Large scale longitudinal research (Morgan, Farkas, & Wu, 2009) and a meta-analysis (Duncan et al., 2007) have found that early mathematics achievement is a strong predictor of later mathematics achievement. In fact, end of Kindergarten and end of grade 1 mathematics achievement on ECLS-K and similar mathematics proficiency measures tends to be a stronger predictor than early measures of reading or reading-related skills such as phonemic awareness. Yet, despite increasing interest, there is little research on the effectiveness of recommended best practices in Response to Intervention Models (RtI) in mathematics (Gersten et al. 2009). This study (Rolfhus, et al. 2012) replicates the Fuchs et al. (2005) study as the first large-scale effectiveness trial (one intended to approximate real-world implementation) of "Number Rockets". While the Fuchs et al. study used interventionists experienced with at-risk students, the current study employed interventionists with a range of experience who were selected from the local community. While the Fuchs et al. study provided interventionists with substantial monitoring and support, the current study provided professional development and a support program similar to those provided by publishers of curriculum products (Agodini et al. 2009). Different measures were used to identify at-risk students, and measure outcomes. Finally, the district in the Fuchs et al. study used just one curriculum; each of the four urban districts in the current study used a different one, which may have provided a more heterogeneous instructional context. This study is the first effectiveness evaluation (and first replication) of "Number Rockets" and builds on the positive findings of the Fuchs et al. (2005) efficacy study. Tables are appended.