DOI: 10.1111/jpn.12156
ORIGINAL ARTICLE
Effect of genotype on chemical composition, ruminal degradability and in vitro fermentation characteristics of maize residual plants F. M. E. Zeller1, B. L. Edmunds2 and F. J. Schwarz1 €nchen, Freising, Germany, and 1 Department of Animal Sciences, Section Animal Nutrition, Technische Universit€at Mu 2 Department of Agriculture and Food Western Australia, Bunbury, WA, Australia
Summary The objective of this study was to determine the changes to residual plant feeding value of early- and latematuring maize varieties. The influence of the cell wall carbohydrate composition, in terms of neutral and acid detergent fibre (NDF and ADF) content, NDF and dry matter (DM) degradability, and in vitro organic matter digestibility and gas production on the feeding value of a range of maize genotypes, was measured. The different genotypes were allotted into two maturity groups (MG I – early to mid-early: S210-S240; MG II – mid-late to late: S 250-S280) and harvested at four different harvest dates (depending on the DM content of the kernels). The maize varieties of MG I had lower NDF and ADF contents and higher ruminal DM degradability, in vitro digestibility and gas production and thus a higher feeding value than MG II at the same stage of physiological maturity. A strong negative relationship between NDF content and the ruminal DM degradability (r = 0.81) was observed. The data indicate that the early-maturing varieties permit a larger flexibility in harvesting due to a longer period of starch inclusion into the kernel whilst simultaneously maintaining a good supply of rumen-available fibre. Conclusively, the higher feeding value of the early-maturing varieties, based on lower NDF and high DM digestibility, permits more flexibility in the harvesting period over the later-maturing varieties. Keywords maize, stover, rumen, harvest, degradation, digestibility €nchen, Liesel-Beckmann-Str. 4, Correspondence F. M. E. Zeller, Department of Animal Sciences, Section Animal Nutrition, Technische Universit€at Mu 85350 Freising, Germany. Tel: +49(0)8161 496441; Fax: +49(0)89 5328383; E-mail: [email protected] Received: 8 August 2013; accepted: 21 November 2013
Introduction Maize silage is one of the most important forage crops in ruminant nutrition in central Europe, providing energy in the form of starch from the cobs and structural fibre from the plant. During maturation, there is a large change in the chemical composition of maize plants in the way of increased starch deposition in cobs, and fibre and lignin deposition in the residual plant. Most research has focused on improving the feeding value of the cob; however, at silage maturity, the residual plant is proportionately 0.40–0.50 of the total dry matter (DM) in maize plants (Irlbeck et al., 1993; Valentin et al., 1999). The residual plant is thus of high importance to the feeding value of maize silage. Large genetic differences have been observed in quality indexes such as in vivo digestibility as well as neutral and acid detergent fibre (NDF, ADF) and acid 982
detergent lignin (ADL) concentrations of the whole maize plant and particularly of the maize residual plant (Deinum, 1988; Cone and Engels, 1993; Flachowsky et al., 1993; Jung and Buxton, 1994; Schwarz et al., 1996; M echin et al., 1998, 2000; Cone et al., 2008; Dobberstein et al., 2008; Zeller et al., 2009). The objective of this study was to examine the influence of early- and late-maturing genotypes on the extent of the changes to structural fibre, lignin, rumen degradability and digestibility in the residual plant using in situ and in vitro methods. Based on previous findings (Schwarz et al., 1996; Ettle and Schwarz, 2003), it was expected that the early-maturing varieties would be superior in feed value due to lower fibre content, and the results were sought to clarify which chemical components of the plant cell walls interact with the degradability and digestibility of maize and which traits are potentially important for plant breeders to improve the feeding value.
Journal of Animal Physiology and Animal Nutrition 98 (2014) 982–990 © 2014 Blackwell Verlag GmbH
F. M. E. Zeller, B. L. Edmunds and F. J. Schwarz
Effect of genotype on feeding value of maize stover
Materials and methods
Agronomic traits
Experimental design
The soil type of the experimental plots was sand clay. The average temperature for the region was 7.5 °C, and the average rainfall per year was 800 mm. Sowing dates for the 3 years were as follows: 27 April 2004; 02 May 2005; and 04 May 2006 respectively. Liquid manure was applied as fertilizer 2 days before sowing (25 cubic metres per ha in 2004 and 20 cubic metres per ha in 2005 and 2006). In all 3 years, the day before sowing, 200 kg of inorganic fertilizer was applied per ha in the form of urea (46% N). Pest management was applied 4 weeks after sowing and was different each year; 2004, 1 l Motivell and 2.5 l Artett per ha; 2005, Motivell/Zintan Gold Pack 1 l and Artett 2.5 l per ha; 2006, 3.6 l Gardo Gold and 0.9 l Calisto per ha.
Fourteen different maize hybrids were included in the investigation to adequately cover genotypic variability. The fourteen hybrids were separated into two maturity groups (MGs). The early- to midearly-maturing varieties (according to the German silage ripening categories: DMK, 2009) were assigned to MG I (S210-S240; n = 7). Varieties cultivated in MG I were as follows: GP 101001 (S210), DK 233 (S220), DKc2942 (S220), EC 2803 (S230), DKc 2949 (S230), ED 3214 (S230) and DK 247 (S 240). The mid-late- to late-maturing varieties were assigned to MG II (S250-S280; n = 7). The varieties cultivated in MG II were as follows: DKc 2960 (S250), ED 3412 (S250), DK 281 (S260), Monumental (S260), DK 291 (S260), DK 287 (S260) and DK 315 (S280). Cultivation of six of the hybrids (DK 233, DKc 2949, DK 247 in MG I and DK 281, Monumental, DK 287 in MG II) was repeated for three harvest years (2004, 2005 and 2006). Varieties DK 291 and DK 315 were cultivated only in 2004. All other hybrids were cultivated in years 2005 and 2006. Harvest time of the varieties within the different maturity groups was based upon the physiological maturity of the plants and not on a fixed date. All hybrids were harvested at four harvest dates (HDs), which were decided by the DM content of the kernel (except for variety ED 3214 which was not available at the first HD). This method of determining the HD was chosen to ensure the same stage of physiological maturity for all varieties of the different maturity groups upon harvest. The selected DM (g/kg) content in the grain for each HDs 1–4, respectively, was as follows: 480–520, 540–580, 600–640 and 650–700. Harvest date 3 corresponds approximately to the typical maturation of maize harvested for ensiling (silage stage). These criteria were kept for all varieties in all 3 years, thereby reproducing the physiological development of the stover. The harvest period for MG I was from 31st of August to 25th of October and for MG II from 6th of September to the 25th of October. Due to weather conditions, the DM in the grain of the last HD for MG II was lower (p < 0.05) than that of MG I in all three harvest years. Around the 25th of October, night frost stopped further physiological development of the plants in all 3 years. When all hybrids from all harvest dates and years are considered, a total number of 67 samples for MG I and 60 samples for MG II were collected.
Harvest of the samples was carried out only during dry weather. Sampling was implemented as follows: each variety was cut at three separate representative locations in the experimental plot. At each location, 10 plants in a row were cut 10–15 cm above ground. Consequently, each sample consisted of 30 plants. After harvesting, the maize cob was discarded and only the stover and husks were used for further investigations. All complete plant samples (n = 127) were chopped, and a subsample (approx. 2 kg) was freezedried and ground to pass a 3-mm screen for determination of ruminal degradability and a 1-mm screen for chemical analyses. The determination of ash and crude protein (CP) was carried out by Weender analysis (VDLUFA, 2007). Determination of the structural cell wall components NDF, ADF and ADL of the plants was conducted using an ANKOM220 Fibre Analyzer (Komarek, 1993). NDF was assayed without heat-stable amylase, and both NDF and ADF are expressed inclusive of residual ash. Entire chemical analyses were accomplished according to the official methods of VDLUFA (2007). The NDF content of residues measuring in situ degradability was determined using near-infrared spectroscopy (NIRS) (NIR System Model 6500 spectrophotometer, FOSS, Hamburg). The samples were scanned in a range from 1100 to 2500 nm. Reflectance data were collected as 1/R. All samples were measured twice and averaged. The calibration set comprised 185 samples of maize plants (stover). NDF ranged from 56.4% to 88.6% DM (mean 67.3% DM). The validation set included 47 samples. NDF
Journal of Animal Physiology and Animal Nutrition © 2014 Blackwell Verlag GmbH
983
Sample preparation and analysis
Effect of genotype on feeding value of maize stover
ranged from 66.4% to 89.5% DM (mean 68.4% DM). Calibration statistics included standard error of calibration (SEC = 0.881), coefficient of determination in calibration (R2 = 0.985), standard error of cross-validation (SECV = 1.049) and coefficient of determination in cross-validation R2CV = 0.978). The slope of the calibration curve was 0.933.
F. M. E. Zeller, B. L. Edmunds and F. J. Schwarz
Only four varieties from each maturity group were analysed for NDF degradation. Selected from MG I were DK 233, DKc 2949 and DK 247 from harvest years 2004 and 2005 and variety EC 2803 from harvest year 2005 (n = 28). Selected from MG II were varieties DK 281, Monumental and DK 287 from harvest years 2004 and 2005 and variety DKc2960 from harvest year 2005 (n = 28).
In situ degradability
For determination of in situ ruminal degradability, six non-lactating dairy cows (Holstein Friesian) fitted with ruminal cannulae were used. Cows were fed twice daily (07:00 and 16:00) with a total amount of (DM basis) 3.0 kg maize silage, 2.72 kg hay, 0.4 kg soy bean meal and 100 g mineral mixture. Approximately 4.0 g of freeze-dried stover, ground to pass a 3-mm screen, was enclosed in pre-weighed nylon bags (10 9 20 cm, average pore size 53 10 lm; Bar Diamond Inc., Parma) and inserted into the rumen. Each sample was incubated in duplicate in three animals (n = 6) for 2, 4, 8, 16, 24, 48, 72 and 96 h. Following removal, bags were submerged in iced water to stop microbial action and then rinsed continuously until the water ran clear. Bags were then further rinsed by machine (Fa. QUELLE; WVA Basic 74) on cold cycle with two water changes for 20 min (without soap or spinning) and then subsequently freeze-dried and weighed to determine DM residue. Washout fraction (0 h) was measured by washing and re-weighing unincubated bags (n = 3) using the same machine washing procedure. The NDF content of residues after 0, 4, 24, 48 and 96 h of incubation was estimated using NIRS. Thus, it was possible to generate ruminal degradation curves for DM and NDF and calculate the potential ruminal degradability. Effective ruminal degradability of DM (EDDM) and NDF (EDNDF) was estimated using the formula of McDonald (1981), modified according to S€ udekum (2005): kt0
ED ¼ a þ ½ðb cÞ=ðc þ kÞ e
where ED is the effective ruminal degradability, a represents the soluble and rapidly degradable fraction, b represents the potentially degradable fraction, c is the degradation rate of b, k is the rate constant measuring the passage of solid particle flow from the rumen to the abomasum and t0 is the lag time during which no microbial degradation occurs. The NLIN procedure of SAS (SAS, 1990) was used to estimate parameters in the model for DM and NDF degradation data. 984
In vitro organic matter digestibility
The digestibility of organic matter (OM) in vitro was measured following the technique described by De Boever et al. (1986). Briefly, 0.3 g of sample was incubated for 24 h in pepsin–HCl solution at 40 °C, then 45 min at 80 °C. After filtration, the contents were rinsed with warm water. A cellulase–buffer mixture (cellulase Onozuka R-10 extracted from Trichoderma viride) was added and the sample returned to the incubator for a further 24 h at 40 °C. After filtration and washing, the residue was dried and weighed. It was then ignited in a muffle furnace, the annealing loss equating to the cellulaseindigestible OM of the sample. Subtraction of the
Table 1 Dry matter content (g/kg) and chemical composition (g/kg DM) of the maturity groups at different stages of maturity Harvest date 1 Dry matter MG I 202.7 MG II 195.8 Ash MG I 57.7 MG II 55.5 Crude protein MG I 63.0a MG II 54.9b Crude fibre MG I 320.9b MG II 347.2a Neutral detergent fibre MG I 542.7b MG II 579.6a Acid detergent fibre MG I 337.6b MG II 368.4a Acid detergent lignin MG I 48.4 MG II 52.9
2
3
4
SEM*
206.9 203.3
229.9 235.1
287.1 274.0
38.1 37.5
58.1 56.8
59.7 58.0
58.3 56.4
7.17 6.83
56.3 52.8
52.2 48.9
47.4 48.6
9.60 7.96
337.4b 354.6a
345.6b 367.7a
351.8b 382.5a
23.4 26.1
570.8 591.9
594.1b 623.6a
619.0 644.4
40.9 45.0
356.0b 375.6a
369.1b 396.0a
385.3b 413.6a
28.0 29.1
51.1 54.1
52.1 56.3
54.7b 59.7a
7.69 6.32
Means from the same harvest date with different superscripts differ between the maturity groups (a,bP 0.05). Crude protein was higher (p < 0.05) for MG I at HD 1. Fibre fractions (NDF, ADF) were lower in MG I at all HDs, although not always significantly. Fibre fractions increased in both MGs during maturation. The group means (g/kg DM) over all HDs for the fibre fractions for MG I and II, respectively, were as follows: NDF 582 vs. 610 (p < 0.01), ADF 362 vs. 388 (p < 0.001). Differences in ADL contents between genotypes were found only in a very late stage of maturity (HD 4; Table 1), although ADL content for MG I tended to be lower at all HDs. As an overall mean, MG I had a lower ADL (g/kg DM) content than MG II (52 vs. 58; p < 0.001). Throughout maturation, ADL increased slightly but this was only significant in MG II (p < 0.05). Dry matter and neutral detergent fibre degradation of residual plant matter in situ
F. M. E. Zeller, B. L. Edmunds and F. J. Schwarz
Table 2 Dry matter degradation parameters in situ of the two maturity groups at the different harvest dates Harvest date 1 a [g/kg DM] MG I 339.3a MG II 301.6b b [g/kg DM] MG I 403.5 MG II 414.2 d [g/kg DM] MG I 742.9a MG II 715.8b c [%/h] MG I 4.83 MG II 4.85 Lag time [h] MG I 1.71 MG II 1.56 EDDM6 (g/kg DM)* MG I 500.4a MG II 467.6b
2
3
4
SEM*
302.5 297.0
286.5 273.2
267.7 254.6
46.5 44.6
420.8 406.1
438.0 425.2
444.8 429.9
43.5 35.4
723.3 703.0
724.5a 698.3b
712.5a 684.5b
29.8 37.7
4.52 5.01
4.22 4.36
3.96 4.31
0.88 0.87
1.44 1.76
2.02 2.08
2.74 2.60
1.46 1.28
467.7 451.6
446.7 430.3
417.7 408.4
47.3 42.6
Means from the same harvest date with different superscripts differ between the maturity groups (a,bP
ORIGINAL ARTICLE
Effect of genotype on chemical composition, ruminal degradability and in vitro fermentation characteristics of maize residual plants F. M. E. Zeller1, B. L. Edmunds2 and F. J. Schwarz1 €nchen, Freising, Germany, and 1 Department of Animal Sciences, Section Animal Nutrition, Technische Universit€at Mu 2 Department of Agriculture and Food Western Australia, Bunbury, WA, Australia
Summary The objective of this study was to determine the changes to residual plant feeding value of early- and latematuring maize varieties. The influence of the cell wall carbohydrate composition, in terms of neutral and acid detergent fibre (NDF and ADF) content, NDF and dry matter (DM) degradability, and in vitro organic matter digestibility and gas production on the feeding value of a range of maize genotypes, was measured. The different genotypes were allotted into two maturity groups (MG I – early to mid-early: S210-S240; MG II – mid-late to late: S 250-S280) and harvested at four different harvest dates (depending on the DM content of the kernels). The maize varieties of MG I had lower NDF and ADF contents and higher ruminal DM degradability, in vitro digestibility and gas production and thus a higher feeding value than MG II at the same stage of physiological maturity. A strong negative relationship between NDF content and the ruminal DM degradability (r = 0.81) was observed. The data indicate that the early-maturing varieties permit a larger flexibility in harvesting due to a longer period of starch inclusion into the kernel whilst simultaneously maintaining a good supply of rumen-available fibre. Conclusively, the higher feeding value of the early-maturing varieties, based on lower NDF and high DM digestibility, permits more flexibility in the harvesting period over the later-maturing varieties. Keywords maize, stover, rumen, harvest, degradation, digestibility €nchen, Liesel-Beckmann-Str. 4, Correspondence F. M. E. Zeller, Department of Animal Sciences, Section Animal Nutrition, Technische Universit€at Mu 85350 Freising, Germany. Tel: +49(0)8161 496441; Fax: +49(0)89 5328383; E-mail: [email protected] Received: 8 August 2013; accepted: 21 November 2013
Introduction Maize silage is one of the most important forage crops in ruminant nutrition in central Europe, providing energy in the form of starch from the cobs and structural fibre from the plant. During maturation, there is a large change in the chemical composition of maize plants in the way of increased starch deposition in cobs, and fibre and lignin deposition in the residual plant. Most research has focused on improving the feeding value of the cob; however, at silage maturity, the residual plant is proportionately 0.40–0.50 of the total dry matter (DM) in maize plants (Irlbeck et al., 1993; Valentin et al., 1999). The residual plant is thus of high importance to the feeding value of maize silage. Large genetic differences have been observed in quality indexes such as in vivo digestibility as well as neutral and acid detergent fibre (NDF, ADF) and acid 982
detergent lignin (ADL) concentrations of the whole maize plant and particularly of the maize residual plant (Deinum, 1988; Cone and Engels, 1993; Flachowsky et al., 1993; Jung and Buxton, 1994; Schwarz et al., 1996; M echin et al., 1998, 2000; Cone et al., 2008; Dobberstein et al., 2008; Zeller et al., 2009). The objective of this study was to examine the influence of early- and late-maturing genotypes on the extent of the changes to structural fibre, lignin, rumen degradability and digestibility in the residual plant using in situ and in vitro methods. Based on previous findings (Schwarz et al., 1996; Ettle and Schwarz, 2003), it was expected that the early-maturing varieties would be superior in feed value due to lower fibre content, and the results were sought to clarify which chemical components of the plant cell walls interact with the degradability and digestibility of maize and which traits are potentially important for plant breeders to improve the feeding value.
Journal of Animal Physiology and Animal Nutrition 98 (2014) 982–990 © 2014 Blackwell Verlag GmbH
F. M. E. Zeller, B. L. Edmunds and F. J. Schwarz
Effect of genotype on feeding value of maize stover
Materials and methods
Agronomic traits
Experimental design
The soil type of the experimental plots was sand clay. The average temperature for the region was 7.5 °C, and the average rainfall per year was 800 mm. Sowing dates for the 3 years were as follows: 27 April 2004; 02 May 2005; and 04 May 2006 respectively. Liquid manure was applied as fertilizer 2 days before sowing (25 cubic metres per ha in 2004 and 20 cubic metres per ha in 2005 and 2006). In all 3 years, the day before sowing, 200 kg of inorganic fertilizer was applied per ha in the form of urea (46% N). Pest management was applied 4 weeks after sowing and was different each year; 2004, 1 l Motivell and 2.5 l Artett per ha; 2005, Motivell/Zintan Gold Pack 1 l and Artett 2.5 l per ha; 2006, 3.6 l Gardo Gold and 0.9 l Calisto per ha.
Fourteen different maize hybrids were included in the investigation to adequately cover genotypic variability. The fourteen hybrids were separated into two maturity groups (MGs). The early- to midearly-maturing varieties (according to the German silage ripening categories: DMK, 2009) were assigned to MG I (S210-S240; n = 7). Varieties cultivated in MG I were as follows: GP 101001 (S210), DK 233 (S220), DKc2942 (S220), EC 2803 (S230), DKc 2949 (S230), ED 3214 (S230) and DK 247 (S 240). The mid-late- to late-maturing varieties were assigned to MG II (S250-S280; n = 7). The varieties cultivated in MG II were as follows: DKc 2960 (S250), ED 3412 (S250), DK 281 (S260), Monumental (S260), DK 291 (S260), DK 287 (S260) and DK 315 (S280). Cultivation of six of the hybrids (DK 233, DKc 2949, DK 247 in MG I and DK 281, Monumental, DK 287 in MG II) was repeated for three harvest years (2004, 2005 and 2006). Varieties DK 291 and DK 315 were cultivated only in 2004. All other hybrids were cultivated in years 2005 and 2006. Harvest time of the varieties within the different maturity groups was based upon the physiological maturity of the plants and not on a fixed date. All hybrids were harvested at four harvest dates (HDs), which were decided by the DM content of the kernel (except for variety ED 3214 which was not available at the first HD). This method of determining the HD was chosen to ensure the same stage of physiological maturity for all varieties of the different maturity groups upon harvest. The selected DM (g/kg) content in the grain for each HDs 1–4, respectively, was as follows: 480–520, 540–580, 600–640 and 650–700. Harvest date 3 corresponds approximately to the typical maturation of maize harvested for ensiling (silage stage). These criteria were kept for all varieties in all 3 years, thereby reproducing the physiological development of the stover. The harvest period for MG I was from 31st of August to 25th of October and for MG II from 6th of September to the 25th of October. Due to weather conditions, the DM in the grain of the last HD for MG II was lower (p < 0.05) than that of MG I in all three harvest years. Around the 25th of October, night frost stopped further physiological development of the plants in all 3 years. When all hybrids from all harvest dates and years are considered, a total number of 67 samples for MG I and 60 samples for MG II were collected.
Harvest of the samples was carried out only during dry weather. Sampling was implemented as follows: each variety was cut at three separate representative locations in the experimental plot. At each location, 10 plants in a row were cut 10–15 cm above ground. Consequently, each sample consisted of 30 plants. After harvesting, the maize cob was discarded and only the stover and husks were used for further investigations. All complete plant samples (n = 127) were chopped, and a subsample (approx. 2 kg) was freezedried and ground to pass a 3-mm screen for determination of ruminal degradability and a 1-mm screen for chemical analyses. The determination of ash and crude protein (CP) was carried out by Weender analysis (VDLUFA, 2007). Determination of the structural cell wall components NDF, ADF and ADL of the plants was conducted using an ANKOM220 Fibre Analyzer (Komarek, 1993). NDF was assayed without heat-stable amylase, and both NDF and ADF are expressed inclusive of residual ash. Entire chemical analyses were accomplished according to the official methods of VDLUFA (2007). The NDF content of residues measuring in situ degradability was determined using near-infrared spectroscopy (NIRS) (NIR System Model 6500 spectrophotometer, FOSS, Hamburg). The samples were scanned in a range from 1100 to 2500 nm. Reflectance data were collected as 1/R. All samples were measured twice and averaged. The calibration set comprised 185 samples of maize plants (stover). NDF ranged from 56.4% to 88.6% DM (mean 67.3% DM). The validation set included 47 samples. NDF
Journal of Animal Physiology and Animal Nutrition © 2014 Blackwell Verlag GmbH
983
Sample preparation and analysis
Effect of genotype on feeding value of maize stover
ranged from 66.4% to 89.5% DM (mean 68.4% DM). Calibration statistics included standard error of calibration (SEC = 0.881), coefficient of determination in calibration (R2 = 0.985), standard error of cross-validation (SECV = 1.049) and coefficient of determination in cross-validation R2CV = 0.978). The slope of the calibration curve was 0.933.
F. M. E. Zeller, B. L. Edmunds and F. J. Schwarz
Only four varieties from each maturity group were analysed for NDF degradation. Selected from MG I were DK 233, DKc 2949 and DK 247 from harvest years 2004 and 2005 and variety EC 2803 from harvest year 2005 (n = 28). Selected from MG II were varieties DK 281, Monumental and DK 287 from harvest years 2004 and 2005 and variety DKc2960 from harvest year 2005 (n = 28).
In situ degradability
For determination of in situ ruminal degradability, six non-lactating dairy cows (Holstein Friesian) fitted with ruminal cannulae were used. Cows were fed twice daily (07:00 and 16:00) with a total amount of (DM basis) 3.0 kg maize silage, 2.72 kg hay, 0.4 kg soy bean meal and 100 g mineral mixture. Approximately 4.0 g of freeze-dried stover, ground to pass a 3-mm screen, was enclosed in pre-weighed nylon bags (10 9 20 cm, average pore size 53 10 lm; Bar Diamond Inc., Parma) and inserted into the rumen. Each sample was incubated in duplicate in three animals (n = 6) for 2, 4, 8, 16, 24, 48, 72 and 96 h. Following removal, bags were submerged in iced water to stop microbial action and then rinsed continuously until the water ran clear. Bags were then further rinsed by machine (Fa. QUELLE; WVA Basic 74) on cold cycle with two water changes for 20 min (without soap or spinning) and then subsequently freeze-dried and weighed to determine DM residue. Washout fraction (0 h) was measured by washing and re-weighing unincubated bags (n = 3) using the same machine washing procedure. The NDF content of residues after 0, 4, 24, 48 and 96 h of incubation was estimated using NIRS. Thus, it was possible to generate ruminal degradation curves for DM and NDF and calculate the potential ruminal degradability. Effective ruminal degradability of DM (EDDM) and NDF (EDNDF) was estimated using the formula of McDonald (1981), modified according to S€ udekum (2005): kt0
ED ¼ a þ ½ðb cÞ=ðc þ kÞ e
where ED is the effective ruminal degradability, a represents the soluble and rapidly degradable fraction, b represents the potentially degradable fraction, c is the degradation rate of b, k is the rate constant measuring the passage of solid particle flow from the rumen to the abomasum and t0 is the lag time during which no microbial degradation occurs. The NLIN procedure of SAS (SAS, 1990) was used to estimate parameters in the model for DM and NDF degradation data. 984
In vitro organic matter digestibility
The digestibility of organic matter (OM) in vitro was measured following the technique described by De Boever et al. (1986). Briefly, 0.3 g of sample was incubated for 24 h in pepsin–HCl solution at 40 °C, then 45 min at 80 °C. After filtration, the contents were rinsed with warm water. A cellulase–buffer mixture (cellulase Onozuka R-10 extracted from Trichoderma viride) was added and the sample returned to the incubator for a further 24 h at 40 °C. After filtration and washing, the residue was dried and weighed. It was then ignited in a muffle furnace, the annealing loss equating to the cellulaseindigestible OM of the sample. Subtraction of the
Table 1 Dry matter content (g/kg) and chemical composition (g/kg DM) of the maturity groups at different stages of maturity Harvest date 1 Dry matter MG I 202.7 MG II 195.8 Ash MG I 57.7 MG II 55.5 Crude protein MG I 63.0a MG II 54.9b Crude fibre MG I 320.9b MG II 347.2a Neutral detergent fibre MG I 542.7b MG II 579.6a Acid detergent fibre MG I 337.6b MG II 368.4a Acid detergent lignin MG I 48.4 MG II 52.9
2
3
4
SEM*
206.9 203.3
229.9 235.1
287.1 274.0
38.1 37.5
58.1 56.8
59.7 58.0
58.3 56.4
7.17 6.83
56.3 52.8
52.2 48.9
47.4 48.6
9.60 7.96
337.4b 354.6a
345.6b 367.7a
351.8b 382.5a
23.4 26.1
570.8 591.9
594.1b 623.6a
619.0 644.4
40.9 45.0
356.0b 375.6a
369.1b 396.0a
385.3b 413.6a
28.0 29.1
51.1 54.1
52.1 56.3
54.7b 59.7a
7.69 6.32
Means from the same harvest date with different superscripts differ between the maturity groups (a,bP 0.05). Crude protein was higher (p < 0.05) for MG I at HD 1. Fibre fractions (NDF, ADF) were lower in MG I at all HDs, although not always significantly. Fibre fractions increased in both MGs during maturation. The group means (g/kg DM) over all HDs for the fibre fractions for MG I and II, respectively, were as follows: NDF 582 vs. 610 (p < 0.01), ADF 362 vs. 388 (p < 0.001). Differences in ADL contents between genotypes were found only in a very late stage of maturity (HD 4; Table 1), although ADL content for MG I tended to be lower at all HDs. As an overall mean, MG I had a lower ADL (g/kg DM) content than MG II (52 vs. 58; p < 0.001). Throughout maturation, ADL increased slightly but this was only significant in MG II (p < 0.05). Dry matter and neutral detergent fibre degradation of residual plant matter in situ
F. M. E. Zeller, B. L. Edmunds and F. J. Schwarz
Table 2 Dry matter degradation parameters in situ of the two maturity groups at the different harvest dates Harvest date 1 a [g/kg DM] MG I 339.3a MG II 301.6b b [g/kg DM] MG I 403.5 MG II 414.2 d [g/kg DM] MG I 742.9a MG II 715.8b c [%/h] MG I 4.83 MG II 4.85 Lag time [h] MG I 1.71 MG II 1.56 EDDM6 (g/kg DM)* MG I 500.4a MG II 467.6b
2
3
4
SEM*
302.5 297.0
286.5 273.2
267.7 254.6
46.5 44.6
420.8 406.1
438.0 425.2
444.8 429.9
43.5 35.4
723.3 703.0
724.5a 698.3b
712.5a 684.5b
29.8 37.7
4.52 5.01
4.22 4.36
3.96 4.31
0.88 0.87
1.44 1.76
2.02 2.08
2.74 2.60
1.46 1.28
467.7 451.6
446.7 430.3
417.7 408.4
47.3 42.6
Means from the same harvest date with different superscripts differ between the maturity groups (a,bP
