values from 1.07 to 2.65 mg ml^1. The
same authors also used^15 N incorporation
as a measure of biomass yield and found
that substrates with proportionally higher
gas production showed lower^15 N incor-
poration. Experimental data thus support
the idea that the biomass yield may depend
on the type of substrate used in vitro.
We must now confront the question of
how to deal with this yield variation in
interpreting gas data. The logistic model
(Schofield et al., 1994) incorporates the
relationship between substrate digested
and gas volume into the specific rate con-
stant S. Sand are inversely proportional
to one another. The model assumes that
different forage fractions can be treated as
separate pools, each having a different, but
constant, specific rate of digestion. A large
value of S may thus mean either a fast
digestion rate or a low value for , or some
combination of the two. Independent
measurements of are required to distin-
guish among these possibilities (Blümmel
et al., 1997a). Other models assume a con-
stant gas yield (France et al., 1993) or
incorporate the gas yield into a biologically
undefined parameter (Groot et al., 1996).
To make good nutrition decisions
based on gas data, we thus need to be
aware that gas yield and microbial yield
are substrate-dependent variables. If the
gas volumefrom digestion of a given feed
is low, we should measure the gas yield
(based on organic matter disappearance)
and the microbial yield(based on microbial
mass produced) before concluding that the
feed value necessarily is low.
Plant Secondary Compounds
Tropical forage plants may contain
secondary compounds such as the poly-
phenolic tannins, the terpene- or steroid-
based saponins and the nitrogenous
alkaloids. Tannins are the major com-
ponents of this group of compounds
(Morris and Robbins, 1997). Chemical
assays for polyphenols (Hagerman and
Butler, 1989) have not shown a close
correlation with biological activity as
measured by inhibition of microbial
growth or by effects on in vitro digestion
(Nelson et al., 1997). In vitro digestion
assays offer an important additional tool to
assess the feed value of tropical forages.
The usual way to investigate tannin
effects has been to digest tannin-containing
forages in vitroin the presence and absence
of binding agents such as polyvinyl-
polypyrrolidone (PVP) or polyethylene
glycol (PEG). These agents are believed to
bind the tannins preferentially and make
them unavailable to react with competing
targets such as microbial cells and extra-
cellular enzymes. The inhibitory effect of
the tannins is thus measured as the differ-
ence in gas produced with and without the
agent. PEG has replaced PVP as the more
effective tannin-binding agent (Khazaal et
al., 1996). Makkar et al. (1995) used a
syringe method and added an approxi-
mately equal weight of PEG (average mol
wt 6000 kDa) to the tannin-rich feed. They
found an increase in gas production after
24 h incubation that correlated well (r=
0.95) with the protein-precipitating
capacity of the plant tannin. Temperate
browse species such as heather (Calluna
vulgaris) may show a substantial (51%)
increase in gas production in response to
PEG treatment (Tolera et al., 1997).Other Applications
Plant soluble componentsThe main advantage of the gas system, other
than the ready automation of readings, is
that it can be applied to study the digestion
of soluble materials. These studies can be
conducted using curve subtraction (e.g. for
the NDS fraction, see Plant Carbohydrate
Fractions and Nutritional Models, p. 223) or
by isolating and digesting the fraction of
interest (e.g. for the water- or ethanol-solu-
ble fraction). The results have tended to
challenge the current assumption that all
soluble material, if digestible, is digested
rapidly (Stefanon et al., 1996).
Forage ensiling converts plant soluble
sugars into acids and would thus beGas Production Methods 227