Although many investigators have studied the morphology of bacterial inflammation in the central nervous system, and the lightmicroscopic appearance of bacterial leptomeningitis is well known, a number of fundamental questions remain unanswered.1It was anticipated that an experimental approach, exploiting the advantages of ultrathin sections of osmium-fixed tissue and the high resolving power of the electron microscope, might result in clarification or partial solution of some of these basic problems. These experiments were designed to allow fine structural observations to be made concerning the manner in which the different cellular elements that make up the brain surface and its coverings react to bacterial infection. As a necessary preliminary, the ultrastructure of the normal leptomeninges and brain surface was examined. In general, these observations coincided with those of Pease and Schultz.2More detailed consideration has already been given to the normal subarachnoid and perivascular spaces3and the process