The study tested the hypothesis that physiological measurements can predict marathon running performance in a top-level homogeneous group of males and females. Ten male, performance: 2:12:04, coefficient of variation (CV)=2.33%, and 8 female marathon runners, performance: 2:34:53, CV=4.54%, performed an increment test on the treadmill (starting speed, 8 km.h-1; increments, 2 km.h-1; increment duration, 3 min to exhaustion). The heart rate (HR), VO2 and the lactate concentrations were measured at the end of each exercise level. During the recovery time, HR and lactate were measured. Furthermore, echocardiographic, anthropometric and haematologic measurements were made. The results of a stepwise multiple regression analysis using marathon time as the dependent variable yielded R2=0.983 for the male group and R2=0.984 for the female group. The model for the male group used the independent variables lactate value at 10 km.h-1, left ventricular telediastolic diameter (LVD) and lactate value at 22 km.h-1. The model for the female group used the independent variables subscapular skinfold, serum ferritin and sum of six skinfolds. Our study demonstrates that in males and females, physiological parameters can explain the variance in marathon time among elite homogeneous groups.