[39] EMBRYOGRAPHY OF OSSEOUS FISHES. 403 



in oi)tic section, normal ova sometimes show the segmentation cavity as 

 a space of considerable depth, as is indicated in the outline sketches of 

 sagittal sections of tvpo early stages of the blastoderm of the cod, Figs. 

 47 and 48. With the further development the depth of the cavity di- 

 minishes, often to such an extent as to apparently vanish in ova in 

 which the yelk has been included by the blastoderm. It is to this fact 

 that we may ascribe the belief, current amongst investigators, that it 

 wholly disappears. 



It is quite impossible to reconcile the account given by Haeckel with 

 the facts as presented by others, when he implies that the whole under 

 surface of the blastoderm is lifted up from the yelk and remains in con- 

 tact with the latter oidy round its margin. The margin at the same 

 time, he says, is reflected inwards, a single layer of cells growing in- 

 wards from all sides, to finally close in the center, forming the hypo- 

 blast in that manner. This is not in accord with what the writer has 

 seen in the egg, of the cod, nor can it be substantiated by the classical 

 researches of (Ellacher made two years prior to Haeckel's,* nor by my 

 own more recent investigations during the past two years. There is no 

 evidence to show that the epiblast of the blastodermic disk is reflected 

 inwards to develop the hypoblast. Haeckel says further, that the 

 clear fluid in the segmentation cavity is resorbed and that the cavity it- 

 self disappears entirely. This statement the writer disputes in toto, 

 with an abundant supijort of facts in his behalf. He will only mention 

 here that not only does this cavity persist, but that it also actually in- 

 creases in size during the later stages of development, as may be ob- 

 served after yelk absorption has begun, as may be seen in Coregonus 

 albus, Elacate canadus, (Jyhium maculatum, FarepMppns faber, etc. 



In Fig. 18 the left half of a blastoderm of a cod's egg of the latter 

 part of the seventh day is seen in median section along the plane of the 

 axis of the embryo from A to B ; the thickened portion or embryonic area 

 has divided into two thick lamina or strata, each several cells deep. 

 The whole of the upper surface of the blastoderm is covered by a very 

 thin single layer of epiblast cells which pertain to the epithelial layer. 

 The bilaminate condition of the blastoderm extends also into the rim 

 or annulus r, which extends around about one-half of the disk and 

 widens just below A, where it is blended and confounded with the em- 

 bryonic shield or area. The thinner portion of the blastoderm, extend- 

 ing from A to the upper border of r, is composed of two layers of cells 

 which roof over the very shallow segmentation cavity. The outer of 

 these is the epithelial layer already referred to, and the inner one com- 

 posed of rounded cells answers to the sensory layer of embryological 

 writers. Such, in brief, is an outline of the history of the blastoderm 

 immediately after it hiis clearly become such and before there is as yet 

 any distinct differentiation of the axis of the embryo which is now marked 

 only by the thickest portion of the b lastoderm in the median region 

 * Jenaische Zeitschrift IX, 1875, 



