Red blood cells labeled with the carbocyanine dyes, 1,1'-dioctadecyl-3,3,3',3'-tetramethylindocarbocyanine perchlorate (DiI) and 3,3'-dioctadecyloxacarbocyanine perchlorate (DiO), were evaluated for use in making microvascular measurements in rat small intestine and spinotrapezius muscle. We determined the minimum concentration of each dye which produced near maximal fluorescent intensity and labeled cell fraction. These dyes, which have excitation and emission spectra similar to fluorescein and rhodamine derivatives, have a number of advantages over the isothiocyanates: (1) the labeling procedure is quicker, easier, and less expensive; (2) the labeled cell fraction and the fluorescent intensity of DiI and DiO cells are stable for long periods of time in the rat circulation; and (3) DiI-labeled cells are brighter and transmit light through overlying erythrocytes better than rhodamine X isothiocyanate. However, in vitro and in vivo evaluations illustrate the potential limiting effects of vessel diameter and cell velocity on the accuracy of microvascular measurements made using this technique. In the small intestine and spinotrapezius muscle preparations, measurements of labeled cell flux were readily reproducible and could be partly automated with image analysis only in capillaries and small venules. Counting labeled cells in larger vessels by human observation or with automation was not reproducible, presumably due to absorption and dispersion of the fluorescent signal by overlying erythrocytes and smearing of the cell image at high cell velocities.