Thirty-eight female subjects (M +/ SD = 33 +/- 3.0 years) had VO2max measured on the cycle ergometer (M +/- SD = 37.3 +/- 6.4 ml.kg-1.min-1) and on the treadmill (M +/- SD = 41.3 +/- 6.6 ml.kg-1.min-1). VO2max was estimated for each subject from heart rate (HR) at submaximal workloads on the cycle ergometer using the Astrand-Rhyming nomogram (A/R) and the extrapolation method (XTP). VO2max was also estimated from three field tests: 1.5-mile run (RUN) (independent variable [IV] = time), mile walk (WALK) (IV = time, age, HR, gender, body weight), and the Queens College Step Test (ST) (IV = HR during 5-20 s recovery). Repeated measure ANOVA revealed significant mean differences between the criterion cycle ergometer VO2max versus A/R and XTP (20 and 12% overestimation). The WALK, RUN, and ST VO2max values were not significantly different from the criterion treadmill VO2max. The correlation between criterion VO2max estimated from the WALK and RUN were r = .73 (SEE = 4.57 ml,kg-1.min-1) and r = .79 (SEE = 4.13 ml.kg-1.min-1), respectively. The ST, A/R, and XTP had higher SEEs (13-13.5% of the mean) and lower r s (r = .55 to r = .66). These results suggest both the WALK and RUN tests are satisfactory predictors of VO2max in 30 to 39-year-old females.