Measures of divergent thinking were administered to 34 7and 8-year-old boys in a permissive testing context and without time limits. Individual differences were reliable across tests and independent of IQ, in contrast to typical findings in studies employing an ability-testing atmosphere. In 87 kindergarten children, similar results were obtained for tests with semantic content but not for one whose content was figural. However, there was intertask consistency in test involvement despite this partial inconsistency in response totals, suggesting that a unitary creativity dimension is present in kindergarten children but is not measured by the figural test at this age. The response style "reflection-impulsivity" was unrelated to creativity, although hypothesized antecedents of this dimension are identical to those which have been proposed for various creativity subgroups. A measure of artistic preference also failed to relate significantly to creativity scores.