Twenty-seven patients with Raynaud syndrome (mean age at onset 11.7 years) were studied to determine the prevalence of primary Raynaud syndrome and to assess the predictive role of antinuclear antibody, nail-fold capillary microscopy, and photoelectric plethysmography in this population. Fourteen patients (52%) had a connective tissue disease, four (15%) had a probable connective tissue disease, and nine (33%) had primary Raynaud syndrome. In all patients with either a connective tissue disease or a probable connective tissue disease, there was a positive reaction to antinuclear antibody, in contrast to patients with primary Raynaud syndrome, in whom antinuclear antibody was not detected. Nail-fold capillary microscopy scores differed significantly between patients with either a connective tissue disease or a probable connective tissue disease and those with primary Raynaud syndrome for both enlarged loop score (p less than 0.025 and less than 0.05, respectively) and avascular score (p less than 0.005 and less than 0.01, respectively). Photoelectric plethysmography scores were reduced in all groups but did not differ significantly between groups. Our findings suggest that in children with Raynaud syndrome, the primary type is more common than was originally suspected, and that both antinuclear antibody and nail-fold capillary microscopy, but not photoelectric plethysmography, can distinguish patients with primary Raynaud syndrome from those with either a connective tissue disease or a probable connective tissue disease.