The objectives of this study were to design a field test based on the Conconi protocol to determine the ventilatory threshold of rowers and to test its reliability and validity. A group of sixteen oarsmen completed a modified Conconi test for on-water rowing. The reliability of the detection of the heart rate threshold was evaluated using heart rate breaking point in the Conconi test and retest. Heart rate threshold was detected in 88.8% of cases in the test-retest. The validity of the modified Conconi test was evaluated by comparing the heart rate threshold data acquired with that obtained in a ventilatory threshold test (VT₂). No significant differences were found for the values of different intensity parameters i.e. heart rate (HR), oxygen consumption (VO₂), stroke rate (SR) and speed (S) between the heart rate threshold and the ventilatory threshold, (170.9 ± 6.8 vs. 169.3 ± 6.4 beats·min^-1; 42.0 ± 8.6 vs. 43.5 ± 8.3 ml·kg^-1·min^-1; 25.8 ± 3.3 vs. 27.0 ± 3.2 strokes·min^-1 and 14.4 ± 0.8 vs. 14.6 ± 0.8 km·h^-1). The differences in averages obtained in the Conconi test-retest were small with a low standard error of the mean. The reliability data between the Conconi test-retest showed low coefficients of variations (CV) and high intraclass correlation coefficients (ICC). The total errors for the Conconi test-retest are low for the measured variables (1.31 HR, 0.87 VO₂, 0.65 SR, and 0.1 S). The Bland- Altman’s method for analysis validity showed a strong concordance according to the analyzed variables. We conclude that the modified Conconi test for on-water rowing is a valid and reliable method for the determination of the second ventilatory threshold (VT₂).