Effects of Neotyphodium lolii infection and sowing rate of perennial ryegrass (Lolium perenne) on the dynamics of ryegrass/subterranean clover (Trifolium subterraneum) swards

There is no clear rationale for use of endophyte-infected or endophyte-free seed of perennial ryegrass when sowing pasture in Australia. This research was undertaken to clarify the importance of ryegrass endophyte in temperat e pasture systems. A field experiment was conducted to study the dynamics o f perennial ryegrass/subterranean clover swards in response to (a) endophyt e infection in perennial ryegrass seed, (b) the sowing rate of this seed, a nd (c) the cultivar of subterranean clover. A higher density of perennial r yegrass seedlings established from seed infected with the endophyte Neotyph odium lolii (E+) than with endophyte-free (E-) seed and there was a signifi cant interaction with the sowing rate. At the highest sowing rate of 16 kg/ ha, about 25% more seedlings emerged in the E+ than the E- treatment. The s ize of individual ryegrass plants at establishment was reduced as sowing ra te increased, but endophyte had no effect on their size. The slopes of the curves describing ryegrass tiller density relationships with sowing rate we re similar for both endophyte treatments in the first 2 years but were diff erent in the third year. Between the first and third years, tiller density in the E+ plots sown at the lowest rate increased by up to 1000/m(2), where as in E- plots the corresponding increase was only 250/m(2). Such changes h ave not previously been demonstrated. This difference is believed to be due to the superior ability of E+ plants to recover after extended dry periods . At higher sowing rates, neither endophyte treatment led to increased till er densities between the first and third years. The presence of N. lolii ha d no effect on plant density of subterranean clover during the 3 years of t he experiment. In the second year, the 3 clover cultivars tested had simila r patterns of reductions in plant density in response to increased sowing r ate of the ryegrass in the previous year, but these patterns did not persis t into the third year. Although dry matter (DM) yield of herbage in the thi rd year was not responsive to initial sowing rate, the E+ treatment was sti ll significantly higher than the E-. The cultivar of subterranean clover ha d little effect on companion ryegrass, and total DM production was insensit ive to cultivar of clover. The use of endophyte-infected seed is desirable in order to maintain long-term density of perennial ryegrass in sheep-produ ction systems in Australia.