This study sought to develop an instrument for measuring teacher knowledge of reading at the secondary school level. Multiple-chioce items were generated from a content outline derived from a review of the literature and were submitted to a panel of experts who categorized the items, answered them, and rated each one's importance. After revisions based on results from the preliminary version, the final version of the test consisted of 45 items. This was administered to 124 teachers and 64 students in secondary education. Reliability based on the KR-20 formula was 0.76. The correlations between four subtests (general knowledge, reading skills, instructional strategies, and measurement and evaluation) were low, suggesting that the subtests measure different abilities; however, their reliability coefficients were not high enough to warrant use as independent subtests. Comparisons between teachers and students indicated that teachers achieved significantly higher scores and that people who had taken a reading course scored higher than those who had not. It was concluded that the test is a valid and reliable instrument for measuring preservice and inservice teacher knowledge of reading. (Author/AA)