260 BULLETIN OF THE 
as the axis of the adult Agalma, although it is not clear that the right 
and left sides of the disk-like elevation correspond with the right and 
left sides of the appendages later found on the adult Aga/ma axis. The 
general appearance of the yolk and the size of the egg is approximately 
the same as in the preceding stages. The right and left sides used for 
figures up to Pl. III. fig. 4 have not the same significance as here 
interpreted. 
The next oldest larva (fig. 9) differs primarily from the last in the 
greater elevation and prominence of the layers formed on the yolk. The 
epiblast and hypoblast are much thicker; the former has a reddish, 
the latter a yellowish color. The constriction around the elevated 
disk between its edges and the surface of the ovum has deepened on 
the distal side of the elevation as seen in profile, but the indentation 
is very slight on the proximal side. 
Within the disk a gelatinous layer, so transparent as to be invisible, 
has formed by a separation of the epiblast and hypoblast. The thickness 
of this layer is greatest near the distal end of the disk. Yellow and 
reddish pigment is found in the epiblast on the surface of the yolk sac, 
It was also noticed that the epiblast at pn. cy., near the proximal end of 
the elevated disk, is much thicker than that near the distal side, and 
that there was a tendency to form a slight epiblastic elevation at that 
point. If the reader will compare the figure of this stage with one of 
about the same age by Metschnikoff, he will find a great difference in 
external shape between the two. My larva is approximately the same 
as Pl. VIII. fig. 5 in the oft-quoted work by that author, who says that 
his larva is five days old. My adult Agalma was put in the aquaria 
on August 6, and the stage represented in fig. 8 was found free in 
the water on August 8, or two days later. I likewise picked out of the 
same water three days after, or five days after the adults were put there, 
larvee of the same age, while with these were still others much farther 
advanced, and some which were just passing through the early stages of 
segmentation of the egg. 
I find a discrepancy, which may be a generic difference, in the rate of 
growth day by day recorded in Haeckel’s observations on the develop- 
ment of Crystallodes, and Metschnikoff’s of Agalma. In larve of Crys- 
tallodes four days old the float was as far advanced as in the Aga/ma six 
days old of Metschnikoff, while on the second day both the Aga/ma and 
Crystallodes larvee were still in a morula stage. These discrepancies 
arise from the difference in the mode of growth of the float in the genera, 
or from the fact that different clusters of eggs, or different members 
