EM13RYOLOGY OI' CLEPSINE. 273 



forward along the lateral edges of the germinal disc. The cells 

 composing the central area of the disc are in a process of rapid 

 multiplication hy division, resulting in centrifugal expansion. 

 This exjjansion takes place, to a certain extent, in all directions: 

 but predominantly in a transverse direction, inasmuch as the 

 chief points of resistance are offered by the cephahc and anal 

 regions. One of the important results of this laternal expan- 

 sion in which the germ- bands of course participate, is seen in the 

 movements of the blastomeres. The lateral blastomeres {a 

 and b) are pressed outward and downward, and the ventral blas- 

 tomere (<?) moves necessarily upward. This movement is fol- 

 lowed without difficulty on the living egg, and it was thus that I 

 first became aware of it. Figs. 37, 38, and 39 show the relative 

 positions of the blastomeres at about the time the movement 

 begins. The ventral blastomere [c] has still the cuneate form 

 form seen in fig. 79. Twelve hours later (figs. 41, 42, 43, 84, 

 48 hours) the upper face of this blastomere, already visible from 

 above in stage 40 as a narrow area tapering backwards, has 

 attained a considerable breadth at the expense of the lower. In 

 stage 44, 45, (54 h.) the upper face of c is much broader than 

 the lower, as is best seen in section (fig. 86). This movement 

 culminates in stage 46, 47. The full extent of the change in 

 position which takes place between stages 37 and 47 is at once 

 seen by comparing fig. 78 with fig. 86. The vfi^^xS^e [c) is inver- 

 ted. The successive positions of the mesoblasts (figs. 84-86) 

 show how they are involved in the same movement. 



In C. complanata, the object studied by Robin, this solid form 

 of invagination is quite as marked as in C. marg'mata. Clepsine 

 thus furnishes a beautiful illustration of the fact that epiboly 

 (Selenka) is only a modified form of emboly. Abolish the limits 

 between a, b, and c, or divide them into small spheres, and it is 

 easy to see that the invaginatory movement might still take place 

 although it might be impossible to recognise it. 



In order to understand the form and the movement of the 

 germ- bands, it is necessary to bear in mind that their two ends 

 are, approximately speaking, fixed. Their anterior ends abut 

 against the cephalic portion and their posterior ends are sup- 

 ported by the neuroblasts. As the bands lengthen, the central 

 field simultaneously expanding, the slight outward curve, which 

 they exhibit in stage 38, is rapidly increased. In this way the 

 central area of the blastodisc soon takes the dumb-bell form seen 

 in fig. 40 (42 h.). At this time it has become so thin that it is 

 easy to recognise the limits of the large blastomeres. 



While the expanding disc contributed strongly at first to the 

 outward bending of the embryonic bands, it is plain that from 

 this time (fig. 4U) forward, the form of the baud-curves is mainly 



