In light of the failure of most attempts to replicate the original demonstration of potent teacher expectancy effects by Rosenthal and Jacobson (1968), the following study was performed. Within each of ten first grade classes for educationally deprived students, one half of the pupils were pretested with the Cognitive Abilities Test (CAT) and were then randomly assigned to an experimental or control group. Three weeks after the pretesting, teachers received reports stating that pupils in the experimental group had hidden academic potential and might show academic gains during the school year. Pupils were retested with the CAT six months later. Although the experimental group tended to make greater gain scores (as measured by IQ and raw scores) on the CAT (p < .10), no significant differences were found between the two groups on the following measures: reading grade, arithmetic grade, social skills, and reading level. The results of the study were interpreted as not supporting the Rosenthal and Jacobson findings. It was recommended that further exploration of teacher expectancies include, (a) relating expectancies to the overt interaction between students and teacher and (b) investigating teacher expectancies that result from more credible naturalistic inputs.