Much information of interest to bilingual education policymakers has emerged from basic research on language learning, bilingualism, and cognition. Among the conclusions drawn from it are the following: (1) bilingual education is superior to submersion education in many contexts;(2) one major bilingual education goal should be development of the full repertoire of English linguistic skills, in preparation for mainstreaming; (3) time spent learning the native language in bilingual programs is not time lost in developing English skills; (4) children can become fluent in a second language without losing the first language, and maintenance of the first language does not retard second language development; (5) there is no cognitive cost, and there may be advantages, in the development of bilingualism in children; (6) bilingual education programs should have the flexibility of adjusting to the large individual and cultural differences among children, including need for lengthy bilingual education; (7) educators should understand that for primary grade children second language learning may be a slow process and that older learners can learn languages quickly and efficiently; (8) reading should be taught in the native language; and (9) actions upgrading the status of limited-English-speaking children's culture and language contribute to their opportunities for friendship with native English-speaking children. (MSE)