Late-season non-selective herbicide application reduces Lolium rigidum seed numbers, seed viability, and seedling fitness

Abstract

Experiments were conducted to investigate the effect of Lolium rigidum (annual ryegrass) seed developmental stage and application rate of glyphosate and SpraySeed (paraquat 135 g/L + diquat 115 g/L) on the number, germinability, and fitness of seeds produced. Glyphosate (450 g/L) was most effective when applied at a rate of 0.5–1 L/ha during heading and anthesis, reducing the number of filled seeds produced compared with unsprayed plants. Application post-anthesis, when seeds were at the milk to soft dough stage, was less effective. SpraySeed was most effective when applied post-anthesis, during the milk and early dough stages of seed development at a rate of 0.5–1 L/ha, resulting in the production of few viable seeds. Although some filled seeds were produced, most of the seeds were dead. Application during anthesis or once the seeds reached soft dough stage was less effective. For both herbicides, those seeds that were capable of germinating were smaller and had slower radicle and coleoptile growth, resulting in slower early seedling growth and reduced biomass production within the first month of growth. Additionally, glyphosate application reduced the proportion of seeds exhibiting dormancy. The anticipated reduction in seed competitive ability and altered emergence timing resulting from late-season herbicide application, even when application timing is not optimal, could be exploited to reduce the likelihood of successful L. rigidum establishment in the following season.