The study reports the results of the analysis of Comprehension-Production Test in the kindergarten of two bilingual schools (L-1 English/L-2 French, L-1 Spanish/L-2 English) in the San Francisco Bay area. The tests used covered 14 formal contrasts of English. Parallel tests for Spanish and French were developed by translating the English test. Tests were administered in L-1 once at the beginning of the school year and in L-2 three times at regular intervals during the year. Various types of analysis are presented: (1) Correlations between L-1 and L-2 scores; (2) Correlations between Comprehension and Production scores; (3) Significant rank order of difficulty of the 14 contrasts in L-1 and L-2; (4) Determination of (a) significant differences between Comprehension and Production scores, (b) significant gains made in L-2 during the year, (c) significant differences between L-1 scores at the two schools, (d) significant difference between L-2 scores at the two schools. The main conclusions advanced are that (1) in spite of an overall similarity between factors accounting for difficulty in first and second language acquisition interference from L-1 cannot be ruled out as playing a role in early childhood L-2 acquisition. (2) Comparative studies of second language acquisition can furnish an empirical basis for psycholinguistic and perhaps also linguistic theory. (Author)