A study was conducted to test whether three, four, and five-year-old children would be better able to use either static or dynamic properties for grouping objects, and whether performance under these conditions would be better than when no property was given. One of the two study tasks, the free sort, also used by Rosch et al. (1976), asked children to select from pictures those that showed the same kind of thing. In the other task, the property sort, the children were shown sets of pictures and asked to select those that had some specific property in common, which was either static (something the object had) or dynamic (something the object did or that could be done to it). Sorting trials were also divided into superordinate and basic level sorts to see if any difference between property types might vary as a function of the level of abstraction. At least at the superordinate level of abstraction, the younger subjects were better able to categorize pictures when they were given a property by which to group them than when they were not. Of the two kinds of properties, dynamic seemed to aid performance more than static. This superiority of dynamic properties was present only in the younger subjects. (SW)