a fantastic emphasis upon sexual selection. That is, much
progress was accounted for through the selection by the females of the
superior males. This, as a prime factor in evolution, has since been
almost "wholly discredited" (Kellogg's phrase) by the careful
experiments of Mayer, Soule, Douglass, Duerigen, Morgan and others. The
belief in sexual selection involved a long string of corollaries, of
which biology has about purged itself, but they hang on tenaciously in
sociological and popular literature. For instance, Ward believed in the
tendency of opposites to mate (tall men with short women, blonds with
brunettes, etc.), although Karl Pearson had published a statistical
refutation in his _Grammar of Science_, which had run through two
editions when the _Pure Sociology_ appeared. The greater variability of
males than females, another gynaecocentric dogma had also been attacked
by Pearson on statistical evidence in 1897 (in the well-known essay on
Variation in Man and Woman, in _Chances of Death_) and has become
increasingly unacceptable through the researches of Mrs.
Hollingworth[6,7,8]. The idea of a vanished age of mother-rule in human
society, so essential to the complete theory, has long since been
modified by anthropologists.

De Vries' experiments showed that a moderately simple fact practically
makes all these complicated theories unnecessary. No two living things
are exactly alike--that is, all living matter is more or less variable.
Some variations are more fortunate than others, and these variants are
the ones which survive--the ones best adapted to their environment.
Given this fact of the constant variation of living matter, natural
selection (i.e., survival of the fittest and elimination of the unfit)
is the mechanism of evolution or progress which best accounts for the
observed facts. Such variation is called "chance variation," not because
it takes place by "chance" in the properly accepted sense of the term,
but because it is so tremendously varied--is evidently due to such
complicated and little-understood circumstances--that it can best be
studied mathematically, using statistical applications of the "theory of
probabilities."

The fine-spun, elaborate theories about sex, so current twenty years
ago, have fallen into almost complete desuetude among scientists. With
the discovery of the place of the chromosomes in inheritance, biologists
began to give their almost undivided attention to a rigid labor