The purpose of this study was to examine which of three instructional modalities was more effective in enhancing the ability of nonnative English speaking children to read during the first grade. In this study, sixty-three first-grade children were randomly selected from four first grade classes from two primary schools in a university town in a nonnative English-speaking country during the second term of the academic year. The group was randomly placed in groups of 21. The first two served as the experimental groups (identity, and blending/manipulation) and the third group served as the language experience (control). Both the identity and blending/manipulation groups were instructed using different instructional modalities. The third group was not exposed to any direct instruction. The results of the study indicate that after the eight weeks of instruction, the children in the identity group made modest gains in posttest scores over those in the blending/manipulation group on two of the measures--the test of phoneme cue reading (TPCR) and the test of vocabulary (PPVT). While the identity group made statistically significant gains over the blending/manipulation group, it was observed that the implicit associations between orthography and phonology that the subjects made that may have aided them to gradually construct an orthographic-phonologic to decode words. Educators of limited English proficiency learners need to adopt instructional modalities that allow for the interplay of models which can have more noticeable effects on the reading programs in the early elementary years. (Contains 8 tables.)