THE DETELOPMENT OE MOTELLA MTTSTELA. 299 



and curiously endugh the eggs of this batch showed more irre- 

 gularities in the early stages of development than any others 

 that were observed. In any case eggs with a number of oil- 

 globules either developed so irregularly as to die before hatching 

 out, or the small oil-globules gradually coalesced to form one 

 large one before the embryo left the shell. There was no ex- 

 ception to this rule, so that directly or indirectly an abnormal 

 development of the oil-globules has its influence on the develop- 

 ment of the embryo. 



The earliest stage observed was that in which the disk is 

 divided into sixteen cells. At this time the disk is oval in shape 

 and measures about •■iST millim. by "SSI millim. ; but these 

 measurements are only approximate, as the form and size of the 

 disk varies considerably in different eggs. Sometimes in this 

 stage, at others not until the thirty-two cell stage, the disk 

 becomes somewhat square in outline, and then measures "441 

 millim. in diameter. The disk is then somewhat concave beneath, 

 and its position is as usual a little eccentric. The disk then, with 

 increasing cell- division, becomes more and more rounded in out- 

 line until about five hours after formation of sixteen-cell stage it 

 has the appearance shown in section in fig. 1. Here the epi- 

 dermal layer of the epiblast is well defined, and the surface of 

 the disk lying on the yolk is perfectly flat. 



It is about, or a little earlier than, this stage that the first 

 collections of granules and nuclei are observed to form the 

 periblast (Agassiz and Whitman *) =parablast of Klein t. In 

 this respect the egg of Motella differs considerably from that of 

 TracMnus. In the latter a minute collection of granules is to 

 be found around the disk, even in the two-cell stage, that is to 

 say on the completion of the first segmentation-process. These 

 granules increase in size and number with each segmentation- 

 process, until in the sixteen-cell stage they form quite a striking 

 feature of the egg, as shown in plate 3. fig. 7 (Linn. Soc. Journ. 

 Zool. vol. xviii.). In Motellano such gradual development occurs ; 

 and it is not until the segmenting disk presents the characteristic 

 morula appearance that the first granules are observed. Again 

 in Tracliinus the first row of cells ' 'j 'Arming the periblast is 

 always uniform and complete before any cells of the second row 

 are formed ; whereas in Motella it is quite usual to find parts of 

 a second and third row in their places before the first row is 



* Proc. Amer. Acad. Arts and Sciences, xx. (1884). 

 t Quart. Journ. Micr. Sci. xvi. (1876). 



