In Vitro Methods for Measuring the Dry Matter Digestibility of Ruminant Feedstuffs: Comparison of Methods and Inoculum Source

Abstract

We conducted this study to evaluate the new in vitro system, DAISY", to determine dry matter (DM) digestibility in ruminant feedstuffs. Results from the DAISY" were compared to those obtained by the traditional Tilly and Terry method. The traditional method buffer was used for both methods. We also compared two sources of rumen inoculum from sheep and dairy cows. Seventeen different feeds were tested, grouped into roughage, concentrate, and CP supplements. The experiment was replicated on two different occasions for all feeds and the two sources of inoculum. The source of inoculum and the time at which it was collected had no effect on the in vitro DM digestibility of the feedstuffs in either of the methods. The DAISY" DM digestibility value compared well with the traditional method values for the roughage group; however, for some feedstuffs in the concentrate and CP supplement groups, the DAISY" values were significantly higher than the traditional method values. Regression analysis of the feeds that resulted in similar values with the two methods revealed that the DAISY" method can be used to predict in vitro digestibility with relatively small variation.

Full text

NUTRITION, FEEDING, AND CALVES
In Vitro Methods for Measuring the Dry Matter Digestibility
of Ruminant Feedstuffs: Comparison of Methods
and Inoculum Source
S. J. Mabjeesh, M. Cohen, and A. Arieli
Department of Animal Sciences,
The Faculty of Agricultural, Food,
and Environmental Quality Sciences,
P.O. Box 12, Rehovot 76100, Israel
ABSTRACT
We conducted this study to evaluate the new in vitro
system, DAISY
II
, to determine dry matter (DM) digest-
ibility in ruminant feedstuffs. Results from the DAISY
II
were compared to those obtained by the traditional Tilly
and Terry method. The traditional method buffer was
used for both methods. We also compared two sources of
rumen inoculum from sheep and dairy cows. Seventeen
different feeds were tested, grouped into roughage, con-
centrate, and CP supplements. The experiment was
replicated on two different occasions for all feeds and
the two sources of inoculum. The source of inoculum
and the time at which it was collected had no effect on
the in vitro DM digestibility of the feedstuffs in either
of the methods. The DAISY
II
DM digestibility value
compared well with the traditional method values for
the roughage group; however, for some feedstuffs in the
concentrate and CP supplement groups, the DAISY
II
values were significantly higher than the traditional
method values. Regression analysis of the feeds that
resulted in similar values with the two methods re-
vealed that the DAISY
II
method can be used to predict
in vitro digestibility with relatively small variation.
(Key words: in vitro, dry matter digestibility, inocu-
lum, rumen)
Abbreviation key: IVDMD =in vitro dry matter di-
gestibility, TT =Tilly and Terry.
INTRODUCTION
The in vitro dry matter digestibility (IVDMD)
method has been extensively used to evaluate the nutri-
tional value of ruminant feeds. Since its introduction,
the Tilly and Terry (TT; 1963) IVDMD method has been
largely used to analyze feedstuffs and has been the most
Received December 28, 1999.
Accepted April 20, 2000.
Corresponding author: S. J. Mabjeesh; e-mail: mabjeesh@
agri.hugi.ac.il.
2000 J Dairy Sci 83:2289–2294 2289
accurate and practical laboratory method available for
predicting digestibility data for ruminants (Goldman
et al., 1987; Stern et al., 1997). The method has been
modified and adapted for starch feedstuff analysis (Au-
frere and Michalet-Doreau, 1988), and various workers
have improved its accuracy of prediction. Different dilu-
tion buffers for the rumen liquor have been developed
to adjust the pH of the inoculum (Grant and Mertens,
1992). Many laboratories have proven the method to
be simple, highly repeatable, and reproducible (Tilly
and Terry, 1963; Goldman et al., 1987; Aufrere and
Michalet-Doreau, 1988; Stern et al., 1997). However,
despite its accuracy at predicting DMD, the method’s
two-stage procedure is time- and labor-consuming, and
each feedstuff has to be incubated separately.
The search for better labor efficiency has led to the
development of the DAISY
II
apparatus (ANKOM Tech-
nology Corp., Fairport, NY), which allows simultaneous
incubation of different feedstuffs in sealed polyester
bags in the same incubation vessel. With this method,
the material that disappears from the bag during the
incubation is considered digestible. Recently, Holden
(1999) compared the TT and DAISY
II
methods for pre-
dicting DMD with the buffer recommended by ANKOM
for both systems. Results from that experiment showed
good agreement between the two systems, proving that
DAISY
II
could be used to predict the IVDMD of forages
and grains. The present study was conducted to deter-
mine the reliability of the DAISY
II
system for predicting
the IVDMD of different feedstuffs commonly used in
dairy rations, with different inoculum sources (dairy
cows and sheep) and the buffer used in the TT method.
The IVDMD values obtained with the TT method for
the same feedstuffs were used as a reference for com-
parison.
MATERIALS AND METHODS
Experimental Protocol
Three groups of feedstuffs used regularly in dairy
rations on Israeli dairy farms were selected for this

MABJEESH ET AL.2290
Table 1. Chemical composition of the feedstuffs on a DM basis.
%ofDM
Feedstuff OM CP NDF
Forages
Grass hay 89.5 11.7 58.1
Alfalfa cubes 84.9 16.7 42.9
Alfalfa hay 88.4 21.8 50.0
Clover hay 89.9 10.8 51.0
Concentrates
Corn grain 94.0 9.5 11.5
Barley grain 96.2 11.3 19.0
Wheat grain 96.2 13.5 14.0
Sorghum grain 97.4 8.8 11.0
Concentrate mix
1
93.5 18.2 27.6
Wheat bran 94.1 15.5 47.5
CP supplements
Soybean meal 93.0 49.3 14.2
Sunflower meal 91.2 34.5 26.3
Corn gluten feed 91.1 29.9 39.0
Cottonseed meal 93.4 38.5 37.2
Rapeseed meal 92.2 39.4 27.9
Fish meal 79.3 68.2 —
Whole cottonseed 96.2 21.4 45.0
1
Mixture 1471 Matmor Ltd., Ashdod, Israel.
study (Table 1). Two methods for measuring IVDMD
were compared: the traditional in vitro method (TT)
and DAISY
II
system. The TT method reagents were
used for both systems. All feeds were ground in a Wiley
mill (Arthur H. Thomas, Philadelphia, PA) to pass a 2-
mm pore-size screen, except whole cottonseed (linted),
which was manually cracked, resulting in a large parti-
cle size (>2 mm).
The donor animals, two Merino rams and two Israeli
Holstein dairy cows, were fitted with a ruminal cannula
(inner diameter 10 cm). The cows were in midlactation
and consumed a TMR formulated to meet their mainte-
nance and production requirements (Table 2; NRC,
1989). The TMR contained a 35:65 ratio of forage:con-
centrate with 17% CP, 31% NDF, and 1.72 Mcal NE
L
(on a DM basis). The rams were maintained on diets
sufficient for their maintenance allowance, which con-
tained a 50:50 chopped alfalfa hay and grain mixture
(Table 2, 16% CP, 29.1% NDF, and 1.6 Mcal NE
L
on a
DM basis, mixture 1471, Matmor Ltd., Ashdod, Israel).
The composition of the grain mixture offered to sheep
contained ingredients similar to the TMR. The TMR
was offered once daily at 0900 h, while the rams’meal
was offered twice daily at 0900 and 1700 h. Rumen
digesta was collected before the morning meal from the
reticulum near the reticuloomasal orifice by vacuum
pump. The digesta from each animal was filtered
through eight layers of gauze cloth, mixed on a volume
basis for each species, purged with CO
2
, and kept in a
prewarmed thermos until use (within approximately
20 min).
Journal of Dairy Science Vol. 83, No. 10, 2000
For both methods, the incubation inoculum was pre-
pared by diluting the digesta inoculum with the buffer
(Tilly and Terry, 1963) in a 1:4 (vol/vol) ratio and stir-
ring in a water bath at 39°C with purging CO
2
until
its use (10 to 15 min later). In the TT trial, 250 mg of
each sample was placed into 50-ml sterile plastic tubes
(Miniplast Ein-Shemer, 37845, Israel), and 20 ml of the
incubation inoculum was added. The tube was stop-
pered with a Bunsen valve and incubated for 48 h at
39°C. Tubes were gently swirled by hand four times
every 12 h. For the DAISY
II
incubation, 500 mg of each
sample was inserted into each bag (F57; 5 ×5.5 cm
2
,
ANKOM Technology Corporation), which was then
heat-sealed and incubated in a digestion vessel (25 bags
per vessel) for 48 h at 39°C. For both methods, each
sample was incubated in six replicates for each source
of inoculum at two different occasions (runs; at least 2
wk apart).
At the end of the 48-h incubation period, tube and
vessel contents were acidified by adding 6 MHCl to
reach a final pH of 1.3 to 1.5. After a few seconds, when
the foam subsided, pepsin powder (EC 3.4.23.1) was
added to a final concentration of 0.2% (wt/vol). The
tubes and the vessels were reincubated for an addi-
tional 48 h. At the end of this stage, the bags were
rinsed thoroughly with tap water until the rinse was
Table 2. Composition of diet and concentrate pellets given to cows
and sheep.
%ofDM
1
Grain
Ingredient TMR mixture
Corn grain 21.7 17.0
Barley grain 10.9 30.0
Oat grain 10.0
Soybean meal 4.4 2.0
Rapeseed meal 4.4 5.5
Sunflower meal 3.2 5.0
Corn gluten meal 1.3 3.0
Whole cottonseed 6.8 10.0
Wheat silage 8.0
Corn silage 19.8
Wheat bran 6.9 10.0
Citrus pulp 3.6
Pea hay 3.1
Oats hay 3.1
NPN (mixture)
2
0.5
Soybean oil 0.1 4.5
Salt/calcium 1.6
Fish meal 0.4
Vitamins and minerals
3
0.18 3.0
1
TMR =Diet given to cannulated cows and grain mixture 1471,
Matmore Ltd., Ashdod, Israel, given to rams.
2
Contained 80% urea and 20% ammonium sulfate.
3
Contained 20,000,000 IU of vitamin A/kg, 2,000,000 IU of vitamin
D/kg, 15,000 mg of vitamin E/kg, 6000 mg/kg of Mn, 6000 ppm of
Zn, 2000 mg/kg of Fe, 1500 mg/kg of Cu, 120 mg/kg of I, 50 mg/kg
of Se, and 20 mg/kg of Co.

IN VITRO DIGESTIBILITY METHODS 2291
clear. The tubes were centrifuged at 2500 ×gfor 15
min, and the supernatant was discarded. To the pellet,
50 ml of H
2
O was added and the tubes were recentri-
fuged to wash out the residual acid. The tubes con-
taining the pellets and the bags were dried in a forced-
air oven at 60°C for 48 h to determine the residual DM
weights. For both methods, IVDMD was calculated as
the DM which disappeared from the initial weight in-
serted into the bag or tube.
Statistical Analyses
Results were analyzed using the general linear model
procedure of SAS (1985). The linear model used in-
cluded the main effects of run, inoculum source (sheep
vs. cow), the IVDMD method (TT vs. DAISY
II
), their
interaction, and the residual error term. There were no
interaction effects, so the model was reduced to include
the main effects and the error term. Means were consid-
ered significantly different at P<0.05.
RESULTS
Chemical Analysis of Feeds
The chemical composition of the feeds tested (Table
1) compared well with previous reports from our labora-
tory (Mabjeesh et al., 1997; Shabi et al., 1998; Arieli et
al., 1999) and others (NRC, 1989; Holden, 1999). The
CP content of the CP supplements ranged from 30% in
the corn gluten feed to 68% in the fish meal. The CP
content of the concentrate grains ranged between 9 and
14%, the highest being wheat grain. The CP content of
roughages ranged from 22 to 11%, the highest being
alfalfa hay. Whole cottonseed was not included in any
of the feedstuff groups because of its relative high con-
tent of CP and fiber (>18%). The NDF content of the
feeds ranged from 11 to 58%, the highest being in the
roughages, intermediate in the CP supplements, and
lowest in the grains. The OM content ranged from 79
to 97%, the lowest being for fish meal.
IVDMD
The IVDMD values are presented in Table 3. The
IVDMD of roughage feeds ranged from 47 to 61%, being
highest for grass hay and lowest for clover hay. The
IVDMD value for the grains ranged from 63 to 92%,
being highest for corn. The CP supplements also exhib-
ited a wide rage of IVDMD values, 55 to 91%, being
lowest for cottonseed meal and highest for fish meal.
The IVDMD of whole cottonseed was low relative to the
other feedstuffs, averaging 38%. The IVDMD values
of all feedstuffs were similar for the two sources of
inoculum: sheep and cow (data not shown).
Journal of Dairy Science Vol. 83, No. 10, 2000
Table 3. Means of in vitro DM digestibility (percentage) values using
two methods: traditional method (TT) and Daisy
II 1
for different feed-
stuffs.
Main effect
3
TT Daisy
II
Feedstuff
2
Method
XSDXSDP<
Forages
Grass hay 61.3 10.5 58.5 5.1 NS
4
Alfalfa cubes 48.3 3.4 58.4 3.6 0.03
Alfalfa hay 58.0 5.5 53.6 4.6 NS
Clover hay 54.8 6.7 46.7 5.0 NS
Concentrates
Corn grain 92.2 14.9 85.9 11.3 NS
Barley grain 68.1 5.9 79.1 4.1 0.02
Wheat grain 79.8 10.8 89.1 7.6 0.01
Sorghum grain 80.0 13.9 77.2 6.8 NS
Concentrate mix
5
71.6 1.5 74.4 1.3 0.05
Wheat bran 67.9 12.5 63.4 2.5 NS
CP supplements
Soybean meal 77.5 4.7 76.3 2.0 NS
Sunflower meal 54.0 4.8 62.8 3.7 0.00
Corn gluten feed 66.6 6.4 70.0 3.4 NS
Cottonseed meal 56.6 13.4 54.8 3.5 NS
Rapeseed meal 66.5 2.9 75.0 4.2 0.001
Fish meal 81.3 2.1 91.0 0.5 0.001
Whole cottonseed 36.5 5.7 38.9 3.7 NS
1
ANKOM Technology Corporation, Fairport, NY.
2
n=12.
3
Method: TT versus Daisy
II
. There was no significant effect of ru-
men inoculum source (sheep vs. cow) or run.
4
NS =Not significant (P<0.05).
5
Mixture 1471, Matmore Ltd., Ashdod, Israel.
The IVDMD values differed between the TT and
DAISY
II
methods for some feeds from each group. The
IVDMD of alfalfa cubes was 21% higher in DAISY
II
than in the TT method. The grains exhibited similarity
between the two methods for corn, sorghum, and wheat
bran (averaging 89, 79, and 66%, respectively). Barley
grain, wheat grain, and the concentrate mix had higher
IVDMD values with the DAISY
II
method than with
TT. The same trend was apparent in some of the CP
supplement feeds. Sunflower meal showed the highest
difference at +17% from DAISY
II
relative to TT, versus
+13% and +12% for rapeseed and fish meal. Whole cot-
tonseed had the lowest IVDMD (38%) relative to all
feed categories.
DISCUSSION
The in vitro method of evaluating the digestibility of
ruminant feeds is used worldwide. The method is easier
than in vivo analysis and avoids the need to surgically
prepared animals in different locations in the gastroin-
testinal tract. The TT method has also been proven
more accurate than digestibility predictions based on
the chemical compositions of feeds (Van Soest, 1994).
The TT method was first introduced to predict forage

MABJEESH ET AL.2292
digestibility and later modified for starchy feedstuffs.
Because of its similarity to in vivo values (Tilly and
Terry, 1963; Aufrere and Michalet-Doreau, 1988), it
is considered a reference method for the prediction of
ruminant feed digestibility. In this study, we compared
the DAISY
II
method to the TT method with two sources
of rumen inoculum, focusing on the reliability of the
DAISY
II
method for predicting digestibility of differ-
ent feedstuffs.
The source of the inoculum did not affect the IVDMD
of any of the feedstuffs used in this study. Cone et al.
(1989) found that the type of diet fed to the donor animal
affects the values of in vitro degradability. The starch
degradability from different feedstuffs was higher when
the donor cow was fed a ration containing 1:1
hay:concentrate (on a DM basis) than when the diet
was based solely on hay. However, the composition of
the concentrate mixture had only a minor effect on
degradability values (Richards et al., 1995; Stern et al.,
1997). In the current study, both sheep and cows appear
to have consumed diets that allowed sufficient growth
of similar mixed microflora in both species. The level
of feeding may also affect activity in the rumen liquor.
Cone et al. (1989) showed that the activity of rumen
microflora, measured at different times after feeding,
was higher in liquor taken from cows fed a higher level
of DM when feeds were incubated for 6 h. Since this
effect is time-dependent, it might be irrelevant to the
extent of the TT and DAISY
II
measurements of feed
digestibility. Similarly, others (Goldman et al., 1987)
had shown a negligible effect on the IVDMD values of
feeds, whether rumen liquor was taken from heifers,
dry cows, or lactating cows (>30 L of milk/d).
In general, the IVDMD values obtained in this study
were within the range reported by the NRC (1989) for
the TDN of these feeds. Moreover, the IVDMD values
compared well with those measured by the TT method
(Aufrere and Michalet-Doreau, 1988). However, the
barley, soybean meal, and gluten feed IVDMD values
observed in this study by the TT method were somewhat
lower than the those reported by Aufrere and Michalet-
Doreau (1988) of 83, 90, and 81%, respectively. This
discrepancy could be due to the method of commercial
feed processing and the different chemical compositions
of the feeds. The DAISY
II
system IVDMD values for
alfalfa hay and corn grain compared well with the re-
sults reported by Holden (1999) using the same method.
It should be noted that the IVDMD value of whole cot-
tonseed in this study was much lower (40%) than the
one reported in NRC tables (1989). This lower value
can be explained by the way cottonseed was prepared
in our study for in vitro incubation: the whole cotton-
seed was manually cracked with pliers to prevent the
extraction of oil, which could mix with the lint; if the
Journal of Dairy Science Vol. 83, No. 10, 2000
cottonseed had been ground in a Wiley mill, such mixing
would have made obtaining a homogeneous sample im-
possible.
In general, relative to TT method, the DAISY
II
system
gave similar IVDMD values for roughages and higher
values for concentrates and some CP supplements. The
IVDMD determined by the DAISY
II
method might be
affected by several factors related to the bags used, feed
characteristics, and the instrument itself. One obvious
advantage of the DAISY
II
instrument over the TT
method is the continuous rotation of the fermentation
vessels, which thoroughly mixes the digestion inoculum
during the incubation period and eliminates the need
for a time-consuming centrifugation step at the end of
incubation. On the other hand, the porosity of the bags
and the sample weight per bag surface area may affect
the IVDMD values. These factors have been extensively
reviewed for the in situ method (Nocek, 1998; Vanzant
et al., 1998). However, the repeatability of measure-
ments in DAISY
II
seems to be higher, as indicated by
overall lower SD across all feeds measured (Table 3).
The pore size of the bags (50 ±15 µm) was within the
range summarized by Vanzant et al. (1998) for many
studies reported in the literature. The ratio of sample
size to bag surface area was 8.8 mg/cm
2
, similar to that
suggested (10 mg/cm
2
; Vanzant et al., 1998) to increase
the accuracy of degradability predictions relative to in
vivo ruminal disappearance. However, in this study
IVDMD values measured by the DAISY
II
method were
consistently higher than those obtained with TT. This
could be related to the physical properties of the feeds
used in this study. Higher IVDMD values were ob-
served in the concentrate and CP supplement groups.
These observations may be explained partially by the
fact that grains become floury when ground and by the
protein solubility versus degradability in CP supple-
ments. Sample processing, particularly grind size, in-
teracts with pore size of the bag and affects the extent
of feed disappearance. Most of the differences in re-
sponse to grind size are explained by differences in
particle size and distribution (Michalet-Doreau and
Cerneau, 1991; Michalet-Doreau and Ould-Bah, 1992;
Norzie
`re, 1996). Moreover, Malestein et al. (1982) sug-
gested that simultaneous incubation of different
starchy feedstuffs might enhance the degradation of
separate feedstuffs. This is unlikely to be the case in the
current experiment. However, in the DAISY
II
system,
incubation of different feeds in the same vessel has
been shown not to affect the IVDMD values for the
different feeds (Holden, 1999).
The effects of size distribution of feed particles may
not be the sole reason for the higher IVDMD values
with the DAISY
II
system. It is worth noting that these
effects are more critical in the in situ method because

IN VITRO DIGESTIBILITY METHODS 2293
the disappearance of nutrients in a particular feed is
measured on a time function, which in turn affects de-
gradability rate. However, the time criteria for both TT
and DAISY
II
are irrelevant because the 48-h incubation
time for the two stages is considered to be sufficient
for complete digestion of feed matter (Tilly and Terry,
1963). Hence, the higher values observed in the DAISY
II
method might be related to the nonrecovered material
that escaped from the bag and was not digested by the
two stages of the method. This phenomenon was very
well demonstrated by Dewhurst et al. (1995), who com-
pared the extent and rate of OM digestibility measured
in situ with those determined in vitro. They observed
that feeds like soybean hulls and corn germ have higher
OM digestibility in situ than in vitro because of the
passage of small particles of unfermentable material
through the bag pores in the former method. For exam-
ple, the fish meal in this study contains ∼20% ash and
other components (e.g., bones and shells), part of which
might not be water-soluble but might escape from the
bags during incubation. Furthermore, the method of
meal processing may also create insoluble compounds
(e.g., Maillard reactions) whose particle sizes may be
suitable for escape from the bags during incubation but
which are not available to the microbial enzymes or
pepsin hydrolysis. This might also be the case with the
alfalfa cubes, which had a lower IVDMD value in the
TT than alfalfa hay, while in the DAISY
II
method the
IVDMD was 21% higher than the TT values for alfalfa
cubes only. During the preparation of alfalfa cubes,
freshly cut plants are heated to dryness, which appears
to produce indigestible material. This is supported by
the higher ADIN content measured in the alfalfa cubes
versus the alfalfa hay (2.24 and 1.60 mg N/g of DM,
respectively). This difference between the two methods
was not observed in this study (1999) for any of the
feedstuffs used. This could be partially related to the
large apparent variation between and within the
means, especially for the grains group.
When IVDMD obtained from TT was plotted against
the DAISY
II
data, the following linear equation was ob-
tained:
IVDMD
TT
=0.84(±0.10)IVDMD
DAISYII
+8.88
(±7.24), P<0.00001, n =17, SD =6.24, R
2
=0.81.
This SD of the regression and SE of the intercept and
slope point to a large scatter around the regression line.
However, when the data from the feeds that had higher
values in the DAISY
II
method were excluded from the
regression analysis, the R
2
improved to 0.95 and the
following regression was calculated:
Journal of Dairy Science Vol. 83, No. 10, 2000
IVDMD
TT
=1.02(±0.08)IVDMD
DAISYII
+1.11
(±5.43), P<0.00001, n =10, SD =3.76.
This regression shows that the DAISY
II
method has
good potential to predict digestibility of certain feed-
stuffs. Because the difference between the methods is
mainly related to the particle size of the indigestible
material in the feeds, we suggest that these groups of
feed be ground to a larger particle size, which may
prevent the escape of indigestible particles from these
components from the bags.
CONCLUSION
The DAISY
II
system is an easier, less time-consuming
method of measuring IVDMD of ruminant feed than
the method of TT. However, its use in the current proce-
dure is limited to feeds that do not contain a high level
of small indigestible derivatives that might escape from
the bags during the incubation. Rumen liquor taken
from sheep maintained on a diet with a 1:1 forage:con-
centrate ratio was as efficient for IVDMD measure-
ments as inoculum taken from the rumen of dairy cows.
REFERENCES
1 Arieli, A., K. Shahar, S. J. Mabjeesh, S. Zamwell, and D. Sklan.
1999. Determination of the digestible energy of ruminant feed-
stuffs by the combined bag technique. J. Dairy Sci. 82:566–573.
2 Aufrere, J., and B. Michalet-Doreau. 1988. Comparison of meth-
ods for predicting digestibility of feeds. Anim. Feed Sci. Technol.
20:203–218.
3 Cone, W. J., W. Cline
´-Theil, A. Malestein, and A. T. Van’t Klooster.
1989. Degradation of starch by inoculum with rumen fluid. A
comparison of different starch sources. J. Sci. Agric. 49:173–183.
4 Dewhurst, R. J., D. Hepper, and A.J.F. Webster. 1995. Compari-
son of in sacco and in vitro techniques for estimating the rate
and extent of rumen fermentation of a range of dietary ingredi-
ents. Anim. Feed Sci. Technol. 51:211–229.
5 Goldman, A., A. Genizi, A. Yuzari, and N. G. Seligman. 1987.
Improving the reliability of the two-stage in vitro assay for rumi-
nant feed digestibility by calibration against in vivo data from a
wide range of sources. Anim. Feed Sci. Technol. 18:233–245.
6 Grant, R. J., and D. R. Mertens. 1992. Development of buffer
systems for pH control and evaluation of pH effects upon fiber
digestion in vitro. J. Dairy Sci. 75:1581–1587.
7 Holden, L. A. 1999. Comparison of methods of in vitro dry matter
digestibility for ten feeds. J. Dairy Sci. 82:1791–1794.
8 Mabjeesh S. J., A. Arieli, I. Bruckental, S. Zamwell and H. Tagari.
1997. Effect of ruminal degradability of crude protein and non-
structural carbohydrates on the efficiency of bacterial crude pro-
tein synthesis and amino acid flow to the abomasum of dairy cow.
J. Dairy Sci. 80:2939–2949.
9 Malestein, A., A. T. Van’t Klooster, G.H.M. Counotte, and R. A.
Prins. 1982. Concentrate feeding and ruminal fermentation. 2.
Influence of concentrate ingredients on pH and on L-lactate con-
centration in incubations in vitro with rumen fluid. Neth. J. Agric.
Sci. 30:259–273.
10 Michalet-Doreau, B., and P. Cerneau. 1991. Influence of foodstuff
particle size on in situ degradation of nitrogen in the rumen.
Anim. Feed Sci. Technol. 35:69–81.
11 Michalet-Doreau, B., and M. Y. Ould-Bah. 1992. In vitro and in
sacco methods for the estimation of dietary nitrogen degradability
in the rumen: a review. Anim. Feed Sci. Technol. 40:57–86.

MABJEESH ET AL.2294
12 National Research Council. 1989. Nutrient Requirements of Dairy
Cattle. 6th rev. ed. Natl. Acad. Sci., Washington, D.C.
13 Nocek, J. E. 1988. In situ and other methods to estimate ruminal
protein and energy digestibility: a review. J. Dairy Sci.
71:2051–2069.
14 Norzie
`re, P., and B. Michalet-Doreau. 1996. Validation of in sacco
method: influence of sampling site, nylon bag or rumen contents,
on fibrolytic activity of solid-associated microorganisms. Anim.
Feed Sci. Technol. 57:203–210.
15 Richards, C. J., J. F. Pederson, R. A. Britton, R. A. Stock, and
C. R. Krehbiel. 1995. In vitro starch disappearance procedure
modifications. Anim. Feed Sci. Technol. 55:35–45.
16 Shabi, Z., A. Arieli, I. Bruckental, Y. Aharoni, S. Zamwel,
A. Bor, and H. Tagari. 1998. Effect of the synchronization of
Journal of Dairy Science Vol. 83, No. 10, 2000
the degradation of dietary crude protein and organic matter and
feeding frequency on ruminal fermentation and flow of digesta
in the abomasum of dairy cows. J. Dairy Sci. 81:1991–2000.
17 Statistical Analysis Systems Institute. 1985. SAS User’s Guide:
Statistics. Statistical Analysis Systems Institute Inc., Cary, NC.
18 Stern, M. D., A. Bach, and S. Calsamiglia. 1997. Alternative tech-
niques for measuring nutrient digestion in ruminants. J. Anim.
Sci. 75:2256–2276.
19 Tilly, J. M., and R. A. Terry. 1963. A two-stage technique for the
in vitro digestion of forage crops. J. Br. Grassl. Soc.18:104–111.
20 Van Soest, P. J. 1994. Nutritional Ecology of the Ruminant. 2nd
ed. Cornell University Press, Ithaca, NY.
21 Vanzant, E. S., R. C. Cochran, and E. C. Titgemeyer. 1998. Stan-
dardization of in situ techniques for ruminant feedstuff evalua-
tion. J. Anim. Sci. 76:2717–2729.