Effective early language and literacy instruction to remediate language deficits and to prevent problems in learning to read is an important area for intervention research. Children with early language deficits who are growing up in poverty are dually at risk. Early deficits in language development predict both continued delays in language development and problems in acquisition of reading-related skills (Dickinson, McCabe, Anastasopoulos, Peisner-Feinberg & Poe, 2003; Snow, Burns & Griffen, 1998; Spira, Storch & Fischel, 2005; Storch & Whitehurst, 2002; Walker, Greenwood, Hart & Carta, 1994). Early emergent problem behavior, particularly in children with low language skills, also predicts difficulties in reading, academic performance and peer relationships (Hester & Kaiser, l998). Without effective early intervention to teach language and emergent literacy skills, many of these children will require intensive, long-term special education. This paper reports findings from a large randomized field trial examining the effects of three variations of early literacy curricula implemented in Head Start classrooms on children with low language and matched children with average language skills. This project involved the comparison of three conditions: "Opening the World of Learning" ("OWL"), "OWL" combined with "Enhanced Milieu Teaching" ("EMT") for low language children, and an enhanced version of "Creative Curriculum" ("CC"), the existing literacy program used by the Head Start center. Effects of the preschool curricula on children's end of preschool, end of kindergarten, and end of first grade outcomes were examined. This intervention took place in Head Start preschool classrooms in Birmingham, Alabama. Overall, there were few systematic differences in language and literacy outcomes across time for low language and matched typical language children. Outcomes varied by child language status (low vs. matched language) and assessment time (end of preschool vs. end of K vs. end of 1st). The results were not those anticipated. Four potential explanations are being considered and examined in follow up analyses of predictors of child outcomes, teacher implementation fidelity, and "active ingredients" (specific teacher behaviors associated with positive language and literacy outcomes). The authors have not yet completed the full set of moderator analysis and they anticipate these analyses will aid in understanding the treatment outcomes. (Contains 1 table.