Abstract: In order to find out the difference of rheotaxis between narrow range species Pseudorasbora elongata and widespread species Pseudorasbora parva, and whether differences in flow sensitivity contribute to their contrasting distribution patterns, with the aim of informing conservation measures for P. elongata. In this study, we examined the induction velocity for P. elongata and P. parva with the increasing velocity protocol. The results showed that absolute induction velocity (AIF) was (2.88±1.00), (6.81±1.20), (9.63±2.28) cm/s, and relative induction velocity (RIF) was (0.47±0.11), (0.83±0.11), (1.09±0.24) BL/s for body length of 5.0—7.0, 7.0—8.5, 8.5—9.0 cm for P. elongate, respectively. Both AIF and RIF differed significantly among length groups (P<0.05). The AIF and RIF for body length of 5.0—7.0, 7.0—8.0, 8.0—9.0 cm was (5.25±0.82), (7.34±0.46), (7.91±1.03) cm/s, and (0.83±0.10), (0.98±0.06), (0.94±0.10) BL/s for P. parva, respectively. Except for the groups between 7.0—8.0 cm and 8.0—9.0 cm, there were significant differences in AIF and RIF between the other body length groups (P<0.05). The AIF and RIF were positively correlated with body length for both two species (P<0.001). Independent-samples t test showed that the AIF and RIF of P. elongata were significantly lower than that of P. parva (P<0.05). The greater sensitivity of the narrow-ranged P. elongata to water flow, compared to the widespread P. parva, is likely a key factor constraining its distribution. Therefore, conservation of P. elongata habitats should fully account for its rheotactic characteristics. River realignment should follow natural variations in channel width and morphology while meeting flood-discharge safety requirements. In addition, in the sections where the P. elongata has been distributed, the density of lower head dams should be reduced to maximize free-flowing intervals, and removal should be considered where feasible.