Aerobic training results in both increases in VO2max as well as changes to cardiac structure and function. However, the degree to which changes in VO2max are linked to changes in cardiac dimensions or to improvements in cardiac function is not clear. The purpose of this study was to establish which individual, or combination of, cardiac parameters best predict changes in VO2max. Seventy one young, healthy, male participants (22 ±5 yrs; 1.79 ±0.07 m; 79 ±13 kg) volunteered to participate in the study, and represented a broad spectrum of cardiorespiratory fitness (athlete to sedentary). This facilitated the cross-sectional analysis of relationships between cardio-respiratory fitness with cardiac structure and function. Participants visited the laboratories on two separate occasions within 5 days. During the initial visit body composition, resting heart rate, resting blood pressure and echocardiographic examinations were performed. A maximal VO2max ramp protocol was employed during a second visit. Echocardiographic assessment included structural measurements of interventricular septum wall thickness (IVS), left ventricular diameter (LVD) and posterior wall thickness (PW) in both systole and diastole, from M-mode images. Left ventricular (LV) function was assessed by transmitral Doppler, LV volumes, Ejection Fraction (EF), TDI and longitudinal peak strain, from the apical 4 chamber view. One tailed Pearson correlation coefficients were established between cardiac parameters and absolute VO2max, relative VO2max and VO2max relative to FFM. Four predictor variables were subsequently entered into a least-squares regression analysis. The strongest correlations with absolute VO2max were for LVM (r = 0.692, p < 0.01) and SV (r = 0.692, p < 0.01) but the regression analysis found EDV, SV and LA to predict 65% of the variance in absolute VO2max (R = 0.81, F (3, 25) = 15.54, p < 0.01). LVM/BM was the strongest correlate to relative VO2max (r = 0.684, p < 0.01) and SV had the next strongest correlation (r = 0.483, p < 0.01), but was removed by the regression analyses due to collinearity, therefore LVM/BM was found to predict 47% of the variance in relative VO2max (R = 0.68, F (1, 28) = 24.67, p < 0.01). SV was found to have the strongest correlation with VO2maxrelative to FFM (r = 0.605, p < 0.01) and LVM/FFM had the next strongest correlation with VO2max relative to FFM (r = 0.598, p < 0.01), however LVM/FFM was removed by the regression analyses finding SV and PWd to predict 52% of the variance in VO2max relative to FFM (R = 0.75, F (2, 22) = 13.76, p < 0.01). The main findings of the current study were that cardiac structural measures were the strongest predictor of VO2max. These results suggest that increased cardiac size may make significant contributions to the concomitant increases in VO2max.