In two latin square designs, four sheep were given either unsupplemented rice straw or rice straw supplemented with 20 g/kg of urea. Within each latin square four treatments were imposed with four levels of glyricidia forage supplementation to provide either 0, 50, 150 or 300 g/kg of the straw. Rice straw intake increased with increasing supplementation of glyricidia forage. With untreated straw the daily intakes of straw were 412, 467, 504 and 498 g/d and with urea-treated straw the intakes were 463, 482, 507 and 510 g/d for the four levels of glyricidia supplementation respectively. It is postulated that for the untreated straw, glyricidia provided both limiting N, easily fermentable fibre and bypass protein while with the urea- supplemented straw the responses were due to easily fermentable fibre and bypass protein.

It is well known that diets based on untreated straw are too deficient in N to sustain maximal rumen microbial activity and optimum degradation rate of the straw. Silva et al (1988), however, showed that even when rumen ammonia concentrations were adequate the fermentation rate of untreated straw was less than maximal. Untreated straw incubated in the rumen of sheep receiving untreated straw, supplemented with urea, was fermented more slowly than when untreated straw was fermented in the rumen of sheep receiving ammoniated straw. They identified that the reason was insufficient rumen microbes in solution to ensure rapid attachment and degradation as attachment appeared to be more rapid in the rumen of animals receiving ammoniated straw. These authors also showed (Silva et al 1989) that this could be corrected by feeding sugar beet pulp as a supplement. Intake of straw was shown to increase by 25% when 15% sugar beet pulp was included in the diet. The contentof easily fermentable fibre in sugar beet pulp appeared to be the reason.

Leguminous forages are likely to provide an excess of N for rumen microbes relative to fermentable carbohydtrate and also a source of easily fermentable fibre. Supplementing straw diets with leguminous forage should therefore contribute both a N source and a source of easily fermentable fibre to untreated straw diets.

The experiment reported here was conducted in Sri Lanka using rice straw supplemented with leaves and petioles from the leguminous multi-purpose tree, Glyricidia maculata, to investigate whether the effect of glyricidia supplementation was the same whether or not urea was added to the basal diet of rice straw.

Eight wethers weighing about 32 kg liveweight were used. They were divided in two groups of four animals and allocated to receive rice straw with or without 20 g urea/kg of straw. The urea was sprayed in a water solution on the straw. Within each group a latin square design was used with four periods and four levels of glyricia supplementation.The length of each period was 29 days during the last 8 days of which the faeces were collected. The levels of glyricidia were equivalent to 0, 5, 15 or 30% of the dry weight of straw. The glyricidia consisted of leaves and petioles. Straw was offered at about 10% above the amount consumed. The chemical composition of the rice straw and the glyricidia is given in Table 1. The analyses were carried out according to AOAC (1975) and Goering and Van Soest (1970). In addition to the described diets the animals received 15 g/d of a vitamin/mineral mixture. At the end of each period rumen fluid samples were obtained by stomach pump four hours after morning feeding for analysis of pH and ammonia concentration.

Table 1: Chemical composition of rice straw and Glyricidia maculata
	N x 6.25	Ash	NDF	ADF
	----- g/kg dry matter ----
Rice straw	37	172	728	423
G. maculata	238	84	423	244

The results are given in Table 2. The digestibility of organic matter is calculated for the rice straw alone and for the whole diet. The digestibility of the organic matter of glyricidia was taken to be 613 g/kg from estimates of degradability of the same material (Premaratne et al 1991).

Table 2: Effect of urea and Glyricidia supplementation of rice straw on intake, digestibility, rumen pH and NH3 concentration in sheep
	G. maculata, g/kg diet DM
	0	50	150	300
OM dig of straw (g/kg)
No urea	412	467	504	498
20 g urea/kg diet	463	463	507	510
OM dig total diet (g/kg)
No urea	463	473	519	548
20 g urea/kg diet	482	488	522	542
DM intake of straw (g/day)
No urea	574	574	848	721
20 g urea/kg diet	690	727	880	715
DM intake total diet (g/day)
No urea	673	708	998	1030
20 g urea/kg diet	690	765	1035	1021
Rumen pH
No urea	6.8	6.8	6.7	6.4
20 g urea/kg diet	6.9	6.9	6.8	6.8
Rumen NH3 (mg/litre)
No urea	43	103	185	202
20 g urea/kg diet	162	136	183	221

Adding urea to the diet increased organic matter digestibility (by 10%)in the absence of glyricidia but had no effect on this parameterwhen the straw was supplemented with glyricidia foliage.

The organic matter digestibility of the straw was increased by 20% as the level of glyricidia was increased up to 150 g/kg of the diet in the absence of urea. The effect was slightly less marked (10% increase) when urea was added. Total diet digestibility continued to increase up to 300 g glyricidia/kg diet due to the higher digestibility of glyricidia in comparison to rice straw. As for the straw digestibility the effects were slightly less when urea was present.

There were large increases (48%) in intake of rice straw and of the total diet as the glyricidia level was raised to 150 g/kg diet, both in presence and absence of urea. The degree of increase was slightly less when urea was added. Urea per se also increased intake especially in absence of glyricidia foliage.

Rumen ammonia concentration was increased significantly by glyricidia when no urea was added; in the presence of urea the tendency was the same but but less consistent.

The results indicate that three different principles can influence the rumen environment and animal response (feed intake) on straw-based diets. The increases in digestibility and intake caused by the urea, in absence of glyricidia, were obviously due to the provision of fermentable nitrogen. The effect of the glyricidia on both these parameters was greater than observed for the urea and shows that the leguminous foliage was providing factors other than fermentable nitrogen. One of these could have been ferrmentable fibre as observed for the addition of sugar beet pulp to a straw diet (Silva et al 1989) resulting in increased attachment rate of microorganisms to the straw. A similar effect was noted by Maniuchi et al (1992) when they used ammoniated straw as a supplement to untreated but urea- supplemented straw and achieved a greater intake. The fact that the glyricidia had a marked effect in stimulating intake suggests that probably it also provided "bypass" or "escape" protein. This is in line with the results reported by Preston and Leng (1987) for steers fed a basal diet of "King" grass when addition of glyricidia foliage tripled the rate of liveweight gain.

The different responses are illustrated in Figures 1 and 2 and show that low levels of suitable supplements can have quite large effects on the rumen environment and stimulate both intake and digestibility of the basal feed.

The results suggest that supplementation with glyricidia foliage at levels of 15 to 30% (DM basis) provides adequate fermentable nitrogen as well as fermentable fibre and that addition of urea may not be necessary in this case.It is also probable that the marked effects of glyricidia on intake are because not all the protein is degraded and that some escapes intact from the rumen.

This work was supported by a grant from the International Atomic Energy Agency, Vienna.

(Received 1 September 1995)