Designed a metacognitive intervention program to remediate the failures of 42 4th-grade boys in using metacognitive skills to aid their reading comprehension. The program consisted of 2 components: story grammar training, designed to increase comprehension monitoring; and attribution training, designed to increase awareness of effort in efficient reading. Ss were assigned to 3 groups: 1 group received both components and the other 2 groups each received one component alone. 14 skilled 4th-grade male readers served as a contrast group. Maintenance was assessed through free and probed recall; generalization was assessed through a metareading test and an error detection and correction task. Results indicate that strategy training produced dramatic gains in comprehension. Only Ss receiving attribution training alone showed poorer performance than skilled readers. Partial support was obtained for generalization on the metareading assessment. It is concluded that strategy training improved poor readers' comprehension by providing them with metacognitive skills.