214 MR GEORGE BROOK ON THE 



any trace of karyoldnctic figure. So far as the herring is concerned, other 

 phases of the process may be followed, which give a clue not only to the 

 direction of the early planes of cleavage, but also show in a most decided 

 manner the true nature and mode of origin of the two primary germinal areas 

 — the archiblast and parablast. 



According to Kupffer, the first furrow is meridianal in direction — that is, 

 in the direction of the axis of the egg. The second is equatorial, and with the 

 completion of this the germinal area is divided into archiblast and parablast. 

 The third furrow is meridianal, but at right angles to the first, and after this 

 the segmentation proceeds in the usual way. 



Hoffmann, on the other hand, maintains that in the fish ovum the first 

 segmentation furrow takes an equatorial direction, so that instead of the first 

 cleavage process resulting in the formation of two segmentation spheres, the 

 germinal area is divided into two layers corresponding to the archiblast and 

 parablast, each of which contains half of the original segmentation nucleus. 

 According to this arrangement, the parablast is given an equal value with the 

 archiblast at the outset, and yet Hoffmann denies that it takes any part in the 

 formation of the tissues of the embryo. 



The majority of observers have not described an equatorial furrow in the 

 Teleostean fish ovum until a considerably later stage, and it has been generally 

 accepted that there is no furrow in the fish ovum which is equivalent to the 

 first equatorial furrow (the third of the series) in the amphibian ovum. 



Agassiz and Whitman (1) differ from other investigators in attributing the 

 whole of the germinal protoplasm in the first instance to the archiblast, and 

 derive the nuclei of the parablast from the archiblast as secondary products 

 which are derived from its marginal cells. 



From my own investigations, I conclude that in the herring, as in so many 

 other forms, the first furrow which takes an equatorial direction is the third 

 of the series. This furrow is therefore comparable with the third furrow in the 

 amphibian ovum. 



After the protoplasm in the germinal area has increased considerably in 

 bulk at the expense of the yolk, there is usually a comparatively quiescent 

 period. During this stage a part of the germinal protoplasm collects at the 

 yolk pole, and there forms a small mound. As soon, however, as the first traces 

 of a furrow are to be seen this mound gradually disappears, and the protoplasm 

 of which it was formed slowly travels along the surface of the yolk to join in 

 the approaching period of activity in the germinal area. There remains, how- 

 ever, a thin film of protoplasm around the yolk both in this and in later stages. 

 This does not form a flat layer, but is seen in section to follow the outline of 

 the yolk spheres, and frequently to fill in the spaces between them. 



The surface of the germinal protoplasm, which has hitherto been much 



