Genetic variation in the feeding value of barley and wheat straw

B.S. Capper¹, E.F. Thomson² and F. Herbert³

1. Animal Feeds Section, Overseas Development and Natural Resources Institute, 56/62 Grays Inn Road, London, WC1X 8LU, United Kingdom

2. Pasture, Forage and Livestock Program, International Center for Agricultural Research in the Dry Areas (ICARDA), P.O. Box 5466, Aleppo, Syria

3. Wye College, University of London, Wye, Ashford, Kent, TN25 5AH, United Kingdom

Introduction
Materials and methods
Results
Discussion
References
Discussion

Introduction

Straw is one of the most important feeds for sheep in West Asia. Barley and wheat straw may constitute half the dry matter of the diets of pregnant and lactating ewes in winter and stubbles make an important contribution to the maintenance of sheep flocks in the summer. In Syria, farmers have rejected an improved barley variety because its straw was less palatable to sheep than straw from a barley landrace (Nygaard, 1983). Voluntary intake and digestibility studies have confirmed farmers' observations that feeding value of straws differs among varieties (Capper et al, 1986).

It has been suggested that the proportions of leaf and stem in the straw, together with variations in chemical composition and microstructure of morphological fractions, are responsible for genetic variation in the feeding value of straw (Capper, in press). Environment also appears to affect straw quality, such that material from higher rainfall locations contains less leaf than that from semi-arid locations.

This paper reports observations on the morphological composition and in vitro digestibility of straw and examines the possibility of relating straw quality to varietal characteristics such as plant height, time to grain maturity and grain yield. The effects of environment on straw quality and the size of variety x environment interactions are considered. Data are presented on the use of the nylon bag technique for predicting the digestible dry matter intake of straw from different barley varieties and comparisons are made with the use of chemical composition and varietal characteristics for predictive purposes. Since ewe feeding in winter utilises a major part of the available feed resources in Syria, experiments are described on the effects of variety on straw consumption and milk yields in Awassi ewes.

Materials and methods

In 1985 samples of barley and wheat were taken from breeding and variety trials at Tel Hadya, Breda and Bouider, situated between 25 and 100 km south of Aleppo in northwest Syria. Plant height, days to maturity or heading, and grain yield were measured on individual plots. There were two replicates for F3 bulk selections of barley, three for barley landraces and four for wheat variety trials. Mid-plot samples, which weighed approximately 100 g, were separated into heads, leaf blade, leaf sheath and stem fractions. Samples of barley landraces and F4 bulk selections, harvested from trials with three replications in 1986, were separated into heads and straw. In vitro digestibility analysis (Tilley and Terry, 1963), standardised using samples of straw of known in vivo digestibility, was carried out on leaf blade, leaf sheath and stem fractions collected in 1985 and on straw harvested in 1986.

Material for feeding trials using unsupplemented barley straw was grown at Tel Hadya (35°-1 55' N. 36°55' E) using a seed rate of 100 kg ha and fertilizer levels of 50 kg N ha^-1 and 50 kg P₂O₅ ha . Rainfall was 229 mm in 1984 and 373 mm in 1985. Prior to harvest, plant height and days to maturity were recorded. Quadrats were cut to estimate grain and straw yields and samples of the crop were fractionated to determine the proportion of leaf. In 1984 the crops were harvested by machine at a cutting height of 20 cm but in 1985 the crops were hand-pulled. Material from each variety was passed through a stationary thresher to give chopped straw with a stem length of 2.5 cm. Digestible dry matter intake was measured with Awassi wethers in digestibility crates. Four measurements were made on each straw. A refusal level of 20% was used. Trials lasted 28 days with voluntary intake being measured and faeces collected over the final 14 days. The crude protein (MAFF, 1981), modified acid-detergent fibre (MAFF, 1973) and neutral-detergent fibre (Goering and Van Soest, 1970) contents of the straws were determined prior to feeding. Nylon bag degradability techniques (Ørskov et al, 1980) were applied to straws evaluated in feeding trials using wethers.

Straw for feeding trials with ewes was obtained from crops grown at Tel Hadya in 1986 using similar methods to those described above. During weeks 2 to 6 of lactation, 36 individually penned ewes were randomly allocated to four straws from different varieties. Straw intake was recorded using a refusal level of 20%. A fixed level of concentrate was provided. The concentrate contained 78% barley grain, 10% cottonseed meal, 10% wheat bran, 1% salt and 1% dicalcium phosphate. Animals were weighed weekly. Milk yields were determined by hand milking and by subsequently weighing lambs before and after suckling. During weeks 8 to 12 of lactation ewes on straw from each variety were allocated to three subtreatments with different levels of concentrate feeding according to previous levels of straw consumption. Straw intake, liveweight changes and milk yields were recorded as described above.

Results

Morphological composition of straws

The mean morphological composition of straw of barley and wheat varieties grown at ICARDA is given in Table 1. Leaf blade and sheath made up a higher proportion of the mature plant than stem.

Table 1. Proportions of morphological fractions in barley and wheat straw harvested in 1985 (mean ± SE).

Crop	Number of varieties	Leaf blade		Leaf sheath		Stem
		Mean	SE	Mean	SE	Mean	SE
Barley F3 bulk	38	0.42	0.02	0.24	0.01	0.34	0.01
Barley landraces	20	0.39	0.02	0.24	0.01	0.37	0.01
Wheat (regional trial)	24	0.29	0.01	0.24	0.01	0.48	0.01
Wheat (elite trial)	29	0.37	0.01	0.26	0.01	0.38	0.01

Correlation coefficients ® given in Table 2 show that the proportion of leaf blade was determined by plant height, straw from taller varieties of barley and wheat containing less leaf blade than that of shorter varieties. In the F₃ barleys, the higher yielding varieties had significantly less leaf blade in the straw. In barley landraces and the wheat trials, days to maturity or heading were associated with an increase in the proportion of leaf blade and a reduction in the proportion of stem. These effects were not statistically significant because of the dominant influence of plant height in determining proportion of leaf blade in the straw.

Table 2. Relationships ® between barley and wheat characteristics and proportions of morphological fractions in straw harvested in 1985.

Crop	Characteristic	Leaf blade	Leaf sheath	Stem
Barley F3 bulk (n=38)	Plant height	-0,75***	0.25	0.74***
	Days to maturity	-0.26	0.09	0.27
	Grain yield	-0.65***	0.14	0.69***
Barley land - races (n=20)	Plant height	-0.58**	0.11	0.60**
	Days to maturity	0.13	0.17	-0.18
	Grain yield	-0.15	-0.05	0.20
Wheat (regional trial) (n=24)	Plant height	-0.29	0.40	0.21
	Days to maturity	0.21	0.37	-0.31
	Grain yield	0.19	0.24	-0.24
Wheat (elite trial) (n=29)	Plant height	-0.78***	0.19	0.76***
	Days to maturity	0.21	-0.08	-0.20

**=P<0. 01; ***=P<0. 001.

In vitro digestibility of morphological fractions

In barley varieties, leaf blade was more digestible than leaf sheath which was, in turn, more digestible than stem (Table 3). In wheat, leaf blade was more digestible than stem. Digestibility of leaf sheath was similar to that of leaf blade in one wheat trial but was similar to stem digestibility in another wheat trial. The effects of varietal characteristics on the digestibility of the different fractions were, generally, non-significant and variable (Table 4). In barley landraces stem digestibility was lower in taller varieties. Higher grain yields tended to be associated with lower digestibility of the fractions. In some cases a positive relationship was observed but, with one exception, these were not statistically significant.

Table 3. In vitro digestibility (%) of morphological fractions of barley and wheat straw harvested in 1985.

Crop	Number of varieties	Leaf blade		Leaf sheath		Stem
		Mean	SE	Mean	SE	Mean	SE
Barley F3 bulk	38	55.5	0.6	45.4	0.6	39.4	0.5
Barley landraces	20	48.1	0.8	42.4	0.7	36.6	0.9
Wheat	24	39.0	0.4	38.7	0.2	30.6	0.2
Wheat	50	42.2	0.2	30.5	0.3	29.6	0.2

In vitro digestibility of straw from different locations

Data presented in Table 5 show that the digestibility of straw from a given variety can be affected very considerably by environment. Barley straw from Tel Hadya was less digestible than that from the drier sites, Breda and Bouider. Table 6 summarises analyses of variance associated with the data in Table S. Most of the variation in straw digestibility was associated with location, but variety had a greater effect than the interaction between location and variety. In the ease of landraces of the black, or Arabic Aswad, type the effects of variety were significant.

Table 4. Relationships ® between barley and wheat characteristics and in vitro digestibility of morphological fractions in straw harvested in 1985.

Crop	Characteristic	Leaf blade	Leaf sheath	Stem
Barley F3 bulk (n=38)	Plant height	-0.34*	-0.26	-0.25
	Days to maturity	0.01	-0.03	-0.08
	Grain yield	-0.11	-0.12	-0.33*
Barley land - races (n=20)	Plant height	0.14	0.14	-0.45*
	Days to maturity	0.09	0.12	0.18
	Grain yield	-0.06	0.10	-0.36
Wheat(n=24)	Plant height	-0.14	0.20	0.28
	Days to maturity	-0.02	-0.11	0.09
	Grain yield	-0.08	-0.37	-0.13
Wheat(n=50)	Plant height	0.02	0.04	-0.19
	Days to maturity	0.06	0.01	0.16
	Grain yield	0.09	-0.20	-0.01

*=P<0 .05.

Table 5. In vitro straw digestibility for barley varieties harvested at three locations in 1986.

Trial	Site (average rainfall, mm)
	Tel Hadya (330)		Breda (283)		Bouider (205)
	Mean	SE	Mean	SE	Mean	SE
Landraces, Arabic Aswad type	37.2	0.26	43.3	0.25	52.9	0.37
Landraces, Arabic Abiad type	40.8	0.23	46.5	0.28	51.7	0.44
F4 bulk selections	36.7	0.29	45.9	0.33	55.3	0.32

Table 6. Summary of analyses of variance for location, variety and interaction effects on the In vitro digestibility of barley straw harvested in 1986.

Trial	Number of entries	Location	Variety	Inter. action
Landraces, Arabic Aswad type	25	4603.8***	11.5***	4.2
Landraces, Arabic Abiad type	20	1797.8***	6.8	6.3
F4 bulk selections	25	4331.4***	7.6	3.8

***=P<0.001.

Feeding trials with unsupplemented barley straw

The digestible dry matter intake (DDMI) of straw by Awassi sheep was affected by barley variety (Table 7). However, the ranking of straws for DDMI varied between 1984 and 1985. The variety Arar, for example, had the lowest DDMI in 1984 and the second highest DDMI in 1985. In 1985 DDMI was closely and positively associated with dry-matter digestibility and nylon-bag degradability (NBD) of the straw but DDMI was not associated with these parameters in 1984 (Table 8). Crude-protein content was positively associated with DDMI and acid-detergent fibre content was negatively associated with DDMI. Leaf proportion and days from planting to maturity were positively associated with straw feeding value but stem height did not influence DDMI. Increase in DDMI was associated with a reduction in grain yield but the relationships were non-significant. The regression coefficients for these relationships (Table 9) suggest that for each 1.0 g kg^-1 W^0.75 day^-1 increase in DDMI, grain yields will be reduced by around 70 kg ha^-1.

Table 7. Digestible dry matter intake (g kg^-1 W^0.75 day^-1) of straw from seven barley varieties grown in successive seasons.

Variety	Year of harvest
	1984	1985
C-63	21.89 (1)	21.41 (4)
Antares	18.35 (2)	27.31 (1)
Rihane	18.26 (3)	21.48 (3)
Badia	17.70 (4)	16.42 (7)
ER/Apam	16.73 (5)	16.59 (6)
Beecher	16.57 (6)	16.98 (5)
Arar	15.49 (7)	22.09 (2)

Straw harvested by machine in 1984 and by hand in 1985.
Numbers in parentheses are ranks.

Table 8. Prediction ® of digestible dry matter intake (g kg^-1 W^0.75 day^-1) of straw from various characteristics of seven barley varieties grown in successive seasons.

Characteristic		Year of harvest
		1984	1985
Digestibility (%)
	In vivo dry matter	-0.12	0.97***
Nylon-bag degradability (%)
	48-hour dry matter	-0.30	0.76*
	72-hour dry matter	n.d.	0.84*
Chemical composition (%)
	Crude protein	0.90**	0.94**
	Acid-detergent fibre	-0.40	-0.83*
	Neutral-detergent fibre	0.24	-0.54
Varietal characteristics
	Leaf proportion	0.58	0.64
	Days to maturity	0.61	0.81*
	Stem height (cm)	0.20	-0.26
	Grain yield (kg ha-1)	-0.27	-0.51
	Straw yield (kg ha-1)	0.67	0.01

*=P<0.05; **=P<0.01; ***=P<0.001; n.d.= not determined.

Table 9. Relationships between grain yields (kg ha) and intake of digestible dry matter (g kg^-1 W^0.75 day^-1) of barley straw.

Year of harvest	Regression coefficient	Intercept
1984	-62.3	3141.3
1985	-80.1	3401.2

Feeding trials with Awassi ewes

During weeks 2 to 6 of lactation ewes consumed more straw from the 2-rowed varieties Arabic Abiad and ER/Apam than from the 6-rowed varieties Beecher and C-63 (Table 10). The animals performed best when fed straw from the local landrace Arabic Abiad. Animals fed Beecher straw lost most liveweight and those fed straw from C-63 had low daily milk yields. In weeks 8 to 12 of lactation, during which each straw was fed with three levels of concentrate, milk production was highest on Arabic Abiad straw and lowest on straw from C-63 (Table 11). Ewes fed lower levels of concentrate consumed more straw. Ewes consumed more straw from the 2-rowed varieties Arabic Abiad and ER/Apam than from Beecher and C-63.

Table 10. Concentrate consumption, voluntary intake of straw from contrasting barley varieties, liveweight changes and milk yields of Awassi ewes in weeks 2 to 6 of lactation.

Barley variety	Concentrate consumption1	Straw intake1	Liveweight change2	Milk yield2
Arabic Abiad	45.1	44.9	-39.7	1634.7
ER/Apam	45.6	44.3	-50.3	1583.6
Beecher	43.1	37.2	-95.2	1580.5
C-63	42.6	36.2	-52.9	1263.9
Mean	44.1	40.6	-59.5	1515.7
SE +	1.7	2.3	24.1	109.4

1. g kg^-1 W^0.75 day^-1.
2. g day^-1.

Discussion

The proportions of leaf, leaf sheath and stem vary considerably among barley and wheat straws and this variation has been shown to be caused mainly by variations in plant height and, to a lesser extent, days to maturity or heading. Considerable variation also exists in the digestibility of fractions but it does not appear possible to relate this variation readily to varietal characteristics. Ramazin et al (1986), using the nylon bag technique, found that differences in degradability of fractions among varieties were more important than differences in morphological composition in determining barley straw quality.

Table 11. Concentrate consumption, voluntary intake of straw from contrasting barley varieties, liveweight changes and milk yields of Awassi ewes in weeks 8 to 12 of lactation.

Barley variety	Concentrate consumption1	Straw intake1	Live weight change2	Milk yield2
Arabic	48.1	40.9	12.4	549.0
Abiad	34.6	50.2	1.9	463.3
	18.3	51.2	0.9	387.3
ER/Apam	47.4	36.9	4.3	478.3
	35.4	41.4	9.5	489.0
	19.9	49.5	3.6	307.3
Beecher	48.2	34.4	3.6	535.5
	34.4	38.0	6.7	387.7
	17.4	43.1	-5.0	355.8
C-63	43.2	31.5	6.4	314.5
	31.8	42.8	2.1	281.5
	14.2	39.1	-8.0	386.8
Mean	32.2	41.2	2.6	405.8
SE	2.0	1.1	1.2	72.9

1. g kg^-1 W0 75 day^-1.
2. g day^-1.

However, the varieties they tested had leaf proportions of 0.44 and 0.53 whereas leaf proportions reported in Syria for barley ranged from 0.50 to 0.81 and from 0.45 to 0.74 for wheat. This suggests that the proportion of leaf in the straw is an important factor in determining variation in straw quality. Whereas Ramazin et al (1986) suggested that only 20% of variation in straw quality can be attributed to variation in morphological composition, the results reported here on DDMI of barley straws suggest that about 40% of the variation in straw feeding value relates to variation in morphological composition. The causes of the remaining variation in straw value are not readily apparent but may relate to the chemical composition and microstructure of morphological fractions. The only relationship found between a varietal characteristic and the digestibility of a morphological fraction is that between plant height and stem digestibility in barley.

Feeding trials with lactating ewes support the contention that straw from shorter, 2-rowed varieties have higher feeding value than straw from taller, 6-rowed varieties (Capper et al, 1986). The 2-rowed varieties usually have a larger proportion of leaf in the straw than 6-rowed varieties. However, trials with unsupplemented barley straws fed to Awassi wethers suggest that there is a positive relationship between days to maturity and DDMI of straw. The results suggest overall that shorter or later-maturing varieties are likely to have better quality straw than tall or early-maturing varieties. Crude-protein content was found to be closely related to straw DDMI but this may be a consequence of higher crude protein levels in leaf material. Where animals receive protein supplements there may be no direct relationship between straw crude protein content and feeding value. The significant relationship between nylon bag degradability and DDMI in straw harvested in 1985 suggests that the technique may be of value in routine screening of straws for feeding value.

The relationship between straw feeding value and grain yield, although not significant, is of considerable importance to the farmer in deciding which variety to plant. Feeding trials with unsupplemented barley straw suggested that for each 1.0 g kg^-1 W^0.75 day^-1 increase in straw feeding value (DDMI), grain yields would be reduced by about 70 kg ha^-1. In order to increase straw feeding by about 50%, from 15 g kg^-1 W^0.75 day 1 to 22.5 g kg^-1 W^0.75 day^-1, grain yields would be reduced from 2200 kg ha^-1 to about 1700 kg ha^-1. In semi-arid areas, where harvest indices are low, the improved quality of the straw may more than offset the reduction in grain yield. Feeding trials with lactating Awassi ewes suggest that feeding lower quality straw could reduce milk production by about one third and result in lower liveweights at the end of lactation, which could affect subsequent breeding performance.

The correlation coefficients for relationships between straw feeding value and grain yield in barley were not significant, suggesting that varieties could be chosen which combine good grain yield with superior straw feeding value. The selection or breeding of varieties of barley and wheat with superior straw feeding value depends upon stability in various measures of straw quality across years or locations. It has been demonstrated that variety has a greater effect on in vitro digestibility of straw than the interaction between variety and location. Thus plant breeders could select varieties that will give better quality straw in a range of environments. However, in vitro digestibility methods, current chemical methods and the use of varietal characteristics are not infallible in ranking varieties for straw quality and alternative methods need to be investigated. At present the use of nylon bag degradability techniques appears to hold promise.

References

Capper B S. (in press). Genetic variation in the feeding value of straw. Animal Feed Science and Technology.

Capper B S. Thomson E F. Rihawi S. Termanini A and McRae R. 1986. The feeding value of straw from different genotypes of barley when given to Awassi sheep. Animal Production 42:337-342.

Goering H K and Van Soest P J. 1970. Forage fiber analysis. Agriculture Handbook No. 379. Agricultural Research Service, United States Department of Agriculture, Washington DC, USA.

MAFF (Ministry of Agriculture, Fisheries and Food). 1973. The analysis of agricultural materials. Technical Bulletin 27. Her Majesty's Stationery Office, London, UK.

MAFF (Ministry of Agriculture, Fisheries and Food). 1981. The analysis of agricultural materials. Technical Bulletin RB427. Her Majesty's Stationery Office, London, UK.

Nygaard D. 1983. Tests on farmer's fields: The ICARDA experience. Proceedings of the First Farming Systems Research Symposium, Kansas State University, Manhattan, Kansas, USA. pp. 76-98.

Ørskov E R. Hovell F D DeB and Mould F. 1980. The use of the nylon bag technique for the evaluation of feedstuffs. Tropical Animal Production 5: 195-213.

Ramazin M, Ørskov E R and Tuah A K. 1986. Rumen degradation of straw. 2. Botanical fractions of straw from two barley cultivars. Animal Production 43:271-278.

Tilley J M A and Terry R A. 1963. A two-stage technique for the in vitro digestion of forage crops. Journal of the British Grassland Society 18:104-111.

Discussion

Ørskov: Is there a correlation between straw quality and grain yield?

Capper: In in vitro digestibility trials both in Syria and the United Kingdom you get the same result: correlation coefficients are very low, even though sometimes they are negative.

Schildkamp: I would like to comment on rice in the Philippines where, because of taste differences, farmers grow a very old variety of rice with reddish grain for home consumption and IR36 for commercial production. Farmers preferred to feed the straw of the older variety to their water buffalo. We did not look at chemical composition but farmers said that the buffalo ate more of the straw from the older variety. They preserved some IR36 straw in case they had shortages of straw from the older variety.

Capper: In response to your comment, it appears that from chemical composition and in vitro analysis, selection for increased grain yield in varieties such as IR36 has not changed straw quality a great deal. I am not claiming that selection for higher harvest index has improved straw quality but at least it has not changed.

Pearce: I am quite sure that there are palatability differences between straws and between varieties that we know very little about.

Capper: Much more work is needed to determine factors that lead to these differences.