Cortical surface vessels were monitored through closed cranial windows with an epifluorescence microscope and SIT or ICCD cameras. Fluorescent dextrans or 1.3 microns latex beads were injected into the contralateral jugular vein for plasma labeling and for vascular transits. For close arterial transits, these tracers or physiological saline were injected into the ipsilateral external carotid artery. AVTTs were calculated from intensity differences of tracers between a branch of the MCA and a vein draining the same cortical region over time. AVTTs for saline dilutions of RBCs were significantly shorter (0.73 times) than for dextrans. Both dextrans and beads distributed with plasma. With FITC-dextran, inner diameters of arterioles and venules averaged 6 microns larger than hemoglobin under green light. This difference was likely due to the segregation of red blood cells and plasma during flow. Velocities of individual fluorescent beads were measured in pial vessels by strobe epi-illumination. Plots of bead velocities against radial position in arterioles were blunted parabolas. Peak shear rates in the marginal layer next to the vessel walls were determined directly from bead tracks in arterioles (D = 21-71 microns) and were 1.32 times the Poiseuille estimate. The calculated peak wall shear stress was 39 +/- 14 dyn/cm2 (mean +/- SD) for these arterioles but was probably severalfold greater in the smallest terminal pial arterioles. Vmax near the axes of arterioles increased with D+0.5. The calculated peak wall shear rate was highest in small arterioles and decreased with D-0.5. The calculated flow Q increased with D+2.5. These methods permit direct, simultaneous, dynamic measurements on multiple identified cerebral microvessels.