The objective of this study was to determine the relative motion of a quadruple hamstring graft within the femoral bone tunnel (graft-tunnel motion) under tensile loading. Six graft constructs were prepared from the semitendinosus and gracilis tendons of human cadavers and were fixed with a titanium button and polyester tape within a bone tunnel in a cadaveric femur. Three different lengths of polyester tape (15, 25, and 35 mm loops) were evaluated. The femur was held stationary and uniaxial tensile loads were applied to the distal end of the graft using a materials testing machine. Each construct was subjected to loading for ten cycles with upper limits of 50 N, 100 N, 200 N and 300 N. Graft-tunnel motion was then determined using the distances between reflective tape markers placed on the hamstring graft and at the entrance to the femoral bone tunnel, which were tracked with a high-resolution video system. Graft-tunnel motion was found to range from 0.7 +/- 0.2 mm to 3.3 +/- 0.2 mm, and significant increases in graft-tunnel motion were observed with increasing tensile loads (P < 0.05). Shorter tape length (15 mm) resulted in significantly less motion when compared to longer tape length (35 mm) (P < 0.05). We conclude that graft-tunnel motion is significant and should be considered when using this fixation technique. Early stress on the graft, as seen in postoperative rehabilitation exercises and athletic activities, may cause large graft-tunnel motion before graft incorporation is complete. A shorter distance between the tendon tissue and the titanium button is recommended to minimize the amount of graft-tunnel motion. Alternative fixation materials to polyester tape, or different fixation techniques, need to be developed such that graft-tunnel motion can be reduced. Further studies are needed to evaluate the effect of graft-tunnel motion on graft incorporation in the bone tunnel.