Joubert syndrome (JS) is a rare autosomal recessive malformation syndrome involving agenesis or dysgenesis of the cerebellar vermis with accompanying brainstem malformations. JS is further characterized by hypotonia, developmental delay, intermittent hyperpnea, and abnormal eye movements. The biochemical and molecular basis of JS remains unknown, although several genes that are crucial in the development of the cerebellum have been proposed as attractive candidate genes. JS is clinically heterogeneous; this, together with previous linkage analyses, suggests that there may also be genetic heterogeneity. A locus for JS was previously identified on chromosome 9q34 by linkage analysis in a consanguineous family of Arabian origin. A putative second JS locus was recently suggested when a deletion on chromosome 17p11.2 was observed in a patient with Smith‐Magenis syndrome and JS phenotype. We have investigated a cohort of apparently unrelated North American JS pedigrees for association with the loci on chromosomes 9q34 and 17p11.2 and excluded them in all cases where data were informative. Analysis of an additional 21 unrelated JS patients showed no evidence of homozygosity at the 9q34 and 17p11.2 loci that would suggest inheritance of founder JS mutation(s) or unreported consanguinity. Together, these data suggest that one or more major loci for JS remain to be identified. Consequently, we undertook mutation analysis of several functional candidate genes, EN1, EN2, and FGF8, in a total of 26 unrelated JS patients. Our data suggest that all of these genes may be excluded from a direct pathogenic role in JS. The BARHL1 gene, which localizes to chromosome 9q34 and has previously been proposed as a strong positional candidate gene for JS, was also investigated and excluded from involvement in JS that is linked to chromosome 9q34. © 2002 Wiley‐Liss, Inc.