This statistical comparison between patients with cervical myelopathy secondary to horizontal atlantoaxial subluxation and those with vertical translocation is designed to elucidate the mechanisms responsible for cranial settling and the effect of translocation on the development of spinal cord compression. In a 10-year study of a cohort of 256 patients, 186 suffered from myelopathy and 116 (62%) of these exhibited vertical translocation according to the Redlund-Johnell criteria. Vertical translocation occurred after a significantly longer period of disease than atlantoaxial subluxation (p < 0.001). Translocation was characterized clinically by a high cervical myelopathy with features of a cruciate paralysis present in 35% of individuals compared with 26% who exhibited horizontal atlantoaxial subluxation (p = 0.29), but there was a surprising paucity of cranial nerve problems. The patients with vertical translocation had a greater degree of neurological disability (p = 0.002) and poorer survival rates (p = 0.04). Radiologically, vertical translocation was secondary to lateral mass collapse and associated with a progressive decrease in the atlantodens interval ([ADI], r = 0.4; p < 0.001) and pannus (p = 0.003). Thirty percent of patients exhibited an ADI of less than 5 mm. This phenomenon has been termed pseudostabilization. The authors' studies emphasize that the ADI (frequently featured in the literature) is totally unreliable as an indicator of neuraxial compromise in the presence of vertical translocation.