48o 
The An 
Naturalist. 
[Jui 
mass of cells, ini, which are rounded or polyzonal and 
densely g^ranular. At one point the outer layer is inter- 
rupted and the space is filled by one of the granular segments 
of the inner mass (Fig. 8). The nuclei of all the cells 
are somewhat nodulated, 
and have several highly 
refractible granules each. 
The granules in the 
bodies ofthe cells of the 
outer layer are somewhat 
concentrated around the 
nucleus, leaving the cor- 
of the ceils clear, 
Beneden, 9, 28-29, 
observed that some- 
;s(2i oval out of 29) 
first two segmenta- 
the 
lilar variability 
in the mloe, 
Heape, 23, 165 ; Tafani, on the other hand, expressly denies 
its occurence in white mice. It is, I think, very improbable 
that this difference, which sometimes occurs and sometimes 
does not. has any fundamental significance ; van Beneden, 
however, has maintained that the small cell gives rise in the 
rabbit to the inner mass of cells, (see below) which he terms 
the entoderm, but which must, it seem to me, be homologized 
with the ectoderm, as explained below. That van Beneden 
is in error, as to the genetic relation ofthe small cell to the 
inner mass has been demonstrated by Heape^ 23, 166. 
The second cleavage plane is probably also meridional, and 
is certainly at right angles to the first, so that four similar 
cells are produced as in the primitive type of segmentation, 
(Fig. 9) those four cells are also rounded at first, and prob- 
ably become fitted against one another so as to produce the 
