BULLETIN OF THE UNITED STATES FISH COMMISSION. 9 



that tlic disk cliaiigi^s form somewhat iu the same way as does the seg- 

 iiientiiig blastodisk of a Teleosteaii ovum. This conclusion is supported 

 by the following data : The disk of an apparently unfertilized egg re- 

 moved from a recently-captured female, was perfectly discoidal, whereas 

 the disk of the nearly completed 16-celled stage was elongated and had 

 a decidedly squarish outline when viewed from above. The latter meas- 

 ured 1.71 millimeters in width and 2.37 milliuieters in length. Its thick- 

 ness in the center was about .C of a millimeter, and thinned out at the 

 margin into a very thin layer of protoplasm which is obviously homol- 

 ogous with the periblastic layer of the Teleostean egg. 



Judging from the arrangement and depth of the segmentation furrows, 

 the first one, I, divided the disk into two halves, and in the process of 

 segmentation the disk became narrowed at right angles to the plane of 

 the first cleavage. This modification seems to have influenced the char- 

 acter of the subsequent cleavages, as the next furrow, II, is about at 

 right angles to the first, and traverses the middle of the disk through 

 its longest diameter. The disk is now segmented into four large cells. 

 The next cleavage furrows, III, III, cut through the disk transversely 

 across its least diameter, and the two of them divide the four cells re- 

 sulting from the first and second cleavages into eight. These are 

 arranged in two parallel rows of four each and embrace the long diam- 

 eter of the disk, just as in the blastodisk of the Teleostean egg at the 

 end of the third cleavage. Some irregularity now becomes apparent 

 upon the advent of the fourth cleavage upon one side, as a result of 

 which it is clear that the disk will soon lose its oblong, subquadrate 

 form and become discoidal in the same way in which this happens in 

 the eggs of bony fishes. On the right side of the blastodisk of Eaia, 

 the furrow, IV, of the fourth cleavage is quite regular, on the left very 

 irregular, as shown in the figure. 



Sections through the disk display the relation of the segmentation 

 spheres to the vitellus. The furrows do not appear to cut quite through 

 the less granular protoplasm of the disk, so that probably a thin peri- 

 blastic stratum is left underlying the latter and immediately overlying 

 the yelk, the coarse granules and corpuscles of which are apparent just 

 below. The marginal cells in section have a thin border, and the outer 

 twelve cells are wider than the four central ones when viewed from 

 above. The thin borders of the marginal cells are directly continuous 

 with the thin periblastic envelope which invests the vitellus. 



The nuclei are relatively small, and contain rather dense single, but 

 somewhat irregular, masses of chromatin. 



Somewhat more advanced stages show the disk divided irregularly 

 into cellular areas ; the whole disk is also more nearly discoidal in form. 



No later stages were observed in which the number of cellular layers 

 had multiplied, or where the disk showed two or more superimposed 

 strata of cells ; and in none of those examined by me did I find any 

 evidence of the presence of a developing cleavage cavity, nor was it 



