104 BULLETIN OF THE 



in a small elevation lying directly over the germ, and concluded, as the 

 result of experiments (pp. 459-462) made on Gasterosteus pungitius, 

 that " the function of the micropyle is to admit the spermatozoids to the 

 surface of the yolk " ; and His ('73, pp. 3, 4) described with some detail 

 the structure of the micropyle in Salmo salar and S. fario, in both of 

 which he recognized a shallow depression (" Mulde ") surrounding the 

 crater, which in S. fario terminated in a deep funnel, and this in the 

 canal. He also showed that ouly one spermatozoon at a time can pass 

 the micropylar canal, which terminates somewhat eccentrically over the 

 germinal disk. 



Hoffmann ('81, pp. 33-36) has confirmed for a large number of fishes 

 the observations that the inner end of the micropyle terminates in a 

 papillary elevation of the zona radiata, and that in the ovarian egg it 

 lies directly over the germ. From a comparison of the dimensions of 

 the spermatozoa and the calibre of the canal, he also draws the conclu- 

 sion that not more than one spermatozoon can traverse the micropyle 

 at a time. In nearly all the micropyles figured by Hoffmann the canal 

 is a tubular passage without any special enlargement ; but in the case 

 of the herring's egg — which is the one most carefully described — the 

 outer half of the passage is enlarged into a conspicuous bulbous cavity 

 (Taf. I. Fig. 19), which, so far as I recall, has been seen in only one 

 other instance, that of Petromyzon as figured by Calberla. But the 

 greatest interest attaches to the conditions figured for Leuciscus rutilus 

 (Taf. I. Fig. 20). In this case there exists a distinct plug of granulosa 

 cells occupying the depression in the egg membrane at the micropylar 

 region. Since the author does not mention the fact in the text, it is 

 probable that he attached to it no importance. I think, hcwever, it is 

 the first clear proof published of the existence of a specialization in the 

 granulosa of the micropylar region in any teleost. On account of the 

 rather diagrammatic rendering of the granulosa cells, it is not possible 

 to be very confident about the existence of a specialized micropylar cell, 

 but the fact that a single cell forms the apex of the plug favors that 

 view, and I shall be surprised if such a structure is not hereafter demon- 

 strated in this European fish. 



Of the more recent writers on the micropyle, Owsjannikow ('85, 

 pp. 11-13, Taf. I. Figs. 5-7) describes for Osmerus eperlanus a micro- 

 pylar apparatus composed of two portions, an external and an in- 

 ternal, corresponding respectively to the two membranes which envelop 

 the egg, — the " external zona radiata" (which corresponds, in his 

 opinion, to the outer [capsular] layer in Perca) and the " internal zona 



