The purpose of this study was to compare two methods for determination of anaerobic threshold from two different treadmill protocols. Forty-eight Croatian runners of national rank (ten sprinters, fifteen 400-m runners, ten middle distance runners and thirteen long distance runners), mean age 21.7 +/- 5.1 years, participated in the study. They performed two graded maximal exercise tests on a treadmill, a standard ramp treadmill test (T(SR), speed increments of 1 km x h(-1) every 60 seconds) and a fast ramp treadmill test (T(FR), speed increments of 1 km x h(-1) every 30 seconds) to determine and compare the parameters at peak values and at heart rate at the deflection point (HR(DP)) and ventilation threshold (VT). There were no significant differences between protocols (p > 0.05) for peak values of oxygen uptake (VO(2max), 4.48 +/- 0.43 and 4.44 +/- 0.45 L x min(-1)), weight related VO(2max) (62.5 +/- 6.2 and 62.0 +/- 6.0 mL x kg(-1) x min(-1)), pulmonary ventilation (VE(max), 163.1 +/- 18.7 and 161.3 +/- 19.9 L x min(-1)) and heart rate (HR(max), 192.3 +/- 8.5 and 194.4 +/- 8.7 bpm) (T(FR) and T(SR), respectively). Moreover, no significant differences between T(FR) and T(SR) where found for VT and HR(DP) when expressed as VO2 and HR. However, there was a significant effect of ramp slope on running speed at VO(2max) and at the anaerobic threshold (AnT), independent of the method used (VT: 16.0 +/- 2.2 vs 14.9 +/- 2.2 km x h(-1);HR(DP): 16.5 +/- 1.9 vs 14.9 +/- 2.0 km x h(-1) for T(FR) and T(SR) respectively). Linear regression analysis revealed high between-test and between-method correlations for VO2, HR and running speed parameters (r = 0.78-0.89, p < 0.01). The present study has indicated that the VT and HR(DP) for running (VO2, ventilation, and heart rate at VT/HR(DP)) are independent of test protocol, while there is a significant effect of ramp slope on VT and HR(DP) when expressed as running speed. Moreover, this study demonstrates that the point of deflection from linearity of heart rate may be an accurate predictor of the anaerobic threshold in trained runners, independently of the protocol used.