Forage grass cultivars must have adequate grazing tolerance for use in the grazing systems for which they are intended. Response to 2 cycles of selection for persistence under heavy grazing pressure was examined in 3 winter-active breeding populations of the productive perennial grass, phalaris (Phalaris aquatica L.), from 2000 to 2003, at Bulart in western Victoria and Rye Park on the Southern Tablelands and Tamworth on the North-West Slopes of New South Wales. There was one continuously grazed and one rotationally grazed set of plots at Bulart to examine the effect of grazing management.

All sites were affected by drought in later years but drought stress was most severe at Tamworth. A strongly positive linear response to selection was observed in an analysis of persistence measured as frequency of phalaris plant base across the 3 sites (excluding the rotational treatment at Bulart), but response interacted with site. By 2003, linear response averaged 14% frequency units or 34–40% proportional response per cycle at Rye Park and the continuously grazed plots at Bulart, sites that were environmentally suited to survival of phalaris. In contrast, response to selection was absent or slightly negative at Tamworth where conditions were drier and hotter. Populations responded similarly in analyses across all 3 sites but 1 population was less responsive when analyses were restricted to Bulart and Rye Park. Herbage mass measurements in 2001 at Bulart and 2002 at Rye Park indicated positive responses to selection, which were increasingly linked to frequency over time. Rotationally grazed plots at Bulart displayed higher frequency than continuously grazed plots after 3 years of grazing.

The experiment showed that grazing tolerance was a heritable trait in the populations tested when environmental constraints were not limiting and that useful improvements in persistence had been obtained compared with existing cultivars.