scale specialized facilities and resources, there has been
growing interest in using in vitro techniques to simulate
Developing countries in general and African nations in
the in vivo process (Blümmel et al. 2005; Bhatta et al.
particular are increasingly becoming victims of climate
2008; Soliva et al. 2008; Melesse et al. 2013). Use of in
change as global temperatures rise. The Intergovern-
vitro gas-production techniques allows the screening of
mental Panel on Climate Change (IPCC) has attributed
significant numbers of species rapidly and at relatively
the temperature increases to human activities, including
low cost (Soliva et al. 2008; Singh et al. 2012).
releases of the greenhouse gases, carbon dioxide, methane
We used in vitro techniques to assess: a) the chemical
and nitrous oxide into the atmosphere. They have
and mineral compositions; and b) ruminal fermentation
requested nations to quantify the amounts of gases they
characteristics and CH4 emission potentials, of some
produce and to develop research to limit further gaseous
common Ethiopian green forages (9 grasses and 3
emissions (Moss et al. 2000).
legumes) for their subsequent use in formulating diets for
Ruminants are a major source of methane (CH4)
ruminants with lower potential emissions of CH4.
emissions, and France et al. (1993) estimated that the
world’s cattle emit about 100 Mt of CH4 into the
Materials and Methods
atmosphere annually, constituting 12.5‒20% of the total
global CH4 emissions. More than half of the global cattle
Feed sample collection
population are located in the tropics (McCrabb and
Hunter 1999), a large proportion of which are supported
Samples of grasses and legumes were collected during
on relatively low-quality, highly fibrous feed resources.
the small rainy season (March‒May) in 2013. Samples of
This constitutes a significant source of global CH4
Avena sativa and Vicia sativa were collected from the first
emissions. Moreover, enteric CH4 emissions in ruminants
stage of growth on the forage farms of College of
represent a loss of 2‒12% of gross energy of feeds
Agriculture, Hawassa University, Hawassa (7°03'43.38"
(McCrabb and Hunter 1999). As a result, CH4 emissions
N, 38°28'34.86" E; 1,700 masl). Samples of Pennisetum
from livestock have become a focus of research activities,
purpureum, Chloris gayana, Panicum maximum,
especially in countries where agriculture is an important
Panicum coloratum, Hyparrhenia cymbaria, Desmodium
economic sector.
intortum and Medicago sativa were collected at the pre-
A wide diversity of forage sources are used in feeding
flowering stage of plants from ILRI’s (International
livestock in the tropics. Improving the feed resource base
Livestock Research Institute) Forage Seed Multiplication
by identifying alternative and more nutritious feeds with
Center located at Debre-Zeit (8°45'8.10" N, 38°58'42.46"
low CH4 production would both reduce greenhouse gas
E; 2,006 masl). Samples of Brachiaria mutica, Cenchrus
emissions and increase the efficiency of energy utilization
ciliaris and Cynodon dactylon were collected at the pre-
in forage. There is a lack of data describing and
flowering stage of plants from ILRI’s Forage Seed
identifying those tropical grass and legume forages with
Multiplication Center located at Zeway (7°55'59.99" N,
low CH4 production potential when fed to ruminant
38°43'0.01" E; 1,640 masl). All samples were dried on
animals.
plastic sheets kept in shade, ground to pass a 1 mm sieve
Since in vivo studies of methanogenesis by ruminants
and transported in air-tight plastic containers to the
are time-consuming and expensive, and require large-
University of Hohenheim, Germany, for analyses.
Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775)
Nutrient composition and methane production of forages in Ethiopia 165
Chemical analyses
sample was weighed and transferred into 100 ml
calibrated glass syringes, fitted with white Vaseline-
Chemical analyses of proximate nutrients, fiber fractions
lubricated glass plungers.
and minerals were performed as outlined by Verband
A buffer solution was prepared and maintained in a
Deutscher Landwirtschaftlicher Untersuchungs- und
water bath at 39 °C under continuous flushing with CO2.
Forschungsanstalten (VDLUFA 2007). The samples were
Rumen fluid was collected before the morning feeding
analyzed at the Institute of Animal Science, University of
from 2 rumen-cannulated, lactating Jersey cows, fed a
Hohenheim, for dry matter (DM, method 3.1), ash (method
total mixed ration consisting (DM basis) of 20% maize
8.1), crude protein (CP, method 4.1.1; N x 6.25), petroleum
silage, 20% grass silage, 20% hay and 40% dairy
ether extract (EE, method 5.1.1) and crude fiber (CF,
concentrate. The rumen fluid from both cows was mixed,
method 6.1.1). Neutral detergent fiber (aNDFom) was
filtered and added to the buffer solution (1:2 v/v) under
assayed on an organic matter basis after amylase treatment
constant stirring. Thirty mL of buffered rumen fluid was
(method 6.5.1) and acid detergent fiber on an organic
injected into each syringe, which was then immediately
matter basis (ADFom, method 6.5.2). Acid detergent lignin
placed into a rotating disc and oven-incubated at constant
(ADL) was analyzed according to method 6.5.3. Cellulose
temperature of 39 °C. Three syringes with only buffered
and hemicellulose were computed as ADFom minus ADL
rumen fluid, termed as blanks, plus 3 syringes with hay
and aNDFom minus ADFom, respectively. Non-fiber
standard and 3 with concentrate standard with known GP
carbohydrate (NFC) concentration was calculated as 100 ‒
were included in each run. The GP of samples, blanks and
(aNDFom + CP + crude fat + ash) according to NRC
standards was recorded at 2, 4, 6, 8, 12, 24, 32, 48, 72 and
(2001). Nitrogen free extract (NFE) was computed as OM
96 hours of incubation. The plunger of the syringe was re-
– (CF + EE + CP). Minerals [Ca, P, magnesium (Mg),
set to 30 ml after the 6 and 24 hour readings. For
potassium (K), sodium (Na), iron (Fe), copper (Cu), man-
metabolizable energy (ME) estimation, the GP of the feed
ganese (Mn) and zinc (Zn)] were determined according to
samples was recalculated as 24 h GP on 200 mg DM using
methods 10 and 11 of VDLUFA (2007) using an Induc-
results from the blanks, with the corrections determined
tively Coupled Plasma spectrometer (ICP-OES).
by the standards of hay and concentrate, the sample
Four species with possible anti-nutritional factors
weight and its DM concentration.
( Chloris gayana, Desmodium intortum, Medicago sativa
The estimations of organic matter digestibility (OMD)
and Vicia sativa) were selected from the collection and
and ME were carried out according to Menke et al. (1979)
analyzed for concentrations of total phenols and non-tannin
and Menke and Steingass (1988) by using the following
phenols using the Folin-Ciocalteu method [Jayanegara et
equations:
al. (2011) with modifications as described by Wischer et al.
ME (MJ/kg DM) = 1.68 + 0.1418*GP + 0.0073*CP +
(2013)]. Extractable condensed tannins were analyzed
0.0217*XL – 0.0028 XA
according to Jayanegara et al. (2011). Concentrations of
OMD (%) = 14.88 + 0.889*GP + 0.0448*CP +
tannin phenols were then calculated as differences between
0.0651*XA
total phenol and non-tannin phenol concentrations. The
where: GP, CP, XL and XA are 24 h gas production
absorbance of total phenols and non-tannin phenols was
(ml/200 mg DM), crude protein, crude fat and ash (g/kg
recorded at 725 nm using a UV-VIS spectrophotometer
DM) of the incubated feed samples, respectively.
(Perkin Elmer Instruments, Norwalk, CT, USA). Con-
The corrected GP recorded between 2 and 96 h of
densed tannins were analyzed by the butanol-HCl-iron
incubation and the kinetics of GP were described by using
method according to Jayanegara et al. (2011). The
the exponential equation: y = b*(1−e(−c(t−lag))), which
absorbance was read at 550 nm using the same UV-VIS
assumed one pool of asymptotic GP (b, ml/200 mg DM)
spectrophotometer as for total phenols and non-tannin
with a constant fractional rate of GP (c, per hour) with a
phenols and was expressed as leucocyanidin equivalents.
lag phase (lag, hours) in the onset of GP; parameter “y” is
All analyses were run in duplicate and were averaged.
GP at time “t” (Blümmel et al. 2003; 2005).
If deviation between duplicates was above the level
specified for each analysis, the analyses were repeated.
Methane production
In vitro gas production
For CH4 determination, 6 separate in vitro runs were
performed. Based on the previous in vitro GP results for
Gas production (GP) was determined according to the
each feed sample, we calculated the quantity of each feed
VDLUFA official method (VDLUFA 2007, method 25.1)
sample to be incubated for 24 h without having to remove
(Menke and Steingass 1988). About 200 mg of feed
the gas produced in the syringes during the incubation
Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775)
166 A. Melesse, H. Steingass, M. Schollenberger and M. Rodehutscord
period. After 24 h of incubation, total GP was recorded,
material and eij is the residual error. All multiple
and the incubation liquid was carefully decanted, while
comparisons among means were performed with
leaving the gas inside the syringes. The CH4 content of
Duncan’s multiple range tests.
the total gas in the syringes was then analyzed using an
infrared methane analyzer (Pronova Analysentechnik,
Results
Berlin, Germany) calibrated with a reference gas
(13.0% CH4 by volume, Westfalen AG, Münster,
Crude nutrients and anti-nutritional factors
Germany). Syringes were directly connected to the
analyzer and about 20 ml of gas was injected for about 20
Crude nutrient concentrations in the studied grass and
seconds until the displayed CH4 concentration was
legume plants are presented in Table 1. The CP
constant. The CH4 produced by each sample was
concentrations in grass species ranged from 70 g/kg DM
corrected by the amount of CH4 produced by blank
( Panicum coloratum) to 220 g/kg DM ( Avena sativa),
syringes (containing only the rumen fluid) and by the
while those in legumes ranged from 257 g/kg DM
factors of reference hay and concentrate feed, which were
( Medicago sativa) to 346 g/kg DM ( Vicia sativa). For
included in each run.
grasses, CF concentrations ranged from 281 g/kg DM
( A. sativa) to 322 g/kg DM ( Cenchrus ciliaris and Chloris
Statistical analyses
gayana), while values for legumes ranged from 213 g/kg
DM ( Desmodium intortum) to 249 g/kg DM ( M. sativa).
Results of chemical and mineral composition are
expressed as means of duplicate analyses of a bulked
Similarly, grasses contained more aNDFom (484‒651
g/kg DM) than legumes (364‒404
sample. Model fitting for gas production kinetics and
g/kg DM). Concen-
parameter estimation was done according to Blümmel et
trations of ADL in grasses (25.3‒41.3 g/kg DM) were
al. (2003) by using the computer program GraphPad
lower than those in legumes (48.2‒89.8 g/kg DM).
Prism 5.0 (2007) for Windows (GraphPad Software Inc.,
As shown in Table 2, concentrations of total phenols
La Jolla, CA, USA). Data on 24 h gas and methane
were comparable for M. sativa, V. sativa and C. gayana,
productions were subjected to the GLM of the Statistical
while those for D. intortum were higher by a factor of 10.
Analysis System (SAS 2010). Analysis of variance was
No tannin phenols or extractable condensed tannins were
conducted according to the following model: yij = μ + Pi
detected in either M. sativa or V. sativa, while C. gayana
+ Rj + eij, where: yij is the independent variable, μ is the
contained very low concentrations of these compounds.
overall mean, Pi is the effect of the i th plant material, Rj
Both tannin phenols and extractable condensed tannins
is the effect of the j th experimental run of the i th plant
were at high concentrations in D. intortum.
Table 1. Crude nutrient concentrations (g/kg DM) in some common grass and legume forages grown in Ethiopia.
Forage species
Ash
CP1
EE
CF
NFE aNDFom ADFom ADL Cellulose Hemi-
NFC
cellulose
Grasses
Avena sativa
133
226
31.2
281
263
484
326
25.3
301
159
126
Brachiaria mutica
167
159
12.2
229
382
504
277
29.8
247
227
158
Cenchrus ciliaris
154
82
14.5
322
380
601
373
26.7
346
228
149
Chloris gayana
131
135
13.3
322
341
651
370
41.1
329
281
70
Cynodon dactylon
125
187
14.7
272
357
609
323
41.3
282
286
64
Hyparrhenia cymbaria
105
156
11.6
299
375
605
333
33.7
299
272
122
Panicum coloratum
102
70
21.0
292
461
633
322
26.3
296
312
174
Panicum maximum
140
140
12.5
284
373
566
344
27.1
317
222
142
Pennisetum purpureum
173
121
10.9
315
325
599
372
33.4
339
227
96
Legumes
Desmodium intortum
91.7
258
9.3
213
365
396
319
89.8
229
77.2
245
Medicago sativa
151
257
13.3
249
282
364
302
48.2
254
61.7
215
Vicia sativa
147
346
17.2
218
214
404
336
56.6
279
67.9
86
1CP = crude protein; EE = crude fat; CF = crude fiber; NFE = nitrogen free extract; aNDFom = neutral detergent fiber on organic matter basis after amylase treatment; ADFom = acid detergent fiber on organic matter basis; ADL = acid detergent lignin; NFC =
non-fiber carbohydrates.
Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775)
Nutrient composition and methane production of forages in Ethiopia 167
Table 2. Concentrations (g/kg DM) of total phenols, tannin phenols and extractable condensed tannins in some selected forage species in Ethiopia.
Species
Type of forage
Total phenols
Tannin phenols
Extractable condensed tannins
Medicago sativa
Legume
6.7
nd1
nd
Desmodium intortum
Legume
77.7
57.6
77.6
Vicia sativa
Legume
7.6
nd
nd
Chloris gayana
Grass
6.8
1.3
0.2
1nd = not detected.
Minerals
in legumes from 8.1 MJ/kg DM in D. intortum to 10.2 MJ/
kg DM in V. sativa. Organic matter digestibility in grasses
As presented in Table 3, among the grasses Brachiaria
ranged from 56.1% in P. coloratum to 79.6% in A. sativa,
mutica had the highest concentrations of Ca, Mg, Fe and
and from 64.6 to 82% in legumes. The highest asymp-
Mn, while P concentration was highest in A. sativa,
totic GP (parameter b) values for grasses were observed
Pennisetum purpureum and Panicum maximum. In
in
Hyparrhenia.
cymbaria
(58.6
ml)
and
leguminous forages, M. sativa and V. sativa had Ca
P. maximum (59.2 ml) with the lowest in B. mutica (49.2
concentrations of about 9 g/kg DM, while V. sativa had
ml). Values for legumes were generally lower with
the highest P concentration (5.6 g/kg DM). Sodium
a range of 39.0‒45.2 ml. The fractional rates of GP per
concentrations varied widely in both legumes and grasses,
with ranges of 0.06‒8.01 g/kg DM for grasses and 0.10‒
hour (parameter c) for grasses ranged from 0.0387
( P. coloratum) to 0.0667 ( A. sativa). The range for
4.26 g/kg DM for legumes.
legumes was 0.0537 ( D. intortum) to 0.0851 ( V. sativa).
In vitro gas production profiles and fermentation kinetics
As shown in Table 4, the values for the goodness of fit
(R2) of the exponential model were above 94% for all
As shown in Table 4, metabolizable energy (ME)
species.
concentrations in grasses ranged from 7.4 MJ/kg DM
Patterns of gas production for the grasses are shown in
in Panicum coloratum to 10.6 MJ/kg DM in A. sativa and
Figure 1 and for legumes in Figure 2.
Table 3. Mineral composition of some common grass and legume forages grown in Ethiopia.
Forage species
Major (g/kg DM)
Trace (mg/kg DM)
Ca1
P
Mg
K
Na
Fe
Cu
Mn
Zn
Grasses
Avena sativa
3.69
5.24
2.18
38.0
8.01
104
2.44
46.9
78.4
Brachiaria mutica
8.55
3.49
4.63
26.0
1.83
716
7.02
84.7
29.9
Cenchrus ciliaris
4.30
2.54
2.95
24.3
0.45
430
4.21
31.7
18.6
Chloris gayana
3.88
2.94
1.98
35.6
0.50
210
4.75
68.7
28.9
Cynodon dactylon
5.11
2.07
2.46
24.9
0.15
181
5.09
57.5
35.0
Hyparrhenia cymbaria
4.27
1.40
2.41
20.6
0.06
150
5.87
44.3
28.1
Panicum coloratum
3.20
2.78
3.66
11.0
1.82
191
3.98
21.7
15.2
Panicum maximum
4.14
4.56
4.38
23.9
3.54
420
8.27
40.6
26.3
Pennisetum purpureum
3.09
4.49
3.49
38.9
0.16
267
7.50
27.5
25.5
Legumes
Desmodium intortum
6.85
2.37
6.25
18.8
0.10
486
7.14
58.7
34.8
Medicago sativa
9.32
3.24
3.02
45.9
4.26
494
2.42
55.2
65.3
Vicia sativa
9.05
5.57
3.12
46.0
2.64
441
3.37
61.3
388
1Ca = calcium; P = phosphorus; Mg = magnesium; K = potassium; Na = sodium; Fe = iron; Cu = copper; Mn = manganese; Zn =
zinc.
Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775)
168 A. Melesse, H. Steingass, M. Schollenberger and M. Rodehutscord
Table 4. In vitro estimates of metabolizable energy (ME), organic matter digestibility (OMD) and kinetics of gas production (ml/200
mg DM) in some common grass and legume forages grown in Ethiopia.
Forage species
ME (MJ/kg DM)
OMD (%)
b1
c
Lag time (h)
R2
Grasses
Avena sativa
10.6
79.6
54.7
0.0667
1.17
97.3
Brachiaria mutica
8.0
66.7
49.2
0.0448
0.20
97.3
Cenchrus ciliaris
8.5
66.9
57.7
0.0469
0.94
98.2
Cynodon dactylon
9.7
75.2
56.9
0.0537
1.47
97.7
Chloris gayana
9.0
70.3
56.4
0.0511
1.53
97.7
Hyparrhenia cymbaria
10.0
75.6
58.6
0.0613
1.39
97.7
Panicum coloratum
7.4
56.1
55.4
0.0387
0.81
96.2
Panicum maximum
9.7
75.2
59.2
0.0548
0.96
97.8
Pennisetum purpureum
8.4
68.8
54.4
0.0475
1.25
97.4
Legumes
Desmodium intortum
8.1
64.6
39.0
0.0537
0.37
94.7
Medicago sativa
9.5
77.2
45.2
0.0839
0.94
97.9
Vicia sativa
10.2
82.0
45.0
0.0851
0.89
96.1
1b = total asymptotic gas production (ml/200 mg DM); c = the rate at which b is produced per hour with a lag phase in the onset of gas production.
Methane production
gas production followed a similar pattern with highest
values for H. cymbaria, P. maximum, A. sativa and
Most grass species produced significant amounts of
Cynodon dactylon and lowest for P. coloratum (P<0.05).
methane during digestion, but P. coloratum produced
Methane:total gas ratios (CH4:GP) ranged from 0.18:1
about half that of other species (P<0.05) (Table 5). Total
( B. mutica) to 0.11:1 ( P. coloratum) (P<0.05).
Figure 1. Patterns of gas production of some tropical grass forages during in vitro incubation for 96 h.
Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775)
Nutrient composition and methane production of forages in Ethiopia 169
Figure 2. Patterns of gas production of some tropical legume forages during in vitro incubation for 96 h.
Desmodium intortum produced much less CH4
ratio was observed in D. intortum (0.12:1), which
than V. sativa and M. sativa (P<0.05) differing
differed significantly from those for the other 2
significantly from these other legumes (Table 5).
legumes (0.16:1). In general, grasses produced
While differences between species for total gas
comparatively higher (P<0.05) GP, CH4 and CH4:GP
production were not so marked, the lowest CH4:GP
ratios than legume forages.
Table 5. In vitro methane and total gas production profiles (±SD) in some common grass and legume forages grown in Ethiopia.
Forage species
CH4 (ml/200 mg DM)
GP (ml/200 mg DM)
CH4:GP (v:v)
Grasses
Avena sativa
6.16±1.00ab1
44.0±5.34abc
0.140d
Brachiaria mutica
5.80±0.53b
33.5±1.41e
0.178a
Cenchrus ciliaris
6.92±0.45a
40.9±1.62cd
0.169b
Chloris gayana
6.44±0.68ab
41.8±3.20bcd
0.159bc
Cynodon dactylon
7.01±0.53a
43.6±4.38abc
0.161bc
Hyparrhenia cymbaria
6.78±0.68a
47.0±2.10a
0.145d
Panicum coloratum
3.18±0.47c
31.4±1.02e
0.107e
Panicum maximum
7.01±1.06a
44.9±2.00ab
0.159bc
Pennisetum purpureum
6.94±1.06a
39.7±2.15d
0.165bc
Legumes
Desmodium intortum
3.67±0.39b
29.9±1.03b
0.123b
Medicago sativa
5.90±0.72a
37.5±2.86a
0.157a
Vicia sativa
5.73±0.51a
36.1±3.25a
0.159a
Grasses vs. legumes
Grasses
6.23±1.42a
40.3±5.52a
0.156a
Legumes
5.34±1.21b
36.9±6.15b
0.144b
1Means within columns and plant types followed by different letters differ significantly (P<0.05).
CH4 = methane production; GP = total gas production at 24 h incubation of feed samples.
Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775)
170 A. Melesse, H. Steingass, M. Schollenberger and M. Rodehutscord
Discussion
where a cut-and-carry system operates, they become quite
relevant. All leguminous forages contained less fiber than
Crude nutrient and mineral concentrations
grasses, which might be explained in part by lower
hemicellulose concentration in the legumes at comparable
The generally lower CP concentrations in the grasses than
levels of cellulose (Table 1). Cellulose and hemicellulose
in the legumes were consistent with the reports of Singh
in forages represent the main sources of energy to
et al. (2012) for Indian green forages. Consistent with the
ruminants (Merkel et al. 1999).
reports of Tessema and Baars (2006), all forages studied
The aNDFom, ADFom and ADL concentrations in
had protein concentrations above 8%, suggested by Van
V. sativa were comparable with those reported by Berhane
Soest (1982) as the critical level, below which intake may
et al. (2006) from the lowlands of northern Ethiopia.
fall due to lack of sufficient nitrogen for effective
While the CP concentration in P. coloratum was similar
proliferation of rumen micro-organisms. Higher CP
to the observation of the same authors, they reported
values in leguminous forages than in grasses might be
higher NDF, ADF and ADL values than those we found.
related to the N-fixing abilities of the legumes. Our
Such variations might be induced by the stage of maturity
current findings are in good agreement with those of
of the forage at harvest as grasses increase stem
Tessema and Baars (2006) from Ethiopia, that pure
proportions with age, resulting in higher NDF, ADF and
legume stands and grass-legume mixtures produced
lignin and lower CP values (Mero and Udén 1997; 1998).
forage with higher CP and lower fiber concentrations than
Except for P. coloratum, lipid concentrations in the
pure stands of grass. The CP concentrations in C. gayana
forages investigated here were much lower than
and P. maximum in the present study are lower than those
observations reported by Singh et al. (2012) and Pamo et
reported by Tessema and Baars (2006). However, those
al. (2007). These variations might be attributed mainly to
authors also reported lower CP for M. sativa than found
stage of maturity of the forage at the time of sampling and
in the present study. These differences in CP
different environmental conditions. While lipids do not
concentrations could be explained due to stage of
constitute a major source of energy from forages, forages
maturity, N profile of the soils where they had been grown
with high lipid concentrations may be a tool to modify
and differences in efficiency of protein accumulation
milk fatty acid profile towards more long-chain and
during growth. Moreover, differences in nutrient
unsaturated fatty acids (Elgersma 2015).
concentrations in the feeds may be due to variations in the
Phosphorus is one of the most important minerals for
stage of growth and plant parts (i.e. twigs, leaves, soft
many metabolic processes in animals and a deficiency of
stem) when sampled.
P in the diet can retard growth and reproductive
All leguminous forages had higher lignin con-
performance of livestock (Paterson et al. 1996). While
centrations than grasses as reported by Singh et al. (2012).
V. sativa and A. sativa were found to be the richest sources
This might be explained by the fact that the leguminous
of P in our study (>0.5% P), all forages had P concen-
forages synthesize lignin for strength and rigidity of
trations above 0.2%. Brachiaria mutica, V. sativa and
plant cell walls. Singh et al. (2012) reported 310 and 58.8
A. sativa proved to be the richest sources of Ca, which is
g/kg DM for ADF and ADL concentrations, respectively,
closely related to P metabolism in the formation of bones.
in M. sativa, which are comparable with the present
The calculated average Ca:P ratio for legume forages in
findings.
the present study was 2.0:1, while for grasses it was 1.7:1,
Except for B. mutica, grass species in the present study
both of which fall within the recommended range for Ca:P
had higher aNDFom, ADFom, cellulose and hemi-
ratio in feedstuffs of 1:2 to 2:1 (NRC 2001), indicating
cellulose concentrations than legumes, which is in
that the studied forages are likely to be a well-balanced
accordance with the findings of Tessema and Baars
source of both minerals.
(2006) and Singh et al. (2012). The threshold level of
NDF in tropical grasses, beyond which DM intake of
In vitro gas and methane production
cattle is affected, is suggested to be 600 g/kg DM
(Meissner et al. 1991) and all legumes and some grasses
The study has shown that methane production from all
( B. mutica, P. maximum, P. purpureum and C. ciliaris)
forages tested was relatively uniform, with the exception
had lower NDF values than this critical level. Since
of P. coloratum and D. intortum, which produced much
animals, when allowed to selectively graze, can select a
less CH4 than the remaining species. The observed low in
better quality diet than feed on offer, these issues may not
vitro GP pattern in D. intortum (Figure 2) might be
be a major problem under a grazing situation. However,
explained by the presence of high concentrations of total
Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775)
Nutrient composition and methane production of forages in Ethiopia 171
phenols (77.7 g/kg DM) and condensed tannins (77.6 g/kg
Fermentation of cell wall carbohydrates produces
DM), which have the ability to complex with protein and
more CH4 than fermentation of soluble sugars, which
are a major cause of the resistance of this legume to
produce more CH4 than fermentation of starch (Johnson
bacterial decomposition.
et al. 1996) and legume forages are digested more quickly
Consistent with the current observations, Mero and
than grasses. This was demonstrated for M. sativa and
Udén (1998) reported in vivo OMD values between 61.5
V. sativa in the current study, which means that intake and
and 64.8% for C. ciliaris hay harvested at 6 weeks of
productivity on leguminous pasture can be higher than on
age. They also reported comparable OMD values for
grasses. In tannin-containing forages, excess plant
P. coloratum harvested at 6 weeks of age. Berhane et al.
proteins that become bound to tannins leave the rumen
(2006) reported values of 65.5 and 68.3 ml for in vitro GP
without being digested. However, some leguminous
(parameter b) of fresh-cut V. sativa and P. coloratum,
forages containing tannins, such as D. intortum, can
respectively, which were higher than those observed in
release these proteins in the abomasum in response to low
the current study. Panicum coloratum in the present study
pH. This allows the protein to be digested and absorbed
had the lowest ME and OMD values. Except for
in the small intestine (Waghorn et al. 1987), resulting in
D. intortum, legumes produced more gas than grasses
high productivity in both sheep (Douglas et al. 1995) and
within 96 h of incubation, which is consistent with the
cattle (Wen et al. 2002).
findings of Singh et al. (2012).
Legumes contain higher CP than grasses at the same
The observed variations in CH4 production among the
stage of maturity and protein fermentation in vitro has
investigated forages may be due to variations in their
been shown to be associated with lower CH4 production
chemical composition. Such variations in in vitro CH4
than fermentation of carbohydrates (Cone and Van Gelder
production have been observed in straws, forages and
1999). In vitro studies conducted by Soliva et al. (2008),
food industry byproducts (Santoso and Hariadi 2009;
Tiemann et al. (2008), Bekele et al. (2009) and Archimède
Singh et al. 2012). In the current study, except for
et al. (2011) have shown that a large portion of the
P. coloratum and B. mutica, all investigated grasses had
variability of CH4 production in legumes can be
higher CH4 values than leguminous forages, which is in
associated with the presence of secondary metabolites
agreement with the findings of Boadi et al. (2004) and
(condensed tannins, saponins) in some legume species,
Navarro-Villa et al. (2011). At 12 h fermentation,
which can inhibit CH4 formation (Beauchemin et al. 2007;
Widiawati and Thalib (2007) found that in grasses CH4
Jouany and Morgavi 2007). In the present study,
production per unit of OM degraded was twice that in
D. intortum had the highest phenols and extractable
legume forages. Moreover, hydrolysis of legumes such as
condensed tannins, which possibly contributed to the
lucerne and red clover generates less CH4/g DM than
reduction of CH4 production in this species. In other
hydrolysis of grasses (Ramirez-Restrepo and Barry
studies, prolonged feeding of tanniniferous forage
2005). The lower CH4 values in legumes vs. grasses might
legumes showed that animals receiving D. intortum had
be attributed to less extensive in vitro rumen fermentation
the lowest total worm burdens, the lowest female:male
of legumes as suggested by Navarro-Villa et al. (2011).
parasite ratios, the lowest numbers of eggs in the uterus
When CH4 emissions are expressed as a proportion of
of each female worm and the lowest per capita fecundity
gross energy intake (Waghorn et al. 2006), values are
(Debela et al. 2012). There is high variability among
lower for animals fed forage legumes (Waghorn et al.
legumes, particularly regarding the presence of secondary
2002) than for those receiving a predominantly grass diet.
metabolites such as tannins, which are more common in
Beauchemin et al. (2008) proposed that the lower CH4
tropical legumes (Waghorn 2008).
emissions of legume-fed animals is a result of a
Tropical legumes show promise as a means of
combination of factors including the presence of
reducing CH4 production, partly because of their lower
condensed tannins, lower fiber concentration, higher DM
fiber concentration and faster rate of passage than grasses,
intake and an increased passage rate from the rumen. In
and in some cases, the presence of condensed tannins as
the current study, no extractable condensed tannins were
observed in D. intortum in this study. Various studies
detected in M. sativa and V. sativa. Beauchemin et al.
have reported that condensed tannins in legume forages
(2008) also reported that, although differences in CH4
are able to suppress ruminal methanogenesis directly
emissions reflect compositional differences between
through their antimethanogenic activity and indirectly
grasses and legumes, stage of maturity at the time of
through their antiprotozoal activity (Goel and Makkar
harvest can be a confounding factor.
2012). Patra and Saxena (2010), Pellikaan et al. (2011)
Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775)
172 A. Melesse, H. Steingass, M. Schollenberger and M. Rodehutscord
and Goel and Makkar (2012) indicated that condensed
could be fed alone or in combination to supplement
and hydrolyzable tannins extracted from a diverse array
tropical feed resources for practical mitigation of CH4
of plant materials reduced CH4 production in vitro.
emissions from ruminants. We recommend animal-based
Similarly, Puchala et al. (2005) demonstrated that the
experiments to validate the actual feeding values of these
presence of condensed tannins in forages can decrease
forages, which showed reduced CH4 production in vitro,
CH4 production in vivo. This was confirmed by Animut
and to assess their production potential.
et al. (2008), who observed decreased CH4 emissions in
sheep fed a ration supplemented with different condensed
Acknowledgments
tannin sources.
The CH4 values measured at 24 h in vitro for M. sativa,
This research work was sponsored by Alexander von
P. purpureum and P. maximum reported by Singh et al.
Humboldt Foundation (Germany) under grant category
(2012) are generally higher than those obtained from the
‘Research Fellowship for Experienced Researchers’, for
current study. These variations could be due to quality and
which the authors are highly grateful. We acknowledge
maturity stage of the forages, soil type and climate in
the Southern Agricultural Research Institute, Holleta
which forages have been grown.
Agricultural Research Center and Forage Seed Production
Enteric CH4 production could be influenced by the
Center (Genebank) of ILRI for allowing the collection of
nature of carbohydrates fermented, such as cellulose,
some feed samples used in this study. We very much
hemicelluloses and soluble residues of the diets. In the
appreciate the excellent technical support provided by
present study, grasses had higher aNDFom, ADFom,
Mrs. Sibylle Rupp and Miss Julia Holstein in chemical
cellulose and hemicellulose concentrations than legumes
and methane gas analyses, respectively.
and produced more CH4 per unit weight. Similarly, Moss
et al. (1994) reported that digestible ADF, cellulose and
References
hemicellulose are important fiber fractions influencing
CH4 production in the rumen.
Many studies (Santoso et al. 2003; Santoso and
Animut G; Puchala R; Goetsch AL; Patra AK; Sahlu T; Varel
VH; Wells J. 2008. Methane emission by goats consuming
Hariadi 2009; Singh et al. 2012) have reported correla-
diets with different levels of condensed tannins from
tions between chemical constituents and CH4 production.
lespedeza. Animal Feed Science and Technology 144:212–
In the current study, CH4 was negatively correlated with
227. DOI: 10.1016/j.anifeedsci.2007.10.014
non-fiber carbohydrates (NFC) only (data not shown) as
Archimède H; Eugène M; Marie Magdeleine C; Boval M;
fermentation of NFC produces less hydrogen due to
Martin C; Morgavi DP; Lecomte P; Doreau M. 2011.
relatively higher propionate production. Accordingly,
Comparison of methane production between C3 and C4
increments in NFC in forages should depress CH
grasses and legumes. Animal Feed Science and Technology
4
production. This has been clearly observed in the current
167:59–64. DOI: 10.1016/j.anifeedsci.2011.04.003
study, in which both P. coloratum and D. intortum had
Beauchemin KA; McGinn SM; Martinez TF; McAllister TA.
high NFC values and produced lower CH
2007. Use of condensed tannin extract from quebracho trees
4 levels than
to reduce methane emissions from cattle. Journal of Animal
other forage species. These results are consistent with the
Science 85:1900–1906. DOI: 10.2527/jas.2006-686
observations of Grainger and Beauchemin (2011), who
Beauchemin KA; Kreuzer M; O’Mara F; McAllister TA. 2008.
reported that increasing NFC levels in feeds reduces CH4
Nutritional management for enteric methane abatement: A
production by lowering pH and increasing rate of ruminal
review. Australian Journal of Experimental Agriculture
passage to favor propionate production, and reduce rumen
48:21–27. DOI: 10.1071/EA07199
protozoal populations.
Bekele AZ; Clement C; Kreuzer M; Soliva CR. 2009.
Efficiency of Sesbania sesban and Acacia angustissima in
Conclusions
limiting methanogenesis and increasing ruminally available
nitrogen in a tropical grass-based diet depends on accession.
The CP concentrations were lower in the grasses than in
Animal Production Science 49:145–153. DOI: 10.1071/
the legumes, while the reverse was the case for aNDFom,
Berhane G; Eik LO; Tolera A. 2006. Chemical composition and
ADFom and cellulose. Methane production was
in vitro gas production of vetch ( Vicia sativa) and some
numerically higher in grasses than legumes. Thus, feeding
browse and grass species in northern Ethiopia. African
of grasses in combination with legumes should result in
Journal of Range and Forage Science 23:69–75. DOI:
enhanced productivity, while reducing CH4 emissions by
ruminants, especially per unit of product. Despite their
Bhatta R; Enishi O; Takusari N; Higuchi K; Nonaka I; Kurihara
lower OMD, it appears that P. coloratum and D. intortum
M. 2008. Diet effects on methane production by goats and a
Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775)
Nutrient composition and methane production of forages in Ethiopia 173
comparison between measurement methodologies. The
Johnson DE; Ward GW; Ramsey JJ. 1996. Livestock methane:
Journal of Agricultural Science 146:705–715. DOI:
Current emissions and mitigation potential. In: Kornegay
ET, ed. Nutrient management of food animals to enhance
Blümmel M; Zerbini E; Reddy BVS; Hash CT; Bidinger F;
and protect the environment. Lewis Publishers, New York,
Ravi D. 2003. Improving the production and utilization of
USA. p. 219–234.
sorghum and pearl millet as livestock feed: Methodological
Jouany JP; Morgavi DP. 2007. Use of ‘natural’ products as
problems and possible solutions. Field Crops Research
alternatives to antibiotic feed additives in ruminant
84:123–142. DOI: 10.1016/S0378-4290(03)00145-X
production.
Animal
1:1443–1466.
DOI:
Blümmel M; Givens DI; Moss AR. 2005. Comparison of
methane produced by straw fed sheep in open-circuit
McCrabb GJ; Hunter RA. 1999. Prediction of methane
respiration with methane predicted by fermentation
emissions from beef cattle in tropical production systems.
characteristics measured by an in vitro gas procedure.
Australian Journal of Agricultural Research 50:1335–1340.
Animal Feed Science and Technology 124:379–390. DOI:
10.1016/j.anifeedsci.2005.06.001
Meissner HH; Köster HH; Nieuwoudt SH; Coertze RJ. 1991.
Boadi D; Benchaar C; Chiquette J; Massé D. 2004. Mitigation
Effects of energy supplementation on intake and digestion
strategies to reduce enteric methane emissions from dairy
of early and mid-season ryegrass and Panicum/ Smuts finger
cows: Update review. Canadian Journal of Animal Science
hay, and on in sacco disappearance of various forage
84:319–335. DOI: 10.4141/A03-109
species. South African Journal of Animal Science 21:33–42.
Cone JW; Van Gelder AH. 1999. Influence of protein
fermentation on gas production profiles. Animal Feed
Melesse A; Steingass H; Boguhn J; Rodehutscord M. 2013. In
Science and Technology 76:251–264. DOI: 10.1016/S0377-
vitro fermentation characteristics and effective utilizable
crude protein in leaves and green pods of Moringa
Debela E; Tolera A; Eik LO; Salte R. 2012. Condensed tannins
stenopetala and Moringa oleifera cultivated at low- and
from Sesbania sesban and Desmodium intortum as a means
mid-altitudes. Journal of Animal Physiology and Animal
of Haemonchus contortus control in goats. Tropical Animal
Nutrition 97:537–546. DOI: 10.1111/j.1439-0396.2012.
Health and Production 44:1939–1944. DOI: 10.1007/
Menke KH; Raab L; Salewski A; Steingass H; Fritz D;
Douglas GB; Wang Y; Waghorn GC; Barry TN; Purchas RW;
Schneider W. 1979. The estimation of the digestibility and
Foote AG; Wilson GF. 1995. Liveweight gain and wool
metabolizable energy content of ruminant feeding stuffs
production of sheep grazing Lotus corniculatus and lucerne
from the gas production when they are incubated with
( Medicago sativa). New Zealand Journal of Agricultural
rumen liquor in vitro. Journal of Agricultural Science
Research
38:95–104.
DOI:
93:217–222. DOI: 10.1017/S0021859600086305
Menke KH; Steingass H. 1988. Estimation of the energetic feed
Elgersma A. 2015. Grazing increases the unsaturated fatty acid
value obtained from chemical analysis and in vitro gas
concentration of milk from grass-fed cows: A review of the
production using rumen fluid. Animal Research for
contributing factors, challenges and future perspectives.
Development 28:7–55.
European Journal of Lipid Science and Technology
Merkel RC; Pond KR; Burns JC; Fisher DS. 1999. Intake,
117:1345–1369. DOI: 10.1002/ejlt.201400469
digestibility and nitrogen utilization of three tropical tree
France J; Beever DE; Siddons RC. 1993. Compartmental
legumes: I. As sole feeds compared to Asystasia intrusa and
schemes for estimating methanogenesis in ruminants from
Brachiaria brizantha. Animal Feed Science and Technology
isotope dilution data. Journal of Theoretical Biology
82:91–106. DOI: 10.1016/S0377-8401(99)00094-2
164:207–218. DOI: 10.1006/jtbi.1993.1149
Mero RN; Udén P. 1997. Promising tropical grasses and
Goel G; Makkar HPS. 2012. Methane mitigation from
legumes as feed resources in Central Tanzania. II. In sacco
ruminants using tannins and saponins. Tropical Animal
rumen degradation characteristics of four grasses and
Health and Production 44:729–739. DOI: 10.1007/s11250-
legumes. Animal Feed Science and Technology 69:341–
352. DOI: 10.1016/S0377-8401(97)85314-X
Grainger C; Beauchemin KA. 2011. Can enteric methane
Mero RN; Udén P. 1998. Promising tropical grasses and
emissions from ruminants be lowered without lowering their
legumes as feed resources in Central Tanzania. III. Effect of
production? Animal Feed Science and Technology
feeding level on digestibility and voluntary intake of four
166/167:308–320. DOI: 10.1016/j.anifeedsci.2011.04.021
grasses by sheep. Animal Feed Science and Technology
Jayanegara A; Wina E; Soliva CR; Marquardt S; Kreuzer M;
70:79–95. DOI: 10.1016/S0377-8401(97)00073-4
Leiber F. 2011. Dependence of forage quality and
Moss AR; Givens DI; Garnsworthy PC. 1994. The effect of
methanogenic potential of tropical plants on their phenolic
alkali treatment of cereal straws on digestibility and
fractions as determined by principal component analysis.
methane production by sheep. Animal Feed Science and
Animal Feed Science and Technology 163:231–243. DOI:
Technology 49:245–259. DOI: 10.1016/0377-8401(94)
10.1016/j.anifeedsci.2010.11.009
Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775)
174 A. Melesse, H. Steingass, M. Schollenberger and M. Rodehutscord
Moss AR; Jouany JP; Newbold J. 2000. Methane production by
Singh S; Kushwaha BP; Nag SK; Mishra AK; Singh A; Anele
ruminants: Its contribution to global warming. Annales de
UY. 2012. In vitro ruminal fermentation, protein and
Zootechnie 49:231–253. DOI: 10.1051/animres:2000119
carbohydrate fractionation, methane production and
Navarro-Villa A; O’Brien M; López S; Boland TM; O’Kiely P.
prediction of twelve commonly used Indian green forages.
2011. In vitro rumen methane output of red clover and
Animal Feed Science and Technology 178:2–11. DOI:
perennial ryegrass assayed using the gas production
10.1016/j.anifeedsci.2012.08.019
technique (GPT). Animal Feed Science and Technology
Soliva CR; Zeleke AB; Clément C; Hess HD; Fievez V;
168:152–164. DOI: 10.1016/j.anifeedsci.2011.04.091
Kreuzer M. 2008. In vitro screening of various tropical
NRC (National Research Council). 2001. Nutrient requirements
foliages, seeds, fruits and medicinal plants for low methane
of dairy cattle. 7th rev. Edn. The National Academies Press,
and high ammonia generating potentials in the rumen.
Washington, DC, USA. DOI: 10.17226/9825
Animal Feed Science and Technology 147:53–71. DOI:
Pamo ET; Boukila B; Fonteh FA; Tendonkeng F; Kana JR;
10.1016/j.anifeedsci.2007.09.009
Nanda AS. 2007. Nutritive value of some grasses and
Tessema Z; Baars RMT. 2006. Chemical composition, dry
leguminous tree leaves of the Central region of Africa.
matter production and yield dynamics of tropical grasses
Animal Feed Science and Technology 135:273–282. DOI:
mixed with perennial forage legumes. Tropical Grasslands
10.1016/j.anifeedsci.2006.07.001
40:150–156. https://goo.gl/rDrbTA
Paterson RT; Roothaert RL; Nyaata OZ; Akyeampong E; Hove
Tiemann TT; Lascano CE; Kreuzer M; Hess HD. 2008. The
L. 1996. Experience with Calliandra calothyrsus as a feed
ruminal degradability of fibre explains part of the low
for livestock in Africa. In: Evans DO, ed. Proceedings of
nutritional value and reduced methanogenesis in highly
international workshop on the genus Calliandra. Winrock
tanniniferous tropical legumes. Journal of Science of Food
International, Bogor, Indonesia. p. 195–209. https://goo.gl/
and Agriculture 88:1794–1803. DOI: 10.1002/jsfa.3282
Van Soest PJ. 1982. The nutritional ecology of the ruminant.
Patra AK; Saxena J. 2010. A new perspective on the use of plant
O and B Books, Corvallis, OR, USA.
secondary metabolites to inhibit methanogenesis in the
VDLUFA (Verband deutscher landwirtschaftlicher Untersu-
rumen. Phytochemistry 71:1198–1222. DOI: 10.1016/
chungs- und Forschungsanstalten). 2007. Handbuch der
Landwirtschaftlichen
Versuchs-
und
Untersuchungs-
Pellikaan WF; Stringano E; Leenaars J; Bongers DJGM; Van
methodik
(VDLUFA-Methodenbuch),
Bd.
III:
Die
Laar-van Schuppen S; Plant J; Mueller-Harvey I. 2011.
chemische Untersuchung von Futtermitteln. VDLUFA-
Evaluating effects of tannins on extent and rate of in vitro
Verlag, Darmstadt, Germany.
gas and CH4 production using an automated pressure
Waghorn GC. 2008. Beneficial and detrimental effects of
evaluation system (APES). Animal Feed Science and
dietary condensed tannins for sustainable sheep and goat
Technology 167:377–390. DOI: 10.1016/j.anifeedsci.
production: Progress and challenges. Animal Feed Science
and Technology 147:116–139. DOI: 10.1016/j.anifeedsci.
Puchala R; Min BR; Goetsch AL; Sahlu T. 2005. The effect of
condensed tannin-containing forage on methane emission
Waghorn GC; Ulyatt MJ; John A; Fisher MT. 1987. The effect
by goats. Journal of Animal Science 83:182–186. DOI:
of condensed tannins on the site of digestion of amino acids
and other nutrients in sheep fed on Lotus corniculatus L.
Ramírez-Restrepo CA; Barry TN. 2005. Alternative temperate
British Journal of Nutrition 57:115–126. DOI: 10.1079/
forages containing secondary compounds for improving
sustainable productivity in grazing ruminants. Animal Feed
Waghorn GC; Tavendale MH; Woodfield DR. 2002.
Science and Technology 120:179–201. DOI: 10.1016/
Methanogenesis from forages fed to sheep. Proceedings of
New Zealand Grassland Association 64:167–171. https://
Santoso B; Kume S; Nonaka K; Kimura K; Mizokoshi H; Gamo
Y; Takahashi J. 2003. Methane emission, nutrient
Waghorn GC; Woodward SL; Tavendale M; Clark DA. 2006.
digestibility, energy metabolism and blood metabolites in
Inconsistencies in rumen methane production ‒ Effects of
dairy cows fed silages with and without galacto-
forage composition and animal genotype. International
oligosaccharides
supplementation.
Asian-Australian
Congress Series 1293:115–118. DOI: 10.1016/j.ics.2006.
Journal of Animal Sciences 16:534–540. DOI: 10.5713/ajas.
Wen L; Kallenbach RL; Williams JE; Roberts CA; Beuselinck
Santoso B; Hariadi BT. 2009. Evaluation of nutritive value and
PR; McGraw RL; Benedict HR. 2002. Performance of steers
in vitro methane production of feed stuffs from agricultural
grazing rhizomatous and nonrhizomatous birdsfoot trefoil in
and food industry by-products. Journal of Indonesian
pure stands and in tall fescue mixtures. Journal of Animal
Tropical Animal Agriculture 34:189–195. DOI: 10.14710/
Science 80:1970–1976. DOI: 10.2527/2002.8071970x
Widiawati Y; Thalib A. 2007. Comparison fermentation
SAS. 2010. SAS Users’ guide, version 9.3. Statistical Analysis
kinetics (in vitro) of grass and shrub legume leaves. The
System (SAS) Institute Inc., Cary, NC, USA.
pattern of VFA concentration, estimated CH4 and microbial
Tropical Grasslands-Forrajes Tropicales (ISSN: 2346-3775)
Nutrient composition and methane production of forages in Ethiopia 175
biomass production. Indonesian Journal of Animal and
extracts and rapeseed tannin monomers on methane
Veterinary Sciences 12:96–104. https://goo.gl/su8jUS
formation and microbial protein synthesis in vitro. Animal
Wischer G; Boguhn J; Steingass H; Schollenberger M;
7:1796–1805. DOI: 10.1017/S1751731113001481
Rodehutscord M. 2013. Effects of different tannin-rich
(Received for publication 23 March 2016; accepted 15 May 2017; published 30 September 2017)
© 2017
Tropical Grasslands-Forrajes Tropicales is an open-access journal published by Centro Internacional de Agricultura Tropical (CIAT). This work is licensed under the Creative Commons Attribution-NonCommercial-ShareAlike 3.0
Unported license. To view a copy of this license, visit http://creativecommons.org/licenses/by-nc-sa/3.0/