The cotiipet(ti.ve abilities of a wide range of genotypes of wheat {Trilicwnaestivum L.) and durum wheat {Triticum durum Desf.) against Lolium rigidunt Gatud. (annual ryegrassj were examined lo determine the potential for breeders to select strongly competiti-ve varieties. Considerable potential within the wheat genonie to hreed varieties with greater competitive ability was demonstrated. In 1993. 250 genotypes from aroimd the world were screened and in 1994 st subset of 45 (mainly Australian) genotypes were further examined. A uniform deasity of L. rigidum reduced grain peld of wheat by up to about 80% in 1993 and to 50% in 1994, depending on wheat genotype. Reduction in grain yield was correlated with i. rigidum dry matter. Wheats varied in competitive ability with source, and durum wheats were less competiifive than T. aestivum. The old' standard wheat varieties (released between 1880 and 1950) suppressed the weed more than all the current varieties, with the exception of eight Fi hybrids. A doubling of the crop seeding rate of 10 of the genotypes in 1994 reduced the biomass of L. rigidum by an average of 25% compared with the standard seeding rate. Ranking of competitive ability of varieties at high deosity was consistent at both seeding rates. The strongly competitive genotypes had high early biomass accumulation, large numbers of tillers, and were taU with extensive leaf display. The potential for breeding enhanced competitive ability in wheat is discussed.

The relative competitive advantage of 12 commercially available wheat varieties was examined against Lolium rigidum Gaud. at a number of sites from 1995 to 1997 in south‐eastern Australia. Nearly all the variation in crop grain yield was attributable to the variety × environment effects (81%), with only 4% due to variety × weed × environment effects. Some varieties exhibited an environment‐specific competitive advantage, for example Katunga, Dollarbird and Hartog, whereas others like Shrike, Rosella and Janz were relatively poorly competitive in some situations. The introduction of greater genetic variability into wheat is required to significantly increase competitiveness. Alternatively, manipulating crop agronomy, such as increasing crop seeding rate, may be a practical alternative. The grain yield of weed‐free wheat was highly positively correlated with grain yield of the weedy plots, suggesting that local adaptation is important for strong competitiveness, and that wheat breeders in southern Australia may be inadvertently selecting for competitive advantage with weeds when selecting for other traits such as early vigour. The varieties which showed competitive yield advantage also suppressed L. rigidum. A combination of short‐term agronomic manipulations and a longer‐term breeding effort is needed for increasing wheat competitiveness, and the increasing importance of herbicide‐resistant weeds may facilitate this process.

SummaiyThe competitive abilities of eight winter crops were compared against Lolium rigidum Gaud, (annual ryegrass), an important weed of southem Australia, as a potential strategy to suppress weeds and reduce dependence on herbicides. Two cultivais of each species were chosen to represent the range of competitive ability within each crop and grown in field experiments in 1992 and 1993. The order of decreasing competitive ability (with the ranges of percentage yield reduction from L. rigidum at 300 plants m"-^ in parenthesis) was as follows: oats {Avena saliva L.), 2-14%; cereal rye {Secale cereale L.), 14-20%; and triticale (X Triticosecale), 5-24%; followed by oilseed rape, (Brassica napus L.). 9-30%; spring wheat {Triticum aestivum L.), 22-40%; spring barley {Hordeum vulgarel..), 10-55%; and, lastly, field pea {Pisum sativum L.), 100%, and lupin (Lupinus angustifolius L.), 100%. Differences in competitive ability of cultivars within each species were identified, but competition was strongly influenced by seasonal conditions.'Competition for nutrients (N, P and K) and hght was demonstrated. L. rigidum dry matter and seed production were negatively correlated with grain yield of the weedy crops. More competitive crops offer the potential to suppress grass weeds while maintaining acceptable grain yields. Ways of improv-© 1995 European Weed Research Society ing the competitive abilities of grain legume crops are discussed. LoUum rigidum Gaud, (annual ryegrass) on yield of wheat.

Total reliance on herbicides for weed control is unsustainable with the spread of herbicide resistance and the environmental need to reduce pesticide use. Strongly competitive wheat crops that have high tolerance to weed pressure and therefore maintain high yields in the presence of weeds are a low-cost option for reducing dependence on herbicides. We examined the feasibility of selecting for wheat tolerance to weeds by crossing varieties differing for traits associated with competitiveness. Competitive ability and yield potential must be treated as separate traits for selection. Current measures of crop tolerance to weed competition do not separate the two traits so that selection based on these measures is often synonymous with selection for yield potential rather than pure tolerance. We propose a new measure, termed Incremental Crop Tolerance (ICT) that reflects the incremental yield difference between genotypes associated with tolerance, over and above differences in underlying yield potential.

The influence of wheat variety on the dose-response of annual ryegrass to diclofop-methyl (POST) was examined in the field in 1992 and 1993 in southern New South Wales, Australia. The aim was to determine if planting a strongly competitive variety of wheat improved control of annual ryegrass at reduced doses of diclofop-methyl. Suppression of ryegrass was dependent on herbicide dose, season, and wheat variety. In the absence of herbicide, dry matter (DM) production of annual ryegrass at 300 plants m−2at anthesis was 500 g ha−1with Dollarbird and Katunga compared to 1000 g ha−1with Rosella or Shrike in 1992. In 1993, DM was approximately 150 g ha−1with Dollarbird or Katunga, and 350 g ha−1with Shrike or Rosella. Ryegrass DM was reduced by diclofop-methyl to a greater extent, relative to the weedy unsprayed controls, with less competitive varieties Rosella and Shrike than with the more competitive Dollarbird or Katunga. Diclofop-methyl at 0.28 kg a.i. ha−1reduced DM of ryegrass growing with Katunga to less than 100 g m−2in 1992, compared to more than 200 g m2with the other varieties. In 1993, diclofop-methyl was more effective on ryegrass, and the same dose reduced ryegrass DM to almost zero in all varieties. Grain yields in unsprayed weedy controls of Dollarbird and Katunga were reduced approximately 20% by annual ryegrass compared with yields achieved with herbicides in both years. Yields of Rosella and Shrike in the unsprayed controls were reduced about 40% in 1992 and 60% in 1993. Only small increases in grain yields of all varieties occurred from diclofop-methyl doses above 0.13 kg a.i. ha−1. Poorly competitive varieties were dependent on herbicides to achieve grain yield potential and had a greater risk of weed survival when herbicide efficacy was reduced. In contrast, strongly competitive varieties, likely to retard build-up of weed seed in the soil, are less dependent on herbicides to achieve grain yield potential, and therefore result in reduced weed control cost.

The variation in field pea grain yield and competitiveness with annual ryegrass due to crop density, row spacing and cultivar was determined to enable farmers to better manage weeds with cultural control tactics. Crop density varied with seeding rate, cultivar, row spacing and year. Higher seeding rates were required to reach equivalent plant densities in cv. Dinkum (short, semi-leafless) compared with cv. Dundale (tall, conventional-leaf), and at 36 cm compared with 18 cm row spacing. Field pea grain yield was reduced more at low crop densities, in Dinkum, at 36 cm row spacing, and in the presence of weeds. Percentage yield losses from weed competition were similar in both cultivars (about 70–80%) at a low density of 10 plants/m2 in 2 seasons. At higher crop densities Dinkum had a larger loss than Dundale (i.e. at 30 plants/m2 losses were 60 and 35%, respectively, compared with 50 and 5% at 60 plants/m2). Seasonal variation influenced the effect of crop density on yield loss from weeds. The percentage yield loss from weeds in 1993 ranged from about 90 to 40% at plant densities of 10 to 40 plants/m2, in contrast to 1995 when 40% yield loss occurred at all these densities. Ryegrass dry weight was reduced with increasing field pea density in both years, and in the tall more than the short cultivar in 1 year. Maintaining recommended field pea seeding rates has considerable financial benefits in both weed-free (around $100/ha) and weed-affected crops (to $400/ha). In weedy situations, integrating cultural practices such as higher seeding rates and choice of cultivar that enhance crop competitiveness will improve weed management in south-eastern Australia.

Vulpia is a widespread weed of temperate Australian pastures, and readily replaces more productive species. Short-term management of vulpia is possible with herbicides but densities rapidly increase in poorly competitive pastures after herbicide application. A field experiment at Wagga Wagga, NSW, examined the effect of 2 fertility levels and 4 pasture types [subterranean clover sown at 1, 25, 100 kg/ha, and subterranean clover (25 kg/ha) + annual ryegrass (20 kg/ha)] on 2 densities of vulpia (50,5 500 plants/m2) from 1990 to 1994. Initially vulpia plant density was inversely related to sowing rate of subterranean clover, but over time this effect declined as the subterranean clover populations converged. Presence of annual ryegrass always resulted in lower vulpia plant, panicle and seed densities compared with treatments where subterranean clover only was present. Respective densities per m2 in 1993 for the average of the subterranean clover monocultures and for annual ryegrass plus subterranean clover were: plant 1315 v. 265; panicle 6700 v. 130; seed 542 400 v. 3460. The effect of drought in 1994 and presence of annual ryegrass were shown to significantly lower the sustainable population of vulpia at Wagga Wagga from 5000–6000 to <1000 plants/m2. The short-term nature of herbicide application for control, and the need to ensure that competitive species were present to slow recruitment of vulpia in any long-term management strategy, were highlighted.