structural carbohydrates. In the second part
of this chapter, we will survey the different
uses of this technique and their limits.
Sources of Variation in the
in SaccoTechnique
The key to the usefulness of this technique
focuses on the ability to standardize it. We
will investigate the factors associated with
the use of this technique, especially the
factors likely to introduce a bias in the
range of feeds according to their rumen
degradability. In order to make discussion
of pertinent sources of variation clearer,
this chapter has been divided into three
parts, bag and sample characteristics,
ruminal environment and exchanges
between bag and rumen, and modelling
degradation kinetics.
Bag and sample characteristicsMilling
The types of cloth generally used are
polyester, nylon or dacron, with the latter
two being used most often (see review by
Huntington and Givens, 1995). However,
the weave structure of the cloth, multi-
filamentous or monofilamentous, is of
greater importance. With monofilamentous
woven cloth, the aperture size is defined
with precision, which means that the pores
are uniform and do not change with
mechanical stress. Conversely, the aperture
size of dacron threads is more variable and
seems to be affected by physical pressure
during incubation (Marinucci et al., 1992).
Another important characteristic of bags is
their pore size. It must permit the influx of
digesting agents and buffers, but prevent
the efflux of undegraded sample whilst
allowing the removal of degradation end-
products. Generally speaking, the degrada-
tion of dietary components increases with
pore size (Weakley et al., 1983). These
results were obtained with large variations
in pore size (5–50 μm). Now researchers
use bags whose pore size is between 35 and
55 μm (Huntington and Givens, 1995), and
the pore size proposed in the recom-
mended experimental procedure is
between 30 and 50 μm.
Choice of bag porosity is also condi-
tioned by the processing of the sample, and
more particularly by the fineness of grind-
ing. Grinding is carried out in order to
obtain a homogeneous sample and also
to mimic the effect of mastication.
Mastication decreases particle size and
thereby exposes more surface area.
Additionally, the crimping and crushing
action of the mastication process exposes
more area of digestible tissue within a
given particle size, and facilitates rupture
of the anatomical barriers of plant
structures to allow subsequently increasing
microbial enzyme accessibility. In a com-
parison between three forages, masticated
or not, Olubobokun et al.(1990) showed
that the mastication allowed a greater
release of soluble nutrients. For cereals,
ingestive mastication increased DM diges-
tion from 16, 26 and 30% to 53, 69 and
66% for barley, maize and wheat, respec-
tively, in comparison with whole grain
(Beauchemin et al., 1994). Incubation of
masticated feeds would be the best method
for in saccostudies, but it is also a very
laborious method. A less perfect but easier
simulation of mastication can be obtained
by grinding of feedstuff prior to ruminal
incubation.
Feed degradation rate tends to increase
with fineness of grinding, and variations
depend on rumen incubation time and
feed. The influence of grinding increases
when the incubation time is short (see
review by Michalet-Doreau and Ould-Bah,
1992), and differences between feeds are
also important. When the fineness of grind-
ing increased from 1 to 3 mm, the DM
and/or N degradability of barley was not
modified, whereas that of maize, oat, pea,
fababean or soybean meal decreased
(Nordin and Campling, 1976; Michalet-
Doreau and Cerneau, 1991). For forages,
when the grinding screen was increased
from 0.8 to 6 mm, N degradability was not
modified (Michalet-Doreau and Cerneau,
1991), but the kinetic parameters varied:
compared with the chopped material,234 P. Nozière and B. Michalet-Doreau