DESCRIPTION (provided by applicant): All subcortical arrangements are primarily nuclear in type. The cortex was the first part of the brain to evolve a radial and laminar arrangement of cells. The resultant modular arrangement is based on the cell minicolumn: a self-contained ecosystem of connectivity linking afferent, efferent and interneuronal connections. Our preliminary study of the fundamental processing unit of the cortex, the minicolumn, indicates that the modular neocortical organization in the brains of autistic individuals differs from that of controls. More specifically, a study of 3 neocortical sites in 9 autistic brains and an equal number of normal controls has shown significant differences in the horizontal spacing that separates minicolumns. It also revealed differences in their internal structure: less neuropil space in the periphery of the minicolumn and increased mean cell spacing. Similar abnormalities have now been reported in Asperger's syndrome but appear absent in Down syndrome (a control group added for mental retardation). This study will examine differences in minicolumnar morphometry between 24 autistic brains and 24 controls, matched for age and sex, from both the Autism Tissue Program and the Yakovlev-Haleem Collection (Armed Forces Institute of Pathology). The patients provide a new cohort not previously examined by the investigator. Ten Down syndrome brains will provide a second control group. Sampled cortical regions include Brodmann's areas 4, 9, 10, 17, 21, 22, 30, 33, and 39, which represent sensory, motor, association and limbic cortex obtained from 4 different brain lobes. The study will utilize quantified computer imaging and stereology in both Nissl and myelin-stained material. The parameters tested include the width of minicolumns, the amount of neuropil space within them, the mean cell density within each minicolumn, individual lamina depth, cell size and type (pyramidal versus non pyramidal), and the distribution of cell types among different lamina. The study will employ inter-myelinated bundle measurements as a way of corroborating our minicolumnar width parameters. We will compare the results within multiple regions of the cortex as well as for lateralization. Finally, we will correlate our minicolunmar parameters with different dimensions on the ADI. This study aims at better understanding the neuroanatomical basis of autism. If validated, positive results from our studies will yield a great understanding of autism. For example, many alleged animal models of autism lack the human neuroanatomic information to justify their utility. Candidate gene studies could also be refined if neuroanatomic development in autism was better understood. Treatment modalities, especially neuropharmaceutical agents, will prove more specific as their targets are better understood.