The aim of this research was to learn whether silent reading fluency was the predictor of reading comprehension and which variable including silent or oral reading fluency was the best predictor of reading comprehension. With this aim, the study used correlational design and the study sample consisted of total 100 fifth-grade Turkish elementary students studying in two elementary schools. The schools were located in low socioeconomic status and the students` families had low socioeconomic level. The informed consent obtained from all participants before the study began. For silent and oral reading fluency assessments, an appropriate grade level text was chosen and the students` silent and oral reading fluency were measured through one-onone sessions in the suitable place in the elementary schools provided by the school principals. After this process, the reading comprehension test related to the grade level text read was administered to all students. The data obtained from testing process were analyzed and the findings were presented in respond to research questions. The research findings showed that silent and oral reading fluency were moderately related to each other and had significant correlations with reading comprehension. They both explained together 23% of the variance in reading comprehension and silent reading fluency had more significant contribution to prediction of reading comprehension than oral reading fluency. Additionally, the total variance of reading comprehension explained by silent and oral reading fluency varied according to gender of the students.