Self-reported population ancestry data for the 19 census divisions (CDs) of Alberta, Canada, were correlated with multiple sclerosis (MS) prevalence rates in those divisions, for men and women separately; and parental ancestry was compared between a group of MS patients and controls attending the University of Alberta MS Clinic. At the CD level, there was a positive correlation between single Scandinavian ancestry and MS prevalence in men, but this was not confirmed in the case-control comparison. The case-control comparison indicated an excess risk of MS associated with single non-specific European as opposed to British ancestry in men only. When paternal versus maternal ancestry was considered separately, there was an excess risk of MS associated with non-specific European as opposed to British ancestry for both men and women, but on the father''s side only. Aboriginal ancestry was negatively associated with MS prevalence at the CD level in both men and women; and no MS patients with aboriginal origin were among cases assembled through the MS clinic.