AIM: To compare four different atlantoaxial stabilization techniques. MATERIAL and METHODS: Eight human cervical spines (segments C0-C3) were tested in flexion/extension, lateral bending and axial rotation. Range of Motion (ROM) at a 1.5 Nm load was recorded. After native testing, the Harms (HARMS), pars screw (PARS), intralaminar screw (INTRA) and anterior transarticular screw (ATA) constructs were applied in a random order. RESULTS: FLEXION/EXTENSION: mean ROM (±SD) in native state was 15.9° (± 7.6°); HARMS 3.6° (± 2.0°); INTRA 5.5° (± 2.7°); PARS 2.8° (± 1.6°); ATA 3.7° (± 1.3°). A significant difference was found for all techniques compared to the native spine. LATERAL BENDING: ROM in native state was 3.2° (± 1.9°); HARMS 1.4° (± 0.4°); INTRA 2.5° (± 1.4°); PARS 1.3° (± 0.7°); ATA 0.9° (± 0.6°). There were no significant differences compared to native spine, although ATA and PARS showed a strong tendency. INTRA had a significantly higher ROM than ATA. AXIAL ROTATION: ROM in native state was 15.7° (± 6.6°); HARMS 1.5° (± 0.7); INTRA 2.7° (± 2.1°); PARS 1.7° (± 0.7); ATA 1.1° (± 0.3°). All instrumentation techniques significantly reduced ROM; there was no significant difference between the techniques. All instrumentation techniques were able to reduce ROM for most of the motions. The differences between the techniques were small. Nevertheless, the intralaminar screw showed deficits in lateral bending. CONCLUSION: Screw positioning seems to be of minor influence on stability in atlantoaxial stabilization. Therefore, the pars screw is a sound alternative to the established techniques from a biomechanical point of view. Anatomical considerations for screw placement should be kept in mind as a superior priority.