George Lefevre and Caroline McGill 479 
Our observations, however, throw no light upon the important prob- 
lem of reduction, yet certain conditions in the formation of the tetrads 
and the distribution of their component elements may be briefly indi- 
cated. A slightly schematic representation of the formation of the tet- 
rads, showing also the axial relations of the longitudinally split bivalent 
threads of the growth-period and of the tetrads on the spindle is given 
in Fig. 5. In each of the successive figures from A to J, it is the largest 
chromosome that is shown, while all are drawn under the same magnifi- 
cation. Many of the chrosomes in this diagram are redrawn with only 
unimportant modification from the macro-chromosome already repre- 
sented in previous figures. For example, D is taken from Fig. 2, K; H, 
from Fig. 3, F’; etc. 
It has already been pointed out that the long axis of the tetrad is iden- 
tical with the long axis of the chromatin threads of the growth period, 
and since the figures have clearly shown that the tetrads are placed upon 
the spindle lengthwise, it follows that the first maturation division must 
separate univalent chromosomes and be therefore a reducing division on 
the assumption that an end to end conjugation takes place between indi- 
vidual chromosomes at synapsis. And continuing this assumption, the 
second division must be equational, since the chromosomes are oriented 
in such a way as to distribute the halves of the dyads which are separated 
by the longitudinal split. This conclusion was drawn by Miss MeGill in 
her original paper. If, however, it should prove true of this form that a 
parallel conjugation occurs, as has been suggested, the first division would 
still be a reducing one, since the axes of the crosses are not reversed by 
the drawing out of the transverse arms and the attachment to the spindle 
fibers is at the end of the longitudinal arms, as seen in Fig. 5, #. But 
in the event of the chromosomes conjugating side to side, it would still 
have to be shown that they separate after fusion as the originally distinct 
chromosomes without loss of identity before a real reduction could be 
established. The crucial problem, then, in the whole question of reduc- 
tion lies at this point, and until further light is thrown upon it by future 
investigation, it is futile to indicate possibilities and frame speculations. 
BIBLIOGRAPHY. 
BoNNEVIE, Kr., 05.—Das Verhalten des Chromatins in den Keimzellen von 
Enteroxenos oOstergreni. Anat. Anz., 26. 
o6.—Untersuchungen tiber Keimzellen. I. Beobachtungen an den 
Keimzellen yon Enteroxenos oOstergreni. Jen. Zeitschr., 41. 
——— o7.—“ Heterotypical”’ Mitosis in Nereis limbata (Ehlers). Biol. Bull., 
il}, 
