Naming tasks were used in two studies to test for conceptual organization in young children by comparing the latencies for naming objects in primed vs. non-primed conditions. In the primed condition, a taxonomic category was primed by prior activation of a coordinate member of the same category. In Study 1, 54 kindergarten children were randomly assigned to a primed or non-primed condition. All subjects had to name as quickly as possible each of a series of five toy objects. Each of the first four was from a distinctly different semantic category. In the primed condition the fifth object was a member of the same taxonomic category as the fourth object. In the non-primed condition the fifth object was of a different semantic category from all the preceding objects. Results indicated that the mean latency was greater in the primed than in the non-primed condition. Study 2 attempted to replicate Study 1 using a within subjects design and a stronger priming manipulation to increase the strength of the effect. Subjects were 16 preschool and 2 kindergarten children. In the primed condition the fifth object in each series was from the same semantic category as the preceding two objects. All objects were from different categories in the non-primed condition. In both conditions a sixth object was added that was always from a different category than the fifth object. Again, the results indicated greater latencies for the primed condition. Three explanations of this hindrance effect are discussed: (1) a false set explanation, (2) a categorization decrement effect; and (3) a naming interference explanation. (Author/SB)