A South Texas survey conducted in 1975 investigated the reading performance of Mexican American students enrolled in a bilingual program to determine whether or not students achieved significantly different reading scores on parallel Spanish and English versions of an appropriate test (Guidance Testing Associates Inter-American Test of Reading). Designed especially for the control group so that the socio-cultural and dialectical sources of bias in either language version should be negligible, the test was given in April and May to a selection of students in grades three, five, and eight in order to determine whether the fit, or lack thereof, between English and Spanish scores was attenuated across grade levels. Mean scores for all students were found to be significantly higher on the English than the Spanish subtest. These differences were interpreted as casting considerable doubt on the commonly held assumption that Spanish language tests are "ipso facto" superior to English language tests for Mexican American, Spanish-surnamed bilingual pupils. Implications were that a re-examination of bilingual programs with respect to their philosophical premises and pedagogical practices appears warranted, and that additional empirical studies are needed in the area of Spanish-English language dynamics with respect to student achievement outcomes. (JD)