OBJECTIVE The purpose of this investigation was to develop a sub-maximal exercise test for estimating VO2max utilizing aerobic dance. METHODS One hundred females between the ages 18 to 40 yr served as the subjects for test validation. The subjects completed a treadmill test to determine VO2max and were assessed for heart rate (HR) response to a bout of aerobic dance. The data associated with responses to treadmill exercise and the aerobic dance test, in conjunction with descriptive variables (e.g., age, BMI) were utilized in the validation of the multiple regression model. METHODS Reliability was determined by correlation and paired "t"-tests of the aerobic dance routine test and retest trials. The construction of the multiple regression equation, via forward entry analysis, and the cross-validation of the regression equation were completed to ensure the validity and reliability of the protocol in accurately estimating VO2max. RESULTS Test, retest reliability for the dance-exercise routine was demonstrated (r = 0.98). Moreover, no significant differences were shown between the HR responses for the test and retest trials. The multiple regression analysis yielded a three variable multiple prediction equation for estimating VO2max (R = 0.84; SEE, 5.5 ml.kg-1.min-1). The three variables were the HR response to four min of aerobic dance (HR4), body mass index (BMI), and age (years). Cross-validation of the aerobic dance test was determined with a second group of 50 female subjects (R = 0.83; SEE, 5.5 ml.kg-1.min-1). Additionally, the application of the validation group regression equation to the cross-validation group yielded a comparable R of 0.82. Comparison of the predicted values for VO2max from both equations also yielded a highly significant invariance coefficient of 0.96. Finally, the results of "t"-tests between the observed and predicted mean values for VO2max revealed no significant difference (p > 0.05). Therefore, the final prediction equation, based on collapsing the means associated with the validation and cross validation samples (n = 150), was: VO2max (ml.kg-1.min-1) = 130.18-(0.38 * HR4)-(0.81 * BMI)-(0.27 * Age); R = 0.84, SEE = 5.5 ml.kg-1.min-1. CONCLUSIONS These results indicate that a four minute aerobic dance test provides a valid and reliable sub-maximal protocol for estimating VO2max and providing an index of aerobic fitness in apparently healthy 18 to 40 yr old females.