- Email: [email protected]
Genetic variability of tropical maize stover quality and the potential for genetic improvement of food-feed value in India
Field Crops Research 153 (2013) 94–101
Contents lists available at ScienceDirect
Field Crops Research journal homepage: www.elsevier.com/locate/fcr
Genetic variability of tropical maize stover quality and the potential for genetic improvement of food-feed value in India P.H. Zaidi a,∗ , M.T. Vinayan a , M. Blümmel b a b
Centro Internacional de Mejoramiento de Maíz y Trigo (CIMMYT), c/o ICRISAT, Patancheru 502324, India International Livestock Research (ILRI), c/o ICRISAT, Patancheru 502324, India
a r t i c l e
i n f o
Article history: Received 21 June 2012 Received in revised form 29 November 2012 Accepted 29 November 2012 Keywords: Maize stover fodder quality Stover genetic enhancement Heterosis effect on stover fodder quality
a b s t r a c t Sixty elite maize inbred lines were selected from CIMMYT-Asia maize program for stover fodder quality analysis. These lines were selected based on high per se and cross performance across several locations in the Asian tropics and have wide adaptability across the region. The line evaluation trials were conducted using recommended agronomic practices during the off-season of 2009 at ICRISAT farm in Hyderabad, India. Data were recorded on various agronomic traits including yields. At harvest the stover was analyzed for a range of fodder quality traits. On the basis of high stover in vitro digestibility (IVOMD) (a trait chosen because it correlated well with pricing in fodder trading of another coarse stover, namely sorghum) and high grain yield, the top ranking 10 inbred lines were selected and crossed in all possible combinations (excluding reciprocals) using diallel mating design in the rainy season of 2010. A total of 26 crosses along with four commercial hybrid checks were selected and planted in two replicates for evaluation during the dry season of 2010. A wide range in performance was observed in grain and stover traits. Grain and stover yields ranged from 0.94 to 7.3 t/ha (P < 0.0001) and from 2.0 to 8.5 t/ha (P < 0.0001), respectively. Stover IVOMD ranged from 46.9 to 55.5% (P = 0.08). Grain and stover traits showed a considerable degree of independency. Grain and stover yield were significantly positively correlated (r = 0.53; P = 0.003), however, grain yields at around 7.0 t/ha showed varied stover yields ranges from 3.5 to 6 t/ha. No significant correlation was observed between IVOMD and grain yield (r = −0.12, P = 0.52) or IVOMD and stover yield (r = 0.03; P = 0.98). It is striking that the cross with the third highest grain yield (7.1 t/ha) had the second highest stover digestibility (about 55%). © 2012 Elsevier B.V. All rights reserved.
1. Introduction Maize has been grown in India for centuries and has long been a popular crop in the low- and mid-altitude areas of the horizontal belt across north-central India (Joshi et al., 2005). In recent decades, maize has emerged as an increasingly important feed and food crop in India and its production is increasingly dominated by high yielding proprietary hybrids from a wide range of large and small private seed companies (Spielman et al., 2011), primarily to meet the needs of an expanding poultry feed market. Maize is now the third most important crop after rice and wheat and it is cultivated on over 8.7 million ha with 20.03 million t of maize grain produced with an average grain yield of 2.3 t/ha (Directorate of
Abbreviations: ADF, acid detergent fiber; ADL, acid detergent lignin; IVOMD, in vitro organic matter digestibility; ME, metabolizable energy; NDF, neutral detergent fiber. ∗ Corresponding author. E-mail address: [email protected] (P.H. Zaidi). 0378-4290/$ – see front matter © 2012 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.fcr.2012.11.020
Maize Research, 2011). Most of the increase has been recorded in non-traditional maize growing areas where it is grown as an irrigated crop for commercial purpose during the dry season (Joshi et al., 2005). In these areas maize is largely replacing sorghum, an important dual-purpose crop, the stover of which is highly valued by livestock keepers and fodder traders (Blümmel and Rao, 2006; Sharma et al., 2010). Given the prevalent fodder shortage in India, maize stover would need to substitute for the loss in sorghum stover but maize stover still appears to be underutilized in India (Erenstein et al., 2011). Farmers in regions, such as Karnataka, consider maize stover to be wasteful and unpalatable for their livestock and the negative perceptions about inferiority of maize compared to sorghum stover is widespread (Biradar, 2004). However maize stover has started appearing at fodder shops in Hyderabad which traditionally sold only sorghum stover (Khan et al., unpublished observations). In addition, on-station livestock productivity trials have shown maize stover to be generally at par with sorghum when used as major component (50%) of total mixed dairy rations (Prasad et al., 2007). A recently completed multidimensional maize improvement project
P.H. Zaidi et al. / Field Crops Research 153 (2013) 94–101
in East Africa suggested that grain and stover traits can be improved simultaneously (Berhanu et al., 2012; Blümmel et al., 2012). Thus even if stover from modern maize cultivars may not be at par with sorghum stover, opportunities still exist to improve maize stover fodder traits. In the present study we assessed available genotypic variability for stover traits in maize, which is prerequisite for further genetic improvement, to explore the opportunities of developing maize cultivars with high stover fodder quality along with improved grain yield of maize cultivars for South Asia.
95
Table 2 Selected bi-parental crosses of maize with combined high grain yield and stover quality traits. S. no.
Pedigree
1 2 3 4 5 6 7
(CML161-BB/CA14514-9-6-3-BB)-B (CML481-BB/CA14514-9-6-3-BB)-B (CML481-BB/CA14514-7-B-2-BB)-B (CLRCY038-BB/CA14514-9-6-3-BB)-B (SW92145-2EV-7-3-B*5-5-B-1-BB/SW3-17-BB-2-BBB-2-B-B1)-B (CML481-BB/CLRCY038-BB)-B (SW92145-2EV-7-3-B*5-5-B-1-BB/CA14514-9-6-3-BB)-B
2. Materials and methods 2.1. Cultivar background and crop management Sixty elite maize inbred lines from the CIMMYT-Asia maize program were evaluated for agronomic performance and stover fodder quality traits during the dry season (Rabi) of 2009 and 10 best performing lines based on in vitro organic matter digestibility (IVOMD) and grain and stover yields were selected (Table 1) and crossed using a diallel mating design during the rainy season (Kharif) of 2010. Twenty-six F1 ’s formed from these crosses were planted along with four commercial hybrids for evaluation during the 2011 Rabi season. The trial was planted following an Alphalattice design with two replications. Each entry was planted in a single row plot, 4 m long. Before planting 60 kg nitrogen (N) ha−1 in the form of urea, 60 kg phosphorous ha−1 as single super phosphate, 40 kg potassium ha−1 as muriate of potash and 10 kg zinc as zinc sulfate were applied as a basal dose. Second and third doses of N (each 30 kg N ha−1 ) were side-dressed when plants were about knee-high and at tasseling, respectively. Pre-emergence application of pendimethalin and atrazine (both at 0.75 kg/ha a.i., tank mixed) were used for weed control. Out of the 26 hybrids used in this study, seven top-ranking crosses in terms of high grain yield and high stover quality were selected and carried forward as bi-parental breeding populations to develop new maize inbred lines with combined high yield and stover fodder quality traits (Table 2). The populations were
Table 1 Description of the selected maize inbred lines with high grain yield and stover fodder quality. S. no.
Pedigree
Description
1
SW92145-2EV-7-3-B*5-5-B-1-B
2
SW1-11-1-B-3-BBB-1-B
3
EY-DMR-C5-S2-BB-3-2-B*6-1-B
4
CML481-B
5
CA14514-7-B-2-B
6
CLRCY038-B
7
CA14514-8-1-2-B
8
CML161
CIMMYT-Asia early yellow line derived from Population 145 CIMMYT-Asia late yellow line derived from Suwan1 cycle 11 CIMMYT-Asia early yellow line derived from early yellow downy mildew resistant population CIMMYT-Asia lines derived from Suwan1 cycle 11 CIMMYT-Asia early yellow line derived from CIMMYT Population 145 CIMMYT-Asia early yellow recycled line CIMMYT-Asia early yellow line derived from Population 145 CIMMYT quality protein maize line derived from pool G25Q c18 CIMMYT-Asia early yellow line derived from CIMMYT Population 145 CIMMYT-Asia late yellow line derived from Suwan3 population
9
10
CA14514-9-6-3-B
SW3-17-BB-2-BBB-2-B
advanced to F2 stage and planted during the Kharif season of 2012 to extract at least 50 F3 progenies from each population.
2.2. Maize stover fodder quality analysis Ten maize plants per treatment were sampled for stover fodder quality analysis, manually chopped and dried at to 60 ◦ C for 48 h and then ground to pass though a 1 mesh. Samples were analyzed using Near Infrared Spectroscopy (NIRS), calibrated for this experiment though use of conventional laboratory analysis. The NIRS instrument used was a FOSS Forage Analyzer 5000 with software package WinISI II. Out of a total of 690 stover samples (from a wider range than the stover investigated in the present work), 345 each were selected for calibration and validation procedures using the WinISI II samples selection programs. Validation procedures were based on blind-predictions of laboratory measurements by the NIRS equations developed during the calibration procedures. Relationships between blind-predicted and measured variables were described by R2 and standard error of prediction (SEP). Relationships between laboratory values and NIRS blind-predicted values were R2 = 0.94 (SEP = 0.06) for nitrogen (N) concentration, R2 = 0.94 (SEP = 1.6) for neutral detergent fiber (NDF), R2 = 0.96 (SEP = 1.2) for acid detergent fiber (ADF), R2 = 0.79 (SEP = 1.6), R2 = 0.82 (SEP = 0.4) for acid detergent lignin (ADL), R2 = 0.81 (SEP = 2.5) for IVOMD and R2 = 0.82 (SEP = 0.4) for ME content. Stover nitrogen (N) was determined (Technicon Auto Analyzer) in duplicate samples and corrected for percentage dry matter (DM). For analysis of IVOMD and metabolizable energy (ME) content, duplicate 200 mg air-dry stover samples were weighed into 100 ml calibrated glass syringes (Menke and Steingass, 1988) and incubated according to the procedure of Blümmel and Ørskov (1993). Rumen inoculum for the incubations was obtained from two ruminally cannulated steers of a local Indian breed maintained on stover supplemented with concentrate. In vitro digestibility and ME content were calculated based on gas volumes produced after 24 h of incubation according to method of Menke and Steingass (1988). Neutral detergent fiber (NDF), acid detergent fiber (ADF) and acid detergent lignin (ADL) were analyzed according to Van Soest et al. (1991).
Table 3 Ranges in grain and stover yields and stover fodder quality traits of 60 inbred maize lines used in the study. Trait
Mean
Minimum
Maximum
SE
Grain yield (GY: kg/ha) Stover yield (SY: kg/ha) Nitrogen (N: %) Neutral detergent fiber (NDF: %) Acid detergent fiber (ADF: %) Acid detergent lignin (ADL: %) In vitro digestibility (IVOMD: %) Metabolizable energy (ME: MJ/kg)
2753 3813 1.20 73.1 35.4 4.0 52.3 7.10
1440 1493 0.79 65.3 31.3 2.9 44.4 5.87
5834 7275 1.64 80.3 39.2 5.3 58.2 7.91
16.3 161.3 0.02 0.33 0.26 0.07 0.33 0.05
96
P.H. Zaidi et al. / Field Crops Research 153 (2013) 94–101
Table 4 Mean performances of sixty maize inbred lines in each sub-cluster. Groups
No. of genotypes
Grain yield (kg/ha)
Stover yield (kg/ha)
N%
NDF %
ADF %
ADL %
ME MJ/kg
IVOMD %
1 2 3 4 5 6
11 8 8 12 10 11
2583 3305 2643 2696 2713 2704
735 994 586 450 810 789
1.37 1.24 1.22 1.17 1.03 1.19
70.6 71.1 75.5 72.9 75.0 73.9
34.0 33.0 38.5 34.9 36.6 35.6
3.4 3.7 4.0 3.8 4.7 4.3
7.55 6.99 7.23 7.28 6.61 6.87
55.6 51.9 53.2 53.2 48.7 51.1
2.3. Statistical analysis Analysis of variance was conducted using the General Linear Models procedure of SAS (2008) for a Complete Randomized
Block Design Model: Yij = + ti + bj + eij, where is overall mean effect, yij denotes the value of observed trait for the i-th treatment (i = 1, 2, . . ., t), received in the jth block (j = 1, 2, . . ., r) with a total number of observations n, ti is fixed effect of the i-th treatment, bj is
Fig. 1. Diversity analysis for stover quality traits of 60 elite inbred lines from CIMMYT-Asia Maize germplasm grouped into six clusters.
P.H. Zaidi et al. / Field Crops Research 153 (2013) 94–101 7000
3. Results
P = 0.16 [n = 59: P = 0.14]
Grain yield (kg/ha)
6000
3.1. Parental lines mean performance and cluster analysis
5000 4000 3000 2000 1000 0 0
1000
2000
3000
4000
5000
6000
7000
8000
Stover yield (kg/ha) Fig. 2. Relationships between stover and grain yields in 60 parental maize lines.
the effect of the j-th block and eij is an experimental error associated with observation of the i-th treatment in the j-th block. Trait relationships were evaluated using linear regression and correlation. An “Unweighted Neighbor-joining dendogram based on dissimilarities of stover quality values between 60 CIMMYT Asia maize inbred lines using Darwin software (Perrier and Jacquemoud-Collet, 2006).
a
Result of the analysis of grain and stover yields and stover fodder quality traits of sixty elite lines from CIMMYT-Asia germplasm are presented in Table 3. The range in grain and stover yields were 4-fold, while for cell wall (NDF), cellulose (ADF) and lignin (ADL), estimates varied by 15 and 7.9 and 2.4 percentage units, respectively. In vitro digestibility and ME contents varied by 13.8 and 2.04 percentage units, respectively. Cluster analysis was used to identify the top ranking lines among the elite lines from CIMMYT-Asia program with respect to high stover IVOMD for use in stover quality breeding activities (Fig. 1 and Table 4). Substantial diversity for stover IVOMD was observed among the elite maize lines used in the study. The analysis divided the entries into 3 major clusters and 6 sub clusters, each consisting of entries with similar quality parameters. Entries in cluster 1 had higher mean values for IVOMD and ME and low trait values for lignin and fiber content, entries of cluster 3 and 4 had moderate levels of IVOMD and metabolic energy and entries in cluster 2, 5 and 6 had lower values for these traits. Most of the entries with favorable trait values for stover fodder quality were placed in cluster 1 (Table 4). Out of eleven lines grouped in cluster 1, nine were used in breeding for enhanced stover fodder quality traits. The remaining two lines were not used because of
b
7000 P = 0.21 [n = 59: y = 3 994 - 1 077x; r = -0.29; P = 0.02]
8000
P = 0.82
7000
Stover yield (kg/ha)
6000
Grain yield (kg/ha)
97
5000 4000 3000 2000
6000 5000 4000 3000 2000
1000
1000
0 0.5 0.6 0.7 0.8 0.9 1.0 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8
0 0.50
0.75
1.00
Stover nitrogen (%)
1.25
1.50
1.75
Stover nitrogen (%)
Fig. 3. (a) Relationships between stover nitrogen and grain yield in 60 parental maize lines and (b) relationships between stover nitrogen and stover yield in 60 parental maize lines.
a
7000
b
P = 0.97 [n = 59: P = 0.82]
Stover yield (kg/ha)
Grain yield (kg/ha)
y = 11 160 - 140.4x; r = -0.26; P = 0.03
7000
6000 5000 4000 3000 2000
6000 5000 4000 3000 2000 1000
1000 0 40.0
8000
42.5
45.0
47.5
50.0
52.5
55.0
57.5
Stover in vitro digestibility
60.0
62.5
0 40.0
42.5
45.0
47.5
50.0
52.5
55.0
57.5
60.0
62.5
Stover in vitro digestibility (%)
Fig. 4. (a) Relationships between stover digestibility and grain yield in 60 parental maize lines and (b) relationships between stover digestibility and stover yield in 60 parental maize lines.
98
P.H. Zaidi et al. / Field Crops Research 153 (2013) 94–101
Table 5 Means and ranges of grain and stover yields and stover fodder quality traits of maize 30 maize hybrids. Trait
Mean
Minimum
Maximum
P>F
Grain yield (GY: kg/ha) Stover yield (SY: kg/ha) Nitrogen (N: %) Neutral detergent fiber (NDF: %) Acid detergent fiber (ADF: %) Acid detergent lignin (ADL: %) In vitro digestibility (IVOMD: %) Metabolizable energy (ME: MJ/kg)
5113 4603 0.98 71.7 38.0 3.7 50.6 7.40
935 1993 0.83 67.2 32.9 2.8 46.7 6.76
7343 8500 1.28 75.0 41.7 4.3 55.5 8.17
0.0001 0.0001 0.25 0.07 0.01 0.26 0.08 0.07
8000
significantly associated with grain yield and association was significant when the outlier was omitted from the analysis and when stover N accounted for 8% of the variation in grain yield (Fig. 3a). No relationship was observed between stover N and stover yield (Fig. 3b). The relationships between stover IVOMD and grain and stover yields are presented in Fig. 4a and b. Stover IVOMD was not related to grain yield (Fig. 4a) but was inversely related to stover yield, accounting for 7% in the variation (Fig. 4b).
y = 1 994 + 0.68x; r = 0.53; P = 0.003 Checks
Grain yield (kg/ha)
7000 6000 5000 4000 3000 2000
3.3. Grain and stover yield and stover quality traits in crosses 1000 0 0
1000
2000
3000
4000
5000
6000
7000
8000
9000
Stover yield (kg/ha) Fig. 5. Relationships between stover and grain yields in 30 maize hybrids.
poor agronomic performance and low grain yield. In addition, one line from cluster 4 was used in the stover forage breeding program, because of its good agronomic performance and high values for stover quality.
3.4. Relationships between grain and stover traits in crosses
3.2. Grain and stover yield and stover quality traits in parental line The relationship between stover and grain yield of the sixty parental lines is presented in Fig. 2. One of the lines had exceptionally high grain yield and relations are reported with and without (in bracket) this visually identified outlier. The association between stover and grain yield was weak and non-significant. The relationships between stover N and grain and stover yields are presented in Fig. 3a and b. Stover N was negative and
8000
Grain and stover yield were significantly and positively related and the variation in one accounting for about 28% of the variation (Fig. 5). Stover N content was inversely correlated to grain yield (P = 0.008) and stover yield (P = 0.06) but the relationships were in both cases influenced by two to three outliers (Fig. 6a and b). The correlation between stover IVOMD and grain and stover yield were statistically non-significant (Fig. 7a and b).
b
9000
7000
8000
6000
7000
Stover yield (kg/ha)
Grain yield (kg/ha)
a
Grain and stover yields and stover fodder quality traits of crosses are presented in Table 5. Observed ranges among the crosses for grain and stover yields were more than 4-fold and statistically highly significant (P < 0.0001). No significant difference among the crosses were observed for stover N (P = 0.25) and lignin content (P = 0.26), while difference in stover cell wall (NDF), cellulose (ADF), IVOMD and ME were significant (P < 0.01 to P = 0.08). In vitro digestibility, one of the major traits selected for in the parental lines, and ME contents varied by 8.8% and 1.41 MJ ME, respectively (Table 5).
5000 4000 3000 2000
6000 5000 4000 3000 2000
Checks
1000
1000
y = 12 832 - 7 855x; r = -0.58; P = 0.0008 0 0.7
0.8
0.9
1.0
1.1
Stover nitrogen (%)
1.2
1.3
1.4
0 0.7
Checks P = 0.06 0.8
0.9
1.0
1.1
1.2
1.3
1.4
Stover nitrogen (%)
Fig. 6. (a) Relationships between stover nitrogen and grain yield in 30 maize hybrids and (b) relationships between stover nitrogen and stover yield in 30 maize hybrids.
P.H. Zaidi et al. / Field Crops Research 153 (2013) 94–101
a
8000
b Stover yield (kg/ha)
Grain yield (kg/ha)
9000 8000
7000 6000 5000 4000 3000 2000 1000
99
7000 6000 5000 4000 3000 2000 1000
P = 0.52 Checks
0 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59
Stover in vitro digestibility (%)
P = 0.88 Checks
0 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59
Stover in vitro digestibility (%)
Fig. 7. (a) Relationships between stover digestibility and grain yield in 30 maize hybrids and (b) relationships between stover digestibility and stover yield in 30 maize hybrids.
4. Discussion 4.1. Maize stover laboratory fodder quality traits Selection for maize stover fodder quality in the present study was largely based on IVOMD. In earlier multi-dimensional crop improvement of coarse cereals such as sorghum and pearl millets, ex-ante impact assessments were based on increasing digestibility (Kristjianson and Zerbini, 1999) of which IVOMD is a close proxy. Kristjianson and Zerbini (1999) proposed that a one-percentage unit increase in digestibility in sorghum or pearl millet stover would result in increases in milk outputs ranging from 6 to 8% (Kristjianson and Zerbini, 1999). This ex-ante assessment was later found to be in general agreement with the result of market studies on the pricing of sorghum stover fodder (Blümmel and Rao, 2006). In survey from 2004 to 2005 involving monthly stover collections from six major sorghum stover traders in Hyderabad, Blümmel and Rao (2006) observed that a mean cultivar-dependent difference in IVOMD of five percentage units (46–51%) was associated with price premium of about 1 Indian Rupee (4 Indian Rupees per kg of dry stover compared to 3 Rupees). Price premiums per one percent increase in IVOMD were in the neighborhood of 6% and variations in IVOMD accounted for about 75% (P = 0.03) of the variation in sorghum stover price while stover N was unrelated (P = 0.63) to pricing (Blümmel and Rao, 2006). In addition, “digestibility” describes the proportion of the feed usable by the animal and is an easier concept to grasp for non-livestock specialists working on crop improvement than more indirect fodder quality traits such as structural carbohydrates measurements such as NDF, ADF and ADL or a quite complex in vitro quality trait such as ME. In other words, stover IVOMD seems to be an appropriate target trait for stover fodder quality improvement in multi-dimensional crop improvement. For the laboratories having no access to ruminally cannulated animals for IVOMD determinations, chemical analysis of structural carbohydrates such as NDF, ADF and ADL, which are all negatively associated with stover fodder quality, can still be used for simple rankings or to predict IVOMD in a socalled summative equation based on regression analysis (Van Soest, 1994). Low N content is often considered the major limiting factor in fodder quality of cereal straws and stover (Sundsto and Owen, 1984) because they do not meet the minimum rumen microorganism requirement for N of 1–1.2% of feed dry matter (Van Soest, 1994). As indicated above, in sorghum stover trading N content was not significantly related to pricing (Blümmel and Rao, 2006) and similar observations have been reported in wheat and rice
straw trading surveys (Teufel, personal communication). Commercial urban and peri-urban dairies using these crop residues might find it easier to supplement stover and straws with other feed N sources than to pay for higher residue N content. Low N content in stover might still be a concern in less intensified livestock systems and we propose maize stover N as possible secondary target trait after IVOMD. 4.2. Genotypic variations in maize stover fodder quality traits The maize stover IVOMD ranged between 44.4 and 58.2% in the parental lines (Table 3) and from 46.7 to 55.5% in the crosses (Table 4). These ranges are greater ranges than traded sorghum stover (46–51%) reported by Blümmel and Rao (2006). Furthermore, maize stover mean IVOMD in the present study was 52.3 and 50.6% for the parental lines (Table 3) and crosses (Table 4), respectively, suggesting that the maize stover fodder quality traits are at least at par with marketed sorghum stover, which is contrary to popular perceptions (Biradar, 2004). This is quite an important finding since sorghum stover marketed as fodder has now a monetary value of half and more than that of the sorghum grain (Sharma et al., 2010). Scaled up on-farm use and marketing of maize stover as fodder would mitigate feed shortages and increase farmers benefit from maize cropping. When the parental lines were clustered based on stover fodder quality traits (Fig. 1) mean sub-cluster values for IVOMD ranged from 48.7 to 55.6% (Table 4). A comparison of the IVOMD of the crosses did not reveal any significant deviation of these traits from the mid-parental values suggesting a simple genetic basis with predominantly additive effects for the traits. This finding is not surprising as the parental lines involved in the hybrids had favorable trait values for the quality parameters resulting in favorable × favorable allelic combinations. Predominance of additive gene action has been reported for several quality traits in sorghum (a close relative of maize) and maize (Pedersen et al., 1982; Roth et al., 1970), hence breeding for these traits is expected to be simple in sorghum. However, a detailed study to understand the genetic basis for the inheritance of these traits in maize is essential for validation of this statement. In Table 6 the performance of bi-parental crosses for stover quality parameters were compared to a leading commercial hybrid in India (900 M Gold from Monsanto). Most of the crosses, except two, showed favorable deviation from the commercial hybrid for quality parameters. Similarly, these crosses showed substantial favorable deviation for other quality traits such as ADF and ADL. Selected
100
P.H. Zaidi et al. / Field Crops Research 153 (2013) 94–101
Table 6 Standard heterosis over one of the leading commercial maize hybrids (900 MG) in India. Hybrids
Pedigree
N%
NDF %
ADL %
ME MJ/kg
ADF %
IVOMD %
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26
SW92145-2EV-7-3-B*5-5-B-1-BB/SW1-11-1-B-3-BBB-1-BB SW92145-2EV-7-3-B*5-5-B-1-BB/CLRCY038-BB SW92145-2EV-7-3-B*5-5-B-1-BB/CA14514-8-1-2-BB SW92145-2EV-7-3-B*5-5-B-1-BB/CA14514-9-6-3-BB SW92145-2EV-7-3-B*5-5-B-1-BB/SW3-17-BB-2-BBB-2-B-B1 SW1-11-1-B-3-BBB-1-BB/EY-DMR-C5-S2-BB-3-2-B*6-1-BB SW1-11-1-B-3-BBB-1-BB/CML481-BB SW1-11-1-B-3-BBB-1-BB/CA14514-7-B-2-BB EY-DMR-C5-S2-BB-3-2-B*6-1-BB/CLRCY038-BB EY-DMR-C5-S2-BB-3-2-B*6-1-BB/CA14514-8-1-2-BB EY-DMR-C5-S2-BB-3-2-B*6-1-BB/CML161-BB EY-DMR-C5-S2-BB-3-2-B*6-1-BB/SW3-17-BB-2-BBB-2-B-B1 CML481-BB/CA14514-7-B-2-BB CML481-BB/CLRCY038-BB CML481-BB/CML161-BB CML481-BB/CA14514-9-6-3-BB CML481-BB/SW3-17-BB-2-BBB-2-B-B1 CLRCY038-BB/CA14514-8-1-2-BB CLRCY038-BB/CML161-BB CLRCY038-BB/CA14514-9-6-3-BB CLRCY038-BB/SW3-17-BB-2-BBB-2-B-B1 CA14514-8-1-2-BB/CML161-BB CA14514-8-1-2-BB/CA14514-9-6-3-BB CA14514-8-1-2-BB/SW3-17-BB-2-BBB-2-B-B1 CML161-BB/CA14514-9-6-3-BB CA14514-9-6-3-BB/SW3-17-BB-2-BBB-2-B-B1
25.0 25.0 25.0 12.5 50.0 25.0 0.00 12.5 12.5 12.5 25.0 12.5 37.5 37.5 0.0 12.5 0.0 12.5 12.5 25.0 37.5 0.0 62.5 37.5 25.0 75.0
−4.3 −3.2 1.4 −2.2 −1.0 −4.9 2.5 2.7 4.2 −0.8 −3.2 0.1 −2.1 −4.2 1.4 3.2 6.7* −1.4 0.3 −0.8 −0.1 0.4 0.4 −2.1 0.6 −2.4
−17.1 −7.3 −4.9 −14.6 −31.7** −24.4* −7.3 −17.1 −4.9 7.3 2.4 −17.1 2.4 −26.8* −4.8 −14.6 −14.6 0.0 0.0 −12.2 −2.4 −7.3 −7.3 −9.8 −22.0 0.1
7.3 8.7 4.4 8.7 14.5** 20.3** 10.2** 5.8 4.4 0.0 −1.5 13.0** −2.9 10.1* 7.3 7.3 7.3 2.9 5.8 11.6* 2.9 5.8 4.4 8.7 18.8** 11.6*
−3.4 −1.0 2.1 −0.3 −9.1 −10.1 5.7 4.4 2.9 1.8 2.6 −2.6 3.6 −6.5 −0.3 0.0 1.0 −2.6 0.8 -8.6 2.3 −3.4 −6.2 −9.8 −14.8** −6.7
6.7 7.1 3.6 8.0 13.6** 19.5** 7.1 5.7 3.1 0.0 −1.5 10.7* −1.1 8.6 4.4 4.4 5.0 −0.2 5.2 7.6 3.1 2.7 4.2 5.9 15.3** 10.7*
* **
Significance at P = 0.05 respectively. Significance at P = 0.01 respectively.
crosses on the basis of combined agronomic performance and fodder quality trait (food-feed type maize) are being used in breeding program as bi-parental breeding populations for extraction of new elite maize lines with substantial improvement in fodder quality and high grain yield.
ambiguous (see Prasad et al., in this issue). While we do not recommend breeding for maize stover N at this point in time, some livestock nutritionally important variations in stover N content might be exploitable. Thus, in higher yielding crosses (c. 6 t/ha), stover N could vary from <0.9 to >1.2% (Fig. 6a) with the latter capable of providing minimum N content to rumen microbes (Van Soest, 1994).
4.3. Potential trade-offs between grain stover traits While maize stover fodder aspects are becoming more important in farmers perceptions about preferable cultivar traits, grain yield is still the most desired trait and stover traits should not compromise with grain yield (De Groote et al., 2012). The most important trait relation is that between grain yield and stover yield. These two traits were not significantly associated in parental lines (Fig. 2) and the correlation in the crosses and checks was comparatively weak (r = 0.53). At high grain yields (c. 7 t/ha), stover yields could vary by more than 2 t/ha (Fig. 5). Such variations in stover yields are important in mixed crop livestock systems where there could be competition for both biomass usage for livestock feed and conservation agriculture needs (Valbuena et al., 2012). Stover IVOMD and grain and stover yields were statistically unrelated and IVOMD at high grain yields (c. 7 t/ha) could vary from approximately 46 to 55% (Fig. 7a). The large degree of independence between stover IVOMD and grain yields observed agree with a synthesis about key findings from multi-dimensional crop improvement of sorghums and pearl millets by Sharma et al. (2010). Maize stover N was significant and negatively associated with grain yields in parental lines (Fig. 3a) and crosses (Fig. 6a). In general, variations in stover N content accounted for 8% and 34% of the variation in grain yields in parental lines (Fig. 3a) and crosses/checks (Fig. 6a), respectively. Similar inverse relations between stover N and grain and stover yields have been observed in pearl millet (Bidinger and Blümmel, 2007). This result suggests that N limitations in the soil and partitioning of N into either grains or stover might impose competitive relations between N accumulation in grain or stover. However, the underlying mechanisms seem more complex since correlations between grain and stover N are
5. Conclusion Significant variations in stover quantity and quality traits of nutritional importance to livestock productivity exist among maize cultivars that can be exploited without any detrimental effect on grain yield. Phenotyping of parental lines for stover traits with subsequent clustering for food and fodder and preferred use of lines which are superior in both traits appears a promising way to develop dual purpose maize hybrids.
References Berhanu, T., Habtamu, Z. S., Twumasi-Afriyie Blümmel, M., Friesen, D., Mosisa, W., Dagne, W., Legesse, W., Girum, A., Tolera, K., Wende, A., 2012. Breeding Maize for Food-Feed Traits in Ethiopia. In: Mosisa Worku, S. Twumasi-Afriyie, Legesse Wolde, Berhanu Tadesse, Girma Demisie, Gezehagn Bogale, Dagne Wegari and B. M. Prasanna, 2012 (editors). Ethiopia Meeting the Challenges of Global Climate Change and Food Security through Innovative Maize Research. Proceedings of the Third National Maize Workshop of Ethiopia. 18-20 April 2011. Addis Ababa, Ethiopia. EIAR/CIMMYT, p 74-80. Bidinger, F.R., Blümmel, M., 2007. Effects of ruminant nutritional quality of pearl millet [Pennisetum glaucum (L) R. Br.] stover. 1. Effects of management alternatives on stover quality and productivity. Field Crops Res. 103, 129–138. Biradar, N., 2004. Analysis of straw and stover of different crops as a livestock feed by the farmers. Ann. Agric. Res. 25, 377–380. Blümmel, M., Lukuyu, B., Zaidi, P.H., Duncan, A.J., Tarawali, S.A., 2012. Dual-Purpose Crop Development, Fodder Trading and Processing Options for Improved Feed Value Chains. In: Mosisa Worku, S. Twumasi-Afriyie, Legesse Wolde, Berhanu Tadesse, Girma Demisie, Gezehagn Bogale, Dagne Wegari and B.M. Prasanna, 2012 (editors). Ethiopia Meeting the Challenges of Global Climate Change and Food Security through Innovative Maize Research. Proceedings of the Third National Maize Workshop of Ethiopia. 18-20 April 2011. Addis Ababa, Ethiopia. EIAR/CIMMYT, p 80-86.
P.H. Zaidi et al. / Field Crops Research 153 (2013) 94–101 Blümmel, M., Ørskov, E.R., 1993. Comparison of in vitro gas production and nylon bag degradability of roughages in predicting feed intake in cattle. Anim. Feed Sci. Technol. 40, 109–119. Blümmel, M., Rao, P.P., 2006. Value of sorghum stover, and stover price to quality relationships in stover trading for urban and peri-urban dairy production. Int. Sorghum Millet Newsl. 47, 97–100. De Groote, H., Dema, G., Sonda, G., Gitonga, Z., 2012. Maize for food and feed in East Africa - the farmer perspective. Field Crops Res. (in press). Directorate of Maize Research, 2011. DMR Vision 2030. Directorate of Maize Research. Pusa Campus, New Delhi 110012, India, p. 24. Erenstein, O., Samaddar, A., Teufel, N., Blümmel, M., 2011. The paradox of limited maize stover use in India’s small holder crop-livestock systems. Exp. Agric. 47, 677–704. Kristjianson, P.M., Zerbini, E., 1999. Genetic enhancement of sorghum and millet residues fed to ruminants ILRI Impact Assessment Series 3, ILRI Nairobi. Joshi, P.K., Singh, N.P., Singh, N.N., Gerpacio, R.V., Pingali, P.L., 2005. Maize in India: Production Systems, Constraints, and Research Priorities. CIMMYT, Mexico, D.F. Menke, K.H., Steingass, H., 1988. Estimation of the energy feed value obtained from chemical analysis and in vitro gas production using rumen fluid. Anim. Res. Dev. 28, 7–55. Perrier, X., Jacquemoud-Collet, J.P., 2006. DARwin software http://darwin.cirad.fr/ Prasad, K.V.S.V., Gupta, M.D., Shah, L., Blümmel, M., 2007. Potential of maize stover from a new dual-purpose hybrid in substituting for sorghum stover in a commercially produced feed block. In: Bakshi, M.P.S., Wadhwa, M. (Eds.), International Tropical Animal Nutrition Conference, vol. 2. NDRI, Karnal.
101
Pedersen, J.F., Gorz, H.J., Haskins, F.A., Ross, W.M., 1982. Variability for quality and agronomic traits in forage sorghum hybrids. Crop Sci. 22, 853–856. Roth, L.S., Marten, G.C., Compton, W.A., Stuthman, D.D., 1970. Genetic variation of quality traits in maize (Zea mays L.) forage. Crop Sci. 10, 365–367. SAS (Statistical Analysis System), 2008. SAS Institute Inc. Release 6.03 Edition. Cary New York, USA. Sharma, K., Pattanaik, A.K., Anandan, S., Blümmel, M., 2010. Food-feed crop research: a synthesis. Anim. Nutr. Feed Tech. 10S, 1–10. Spielman, D.J., Kolady, D., Cavalieri, A., Rao, N.C., 2011. The Seed and Agricultural Biotechnology Industries in India: An Analysis of Industry Structure, Competition, and Policy Options. IFPRI Discussion Paper 01103, International Food Policy Research Institute, Washington DC, USA, p. 36. Sundsto, F., Owen, E., 1984. Straw and Other Fibrous By-products as Feed. Elsevier, Amsterdam. Valbuena, D., Erenstein, O., Homann-Kee, S., Abdoulaye, T., Claessens, L., Duncan A. J., Gerard, B., Rufffino, M. C., Teufel, N., van Rooyen, A., Van Wijk, M. T., 2012. Conservation Agriculture in mixed crop-livestock systems: Scoping crop residue trade-offs in Sub-Saharan Africa and South Asia. Field Crops Res. (in press). Van Soest, P.J., Robertson, J.B., Lewis, B.A., 1991. Methods for dietary fibre, neutral detergent fibre and non-starch polysaccharides in relation to animal nutrition. J. Dairy Sci. 74, 3583–3597. Van Soest, P.J., 1994. Nutritional Ecology of the Ruminant, 2nd ed. Cornell University Press, Ithaca, New York.
Contents lists available at ScienceDirect
Field Crops Research journal homepage: www.elsevier.com/locate/fcr
Genetic variability of tropical maize stover quality and the potential for genetic improvement of food-feed value in India P.H. Zaidi a,∗ , M.T. Vinayan a , M. Blümmel b a b
Centro Internacional de Mejoramiento de Maíz y Trigo (CIMMYT), c/o ICRISAT, Patancheru 502324, India International Livestock Research (ILRI), c/o ICRISAT, Patancheru 502324, India
a r t i c l e
i n f o
Article history: Received 21 June 2012 Received in revised form 29 November 2012 Accepted 29 November 2012 Keywords: Maize stover fodder quality Stover genetic enhancement Heterosis effect on stover fodder quality
a b s t r a c t Sixty elite maize inbred lines were selected from CIMMYT-Asia maize program for stover fodder quality analysis. These lines were selected based on high per se and cross performance across several locations in the Asian tropics and have wide adaptability across the region. The line evaluation trials were conducted using recommended agronomic practices during the off-season of 2009 at ICRISAT farm in Hyderabad, India. Data were recorded on various agronomic traits including yields. At harvest the stover was analyzed for a range of fodder quality traits. On the basis of high stover in vitro digestibility (IVOMD) (a trait chosen because it correlated well with pricing in fodder trading of another coarse stover, namely sorghum) and high grain yield, the top ranking 10 inbred lines were selected and crossed in all possible combinations (excluding reciprocals) using diallel mating design in the rainy season of 2010. A total of 26 crosses along with four commercial hybrid checks were selected and planted in two replicates for evaluation during the dry season of 2010. A wide range in performance was observed in grain and stover traits. Grain and stover yields ranged from 0.94 to 7.3 t/ha (P < 0.0001) and from 2.0 to 8.5 t/ha (P < 0.0001), respectively. Stover IVOMD ranged from 46.9 to 55.5% (P = 0.08). Grain and stover traits showed a considerable degree of independency. Grain and stover yield were significantly positively correlated (r = 0.53; P = 0.003), however, grain yields at around 7.0 t/ha showed varied stover yields ranges from 3.5 to 6 t/ha. No significant correlation was observed between IVOMD and grain yield (r = −0.12, P = 0.52) or IVOMD and stover yield (r = 0.03; P = 0.98). It is striking that the cross with the third highest grain yield (7.1 t/ha) had the second highest stover digestibility (about 55%). © 2012 Elsevier B.V. All rights reserved.
1. Introduction Maize has been grown in India for centuries and has long been a popular crop in the low- and mid-altitude areas of the horizontal belt across north-central India (Joshi et al., 2005). In recent decades, maize has emerged as an increasingly important feed and food crop in India and its production is increasingly dominated by high yielding proprietary hybrids from a wide range of large and small private seed companies (Spielman et al., 2011), primarily to meet the needs of an expanding poultry feed market. Maize is now the third most important crop after rice and wheat and it is cultivated on over 8.7 million ha with 20.03 million t of maize grain produced with an average grain yield of 2.3 t/ha (Directorate of
Abbreviations: ADF, acid detergent fiber; ADL, acid detergent lignin; IVOMD, in vitro organic matter digestibility; ME, metabolizable energy; NDF, neutral detergent fiber. ∗ Corresponding author. E-mail address: [email protected] (P.H. Zaidi). 0378-4290/$ – see front matter © 2012 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.fcr.2012.11.020
Maize Research, 2011). Most of the increase has been recorded in non-traditional maize growing areas where it is grown as an irrigated crop for commercial purpose during the dry season (Joshi et al., 2005). In these areas maize is largely replacing sorghum, an important dual-purpose crop, the stover of which is highly valued by livestock keepers and fodder traders (Blümmel and Rao, 2006; Sharma et al., 2010). Given the prevalent fodder shortage in India, maize stover would need to substitute for the loss in sorghum stover but maize stover still appears to be underutilized in India (Erenstein et al., 2011). Farmers in regions, such as Karnataka, consider maize stover to be wasteful and unpalatable for their livestock and the negative perceptions about inferiority of maize compared to sorghum stover is widespread (Biradar, 2004). However maize stover has started appearing at fodder shops in Hyderabad which traditionally sold only sorghum stover (Khan et al., unpublished observations). In addition, on-station livestock productivity trials have shown maize stover to be generally at par with sorghum when used as major component (50%) of total mixed dairy rations (Prasad et al., 2007). A recently completed multidimensional maize improvement project
P.H. Zaidi et al. / Field Crops Research 153 (2013) 94–101
in East Africa suggested that grain and stover traits can be improved simultaneously (Berhanu et al., 2012; Blümmel et al., 2012). Thus even if stover from modern maize cultivars may not be at par with sorghum stover, opportunities still exist to improve maize stover fodder traits. In the present study we assessed available genotypic variability for stover traits in maize, which is prerequisite for further genetic improvement, to explore the opportunities of developing maize cultivars with high stover fodder quality along with improved grain yield of maize cultivars for South Asia.
95
Table 2 Selected bi-parental crosses of maize with combined high grain yield and stover quality traits. S. no.
Pedigree
1 2 3 4 5 6 7
(CML161-BB/CA14514-9-6-3-BB)-B (CML481-BB/CA14514-9-6-3-BB)-B (CML481-BB/CA14514-7-B-2-BB)-B (CLRCY038-BB/CA14514-9-6-3-BB)-B (SW92145-2EV-7-3-B*5-5-B-1-BB/SW3-17-BB-2-BBB-2-B-B1)-B (CML481-BB/CLRCY038-BB)-B (SW92145-2EV-7-3-B*5-5-B-1-BB/CA14514-9-6-3-BB)-B
2. Materials and methods 2.1. Cultivar background and crop management Sixty elite maize inbred lines from the CIMMYT-Asia maize program were evaluated for agronomic performance and stover fodder quality traits during the dry season (Rabi) of 2009 and 10 best performing lines based on in vitro organic matter digestibility (IVOMD) and grain and stover yields were selected (Table 1) and crossed using a diallel mating design during the rainy season (Kharif) of 2010. Twenty-six F1 ’s formed from these crosses were planted along with four commercial hybrids for evaluation during the 2011 Rabi season. The trial was planted following an Alphalattice design with two replications. Each entry was planted in a single row plot, 4 m long. Before planting 60 kg nitrogen (N) ha−1 in the form of urea, 60 kg phosphorous ha−1 as single super phosphate, 40 kg potassium ha−1 as muriate of potash and 10 kg zinc as zinc sulfate were applied as a basal dose. Second and third doses of N (each 30 kg N ha−1 ) were side-dressed when plants were about knee-high and at tasseling, respectively. Pre-emergence application of pendimethalin and atrazine (both at 0.75 kg/ha a.i., tank mixed) were used for weed control. Out of the 26 hybrids used in this study, seven top-ranking crosses in terms of high grain yield and high stover quality were selected and carried forward as bi-parental breeding populations to develop new maize inbred lines with combined high yield and stover fodder quality traits (Table 2). The populations were
Table 1 Description of the selected maize inbred lines with high grain yield and stover fodder quality. S. no.
Pedigree
Description
1
SW92145-2EV-7-3-B*5-5-B-1-B
2
SW1-11-1-B-3-BBB-1-B
3
EY-DMR-C5-S2-BB-3-2-B*6-1-B
4
CML481-B
5
CA14514-7-B-2-B
6
CLRCY038-B
7
CA14514-8-1-2-B
8
CML161
CIMMYT-Asia early yellow line derived from Population 145 CIMMYT-Asia late yellow line derived from Suwan1 cycle 11 CIMMYT-Asia early yellow line derived from early yellow downy mildew resistant population CIMMYT-Asia lines derived from Suwan1 cycle 11 CIMMYT-Asia early yellow line derived from CIMMYT Population 145 CIMMYT-Asia early yellow recycled line CIMMYT-Asia early yellow line derived from Population 145 CIMMYT quality protein maize line derived from pool G25Q c18 CIMMYT-Asia early yellow line derived from CIMMYT Population 145 CIMMYT-Asia late yellow line derived from Suwan3 population
9
10
CA14514-9-6-3-B
SW3-17-BB-2-BBB-2-B
advanced to F2 stage and planted during the Kharif season of 2012 to extract at least 50 F3 progenies from each population.
2.2. Maize stover fodder quality analysis Ten maize plants per treatment were sampled for stover fodder quality analysis, manually chopped and dried at to 60 ◦ C for 48 h and then ground to pass though a 1 mesh. Samples were analyzed using Near Infrared Spectroscopy (NIRS), calibrated for this experiment though use of conventional laboratory analysis. The NIRS instrument used was a FOSS Forage Analyzer 5000 with software package WinISI II. Out of a total of 690 stover samples (from a wider range than the stover investigated in the present work), 345 each were selected for calibration and validation procedures using the WinISI II samples selection programs. Validation procedures were based on blind-predictions of laboratory measurements by the NIRS equations developed during the calibration procedures. Relationships between blind-predicted and measured variables were described by R2 and standard error of prediction (SEP). Relationships between laboratory values and NIRS blind-predicted values were R2 = 0.94 (SEP = 0.06) for nitrogen (N) concentration, R2 = 0.94 (SEP = 1.6) for neutral detergent fiber (NDF), R2 = 0.96 (SEP = 1.2) for acid detergent fiber (ADF), R2 = 0.79 (SEP = 1.6), R2 = 0.82 (SEP = 0.4) for acid detergent lignin (ADL), R2 = 0.81 (SEP = 2.5) for IVOMD and R2 = 0.82 (SEP = 0.4) for ME content. Stover nitrogen (N) was determined (Technicon Auto Analyzer) in duplicate samples and corrected for percentage dry matter (DM). For analysis of IVOMD and metabolizable energy (ME) content, duplicate 200 mg air-dry stover samples were weighed into 100 ml calibrated glass syringes (Menke and Steingass, 1988) and incubated according to the procedure of Blümmel and Ørskov (1993). Rumen inoculum for the incubations was obtained from two ruminally cannulated steers of a local Indian breed maintained on stover supplemented with concentrate. In vitro digestibility and ME content were calculated based on gas volumes produced after 24 h of incubation according to method of Menke and Steingass (1988). Neutral detergent fiber (NDF), acid detergent fiber (ADF) and acid detergent lignin (ADL) were analyzed according to Van Soest et al. (1991).
Table 3 Ranges in grain and stover yields and stover fodder quality traits of 60 inbred maize lines used in the study. Trait
Mean
Minimum
Maximum
SE
Grain yield (GY: kg/ha) Stover yield (SY: kg/ha) Nitrogen (N: %) Neutral detergent fiber (NDF: %) Acid detergent fiber (ADF: %) Acid detergent lignin (ADL: %) In vitro digestibility (IVOMD: %) Metabolizable energy (ME: MJ/kg)
2753 3813 1.20 73.1 35.4 4.0 52.3 7.10
1440 1493 0.79 65.3 31.3 2.9 44.4 5.87
5834 7275 1.64 80.3 39.2 5.3 58.2 7.91
16.3 161.3 0.02 0.33 0.26 0.07 0.33 0.05
96
P.H. Zaidi et al. / Field Crops Research 153 (2013) 94–101
Table 4 Mean performances of sixty maize inbred lines in each sub-cluster. Groups
No. of genotypes
Grain yield (kg/ha)
Stover yield (kg/ha)
N%
NDF %
ADF %
ADL %
ME MJ/kg
IVOMD %
1 2 3 4 5 6
11 8 8 12 10 11
2583 3305 2643 2696 2713 2704
735 994 586 450 810 789
1.37 1.24 1.22 1.17 1.03 1.19
70.6 71.1 75.5 72.9 75.0 73.9
34.0 33.0 38.5 34.9 36.6 35.6
3.4 3.7 4.0 3.8 4.7 4.3
7.55 6.99 7.23 7.28 6.61 6.87
55.6 51.9 53.2 53.2 48.7 51.1
2.3. Statistical analysis Analysis of variance was conducted using the General Linear Models procedure of SAS (2008) for a Complete Randomized
Block Design Model: Yij = + ti + bj + eij, where is overall mean effect, yij denotes the value of observed trait for the i-th treatment (i = 1, 2, . . ., t), received in the jth block (j = 1, 2, . . ., r) with a total number of observations n, ti is fixed effect of the i-th treatment, bj is
Fig. 1. Diversity analysis for stover quality traits of 60 elite inbred lines from CIMMYT-Asia Maize germplasm grouped into six clusters.
P.H. Zaidi et al. / Field Crops Research 153 (2013) 94–101 7000
3. Results
P = 0.16 [n = 59: P = 0.14]
Grain yield (kg/ha)
6000
3.1. Parental lines mean performance and cluster analysis
5000 4000 3000 2000 1000 0 0
1000
2000
3000
4000
5000
6000
7000
8000
Stover yield (kg/ha) Fig. 2. Relationships between stover and grain yields in 60 parental maize lines.
the effect of the j-th block and eij is an experimental error associated with observation of the i-th treatment in the j-th block. Trait relationships were evaluated using linear regression and correlation. An “Unweighted Neighbor-joining dendogram based on dissimilarities of stover quality values between 60 CIMMYT Asia maize inbred lines using Darwin software (Perrier and Jacquemoud-Collet, 2006).
a
Result of the analysis of grain and stover yields and stover fodder quality traits of sixty elite lines from CIMMYT-Asia germplasm are presented in Table 3. The range in grain and stover yields were 4-fold, while for cell wall (NDF), cellulose (ADF) and lignin (ADL), estimates varied by 15 and 7.9 and 2.4 percentage units, respectively. In vitro digestibility and ME contents varied by 13.8 and 2.04 percentage units, respectively. Cluster analysis was used to identify the top ranking lines among the elite lines from CIMMYT-Asia program with respect to high stover IVOMD for use in stover quality breeding activities (Fig. 1 and Table 4). Substantial diversity for stover IVOMD was observed among the elite maize lines used in the study. The analysis divided the entries into 3 major clusters and 6 sub clusters, each consisting of entries with similar quality parameters. Entries in cluster 1 had higher mean values for IVOMD and ME and low trait values for lignin and fiber content, entries of cluster 3 and 4 had moderate levels of IVOMD and metabolic energy and entries in cluster 2, 5 and 6 had lower values for these traits. Most of the entries with favorable trait values for stover fodder quality were placed in cluster 1 (Table 4). Out of eleven lines grouped in cluster 1, nine were used in breeding for enhanced stover fodder quality traits. The remaining two lines were not used because of
b
7000 P = 0.21 [n = 59: y = 3 994 - 1 077x; r = -0.29; P = 0.02]
8000
P = 0.82
7000
Stover yield (kg/ha)
6000
Grain yield (kg/ha)
97
5000 4000 3000 2000
6000 5000 4000 3000 2000
1000
1000
0 0.5 0.6 0.7 0.8 0.9 1.0 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8
0 0.50
0.75
1.00
Stover nitrogen (%)
1.25
1.50
1.75
Stover nitrogen (%)
Fig. 3. (a) Relationships between stover nitrogen and grain yield in 60 parental maize lines and (b) relationships between stover nitrogen and stover yield in 60 parental maize lines.
a
7000
b
P = 0.97 [n = 59: P = 0.82]
Stover yield (kg/ha)
Grain yield (kg/ha)
y = 11 160 - 140.4x; r = -0.26; P = 0.03
7000
6000 5000 4000 3000 2000
6000 5000 4000 3000 2000 1000
1000 0 40.0
8000
42.5
45.0
47.5
50.0
52.5
55.0
57.5
Stover in vitro digestibility
60.0
62.5
0 40.0
42.5
45.0
47.5
50.0
52.5
55.0
57.5
60.0
62.5
Stover in vitro digestibility (%)
Fig. 4. (a) Relationships between stover digestibility and grain yield in 60 parental maize lines and (b) relationships between stover digestibility and stover yield in 60 parental maize lines.
98
P.H. Zaidi et al. / Field Crops Research 153 (2013) 94–101
Table 5 Means and ranges of grain and stover yields and stover fodder quality traits of maize 30 maize hybrids. Trait
Mean
Minimum
Maximum
P>F
Grain yield (GY: kg/ha) Stover yield (SY: kg/ha) Nitrogen (N: %) Neutral detergent fiber (NDF: %) Acid detergent fiber (ADF: %) Acid detergent lignin (ADL: %) In vitro digestibility (IVOMD: %) Metabolizable energy (ME: MJ/kg)
5113 4603 0.98 71.7 38.0 3.7 50.6 7.40
935 1993 0.83 67.2 32.9 2.8 46.7 6.76
7343 8500 1.28 75.0 41.7 4.3 55.5 8.17
0.0001 0.0001 0.25 0.07 0.01 0.26 0.08 0.07
8000
significantly associated with grain yield and association was significant when the outlier was omitted from the analysis and when stover N accounted for 8% of the variation in grain yield (Fig. 3a). No relationship was observed between stover N and stover yield (Fig. 3b). The relationships between stover IVOMD and grain and stover yields are presented in Fig. 4a and b. Stover IVOMD was not related to grain yield (Fig. 4a) but was inversely related to stover yield, accounting for 7% in the variation (Fig. 4b).
y = 1 994 + 0.68x; r = 0.53; P = 0.003 Checks
Grain yield (kg/ha)
7000 6000 5000 4000 3000 2000
3.3. Grain and stover yield and stover quality traits in crosses 1000 0 0
1000
2000
3000
4000
5000
6000
7000
8000
9000
Stover yield (kg/ha) Fig. 5. Relationships between stover and grain yields in 30 maize hybrids.
poor agronomic performance and low grain yield. In addition, one line from cluster 4 was used in the stover forage breeding program, because of its good agronomic performance and high values for stover quality.
3.4. Relationships between grain and stover traits in crosses
3.2. Grain and stover yield and stover quality traits in parental line The relationship between stover and grain yield of the sixty parental lines is presented in Fig. 2. One of the lines had exceptionally high grain yield and relations are reported with and without (in bracket) this visually identified outlier. The association between stover and grain yield was weak and non-significant. The relationships between stover N and grain and stover yields are presented in Fig. 3a and b. Stover N was negative and
8000
Grain and stover yield were significantly and positively related and the variation in one accounting for about 28% of the variation (Fig. 5). Stover N content was inversely correlated to grain yield (P = 0.008) and stover yield (P = 0.06) but the relationships were in both cases influenced by two to three outliers (Fig. 6a and b). The correlation between stover IVOMD and grain and stover yield were statistically non-significant (Fig. 7a and b).
b
9000
7000
8000
6000
7000
Stover yield (kg/ha)
Grain yield (kg/ha)
a
Grain and stover yields and stover fodder quality traits of crosses are presented in Table 5. Observed ranges among the crosses for grain and stover yields were more than 4-fold and statistically highly significant (P < 0.0001). No significant difference among the crosses were observed for stover N (P = 0.25) and lignin content (P = 0.26), while difference in stover cell wall (NDF), cellulose (ADF), IVOMD and ME were significant (P < 0.01 to P = 0.08). In vitro digestibility, one of the major traits selected for in the parental lines, and ME contents varied by 8.8% and 1.41 MJ ME, respectively (Table 5).
5000 4000 3000 2000
6000 5000 4000 3000 2000
Checks
1000
1000
y = 12 832 - 7 855x; r = -0.58; P = 0.0008 0 0.7
0.8
0.9
1.0
1.1
Stover nitrogen (%)
1.2
1.3
1.4
0 0.7
Checks P = 0.06 0.8
0.9
1.0
1.1
1.2
1.3
1.4
Stover nitrogen (%)
Fig. 6. (a) Relationships between stover nitrogen and grain yield in 30 maize hybrids and (b) relationships between stover nitrogen and stover yield in 30 maize hybrids.
P.H. Zaidi et al. / Field Crops Research 153 (2013) 94–101
a
8000
b Stover yield (kg/ha)
Grain yield (kg/ha)
9000 8000
7000 6000 5000 4000 3000 2000 1000
99
7000 6000 5000 4000 3000 2000 1000
P = 0.52 Checks
0 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59
Stover in vitro digestibility (%)
P = 0.88 Checks
0 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59
Stover in vitro digestibility (%)
Fig. 7. (a) Relationships between stover digestibility and grain yield in 30 maize hybrids and (b) relationships between stover digestibility and stover yield in 30 maize hybrids.
4. Discussion 4.1. Maize stover laboratory fodder quality traits Selection for maize stover fodder quality in the present study was largely based on IVOMD. In earlier multi-dimensional crop improvement of coarse cereals such as sorghum and pearl millets, ex-ante impact assessments were based on increasing digestibility (Kristjianson and Zerbini, 1999) of which IVOMD is a close proxy. Kristjianson and Zerbini (1999) proposed that a one-percentage unit increase in digestibility in sorghum or pearl millet stover would result in increases in milk outputs ranging from 6 to 8% (Kristjianson and Zerbini, 1999). This ex-ante assessment was later found to be in general agreement with the result of market studies on the pricing of sorghum stover fodder (Blümmel and Rao, 2006). In survey from 2004 to 2005 involving monthly stover collections from six major sorghum stover traders in Hyderabad, Blümmel and Rao (2006) observed that a mean cultivar-dependent difference in IVOMD of five percentage units (46–51%) was associated with price premium of about 1 Indian Rupee (4 Indian Rupees per kg of dry stover compared to 3 Rupees). Price premiums per one percent increase in IVOMD were in the neighborhood of 6% and variations in IVOMD accounted for about 75% (P = 0.03) of the variation in sorghum stover price while stover N was unrelated (P = 0.63) to pricing (Blümmel and Rao, 2006). In addition, “digestibility” describes the proportion of the feed usable by the animal and is an easier concept to grasp for non-livestock specialists working on crop improvement than more indirect fodder quality traits such as structural carbohydrates measurements such as NDF, ADF and ADL or a quite complex in vitro quality trait such as ME. In other words, stover IVOMD seems to be an appropriate target trait for stover fodder quality improvement in multi-dimensional crop improvement. For the laboratories having no access to ruminally cannulated animals for IVOMD determinations, chemical analysis of structural carbohydrates such as NDF, ADF and ADL, which are all negatively associated with stover fodder quality, can still be used for simple rankings or to predict IVOMD in a socalled summative equation based on regression analysis (Van Soest, 1994). Low N content is often considered the major limiting factor in fodder quality of cereal straws and stover (Sundsto and Owen, 1984) because they do not meet the minimum rumen microorganism requirement for N of 1–1.2% of feed dry matter (Van Soest, 1994). As indicated above, in sorghum stover trading N content was not significantly related to pricing (Blümmel and Rao, 2006) and similar observations have been reported in wheat and rice
straw trading surveys (Teufel, personal communication). Commercial urban and peri-urban dairies using these crop residues might find it easier to supplement stover and straws with other feed N sources than to pay for higher residue N content. Low N content in stover might still be a concern in less intensified livestock systems and we propose maize stover N as possible secondary target trait after IVOMD. 4.2. Genotypic variations in maize stover fodder quality traits The maize stover IVOMD ranged between 44.4 and 58.2% in the parental lines (Table 3) and from 46.7 to 55.5% in the crosses (Table 4). These ranges are greater ranges than traded sorghum stover (46–51%) reported by Blümmel and Rao (2006). Furthermore, maize stover mean IVOMD in the present study was 52.3 and 50.6% for the parental lines (Table 3) and crosses (Table 4), respectively, suggesting that the maize stover fodder quality traits are at least at par with marketed sorghum stover, which is contrary to popular perceptions (Biradar, 2004). This is quite an important finding since sorghum stover marketed as fodder has now a monetary value of half and more than that of the sorghum grain (Sharma et al., 2010). Scaled up on-farm use and marketing of maize stover as fodder would mitigate feed shortages and increase farmers benefit from maize cropping. When the parental lines were clustered based on stover fodder quality traits (Fig. 1) mean sub-cluster values for IVOMD ranged from 48.7 to 55.6% (Table 4). A comparison of the IVOMD of the crosses did not reveal any significant deviation of these traits from the mid-parental values suggesting a simple genetic basis with predominantly additive effects for the traits. This finding is not surprising as the parental lines involved in the hybrids had favorable trait values for the quality parameters resulting in favorable × favorable allelic combinations. Predominance of additive gene action has been reported for several quality traits in sorghum (a close relative of maize) and maize (Pedersen et al., 1982; Roth et al., 1970), hence breeding for these traits is expected to be simple in sorghum. However, a detailed study to understand the genetic basis for the inheritance of these traits in maize is essential for validation of this statement. In Table 6 the performance of bi-parental crosses for stover quality parameters were compared to a leading commercial hybrid in India (900 M Gold from Monsanto). Most of the crosses, except two, showed favorable deviation from the commercial hybrid for quality parameters. Similarly, these crosses showed substantial favorable deviation for other quality traits such as ADF and ADL. Selected
100
P.H. Zaidi et al. / Field Crops Research 153 (2013) 94–101
Table 6 Standard heterosis over one of the leading commercial maize hybrids (900 MG) in India. Hybrids
Pedigree
N%
NDF %
ADL %
ME MJ/kg
ADF %
IVOMD %
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26
SW92145-2EV-7-3-B*5-5-B-1-BB/SW1-11-1-B-3-BBB-1-BB SW92145-2EV-7-3-B*5-5-B-1-BB/CLRCY038-BB SW92145-2EV-7-3-B*5-5-B-1-BB/CA14514-8-1-2-BB SW92145-2EV-7-3-B*5-5-B-1-BB/CA14514-9-6-3-BB SW92145-2EV-7-3-B*5-5-B-1-BB/SW3-17-BB-2-BBB-2-B-B1 SW1-11-1-B-3-BBB-1-BB/EY-DMR-C5-S2-BB-3-2-B*6-1-BB SW1-11-1-B-3-BBB-1-BB/CML481-BB SW1-11-1-B-3-BBB-1-BB/CA14514-7-B-2-BB EY-DMR-C5-S2-BB-3-2-B*6-1-BB/CLRCY038-BB EY-DMR-C5-S2-BB-3-2-B*6-1-BB/CA14514-8-1-2-BB EY-DMR-C5-S2-BB-3-2-B*6-1-BB/CML161-BB EY-DMR-C5-S2-BB-3-2-B*6-1-BB/SW3-17-BB-2-BBB-2-B-B1 CML481-BB/CA14514-7-B-2-BB CML481-BB/CLRCY038-BB CML481-BB/CML161-BB CML481-BB/CA14514-9-6-3-BB CML481-BB/SW3-17-BB-2-BBB-2-B-B1 CLRCY038-BB/CA14514-8-1-2-BB CLRCY038-BB/CML161-BB CLRCY038-BB/CA14514-9-6-3-BB CLRCY038-BB/SW3-17-BB-2-BBB-2-B-B1 CA14514-8-1-2-BB/CML161-BB CA14514-8-1-2-BB/CA14514-9-6-3-BB CA14514-8-1-2-BB/SW3-17-BB-2-BBB-2-B-B1 CML161-BB/CA14514-9-6-3-BB CA14514-9-6-3-BB/SW3-17-BB-2-BBB-2-B-B1
25.0 25.0 25.0 12.5 50.0 25.0 0.00 12.5 12.5 12.5 25.0 12.5 37.5 37.5 0.0 12.5 0.0 12.5 12.5 25.0 37.5 0.0 62.5 37.5 25.0 75.0
−4.3 −3.2 1.4 −2.2 −1.0 −4.9 2.5 2.7 4.2 −0.8 −3.2 0.1 −2.1 −4.2 1.4 3.2 6.7* −1.4 0.3 −0.8 −0.1 0.4 0.4 −2.1 0.6 −2.4
−17.1 −7.3 −4.9 −14.6 −31.7** −24.4* −7.3 −17.1 −4.9 7.3 2.4 −17.1 2.4 −26.8* −4.8 −14.6 −14.6 0.0 0.0 −12.2 −2.4 −7.3 −7.3 −9.8 −22.0 0.1
7.3 8.7 4.4 8.7 14.5** 20.3** 10.2** 5.8 4.4 0.0 −1.5 13.0** −2.9 10.1* 7.3 7.3 7.3 2.9 5.8 11.6* 2.9 5.8 4.4 8.7 18.8** 11.6*
−3.4 −1.0 2.1 −0.3 −9.1 −10.1 5.7 4.4 2.9 1.8 2.6 −2.6 3.6 −6.5 −0.3 0.0 1.0 −2.6 0.8 -8.6 2.3 −3.4 −6.2 −9.8 −14.8** −6.7
6.7 7.1 3.6 8.0 13.6** 19.5** 7.1 5.7 3.1 0.0 −1.5 10.7* −1.1 8.6 4.4 4.4 5.0 −0.2 5.2 7.6 3.1 2.7 4.2 5.9 15.3** 10.7*
* **
Significance at P = 0.05 respectively. Significance at P = 0.01 respectively.
crosses on the basis of combined agronomic performance and fodder quality trait (food-feed type maize) are being used in breeding program as bi-parental breeding populations for extraction of new elite maize lines with substantial improvement in fodder quality and high grain yield.
ambiguous (see Prasad et al., in this issue). While we do not recommend breeding for maize stover N at this point in time, some livestock nutritionally important variations in stover N content might be exploitable. Thus, in higher yielding crosses (c. 6 t/ha), stover N could vary from <0.9 to >1.2% (Fig. 6a) with the latter capable of providing minimum N content to rumen microbes (Van Soest, 1994).
4.3. Potential trade-offs between grain stover traits While maize stover fodder aspects are becoming more important in farmers perceptions about preferable cultivar traits, grain yield is still the most desired trait and stover traits should not compromise with grain yield (De Groote et al., 2012). The most important trait relation is that between grain yield and stover yield. These two traits were not significantly associated in parental lines (Fig. 2) and the correlation in the crosses and checks was comparatively weak (r = 0.53). At high grain yields (c. 7 t/ha), stover yields could vary by more than 2 t/ha (Fig. 5). Such variations in stover yields are important in mixed crop livestock systems where there could be competition for both biomass usage for livestock feed and conservation agriculture needs (Valbuena et al., 2012). Stover IVOMD and grain and stover yields were statistically unrelated and IVOMD at high grain yields (c. 7 t/ha) could vary from approximately 46 to 55% (Fig. 7a). The large degree of independence between stover IVOMD and grain yields observed agree with a synthesis about key findings from multi-dimensional crop improvement of sorghums and pearl millets by Sharma et al. (2010). Maize stover N was significant and negatively associated with grain yields in parental lines (Fig. 3a) and crosses (Fig. 6a). In general, variations in stover N content accounted for 8% and 34% of the variation in grain yields in parental lines (Fig. 3a) and crosses/checks (Fig. 6a), respectively. Similar inverse relations between stover N and grain and stover yields have been observed in pearl millet (Bidinger and Blümmel, 2007). This result suggests that N limitations in the soil and partitioning of N into either grains or stover might impose competitive relations between N accumulation in grain or stover. However, the underlying mechanisms seem more complex since correlations between grain and stover N are
5. Conclusion Significant variations in stover quantity and quality traits of nutritional importance to livestock productivity exist among maize cultivars that can be exploited without any detrimental effect on grain yield. Phenotyping of parental lines for stover traits with subsequent clustering for food and fodder and preferred use of lines which are superior in both traits appears a promising way to develop dual purpose maize hybrids.
References Berhanu, T., Habtamu, Z. S., Twumasi-Afriyie Blümmel, M., Friesen, D., Mosisa, W., Dagne, W., Legesse, W., Girum, A., Tolera, K., Wende, A., 2012. Breeding Maize for Food-Feed Traits in Ethiopia. In: Mosisa Worku, S. Twumasi-Afriyie, Legesse Wolde, Berhanu Tadesse, Girma Demisie, Gezehagn Bogale, Dagne Wegari and B. M. Prasanna, 2012 (editors). Ethiopia Meeting the Challenges of Global Climate Change and Food Security through Innovative Maize Research. Proceedings of the Third National Maize Workshop of Ethiopia. 18-20 April 2011. Addis Ababa, Ethiopia. EIAR/CIMMYT, p 74-80. Bidinger, F.R., Blümmel, M., 2007. Effects of ruminant nutritional quality of pearl millet [Pennisetum glaucum (L) R. Br.] stover. 1. Effects of management alternatives on stover quality and productivity. Field Crops Res. 103, 129–138. Biradar, N., 2004. Analysis of straw and stover of different crops as a livestock feed by the farmers. Ann. Agric. Res. 25, 377–380. Blümmel, M., Lukuyu, B., Zaidi, P.H., Duncan, A.J., Tarawali, S.A., 2012. Dual-Purpose Crop Development, Fodder Trading and Processing Options for Improved Feed Value Chains. In: Mosisa Worku, S. Twumasi-Afriyie, Legesse Wolde, Berhanu Tadesse, Girma Demisie, Gezehagn Bogale, Dagne Wegari and B.M. Prasanna, 2012 (editors). Ethiopia Meeting the Challenges of Global Climate Change and Food Security through Innovative Maize Research. Proceedings of the Third National Maize Workshop of Ethiopia. 18-20 April 2011. Addis Ababa, Ethiopia. EIAR/CIMMYT, p 80-86.
P.H. Zaidi et al. / Field Crops Research 153 (2013) 94–101 Blümmel, M., Ørskov, E.R., 1993. Comparison of in vitro gas production and nylon bag degradability of roughages in predicting feed intake in cattle. Anim. Feed Sci. Technol. 40, 109–119. Blümmel, M., Rao, P.P., 2006. Value of sorghum stover, and stover price to quality relationships in stover trading for urban and peri-urban dairy production. Int. Sorghum Millet Newsl. 47, 97–100. De Groote, H., Dema, G., Sonda, G., Gitonga, Z., 2012. Maize for food and feed in East Africa - the farmer perspective. Field Crops Res. (in press). Directorate of Maize Research, 2011. DMR Vision 2030. Directorate of Maize Research. Pusa Campus, New Delhi 110012, India, p. 24. Erenstein, O., Samaddar, A., Teufel, N., Blümmel, M., 2011. The paradox of limited maize stover use in India’s small holder crop-livestock systems. Exp. Agric. 47, 677–704. Kristjianson, P.M., Zerbini, E., 1999. Genetic enhancement of sorghum and millet residues fed to ruminants ILRI Impact Assessment Series 3, ILRI Nairobi. Joshi, P.K., Singh, N.P., Singh, N.N., Gerpacio, R.V., Pingali, P.L., 2005. Maize in India: Production Systems, Constraints, and Research Priorities. CIMMYT, Mexico, D.F. Menke, K.H., Steingass, H., 1988. Estimation of the energy feed value obtained from chemical analysis and in vitro gas production using rumen fluid. Anim. Res. Dev. 28, 7–55. Perrier, X., Jacquemoud-Collet, J.P., 2006. DARwin software http://darwin.cirad.fr/ Prasad, K.V.S.V., Gupta, M.D., Shah, L., Blümmel, M., 2007. Potential of maize stover from a new dual-purpose hybrid in substituting for sorghum stover in a commercially produced feed block. In: Bakshi, M.P.S., Wadhwa, M. (Eds.), International Tropical Animal Nutrition Conference, vol. 2. NDRI, Karnal.
101
Pedersen, J.F., Gorz, H.J., Haskins, F.A., Ross, W.M., 1982. Variability for quality and agronomic traits in forage sorghum hybrids. Crop Sci. 22, 853–856. Roth, L.S., Marten, G.C., Compton, W.A., Stuthman, D.D., 1970. Genetic variation of quality traits in maize (Zea mays L.) forage. Crop Sci. 10, 365–367. SAS (Statistical Analysis System), 2008. SAS Institute Inc. Release 6.03 Edition. Cary New York, USA. Sharma, K., Pattanaik, A.K., Anandan, S., Blümmel, M., 2010. Food-feed crop research: a synthesis. Anim. Nutr. Feed Tech. 10S, 1–10. Spielman, D.J., Kolady, D., Cavalieri, A., Rao, N.C., 2011. The Seed and Agricultural Biotechnology Industries in India: An Analysis of Industry Structure, Competition, and Policy Options. IFPRI Discussion Paper 01103, International Food Policy Research Institute, Washington DC, USA, p. 36. Sundsto, F., Owen, E., 1984. Straw and Other Fibrous By-products as Feed. Elsevier, Amsterdam. Valbuena, D., Erenstein, O., Homann-Kee, S., Abdoulaye, T., Claessens, L., Duncan A. J., Gerard, B., Rufffino, M. C., Teufel, N., van Rooyen, A., Van Wijk, M. T., 2012. Conservation Agriculture in mixed crop-livestock systems: Scoping crop residue trade-offs in Sub-Saharan Africa and South Asia. Field Crops Res. (in press). Van Soest, P.J., Robertson, J.B., Lewis, B.A., 1991. Methods for dietary fibre, neutral detergent fibre and non-starch polysaccharides in relation to animal nutrition. J. Dairy Sci. 74, 3583–3597. Van Soest, P.J., 1994. Nutritional Ecology of the Ruminant, 2nd ed. Cornell University Press, Ithaca, New York.