Purpose: It is well known that individuals with mental retardation (MR), especially those with Down syndrome (DS), have low maximal heart rates (MHR). We evaluated the ability to predict MHR in individuals with MR and DS in comparison with persons without MR. Methods: Subjects completed a maximal exercise test on the treadmill with metabolic and HR measurements. Stepwise multiple regression was used to develop prediction equations for subjects with MR (N=276; 97 with DS) and without (N=296) MR, ranging in age from 9-46 yr. Results: Subjects with MR exhibited significantly lower MHR (177 vs 185 beats-min-1) and VO2peak (33.8 vs 35.6 mL·kg-1·min-1). In subjects with MR, age was a poor predictor of MHR, Y=189 - 0.59 (age) (R=0.30, SEE=13.8 beats·min-1; P<0.01), but age was a better predictor for subjects without MR, Y=205 - 0.64 (age) (R=0.52, SEE=9.9 beats·min-1; P<0.01). A large sample Z test indicated that these regression coefficients were significantly different (P<0.01). However, adding DS to the regression improved the prediction for subjects with MR, Y=210 - (0.56 age) - (15.5 DS) (R=0.57; SEE=11.8 beats·min-1, p<0.01). Conclusion: MHR can be predicted with similar accuracy in subjects with and without MR, provided DS is accounted for in the equation for the subjects with MR.