Measurement of portal blood flow in dogs using duplex-Doppler ultrasonography has potential clinical utility for diagnosis of cirrhosis and portosystemic shunts. Three methods for calculating portal blood flow were compared to identify a consistent method with minimal potential for intraoperator variation. Sixteen unsedated, normal beagles had their mean portal blood flow velocity determined using three methods: method 1 used a small Doppler sample volume positioned over the center of the portal vein and manual selection of data points on the resulting velocity spectra, followed by arithmetic correction of the calculated maximum velocity to mean velocity using the factor 0.57; method 2 used a small centrally-located Doppler sample volume and computed maximum flow determination, again corrected using the factor 0.57; method 3 used a large Doppler sample volume that overlapped the walls of the portal vein followed by computed mean blood flow velocity determination. The calculated mean (SD) portal blood flow velocity was 15.8 (1.8), 15.1 (1.4), and 14.7 (2.5) cm/s using methods 1–3, respectively. There was no significant difference between the results of the three methods; however, method 3 (uniform insonation) was easier to use and produced a higher amplitude Doppler signal and therefore represents a useful alternative to the peak flow methods.