310 PROCEEDINGS OP THE AMERICAN ACADEMY. 



in the number of cells over a certain area, so that they come to lie 

 more closely together, and in consequence give the appearance in sur- 

 face view of a thickened entpdermal area. The question of course 

 concerns the significance of this circum-germinal thickening. An ex- 

 planation has suggested itself to me, which rests, however, on a theory 

 which I do not feel at all sure is established. The theory concerns 

 the origin of the mesoderm. Hubrecht ('90) gives three sources for 

 the mesoderm : the protochordal plate, the primitive streak (" gastrula 

 ridge " and " KoptFortsatz "), and " an annular zone of hypoblast situ- 

 ated just outside the limits of the embryonic shield, and thus enclos- 

 ing — but at the outset independent of — the protochordal plate." The 

 annular zone according to Hubrecht does not arise until after the first 

 stages in the development of the primitive streak, and therefore is 

 a later differentiation in the hypoblast than is the protochordal 

 plate. 



Concerning this third source there has been much dispute. Bonnet 

 ('84) has found it in the sheep, and Robinson ('92) in the rat and the 

 mouse. But many authors find no evidence of such an origin for any 

 part of the mesoderm, as, for example, Kolliker ('82), Heape ('83), 

 Fleischmann ('89), Keibel ('91 and "93), Hertwig ('93), and very 

 many others. It has occurred to me that this circum-germinal thick- 

 ening might be the first evidence of a later (ontogenetically) formation 

 of mesoderm in this region. Since Keibel's ('93) youngest pig embryo 

 has the mesoderm already well advanced in development and covering 

 the area in question in two layers, somatic and splanchnic, it is impos- 

 sible to say what the stages intermediate between his and mine may 

 have been, and I merely mention the above suggestion as a possible 

 explanation of an interesting phenomenon. There is, as I have al- 

 ready said, no trace in my specimens of mesoderm in any region of 

 the vesicle, and no sign, either in surface view or in section, of a 

 thickening of entodermal cells in any part of the germinal disk, like 

 the protochordal plate which Hubrecht ('90) describes for the shrew. 

 To make sure that this was not merely a subjective impression I 

 enumerated the nuclei in each section of the germinal disk of Fig. 6, 

 Plate I., beginning at the broader end, and the result was as follows : 

 3, 8, 10, 10, 14, 15, 18, 18, 20, 23, 28, 24, 23, 25, 24, 31, 30, 23, 22, 

 17, 15, 12, 10, 9, 10, 11, 6. A comparison of these figures with the 

 shape of the disk will show an almost uniform distribution of entodermal 

 cells. 



