Establishment of four pasture grasses and Siratro from seed oversown into dense and open speargrass pastures

Abstract

In two experiments, seed of Gayndah and Biloela buffel grass (Cenchrus ciliaris), green panic (Panicum maximum var. trichoglume), signal grass (Brachiaria decumbens) and Siratro (Macroptilium atropurpureum) was broadcast into a range of seedbeds imposed on native speargrass (Heteropogon contortus) pastures that had either been cleared of trees 4-5 years before sowing (C) or had the trees killed at sowing (K). The seedbeds were: untreated control (NP); mown to 3 cm before and at sowing (M); M followed by regular clipping (MD); burnt (B); B followed by regular clipping (BD); and herbicide (H). Seed was also sown into a cultivated seedbed (P) in C. The number of plants established 15-16 months after sowing was highest in the cultivated seedbed and lowest in seedbeds where competition from the native grasses was greatest. Competition was greater in C than in K, the native pastures in C having about three times more dry matter than those in K. Burning reduced competition but also reduced seedling emergence. It increased establishment slightly in K but not in C. The M and MD treatments failed to increase establishment over that of the NP control. Herbicides reduced competition and increased establishment in C and K, but only for signal grass and Siratro in C when drier conditions occurred in the second experiment. Siratro established better than the grasses in the presence of competition when weather conditions were favourable, but there was little Siratro or grass establishment when water deficits occurred soon after emergence, especially in C