recurrence, is obliged to admit that the hot weather exercises a
propelling influence on suicidal tendencies, and that the cold weather
on the other hand acts in an opposite direction[18].

[18] DISTRIBUTION OF SUICIDES IN LONDON BY MONTHS OF EQUAL
LENGTH PER 10,000, 1865-84:--

January, 732. July, 905.
February, 714. August, 891.
March, 840. September, 705.
April, 933. October, 772.
May, 1003. November, 726.
June, 1022. December, 697.

Dr. Ogle, vol. xlix., 117. _Statistical Society's Journal_.

The influence of temperature is, however, much less powerful on crime
than it is on suicide. It has the effect of raising by one third the
number of persons to whom life becomes an intolerable burden, but
according to the diagram in the Prison Commissioners' Reports the
highest increase in crime between summer and winter does not amount to
more than one twelfth. In other words, between six and eight per cent.
of the crime committed in this country in summer may with reasonable
certainty be attributed to the direct action of temperature. This is
a most important result and I should almost hesitate to state it if
it were supported by my investigations only. But this is far from
being the case. In an important paper contributed to the Revista di
Discipline Carcerarie for 1886, Dr. Marro, one of the most
distinguished students of crime in Italy, has arrived at similar
conclusions. He has shown that in the Italian prisons in the four