MUSEUM OF COMPARATIVE ZOOLOGY. 109 



paratus was once occupied by granulosa cells, then either the true 

 micropylar canal has not been seen in Myxine, or it is formed in a 

 different manner from that of bony fishes, for in no case is it occupied 

 in the latter by a plug of cells from the follicular epithelium. The 

 former alternative is, in my opinion, the more probable. This con- 

 clusion I base partly on the appearance of Cunningham's Figure 3, and 

 partly on the conditions presented by sections of an ovarian ovum of 

 Myxine australis, which I have studied since reading Cunningham's 

 paper. In his Figure 3, the plug of granulosa cells which is sunk 

 into the yolk is completely enveloped in a uniform layer not more than 

 2 /x, thick, which not only separates its deep end from the yolk, but 

 also its sides from the membrane called by him vitelline membrane, or 

 zona radiata. What the significance of the part of this thin mem- 

 brane lining the chimney-like elevation of the zona radiata may be, I 

 cannot say, unless it is reflected at the upper edge of the chimney to 

 form the outermost layer of the zona ; but the portion which separates 

 the yolk from the granulosa plug I regard as the equivalent of the first 

 formed portion of the zona radiata, and believe that the true micropylar 

 canal will be found in the form of a minute passage through that mem- 

 brane. My principal reason for this opinion rests on the condition of 

 this region in the egg of M. australis. This egg was about 22 mm. 

 long and 6 mm. in diameter. It was enveloped by a follicular epithelium 

 composed at the animal pole of a single layer of cells, averaging about 

 10 fx thick. Over the region of the flattened germinative vesicle the 

 granulosa was gradually thickened to about 25 /x, and from the middle 

 of the thickening a solid plug of cells about 35 fx long and 25 fx in di- 

 ameter projected into the yolk. The membrana propria of the follicular 

 theca passed over the micropylar region without being at all infolded, so 

 that the total thickness of the granulosa, measured from the apex of 

 the plug, was about 60 /x. The cells of the plug were not well pre- 

 served, but appeared to be of about the same relative size as in Cun- 

 ningham's Figure 3, — i. e. of one half or one third the diameter of the 

 plug. There was no enlargement at the apex of the ping as seen in 

 his figure. Between the granulosa and the yolk, and in contact with 

 both, was a highly refractive thin (3.5 /x) membrane, which at first 

 appeared homogeneous, but in which I believe I have detected at 

 intervals radial markings. This membrane became thinner at some 

 distance from the pole. 



The whole apparatus had such a striking resemblance to the micropy- 

 lar apparatus in Lepidosteus that I cannot doubt that the granulosa 



