Most analyses of oral reading fluency (ORF) are based on L1 reading, and the norms that have been developed in English are based on first language reading data. This is problematic for developing countries where many children are learning in English as a second language. The aim of the present study is to model the relationship between English reading fluency and comprehension among rural English-second-language learners (ESL) in South Africa. We use data collected in 2013 by the National Education and Evaluation Development Unit in South Africa. This survey tested 4697 Grade 5 students from 214 schools across rural areas in South Africa. A sub-sample--1772 students--were selected for an ORF test. For these students there exist data on both reading comprehension and reading fluency. Although a number of studies have analyzed the relationship between fluency and comprehension, none have been conducted on a large-scale for ESL learners in a developing country such as South Africa. The present research contributes to the literature by analysing the size, significance and uniformity of this relationship for ESL learners in South Africa. Preliminary findings indicate a threshold at 70 words-read-correct-per-minute which is lower than the typically used threshold of 90 words-read-correct-per-minute of English first language readers.