Sabi grass (Urochloa mosambicensis) as a component of Townsville stylo (Stylosanthes humilis) pastures.

Abstract

SABI GRASS (UROCHLOA MOSAMBICENSIS) AS A COMPONENT OF TOWNSVILLE STYLO (STYLOSANTHES HUMILIS) PASTURES L. FALVEY SUMMARY The liveweight changes of steers grazing pastures sown to T.S. and T.S.-Sabi are compared. Liveweight change varied with legume content of the sward most of the year except during the early wet `season. The T.S. pasture produced significantly higher liveweight gains over the whole period of the experiment , probably because of its higher proportion of legume. Despite the difference in animal productivity a T.S.-Sabi pasture is probably-easier to manage to maintain the legume component. 1, INTRODUCTION . The use of Townsville stylo (T,S,) (Stylosanthes humilis) as a pasture species has allowed very much higher weight gains to be achieved over native pastures by increasing liveweight gains per animal and per hectare, (Norman, 1961) o One problem of T,S, pastures is invasion by annual grasses which eventually lower animal production, Norman (1965) suggested that increased grazing pressure would control this invasion on Tippera clay loam at Katherine, and Falvey (1975) determined,a similar requirement for Blain sand in the north Tipperary region, An alternative means of overcoming this problem is the deliberate inclusion of a perennial grass in the sward to occupy the same niche as the invading annual grasses. Such a grass, however must be persistent yet not so vigorous as to compete significantly with the T& (Austin, 1970) 0 This grass could also provide more rapid regrowth after rains which render the dry annual species unpalatable, and promote soil stabi 1 ity during the early heavy showers of the wet season, Austin (1970) reported high levels of animal production from a T,S, pasture including the perennial Sabi grass (Urochloa mosambicensis), This paper reports a comparison between a T,S, pasture and a ToSo-Sabi pasture, 110 METHOD The experiment was conducted at Douglas Daly Experiment Station 160 km south of Darwin which is in a tropical monsoon region receiving a mean annual rainfall of 1100 mm, 90% of which falls between November and March, Two, four hectare paddocks of Townsvil le stylo and two, four hectare paddocks of Townsvil le stylo - Sabi grass were each stocked with eight Brahman cross steers on 20/T/72. The stocking rat& was reduced to lo75 steers per hectare by removing the lightest animal in each treatment on 14/U/73 to compensate for the increased grazing pressure as the steers grew, Steers were weighed at four weekly intervals, Pastures were sampled regularly for botanical composition and dry matter avai labi 1 i tyo Botanical composition was estimated by the dry weight rank method, Dry matter availabil ity was estimated by taking 10, Oo4m2 quadrat cuts per paddock or by harvesting a 30 x 0,2m strip with a small forage harvester at 7 cm above the ground. Aol. E AoB9 Department of Northern Austral ia, Darwin, NOT0 337 Faecal grab samples of similar size were collected per rectum from each animal during the dry season. Samples from the steers in each paddock were bulked, dried at 100�C for 24 hours (Falvey and Woolley, 1974) and analysed for ni trogen and in some cases phosphorus. Results presented in table 1 are divided into season periods. 'Early wet season' refers to a period of about eight weeks after the opening rains while the period called 'All wet' includes this period and the remaining four months of the wet season. The dry season is a period of about six months between May and October. 111. (a) Liveweights The 1 iveweights of steers and the date of the opening rains of each wet season are presented in figure 1. Li vewei ght changes of steers on the T.S.-Sabi pasture are significantly (P(O.05) higher during the early wet season in al 1 three years. Over the whole wet season there was no significant (P(O.05) difference in 1 ivewei g'ht change in two years and In each dry season- the in the other -year the T.S. pasture was superior. steers grazing the T.S. pasture showed significant (P(O.05) advantage over those on the T.S,-Sabi pasture. Over the three year experimental period, T.S. was significantly (P(O.05) superior. (b) Botanical composition and Dry Matter Availability The mean percentages of T.S . and grass and the dry matter production figures for both pastures are presented in table 1. The grass component for the T.S. pasture represents i nvadi ng annual grasses whereas the grass in the T.S. -Sabi pasture is Sabi grass alone. Weed invasion in the T.S.-Sabi pasture was minimi'sed and did not usually include other grasses. The main invading grass species in the Townsvi 1 le stylo pastures were Brachiaria pubigera, Digi taria ci 1 iaris, Panicum affairoides and Setaria dielsi i. The proportion of T.S. was always greater in the T.S. pasture. Correlation coefficients indicate a positive association between the percentage legume in a sward and the liveweight gain during the dry season, and a negative association during the early wet season. Signifi cant correlat ions were only recorded dur i ng the 1973 dry season (r =+ 0.97 9 P< 0.05 ) and the 19731 74 wet season (r = - 0.91, P(O.10). Dry matter a vailabili ty was simi lar on both swa rds and was not limiting at any stage. Faecal crude protein and phosphorus The mean faecal crude protein and phosphorus contents for steers grazing both pastures are also presented in table 1. The faecal crude protein values of steers grazing the T.S. pastures were consistently higher than those grazing the T.S:Sabi pastures. Faecal phosphorus levels were generally higher on the T.S:Sabi pasture. Correlations between faecal crude protein values and liveweight gain were never significant although in the 1974 dry season a correlation coefficient of +0.97 (P(O.05) was calculated between the mean percent legume from three estimates and the mean faecal crude protein values. RESULTS 338 TABLE 1. 1v. DISCUSSION The T.S. pasture showed significant liveweight advantage during each dry s.eason, duriing the latter part of each wet season, and during 1974, over the whole period of the wet season. The T.S:Sabi pasture on the other hand showed a significantly lower liveweight loss after the opening rains of the wet season and significantly greater liveweight gain during the early wet season. The two main differences between the pasture types were the different proportions of T.S. in the swards and the grass species present. Previous discussions have considered the grass species present to be a more important issue (Austin, 1970). This appears to be true only during the early wet season when the perennial nature of the Sab i grass al lows a quicker response to increased moisture. The negative correlation between the percentage legume and. liveweight gain at that time of the year indicates that T.S. is probably not the species being grazed to any extent, most probably due to low accessibility of new growth and unpalatability of the older material. The correlation between percentage legume and liveweight gain in the dry season indicates that nitrogen content of the diet is more impoirtant for the rest of the year. Faecal crude protein determinations reflect the nitrogen content of the diet (Moir, 1960) and in this case steers grazing the T.S.pasture had higher faecal crulde protein levels. Winks and Laing (1972 and pers. comm.) observed that when faecal crude protein falls below 8%, liveweight loss can be expected. This has been. observed in this case where the steers grazing the T.S. pasture with a mean faecal crude protein content of greater than 8% continued to gain weight while those grazing a 339 T.S.-Sabi grass pasture and having a mean faecal crude protein content of less than 8% lost weight. The lack of significant correlation between faecal crude protein and liveweight gain indicates that further interpretation of the relationship is complex. A positive correlation between the mean faecal crude protein content and the mean percentage legume in the sward over the 1974 dry season and between liveweight change and percentage legume in the 1973 dry season indicates that 1 ittle nitrogen is suppl ied by the grass component of the pastures. While a significant difference between these pastures was recorded in this experiment, a major extension barrier exists in convincing property managers of the importance of high stocking rates to maintain the legume component of the swards. Experience has shown that a T.S.-Sabi pasture can withstand greater mismanagement than a T.S. pasture and sti 11 respond to correct management later. Hence pasture recommendations vary with the intended grazing management of the area. v. ACKNOWLEDGEMENT I thank Mr. H. Mollmann for managing this experiment. Vl. REFERENCES Turnoff 2~28 AUSTIN, J.D.A. (1970) Tropi cal=Agr icul ture 5g: 143 FALVEY, L. (1975) FALVEY, L. and WOOLLEY, Anne (1974) Australian Journal of experimental Agriculture and Animal Husbandry -14:716 Queensland Journal of Agricultural MOIR, K.W. (1960) Science =11~361 Journal of the Australian Institute of NORMAN, M.J.T. (1961) Agricultural Science 22:24 =Journal of the Australian Institute of NORMAN, M. J.T. (1965) Agricultural Science 31:311 Proceedings of the Austral ian WINKS, L. and LAING, A.R. (1955) Society of Animal Production 9:253 = 340