KEIMBAHN-DETERMINANTS 227 
posed regarding the réle of these bodies in heredity 
make it necessary to refer to them briefly here. At 
the present time it is difficult to make any definite 
statement regarding the origin, nature, and signifi- 
cance of the various cytoplasmic inclusions that have 
been grouped under the general title of mitochondria. 
It seems probable that we are concerned with a num- 
ber of different sorts of inclusions, and with various 
stages in their evolution. Inthe guinea pig (Ru- 
baschkin, 1910, 1912) and chick (Tschaschkin, 1910) 
the chondriosomes of the cleavage cells are spherical 
and all similar, but, as development proceeds, those 
of the cells which become differentiated to produce 
the germ layers unite to form chains and threads, 
whereas those of the primordial germ cells remain 
in a spherical and therefore primitive condition 
(Fig. 31, B). Swift (1914) has found, however, 
that in the chick the mitochondria in the primordial 
germ cells are not at all characteristic, resembling 
those of the somatic cells. The germ cells neverthe- 
less can be distinguished from the latter by the pres- 
ence of an especially large attraction-sphere (Fig. 
31, D). This distinction between the primordial 
germ cells and the surrounding somatic cells may 
enable us to trace the keimbahn in vertebrates back 
into cleavage stages — something that has not been 
accomplished as yet. 
An examination of the various keimbahn-deter- 
minants listed in the table (p. 212) has led the writer 
to conclude that none of them is of a mitochondrial 
nature, but the results obtained by the special methods 
