Protection Quarterly Vol.11 Supplement 1 1996 195 of wheat and annual ryegrass. Austral- ian Journal of Agricultural Research 31, 649-58. Ecology of annual ryegrass Riley, I.T. and Gill, G.S. (1994). Relation- ship between herbicide resistance in Gurjeet S. Gill, Co-operative Research Centre for Weed Management Systems, rigidum and populations of Department of Agronomy and Farming Systems, The University of Adelaide, Anguina funesta, the vector in Roseworthy Campus, Roseworthy, South Australia 5371, Australia. annual ryegrass toxicity. Australian Jour- nal of Experimental Agriculture 34, 633-5. Wells, G.J. and Lyons, F.J. (1979). Weed Summary ungrazed pasture but a combination of survey in Victorian cereal crops, 1977. Annual ryegrass (Lolium rigidum heavy grazing in spring and spray-top- Victorian Department of Agriculture Gaudin) is a major weed of cropping in ping with paraquat caused a 92% reduc- and Rural Affairs, Research Project southern Australia. It is a prolific seed tion in the seed production of L. rigidum No. 66. producer and figures as high as 45 000 (2000 seeds m-2). L. rigidum sprayed with seeds m-2 have been reported. L. rigidum selective herbicides in wheat and lupins seeds have a short period (about two still managed to produce 2500–9000 seeds months) of innate dormancy. A propor- m-2 (Davidson 1990). Although this might tion of the seed bank can have dark-dor- be substantially lower than the unsprayed mancy which has been suggested as a fac- controls, it is still high enough to cause tor contributing to short term persistence problems in the following crop and lock of buried seeds of L. rigidum in crop-pas- the farmer into using selective herbicides ture rotations. Seed by environment in- every year. teractions allow a large proportion of the seed bank to survive until well after crops Seed dormancy and germination have been sown. Depending on its den- As in most other grass of Mediter- sity and time of emergence, L. rigidum ranean origin, L. rigidum seeds are dor- can be highly competitive with the crop mant at the time of production although and the competition can start as early as this period of innate dormancy is fairly the two-leaf stage of the crop. short. Gramshaw (1972) reported a rapid alleviation of seed dormancy within nine Introduction weeks of harvest. Annual ryegrass (Lolium rigidum Gaudin) Gramshaw (1972) also failed to detect was sown in pastures (and even any significant differences in seed dor- undersown in crops) over large areas of mancy after 18 weeks storage, between southern Australia. Kloot (1983) docu- eight accessions of L. rigidum grown at mented occurrence of this species in all two common sites. However, the acces- Australian states and on many off-shore sions differed significantly in their level of islands. L. rigidum is an outcrossing diploid dark dormancy (7–22% range). Subse- (2n=14) whose wide adaptability and vari- quent studies of Gramshaw and Stern able morphology arise from its genetic in- (1977b) showed that the dark-dormant stability (Cariss 1962, Kloot 1983). L. component can be substantial, at 10–20% rigidum is arguably the most important of the total seed population. This compo- weed of agriculture in Australia, infesting nent of the L. rigidum seed bank may be crops on a wide variety of soil types and important in short term field persistence climatic conditions. Millions of dollars are of seeds buried in the soil. spent annually on herbicides for L. rigidum Burial of L. rigidum seeds by cultivation control. An estimate in 1989 indicated that can affect their germination and establish- in excess of $33 million was spent on herbi- ment. Smith (1968) and Pearce and cides for L. rigidum control in Western Quinlivan (1971) found optimum germi- Australia alone (Madin personal commu- nation and seedling emergence (70-90%) nication). from L. rigidum seeds buried near the sur- This paper will review different aspects face of the soil, at about 2 cm depth. of the ecology of L. rigidum that makes it Gramshaw and Stern (1977b) also re- such a successful weed of field crops in ported a decrease in the germination of L. southern Australia. The biology of the rigidum with increasing burial depth, with weed was last reviewed by Monaghan complete inhibition at 11–14 cm. How- (1980) and the major development since ever, these seeds germinated readily when then has been the evolution of resistance brought up to a depth of 2 cm without soil in L. rigidum to a range of selective herbi- disturbance. This indicated enforced dor- cides (Powles and Howat 1990, Gill 1995). mancy possibly due to an unfavourable gaseous environment in the deeper soil Seed production layers. There are few published data on the fecun- Like other plant species, germination in dity of L. rigidum. Rerkasem et al. (1980a), L. rigidum is highly responsive to tempera- under irrigated conditions, reported seed ture. Under constant temperature re- production figures of 31 000–45 000 seeds gimes, Gramshaw (1976) reported a sharp m-2 when growing in a wheat crop. increase in the rate of germination with Davidson (1990) recorded L. rigidum seed increase in temperature from 8–26°C and set of nearly 26 000 seeds m-2 in an a slight reduction in the rate with further 196 Plant Protection Quarterly Vol.11 Supplement 1 1996 temperature increases to 35°C. Optimum held in seed heads and at break of season, Phenology temperature for the germination of after- this component of L. rigidum seed was According to Aitken (1966), the major in- ripened seeds of L. rigidum is different in found to germinate more slowly than fluence on the rate of development of L. light and dark; in the light the optimum seeds shed to the soil surface (Gramshaw rigidum was a cold requirement (vernaliza- was 27°C, whereas the optimum in the and Stern 1977a). Even in a cropped pad- tion) for flower initiation. Plant develop- dark was much lower at about 11°C dock, a considerable proportion (16%) of ment was most rapid after sowing in the (Gramshaw 1976). Furthermore, ryegrass seed can still be attached to bro- cooler part of the year, when time to head- germinability of L. rigidum seeds at alter- ken segments of seed heads (Gill unpub- ing was only one-half to one-third as long nating temperatures in darkness has been lished data). Swathing the crop a few as when sown in summer. In an earlier shown to depend solely on the lower tem- weeks earlier reduced this component of study in Adelaide, Silsbury (1964) also perature (Gramshaw 1976). In contrast, ryegrass seed bank to 4% of the total. showed that the heading (spike emer- light and alternating temperatures ap- According to Gramshaw and Stern gence) of L. rigidum was related mainly to peared to interact to increase (1977a), seedling emergence of L. rigidum the response of various cultivars to the germinability, although the highest in autumn is regulated mainly by the in- cold treatment. Long photoperiod has also germinability occurred only when the terrelationship between the germination been shown to hasten the rate of develop- maximum temperature was close to the rate of the seed population, the occurrence ment of L. rigidum , both before and optimum constant temperature, i.e. 27°C of summer/early autumn rains, and the after flower initiation (Aitken 1966). The (Gramshaw 1976). The lower temperature period for which favourable moisture winter level of photoperiod delayed requirement for germination in the dark is conditions prevail at the soil surface after flower initiation even if seed was vernal- likely to reduce depletion of the buried rains begin in autumn. In the field, tem- ized before sowing in winter. Because of seed bank of L. rigidum following rainfall perature and light appear to be relatively its short cold requirement and low critical events in summer. unimportant in influencing germination at photoperiod, L. rigidum is well suited to Rainfall events during late summer and the break of season. the short and variable growing season of autumn also have a major influence on the In the field, L. rigidum appears to have a the wheat belt (Aitken 1966). germination behaviour of L. rigidum and protracted emergence pattern. Farmers possibly, on the success of pre-sowing generally do not view delaying sowing as Crop-weed competition control measures. In the cereal belt of an effective strategy against L. rigidum. It L. rigidum can cause large yield losses in Western Australia, one or two substantial is possible that due to its prolific seed pro- wheat (Poole and Gill 1987). In a heavily falls of rain (some 10 mm or more) com- duction capability, a substantial density of infested crop, L. rigidum can compete with monly occur in summer to early autumn. seed is still present even after a large pro- wheat for nitrogen as early as the two-leaf Such rainfall events, although insufficient portion of the seed bank has been ex- stage of crop growth (Smith and Levick to cause germination, do nevertheless hausted by germination prior to sowing. 1974). Time of sowing can have a marked markedly increase the potential germina- As McGowan (1970) found, 12–23% of sea- effect on the competitive ability of L. tion rate of L. rigidum seeds (Gramshaw sonal germination of L. rigidum seed oc- rigidum in wheat (Reeves 1976a), with later 1972, Cocks and Donald 1973). Later stud- curred after June, by which time most sown crops suffering a greater percentage ies of Lush et al. (1981) showed that a single crops would have been sown. The logistics loss in yield. hydration-dehydration cycle approxi- of sowing programs on large farms are Rerkasem et al. (1980a) used the replace- mately doubled the speed of germination such that most farmers would not be pre- ment series of de Wit, and concluded that of L. rigidum by shortening the lag phase pared to wait this long to achieve better increasing the density of wheat does little of germination. In years with a relatively weed control. Furthermore, missed early to overcome the effect of L. rigidum com- dry summer and autumn, germination of sowing opportunities, just to control L. petition on the yield of wheat. Such a re- L. rigidum before seeding may be quite rigidum, can result in large loss of income sult is however, contrary to the conclu- low and protracted. in some seasons. sions of Medd et al. (1985), who suggested Tillage regime of the seed-bed can also increasing crop density as a means for re- Seedling emergence in the field influence the germination pattern of weed ducing competition from L. rigidum. Our McGowan (1970) recorded successive ger- seeds. Davidson (1994) reported a faster unpublished data from Western Australia mination of L. rigidum during autumn and rate of L. rigidum emergence in plots that over the last two growing seasons has also early winter in an undisturbed pasture and received no cultivation as compared to shown considerable benefits of increasing also when the surface soil was cultivated. plots cultivated with various tyne and disc crop density on the grain yield of wheat Measurements in two seasons showed implements. These results contradict those and reduced shoot biomass and seed pro- that 75–80% of the total seasonal germina- of Smith (1968), who had earlier reported duction of L. rigidum. Altering the spatial tion had occurred after the first two sub- a faster rate of germination of L. rigidum arrangement of the crop however, did not stantial falls of rain (>20 mm) at the begin- when seeds were covered with soil as have a significant impact on competition ning of the season. Residual viable seed compared to those on the soil surface. Gill with L. rigidum during a three year study under the undisturbed pasture in late win- (unpublished data) showed large effects in central western New South Wales ter, represented less than 1% of the total of different tillage treatments on the es- (Medd et al. 1985). seasonal germination. There appeared to tablishment of L. rigidum in wheat on a Time of emergence and establishment be short-lived dormancy in the seed popu- sandy loam soil in Western Australia. All of L. rigidum relative to the crop is also lation and this could be a significant factor treatments that received cultivation, either likely to be important in determining the in the survival of some seed through culti- before sowing or with the sowing opera- outcome of competition between the two vations prior to cropping in the cereal-pas- tion, showed much faster and greater species. Rerkasem et al. (1980b) found the ture rotation. final emergence than the uncultivated competitive ability of L. rigidum to be low The grazing history of paddocks over plots. There was also an indication of when it germinates later than wheat. summer was found to influence the ger- greater weed seed carryover in undis- In a preliminary investigation, Reeves mination of L. rigidum after the opening turbed plots. With the growing trend to- (1976b) failed to detect any difference be- rains in autumn; more seeds germinating wards no-till farming, it is important to in- tween four wheat genotypes in their com- in pastures which were heavily grazed vestigate the ecology of L. rigidum, and petitive ability against L. rigidum. Subse- (Gramshaw and Stern 1977a). Heavy graz- other important weeds, under different quently, however, Reeves and Brooke ing reduced the amount of seed firmly tillage regimes. (1977) reported differences between Plant Protection Quarterly Vol.11 Supplement 1 1996 197 wheat cultivars in their ability to compete bank can persist until the break of season. leporinum Link. and Lolium rigidum with L. rigidum, but they could not corre- Reported data indicate that between 40 Gaud.). Australian Journal of Agricultural late this with differences in height, tillering and 80% of viable seed may produce seed- Research 24, 1-10. or dry matter accumulation between the lings at the break of season (Gramshaw Davidson, R.M. (1990). Management of varieties. Later, Lemerle et al. (1979) and Stern 1977a). The residual herbicide resistant annual ryegrass, showed triticale (Triticum × Secale) to be ungerminated (but viable) seeds (20–60%) Lolium rigidum, in crops and pastures. less sensitive to competition from L. may cause major reductions in crop yields, Proceedings of the 9th Australian Weeds rigidum than wheat. Poole (1979) com- especially when seed banks are large. Conference, Adelaide, pp. 230-3. pared barley and wheat for competitive- The inability of all viable seeds to germi- Davidson, R.M. (1994). Biology and con- ness with L. rigidum and found that barley nate at the break of season may be attrib- trol of herbicide resistant Lolium rigidum. was affected less than wheat at equivalent uted partly to the presence of intact seed M.Ag.Sc. thesis, La Trobe University, densities of the weed. More vigorous heads, particularly in pasture that has been Bundoora. tillering and prostrate growth habit of bar- leniently grazed in spring and summer, Gill, G.S. (1995). Development of herbicide ley may explain the difference. and partly to the moisture status of seeds resistance in annual ryegrass Reeves (1976a) reported a consistent re- that are in contact with the soil. Gramshaw populations (Lolium rigidum Gaud.) in lationship between L. rigidum density and and Stern (1977a) found that, at the break the cropping belt of Western Australia. yield loss in his trials and suggested that it of season, soil moisture was sufficient for Australian Journal of Experimental Agri- was possible to predict yield loss in L. only 2–3 days during the seasonal breaks culture 35, 67-72. rigidum infested crops. Gill and Poole monitored, and this clearly precluded ger- Gill, G.S. and Poole, M.L. (1986). Variation (1986) found some evidence to suggest mination of some seeds. Even in cropped in the competitive ability of annual that prolonged dry soil surface conditions paddocks, it is common to find segments ryegrass in wheat crops of Western Aus- after seeding may retard establishment of of spikes on the soil surface after the har- tralia. Australian Weeds Research News- L. rigidum relative to wheat and therefore vesting operation. This component of the letter 35, 15-9. reduce its competitive effect on the crop. seed bank may be partly responsible for Gramshaw, D. (1972). Germination of an- Such dry conditions are not uncommon in extending the period of germination of L. nual ryegrass seeds (Lolium rigidum barley growing areas of Spain where L. rigidum seeds after the break of season. Gaud.) as influenced by temperature, rigidum seedlings in some seasons estab- light, storage environment, and age. lished 3–4 weeks after the crop resulting in Future research Australian Journal of Agricultural Research little or no detectable yield loss (Izquierdo • Germination, emergence and persist- 23, 779-87. personal communication). ence of the L. rigidum seed bank under Gramshaw, D. (1976). Temperature/light Narrow-leaf lupins (Lupinus different tillage regimes. interactions and the effect of seed source angustifolius) are also affected by competi- • Lolium rigidum seed management at har- on the germination of annual ryegrass tion from L. rigidum. Allen (1977) reported vest – seed capture, how can it be im- (Lolium rigidum Gaud.) seeds. Australian large yield reductions in lupins from L. proved e.g. swathing, header trail man- Journal of Agricultural Research 27, rigidum emerging before and with the agement (burning, non-selective spray 779-86. crop. L. rigidum seeds continued to emerge with a single nozzle boom). Gramshaw, D. and Stern, W.R. (1977a). as late as six weeks after the crop, but were • Seed-shedding in L. rigidum – genetic Survival of annual ryegrass (Lolium not competitive with the crop and failed to variability. rigidum Gaud.) in a mediterranean type cause any yield reduction (Allen 1977). In a • Improving crop competitiveness – environment. I. Effect of summer graz- later study, Arnold et al. (1985) also found cultivars, agronomy. ing by sheep on seed numbers and seed L. rigidum to be highly competitive with • Other non-chemical techniques to mini- germination in autumn. Australian Jour- lupins. Ryegrass density of 39 plants m-2 mize seed production/input into the nal of Agricultural Research 28, 81-91. was found to reduce lupin grain yield by seed bank. Gramshaw, D. and Stern, W.R. (1977b). 15%. In this study of Arnold et al. (1985), Survival of annual ryegrass (Lolium lupin biomass production was unaffected References rigidum Gaud.) seed in a Mediterranean by ryegrass until October and reduction in Aitken, Y. (1966). Flowering responses of type environment. II. Effects of short- yield appeared to be due to lower pod-set crop and pasture species in Australia. I. term burial on persistence of viable on the main stem and lateral branches. It is Factors affecting development in the seed. Australian Journal of Agricultural possible that ryegrass was competing with field of Lolium species (L. rigidum Gaud., Research 28, 93-101. lupins for available soil moisture or the L. perenne L., L. multiflorum Lam.). Aus- Kloot, P.M. (1983). The genus Lolium in lower pod-set could also have been due to tralian Journal of Agricultural Research 17, Australia. Australian Journal of Botany 31, shading of lupins by ryegrass. 821-39. 421-35. Weed density-yield loss relationships Allen, J.M. (1977). Weeds in grain-lupins. I. Lemerle, D., Michael, P.W. and Sutton, are guides to assist on-farm decision mak- The effect of weeds on grain-lupin B.G. (1979). The competitive abilities of ing. Economic threshold density of a weed yields. Australian Journal of Experimental wheat and triticale against different den- is not a single fixed value; it can vary con- Agriculture and Animal Husbandry 17, sities of Lolium rigidum. Proceedings of siderably due to changes in factors such as 112-7. the 7th Asian-Pacific Weed Science Soci- grain yield, price of grain and cost of weed Arnold, G.W., Weeldenberg, J. and ety Conference, Sydney, pp. 447-50. control. Consideration of weed infesta- Grassia, A. (1985). Competition between Lush, W.M., Groves, R.H. and Kaye, P.E. tions in the future and costs incurred to Wimmera ryegrass and narrow-leafed (1981). Pre-sowing hydration-dehydra- control them, can further reduce the lupins. Australian Journal of Experimental tion treatments in relation to seed ger- threshold weed density. Therefore, efforts Agriculture 25, 824-31. mination and early seedling growth of to fine-tune weed density-yield loss rela- Cariss, H.G. (1962). Wimmera ryegrass – wheat and ryegrass. Australian Journal of tionships to the last kilo are not likely be major pasture grass of the cereal and Plant Physiology 8, 409-25. rewarding. sheep areas. Journal of Agriculture, West- McGowan, A.A. (1970). Comparative ger- ern Australia (4th series) 3, 854-60. mination patterns of annual grasses in General discussion Cocks, P.S. and Donald, C.M. (1973). The north-eastern Victoria. Australian Jour- In spite of its short innate dormancy, a germination and establishment of two nal of Experimental Agriculture and Ani- substantial fraction of the L. rigidum seed annual pasture grasses (Hordeum mal Husbandry 10, 401-4. 198 Plant Protection Quarterly Vol.11 Supplement 1 1996 Medd, R.W., Auld, B.A., Kemp, D.R. and Murison, R.D. (1985). The influence of wheat density and spatial arrangement Cultural management of annual ryegrass on annual ryegrass, Lolium rigidum Gaudin, competition. Australian Journal J.M. Matthews, Co-operative Research Centre for Weed Management Systems, of Agricultural Research 36, 361-71. Faculty of Natural Resources, University of Adelaide, Glen Osmond, South Monaghan, N.M. (1980). The biology and Australia 5045, Australia. control of Lolium rigidum as a weed of wheat. Weed Research 20, 117-21. Pearce, G.A. and Quinlivan, B.J. (1971). The Introduction control of annual (‘Wimmera’) ryegrass The availability of both selective and 1996, Gill 1995, Pratley et al. 1993). in cereal crops. Journal of Agriculture, non-selective herbicides during the last Changes in cropping systems, namely, in- Western Australia (4th series) 12, 58-62. twenty years has provided a range of her- creased fertilizer use, different crop varie- Poole, M.L. (1979). Competition between bicides to manage ryegrass in crop and ties and crop species along with minimum crops and ryegrass. Proceedings of the pasture paddocks with a consequent de- tillage systems have accentuated the de- Western Australian Weeds Conference, cline of alternative management strate- velopment of herbicide resistance in an- Muresk, pp. 95-7. gies. However, strategies for the manage- nual ryegrass. Cultural methods of Poole, M.L. and Gill, G.S. (1987). Competi- ment of annual ryegrass (Lolium rigidum ryegrass management have become im- tion between crops and weeds in south- Gaudin) without the use of herbicides portant for the prevention and manage- ern Australia. Plant Protection Quarterly have been re-focused in the last decade by ment of herbicide resistance in ryegrass. 2, 86-96. the appearance of populations of annual This review of the cultural management of Powles, S.B. and Howat, P.D. (1990). Her- ryegrass that are not controlled by the use ryegrass assumes that herbicide resistant bicide-resistant weeds in Australia. Weed of selective herbicides. Annual ryegrass is ryegrass is similar to susceptible ryegrass Technology 4, 178-85. widely distributed across the southern in germination, growth and seed output Reeves, T.G. (1976a). The effect of annual Australian cropping regions (Mayfield and (Gill et al. 1996, Matthews unpublished ryegrass (Lolium rigidum Gaud.) on yield Edwards 1992) as are the sites with herbi- data). of wheat. Weed Research 16, 57-63. cide resistant populations (Nietschke et al. Reeves, T.G. (1976b). The effect of growth Ryegrass management in pastures habit on the competition between wheat and annual ryegrass. Proceedings of the Grazing 5th Australian Weeds Conference, Mel- 40,000 Ryegrass is usually reduced

bourne, pp. 4-11 to 4-12. ) in density by grazing over a -2 Reeves, T.G. and Brooke, H.D. (1977). The 30,000 number of seasons. The seed effect of genotype and phenotype on heads are preferentially the competition between wheat and an- grazed by sheep and cattle nual ryegrass. Proceedings of the 6th 20,000 and seed return to the pad- Asian-Pacific Weed Science Society Con- dock is thereby reduced. The ference, Jakarta, Indonesia, pp. 167-72. Tyne grazing intensity needs to be Rerkasem, K., Stern, W.R. and Goodchild, 10,000 high to effect a net reduction Seed bank (seeds m Min.till N.A. (1980a). Associated growth of wheat Control in seed return in any one sea- and annual ryegrass. I. Effect of varying 0 son (Pearce 1975). With in- total density and proportion in mixtures Year 0 Year 1 Year 2 sufficient grazing pressures of wheat and annual ryegrass. Australian other grasses viz. Vulpia spp., Journal of Agricultural Research 31, 649-58. Figure 1. Comparison of seed bank decline Hordeum spp. and Bromus Rerkasem, K., Stern, W.R. and Goodchild, attributed to tillage method at sowing and spp. will increase due to the N.A. (1980b). Associated growth of unpalatability of seeds from compared to a no cultivation control (redrawn wheat and annual ryegrass. II. Effect of these species. Grazing alone varying the time of ryegrass germina- from Davidson 1994). has been largely superseded tion in stands of wheat. Australian Jour- 100 nal of Agricultural Research 31, 659-72. Silsbury, J.H. (1964). The effect of vernalization on the heading of 80 Susceptible ryegrass Wimmera ryegrass (Lolium rigidum) and on five cultivars of perennial ryegrass Resistant ryegrass (). Australian Journal of Ex- 60 perimental Agriculture and Animal Hus- bandry 4, 352-6. Smith, D.F. (1968). The growth of barley 40

grass (Hordeum leporinum) in annual pas- Emergence (%) ture. I. Germination and establishment in comparison with other annual pas- 20 ture species. Australian Journal of Experi- mental Agriculture and Animal Husbandry 8, 478-83. 0 Smith, D.F. and Levick, G.R.T. (1974). The May June July Aug. Sept. Oct. Nov. Jan. Mar. April May June April-July effect of infestations by Lolium rigidum Year 1 Year 2 Year 3 Gaud. (annual ryegrass) on the yield of wheat. Australian Journal of Agricultural Figure 2. Emergence of susceptible and herbicide resistant ryegrass over a Research 25, 381-93. three year period, from Heap (1988).