in forages whose HFkhdiffer (Ellis et al.,
1991, 1994, and Figure 18 of Ellis et al.,
1999). Biological mechanisms involving
buoyancy and the self-regulation nature of
physical regulation have been proposed
whereby HFkhmay regulate^2 UF–e(Ellis et
al., 1991). Another example is the rela-
tively constant degree of colonization of
ruminal efflux (16.3 g microbial CP g^1
ruminal efflux OM; Table 16.4; and Tables
5 and 9 of Ellis et al., 1999). This level of
microbial CP in the ruminal efflux OM
approximates that attained on forage
residues subsequent to an in situincubation
for ruminal residence times comparable
with those reviewed in Fig. 16.4 and else-
where (Van Milgen et al., 1993; Figure 20
of Ellis et al., 1994).
In contrast to earlier proposals of a con-
stant microbial CP in the ruminal efflux OM,
significant trends occur in microbial CP con-
centrations of ruminal efflux OM for the
current data set. These trends appeared most
related to the dietary level of fermentable
carbohydrate and protein in the diet and
may be due to obscuring differences in dilu-
tion of microbial CP by different non-
microbial protein OM components. The
dynamic and complex interactions that
determine intraruminal synthesis of micro-
bial protein are of paramount significance.
However, a less direct but less complex
approach involves estimating ruminal efflux
yield of microbial protein and inferring the
causal intraruminal process.
Ruminal efflux yield of microbial proteinRelationships between ruminal dynamics
and efficiency of ruminal efflux yield of
microbial protein, MCPE (g of ruminal
efflux of MCP g^1 of ruminally hydrolysed
carbohydrate) were investigated. These rela-
tionships were of necessity restricted to data
supporting the calculation of ruminally
hydrolysed carbohydrates. The difference in
NDF intake and NDF effluxing the rumen
was assumed to represent hydrolysed
structural carbohydrates.
Relationships involving MCPE versus
various candidate independent variables
are shown in Fig. 16.9. None of the
candidate independent variables alone
exhibited impressive correlations, including
flux of hydrolysed carbohydrates (Fig.
16.9D), rate of hydrolysis of component
carbohydrates HF and NSC (Figs 16.9B and
C) or effective escape rate from the two
ruminal pools, UF–e (Fig. 16.9A). These
candidate independent variables are
utilized in various models to predict MCPE
and/or yield. In contrast, the proportions of
ruminally hydrolysed protein to ruminally
hydrolysed total carbohydrates (RHP:HCHO)
were more highly correlated with MCPE
than that observed for other candidate
independent variables. This improvement
in correlation was due in part to ruminally
hydrolysed total carbohydrates (g of HCHO
day^1 ) being common to both the dependent
(g of ruminal efflux MCP g^1 of hydrolysed
HCHO) and the independent variable (g of
RHP g^1 of HCHO). However, the relation-
ship MCPE versus (RHP:HCHO) has con-
siderable biological logic as an expression
of flux proportions of two different types of
precursors, molecular precursors and ME.
The linear nature of MCPE versus
(RHP:HCHO) observed in Fig. 16.9F
suggests that RHP provided molecular
precursor(s) that first limited efficiency of
ruminal microbial protein synthesis
throughout the range of 0–0.9 (RHP:HCHO).
The regression equation was MCPE = 0.06
+ 0.46 (RHP:HCHO). It is interesting that,
statistically, a ratio of 0.46 of (RHP:HCHO)
was applicable throughout the entire range
of the data up to the maximal observed
efficiency value of 0.56 g of MCP g^1 of
HCHO.
Dijkstra et al.(1998) reported expected
maximal growth yields within the rumen
of 0.45 and 0.65 g of microbial DM g^1 of
hydrolysed CHO in the absence or
presence of ‘pre-formed monomers’,
respectively. It is clear that a response to
pre-formed precursors would be expected
even though the precise molecular
precursors are unidentified here. It should
also be noted that values for MCPE sum-
marized here are for ruminal efflux yield of
MCP g^1 of ruminally hydrolysed CHO. If
rumen microbial DM is assumed to contain352 W.C. Ellis et al.