G. S. Graham-Smith 
463 
founded on casual observations, that flies die off in the autumn owing 
to cold and the effects of empusa disease comprise all the information 
that is to be found in the literature on the subject. 
The only method of arriving at any satisfactory information regarding 
this subject is to keep flies, confined under conditions as natural as 
possible, and record the deaths which occur daily, and all the available 
meteorological data. During the summer of 1915 the writer obtained 
much information from the records relating to the blow-fly breeding 
experiment just described, and further information was obtained from 
observations on blow-flies which were kept in different situations during 
the winter 1915-16. 
Before any final conclusions can be drawn such experiments will 
have to be repeated in order to ascertain the effects of the weather in 
years with differing meteorological conditions, but these observations 
appear to indicate that blow-flies emerging from pupae which have 
passed the winter in the earth are more hardy than the flies of the 
subsequent summer generations, and that certain individuals of the 
late autumn and winter generations can endure exposure to cold to an 
astonishing extent. 
Blow-flies emerging in the spring from winter pupae. 
From Chart 2 it may be seen that most of the blow-flies of the 
parent generation survived the very heavy rains (1-47 inches) of May 12 
and 13, and that they gradually died off without a great mortality at 
any special period, the mean duration of life being apparently about 
30 days. The spell of cold weather from May 28 to June 1 did not 
perceptibly increase the death-rate and most of these flies probably 
died of old age. Throughout the period covered by the flies of this 
generation the weather was on the whole fine and warm, and therefore 
it would be unsafe, without observations in other less favourable years, 
to assert definitely that the flies which emerge from winter pupae are 
always hardier than those of later generations. 
Blow-flies belonging to the summer generations. 
It will be seen from Chart 2 that after most periods of emergence 
a large proportion of the flies died within a few days. In the attempt 
to find some of the factors which contributed to this mortality charts 
were constructed showing the daily percentage of deaths, the total rain¬ 
fall, the hours during which rain fell, the maximum temperatures in the 
