S U M M A R YTwo steers totally nourished by intragastric infusion of volatile fatty acids and casein were given an abomasal infusion of a microbial protein preparation (Pruteen) at eight rates with a purine input ranging from 0 to 170 mmol/day over 11 successive 5-day periods. The urinary excretion of purine derivatives relative to the purine input was measured. Negligible amounts of xanthine plus hypoxanthine were present in the urine. The relative contributions of allantoin and uric acid to the total excretion were not affected by the rate of purine infusion. Total purine derivative excretion (uric acid and allantoin) was linearly correlated with purine input. Recovery in the urine of the infused purines was 0-77. It is suggested that utilization of exogenous purines may only occur in the intestinal mucosa and that the remaining purines may be completely converted, before entering the liver, to uric acid and allantoin, which are subsequently eliminated by the renal and extrarenal routes. The differences between cattle and sheep in excretion of purine derivatives, and the implications of these differences for the use of purine excretion values in order to estimate microbial protein supply to the ruminant, are discussed.

1. Cattle were maintained by intragastric infusion to see how much nitrogen was excreted on protein-free diets. 2. Minimal N excretion was estimated with two dairy cows in three periods, i.e. when they were non-pregnant and non-lactating, when they were between 117 and 133 d pregnant and when they were between 220 and 233 d pregnant. The minimal N excretion was also estimated on two occasions with two steers when their average live weights were 200 and 350 kg. 3. Average urinary N excretion without protein infusion was 298, 305 and 283 mg/kg metabolic live weight (W0.75) for the non-pregnant cows and for cows during the first and second periods of pregnancy respectively; total N excretion including the faecal N was 340, 329 and 319 g/kg W0.75. 4. For steers the urinary N values were 403 and 295 mg/kg W0.75 at 200 and 350 kg live weight respectively and total N excretion including faecal N was 408 and 320 mg/kg W0.75. 5. Urinary excretion of creatinine was the same for animals given casein via the abomasum as a source of protein or given no protein with mean values for the cows of 13.6 and 14.9 g/d for the first and second stages of pregnancy respectively. Mean values for the steers were 6.5 and 7.6 g creatinine/d at 200 and 350 kg live weight respectively. 6. It is suggested that the so-called metabolic faecal N in ruminants, estimated with N-free diets, is mainly, endogenous N derived from tissue breakdown of protein but incorporated in microbial debris and excreted in the faeces.

1. A method of continuous alimentation of cattle by total infusion of nutrients has been developed. Friesian steers within the weight range 100400 kg live weight and dairy cows were used.2. A multi-channel peristaltic pump was used to infuse solutions of volatile fatty acids (VFA), minerals, and buffer through a cannula in the rumen and a casein-vitamin solution into the abomasum.3. The method described was successfully used with two cows and four steers in a series of trials over intervals of approximately 2 months. The levels of infusion were up to twice maintenance and with various relative proportions of VFA and protein. Blood metabolite levels, rumen osmotic pressure and pH were monitored and effectively controlled.In ruminants where there are two interrelated sets of requirements: those of the microbial population and those of the host animal, interpretation of information collected from studies on normally-fed animals is difficult. These difficulties arise in predicting the fate of nutrients in the rumen, and in determining quantitatively the contribution made by bacteria and protozoa to the host animal's requirements.In order to reduce the contribution of the rumen fermentation to the animal's requirement, diets have variously been supplemented with salts of volatile fatty acids (VFA) added to the diet, or with infusions of VFA directly into the rumen and infusions of protein into the abomasum. Armstrong & Blaxter (1957a,b) used infusions of acetic, propionic and butyric acids to supply part of the energy of a basal diet of dried grass and Martin & Blaxter (1963) infused protein only, in fasting sheep. Hovel1 (1972) conducted studies with lambs using salts of VFA, and Tao & Asplund (1 975) developed a method by which they infused partially-neutralized VFA into the rumen of sheep and intravenously-infused amino acids. Difficulties were sometimes encountered however, primarily with the control of rumen pH and with electrolyte imbalances. Furthermore, with the exception of very low levels of infusion, the approach was generally that of supplementation of a normal basal diet. A method was developed however, using lambs (Orskov, Grubb, Wenham et al. 1979) in which all nutrient requirements were met by infusion. Relatively few difficulties arose except with a VFA mixture containing a high proportion of acetic acid (mmol/mol: 850 C,, 50 C,, 100 C4), when problems of metabolic disturbance and of controlling rumen pH were encountered (Orskov, Grubb, Smith et al. 1979). The technique reported here is an adaptation of the method used for lambs. M A T E R I A L S A N D METHODSRumen and abomasal cannulas. The rumen cannulas fitted to the cattle (Fig.

No abstract

The effects of changing rumen osmotic pressure (OP) upon water kinetics and volatile fatty acid (VFA) absorption in the rumen of sheep were studied in two 4 x 4 Latin square experiments, each using four lambs with a rumen cannula and an abomasal catheter. In both experiments the lambs were sustained by the intragastric infusion of all nutrients (VFA, Ca, P, Mg and a buffer solution into the rumen, and casein, vitamins and trace elements into the abomasum). On experimental days, which were at least 1 week apart, drinking water and the casein infusion were withdrawn, and the ruminal OP was changed and held constant for 9.5 h, by incorporating NaCl at different concentrations in the buffer solution being infused. In Expt 1 the target OP values were 300,340,380 and 420 mosmol/kg, and in Expt 2 were 261 (no saline addition), 350, 420 and 490 mosmol/kg. Using soluble non-absorbable markers (PEG in continuous infusion and Cr-EDTA injected in pulse doses) rumen volume, liquid outflow rates, apparent water absorption through the rumen wall and VFA absorption rates were estimated at six sampling times corresponding to the 1.5 h intervals during the last 7.5 h following the change in rumen OP. Liquid outflow rate (F; ml/h) showed a significant and positive linear relationship with the rumen OP (mosmol/kg), resulting in the equation F = 1.24 OP (SE 0.096)-36.5 (SE 36.6) (v2 0.96). Similarly, water absorption rate ( W ; ml/h) was significantly affected by rumen OP, and this relationship was given by W = 395 (SE 39.9) -1.16 OP (SE 0.105) (v2 0.95), which means that for an OP of 341 mosmol/kg the net movement of water across the rumen wall would be zero, and either a net efflux or a net influx of water would be observed with lower or higher OP respectively. In Expt 2 there was a significant linear effect of OP on rumen volume (P < 0.01), with higher OP being associated with increases in rumen liquid contents of about 1040%. As rumen OP was increased there was also a decline in the absorption rate of VFA (from 232 mmol VFA/h for OP 350 to 191 mmol/h for OP 490 mosmol/kg), resulting in the accumulation of VFA (especially acetate) in the rumen and a consequent fall in rumen pH. Rumen OP seems to be important in defining water movement across the rumen wall and, hence, partitioning between absorption and outflow.