| 
of chromosomes then forms a small nucleus, one of which lies outside 
the cell and forms the “second polar body”. The remains of the 
spindle are for some time conspicuous in the developing spermatid as 
a dark striated body lying near the nucleus (Fig. 5). The nucleus 
of the polar body and that of the spermatid develop similarly for 
Some time, but in the later stages of spermatogenesis the polar nuclei 
are no longer recognisable. 
It appears therefore that in the Bee the nucleus of the primary 
spermatocyte resolves itself into 8 dyads, instead of the usual tetrads; 
there is only one complete maturation division, which separates is 
halves of the dyads so that 8 single elirotrodonies pass to each pole. 
es 
SEE. 2. Fig, 3: Higcit, 5 in 
Fig. q | ee 
=) | Rael | | m 
. FAN 4 “Oy 
- \ oo Be ey : } £ 
Fig. 1. Nucleus of primary spermatocyte becoming resolved into 8 pairs of 
chromosomes. 
Figs. 2 and 3. Metaphase of second spermatocyte mitosis; all the chromosomes 
are shown. 
Fig. 4. Anaphase, second spermatocyte mitosis. 
Fig. 5. Spermatid, with remains of spindle and “polar nucleus”. 
Attention has been drawn to the importance of these facts by Granio- 
Tos (Anat. Anz., Bd. 26, 1905, p. 369), and the work here described 
confirms his pace tee that in the spermatogenesis there is no reduc- 
tion in the ordinary sense, and that the failure of the first maturation 
division is due to the fact that the primitive germ cells of the drone 
contain the reduced number of chromosomes, as might be supposed 
if the drone is produced from an unfertilized egg. 
In divisions of the young ovarian cells of the queen larva, I find 
about 16 chromosomes, but in my material from drone larvae all the 
cells are already in the growth period, and I have been unable to find 
a mitosis, so that I cannot at present compare the premaiotic mitoses 
of male and female. 
Zoological Laboratory, Cambridge, September 1906. 
