EARLY STAGES OF DEVELOPMENT OF THE MOUSE. 69 



Beneden, Heape, and Selenka found the "blastopore"), 

 Duval also figures a gap in the outer layer to which he gives 

 this name. The inner mass is, according to him, endoderm 

 alone, and soon becomes everywhere reduced to a single 

 layer of more or less flattened cells spread over the inner 

 surface of the ectoderm, while the epiblastic knob is produced 

 by an inward proliferation of certain cells of the outer layer 

 (see figs. E, F, G). This is also, it should be said, the 

 account given by Duval of the formation of the embryonic 

 epiblast in the mouse, rat, and guinea-pig. Now, as far as 

 the mouse is concerned, I have little hesitation in saying that 

 Duval's interpretation is wrong. I have already said that 

 his figures are diagrammatic, and I think a comparison of 

 them with Selenka's or mine will convince anyone that this 

 is so. In my series of embryos, which I think it will be 

 admitted is a tolerably complete one, I have the clearest evi- 

 dence that the inner mass, the cells of which are at first all 

 alike, becomes gradually differentiated into a compact epi- 

 blastic knob and hypoblastic cells, many of which are isolated 

 from one another as they creep round the inside of the 

 blastocyst. The trophoblast cells overlying the inner mass 

 are, almost from the first, as Selenka originally described 

 them, very much flattened, and could not well be supposed 

 to have given rise by proliferation to the rounded cells of the 

 embryonic epiblast ; and, besides, all the mitoses that I have 

 found in the ti'ophoblast have the axes of their spindles tan- 

 gential. Finally, though I should be unwilling to lay any 

 great stress upon this, it seems to me a little remarkable that 

 the ratio of the number of cells in the inner mass to that of 

 those of the trophoblast is sensibly equal, in those stages in 

 which the epiblastic knob is already beginning to be differ- 

 entiated, to what it is in the earlier stages, in which no 

 difference is to be detected among the cells of the inner mass. 

 I think, therefore, we may take it that Duval's account of 

 the mode of formation of the embryonic epiblast in the 

 mouse is not correct, and if so we may fairly regard with a 

 certain amount of suspicion his description of the same pro- 



