Studies have shown partial to complete root coverage of denuded root surfaces with the use of thick free gingival autografts (FGGs) or subepithelial connective tissue autografts (CTGs). The purpose of this study was to determine which technique would result in more predictable root coverage of Miller Class I and II marginal tissue recession defects. Paired defects in 10 patients were randomly selected for treatment with either the FGG or the CTG. With stents as reference points, soft tissue recession was measured with a calibrated probe presurgically and 3 and 6 months postsurgically. No significant differences between paired sites in presurgical defect dimensions were found. One patient was dropped from the study for noncompliance with postoperative instructions. The mean percentage of root coverage for the CTG 3 and 6 months postsurgery for the remaining 9 patients was 78% and 80%, respectively. The mean percentage of root coverage for the FGG was 43% at both periods. The difference in root coverage between the 2 techniques was significant (P < 0.03). Complete root coverage was gained in 5 of 9 CTGs but only in one of 9 FGGs. Both techniques resulted in a significant improvement in keratinized tissue and probing attachment level, with most of the changes having occurred during the first three months postoperatively. Results suggest that the CTG may provide a greater percentage of root coverage than the FGG and that both techniques will effectively increase the width of keratinized tissue.