B 
 Desor on the Embryology of Nemertes. 5 
it, it is necessary to compress it more or less, and then it will 
be seen torn open like a mass of jelly.’ 
The absence of a peculiar membrane surrounding the sid. 
lus will not surprise us, if we consider that the transparent 
liquid contained in the flask is not similar to the albumen 
found in the eges of higher animals, but belongs more 
properly to the vitellus itself; so tbat the membrane sur- 
rounding it is the true representative of the vitelline mem- 
brane, or chorion of the higher animals, and by no means 
the representative of the shell-membrane. 
The furrowing of the vitellus begins as early as the third, 
and even sometimes the second day. "The most striking ap- 
pearance, as distinguishing this from other animals, is the great 
irregularity of the divisions. We find nothing of the almost 
geometrical progression and external regularity that exists in 
many mollusks and fishes. At first, we see a few folds form 
on the border of the vitellus; which divide it into irregular 
lobes, as seen in fig. 13. It is important to remark, that in 
this state the clear spot is undivided. It is plain, therefore, 
that the division of the yolk is in no way dependent upon the- 
transparent spot; at least, that it is not occasioned by a pre- 
vious division of the latter. In this respect, my observations 
confirm completely the results obtained by my friend Dr. Vogt, 
as to the development of the Actæon.? 
The furrowing goes on gradually, till the whole mass is 
divided into a number of fragments of irregular size and shape, 
as seen in fig. 14, which shows one of the flasks on the fifth 
day. At this time each fragment shows a clear spot which is 
smaller than the single spot in an undivided yolk. Neverthe- 
less, the sum of these spots seems to represent a larger mass 
than the primitive spot. "Their form is spherical, as is ascer- 
tained by the following experiment. Having compressed the 
1 The vitelline granules being very cohesive in the Nemertes, it often happens, 
. that when a r adhere to each other and form balls, which might easily be 
mistaken for large 
2? See Annales lie: ibd Naturelles. 1846. 
