The effect of defoliation practice in Western Australia on tiller development of annual ryegrass (Lolium rigidum) and Italian ryegrass (Lolium multiflorum) and its association with forage quality

The effect of defoliation on the vegetative, early reproductive and inflorescence stages of tiller development, changes in the dry‐matter yield of leaf, stem and inflorescence and the associated changes in forage quality was determined on plants of annual ryegrass (Lolium rigidum Gaud.) and Italian ryegrass (L. multiflorum Lam.). The field study comprised seventy‐two plots of 1 m × 2 m, sown with one annual ryegrass and seven Italian ryegrass cultivars with a range of heading dates from early to late; defoliation commenced 6 weeks after germination. During the vegetative stage of growth, plots were defoliated when the tillers had three fully expanded leaves (three‐leaf stage). During the early reproductive stage of growth, to simulate a cut for silage, plots were defoliated 6–7 weeks after 0·10 of the tillers displayed nodal development. The subsequent regrowth was defoliated every 3 weeks. Assessments of changes in tiller density, yield and quality were made in the growth cycle that followed three contrasting cutting treatments during the winter–spring period (from 10 July). In treatment 1, this growth cycle (following closing‐up before a subsequent conservation cut) commenced on 7 August following two defoliations each taken when the tillers were at the three‐leaf stage. In treatment 2, the growth cycle commenced on 16 October following: for early‐maturing cultivars, two cuts at the three‐leaf stage, a cut for silage and an additional regrowth cut; for medium‐maturing cultivars three cuts at the three‐leaf stage and a cut for silage; and late‐maturing cultivars, five cuts at the three‐leaf stage. In treatment 3, defoliation up to 16 October was as for treatment 2, but the growth cycle studied started on 27 November following two additional regrowth cuts for early‐ and medium‐maturing cultivars and cut for silage for the late‐maturing cultivars. Tiller development for all cultivars was classified into three stages; vegetative, early reproductive and inflorescence. In treatment 1, in vitro dry‐matter digestibility (IVDMD) and crude protein (CP) content were negatively associated with maturation of tillers. IVDMD ranged from 0·85 to 0·60 and CP ranged from 200 to less than 100 g kg^–1 dry matter (DM) during the vegetative and inflorescence stages respectively. This large reduction in forage quality was due to an increase in the proportion of stem, inflorescence and dead material, combined with a reduction in the IVDMD and CP content of the stem. A high level of forage quality was retained for longer with later‐maturing cultivars, and/or when vegetative tillers were initiated from the defoliation of early reproductive tillers (treatments 2 and 3). However, 15 weeks after the closing‐up date in treatment 1, defoliation significantly reduced the density of inflorescences with means (±pooled s.e_m.) of 1560, 1178 and 299 ± 108 tillers m^–2, and DM yield of inflorescence with means of 3·0, 0·6 and 0·1 ± 0·15 t ha^–1 for treatments 1, 2 and 3 respectively. This study supports the recommendation that annual and Italian ryegrass cultivars should be classified according to maturity date based on the onset of inflorescence emergence, and that the judicious defoliation of early reproductive tillers can be used to promote the initiation of new vegetative tillers which in turn will retain forage quality for longer.     

Effect of dairy slurry application rate and forage type on production,soil nutrient status and nitrogen‐use efficiency

Seven forage types (diploid and tetraploid perennial ryegrass, Italian ryegrass and hybrid ryegrass, a low‐input mixture of perennial ryegrass, cocksfoot, timothy and meadow fescue, a mixture of perennial ryegrass and white clover, and monoculture of red clover) were sown in late July 2004. Each received one of four rates of dairy cattle slurry in three annual applications by trailing shoe, which supplied average nitrogen (N) inputs of 0·0, 114·9, 204·8 and 301·2 kg N ha^?1 annum^?1. Treatments were cut either three or four times annually over four years. Average dry‐matter yield (DM) response to slurry N was 15·6 kg DM kg^?1 N. Lowest recovery of slurry N was in the second application each year (after first cut). The data suggest that slurry applied to Italian ryegrass, and also to swards containing legumes on soils with high phosphorus content, will produce a lower DM response to slurry N and result in a lower slurry N recovery than on swards of perennial ryegrass or cocksfoot‐dominant low‐input mixtures. Apparent recovery of slurry N was low at the second cut, especially when first‐cut yields had been high. To maximize slurry N recovery, application to regrowths with potentially slow rates of growth or high legume content should be avoided.     

Herbage yield, tiller number and root system activity after defoliation of prairie grass (Bromus catharticus Vahl)

The effects of defoliation upon root and shoot systems of prairie grass (Bromus catharticus Vahl) were examined in both field and pot studies. The varieties used were 78–32 (HY), a high-yielding variety; 79–42 (LY), a low-yielding variety; and the commercial variety Grasslands Matua. In the field, the presence of roots in early and late spring was estimated by measuring uptake of [³²P]phosphate by roots; herbage yields and tiller numbers were recorded. In a pot study, root and shoot dry-matter (DM) yields were analysed. In the field, roots were detected to a depth of 1·2 m. After defoliation to a height of 0·1 m, root presence decreased more than 50% at depths of 0·6 m for LY and 1·0 m for Matua in early spring and at several depths for each variety in late spring. After a second defoliation, the apparent growth rate of shoots decreased by 35% in relation to the first regrowth period. In pots, shoot DM and root DM of control plants (undefoliated) had the following allometric relationship of the form: In (shoot DM) = 0·61 + 1·14 ln (root DM) (r²= 0·81). After defoliation, compared with undefoliated controls, the relative growth rate of shoots and total herbage yields were higher, but root and stubble DM were lower in all three varieties. Pooled root DM means were 10·3 and 6·8 g plant^?1 and pooled stubble DM means were 12·7 and 7·6 g plant^?1 for control and defoliated plants respectively. HY produced heavier tillers than LY, pooled means being 0·94 and 0·53 g DM tiller^?1 (field study) and 3·44 and 2·05 g DM tiller^?1 (pot study) for HY and LY respectively. HY had 5–6 green leaves per tiller, whereas LY had 3–4. Developed green leaves were heavier in HY (58 g m^?2) than in LY (48 g m^?2). It is suggested that differences in both leaf parameters may be related to higher herbage yields for HY than LY.     

Nitrogen and winter cover crop effects on spring and summer nutrient uptake

Overseeded winter annuals in bermudagrass [Cynodon dactylon (L.) Pers.] improve annual dry‐matter (DM) yield and capture nutrients in fields receiving manure application. This study determined the DM and nutrient uptake responses of annual ryegrass (Lolium multiflorum L.), cereal rye (Secale cereale), berseem clover (Trifolium alexandrinum L.) and bermudagrass‐winter fallow to 0, 50, 100 and 150 kg N ha^?1 applied approximately 2 months before a single spring harvest, and in addition to swine‐effluent N (258 and 533 kg ha^?1 in summer 2000 and 2001, respectively). Under drought conditions in 2000, DM yield at the spring harvest was highest in ryegrass, and summer DM yield of bermudagrass was greater at 100 and 150 kg N ha^?1 than 50 kg N ha^?1(P < 0·05). The concentration and uptake of N at the spring harvest increased linearly across N rates in both years (P < 0.05). Cover crops differed in N uptake in 2000 (P < 0.01) and values ranged from approximately 141 kg N ha^?1 in berseem clover to 86 kg N ha^?1 in rye. Per unit of N applied, uptake of N increased by approximately 0·409 kg ha^?1 in 2000 and 0·267 kg ha^?1 in 2001; uptake of P increased by 0·029 and 0·014 kg ha^?1 respectively. In 2000, uptake of P was responsive to N rate and this relationship was significant (P < 0·01) in winter fallow (slope = 0·032) and ryegrass (slope = 0·057). Increased uptake of N and P at the single spring harvest was due mainly to higher concentrations in herbage and not higher DM yield.     

Optimizing legume content and forage yield of mown white clover–Italian ryegrass mixtures through nitrogen fertilization and grass row spacing

Mixed swards of white clover–grass mixtures in highly productive environments often fail to reach the minimum recommended annual clover proportion of about 0·30. This study assessed the effect on clover content and total dry matter (DM) yield of two spring N applications (0 and 45 kg N ha^?1) and two distances between drilled grass‐rows (0·18 and 0·36 m) over 3 years for mown swards of white clover–Italian ryegrass (Trifolium repens–Lolium multiflorum) in binary mixtures in northern Italy. An additional aim was to determine the advantage of association of grass–clover compared with grass and clover monocultures. On average, N fertilization of mixtures resulted in almost 9% higher total yield (P < 0·01; mean response = 18·1 kg of total DM per kg of N) but decreased the clover proportion (0·250 vs. 0·312). Wider grass‐row spacing increased clover proportion (0·327 vs. 0·234; P < 0·01) with no reduction of total DM yield. N fertilization × grass‐row spacing interaction occurred only for clover content (P < 0·01). Without N fertilization, mixtures out‐yielded clover and grass pure stands. With N fertilization, at double rate to pure grass, yields from mixtures were greater than from clover and comparable to Italian ryegrass.     

In vitro rumen methane output of perennial ryegrass varieties and perennial grass species harvested throughout the growing season

The selection and feeding of perennial ryegrass (Lolium perenne L.) varieties (PRV) or perennial grass species (PGS) may affect enteric methane (CH₄) output because of changes in the fermentation dynamics in the rumen as a result of differences in herbage chemical composition. The objective of this study was to determine the effects of PRV and PGS harvested throughout the growing season on herbage chemical composition, and in vitro rumen fermentation variables and CH₄ output per unit of feed using a batch culture technique. Seven PRV (Experiment 1: Alto, Arrow, Bealey, Dunluce, Greengold, Malone, Tyrella) and six perennial grasses [Experiment 2: perennial ryegrass (Navan), perennial ryegrass (Portstewart), cocksfoot, meadow fescue, tall fescue, timothy; defined as PGS], managed under a simulated grazing regime, were incubated for 24 h with buffered rumen fluid in two separate experiments. The CH₄ output per unit of feed dry‐matter (DM) incubated was not affected (P > 0·05) by PRV (range of mean values across PRV of 23·9–25·3 (SEM 0·41) mL g^?1 DM) or by PGS (25·6–26·6 (SEM 0·37) mL g^?1 DM). The CH₄ output per unit feed DM disappearing during the in vitro rumen incubation was not affected by PRV (33·9–35·1 (SEM 0·70) mL g^?1 DM), and although there was an overall PGS effect (P < 0·05; 37·2–40·3 (SEM 0·71) mL g^?1 DM), none of the paired contrasts between PGS were significant when analysed using Tukey adjusted comparisons. This outcome reflected either small‐scale or a lack of treatment effects on individual herbage chemical composition (e.g. 454–483 g NDF kg^?1 DM, 215–224 g CP kg^?1 DM and 94–122 g water‐soluble carbohydrate (WSC) kg^?1 DM across PRV; 452–506 g NDF kg^?1 DM, 208–243 g CP kg^?1 DM and 73–131 g WSC kg^?1 DM across PGS) and in vitro rumen fermentation variables. Hence, these results provide no encouragement that choices among the grasses examined, produced within the management regimes operated, would reduce enteric CH₄ output per unit of feed in vivo. However, the technique utilized did not take account of animal × PRV or PGS interactions, such as potential differences in intake between animals, that may occur under farm conditions.     

Characteristics of baled silage made from first and second harvests of wilted and severely wilted forages

First and second harvests of lucerne (Medicago sativa L.), perennial ryegrass (Lolium perenne L.) and a lucerne–perennial ryegrass mixture [80 or 144 g kg^?1 dry matter (DM) of ryegrass] at the first and second harvests were cut and conditioned, wilted to 500 or 700 g DM kg^?1 then baled and stretch‐wrapped for silage on the same dates. Lucerne bales were denser (411 kg m^?3) than bales of perennial ryegrass (331 kg m^?3) (P < 0·05). After an 8‐month storage period, silage made from high DM‐content forage had a higher concentration of neutral‐detergent fibre (NDF) and was less digestible than that made from low DM‐content forage. Daily DM intakes by beef steers, when the silages of the second harvest were fed ad libitum, were 31·2, 31·2 and 22·3 g kg^?1 live weight for lucerne, lucerne–perennial ryegrass mixture and perennial ryegrass silages, respectively (P < 0·01), when the herbage had been wilted to 500 g kg^?1. In vivo digestibility of NDF in the lucerne–perennial ryegrass mixture silage (0·587) was significantly lower than that of perennial ryegrass silage (0·763) but higher than lucerne silage (0·518). Higher intakes of baled lucerne silage tended to offset its lower digestibility values. Lucerne–perennial ryegrass mixture silage had a higher DM and NDF digestibility than lucerne silage, indicating perhaps the presence of associative effects.     

Effect of Lolium perenne sward density on productivity under simulated and actual cattle grazing

The objective of this study was to quantify the effect of perennial ryegrass (Lolium perenne L.; PRG) sward density on seasonal and total DM yield under simulated grazing and animal grazing by cattle, and to assess the effectiveness of visually estimated ground scores (GSs) for predicting sward PRG density. The study incorporated five different seeding rates of PRG, each replicated three times, to simulate swards ranging in PRG density typical of different ages and conditions. There was no significant difference between defoliation managements for total DM yield, but sward PRG density had a significant effect on both the seasonal and total DM herbage yield under both systems. Under simulated grazing, total DM yield ranged from 10·7 to 12·0 t DM ha^?1 with increasing sward PRG density at a GS range of 1·70–4·28 (mean of 2 years’ data, P < 0·01). Under animal grazing, the yield range was from 10·3 to 12·2 t DM ha^?1 for a GS range of 1·50–3·39 (mean of 2 years’ data, P < 0·01). The largest differences in DM yield occurred during the spring period. The relationship between sward DM yield and GS was significant (P < 0·001) for both simulated and animal‐grazed swards. Each unit increase in midseason GS (June) related to an average yield increase of 350 kg DM ha^?1 under simulated grazing and a 721 kg DM ha^?1 increase under cattle grazing. Every unit increase in the GS at the end of the grazing season (December) was associated with a 460 or 1194 kg DM ha^?1 increase under simulated and animal grazing, respectively. These results show that visual estimates of density were an effective tool in describing PRG density and that this could be related to DM yield potential. Further investigations may provide a threshold value below which the renewal of swards could be advised based on a visual GS of PRG.     

Selection by sheep and goats for perennial ryegrass and white clover offered over a range of sward height contrasts

The selection by sheep (six Coopworth ewe hoggets, 44·3 ± 4·6 kg live weight) and goats (six Saanen/Anglo‐Nubian yearling males, 38·1 ± 3·8 kg live weight) for perennial ryegrass (Lolium perenne) and white clover (Trifolium repens) and for sward height was measured in two experiments involving paired turves. Pairs of turves with herbage of differing height and of either the same or different plant species were offered. One sward (fixed height species, FHS) was always offered at 130 mm and the other (variable height species, VHS) at 130, 90 or 50 mm. Turves (450 mm × 220 mm) were cut to a soil depth of 100–150 mm from areas of perennial ryegrass and white clover regrown to the desired height after previously being cut to 30 mm. Each turf in a pair was weighed (±1 g) before and after grazing by penned animals maintained on a barley‐based pelleted diet. The number of prehending bites taken from each turf was recorded over a grazing period (128 ± 12 s). Bite mass, bite rate and intake rate were calculated. As the sward height of the VHS turf declined, an increasing proportion of the diet was selected from the 130 mm turf. When averaged over all height contrasts, both animal species selected a higher proportion (0·776 ± 0·026) of their diet from 130‐mm white clover than from 130‐mm perennial ryegrass (0·591 ± 0·018) turves. On average, goats selected a higher proportion (0·721 ± 0·022) of their dry‐matter (DM) intake from the 130‐mm turf than sheep (0·646 ± 0·019), but the effect was not consistent. In contrasts with perennial ryegrass as the VHS (and both perennial ryegrass and white clover as FHS), the proportion of the diet selected from the 130‐mm turf was very similar for both animal species. However, with white clover as the VHS (and both perennial ryegrass and white clover as FHS), goats selected a higher proportion of their intake from the 130‐mm turf to the extent that in the 130‐mm perennial ryegrass/50‐mm white clover contrast sheep showed as strong selection for 50‐mm white clover as goats did for 130‐mm perennial ryegrass. This lesser selection of goats for white clover as its height in a sward declines is likely to contribute to the higher white clover content observed in swards grazed by goats. Bite mass was greater on white clover (246 ± 5 mg DM bite^–1) than on perennial ryegrass (173 ± 5 mg DM bite^–1) and was greater for goats (255 ± 6 mg DM bite^–1) than for sheep (195 ± 5 mg DM bite^–1). Bite rate was greater on perennial ryegrass (45·9 ± 1·0 bites min^–1) than on white clover (39·9 ± 1·0 bites min^–1) and was greater for sheep (45·5 ± 1·1 bites min^–1) than for goats (42·5 ± 1·1 bites min^–1). Apparent intake rate by both sheep and goats was lower (mean, 5·0 ± 0·29 g DM min^–1) on 130 mm perennial ryegrass/white clover than on 130 mm perennial ryegrass/perennial ryegrass (7·0 ± 0·27 g DM min^–1), but was higher (9·62 ± 0·29 g DM min^–1) on 130‐mm white clover/perennial ryegrass than on 130‐mm white clover/white clover (8·2 ± 0·29 g DM min^–1) combinations.     

Effect of formic acid and a bacterial inoculant on the amino acid composition of grass silage and on animal performance

An Italian ryegrass and hybrid ryegrass sward was harvested on 11 May 1994. The mean dry‐matter (DM) content of the herbage was 197 g kg^–1 fresh matter (FM), and mean nitrogen and water‐soluble carbohydrate contents were 20 and 272 g kg^–1 DM respectively. Approximately 72% of total nitrogen (TN) was in the form of protein‐nitrogen. The herbage was treated with either no additive, formic acid (3·3 l t^–1) (Add‐F, BP) or inoculant (2·3 l t^–1) (Live‐system, Genus) and ensiled in 100 t silos. Changes in effluent composition with time showed that silage fermentation and protein breakdown were delayed by treatment with formic acid. Formic acid and inoculant treatments also inhibited amino acid catabolism during ensilage. All silages were well fermented at opening with pH values < 4·0 and ammonia‐N concentrations of ≤ 50 g kg^–1 TN after 120 d ensilage. Treatment had an effect on protein breakdown as measured by free amino acid concentration, with values of 21·5, 18·2 and 13·2 mol kg^–1 N at opening (191 d) for untreated, formic acid‐treated and inoculated silages respectively. Amino acid catabolism occurred to the greatest extent in untreated silages with significant decreases in glutamic acid, lysine and arginine, and increases in gamma amino butyric acid and ornithine. The silages were offered ad libitum without concentrate supplementation to thirty‐six Charolais beef steers for a period of 69 d (mean live weight 401 kg). Silage dry‐matter intakes and liveweight gains were significantly (P < 0·05) higher on the treated silages. Silage dry‐matter intakes were 7·42, 8·41 and 8·23 kg d^–1 (s.e.d. 0·27) with liveweight gains of 0·66, 0·94 and 0·89 kg d^–1 (s.e.d. 0·058) for untreated, formic acid‐treated and inoculated silage‐fed cattle respectively. In conclusion, additives increased the intake of silage and liveweight gain by the beef steers, and it is suggested that this may be caused in part by the amino acid balance in these silages.     

Effects of irrigation and harvest management on dry-matter yield and seed yield of annual clovers grown in pure stand and in mixtures with graminaceous species in a Mediterranean environment

Forage crops in Mediterranean environments are characterized by variable seed and forage production. Knowledge of the effects of agronomic factors on annual clovers grown in pure stand and in association with grasses is essential for their effective exploitation of the available environmental resources. Herbage and seed production were evaluated in southern Italy in an experiment with winter annual forage crops during the years 1992–95. Clovers (Trifolium alexandrinum L., berseem; T. incarnatum L., crimson; T. resupinatum L., Persian; and T. squarrosum L., squarrosum) and graminaceous forage crops (Hordeum vulgare L., barley; and Lolium multiflorum Lam., Italian ryegrass) were used to examine the agronomic effects of irrigation and harvest management. The clovers were evaluated in pure stand and in mixture, whereas the graminaceous species were evaluated only in mixtures. The mixtures were sown in alternating, equally spaced rows. The parameters evaluated were forage dry-matter yield, seed yield and its components. The results showed wide differences in forage production between clovers in pure stands and in binary mixtures. Mean dry-matter production from forage harvest of pure stands of irrigated clovers taken when 10–15% of the stems were flowering yielded 4·36 t ha^–1, that is 0·67 and 0·55 of that of irrigated mixtures of clovers with either Italian ryegrass or barley, respectively, harvested when 10–20% of the graminaceous components were at the heading stage. The forage yield of non-irrigated pure stands of clovers was 0·60 of that of irrigated plots, whereas non-irrigated mixtures yielded 0·82 and 0·86 of that the irrigated treatments for Italian ryegrass or barley mixtures. The mean seed yield of non-irrigated pure stands of the clovers was 0·51 of that of irrigated stands, which yielded 451 kg ha^?1. Persian clover gave the highest seed yields (732 kg ha^?1 under irrigation). These higher yields were related to a higher number of fructiferous organs per stem in Persian clover. When a forage harvest had previously been taken on irrigated clover plots, the subsequent mean seed yields were greatly reduced to 0·76 for berseem and 0·21 for Persian clover and were almost negligible for crimson and squarrosum clover. When irrigated and grown in mixtures with Italian ryegrass, only berseem produced a worthwhile yield of 0·36 compared with that of pure clover taken for seed without a forage harvest; mixtures with barley gave yields of 0·71, 1·07 and 0·20 for berseem, crimson and Persian clovers, respectively, compared with seed yield from uncut pure clover.     

Herbage productivity of prairie grass, reed canary-grass and phalaris

The herbage production and quality of swards of three grass species, prairie grass (Bromus willdenowii Kunth), reed canary-grass (Phalaris arundinacea L.) and phalaris (Phalaris tuberosa L.) were compared with perennial ryegrass (Lolium perenne L.) and hybrid ryegrass (L. perenne L. ×L. muitiflorum Lam.) under 6–cut (experiment 1) and 4–cut (experiment 2) regimes over 3 years at Ayr; annually, 360 kg ha^-1 fertilizer N were applied. At Edinburgh prairie grass was compared with Italian ryegrass (L. multiflorum Lam.), timothy (Phleum pratense L.) and cocksfoot (Dactylis glomerata L.) under an annual 4–cut regime for 3 years (experiment 3); fertilizer N application totalled 350 kg ha^-1 annually. Prairie grass gave the highest annual dry matter (DM) production at Ayr, averaging 11·99 t ha^-1 in experiment 1 and 15·62 t ha^-1 in experiment 2. Reed canary-grass was much less productive whilst phalaris did not persist after harvest year 1. On average, prairie grass gave 8–10% more DM than the three ryegrasses in the 6-cut system but its advantage was much less under the 4-cut regime. In experiment 3, the DM production of prairie grass and Italian ryegrass were similar in year 1, but following winter damage prairie grass gave the lowest production in subsequent harvest years. Prairie grass had digestibility (OMD) values lower than the ryegrass but higher than reed canary-grass, timothy and cocksfoot. The water soluble carbohydrate (WSC) concentrations in prairie grass were markedly higher than in timothy and cocksfoot but lower than those in Italian ryegrass. Prairie grass had relatively low P and Mg concentrations. Reed canary-grass had relatively low OMD and Ca, but high N, P, K and Mg contents. It is concluded that prairie grass may have potential in the UK as a special-purpose species for conservation management but mainly in the milder climatic areas. The Phalaris species evaluated had disappointing agronomic potential.     

Relationships between morphological and chemical characteristics of perennial ryegrass varieties and intake by sheep under continuous stocking management

Plots of five intermediate‐heading varieties of perennial ryegrass (Lolium perenne L.) [AberDove, Belramo and Glen (diploid); Twins (tetraploid); and AberExcel (tetraploid hybrid)] were continuously stocked with sheep to maintain a target sward surface height of 40–50 mm. Daily dry matter (DM) intake was significantly different (F‐value = 0·032) between the varieties, with the tetraploid hybrid AberExcel having the highest values for daily DM intake and intake rate during eating. Amongst the diploid varieties, intake rate tended to be higher for sheep grazing Glen. The varieties comprised a wide range in potential growth habit, from the relatively prostrate, highly tillered Glen to the more‐erect AberExcel and there were differences between them in the vertical distribution of leaves within the sward canopy. The leaves of AberExcel weighed 3·6 mg DM cm^?2 leaf area in contrast to the other varieties (4·3–5·3 mg DM cm^?2 leaf area) resulting in a high leaf area index (LAI) in relation to the green leaf mass. Intake rate was not significantly correlated with extended tiller and sheath tube lengths, partition of herbage mass, number of tillers per square metre or LAI. However, canonical variates analysis showed that there were significant differences between the varieties for the morphological and chemical factors examined. Other factors also need to be explored to explain these differences in ingestive behaviour in order to identify plant traits that are correlated with herbage intake rate. These are needed for varieties destined for grazing use, both during the breeding programme and their subsequent evaluation.     

Seed variation among annual ryegrass cultivars in south-eastern USA and the relationship with seedling vigour and forage production

Abstract Annual ryegrass (Lolium multiflorum Lam.) is grown on more than one million ha in the south‐eastern USA each year. Recommended and actual seeding rates vary substantially within the region. The objective of this study was to evaluate variation in seed weight, germination, seedling vigour and seasonal yield performance among annual ryegrass cultivars. During 1997, 1998 and 1999, seed from fourteen commercial cultivars was weighed and germinated to determine numbers of pure live seed (PLS) m^?2 before yield evaluation at four locations. Seed from ten cultivars was planted at 0·7 and 2·0 cm depth in a greenhouse study to evaluate relative seedling vigour. Cultivar mean single‐seed weight ranged from 2·4 to 4·8 mg in 1997, 1·8 to 4·5 mg in 1998, and 2·6 to 4·6 mg in 1999. Seed germination ranged from 78·8% to 98·0% in 1997, 82·3 to 98·3% in 1998 and 77·8 to 98·3% in 1999. Seed number, PLS m^?2, ranged from 675 to 1289 in 1997, 710 to 1550 in 1998, and 717 to 1179 in 1999. Among the ten cultivars evaluated for seedling vigour, seedling weight differed between planting depths and a significant cultivar by year interaction was observed. Seedling weight was highly correlated with seed weight at each seeding depth. The effect of increasing number of PLS m^?2 on subsequent yield performance, although small, was consistently negative. These results indicate that target plant populations may be obtained more economically by adjusting seeding rates for seed size differences among cultivars and seed lots of annual ryegrass.     

Increased concentration of water-soluble carbohydrate in perennial ryegrass (Lolium perenne L.): milk production from late-lactation dairy cows

Eight multiparous Holstein–Friesian dairy cows in late lactation were used to investigate the potential of using perennial ryegrass with a high concentration of water‐soluble carbohydrate (WSC) to increase the efficiency of milk production. After a pretreatment period on a common pasture, the cows were each given ad libitum access to one of two varieties of zero‐grazed grass continuously for 3 weeks. Treatments were: high sugar (HS), an experimental perennial ryegrass variety bred to contain high concentrations of WSC; or control, a standard variety of perennial ryegrass (cv. AberElan) containing typical concentrations of WSC. The two grass varieties were matched in terms of heading date. All animals also received 4 kg day^–1 standard dairy concentrate. Grass dry matter (DM) intake was not significantly different between treatments (11·6 vs. 10·7 kg DM day^–1; s.e.d. 0·95 for HS and control diets respectively), although DM digestibility was higher on the HS diet (0·71 vs. 0·64 g g^–1 DM; s.e.d. 0·23; P < 0·01) leading to higher digestible DM intakes for that diet. Milk yield from animals offered the HS diet was higher (15·3 vs. 12·6 kg day^–1; s.e.d. 0·87; P < 0·05) and, although milk constituent concentrations were unaffected by treatment, milk protein yields were significantly increased on the HS diet. The partitioning of feed N was significantly affected by diet, with more N from the HS diet being used for milk production (0·30 vs. 0·23 g milk N g^–1 feed N; s.e.d. 0·012; P < 0·01) and less being excreted in urine (0·25 vs. 0·35; s.e.d. 0·020; P < 0·01). In a separate experiment, using the same grasses harvested earlier in the season, the fractional rate of DM degradation, measured by in situ and gas production techniques, was higher for the HS grass than for the control. It is concluded that increased digestible DM intakes of the HS grass led to increased milk yields, whereas increased efficiency of utilization of the HS grass in the rumen resulted in the more efficient use of feed N for milk production and reduced N excretion.     

The impact of defoliation frequency and nitrogen fertilizer application in spring on summer survival of perennial ryegrass under grazing in subtropical Australia

Abstract This field study investigated the effect of timing of nitrogen (N) fertilizer application in spring on the survival of grazed perennial ryegrass (Lolium perenne cv. Dobson and Yatsyn) over summer in a subtropical environment. There were five N fertilizer treatments: no applied N, 46 kg N ha^?1 on 22 October or 22 November or 22 December, or on 22 October and again on 22 December. Water‐soluble carbohydrate (WSC) concentration of perennial ryegrass plants entering the summer was altered by varying defoliation frequency, with defoliation interval based on the number of leaves per tiller. Frequent defoliation was set at a regrowth level of one leaf per tiller and less frequent defoliation at a regrowth level of three leaves per tiller, over a total of two by three‐leaf per tiller regrowth periods. Application of N fertilizer was found to have no significant effect (P > 0·05) on survival of perennial ryegrass plants over summer. On the other hand, defoliation had a marked effect on perennial ryegrass persistence, with frequent defoliation decreasing ryegrass plant density (51 vs. 88 plants m^?2; P < 0·001) and increasing the density of tropical weed grasses (99 vs. 73 plants m^?2; P < 0·001) by autumn. Frequently defoliated plants had a lower stubble WSC content on a per plant basis than less frequently defoliated plants in spring (103 vs. 201 mg per plant; P < 0·001) and summer (59 vs. 101 mg per plant; P < 0·001). The lower WSC content was associated with a smaller root system in spring (1·50 vs. 2·14 g per plant; P < 0·001) and autumn (1·79 vs. 2·66 g per plant; P < 0·01), and this was reflected in 0·29 more plants being pulled from the soil by livestock between November 1996 and April 1997. Rhizoctonia fungus was associated with roots of pulled plants, but not with roots of seemingly healthy plants, indicating that this fungus may have a role in a weakened root system, which was more prone to sod pulling. Nitrogen applied in October and November resulted in a reduced WSC concentration, although the effect was restricted to 1 month after N application. The present study indicates that survival of perennial ryegrass plants over the summer in a subtropical region is prejudiced by frequent defoliation, which is associated with a lower WSC concentration and a shallower root system. Under grazing, sod pulling is a reflection of this weaker root system and contributes to plant mortality.     

Effect of defoliation management, based on leaf stage, on perennial ryegrass (Lolium perenne L.), prairie grass (Bromus willdenowii Kunth.) and cocksfoot (Dactylis glomerata L.) under dryland conditions. 2. Nutritive value

A field experiment was undertaken between April 2003 and May 2004 in southern Tasmania, Australia, to quantify and compare changes in the nutritive value of perennial ryegrass (Lolium perenne L.), prairie grass (Bromus willdenowii Kunth.) and cocksfoot (Dactylis glomerata L.) under a defoliation regime based on stage of leaf regrowth. Defoliation interval was based on the time taken for two, three or four leaves per tiller to fully expand. At every defoliation event, samples were collected and analysed for acid‐detergent fibre (ADF), neutral‐detergent fibre (NDF) and total nitrogen (N) concentrations and to estimate metabolizable energy (ME) and digestible dry matter (DDM) concentrations. Amounts of crude protein (CP) and metabolizable energy (MJ) per hectare values were subsequently calculated. There was a significantly lower (P < 0·001) NDF concentration for perennial ryegrass compared with prairie grass and cocksfoot, and a significantly lower (P < 0·001) ADF concentration for cocksfoot compared with prairie grass and perennial ryegrass, regardless of defoliation interval. The CP concentration of cocksfoot was significantly greater (P < 0·001) compared with the CP concentrations of prairie grass and perennial ryegrass. The estimated ME concentrations in cocksfoot were high enough to satisfy the requirements of a lactating dairy cow, with defoliation at or before the four‐leaf stage maintaining ME concentrations between 10·7 and 10·9 MJ kg^?1 DM, and minimizing reproductive plant development. The ME concentrations of prairie grass (10·2–10·4 MJ kg^?1 DM) were significantly lower (P < 0·001) than for cocksfoot (as above) and perennial ryegrass (11·4–11·6 MJ kg^?1 DM) but a higher DM production per hectare resulted in prairie grass providing the greatest amounts of ME ha^?1.     

Crown rust affects plant performance and interference ability of Italian ryegrass in the post-epidemic generation

This study investigated the effects of crown rust (caused by Puccinia coronata f.sp. lolii) in Italian ryegrass (Lolium multiflorum) on plant growth and interference ability in the post‐epidemic generation. Rust‐infected Italian ryegrass plants produced seeds with lower weights than healthy plants (on average proportionately 0·36 lower). Furthermore, the early seedling vigour of plants from rust‐infected parents was reduced by 0·50 compared with seed from healthy parents. Under disease‐free conditions in the post‐epidemic generation, mature plants from rust‐infected parents produced 0·13 less shoot dry weight, 0·15 fewer tillers, 0·11 fewer leaves and 0·25 less leaf area than plants from healthy parents. As such, the effects of crown rust in Italian ryegrass persisted for at least one post‐epidemic generation. Under rust‐infected conditions in the post‐epidemic generation, however, the difference in performance between plants from healthy and rust‐infected parents was minimal. This suggests that any advantage plants gain from having healthy parents, compared with rust‐infected parents, might be lost when the post‐epidemic generation is grown under conditions of rust infection. Irrespective of rust infection in the post‐epidemic generation, the magnitude of intra‐specific interference was greater in monocultures of Italian ryegrass from healthy parents than in those from rust‐infected parents. However, the magnitude of inter‐specific interference and the relative interference ability of plants of rust‐infected and healthy parents did not differ. The implications of these results to plant ecology are discussed.     

‘MaxClover’ grazing experiment: I. Annual yields,botanical composition and growth rates of six dryland pastures over nine years

Six dryland pastures were established at Lincoln University, Canterbury, New Zealand, in February 2002. Production and persistence of cocksfoot pastures established with subterranean, balansa, white or Caucasian clovers, and a perennial ryegrass‐white clover control and a lucerne monoculture were monitored for nine years. Total annual dry‐matter (10.0–18·5 t DM ha^?1) and sown legume yields from the lucerne monoculture exceeded those from the grass‐based pastures in all but one year. The lowest lucerne yield (10 t ha^?1 yr^?1) occurred in Year 4, when spring snow caused ungrazed lucerne to lodge and senesce. Cocksfoot with subterranean clover was the most productive grass‐based pasture. Yields were 8·7–13·0 t DM ha^?1 annually. Subterranean clover yields were 2·4–3·7 t ha^?1 in six of the nine years which represented 26–32% of total annual production. In all cocksfoot‐based pastures, the contribution of sown pasture components decreased at a rate equivalent to 3·3 ± 0·05% per year (R² = 0·83) and sown components accounted for 65% of total yield in Year 9. In contrast, sown components represented only 13% of total yield in the ryegrass‐white clover pastures in Year 9, and their contribution declined at 10·1 ± 0·9% per year (R² = 0·94). By Year 9, 79% of the 6.6 t ha^?1 produced from the ryegrass‐white clover pasture was from unsown species and 7% was dead material. For maximum production and persistence, dryland farmers on 450–780 mm yr^?1 rainfall should grow lucerne or cocksfoot‐subterranean clover pastures in preference to ryegrass and white clover. Inclusion of white clover as a secondary legume component to sub clover would offer opportunities to respond to unpredictable summer rainfall after sub clover has set seed.     

Increased concentration of water-soluble carbohydrate in perennial ryegrass (Lolium perenne L.). Evaluation in dairy cows in early lactation

Twelve multiparous Holstein–Friesian dairy cows in early lactation were used to investigate the potential of using perennial ryegrass (Lolium perenne) with a high concentration of water‐soluble carbohydrates (WSC) to increase the efficiency of milk production. Ad libitum access to one of two varieties of zero‐grazed herbage was given continuously for 3 weeks: treatment High Sugar (HS), an experimental perennial ryegrass variety (Ba11353) bred to contain a high concentration of WSC, harvested in the afternoon; or Control, a standard variety of perennial ryegrass (cv. AberElan), harvested in the morning. All dairy cows also received 4 kg d^?1 of a standard dairy concentrate. Dairy cows given the HS diet treatment consumed 2·8 kg dry matter (DM) d^?1 more than Control dairy cows (P < 0·01), and the DM digestibility of the diet on the HS treatment was significantly greater than that of the diet on the Control treatment (0·75 vs. 0·72; s.e.d. 0·010; P < 0·05). Excretion of urinary purine derivatives (PD) tended (P < 0·1) to be higher from dairy cows on the HS treatment, implying increased microbial protein flow to the duodenum, although there was no significant difference in the apparent efficiency of rumen fermentation of either dietary nitrogen (N) or DM expressed as a ratio to urinary PD. Milk yields and milk composition were not significantly affected by dietary treatment, although true protein yields of milk were higher (P < 0·05) from dairy cows given the HS treatment. The proportion of dietary N excreted in urine was significantly lower from HS cows, although the values were low for both treatments (0·20 g g^?1 vs. 0·27 g g^?1; s.e.d. 0·020; P < 0·05). It is concluded that increased DM intakes by dairy cows given the HS treatment led to increased milk protein outputs. With a proportional decrease in urinary N excretion, the use of perennial ryegrass with a high WSC concentration, in the context of the harvesting regime used in this study, may help to reduce N pollution from dairy systems into which it is incorporated.