Improvement in the strength of the transpedicular screw-vertebra interface by increasing the depth of screw insertion may provide improved performance of spinal implants using such screws. Within human cadaveric vertebrae, we measured the failure strength of Vermont Spinal Fixator (VSF) screws under flexion or torsion loads and of Schanz screws under pull-out loads (along the screw axis). Comparisons between opposite pedicles of vertebral specimens were made at 50 vs. 80% and 80 vs. 100% of maximum available insertion depth. Mean failure strength of VSF screws at 50% depth was 75-77% (depending upon load type) of that at 80% depth; strength for screws at 100% ("to-cortex") depth was 124-154% of that at 80%. Reanalysis of the data from Lavaste shows, contrary to his conclusion, a 26% increase in strength from a 5-mm increase in screw depth of insertion. All these differences were significant (p less than 0.05) by the matched-pairs t test. Benefit from the increased strength of deeper screw placement must be balanced against possible increased operative risk. A "near-approach" x-ray view is suggested here to decrease that presumed operative risk.