Abstract:
Proc. Aust. Soc. Anim. Prod. 1994 Vol. 20 THE DIFFERENCE IN MILK PRODUCTION BETWEEN SUBTROPICAL AND TEMPERATE ANNUAL RYEGRASS SELECTIONS K.F. LOWE, G.K. REASON and T.M. BOWDLER Dcpt of Primary Industries, PO Box 96, Ipswich, Qld. 4305 SUMMARY A grazing experiment was conducted to compare, in a subtropical environment, the performance of a subtropical selection of annual ryegrass (Lolium multiflorum cv. Midmar) with 1 selected for temperate situations (cv. Grasslands Tama) and to determine whether cuttin g experiments could predict their relative performance under grazing. Midmar produced higher pasture and milk yields than Grasslands Tama and generally produced milk of a similar or higher standard. There was good agreement between yields under cutting and those obtained as feed on offer under grazing. Milk production differences generally followed those of feed on offer but quality differences, particularly with regard to rust damage Ooerate the cultivar differences. The results suggest that selections of annual in spring, tended to exa,, ryegrasses made in the subtropics which show improved yields under cutting can also demonstrate improved milk yields when grazed. Keywords: annual ryegrass, milk production, subtropics, Midmar, Grasslands Tama. INTRODUCTION Nitrogen fertilised, irrigated ryegrass is a major source of winter feed for dairying in the subtropics. Cutting experiments indicate that there are differences in the dry matter yields produced by, and the damage resulting from leaf rust (Puccirzia coronata) infection of, annual ryegrass (Lolium muZtijZorum) cultivars (Lowe et al. 1983). However, whether the increased production measured in cutting experiments can be converted into increased animal production has long been a concern (Rossiter 1969) because such improvements may call for a completely new approach to livestock management (Campbell 1967). There is little doubt that grazing experiments are the best method of measuring differences between pasture species (Lambourne 1969) but the magnitude of the costs involved require the use of compromise techniques which provide valid comparisons (Rossiter 1969). If it could be shown that differences in ryegrass yields obtained from cutting experiments were of a similar order to those measured by lactating dairy cows, then cutting experiments could reliably be used to estimate the likely animal production increases from superior cultivars. The dry matter production in cutting experiments of Midmar, a diploid cultivar from South Africa, and Grasslands Tama, a tetraploid cultivar selected in New Zealand, are different in subtropical and temperate environments (Oram 1990). This experiment compared the plant and animal production differences between the 2 cultivars of Lolium multiflorum in subcoastal south-east Queensland. MATERIALS AND METHODS Area The evaluation was conducted at Mutdapilly Research Station (27'46' S,152'40' E) on heavy, dark, self mulching, cracking clay soils with dark to grey alkaline subsoils (Powell et al. 1985). Both the internal and surface drainage of these soils are restricted. Phosphorus levels are moderate to high, potassium and sulfur levels are moderate, and pH is around 6.0. Design and treatments Lolium multiflorum cv. Midmar, from South Africa, and cv. Grasslands Tama, selected in New Zealand, represented selections from the 2 environments. Two stocking rates (4.0 and 7.0 lactating cows/ha) were imposed on the 2 grasses in a randomised block design with 2 replications. In 1983 a cutting experiment was conducted by locating a fenced, 5 m x 5 m plot, in each treatment of the grazing area. Pastures Both cultivars were sown into fully prepared seedbeds at 40 kg/ha of commercial seed. A 1 week on, 3 weeks rest, rotational grazing system was used, requiring 4 pasture areas of which 2 were sown in the third week of March, and 2 in the first week of April each year. Follow up irrigation was applied at 5-7 day intervals until grazing commenced 5 weeks after sowing. Regular, unseasonal rainfall in winter was received in 1983 and further irrigation was not required until September when irrigation 120 Proc. Amt. Sot. Anim. Prod. 1994 Vol. 20 (50 mm/ha.application) was supplied on a 2-weekly basis. Irrigation applications were adjusted to allow 1 week drying prior to stock entry. Basal fertiliser (500 kg/ha superphosphate, 125 kg/ha muriate of potash and 280 kg/ha sulphate of ammonia) was applied before the final cultivation and incorporated into the surface soil. Maintenance fertiliser (125 kg/ha of urea) was applied after grazing (4-weekly intervals), and followed by irrigation (50 mm). AnimaZs A mixture of multiparous Holstein-Friesian and Australian Friesian Sahiwal (AFS) cows were used in 1983; in 1984 all animals were Holstein-Friesian heifers. Animals grazed the pasture for 24 hours and were not supplemented. Measurements Feed on offer was measured using an electronic capacitance meter, taking 20 and 30 readings in the 7 and 4 cows/ha paddocks respectively. Calibration curves were constructed from each cultivar for each week by cutting 10,0.5 m x 0.5 m quadrats from each paddock with hand shears after measuring the same area with the meter. A pooled equation for the 4 treatments was found to give the best estimate of yield. The r3 values of these equations ranged from 0.90 at the beginning of the season to 0.72 in November. Nitrogen and phosphorus content of ryegrass was measured monthly in 1984. Yield in the cutting 0 experiment was measured monthly by cutting a 0.9 m x 2.4 m quadrat from the area with a reciprocating blade mower. The remaining material was removed after sampling. Rust damage was assessed on a 0 (no rust) to 4 (more than 25% of individual leaf blades damaged by rust) scale (Lowe et al. 1983). Milk production was recorded weekly and samples were taken for analysis of milk constituents. Animals were weighed monthly. Statistics Pasture on offer was submitted to ANOVA, with cultivar, stocking rate and weekly paddock data used as factors in the analyses. Milk production and milk quality was analysed using GLIM (Baker and Nelder 1985) with cultivars, stocking rate, and pasture rotations as factors. RESULTS Pasture production Feed on offer from Midmar was 22% and 10% higher than Grasslands Tama in 1983 and 1984 respectively. There was no significant difference (P > 0.05) until the September/October period in both years, although the Midmar yields were generally greater (Table 1). At both stocking rates, feed on offer of Midmar increased throughout the experimental period in 1984, and from mid winter onwards in 1983. Yields of Grasslands Tama increased only at the lower stocking rate. The nitrogen content of both forages in 1984 averaged 3.55% and were similar (P > 0.05) in each cycle. The forage at the lower stocking rate (4 cows/ha) had a lower (P c 0.05) nitrogen content (3.24%) 0 than at 7 cows/ha (3.85%). These stocking rate differences were significant after the first 4-week cycle. Phosphorus content differences mirrored those of nitrogen, averaging 0.30% and 0.31% for Midmar and Grasslands Tama respectively. At 4 and 7 cows/ha, the phosphorus contents were 0.29% and 0.32% respectively. In the cutting experiment in 1983, Midmar produced more (P c 0.05) dry matter from June to October than Grasslands Tama (9.9 vs 7.7 t/ha). Yield data for the last period (October/November) was lost when animals broke into the plots. From mid August, rust affected the ryegrass in both cutting plots and grazing areas; Midmar was always less (P c 0.01) affected than Grasslands Tama. Highest rust damage occurred in early November, scoring 0.5 and 2.7 for Midmar and Grassland Tama respectively. Milk production, milk quality and animal weights In 1983, milk production per cow was similar for both cultivars until September, after which production from Midmar was significantly higher (P c 0.05; Table 2). In 1984, milk production from Midmar was higher (P c 0.05) throughout the season. Stocking rate did not affect per cow production until September in each year, but there was a significant response in milk production per hectare throughout. Cultivar differences were similar for both milk production per cow and per hectare. In 1983, fat and protein contents of milk from animals grazing Grasslands Tama (3.94 and 3.33% respectively) were higher (P c 0.05) than from those grazing Midmar (3.80 and 3.25% respectively), while lactose content was higher (P < 0.05) from animals grazing Midmar (4.08 vs 3.82%). Lactose and protein yields were higher (P c 0.05) in cows grazing Midmar whilst fat yields were similar for both cultivars. The higher stocking rate reduced the percentage and yield of all 3 milk components. In 1984, protein content was higher (P < 0.05) from cows grazing Midmar compared with those on 121 Proc. Amt. Sot. Anim. Prod. 1994 Vol. 20 Table 1. The effect of cultivar and stocking rate on feed on offer (t/ha) of annually-sown ryegrasses at Mutdapilly Research Station in south-east Queensland Table 2. The effect of ryegrass cultivar and stocking rate on milk production per cow (L/cow.day) in 1993 and 1984 Grasslands Tama (3.12 vs 3.03% respectively) but fat levels were higher (P c 0.05) from cows on Grasslands Tama (3.94 vs 3.83% respectively). However, there was no difference between the lactose content (4.98 and 5.02% respectively) of the milk. There were no stocking rate effects on milk composition but per cow milk component yields were all significantly lower (P < 0.05) at the higher stocking rate. Fat, lactose and protein yields per cow were higher on Midmar. Liveweight changes were similar (P > 0.05) on the 2 cultivars. Between May and November each year, animals on the higher stocking rate lost weight while those on the lower stocking rate gained weight (the 2 year averages were -9 and +18 kg respectively). DISCUSSION The growth of the 2 annual ryegrass cultivars in a grazed situation was similar to previously published results from small plot cutting experiments (Lowe et al. 1983, 1993). In all experiments, Midmar produced superior growth rates, particularly in spring, and lower rust damage, compared with Grasslands Tama. Our milk production results also demonstrate Midmar' superiority which was most evident in s spring. In 1983, milk production differences between cultivars appeared to closely mirror differences in 122 Proc. Aust. Sot. Anim. Prod. 1994 Vol. 20 feed on offer. However in 1984, Midmar gave higher production throughout the year even though differences in feed on offer only became evident at the beginning of spring. It is postulated that the milk production of animals grazing Grasslands Tama could have been restricted by the high moisture content of the Grasslands Tama, compared with Midmar, in the first 8-12 weeks of grazing (92 vs 88%) during 1984. In 1983, excessively wet conditions early in the season lengthened the interval of grazing so that animals grazed more mature forage increasing the dry matter content of Grasslands Tama. There was no ooest evidence from our results or from previous data (Lowe et al. 1983) to su,, that there were substantial quality differences between the 2 cultivars except in spring when rust caused major damage to Grasslands Tama foliage. Midmar generally produced milk of equal or better quality than that produced off Grasslands Tama. The between-year differences probably reflect the different breeds of animals used. In 1983, about a third of the animals were from an experimental crossbreeding herd of AFS. In 1984 a line of Holstein Friesian heifers was purchased for the experiment. They were poorly fed prior to purchase and proved to have low production potential. Milk production levels were very low, considering the quality of feed available. This was partly due to the fact that the animals were unsupplemented and partly to the type of animals used. Our results suggest that there is good agreement between the relative growth of the 2 annual ryegrass cultivars from cutting and grazing studies. To some extent, milk production reflected differences in pasture growth although other factors such as the effect of rust on palatability in spring appeared to further increase cultivar differences. REFERENCES BAKER, R.J. and NELDER, J.A. (1985). 'The GLIM System Release 3' (Numerical Algorithm Group: Oxford). CAMPBELL, A.G. (1967). P YOC. N.Z. Sot. Anim. Prod. 27: 126. LAMBOURNE, L.J. (1969). Proceedings of the Australian Grasslands Conference 1968, Vol. 2, Perth, pp. 33-48. LOWE, K-F., BOWDLER, T-M., OSTROWSKI, H. and STILLMAN, S.L. (1983). Aust. J. Exp. Agric. Anim. Hub. 23: 294-301. LOWE, K.F., BOWDLER, T.M., REASON, G.K. and MOSS, R.J. (1993). Proc. ht. Grassland Cong. XVII, (in press). ORAM, R.N. (1990). 'Register of Australian Herbage Cultivars' (CSIRO: Australia). POWELL, B., BAKER, D.E. and CHRISTANOS, N.G. (1985). 'Soils of the Mutdapilly Research Station', Research Bulletin QR85001 (Qld. Dept of Primary Industries: Brisbane). ROSSITER, R.C. (1969). Proceedings of the Australian Grasslands Conference 1968, Vol. 2, Perth, pp. 13-24. 123