view that the infant's hand movements in reaching or
grasping are the best index of the kind and intensity of
its sensory experiences," finds that the colors range
themselves in order of attractiveness, blue, white, red,
green, brown. Further corrections lay more emphasis upon
the white. Yellow was not included in the experiments.
Cohn's results, which show a relative dislike of yellow,
are contradicted by other observers, notably Major and
Baker,<2> and (unpublished) experiments of my own, including
the aesthetic preferences of seven or eight different sets
of students at Radcliffe and Wellesley colleges. Experiments
of this kind are particularly difficult, inasmuch as the
material, usually colored paper, varies considerably from
the spectral color, and differences in saturation, hue, and
brightness make great differences in the results, while the
feeling-tone of association, individual or racial, very
often intrudes. But other things being equal, the bright,
the clear, the saturated color is relatively more pleasing,
and white, red, and yellow seem especially preferred.

<1> _Mental Development in the Child and the Race_, 1895,
pp. 39, 50, ff.
<2> E. S. Baker, _Univ. of Toronto Studies, Psychol. Series_,
No. 4; J. Cohn, _Philos. Studien_, vol. X; Major, _Amer. Jour.
of Psychol._, vol. vii.

Now, according to the Hering theory of color, white, red, and
yellow are the so-called "dissimilating" colors in the three
pairs, white-black, red-green, and yellow-blue, corresponding
to three hypothetical visual substances in the retina. These