There is limited knowledge of the distance larvae of Murray-Darling Basin riverine fishes drift or what conditions influence them to strop drifting and settle. Studies in other rivers and the marine environment suggest that the characteristics of the river reach in which larvae are released may influence whether they drift through that reach or are retained. In a previous study, we identified a gradient of ‘retentiveness’ of 12 reaches of the lower Ovens River from very high to very low. The aim of the present study was to quantify the relationship between reach-scale retentiveness of fish larvae of species with contrasting swimming abilities (Murray cod = good, golden perch = poor), by using field-based experimental releases of larvae and passive particles. Larval release experiments were carried out in the lower Ovens River between November 2018 and March 2019 at three discharges and in four reaches of differing retentiveness. A total of 140,000 marked Murray cod and golden perch larvae were released. A greater proportion of golden perch (poor swimmers) than Murray cod (good swimmers) larvae were retained in all reaches, regardless of flow or retentiveness. In the majority of cases, more larvae of both species than passive particles were retained in reaches, regardless of reach retentiveness. At moderate and high discharges, Murray cod larvae are able to determine whether they stay or move through a reach. At low and high discharges, golden perch larvae act at times more like passive particles than do Murray cod larvae, but still have a degree of capacity to determine whether they stay or move through a reach. If managers want to maximise the potential recruits in a section of river, then increasing physical complexity of that reach, or altering flows to increase retentiveness, will allow fish larvae to remain. Maintaining a diversity of hydraulic, structural and geomorphic conditions in river reaches will allow fish larvae to be retained or move through, if they choose.