The purpose of this study was to examine if measurements of submaximal blood lactate may improve the prediction of VO2max from submaximal exercise tests. Subjects (39 men & 21 women, age 20-44 yrs) completed a branching protocol on a cycle ergometer which consisted of 3 submaximal stages of 4 minutes each followed by a maximal effort. During the last minute of each stage heart rate (HR) was recorded from telemetry and a capillary blood sample was obtained from a finger puncture. Maximal power output was estimated from the extrapolation of the 3 submaximal HRs to age predicted HRmax and blood lactate concentration was measured using a Yellow Springs analyzer (Yellow Springs, OH). Data were analyzed using multiple regression procedures. Estimated maximal power output (238 ± 56 W) was significantly related to measured VOzmaX (p < 0.001, r2= .523). Mean values for ACSM estimated, ACSM-revised and measured VOz.X were not significantly different (3.170 vs. 3.140 vs. 3.260 L/min respectively), however the SEE % for the ACSM and ACSM revised estimated VOz,. were 13.4% and 12.1%, respectively. When the change in power output relative to the change in blood lactate concentrations (po/BL) from stage 1 to stage 2 and the estimated maximal power output from 3 submaximal heart rates were added as independent variables to the regression model with measured VO2,t as the dependent variable, po/BL was significantly correlated with VO X , accounting for 7.6% of the variance in VOz... In conclusion, only the relative change in power output to the relative change in blood lactate measure from stage 1 to _2 were significant in improving the estimation of VO2 during submaximal exercise tests using a branching protocol. Other blood lactate measures did not improve the estimation of VOA,. in this study.