[51] EMBEYOGRAPHY OF OSSEOUS FISHES. 505 



oral invagiuation, as is the case with birds, mammals, Elasmobrauchs, 

 and Amphibians. I have a number of such longitudinal sections of 

 Alosa which show the hyj)ophysi8 connected with the oral ephithe- 

 lium by a narrow stalk. In sections of much earlier embryos 1 find it 

 exceediDgly difficult to detect this structure with certainty. I believe, 

 however, that Professor Dohrn is quite right in holding to the belief 

 that it does not originate from an epiblastic involution of the stomo- 

 dseum. In fact it would almost seem to be demonstrated that the mouth 

 of the young fish is developed, as Dohrn shows, from behind forwards 

 and that it really has no stomodseiim as we know that structure in 

 other forms. According to the above mentioned authority the mouth is 

 developed from the anterior part of the mesenteron and that it at first 

 grows out as two narrow, pointed, horizontal clefts which break through 

 at two points on either side of the middle line even before the head has 

 grown out over and beyond the epiblast which is continued over the 

 snout and down over the yelk. This is a very singular state of affairs, 

 but I am not assured from my own investigations that what was observed 

 in vertical longitudinal sections of Hippocampus aud Belone will apply to 

 Alosa. The origin of the mouth of the Teleostean embryo is a very 

 difficult subject to work out. In embryos in which the mouth is just on 

 the eve of opening I cannot convince myself positively that I can see what 

 Dohrn claims to have done. It is true there seems to be a less pronounced 

 development of the oral tract of hypoblast near the point where the mouth 

 ought to open in the middle line, but I cannot convince myself of its total 

 absence. In sections off of the middle line th'e walls of the oral opening 

 are more pronounced, but I can not yet agree that it is decided that the 

 mouth of the Teleostean embryo first opens at two points a little way off 

 of the middle line. In living embryos as well as in hardened ones, treated 

 with chromic acid, the mouth opens as a small tran verse opening, and as 

 development proceeds the rim of the upper jaw is carried forward beyond 

 the line of the lower. But this is digressing again from the subject of the 

 hypophysis, which it appears we cannot regard in the present state of our 

 knowledge at least, as certainly originating from an oral invagination. 

 Although the hypophysis is not a part of the brain, as lias been positively 

 demonstrated by Rathke, Goette, Balfour, Dohrn, and others, its develop- 

 ment naturally falls within the limits of a description of the formation of 

 that organ. It is a very diminutive structure in Teleostean embryos, even 

 after hatching, and is pushed up between the cranial trabeculae, still 

 retaining its connection with the oral epithelium even on the third day 

 after incubation in Alosa. It rests in a bowl-shaped depression on the 

 lower face of the infundibulum and only two or three of a series of thin 

 sections through the mesial region will usually strike it. 



The pineal gland, on the other hand, is clearly a part of the brain ; 

 sections through the middle line often strike it and show it as a de- 

 pressed, biscuit-shaped body with a stalk at its anterior portion con- 

 necting it with the forepart of the roof of the mid-brain. It is very 



