UTAH ACADEMY OF SCIENCES 67 
on the assumption that a test involving acts of dexterity 
would be superior to a test of skill or endurance, since 
such a test would ultimately be given in most cases to 
children between the ages of four and eight years. 
The desired test, in order to be valid for the purpose 
in mind, must be relatively accurate when applied to 
individual cases. 
The test must be simple, not alone of comprehension 
by the subject, but in construction and application; it 
must be designed to meet the needs of school administra- 
tors, teachers, and other educational officers. 
It must choose as a means of diagnosis some type of 
skill which, when tested in young children will reveal 
and express the relative potential dexterity of the two 
hands. 
The ability or dexterity tested must closely resemble 
the principal dexterities called forth in school work and 
in practical life; at the same time the two must not be 
identical. That is, one could not adequately detect native 
handedness by asking a first-grade child to write his 
name, first with the right hand and then with the left, 
and thus determine the superior hand by the superior 
product. A valid test must obviate the practice element 
as much as possible. 
With these basic considerations in mind the writer 
decided on the general plan of trying out certain of the 
existing and most suggestive tests in order, first, to elim- 
inate the least promising, and, secondly, either to perfect 
the remaining ones or to glean suggestions for the con- 
struction of a new test. The essence of the general 
method was to apply certain tests to a large number of 
children of different school ages, whose handedness was 
already known quite accurately, and to determine the 
validity of the tests by the degree or extent to which its 
results corresponded or correlated with the known facts 
of the children’s handedness. 
In all, eight mechanical tests were evaluated exper- 
imentally on a group of 465 school children. Six of 
these tests had already been suggested and used in 
attempts at diagnosis; two of these tests embodied the 
writer’s own ideas and conclusions. The test which I 
have here is the one which has yielded the most accurate 
