Rhodes grass can be a high quality forage for ruminants when grazed or harvested at an early stage of maturity. However, its nutritive value strongly decreases with maturity, especially after the first cut. Hay harvested at a later stage of maturity has a low protein content and a high fibre content, particularly in the stems, and it should be supplemented when fed to ruminants with nutritional requirements higher than those necessary for maintenance.
Young Rhodes grass of 4 weeks of regrowth or less was found to have a high in vitro OM digestibility of 70-80 % (Mbwile et al., 1997a; Mero et al., 1997), which decreased to 50% after 10 weeks of regrowth (Mero et al., 1997). In Kenya and Tanzania, in vivo OM digestibility and intake of Rhodes grass by dairy cows or heifers decreased with increasing maturity after the first cut (Abate et al., 1981; Mbwile et al., 1997b). In Tanzania, in vivo OM digestibility decreased from 76% at 6 weeks of regrowth to 60% at 12 weeks of regrowth in cows fed fresh Rhodes grass (Mbwile et al., 1997b). In Kenya, intake of Rhodes grass decreased with maturity in grazing growing Friesian and Ayshire heifers (Abate et al., 1981). However, after the second cut, the effect of the stage of maturity on intake and in vivo digestibility was less important as these parameters remained high even with mature forage (Mbwile et al., 1997b). OM digestibility in grazing heifers was higher during the wet season than during the dry season (Abate et al., 1981).
Pasture and fresh forage
In Australia, Chloris gayana fed at late stage of maturity could be fed to lactating cows with moderate milk production, generally with supplementation. Dairy cows grazing irrigated Rhodes grass pasture (2 to 6 weeks regrowth, after the 2nd or 3rd cut) mulched at least once a year could produce more than 14.3 kg/d of milk when supplementated with 5 kg of cereal-based concentrate per cow at a stocking rate up to 3.7 cows/ha (Ehrlich et al., 2003a). In a follow-up study, with Holstein-Friesian dairy cows grazing irrigated Rhodes grass pastures (6-week grazing rotation) supplemented with 5 kg of cereal-based concentrate, the stocking rate could be increased from 3.5 to 6.1 cows/ha without altering milk production. However, cows stocked at stocking rates higher than 3.5/ha could not maintain liveweight during the 18-week experiment and it was recommended to increase the level of supplementation for high stocking rates (Ehrlich et al., 2003b).
In Tanzania, lactating Friesian cows could be fed with fresh Rhodes grass that was supplemented only with a small amount of salts. However, the evolution of milk production during the experiment was not reported. The cows produced 8.7 kg milk/d at the beginning of the experiment (Mbwile et al., 1997b). In Kenya, dairy heifers grazing Rhodes grass had an average daily gain of 581 g/day during 1-year experiment with a stocking rate of 2 livestock unit/ha but the average daily gain ranged between 200 and 1100 g/d according to the period of the year. Supplementation was recommended in order to sustain an average daily gain of 550 g/d at least when the forage quality is low (Abate et al., 1981).
Cows not only prefer to eat leaves rather than stems but have the ability to select leaves over stems in Rhodes grass (Mbwile et al., 1997b; Ehrlich et al., 2003a): in order to maximize intake, it can be useful to allow a high level of selection of forage by cows, by offering 10 to 20% excess feed depending on the maturity and cut (Mbwile et al., 1997b).
In order to optimize the harvested biomass, Rhodes grass hay is generally harvested at advanced maturity stages. However, hay alone can hardly meet the nutritional requirements of productive ruminants and must therefore be supplemented (Mero et al., 1998; Mtenga et al., 1990; Mupangwa et al., 2000; Osuga et al., 2012). On the other hand, Rhodes grass hay cut early at 21 days of regrowth had a high nutritive value comparable to that of fresh Rhodes grass (Tagari et al., 1977).
A first limitation of the nutritive value of mature Chloris gayana hay is its low intake by livestock when compared to hays of other tropical grass species. In Tanzania, in Blackhead Persian rams, the intake of Rhodes grass hay harvested at 6 or 10 weeks or regrowth was 20% lower than that of Cenchrus ciliaris and Panicum coloratum, even though the in vivo OM of the hays were comparable. It was recommended to allow for 30 to 50% of refusals depending on maturity so that the animals could select the most digestible parts. Rhodes grass hay was then able to meet sheep maintenance requirements (x 1.1-1.2) (Mero et al., 1998).
Another limitation of mature Rhodes grass hay is its low protein content, particularly during the dry season. Several trials have shown that supplementation with a protein-rich source resulted in higher animal performance (Mtenga et al., 1990; Mupangwa et al., 2000; Osuga et al., 2012). In Tanzanian goats, supplementation of low protein Chloris gayana hay (5.7-7.7 %DM) with a concentrate containing between 10-18% DM protein increased the total feed intake, doubled the growth rate, and improved the feed conversion ratio and the lean+fat:bone ratio (Mtenga et al., 1990). In Zimbabwe, with East African goats, supplementation of a low protein Rhodes grass hay (7% DM) with 100 g/d of maize grain and 25% of legumes (Cassia rotundifolia, Lablab purpureus or Macroptilium atropurpureum, 12% DM of protein) increased daily gain, total intake and the supply of microbial nitrogen for absorption in the lower intestinal tract (Mupangwa et al., 2000). In Kenya, with growing East African goats, supplementation of a poor quality Chloris gayana hay (5% DM protein) with 60 g of maize bran and 15 or 30 % of legumes (Berchemia discolor or Zizyphus mucronata) increased intake, multiplied liveweight gain per 6 to 12 and increased the ammonia content of the rumen above 50 mg/l (Osuga et al., 2012) which is considered as the minimal concentration required to maximize microbial growth in the rumen (Leng, 1990).
Ensiling of Rhodes grass has been little studied, since it is difficult to ensile due to its high moisture coupled with low contents of water-soluble carbohydrates, like other tropical grasses (Parvin et al., 2010). In Australia, with Friesian-Holstein heifers, a comparison of Rhodes grass cut at two stages of maturity (60 and 100 days) and treated with CaO, NaOH or a microbial incoculent before ensiling found that only NaOH treatment allowed a 25% increase of DM intake for mature grass silage as well as a reduction in NDF content and in sacco digestibility. None of the treatments had positive effects on the in vivo nutritive value or storage quality of young Chloris gayana silage (Chaudhry et al., 2001).