An analysis of the benefits of submuscular versus subcutaneous implantation was made on mastectomy patients. Ninety-one breast were reconstructed following mastectomy. In 30 breasts, the implants were placed subcutaneously; in 19, subpectorally, and in 42, beneath both the pectoralis and the serratus. The follow-up averaged 2 to 3 years, and recent cases included postoperative tonometry measurements to quantitate the degree of capsular contraction. In addition, 12 cadaver dissections were done to delineate muscle insertion and origins. Results indicate that (1) submuscular implants are clearly superior to subcutaneous ones; (2) subpectoral implantation requires complete detachment of the muscular origin from the ribs; (3) subserratus implantation provides extra muscular coverage, but dissection is more difficult owing to its firm rib attachment; and (4) the subserratus technique provided the lowest incidence of capsular contracture, although the breast was slightly flatter initially, but improved with time.