rous capsule (zona radiata) of other eggs. The striation is due to pore- 

 canals. These do not traverse the outer layer, which has concentric 

 striations. A boiling ten per cent solution of potash dissolves the po- 

 rous layer quickly, but does not affect the outer layer of the capsule, 

 which is believed to be the same as that described for Esox by Aubert, 

 and for other fishes by Kolliker. This difference in the two layers 

 Kupffer regards as favoring Eimer's view, that one is produced from 

 the egg and the other from the follicular cells. 



On the other hand, Hoffmann ('81, pp. 15-33), who has given the 

 subject of egg membranes in fishes the most extensive treatment of any 

 recent writer, differs materially from Kupffer in his account of the her- 

 ring, although he - offers a partial explanation of their differences in 

 saying that Kupffer examined only fully mature eggs, and such as had 

 been in contact with sea-water. Hoffmann makes the total thickness of 

 the membranes in not quite mature eggs to be 32.5 p. The outer layer, 

 10 fx thick, is not separable from the inner, although a dark line marks 

 the boundary. Both layers are traversed by numerous pore-canals, 

 which, to judge from his figures, appear to be finer (not necessarily 

 nearer together) in the inner than in the outer layer; but whether they 

 are continuous it is impossible to say on account of the sharp line sepa- 



