In 11 normal volunteers and six patients with Parkinson's disease, we compared six different analyses of dopaminergic function with L-3,4-dihydroxy-6-[18F]fluorophenylalanine (FDOPA) and positron emission tomography (PET). The caudate nucleus, putamen, and several reference regions were identified in PET images, using magnetic resonance imaging (MRI). The six analyses included two direct determinations of DOPA decarboxylase activity (k3D, k3*), the slope-intercept plot based on plasma concentration (K), two slope-intercept plots based on tissue content (k3r, k3s), and the striato-occipital ratio [R(T)]. For all analyses, the difference between two groups of subjects (normal volunteers and patients with Parkinson's disease) was larger in the putamen than in the caudate. For the caudate nucleus, the DOPA decarboxylase activity (k3D, k3*), tissue slope-intercept plots (kr3, ks3); and striato-occipital ratio [R(T)] analyses significantly discriminated between the normal volunteers and the patients with Parkinson's disease (p < 0.005) [with least significance for k3* (p < 0.05)], while the plasma slope-intercept plot (K) failed to do so. For the putamen, the values for k3D, k3*, K, k3r, k3s, and R(T) of normal volunteers were significantly higher than those of patients (p < 0.005) [with least significance for K (p < 0.025)]. Linear correlations were significant between k3D and k3s; k3D and k3r; k3D and R(T); and k3D and k3*, in this order of significance. We found no correlation between k3D and K values in the caudate nucleus.