1897,] Embryology. 453 
isolating the first two cells of the egg and ultimately obtained from 
each cell a perfect, symmetrical, free-swimming larva. 
Contrary to what has often been stated for the result of similar ex- 
periments upon other animals these larve are not half the normal size, 
though each arises from half an egg. 
Each larva is larger than half a normal larva. ‘There are also cer- 
tain remarkable facts concerning the size of the organs and the number 
of cells in these half-egg larve. Thus while the intestine and the 
muscle segments appear on transverse section much smaller than in the 
normal larva, the medulla and the notochord are equal in transverse 
section to the normal. In the medulla and in the muscle segments the 
nuclei have the same size in the half-egg larve as in the whole egg 
larvee. 
The number of cells seen in cross section is half as great in the mus- 
cle segments of the half-egg larva while the number of cells in the 
medulla is the same in the half-egg larva as in the whole-egg larva! 
It seems that certain structures may be formed with less than the 
normal number of cells while others have the normal number. 
Do the Astral Rays pull or push ?—Ludwig Rhumbler’ con- 
cludes that the radiations often seen as star-like figures during cell 
division are probably lines of pulling or drawing and not lines of ex- 
tension or pushing. He thinks the only adequate explanation of cell 
division is one based upon Biitschlis’ theory of the foam-like structure 
of protoplasm and in deciding in favor of a contractile rather than an 
expansive action along the astral rays he thinks he brings support to 
the foam theory of protoplasm. 
In a previous paper* he began the first of a series of attempts to ex- 
plain cell division upon a physical basis; he assumed a vesicular or 
foam-like structure for protoplasm and also certain chemical changes 
in the centrosomes leading to periods of great absorbtion of liquid. 
The withdrawal of liquid from the vesicles round about leads to ten- 
sions and, if rapid enough, to the appearance of radiating lines of 
vesicles. Based then chiefly upon phenomena of surface tension in the 
constitutent vesicles of protoplasm, each a viscid bag with more liquid 
contents, this hypothesis seeks to reduce all the complexities of cell 
division to a very few physical laws. 
The present paper gives a few noteworthy figures of sections of snail’s 
` eggs showing a marked vacuolated or vesicular appearance in the pro- 
* Archiv f. Entwicklungsmechanik. IV, März 2, 1897, pps. 659-725. PI. 28. 
*See AMERICAN NATURALIST for January, 1897, p. 84-6. 
