larve of Smerinthus popult. 455 
black and white balls, two of the same colour will be 
drawn together twice as frequently as those of different 
colours. In future observations of a sluggish species 
like Smerinthus populi, it would be desirable to track the 
moth as it lays little groups of eggs on different leaves, 
and to rear such groups separately. We must not 
altogether put aside the remote possibility that the 
parent may possess some power of controlling the sex 
of her offspring. 
Six moths emerged in the hot summer of 1893 ; three 
of these were males and three females. The eggs from 
which these individuals were hatched were scattered 
through the successional series, and were not peculiar 
to any part of it, and the same appears to be true of the 
deaths of larve and deformity of pupe. 
Failures to hatch, however, are confined to the end of 
the series, unless some of the ‘‘missing”’ larve are to 
be explained in this way. ‘This tendency is especially 
clearly seen on the last two days, and yet the very last 
ego laid contained a well-formed larva which was unable 
to hatch. 
Even the small number of larval deaths which took 
place are chiefly to be accounted for by the conditions of 
the experiment. Thus, considering only the eggs laid 
during May (and omitting the later eggs because of the 
frequent failure to hatch), we find out of 40 larve reared 
in separate sleeves (23 to 34 on May 27th, and the 28 on 
May 28th) only two deaths occurred, not including the 
manifestly accidental death of the fourth larva on May 
28th. On the other hand, out of 95 larve reared in 
groups of from 3 to 8 (usually 7) in number, 11 larve 
died or were missing. From this it may be inferred that 
larvee are in some way injured by being reared in groups 
with a much greater relative proximity than in nature. 
I have also found.this to be true of S. ocellatus ; for, 
when groups of more than 10 or 12 small larve were 
enclosed in sleeves of moderate size, it was always found 
that the numbers became reduced to a maximum of 
about a dozen, and often much lower; whereas groups 
of smaller numbers were frequently maintained without 
any deaths. 
The list as a whole strongly impresses upon us the 
comparative rarity of death from internal causes (in- 
cluding disease), and the overwhelming importance of 
