A sample of 217 students (from grades 4 to 8) who were enrolled in an elementary school science-oriented gifted and talented program were compared to 91 mainstreamed subjects in grades 7 through 10. Assessement instruments included: a battery of Piagetian measures designed to assess combinatorial reasoning, probabilistic reasoning, and the ability to isolate and control variables; the Propositional Logic Test; and a variation of the four-card hypothesis testing task. Results show that the gifted and talented sample was accelerated over the comparison group by two to three grade levels, suggesting that the former students are truly gifted in regard to very basic reasoning skills. The criteria for inclusions in gifted and talented programs vary from case to case, and are rarely well defined. Standardized aptitude tests may be appropriate for an academically-oriented acceleration program, but are probably not for local pullout enrichment programs. A more general battery of reasoning tasks, such as those used in this study, should be included in the selection of students for such gifted and talented programs. (Author/JN)