Ruminal degradability of corn forages depending on the processing method employed
Ruminal degradability of corn forages depending on the processing method employed
Stefan JURJANZ 1
Valérie MONTEILS 0
0 Laboratoire de Zootechnie et Qualité des Produits Animaux , ENSAT, BP 107, 31326 Castanet-Tolosan Cedex , France
1 Laboratoire de Sciences Animales, INRA-ENSAIA de Nancy , BP 172, 54505 Vandoeuvre Cedex , France
- The in sacco degradabilities of starch and fibre in corn were compared between kernel grains and the whole corn plant before and after ensiling using the nylon bag technique. The same plants were used, in order to exclude the effects of genotype or maturity at harvest. The incubation time course was carried out over 48 h on four cannulated dairy cows. The effective degradability of starch was lower in kernel grains (70.2%) than in the whole plant before (83.9%) and after (92.3%) ensiling. Starch degradation in whole plants was accelerated compared to kernel grains by a shift from the slowly degradable (from 61.3% to 31.9%) to the rapidly degradable fraction (from 35.9% to 65.6%) without significantly affecting the degradation constant rate (7.7 and 8.0% per h respectively for kernel grains and whole corn plants). The ensiling process improved starch degradation even further compared to whole fresh plants by significantly increasing the rapidly degradable fraction (80.7% versus 65.6%) and by a higher degradation constant rate (12.4% per h versus 8.0% per h). The fibre degradation was similar between grains and whole corn plants despite differences in their content and composition of NDF. However, ensiling significantly increased the rapidly degradable NDF fraction (15.2 versus 9.9%) and doubled its degradation constant rate (from 1.6% per h to 3.2% per h). This effect was probably due to improved hemicellulose degradation, because cellulose and lignin were not degraded differently between corn plants before and after ensiling. Mechanical cracking such as chopping at harvest improves ruminal starch degradation without altering fibre degradation but the ensiling process simultaneously increases the degradability of starch and fibres.
31,9 %) vers la fraction rapidement dégradable (de 35,9 à 65,6 %) sans effet significatif sur le taux
horaire de dégradation (7,7 et 8,0 % par h respectivement pour le grain et la plante entière). La
fermentation de la plante entière améliore la dégradation de l’amidon par une augmentation
significative de la fraction rapidement dégradable (de 65,6 à 80,7 %) et du taux horaire (8,0 contre
12,4 % par h). Les dégradations des fibres ont été similaires entre grains et plante entière malgré des
différences de la teneur en NDF et de sa composition. Par contre, la fermentation de la plante
augmente significativement la fraction rapidement dégradable des NDF (15,2 contre 9,9 %) et double
son taux horaire (de 1,6 à 3,2 % par h). Cet effet semble dû à une augmentation de la dégradation
de l’hémicellulose puisque aucune différence de dégradation avant ou après fermentation pour la
lignine et la cellulose n’est observée. Le broyage mécanique, comme le hachage à la récolte, améliore
la dégradation de l’amidon de maïs dans le rumen sans modifier la dégradation des fibres. Par contre,
la fermentation de la plante entière en ensilage augmente simultanément la dégradabilité de l’amidon
et des fibres.
dégradation ruminale / traitement du maïs / amidon de maïs / fibres de maïs
Corn silage is currently used as forage in
the diet of high yielding dairy cows to assure
an elevated energy supply, mainly because
of its high starch content. The starch of corn
grains harvested at maturity has been reported
as being slowly degraded by
micro-organisms in the rumen [
]. In contrast, the
starch in ensiled corn is rapidly degraded in
the rumen [
21, 23, 25
] and an increased
amount of starch degraded in the rumen has
been shown to depress fibre digestion [
Thus the risk of acidosis, at least in a
subclinical state, is present. Different
hypotheses can be put forward to explain the
acceleration of starch degradation in ensiled corn
when compared to starch from corn grains
harvested at maturity.
A first possible explanation for modified
ruminal degradation of corn is the genetic
variability between cultivars. This has been
shown not only for cell wall degradation in
the rumen [
7, 11, 35
] but also for starch
between grains of dent and flint genotypes
]. Other authors  have noted
increased ruminal starch degradability after
ensiling, independently of corn genotypes.
Secondly, the literature [
3, 8, 14, 24
reported a decreased starch degradability of
corn silage in the rumen when the maturity
of harvested plants increases. Because silage
cultivars are harvested at a less mature stage
than crop cultivars, the starch fraction in
corn silage can be degraded more rapidly in
the rumen than that found in crop cultivars.
Thus the effect of the genotype, i.e. cultivar,
and the vegetation stage at harvest on starch
degradation in the rumen has always been
demonstrated. Another explanation for a
higher starch degradation after ensiling may
be plant processing. The mechanical
cracking that takes place prior to ensiling reduces
particle size and there is an increased
ruminal degradability of ensiled corn starch
compared to the starch of identically
prepared corn plants even before ensiling .
Moreover the solubilisation of endosperm
proteins during corn fermentation has been
], and this may improve enzyme
access to starch, alongside mechanical
cracking. Starch granules (especially in corn) are
embedded in a protein matrix [
] that grows
with the maturation of the kernel [
thus protects starch grains from hydrolysis
by amylolytic enzymes [
]. Mainly made
up of zeins, the core of these protein bodies
is negatively correlated to starch
The first aim of this study was therefore
to determine whether the more rapid
degradation in the rumen of starch in ensiled corn
is due to processing, i.e. not dependent upon
differences in genotype or maturity. Finally,
the supplementation of readily fermentable
carbohydrates is known to reduce fibre
digestion in vivo [
], at least when using
rapidly degradable grains such as barley or
wheat. The second aim of this study was to
determine whether corn processing,
especially fermentation, would affect the
degradation of corn fibres and hence the nutritive
value of corn forages.
2. MATERIALS AND METHODS
2.1. Experimental design and plant growing conditions
The characteristics of ruminal degrada
tion of corn kernel grains, fresh whole corn
plants (FWCP) and ensiled whole corn
plants (EWCP) were compared using the
same raw material throughout the trial.
Corn was grown on a calcic vertic com
bisol (FAO classification). During the spring,
it was necessary to seed two cultivars (2265®
hybrid, Force Limagrain,
Saint-Quentin-Fallavier, 38, France and Etendard® hybrid,
Sémences de France, La Chapelle
d’Armentières, 59, France) because of the different
moisture conditions which prevailed in the
same field. Grains were sown in equivalent
proportions with a corn population density
of 113,000 plants per ha on May 5th and
harvested 152 days later, on October 4th, 2000.
The entire field was fertilised with 252 units
of N per ha and the rainfall during the
growing period was 402 mm. Growth and harvest
characteristics are shown in Table I. Corn
plants at silage maturity (dent stage) were
chopped with a forage harvester to a particle
size of 1 cm.
Fifteen kernels from each cultivar were
randomly sampled from the field one day
before harvest. The grains were manually
separated and weighed. Because both
cultivars were mixed in the same flat silo
(width 13 m, depth 18 m, height 4 m) at
similar proportions (as in the field), we also
mixed the two cultivars together before
incubating the grains (50:50 wt/wt). A
representative sample of the whole fresh corn
plant (FWCP) before fermentation was
collected just after filling the silo, and another
sample was taken from the same place in the
silo after fermentation for 2 weeks (EWCP)
in order to determine the chemical
In sacco measurements were carried out
using four ruminally cannulated lactating
multiparous Holstein cows housed in a free
stall barn. The animals were fed ad libitum
with a total mixed ration composed of corn
silage (55.3% of DM), wheat straw (2.8%),
cracked wheat (29.9%), soybean meal
(10.4%) and minerals (1.6%). Cows had an
average intake of 19.2 (± 2.6) kg DM per d
and a daily yield of 31.4 (± 5.1) kg milk. The
corn silage fed to them was the same as that
used for incubation samples. The animals
were adapted to the experimental diet for
2 weeks before measurements were
All feed samples were oven dried (48 h
at 65 °C) and ground (centrifugal force mill
SK100, Retsch GmbH & Co KG, Haan,
Germany, sieve 1.5 mm). The nylon bag
] was then carried out and
approximately 4 g of each tested feedstuff
were placed separately into bags (internal
dimension 9 × 14 cm, Blutex 120, Saati
France Co., Sailly Saillisel, France) with a
pore size of 50 µm.
The incubated bags were soaked and
then inserted into the rumen just before
giving access to the morning meal. The bags
were removed after 1, 3, 6, 12, 24, and 48 h
of incubation, rinsed in cold water and
washed in a washing machine (2 cycles of
5 min). Tests with different incubation times
were performed on different days to avoid
disturbing ruminal conditions by too many
manipulations on the same day. However
all bags for a given incubation time were
introduced into the rumen at the same day.
Three replicates were carried out for each
feedstuff, each cow and each incubation
time. Experimentally measured solubility
(a0) corresponds to non-degraded dry
matter loss and the soluble fraction, measured
by the difference between non-incubated
feedstuff and samples (n = 4) in nylon bags
spun in cold water in a washing machine
(2 cycles of 5 min).
After incubation, all the bags were
freeze-dried and weighed. The bags of the
same feedstuff and the same incubation time
in all cows were pooled prior to chemical
2.3. Chemical analyses
The samples of each tested feedstuff
before incubation and their pooled residues
after incubation were analysed for their
starch content (enzymatic method, AFNOR
NF V 18–121, 1997). The fibre fraction of all
samples was described by analysing NDF,
] and ADL [
], in order to calculate
cellulose and hemicellulose levels
according to the method described by Staples [
In addition, ash (6 h at 550 °C) and crude
protein (Kjeldahl method) were analysed in
feedstuffs before incubation. The chemical
composition of all feedstuff samples prior
to incubation is shown in Table II.
2.4. Data analyses
The DM before and after incubation was
averaged for each cow, a given feedstuff
and a given incubation time. These DM
amounts were then multiplied by the
contents in starch or fibre of the pooled sample
in order to calculate the disappearance rates
of DM, starch, NDF, cellulose and
hemicel1 cultivars of used kernel grains;
2 FWCP fresh whole corn plant;
3 EWCP ensiled whole corn plant;
4 cellulose = NDF – ADF according to Staples et al.
5 hemicellulose = ADF – ADL according to Staples
et al. [
lulose individually for each cow, each
feedstuff and at each incubation time.
The values for experimentally measured
solubility (a0) were statistically compared
using the Student t test.
The disappearance kinetics obtained for
each feed were then fitted to an exponential
model according to the method described
by Ørskov & McDonald [
(am: rapidly degradable fraction, bm: slowly
degradable fraction, cm: degradation
constant rate exponentially reduced over time)
were estimated by an iterative procedure of
STAT-ITCF (version 5.0, 1991, Institut
Technique des Céréales et des Fourrages –
renamed ARVALIS Institut du Végétal,
Paris) in order to determine the smallest
sum of squares after convergence. A
statistical difference between feedstuffs for a
given parameter was concluded if the
values were outside the calculated confidence
interval of ± 2δ (i.e. P < 0.05). The fibre
disappearance curves (i.e. NDF), differed from
those of starch: a latency period of 3 h was
observed (Fig. 1). Thus, the exponential
model has to be adjusted to the values only
after this lag time, especially for cellulose
and hemicellusose fractions.
The effective degradability (eD) of DM
and all fractions were calculated using
average parameters calculated by the
exponential model for a given feedstuff at a ruminal
outflow rate of 0.06 per h, as is classically
This approach privileged the use of
observed variability to statistically compare
the model parameters (i.e. am, bm, cm). Their
resulting effective degradability is
indicated without statistical comparison.
Before incubation, the dry matter content
of kernel grains ranged from 49.7 to 56.4%
depending on the cultivar, which confirms
that the plants were harvested at silage
maturity. The dry matter of the whole corn
plants before incubation increased from
33.6% before to 37.6% after ensiling (Tab. II).
The kernel grains of both cultivars were
similarly composed, mainly of starch
(respectively 78 and 76.9% of the DM). Moreover,
the kernel grains of both cultivars contained
nearly 10% total fibres (i.e. NDF), 7.8%
crude protein and 1.3% ash (Tab. II). The
starch content of the plants before and after
ensiling was very similar (respectively 37.2
and 37.0% of the DM, Tab. II). But plants
after ensiling had lower contents of total
fibres, i.e. NDF (40.1 vs. 46.3%), crude
protein (6.4 vs. 6.7%) and ash (3.0 vs. 4.8%).
Proportions of cellulose and hemicellulose
within the fibre fractions were similar for
both forage forms (Tab. II).
The incubation kinetics show that
experimentally measured DM solubility of
kernels (37.3%) was similar to that of the whole
fresh corn plant (38.4%) but significantly
lower than the DM solubility of ensiled corn
(57.4%, P < 0.05; Tab. III). These
experimentally measured values (a0) were
confirmed by the values for the rapidly degraded
fraction (am) calculated using the model.
Thus, the rapidly degradable DM of grains
or whole fresh plants was significantly
lower than this fraction in ensiled corn
(36.0% and 39.0% vs. 55.7% respectively,
P < 0.05). The slowly degraded fraction
(bm) was significantly (P < 0.05) higher in
grains (57.6%), followed by FWCP (41.6%)
Degradation parameter values in the same column within the same fraction and without a common
superscript differ significantly at P < 0.05;
FWCP: fresh whole corn plant;
EWCP: ensiled whole corn plant;
a0: experimentally measured solubility (%);
am: rapidly degradable fraction (%) calculated by the model [
bm: slowly degradable fraction (%) calculated by the model [
cm: rate of degradation (per h) calculated by the model [
and then EWCP (27.1%). The degradation
rate (cm) of grain DM was significantly
higher (7.3% per h, P < 0.05) than that in
whole corn plants, ensiled or not (4.2 or
2.8% per h respectively).
The experimentally measured solubility
of starch was the highest in ensiled corn
(81.2%) followed by the whole fresh plant
(63.0%) and was the lowest for starch in
grains (34.7%). This significant hierarchy
of starch solubility between tested
feedstuffs was also observed for their rapidly
degradable starch fraction (am) calculated
by the model (Tab. III). The slowly
degradable fraction (bm) varied significantly in
the opposite order: the highest value was
seen for starch in grains (61.3%) followed
by the whole fresh plant (31.9%) and finally
starch in ensiled corn (17.2%). The
degradation rates (cm) of grain starch and starch
from the whole fresh plant were very similar
(respectively 7.7% and 8.0%) and
significantly (P < 0.05) lower than those seen for
starch in ensiled corn (12.4%).
Nevertheless, the degradation constant rate (cm) for
starch of the three feedstuffs was
diametrically opposite that of the slowly degradable
fraction: the smaller the slowly degraded
fraction, the steeper the slope of the
degradation curve (Tab. III), and the starch in all
forms was almost completely degraded
in sacco after 48 h of incubation (Fig. 1).
FWCP: fresh whole corn plant;
EWCP: ensiled whole corn plant;
a0: experimentally measured solubility (%);
am: rapidly degradable fraction (%) calculated by
the model [
bm: slowly degradable fraction (%) calculated by
the model [
cm: rate of degradation (per h) calculated by the
Since the values of fibre disappearance
kinetics were adjusted to the exponential
model after a lag time of 3 h (see Sect. 2.4),
the alignment between experimentally
measured solubility (a0) and rapidly degraded
fractions (am) was less evident for all fibre
fractions (Tab. III and IV). The large
confidence interval of the model parameters
and the very high standard error, mainly for
NDF in grains (SE = 18.8%, Tab. III),
limited the statistical signification of observed
differences. However, several trends for the
kinetics of fibre degradation could be
reported. The rapidly degraded NDF
fractions of grains (11.3%) and whole fresh
plants (9.9%) seemed to be lower than this
NDF-fraction in ensiled corn plants (15.2%).
This tendency was confirmed by
significantly (P < 0.05) lower NDF-solubility in
grains and fresh plants when compared to
NDF in ensiled plants. In contrast, the
slowly degradable fraction was significantly
(P < 0.05) lower in ensiled plants (53.6%)
than in grains (70.7%) or the whole fresh
plant (74.6%). As seen previously for starch,
the degradation constant rate (cm) of NDF
in grains and in whole fresh plants was quite
similar (respectively 2.2 and 1.6% per h),
and the slope seemed to be lower than for
NDF in ensiled corn (3.2%).
The fact that grains exhibited a low fibre
content (before incubation: <4% ADF and
approximately 1% ADL in the DM, Tab. II)
limited the interest of monitoring the
degradation kinetics of the different fibre
fractions in grains. Therefore, these degradation
kinetics were only determined in forages. In
both forage forms, no increase in ADL
disappearance rates (data not shown) was seen,
since lignin (i.e. ADL), is generally not
degraded in the rumen [
]. They remained
close to the initial values (i.e.
experimentally measured solubility) throughout the
time course, but differed between fresh
(31.2% ± 7.6) and ensiled whole corn plants
(51.8% ± 4.9). The degradation models were
calculated for cellulose and hemicellulose
using the same lag time as that described
above for NDF.
The rapidly degradable fraction of
cellulose appeared to be higher in fresh corn
plants than in ensiled plants (a0 + 5.2%, am
+ 13.3%; Tab. IV) but the threshold of
significance was not reached. The slowly
degradable cellulose fractions of both
forages were close (52.9 and 60.4%
respectively for FWCP and EWCP). Finally, the
degradation constant rate of cellulose in
ensiled corn tended to be slightly higher
than that seen in whole fresh corn plants
(+ 1.4% per h). Thus, the calculated
degradation curves of cellulose in both corn
forms were very similar (Fig. 2) and
differences between model parameters remained
within their confidence intervals. Indeed,
the exponential models for the degradation
of cellulose and particularly for
hemicellulose did not reach their asymptotes during
the time course (Fig. 2). Indeed, the classically
applied exponential model [
these degradation curves less precisely than
a linear model would do. So, the
exponential model for hemicellulose degradation in
corn plants showed only tendencies and no
difference reached the signification
threshold. The experimentally measured
solubility (a0) and slowly degradable fraction (bm)
of hemicellulose were similar in both
forages. The lag time used for the exponential
model enhanced an artefact of negative
values for the rapidly degradable fraction (am)
of hemicellulose in both forms (Tab. IV).
Indeed, these values tended to reflect a very
low baseline point for the fitted model curve
rather than true degradation features. The
degradation constant rate of hemicellulose
in ensiled corn was apparently higher than
that seen in fresh corn plants (3.6 vs. 1.9%
per h respectively), which was well
illustrated by the steeper incline of the
degradation curve for EWCP when compared to
FWCP (Fig. 2).
The effective degradability of DM was
similar in grains and ensiled corn (67.6 and
66.8% respectively, Tab. V) but was higher
than in whole fresh plants (52.1%). The
lower solubility of DM in grains compared
to ensiled corn was counterbalanced with a
smaller slowly degradable fraction and an
increased degradation constant rate. The
effective degradability of starch was the
highest for ensiled corn (92.3%) followed
by whole fresh plants (83.9%). Kernel starch
reached only 70.2% of effective
degradability (Tab. V). The effective degradability of
NDF ranged from 25.2% (whole fresh plants)
to 34.0% (ensiled plants), with an
intermediate value of 30.5% for grains (Tab. V).
Figure 1 illustrates this tendency of lower
effective NDF degradation in FWCP
compared to kernels or ensiled plants. Thus the
effective degradability of cellulose using
the parameters calculated within a broad
confidence interval was slightly higher in
fresh than in ensiled corn plants (+4.9%).
The higher degradation constant rate of
hemicellulose in ensiled corn enhanced an
effective degradability, which was almost
double that seen in fresh corn plants
1: effective degradability (%) calculated using the
equation a + bc/(c + k), where k = outflow rate
assumed to be 0.06 per h [
FWCP: fresh whole corn plant;
EWCP: ensiled whole corn plant.
4.1. Starch degradation
Corn starch disappeared completely after
48 h of incubation in the rumen whichever
processing method was employed.
However, the parameters of the degradation
model differed considerably between corn
In the literature, a broad range of values of
starch degradation parameters are reported
in corn grains depending on genotype [
], maturity at harvest [
7, 14, 24, 26
and sample preparation [
18, 24, 25
observed a high rapidly degradable fraction
but a relatively small slowly degradable
fraction of starch compared to others 
working at similar maturity (37% of DM at
harvest). A high grinding fineness of
samples has been shown [
5, 20, 24, 36
increase starch lost through the bag pores and
the grinding of our samples over a 1.5 mm
sieve may, because of the small particle
size, have favoured these particulate losses.
This methodological weakness affected all
tested corn forms similarly in our trial, and
possibly led to an overestimation of ruminal
starch degradability under our experimental
conditions. We therefore preferred to focus
the discussion of our results on differences
in degradation between corn forms, rather
than their absolute values.
Nevertheless, corn starch in the whole
plant was more rapidly degraded than in
kernel grains. Since the same plants and
vegetation stage at harvest were used, an
explanation due to differences in genotype or
maturity can be excluded. The low starch
content in plant residue (2% of DM, [
us conclude that starch in the whole corn
plant is nearly identical with starch in the
grains. The more rapid degradation in the
FWCP could be due to the chopping of
the plant at harvest as suggested in the
]. The double mechanical
treatment during harvest and grinding prior to
ruminal incubation could have weakened
the protective layers of starch granules and
the fine grinding in our trial could amplify
this effect by increased particulate losses.
Another hypothesis to explain for the more
rapid degradation of starch in both forage
forms was the higher moisture level of the
whole plant when compared to dry kernel
grains. The starch structure would be
softened by moisture and therefore favour
enzymatic hydrolysis, even if the forages were
dried during sample preparation prior to
incubation. Indeed, the literature reports an
increase in the soluble fraction and the
fractional rate of degradation in the rumen for
high moisture corn versus dry cracked [
or dry rolled [
] corn, even though the effects
of the maturity stage may distort these
Thus, double cracking and/or a softened
starch structure lead to a higher rapidly
degradable starch fraction in the whole fresh
plant when compared to the starch in grains.
In our trial, the higher starch degradability
observed for the fresh plant was mainly due
to a shift from slowly degradable starch to
the rapidly degradable fraction which let us
privilege the mechanical hypothesis.
In agreement with other studies [
], starch in ensiled corn was degraded
very rapidly. Ensiling increases ruminal
digestion of cornstarch more than
mechanical treatments such as grinding, rolling or
]. Indeed, ensiling enhanced a
second shift of slowly degradable starch to
the rapidly degradable fraction (Tab. III),
and increases the effective degradability of
]. In our trial we also observed
an increase of 4.4% per h in the degradation
constant rate of starch in ensiled corn when
compared with the whole fresh plant. These
observations should be linked to the
fermentation process, because both forages
underwent identical chopping during
harvest and sample preparation. Moreover,
their moisture contents were also similar
(Tab. II). Baron [
] reported a partial
solubilisation of endosperm proteins in corn
grains after ensiling which involved a
reduction in the protective aptitude of the protein
matrix. It would be interesting to test whether
zeins, the main factor causing reduced starch
], are sensitive to variations in pH
and if starch degradation can so be improved.
This hypothesis is strengthened by the fact
that enzymes used in analytical procedures
that hydrolyse starch usually require an acid
pH to ensure their optimum activity.
4.2. Fibre degradation
The velocity of starch degradation in the
rumen determines the in vivo postprandial
pH drop [
] and a moderate drop in pH
would favour the cellulolytic activity of
]. But this widely reported
negative relationship between starch and
fibre degradation [
2, 10, 30
], especially in
corn silage [
], cannot be studied in sacco.
Nevertheless, determining the effects of plant
processing on corn fibre degradation in a
given ruminal environment would clarify
their nutritive value.
Unlike the starch fraction, fibres (i.e.
NDF) in grains and in the whole corn plant
differ in terms of both their quantity (NDF
content) and their quality (fibre
composition). In our study, fibres in kernel grains
represented less than 10% of the dry matter
in this organ, versus approximately 4-fold
more in whole corn plant forms (Tab. II).
The indigestible lignin (i.e. ADL)
represented nearly 10% of the fibre fraction (i.e.
NDF) regardless of the considered corn
form (Fig. 3). But kernel fibres of both
cultivars were mainly composed of cellulose
and in the whole plant cellulose and
hemicellulose were present at similar
proportions in the fibre fraction (Fig. 3).
The experimentally measured solubility
data (a0) for all fibre fractions were higher
than their corresponding rapidly degradable
fraction as determined by the model (am;
Tab. III, Tab. IV). This difference may have
been due firstly, to an overestimation of a0
because of particulate losses (see above),
and secondly, to the lag time applied in the
model, which lowered the calculated value
of the rapidly degradable fraction (am).
The degradable fractions of NDF in
grains were similar to those found in the
whole fresh plant, despite the different
composition of NDF in both feedstuffs.
Chopping at harvest did not seem to affect fibre
degradation. It can be concluded that
mechanical effects such as chopping will similarly
affect cellulose and hemicellulose
degradation in the rumen. An evoked effect of the
higher moisture content in the whole plant
compared to grains on fibre degradation is
unknown. The tendency of a slightly lower
effective degradability in the plant before
ensiling (–5.3%) should be considered with
caution because of the lack of significance.
Thus, no difference of fibre degradation
between kernels and the whole fresh plant
has been observed in sacco despite initial
variations in its composition.
It was surprising that before incubation
all the characteristics of the fibre fraction in
EWCP exhibited lower initial values than in
FWCP (Tab. II). The fibre levels in EWCP
were low but within the usual range
observed for this site (38 to 48% NDF in
DM) and in line with the high starch
contents seen in corn silage harvested at later
], such as at the dent stage. No
data were found in the literature to explain
for any disappearance of fibre during
ensiling. FWCP values therefore seemed to be
quite high [
]. In spite of the numerous
precautions taken during sampling, we
advance the hypothesis that some sampling
errors, linked to the heterogeneity of feed
stock, were the most likely explanation for
the higher fibre contents found in FWCP
when compared with the levels seen after
Fibre degradation (i.e. NDF) in ensiled
plants was higher than in whole fresh plants:
there was a significantly increased
solubility (a0), confirmed by a similar trend with
respect to rapidly degradable NDF (am).
The lower slowly degradable fraction (bm)
in ensiled plants reflected the initially more
rapid disappearance of NDF at the
beginning of the time course for ensiled plants.
This difference between fresh and ensiled
plants was maintained throughout the time
course (Fig. 1). The doubled degradation
constant rate slightly increased the fibre
degradability in favour of ensiling, when
compared to fresh plants. The literature [
has generally associated a more acid
environment in the rumen with a reduced fibre
degradation via a shift in the microbial
population balance [
] and a more rapid
passage rate by the small size of concentrate
]. On the contrary, we observed
an increased fibre degradation in the
acidified (i.e. ensiled) compared to fresh plants.
Passage rate and particulate size were
standardised in sacco. These factors can
therefore not explain in sacco differences in
ruminal fibre degradation between plants
before and after ensiling, in agreement with
]. Nevertheless, it is possible that
the acidity provoked by microbial
fermentation of fresh plants in the rumen would
affect fibre structures differently than the
acidity created by fermentation in a silo.
Perhaps fibre structures would be weakened
in the silo prior to degradation in the rumen.
Indeed, acid chemicals such as ADF [
enable the solubilisation of some fibre
Within the fibre fraction, the
disappearance of lignin was too small to conclude as
to true degradation. Cellulose degradation
varied little and in a non-significant manner
between the two forage forms. The
calculated effective degradation of cellulose in
our corn silage of 28.7% was close to the
24% reported by Jochmann [
] in similar
conditions (36% DM at harvest, kp = 6%
per h) but for a more fibrous silage (49%
NDF, 25% ADF). The main difference in
fibre degradation in our study was due to
the hemicellulose fraction that tended to be
degraded at an increased degradation
constant rate in ensiled rather than fresh plants,
leading to a higher effective degradability.
Indeed, degradation of the very regular
cellulose structure (glucose monomers with
β-glucosidic links only) did not vary
notably between fresh and ensiled plants, but the
very heterogeneous structure of
hemicellulose (varying monomer composition and
numerous lateral chains) seemed to be more
sensitive to weakening by the fermentation
in the silo.
Thus the ensiling process increased the
total fibre degradation in the rumen. This
difference in favour of ensiling seemed to
be due to a higher degradation of
hemicellulose since cellulose and lignin
degradation did not differ between fresh and ensiled
plants. Complementary studies in vivo should
be focused on the question if an apparently
improving effect of ensiling on starch and
fibre degradation of corn plants in the rumen
would be lowered by changes in microbial
populations and modified passage rate.
Cornstarch in the plant before and espe
cially after ensiling was more rapidly
degraded than that found in kernel grains.
The higher degradability of starch in the
whole corn plant compared to kernel grains
was mainly due to the increased rapidly
degradable fraction. More work is
necessary to elucidate whether the mechanical
cracking of grains during harvest or a
softened starch structure by higher moisture
level in the whole fresh plant were
responsible for improved starch degradability. The
fermentation of cornstarch during ensiling
increased its degradation. Acidification
probably solubilises the protective protein
layer surrounding the starch granules. The
sensitivity of endosperm proteins to pH
variations should be tested to see whether
acidification could improve starch access, even
in dry corn grains.
Fibre degradation did not vary between
kernels and whole plants despite the
different composition of NDF. Thus chopping at
harvest does not seem to modify in sacco
degradation. These results should be
confirmed using more roughly ground samples
so as to reduce particulate losses and less
variable parameters of the degradation model.
Nevertheless, ensiling increased in sacco
NDF degradation in the rumen and mainly
the degradation of the hemicellulose
fraction was higher in plants after than before
ensiling. The responsible mechanisms should
be determined in studies focusing on the
resistance of hemicellulose in fresh plants
to pH variations or on the effect of
microorganisms specific of the ensiling process.
Mechanical cracking such as chopping at
harvest improves ruminal starch
degradation without altering fibre degradation.
Even though these results need
confirmation in vivo, higher starch availability by
cracking and similar fibre degradation can
be a way to improve the energy furniture to
the animals while limiting the risk of
acidosis. The ensiling process increases the
degradability of starch and fibres. In spite
of the better nutritive value of ensiled corn
plants, this process would also enhance a
higher risk of acidosis. Therefore, the fibre
supply of corn silage to the diet should not
 Alvarez H.J. , Santini F.J. , Rearte D.H. , Elizalde J.C. , Milk production and ruminal digestion in lactating dairy cows grazing temperate pastures and supplemented with dry cracked corn or high moisture corn , Anim. Feed Sci. Tech . 91 ( 2001 ) 183 - 195 .
 Archimède H. , Sauvant D. , Schmidely P. , Quantitative review of ruminal and total tract digestion of mixed diet organic matter and carbohydrates , Reprod. Nutr . Dev. 37 ( 1997 ) 173 - 190 .
 Bal M.A. , Coors J.G. , Shaver R.D. , Impact of the maturity of corn for use as silage in the diets of dairy cows on intake, digestion, and milk production , J. Dairy Sci . 80 ( 1997 ) 2497 - 2503 .
 Baron V.S. , Stevenson K.R. , Buchanan-Smith J.G. , Proteolysis and fermentation of corngrain ensiled at several moisture levels and under several stimulated storage methods , Can. J. Anim. Sci . 66 ( 1986 ) 451 - 461 .
 Cerneau P. , Michalet-Doreau B. , In situ starch degradation of different feeds in the rumen , Reprod. Nutr . Dev. 31 ( 1991 ) 65 - 72 .
 Cooper R.J. , Milton C.T. , Klopfenstein T.J. , Scott T.L. , Wilson C.B. , Mass R.A. , Effect of corn processing on starch digestion and bacterial crude protein flow in finishing cattle , J. Anim. Sci . 80 ( 2002 ) 797 - 804 .
 Flachowsky G. , Peyker W. , Schneider A. , Henkel K. , Fibre analyses and in sacco degradability of plant fractions of two corn varieties harvested at various times , Anim. Feed Sci. Tech . 43 ( 1993 ) 41 - 50 .
 Flachowsky G. , Einsatz von maisstärkehaltigen Futtermitteln, Kraftfutter 6 ( 1994 ) 224 - 232 .
 Gallais A. , La physiologie du maïs , INRA (Ed.), Versailles, France, 1983 , 574 p.
 Grant R.J. , Mertens D.R. , Influence of buffer pH and raw corn starch addition on in vitro fiber digestion kinetics , J. Dairy Sci . 75 ( 1992 ) 2762 - 2768 .
 Hunt C.W. , Kezar W. , Vinande R. , Yield, chemical composition, and ruminal fermentability of corn whole plant, ear, and stover as affected by hybrids , J. Prod. Agric . 75 ( 1992 ) 852 - 867 .
 Huntington G.B. , Starch utilization by ruminants: from basics to the bunk , J. Anim. Sci . 75 ( 1997 ) 852 - 867 .
 Jarrige R. , Alimentation des bovins, ovins et caprins , INRA (Ed.), Versailles, France, 1988 , 471 p.
 Jochmann K. , Lebzien P. , Daenicke R. , Flachowsky G. , Zum Einfluss des Reidestadiums von Maispflanzen und des Einsatzes von Milchsäurebakterien bei ihrer Einsilierung auf die Nährstoffumsetzungen im Verdauungstrakt von Milchkühen , J. Anim. Physiol. An. N. 82 ( 1999 ) 178 - 192 .
 Martin C.B. , Michalet-Doreau B. , Variations in mass and enzyme activity of rumen microorganisms: Effect of barley and buffer supplements , J. Sci. Food Agr . 67 ( 1995 ) 223 - 228 .
 McAllister T.A. , Phillippe R.C. , Rode L .M., Cheng K.J., Effect of the protein matrix on the digestion of cereal grains by ruminal microorgansims , J. Anim. Sci . 71 ( 1993 ) 205 - 212 .
 Michalet-Doreau B. , Vérité R. , Chapoutot P. , Méthodologie de mesure de la dégradabilité in sacco de l' azote des aliments dans le rumen , Bull. Tech. CRZV INRA Theix 69 ( 1987 ) 5 - 7 .
 Michalet-Doreau B. , Philippeau C. , Doreau M., In situ and in vitro ruminal starch degradation of untreated and formaldehyde-treated wheat and maize , Reprod. Nutr . Dev. 37 ( 1997 ) 305 - 312 .
 Monteils V. , Jurjanz S. , Colin-Schoellen O. , Blanchart G. , Laurent F. , Kinetics of ruminal degradation of wheat and potato starches in total mixed rations , J. Anim. Sci . 80 ( 2002 ) 235 - 241 .
 Nozière P. , Flors S. , Rémond D. , Philippeau C. , Michalet-Doreau B. , Prise en compte de la taille des particules dans la détermination de la dégradation in sacco de l'amidon, 9e Renc . Rech. Ruminants, 2002 , p. 335 .
 Offner A. , Bach A. , Sauvant D. , Quantitative review of in situ starch degradation in the rumen , Anim. Feed Sci. Tech . 106 ( 2003 ) 81 - 93 .
 Ørskov E.R. , McDonald J., The estimation of protein degradability in the rumen from incubation measurement weighted according to rate of passage , J. Agr. Sci . 92 ( 1979 ) 499 - 503 .
 Owens F.N. , Zinn R.A. , Kim Y.K. , Limits to starch digestion in the ruminant small intestine , J. Anim. Sci . 63 ( 1986 ) 1634 - 1648 .
 Philippeau C. , Michalet-Doreau B. , Influence of genotype and stage of maturity of maize on rate of ruminal starch degradation , Anim. Feed Sci. Tech . 68 ( 1997 ) 25 - 35 .
 Philippeau C. , Michalet-Doreau B. , Influence of genotype and ensiling of corn grain on in situ degradation of starch in the rumen , J. Dairy Sci . 81 ( 1998 ) 2178 - 2184 .
 Philippeau C. , Le Deschault de Mondredon F ., Michalet-Doreau B. , Relationship between ruminal starch degradation and the physical characteristics of corn grain , J. Anim. Sci . 77 ( 1999 ) 238 - 243 .
 Philippeau C. , Landry J. , Michalet-Doreau B. , Influence of the protein distribution of maize endosperm on ruminal starch degradability , J. Sci. Food Agr . 80 ( 2000 ) 404 - 408 .
 Sauvant D. , Chapoutot P. , Archimède H. , La digestion des amidons par les ruminants et ses conséquences , INRA Prod. Anim . 7 ( 1994 ) 115 - 124 .
 Staples C.R. , Fernando R.L. , Fahey G.C. , Berger L.L. , Jaster E.H. , Effects of intake of a mixed diet by dairy steers on digestion events , J. Dairy Sci . 67 ( 1984 ) 995 - 1006 .
 Tamminga S. , Van Vuuren A.M. , Van Der Keulen C.J. , Ketelaar R.S. , Van Der Togt P.L. , Ruminal behaviour of structural carbohydrates, non-structural carbohydrates and crude protein from concentrate ingredients in dairy cows , Neth. J. Agr. Sci . 38 ( 1990 ) 513 - 526 .
 Valentin S.F. , Williams P.E.V. , Forbes J.M. , Sauvant D. , Comparison of the in vitro gas production technique and the nylon bag degradability technique to measure short- and longterm processes of degradation of maize silage in dairy cows , Anim. Feed Sci. Tech . 78 ( 1999 ) 81 - 99 .
 Van Soest P.J. , Use of detergents in analysis of fibrous feeds. 2. A rapid method for the determination of fiber and lignin , J. Assoc. Off. Agr. Chem . 46 ( 1963 ) 829 - 835 .
 Van Soest P.J. , Wine R.H. , Use of detergents in analysis of fibrous feeds. Determination of plant cell-wall constituents , J. Assoc. Off. Agr. Chem . 50 ( 1967 ) 50 - 55 .
 Van Soest P.J. , Nutrional Ecology of the Ruminant, Durham and Downey Inc . (Eds.), Portland OR , USA, 1982 , 374 p.
 Verbic J. , Stekar J.M.A. , Resnik-Cepon M. , Rumen degradation characteristics and fibre composition of various morphological parts of different maize hybrids and possible consequences for breeding , Anim. Feed Sci. Tech . 54 ( 1995 ) 133 - 148 .
 Wadhwa M. , Dharam P. , Kataria P. , Bakshi M.P.S. , Effect of particle size of corn grains on the release of nutriments and in sacco degradability , Anim. Feed Sci. Tech . 72 ( 1998 ) 11 - 17 .