The aim of the present study was to investigate the relationship between involvement of specific areas of the brain and the occurrence of depression and anxiety in patients with multiple sclerosis. We studied 95 patients (62 women and 33 men, mean age 39.5 years, SD 11.2) with definite MS, 97 patients (65 women and 32 men, mean age 40.7, SD 11.9) suffering from chronic rheumatoid diseases and 110 healthy subjects (71 women and 39 men, mean age 40.1, SD 12.7). The disability, the independence, the cognitive performances, the depressive and anxiety symptoms were assessed. The diagnosis of major depression was made according to the DSM-IV. The patients with multiple sclerosis underwent a 1.5 Tesla magnetic resonance examination including T1 and T2 weighted images. Calculation of regional and total lesion loads and brain volumes were performed. The number (%) of subjects with a diagnosis of major depression was 18 (18.9) among MS cases, 16 (16.5) among controls with chronic disease (p=NS), and 4 (3.6) among healthy volunteers (p < 0.0001). The Hamilton Depression and Anxiety rating scales median scores were 5 and 18, respectively in the MS patients, 5 (p= NS) and 14 (p= NS) in the chronic rheumatoid diseases controls, and 3 (p= < 0.0001) and 6 (p= < 0.0001) in the healthy controls. Both severity of depressive symptoms and diagnosis of major depression correlated, albeit weakly, with right frontal lesion load (r=0.22, p=0.035, and r=0.23, p=0.026, respectively) and right temporal brain volume (r=0.22, p=0.005 and r=0.22, p=0.036, respectively). The severity of depression was related significantly also with total temporal brain volume (r=0.26, p=0.012), right hemisphere brain volume (r=0.25, p=0.015), disability (r=0.30, p=0.003) and independence of MS cases (r=-0.26, p=0.01). The anxiety did not correlate significantly with any of the measures of regional and total lesion loads and brain volume or with any of the considered clinical variables. The similar frequency of depression and severity of depressive symptoms in MS patients and in chronic disease patients, the significant difference in this respect with the normal controls, and the significant correlation between depression and the disability measures would suggest a psychological reaction to the impact of the disease but the relationship between depression and the alterations in the frontal and temporal lobes of the right hemisphere supports, on the contrary, the causative role of organic brain damage. The lack of any significant association between symptoms of anxiety and either MRI abnormalities or clinical variables led us to the opinion that anxiety is a reactive response to the psychosocial pressure put on the patients.