INTELLIGENCE TEST constructors are usually guided by a combination of both subjective and empirical criteria in their selection of test items. Intelli gence is frequently conceived as general ability to learn and to solve prob lems, or more precisely as abilities demanded in solution of problems which require comprehension and use of symbols (3:372). Test constructionists have included measures of such mental processes as making associations, log ical classifications, discriminations, seeing analogies, drawing conclusions, organizing elements into meaningful patterns, and achieving insights (2:3). Ideas and are often symbolized by words, numbers, pictures, diagrams, equations, and formulas (3:372). Since it is generally deemed desirable to have tests which measure capacity for learning and problem-solv ing rather than training and experience, intelligence test makers usually at tempt to control these factors by selecting items on which subjects for whom they are intended have had ample, unrestricted and approximately equal opportunities for learning (1). Another common assumption inte 1 ligence is that it is a product of development, of mental maturation, and asa consequence level of a child's intelligence increases gradually from birth to maturity. Thus, another criterion often employed in selection of tes t items is relatively steep increments in percentages of children passing items at successive ages. Finally, among these commonly accepted criteria, such abstract, culture-fair, and developmental items are expected to correlate significantly with school achievement. Not all inventors of tests, however,ad here to all of these criteria; and because they don't, comparisons of their dif ferently constructed tests permit testing validity of these princ iples of test construction. The Davis-Eells Test of General Intelligence or Problem Solving Ability (2) and Kuhlmann-Finch Tests (5) are, for example, simi lar in their problem-solving requirements and in their attempts to achieve cul ture-fairness. Both tests, directly or indirectly, require comprehension of verbal symbols, although in neither is reading emphasized. Both require sub jects to discriminate, to achieve inductive generalizations, to comprehend analogies, and to draw and apply conclusions. The Davis-Eells test attempts to achieve culture-fairness by using situations, pictures, and oral language about equally familiar to all American urban cultural groups (2:8). The Kuhlmann-Finch Tests attempt to eliminate possible cultural bias by using simple vocabulary and pictures which express relationships in terms of form, position, size, and number (5:13). The reported reliaoilities at elemen tary level are also similar for two tests (2, 5). But these tests differ inap plication of criterion of age discrimination. In Kuhlmann-Finch Tests, the prime requisite for every item's inclu