People with schizophrenia and primary negative symptoms (deficit schizophrenia) differ from those without such symptoms (nondeficit schizophrenia) on risk factors, course of illness, other signs and symptoms, treatment response, and biological correlates. These differences suggest that the 2 groups may also have developmental differences. A previous study found that people with schizophrenia have a wider palate than comparison subjects. We tested the hypothesis that those with deficit and nondeficit schizophrenia would differ on palate width. A dentist made blinded measurements of palate shape in deficit (N = 21) and nondeficit (N = 25) patients and control subjects (N = 127), matched for age and gender. The deficit group had significantly wider palates than either nondeficit or control subjects (respective means [standard deviation] 37.5 [3.9], 33.7 [3.1], and 34.0 [2.9]; P < .001 for both deficit/nondeficit and deficit/control comparisons, respective effect sizes 1.08 and 1.01). The nondeficit/control difference in width was not significant (P = .83), and there were no significant group differences in length or depth. The power to detect a nondeficit/control difference in width equal in size to that of the deficit/control difference in width (3.5 mm) was 0.99 and 0.92 for a 2.0-mm difference. This difference in palate width may reflect a divergence in development between deficit and nondeficit patients that occurs by the early second trimester and is consistent with the hypothesis that deficit schizophrenia is a separate disease within the syndrome of schizophrenia.