Estimating ruminal microbial efficiencies in silage‐fed cattle: comparison of an in vitro method with a combination of in situ and in vivo measurements

The achievement of maximum ruminal feed conversion into microbial biomass is a widely accepted concept of ruminant nutrition because high microbial efficiency improves microbial protein supply to the small intestine and, proportionally, reduces fermentative gaseous carbon losses (Beever 1993; Leng 1993). It has recently been demonstrated that differences between forages in in vitro microbial efficiencies, i.e. differences in the proportion of fermented substrate incorporated into microbial biomass, could be determined by a combination of in vitro gas volume measurements with a concomitant evaluation of the amount of substrate truly degraded during 24 h of incubation (for review see Blümmel et al. 1997a). It has been pointed out that these two measurements are not synonymous but complementary. The measurement of degradability is a modification by Goering and van Soest (1970) of the Tilley and Terry (1963) method to remove any residual microbial biomass from the undegraded substrate, thus allowing the calculation of the total amount of substrate dissimilated into all fermentation products, i.e. microbial biomass, short chain fatty acids (SCFA) and gases. In contrast, the gas volume measurement indicates how much of the fermented substrate was used for the formation of SCFA and gases since these two fermentation products are stoichiometrically very closely associated (Blümmel et al. 1997a). The measurements described above were performed after 24 h of incubation because the analytical approach used required all substrate solubles to already be fermented at the time of the residue determination; they would otherwise be removed by the treatment, leading to an overestimation of substrate degradability. On the other hand the determination should not be conducted too far beyond the microbial peak yield in order to minimize the possible contribution of microbial lysis to gas production. The ratio of truly degraded substrate to the gas volume thereby produced in 24 h was termed partitioning factor (PF). This factor denotes the substrate specific variation of in vitro microbial efficiency. The concept of the PF value was derived from, and applied to, crop residues of temperate and tropical origin (Blümmel et al. 1997b) and to Mediterranean grass and legume hays (Blümmel and Bullerdieck 1997). Forages with a high PF value, i.e. proportionally low gas production per unit of substrate degraded, were related to higher voluntary feed intakes than those with a low PF value. High in vitro PF values were also associated with high excretion of renal purine derivatives in Malawian goats fed maize stover leaves with different PF values but similar digestibilities (Mgomezulu and Blümmel 1996). For 61 straws and hays examined, the production of 1 ml of gas was associated with the concomitant in vitro true degradability of 2.74 to 4.65 mg of substrate i.e. PF values ranged from 2.74 to 4.65 mg/ml of which a minimum of 2.20 mg (Blümmel et al. 1997a) were required for the formation of acetate, propionate, butyrate and fermentative CO 2 , CH 4 and H 2 O, the latter produced upon the reduction of CO 2 to CH 4 as follows: CO 2 +8H > CH 4 +2H 2 O. Carbon dioxide also arises from buffering the SCFA but this CO 2 is, contrary to the fermentative CO 2 , not derived from the incubated substrate but from the buffer medium. Substrate degraded on top of these requirements (2.20 mg/ml) was available for microbial cell synthesis; proportionally more microbial biomass has been synthesized in fermentations having high PF values (Blümmel et al. 1997a). The objective of this work was to further investigate the relationship between in vitro microbial efficiencies as estimated by the PF value and in vivo microbial efficiencies using four perennial rye‐grass silages harvested at different stages of maturity. These silages were fed to steers and the in vivo microbial protein synthesis was estimated from renal allantoin excretion, thus providing the opportunity for the comparison of in vitro and in vivo microbial efficiencies. Some aspects of the in vivo work have been reported by Philipczyk et al. (1996) and Schröder et al. (1997).