Localized cutaneous infantile hemangioma acts like a tissue expander. This rapidly growing tumor can destroy elastic fibers or cause ulceration resulting in telangiectases, cutaneous laxity, scarring, and fibrofatty residuum. Although surgeons may dispute indications and timing, most would agree that the scar of resection should be minimized. For this reason, circular excision and purse-string closure is particularly applicable for hemangioma at any stage of its evolution. The purposes of this study were to: (1) analyze the results of circular excision/purse-string closure in all three phases of the life cycle of hemangioma; (2) quantify dimensional changes after resection; and (3) compare the scars after theoretical single-stage lenticular excision with those after staged circular excision/purse-string closure. The authors retrospectively analyzed their experience in 25 children with localized hemangioma who underwent circular excision/purse-string closure from 1997 to 2000. Each hemangioma was measured preoperatively and the scars were measured at most recent follow-up (minimum, 6 months). Preoperative and postoperative dimensions were analyzed using SPSS statistical software. The study included 22 girls and three boys, with an average time to follow-up evaluation of 13.1 months. Twenty-one lesions were in the face and scalp, and five were in the extremity. Five tumors were resected in the proliferative phase (either because of ulceration, bleeding, or visual complications) and 21 were excised in the involuting or involuted phase. Six patients had a second-stage procedure: three had another circular excision and three had later lenticular excision. After single circular excision/purse-string closure, the mean long-axial diameter (length) decreased by 45 percent, the mean short-axial width (width) decreased by 73 percent, and the mean scar area was only 15 percent of the original area. All these differences were statistically significant (p = 0.001). The average width/length ratio decreased by 50 percent, indicating a tendency for scars to linearize. There was no difference in linearization for the three phases of hemangioma (p > 0.05); extremity scars became more linear that those on the face (p = 0.01). The authors devised a formula for scar length after lenticular excision/linear closure, assuming a conventional excisional ratio of 3:1 for a circular lesion. Using this equation, the authors predicted that mean scar length after circular excision, followed by lenticular excision, would be 72 percent shorter than the calculated scar that would result from conventional lenticular excision. In three patients who underwent this two-stage approach, the resultant scar was 69 percent shorter. Circular excision of hemangioma and purse-string closure reduces both the longitudinal and transverse dimensions and converts a large circular lesion into a small ellipsoid scar. If subsequent revision to a linear scar is desirable, its length will be the same or slightly less than the diameter of the original lesion. No other excision and closure technique results in a smaller scar. Another advantage of this method is minimal distortion of surrounding structures.