This longitudinal study investigated the facilitative effects of four different program models on both acquisition of English and maintenance of Spanish by native Spanish-speaking kindergarten children. The four program models or treatments were: (1) traditional or regular kindergarten, taught in English with no formal English as Second Language (ESL) training, (2) traditional, with ESL daily, (3) bilingual kindergarten, taught in English and Spanish with no ESL, and (4) bilingual kindergarten with ESL daily. Criterion language measures were administered on a pre and post basis, in both Spanish and English by peer testers. These measures included oral comprehension, vocabulary for home, school, and street, and oral production. An instrument to measure communicative competence, designed by the investigator, was also given. The two groups (3 and 4) taught bilingually were further divided on the basis of balanced or unbalanced input of Spanish and English by the teaching staff. Using multivariate analysis of variance on gain scores, it appears that treatment 3B, bilingual with balanced language, and no ESL produced significantly greater gains in communicative competence in Spanish as well as English, though to a lesser extent, than did the other treatments. This study indicated that the balanced bilingual treatment with no formal ESL training enhanced communicative competence in Spanish and English while facilitating comprehension of English, as well as vocabulary in both Spanish and English. (Author/AM)