The purpose of this study is to evaluate ultrasonic suture welding of monofilament suture in an animal model of rotator cuff repair with biomechanical and histologic analyses. We randomly assigned 46 shoulders in 23 rabbits to 1 of 3 treatment groups: sham-operated (n = 15), knotted (n = 15), and welded (n = 16). Supraspinatus defects were surgically created and acutely repaired with suture anchors loaded with either No. 2-0 Ethibond for knotted group or No. 2-0 nylon for welded shoulders. Eighteen weeks postoperatively, all animals were killed, and the shoulders underwent either biomechanical testing or histologic analysis. The maximum stress of the sham-operated group (20.6 N/mm2) was significantly greater than that of both the knotted (10.2 N/mm2) and welded (8.3 N/mm2) groups (P < .05), but no differences were observed between the knotted and welded groups. Although some histologic changes were noted, none was considered to be significant to distinguish either group.