a low
death-rate in succeeding years, then there will be grounds for believing
that natural selection is really cutting off the weaker and allowing the
stronger to survive.

E. C. Snow[56] analyzed the infant mortality registration of parts of
England and Prussia to determine whether any such conclusion was
justified. His investigation met with many difficulties, and his results
are not as clear-cut as could be desired, but he felt justified in
concluding from them that "the general result can not be questioned.
Natural selection, in the form of a selective death-rate, is strongly
operative in man in the early years of life. We assert with great
confidence that a high mortality in infancy (the first two years of
life) is followed by a correspondingly low mortality in childhood, and
vice-versa.... Our work has led us to the conclusion that infant
mortality _does_ effect a 'weeding out' of the unfit."

"Unfitness" in this connection must not be interpreted too narrowly. A
child may be "unfit" to survive in its environment, merely because its
parents are ignorant and careless. Such unfitness makes more probable an
inheritance of low intelligence.

Evidence of natural selection was gathered by Karl Pearson from another
source and published in 1912. He dealt with material analogous to that
of Dr. Snow and showed "that when allowance was made for change of
environment in the course of 50 years, a very high association existed
between the deaths in the first year of life and the deaths in childhood
(1 to 5 years). This association was such that if the infantile
death-rate _increased_ by 10% the child death rate _decreased_ by 5.3%
in males, while in females the _fall_ in the child death-rate was almost
1% for every 1% _rise_ in the infantile death-rate."

To put the matter in the form of a truism, part of the children born in
any district in a given year are doomed by heredity to a premature
death; and if they die in one year they will not be alive to die in some
succeeding year.

Lately a new mathematical method, which is termed the Variate Difference
Correlation method, has been invented and gives more accurate results,
in such an investigation as that of natural selection, than any hitherto
used. With this instrument Professor Pearson and Miss Elderton have
confirmed the previous work. Applying it to the registered births in
England and Wales between 1850 and 1912, and the deaths during the first
five years of life in