The purpose of this study is to examine whether subject-specific and/or across subject metacognition relates positively with academic achievement and determine if partialling IQ from metacognition-achievement relations should influence the magnitude of those correlations. The subjects of the study were the first grade from 4 Junior-high schools located in kwang-Ju city. The total number of the subjects was 414. Each of subject-specific metacognition and across subject metacognition was measured by the Comprehension Monitoring Test made by the present researcher. A concurrent validity between IQ and metacognition was moderate(r＝.59), and a test-retest reliability was acceptable(r＝.83). Academic achievements were assessed by the national sham examination prepared by Dong-A publishers co.. IQ was measured by standarized intelligence scale (Chung Bum-Mo, Kim Ho-Kwon). The datum of the study were computed with Pearson r. and t - value. The results revealed that subject-specific metacognition related positively with academic achievement and across-subject metacognition also related positively metacognition and academic achievements were reduced to near a half of the magnitude of the correlation, when IQ was statistically taken from the relation, however the remained correlations were still significant statistically. It was found that the high group in metacognition and that high in IQ did not show difference in academic achievement, even though difference IQ 10 points between high metacognition and the high IQ group. It was concluded that subject-specific and across-subject metacognition should be a promising cognitive variable for effecting academic achievement, even though IQ is highly correlated with these metacognition.