SEARCH

SEARCH BY CITATION

Keywords:

  • silage;
  • rumen;
  • crude protein degradation;
  • dry matter;
  • utilizable crude protein;
  • crude protein fractions;
  • amino acids

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results and discussion
  6. General discussion
  7. Conclusions
  8. Acknowledgments
  9. References

Wilting grass prior to ensiling generally increases the dry matter (DM) intake but the effect of wilting on animal performance is still poorly understood. There is a need to improve understanding of the effects of wilting on the nutritional components and chemical composition of grass silage. This study focused on the effects of the extent and rate of wilting on N components of grass silage. Meadow grass was wilted to four DM contents (200, 350, 500, 650 g kg−1) at two different rates (fast, slow), creating a total of eight silages. Crude protein (CP) fractions were measured using the Cornell Net Carbohydrate and Protein System. Utilizable CP at the duodenum (uCP), a measure of feed protein value, was estimated using the modified Hohenheim gas test. Ruminally insoluble, undegraded feed CP (RUP) was measured using an in situ technique. Amino acid (AA) composition prior to and after rumen incubation was also investigated. Utilizable CP at the duodenum, RUP and true protein fractions B2 and B3 were increased by rapid wilting and high DM content (DM > 500 g kg−1), although the increase with DM was only mild for uCP, probably due to lower ME content in the DM-650 silages. Non-protein-N decreased with increasing DM and rapid wilting. The higher RUP content from both DM-650 silages leads to a higher total AA content after rumen incubation. Treatment also influenced the AA composition of the ensiled material, but the AA composition after rumen incubation was similar across treatments. Rapid and extensive wilting (DM > 500 g kg−1) improved protein value and reduced CP degradability. Increased uCP may result in higher milk protein yield, while reduced degradability may reduce N lost from urinary excretion. The primary effect of wilting on post-ruminal AA supply from RUP appeared to be quantitative, rather than qualitative.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results and discussion
  6. General discussion
  7. Conclusions
  8. Acknowledgments
  9. References

Wilting grass prior to ensiling is common practice in many countries. The main reasons for wilting are to improve fermentation quality (Marsh, 1979) and to reduce environmental pollution and nutrient loss through effluent. Wilting generally also increases the dry matter (DM) intake (DMI: Dawson et al., 1999; Wright et al., 2000; Huhtanen et al., 2007) although this does not always translate into improved animal performance and, despite a large number of studies comparing wilted and unwilted silage, the effect of wilting on animal performance is still poorly understood. Generally, responses in terms of DMI and performance of both dairy and beef cattle to wilting are mainly a reflexion of the fermentation quality of unwilted silage (Wilkins, 1984), particularly ammonia N concentration (Wright et al., 2000), relative to its wilted counterpart. In a review of the literature, Wright et al. (2000) also established that responses in DMI and animal performance are related to the extent and rate of moisture loss in the field. Considering the benefits of wilting in terms of increased DMI and difficulty in predicting production responses, as well as the need for improvement in efficiency of feed utilization and reduction in environmental emissions, it is important to gain a better understanding of the effects of wilting on the nutritional components and chemical composition of grass silage. This study has focused on the effect of extent and rate of wilting on N components.

Meadow grass was wilted to four concentrations of DM (200, 350, 500 and 650 g kg−1 fresh matter) at two rates of moisture loss (fast and slow). An intensive examination of N components was performed including analysis of the following: utilizable crude protein at the duodenum (uCP) via the modified Hohenheim gas test (modHGT; Steingass et al., 2001), ruminally insoluble, undegraded crude protein (RUP) via an in situ technique, crude protein (CP) fractions via the Cornell Net Carbohydrate and Protein System (CNCPS: Sniffen et al., 1992) and amino acid (AA) composition both prior to and after rumen incubation. While animal production trials are the ultimate test of feeding value, this initial study focused on chemical composition and analysis of non-ammonia-N (NAN) supply and CP degradation using in vitro and in situ techniques respectively.

Materials and methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results and discussion
  6. General discussion
  7. Conclusions
  8. Acknowledgments
  9. References

Preparation of silages

Eight silages were derived from the same parent material (meadow grass: approximately 0·85 perennial ryegrass, 0·08 legumes and 0·07 herbs, second harvest, heading) from the 2008 harvest in Aulendorf, Baden-Württemberg, Germany. The grass was subdivided and either wilted thinly spread on black plastic in the sun (fast; F) or on white plastic in the shade (slow; S) to a DM content of approximately 200, 350, 500 and 650 g kg−1, thus providing material for eight silages. Weather conditions during the wilting period were sunny and hot, with maximum temperatures around 30°C over the 2 d required to achieve all treatment DM targets. Wilting times of each treatment are listed in Table 1. Upon reaching the desired DM, the grass was chopped at a 20-mm setting and ensiled in triplicate, without additives, in 1·75-l glass jars according to the scheme for silage testing in Germany (Bundesarbeitskreis Futterkonservierung, 2006). The average amount of DM ensiled per glass for each treatment is given in Table 1. Ninety-day fermentation was allowed in a temperature-controlled storage room at 25°C. The treatments are referred to as: F-200, S-200, F-350, S-350, F-500, S-500, F-650 and S-650.

Table 1. Wilting time (h), amount of dry matter (DM; g) ensiled per 1·5-l glass, DM content (g kg−1) at ensiling and proximate variables of silages wilted to various DM contents at two rates (R) of moisture loss (F = fast, S = slow)
   SilageUnensiled grass
TreatmentWilting timeDM ensiledDMCPNDFomADFomMEBuffering capacitySugar
  1. NS, not significant; CP, NDFom, ADFom (g kg−1 DM) and ME (MJ kg−1 DM) are respectively crude protein, neutral and acid detergent fibre (exclusive of residual ash) and metabolizable energy. Buffering capacity (g lactic acid kg−1 DM) and sugar content (g kg−1 DM) of the unensiled material are also included. *< 0·05, **< 0·01, ***< 0·001. †ME = 7·81 + 0·07559 GP − 0·00384 ash + 0·00565 CP + 0·01898 fat − 0·00831. ADFom (GfE, 2008). All variables are expressed as g kg−1 DM except GP, which is gas production at 24 h (mL 200 mg−1 DM).

F-200317919418841724411·05394
S-200517819318940622711·25396
F-350732538118943424711·248104
S-3503130837319144425410·85184
F-500939049918645025610·847114
S-5003336846619543624910·84693
F-6502647469217948427510·146117
S-6505048666919147227510·045118
DM   NS * * * ** ***
R   NSNSNSNSNS ***
DM × R   NSNS ***

General analysis

The silages were pooled, freeze-dried and milled through a 3-mm screen for the in situ trial and a 1-mm screen for all other analyses. Proximate analysis was performed according to VDLUFA (2007), and method numbers are given. The DM of the forages and incubation residues were determined by oven-drying of a subsample at 105°C (3·1). Ash and fat were analysed using methods 8·1 and 5·1·1 respectively. CP was determined by Dumas combustion (4·1·2) for original silage material, in situ residues and for forage fed as part of the in situ diet. The Kjeldahl method (4·1·1) was used for CP fractionation using a Vapodest 50s carousel (Gerhardt, Königswinter, Germany) for automated distillation and titration. The same equipment was also used to measure ammonia, by distillation, after incubation in rumen fluid as part of the modHGT. Neutral and acid detergent fibre of the samples were estimated using NIRS and were expressed exclusive of residual ash (NDFom, ADFom). For forages fed as part of the in situ diet, NDF was analysed using an ANKOM220 fibre analyser and is expressed inclusive of residual ash. Metabolizable energy (ME) was calculated using the GfE (2008) equation for grass silage (see Table 1 for calculation). Silages were not corrected for DM losses associated with drying (Weissbach and Kuhla, 1995) as the correction is based on oven-dried, not freeze-dried material. Grass after harvest was also freeze-dried and analysed for buffering capacity (Weissbach, 1992) and sugar concentration as an estimate of water-soluble carbohydrates (Anonymous, 2009).

Modified Hohenheim gas test

Methods describing the modHGT (Steingass et al., 2001) are presented in detail in Edmunds et al. (2012a). Briefly, the method followed basic procedures of the original Hohenheim gas test (Menke and Steingass, 1988), the modifications being a 2 g l−1 increase in (NH4)HCO3 and a 2 g l−1 decrease in NaHCO3 in the buffer solution and measurement of ammonia at the end of each incubation. Rumen fluid was collected from two to three fistulated sheep receiving a 50:50 grass hay/pelleted compound maintenance ration twice daily. One-third of the ration was given at 07:00 and two-thirds at 15:30 h. Approximately 200-mg DM sample was incubated for 8 and 48 h in the rumen fluid/buffer solution over two runs (i.e. using two different batches of rumen fluid). Runs were used as statistical replicates. At the termination of the incubation, the entire contents of the syringe (30 mL) were analysed for ammonia N and uCP was calculated as follows:

  • display math

where NH3N is in mg 30 mL−1, ‘blank’ refers to rumen fluid/buffer solution without added substrate, ‘sample’ is the solution with added substrate, Nsample is N added to the syringe from the measured amount of substrate (mg) and weight is the amount of substrate weighed into the syringe and calculated to DM.

Biological between-run fluctuations were corrected using a protein standard (provided by the University of Hohenheim), which was analysed with every run. The correction follows the same method as that used for gas production (Menke and Steingass, 1988), with deviations higher than 10% from the reference mean of the standard requiring repetition of that run. Following correction of uCP, values from the two incubation time points were plotted against a log (ln(time)) scale and the resulting regression equation was used to calculate effective uCP to passage rates of 0·02, 0·04 and 0·06 h−1, which will hereafter be referred to as: uCP2, uCP4 and uCP6. The passage rates are assumed to represent whole content flow (i.e. passage of solid and liquid phases through the rumen) of animals at different levels of production.

In situ procedure

Methods describing the in situ trial are presented in detail in Edmunds et al. (2012b). Briefly, the procedure followed basic guidelines of Madsen and Hvelplund (1994) with incubation periods of 2, 4, 8, 16, 24, 48 and 96 h and using three non-lactating German Holstein cows, fitted with rumen cannula, to give three statistical replicates per feedstuff. Cows received a diet of proportionately and approximately (DM basis) 0·22 soybean meal and mineral concentrate (approximately 4:1 soybean meal/mineral mix), 0·53 maize silage (CP and NDF 75·5 and 461 g kg−1 DM respectively) and 0·26 grass hay (CP and NDF 137 and 586 g kg−1 DM respectively) at 07:00 and 16:00 h daily in two equal meals meeting ME maintenance requirements. The number of bags used as replicates changed with incubation time depending on expected degradability and amount of residue required for subsequent analysis. Replicate numbers were three, four and six bags for hours 2–8, 16–24 and 48–96 respectively. For each incubation time, bags were inserted directly before the morning feed and were immediately immersed in ice water upon removal. All bags underwent machine washing in cold water and were subsequently freeze-dried. Incubation residues were pooled per cow and incubation time. Three bags that had not undergone any incubation were also machine washed to calculate the washout fraction. Water-soluble material was estimated by mixing duplicate samples in 100 mL, 40°C distilled water and then filtering through No. 5951/2, diameter 270 mm filter paper (Schleicher and Schuell, Dassel, Germany). The equation of Hvelplund and Weisbjerg (2000) was used to correct CP disappearance for small-particle loss at each incubation time point. Correction for microbial attachment (MA: g kg−1 residue CP) to undegraded feed particles was carried out as described by Edmunds et al. (2012b) using the exponential equation of Krawielitzki et al. (2006). The Amax parameter of the equation, describing maximum MA at time t ≈ ∞, was estimated by boiling a subsample of the residue (t ≥ 16 h) in neutral detergent (ND) solution to extract microbes (Mass et al., 1999). Time-specific degradation of CP was calculated as effective degradability of CP (EDP: g kg−1 CP) according to McDonald (1981) but with the assumption that no degradation occurred during the lag phase (Wulf and Südekum, 2005). Finally, RUP was calculated as 1000-EDP and is presented at an assumed passage rate of 0·03 h−1 (Hristov et al., 2003), representing the passage of particulate matter through the rumen.

Crude protein fractionation

Division of CP into five fractions (A, B1, B2, B3 and C) based on characteristics of degradability was performed according to the CNCPS (Sniffen et al., 1992) using standardizations and recommendations of Licitra et al. (1996). All fractions, including CP, were analysed in triplicate. See Edmunds et al. (2012b) for a more detailed description.

Amino acid analysis

The original material (i.e. not incubated in the rumen) and 16-h in situ residues underwent a complete AA profile analysis, performed in the CARAT laboratory, Adisseo, Commentry, France. The 16-h time point was chosen on the basis that all rumen-soluble materials were assumed to have been solubilized and that sufficient material remained after rumen incubation for subsequent analysis. The AA contents were measured by cation-exchange chromatography after acid hydrolysis for 24 h (Anonymous, 1998; Directive 98/64/EC, 3/09/99 – Norme NF EN ISO 13903, Antony, France). Analysis of methionine was performed after initial oxidation of samples with performic acid. Phenylalanine was analysed without oxidation.

Correction for added AA coming from microbial colonization of in situ residues was performed using the following procedure. The amount of microbial matter was estimated for each residue as described in the section: ‘in situ procedure’. Next, the AA composition of the microbes had to be estimated. Microbial matter has been shown to be relatively consistent in its AA composition (Storm and Ørskov, 1983; Chamberlain et al., 1986; Hildebrand et al., 2011), and although this assumption has been debated (Clark et al., 1992), it was maintained in this study and values published by Storm and Ørskov (1983) were used to correct each sample. As not all N in microbial matter is AA-N, the calculated extent of microbial contamination was multiplied by a factor of 0·8 (Storm and Ørskov, 1983). The resulting number was then multiplied by the published value for each individual AA and subtracted from the measured concentration of the residue AA as in the following calculation:

  • display math

where AAi is the measured concentration of the ith AA from the residue (g kg−1 DM), MA is the amount of estimated microbial CP of the residue (g kg−1 residue CP) as calculated from the equation of Krawielitzki et al. (2006), MAAi is the concentration of the ith AA in microbial matter (g kg−1 AA: Storm and Ørskov, 1983) and CP is the concentration of residue CP (g kg−1 DM). Summation of individual, corrected AA provided the corrected total AA content.

Statistics

All statistical analyses were performed using the GLM procedure of SAS version 9·1 (SAS, 2002) using least squares means and the following model:

  • display math

where DM is DM category i (200–650) and R is the rate of moisture loss category j (fast, slow). Differences were deemed significant at P < 0·05. Due to insufficient material from the S-200 treatment for analysis by CP fractionation and modHGT, statistics analysing results from these procedures include treatments with DM 350–650 only.

Results and discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results and discussion
  6. General discussion
  7. Conclusions
  8. Acknowledgments
  9. References

Silage chemical composition

Results of the proximate analysis and exact DM at ensiling are presented in Table 1. CP did not change between treatments (P > 0·05), NDFom and ADFom increased (P < 0·05) with increasing DM by approximately 70 and 45 g respectively, and ME, which was estimated from changes in chemical composition and gas production, decreased by approximately 1 MJ kg−1 DM in the DM-650 silages (P < 0·05). Buffering capacity was low, at less than 100-g lactic acid per kg DM, and sugar level at ensiling was above 80 g kg−1 DM, indicating good potential for rapid lowering of the pH (Table 1). Sugar content after wilting was strongly affected by DM and wilting speed. More sugar remained due to rapid (all DM levels) and extensive (DM > 500 kg−1) wilting. An interaction was also present (P < 0·001) whereby sugar decreased between DM-200 and DM-500 in the fast treatment, but did not change in the slow treatment. Upon opening the silages, no visible evidence of moulding or warming was present and the aroma was typical of silage fermented primarily by lactic acid bacteria.

Utilizable crude protein

Utilizable CP (uCP) is defined as the sum of microbial CP and RUP at the duodenum (i.e. NAN supply to the animal) and is used in Germany as a measure of protein value (GfE, 2001). In this study, uCP was estimated in vitro using the modHGT. Due to insufficient material, only DM-350 to DM-650 were analysed. Initially, some results went against expectations, e.g. F-650 was lower than S-650 at uCP2 (effective uCP using the passage rate 0·02), which was due to the differences in CP concentration (Table 1). Although statistically the same, CP varied more than would be expected for forage coming from the same parent material and reanalysis reduced the between-treatment standard deviation. These discrepancies in CP content made it difficult to obtain a clear picture of the effects of treatment on uCP, and because CP is calculated from total N and total N should not change under such controlled experimental conditions, a singular CP concentration of 186 g kg−1 DM (the average CP concentration after reanalysis) was assigned to all samples and uCP was recalculated. The results are presented in Table 2.

Table 2. Ammonia N (mg L−1) after 8- and 48-h incubation in vitro and mean ± s.d. utilizable crude protein (uCP; g kg−1 DM) of grass silage wilted to different DM contents at two rates (R) of moisture loss (fast, slow) and calculated to three assumed rates of passage (0·02, 0·04 and 0·06 h−1) using a fixed CP value of 186 g kg−1 DM
 NH3-NuCP
8480·020·040·06
  1. NS, not significant.

  2. *< 0·05, **< 0·01, ***< 0·001.

DM
 35021934598 ± 5·6128 ± 7·6146 ± 11·0
 50021734599 ± 8·5130 ± 9·1148 ± 10·4
 650205339100 ± 3·0136 ± 6·6156 ± 9·9
R
 Fast203340102 ± 3·9138 ± 3·0159 ± 5·4
 Slow22434696 ± 5·3124 ± 4·3141 ± 5·7
P
 DM  NS *** **
R  NS *** ***
 DM × R  NSNSNS

Both wilting extent and speed affected uCP content, although there was no interaction. Due to faster wilting, uCP was significantly higher at passage rates 0·04 and 0·06 (< 0·05) and tended to be higher at uCP2 (< 0·075). The trend existed at all levels of DM (350–650). The lower uCP from the slow-wilted silages was entirely due to a higher ammonia concentration at the end of the in vitro incubation (Table 2). A probable explanation for this effect is a better preservation of substrates, e.g. water-soluble carbohydrates, required for microbial protein synthesis, from fast wilting and/or a higher proportion of non-protein-N (NPN) from the slow-wilted silages (Figure 1b). An effect of DM was also present (uCP4 and 6 only), whereby uCP increased at DM-650. There was no difference between DM-350 and 500.

image

Figure 1. Effect of dry matter (DM g kg−1; x-axis) at ensiling and wilting speed (▲ fast, ■ slow) on: (a) Ruminally insoluble, undegraded feed protein (RUP) calculated to a passage rate of 0·03 h−1. (b) Non-protein N (NPN, A fraction). (c) Intermediately degraded true protein (B2 fraction). (d) Slowly degraded true protein (B3 fraction).

Download figure to PowerPoint

In vivo trials generally agree that wilting increases NAN flow at the duodenum (Narasimhalu et al., 1989; Teller et al., 1992) and that this is accentuated by extensive wilting (Merchen and Satter, 1983; Lebzien and Gädeken, 1996). The latter two studies also observed significant increases in duodenal NAN at DM ≥ 600 g kg−1 and that this was mainly due to an increased proportion of RUP. In contrast, Verbič et al. (1999) revealed a larger amount of microbial N reaching the duodenum from sheep fed highly wilted silage (DM 520 g kg−1) and hay than from unwilted (210 g kg−1) and moderately wilted (430 g kg−1) silage. However, they used the in situ technique combined with urinary purine-derivative excretion to estimate NAN flow and metabolizable protein, rather than in vivo procedures.

In this study, a large part of these results can surely be explained by the previously described effect of wilting on sugar content at ensiling. Additionally, higher uCP in the DM-650 silages may also be due to restrictive fermentation. That a more pronounced difference was not observed in the DM-650 silages is probably due to the 1 MJ kg−1 reduction in ME. The lack of difference in uCP between DM-350 and DM-500, which are within the recommended range for ensiling in Germany (Thaysen, 2004), implies that the emphasis may best be placed on rapid wilting to increase protein supply to the animal from grass silage. Results of production trials do suggest that extent of wilting is positively related to animal performance; however, this is mainly due to increased ME intake (Wright et al., 2000) and results are not unanimous between studies. (This will be discussed in more detail in the section headed ‘General discussion’.)

Rumen undegraded crude protein

The RUP concentrations of the DM-200 to DM-500 silages were below the tabulated range for grass silage in Germany (150–200 g kg−1 CP: Universität Hohenheim – Dokumentationsstelle, 1997; Figure 1a), although both DM-650 silages were slightly above 150 g kg−1 CP. Quantitative comparisons of RUP are difficult to make with published data as so many factors affect degradability and its estimation. Such factors include CP composition of the plant material itself, variations in methodological techniques and errors associated with the in situ analysis. Advances in knowledge suggest that RUP is generally overestimated from the in situ technique, and this is not only due to the failure to acknowledge soluble NAN escaping rumen degradation but also failure to correct for CP added by microbial attachment to in situ residues. This study corrected for the latter source of error. Due to the aforementioned problems, analysis of the results in this study focused on trends rather than values.

With increasing DM, RUP increased quadratically (< 0·001). Figure 1a clearly shows that there was no effect of DM on RUP up to and including DM-500. From DM-500 to DM-650, RUP increased (< 0·001) by 46 and 61 g for slow- and fast-wilted treatments respectively. Rapid wilting also increased RUP (< 0·001) and there was a weak interaction with DM (< 0·025), whereby the difference between wilting rate was enhanced at DM-650. The increase in RUP with DM supports findings from Van Vuuren et al. (1990) and Lebzien and Gädeken (1996). However, Van Vuuren et al. (1990), who analysed silages with DM contents of 220, 300 and 450 g kg−1 in situ, observed RUP increases between all DM levels. Lebzien and Gädeken (1996) also observed a curvilinear increase in RUP (in vivo) with increasing DM in grass silage; the sharp decline in degradability occurring between DM 390 and 600 g kg−1. The threshold DM at which the sharp decline in degradability occurs is not clear, and although this study suggests this is at DM > 500 g kg−1, it is most probably influenced by multiple factors.

Although still frequently analysed, RUP does not appear to improve the prediction of milk protein yield (Huhtanen and Hristov, 2009). This may be due to error in the measurement of RUP via the in situ technique as well as failure to account for soluble NAN escaping rumen degradation. Measurement of degradability, however, may still be useful in reducing N loss through excretion. Although N intake is the main factor affecting N-use efficiency and excretion, lower degradability has been clearly shown to reduce N lost in the urine (Castillo et al., 2001; Kebreab et al., 2001). This was further demonstrated by Nguyen et al. (2005) who found a positive correlation between urinary N and N degradability of orchard grass (Dactylis glomerata) silage wilted to low (240 g kg−1), medium (350 g kg−1) and high (600 g kg−1) DM content.

Crude protein fractions

To understand more clearly the effect of wilting on protein composition, CNCPS CP fractionation was performed. Insufficient material remained from S-200; however, trends between fast and slow were similar between the higher DM treatments for the A, B2 and B3 fractions. Therefore, if it is assumed that the same pattern would have occurred with S-200 as was observed with F-200, a rough idea can be obtained through results of F-200.

Non-protein nitrogen

Fraction A, which represents NPN, decreased quadratically with increasing DM (< 0·001; Figure 1b). As is characteristic of silages, NPN was high (approximately 600 g kg−1 CP) in F-200 and both DM-350 silages. This was also true for the DM-500 treatments. In both DM-650 silages, NPN was largely reduced (< 0·001), indicating decreased proteolysis. Rapid wilting also reduced NPN (< 0·001) implying reduced proteolysis during wilting due to a shorter wilting time. A strong interaction between DM and wilting speed was also present (< 0·001), whereby the difference between wilting rate was enhanced at DM-650. One might expect a higher increase in NPN with wilting time due to the longer exposure to proteolytic plant enzymes. For example, although there was a difference in wilting time of 24 h between F-500 and S-500, NPN was only higher by 66 g kg−1 CP in the slow treatment. The moderate difference between fast and slow treatments supports the knowledge that most proteolysis occurs in-silo. It may also be, for the fast-wilted treatments, that the decreasing rate of proteolysis with advancing moisture loss was slightly offset by an initial increase in rate caused by higher temperature (Muck, 1988), a result of wilting in direct sunlight.

What happens to NPN once ingested is of great importance. This fraction makes up the larger part of silage-soluble N (fractions A+B1) and is composed of peptides, AA and ammonia. A portion of soluble NAN may escape rumen degradation and be available for absorption in the small intestine (Choi et al., 2002; Volden et al., 2002). These authors estimated silage-soluble NAN escaping degradation to be approximately 10%; thus, differences in total feed CP (soluble and insoluble) reaching the duodenum may be not be as pronounced as the previously described RUP data suggest. According to a meta-analysis by Huhtanen et al. (2008), the main component affecting milk protein yield and milk urea-N is the ammonia-N content of the silage and not soluble N itself. However, while production might not be directly affected, a large portion of soluble NAN may be converted to ammonia and, after urea has been synthesized in the liver, excreted in the urine. Indeed, Nguyen et al. (2005) found a positive correlation between fraction A + B1 and urinary N excretion in sheep. In a review by Givens and Rulquin (2004), it was concluded that N utilization from grass silage is poor, probably due to the high portion of soluble N and relatively low rapidly fermentable carbohydrates, and that a greater understanding of carbohydrate/protein relationship was required.

True protein fractions

Treatment had no effect on the B1 (soluble TP) fraction, and the concentration was low (39 ± 12 g kg−1 CP) in all treatments. Fresh, unensiled forage usually contains B1 at above 100 g kg−1 CP (Kirchhof et al., 2010; Edmunds et al., 2012b), and the low B1 values of the silages in this study indicate its susceptibility to proteolysis. Likewise, the B2 fraction of the silages was reduced compared to what could be expected from unconserved grass; a study comparing conservation methods and their effects on CP fractions revealed an average B2 fraction of 520 ± 75 g kg−1 CP in fresh forage (n = 12; Edmunds et al., 2012b), whereas silages in this study contained an average of 277 ± 63 g kg−1. The concentration of B2 followed a quadratic trend with the highest concentration remaining at DM-650 (< 0·001: Figure 1c), which is in contrast to Nguyen et al. (2005) who found no difference. Fast wilting retained more B2 than slow wilting (< 0·001), and this trend occurred at all levels of DM. No interaction between DM and wilting speed was present.

The same trend as for RUP and B2 was observed for fraction B3; i.e. increasing DM concentration and fast wilting provided higher B3 values (< 0·001: Figure 1d), with the main increase occurring at DM-650. A strong interaction (< 0·001) also occurred whereby rapid wilting appeared to have a greater influence on B3 with increasing DM. The sharp rise in B3 between DM-500 and DM-650 resulted in B3 concentrations above those found in fresh meadow grass of a similar composition and maturity (Edmunds et al., 2012b). This is in agreement with Nguyen et al. (2005) who observed an increase in B3 of 65 g kg−1 CP above its fresh counterpart in a D. glomerata silage with a DM content of 600 g kg−1. The reason for the increase in the B3 fraction at low moisture contents is not yet clear. A possible explanation could be a combination of decreased in-silo proteolysis and denaturation of proteins caused by heating: a result of respiration using trapped air. Nguyen et al. (2005) suggested that sunlight exposure during wilting alters the properties of proteins and forms bonds between proteins and carbohydrates. However, the B3 fraction also increases through conserving forage as hay or through rapid, artificial drying at high temperatures. Artificially dried forage often has limited sunlight exposure; therefore, the increase in B3 may simply be a case of decreased solubility caused by precipitation of proteins during drying. It may also be due to the increased NDFom concentration. Fraction B3 is, by definition, ND-insoluble N so it was not surprising that a strong relationship was found between B3 and NDFom (n = 7, R= 0·88).

There was a strong relationship between B3 and RUP (n = 7, R2 = 0·85, < 0·001: Figure 2). Generally, RUP is composed mainly of B3 and C fractions, with some remaining B2. In this study, regression of B3 + C against RUP revealed a strong linear relationship. There were no differences between the treatments in the C fraction; thus, the linearity was entirely due to B3. Only F-200 deviated from the trend line and removal of this sample increased the R2 to 0·99. This strong trend is more likely to occur in conserved forages due to the depleted B2 fraction. The same regression using data from unconserved forages revealed no relationship (n = 12; Edmunds, B., Institute of Animal Science, University of Bonn, Germany; unpublished results). If applicable to a wider range of silages, this information could serve as a rapid and useful estimation of in situ RUP.

image

Figure 2. Relationship between ruminally insoluble, undegraded feed crude protein (RUP) and slowly degraded true protein (fraction B3).

Download figure to PowerPoint

The C fraction remained stable (26 ± 3 g kg−1 CP) across all treatments. This is in contrast to the results from Merchen and Satter (1983) who observed significant increases in the C fraction in silage with a DM content of 660 g kg−1. Nguyen et al. (2005) also observed a slight increase in the C fraction in silages containing 600 g DM in comparison with moister silages (450 and 240 g DM). The stability of the C fraction in this study probably indicates that heat accumulation during ensiling was not enough to initiate the reaction causing indigestible Maillard products, which is attributable to good compaction and air-tight sealing and is easy to achieve in experimental glass silos. The results may not be reflected in practice, which is suggested by the results from Merchen and Satter (1983) who used tower silos.

Amino acid composition

Effects of ensiling

Clear treatment effects on the AA composition of the silages were observed. However, with the knowledge that AA composition of most feeds changes during rumen incubation (Erasmus et al., 1994; Van Straalen et al., 1997; Von Keyserlingk et al., 1998; González et al., 2001), these ensiling-related changes will only be discussed briefly. The mean AA content of sample CP was 695 ± 31 g kg−1. There was no difference between treatments for total AA content (mean = 131 ± 4 g kg−1 DM; > 0·05). Generally, for most AA, concentrations were lower in slow-wilted silages and the lack of difference in total AA can be explained by the higher level of proline, which supports the findings of Kemble and Macpherson (1954) and is a typical sign of water stress (Boggess and Stewart, 1976). Treatment-related changes for individual AA are shown in Figure 3a. Only DM-350 and DM-650 have been presented for reasons of simplicity. The changes were calculated (using values expressed as g per 100 g total AA) as per cent change from a fresh, unensiled meadow grass sample (Fr-0; second harvest, 2008). The Fr-0 sample was chosen as a representative, based on the closeness of its AA profile to published values (Degussa Feed Additives, 1996; Misciattelli et al., 2002), and was preferred to published values based on geographical location and harvest, and because it was analysed using the same procedure as the silages. Worthy of remark is the higher retention of arginine both with increasing DM and fast wilting, and the overall decrease in cysteine and methionine.

image

Figure 3. Change (%) in amino acid (AA) composition of silages wilted to various contents of dry matter at two rates (F = fast, S = slow). (a) As compared to a representative fresh grass sample. (b) From 0 to 16 h rumen exposure.

Download figure to PowerPoint

Effects of rumen exposure

While the previously described changes are interesting, they may be of little use when trying to establish supply of AA to the duodenum of ruminants. González et al. (2009), who also reported changes in the AA profile after ensiling, observed that these effects were lost during rumen incubation. Other studies also support this observation (Van Straalen et al., 1997; Von Keyserlingk et al., 1998). Therefore, it is of interest to know how wilting speed and DM at ensiling affect the AA profile after rumen exposure. Residues following 16-h rumen incubation were analysed for AA content. All values were corrected for AA originating from microbial attachment to in situ residues using the previously described method. The averaged concentrations (g kg−1 DM) of total AA of the original silage sample (GS-0) and after 16-h rumen exposure (GS-16) were 130·7 and 41·1 respectively. The DM-650 silages held a higher total AA content (< 0·05) than all other levels of DM by approximately 44% (61·1 and 58·3 g kg−1 DM for F-650 and S-650 respectively), which can be directly related to the higher level of RUP.

Changes to the AA profile from GS-0 to GS-16 for treatments DM-350 and DM-650 are highlighted in Figure 3b. As in Figure 3a, data are presented as per cent change, which is calculated from data expressed as g per 100 g total AA. There appeared to be a mirror effect for many of the changes in AA observed after ensiling (Figure 3a); from the effect of ensiling, arginine, cysteine and methionine decreased and proline increased, while the opposite is true after rumen exposure. Generally, all major effects from ensiling were lost in the rumen.

To obtain a clearer picture of the combined effects of ensiling and rumen incubation on the AA composition of RUP, a net change in AA was calculated. Net change is simply the per cent change in AA directly calculated from Fr-0 to GS-16. From this calculation, it was immediately apparent that most treatment effects had been lost (data not shown). The Fr-0 sample had also been incubated in the rumen as part of a larger study using the same materials and methods. The changes from the Fr-0 to Fr-16 (AA composition of unensiled sample after 16-h rumen exposure) and Fr-0 to GS-16 (net changes after ensiling and rumen incubation) are included in Figure 4, where the silages have been presented as a mean (for improved clarity). The results are remarkably similar. For all individual AA, the per cent change from Fr-0 followed the same trend for both Fr-16 and GS-16. González et al. (2009) also compared AA composition of fresh grass (Italian ryegrass) and its silage before and after rumen exposure and found that although changes in the AA profile after ensiling were observed, these effects were lost during rumen incubation. They concluded that the AA profile of RUP of both forages was similar. The same observation was made in this study.

image

Figure 4. Change (%) in amino acid (AA) after 16-h rumen exposure for fresh grass (Fr-16, n = 1) and silage (GS-16, n = 6). The AA composition of the fresh grass sample at 0-h rumen exposure (Fr-0) was used to calculate net changes in silage (i.e. the total effect of ensiling and rumen exposure on the AA profile).

Download figure to PowerPoint

General discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results and discussion
  6. General discussion
  7. Conclusions
  8. Acknowledgments
  9. References

Improving understanding of the effect of wilting on chemical composition and feeding value of silage is important to further understanding of DMI, animal production and N excretion from grass silage-based diets. Animal production is closely related to feed intake, and it has been generally well accepted that wilting increases DMI. Response in DMI has been shown to be positively and linearly associated with both the extent and rate of field wilting (Wright et al., 2000) as well as to fermentation characteristics, D-value (digestible organic matter in silage DM), botanical composition, harvest characteristics and NDF (Huhtanen et al., 2007) and to the ration with regard to the proportion of grass silage and type of concentrate.

Despite a general increase in DMI due to wilting, production responses in terms of milk and protein yield are unpredictable. Teller et al. (1992), Patterson et al. (1998) and Romney et al. (2000) all observed increases in milk and protein yield due to wilting. Purwin et al. (2009) did not observe any difference in milk and protein yield until after 200 d in milk, which they explained was probably due to reduced concentrate supply. Gordon et al. (2000) and Kokkonen et al. (2000) observed a decrease in milk yield and no change in milk protein yield due to wilting. Conflicting results in production responses due to wilting, despite increased intake, are an indication of the complexity of the interacting effects between metabolism processes in the ruminant, silage quality and whole-ration composition. Results from studies may be confounded by: (i) a high level of concentrate in the ration, as suggested by Ferris et al. (2001) who demonstrated that silage quality considerably influenced milk yield in high-producing dairy cows only when the concentrate content of the diet was low; (ii) varying feed components (e.g. concentrate type, inclusion of maize or legume silage) within the ration; (iii) spreading the crop during wilting, as opposed to wilting in swaths, thus increasing the rate of moisture loss (Wilkinson et al., 1999); (iv) partitioning of energy gained from increased DMI into live weight gain rather than milk (Gordon et al., 2000; Kokkonen et al., 2000); (v) the greater influence of dirt contamination in wetter silages, leading to higher butyric acid and ammonia production (McDonald et al., 1991); (vi) silage management and type of silo, e.g. bunker silage is prone to warming after opening, particularly if grass is high in DM (Spiekers et al., 2009); and (vii) other factors affecting nutritive value and fermentation characteristics, DMI and milk production. The uCP results of this study suggest that rapid wilting has the potential to increase milk protein yield. Highly wilted silages may have a similar effect, although this will depend largely on factors previously described.

Regarding N emissions, Purwin et al. (2009) demonstrated that increasing DM from 250 to 375 g kg−1 decreased milk urea in mid- to late-lactating cows. Kokkonen et al. (2000) also observed lower milk urea in primi- and multiparous cows fed wilted silages. As discussed earlier, wilting decreases CP degradability and decreasing degradability is associated with reduced urinary N excretion (Castillo et al., 2001; Kebreab et al., 2001). With the increased focus on precision-farming practices and the call for reduced N emissions, the information presented in this paper suggests that rapid and/or extensive (DM > 500 g kg−1) wilting may be beneficial in reducing both urinary N excretion and the amount of CP included in the ration. Animal trials testing these hypotheses will be beneficial.

Conclusions

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results and discussion
  6. General discussion
  7. Conclusions
  8. Acknowledgments
  9. References
  1. Utilizable CP at the duodenum, in situ determined RUP and true protein content were positively influenced by rapid wilting and high DM (650 g kg−1).
  2. The measurement of uCP may be a useful tool in predicting animal performance as, unlike the estimation of RUP using the in situ technique, it considers the interaction of energy and protein metabolism and measures the end product of fermentation to provide a direct measurement of NAN after exposure to rumen fluid.
  3. Ensiling altered the AA profile, although the net effect of ensiling and rumen exposure on the AA profile was not greatly influenced by treatment and the total changes were similar to that of a fresh, unensiled grass of similar botanical composition. This implies that the AA composition of silage RUP is determined by rumen exposure and not ensiling.

Acknowledgments

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results and discussion
  6. General discussion
  7. Conclusions
  8. Acknowledgments
  9. References

This study was financially supported by the Bavarian Ministry for Agriculture, the Institute of Animal Science, University of Bonn and Adisseo, Antony, France. Thanks go out to Dr. M. Schuster and laboratory staff at LfL-Poing for the proximate analysis, E. Devillard, M. Gobert and technical staff at Adisseo for AA analysis and staff from LAZBW, Aulendorf, Germany, who assisted in the preparation of the silages.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results and discussion
  6. General discussion
  7. Conclusions
  8. Acknowledgments
  9. References
  • Anonymous (1998) Commission directive 98/64/EC of 3 September 1998 establishing community methods of analysis for the determination of amino-acids, crude oils and fats, and olaquindox in feedingstuffs and amending directive 71/393/EEC. Official Journal of the European Communities, L257, 1427.
  • Anonymous (2009) Commission Regulation (EC) No 152/2009 of 27 January 2009 laying down methods of sampling and analysis for the official control of feed. Official Journal of the European Communities, L54, 1130.
  • Boggess S.F. and Stewart C.R. (1976) Effect of water stress on proline synthesis from radioactive precursors. Plant Physiology, 58, 398401.
  • Bundesarbeitskreis Futterkonservierung (ed.) (2006) Praxishandbuch Futterkonservierung. Silagebereitung, Siliermittel, Dosiergeräte, Silofolien. [Handbook of feed preservation in practice. Silage making, silage additives, dosage apparatusses, silage plastic films], 7th edn. Frankfurt/Main, Germany: DLG-Verlag.
  • Castillo A.R., Kebreab E., Beever D.E., Barbi J.H., Sutton J.D., Kirby H.C. and France J. (2001) The effect of protein supplementation on nitrogen utilization in lactating dairy cows fed grass silage diets. Journal of Animal Science, 79, 247253.
  • Chamberlain D.G., Thomas P.C. and Quig J. (1986) Utilization of silage nitrogen in sheep and cows: amino acid composition of duodenal digesta and rumen microbes. Grass and Forage Science, 41, 3138.
  • Choi C.W., Ahvenjärvi S., Vanhatalo A., Toivonen V. and Huhtanen P. (2002) Quantitation of the flow of soluble non-ammonia nitrogen entering the omasal canal of dairy cows fed grass silage based diets. Animal Feed Science and Technology, 96, 203220.
  • Clark J.H., Klusmeyer T.H. and Cameron M.R. (1992) Symposium: nitrogen metabolism and amino acid nutrition in dairy cattle. Journal of Dairy Science, 75, 23042323.
  • Dawson L.E.R., Ferris C.P., Steen R.W.J., Gordon F.J. and Kilpatrick D.J. (1999) The effect of wilting grass before ensiling on silage intake. Grass and Forage Science, 54, 237247.
  • Degussa Feed Additives (1996) The amino acid composition of feedstuffs. Hanau, Germany: Degussa.
  • Edmunds B., Südekum K.-H., Spiekers H., Schuster M. and Schwarz F.J. (2012a) Estimating utilizable crude protein from forages using a modified gas test. Animal Feed Science and Technology, 175, 106113.
  • Edmunds B., Südekum K.-H., Spiekers H. and Schwarz F.J. (2012b) Estimating ruminal crude protein degradation of forages using in situ and in vitro techniques. Animal Feed Science and Technology, 175, 95105.
  • Erasmus L.J., Botha P.M., Cruywagen C.W. and Meissner H.H. (1994) Amino acid profile and intestinal digestibility in dairy cows of rumen-undegraded protein from various feedstuffs. Journal of Dairy Science, 77, 541551.
  • Ferris C.P., Gordon F.J., Patterson D.C., Kilpatrick D.J., Mayne C.S. and McCoy M.A. (2001) The response of dairy cows of high genetic merit to increasing proportion of concentrate in the diet with a high and medium feed value silage. Journal of Agricultural Science, Cambridge, 136, 319329.
  • GfE (2001) Empfehlungen zur Energie- und Nährstoffversorgung der Milchkühe und Aufzuchrinder [Recommendations for the energy and nutrient supply of dairy cows and heifer calves]. Frankfurt/Main, Germany: DLG-Verlag.
  • GfE (2008) New equations for predicting metabolisable energy of grass and maize products for ruminants. Proceedings of the Society for Nutrition Physiology, 17, 191197.
  • Givens D.I. and Rulquin H. (2004) Utilization by ruminants of nitrogen compounds in silage-based diets. Animal Feed Science and Technology, 114, 118.
  • González J., Centeno C., Lamrani F. and Rodríguez C.A. (2001) In situ rumen determination of amino acids from different feeds corrected for microbial contamination. Animal Research, 50, 253264.
  • González J., Centeno C., Morujo N., Faría-Mármol J. and Martínez A. (2009) In situ ruminal amino acid degradability of green and ensiled Italian rye-grass according to particle transit model and microbial contamination correction. Livestock Science, 123, 209214.
  • Gordon F.J., Patterson D.C., Porter M.G. and Unsworth E.F. (2000) The effect of degree of grass wilting prior to ensiling on performance and energy utilization by lactating dairy cattle. Livestock Production Science, 64, 291294.
  • Hildebrand B., Boguhn J. and Rodehutscord M. (2011) Effect of maize silage to grass silage ratio and feed particle size on protein synthesis and amino acid profile in different microbial fractions in a semi-continuous rumen simulation. Animal, 5, 537546.
  • Hristov A.N., Ahvenjärvi S., Mcallister T.A. and Huhtanen P. (2003) Composition and digestive tract retention time of ruminal particles with functional specific gravity of greater or less than 1.02. Journal of Animal Science, 81, 26392648.
  • Huhtanen P. and Hristov A.N. (2009) A meta-analysis of the effects of dietary protein concentration and degradability on milk protein yield and milk N efficiency in dairy cows. Journal of Dairy Science, 92, 32223232.
  • Huhtanen P., Rinne M. and Nousiainen J. (2007) Evaluation of the factors affecting silage intake of dairy cows: a revision of the relative silage dry-matter intake index. Animal, 1, 758770.
  • Huhtanen P., Rinne M. and Nousiainen J. (2008) Effects of silage soluble nitrogen components on metabolizable protein concentration: a meta-analysis of dairy cow production experiments. Journal of Dairy Science, 91, 11501158.
  • Hvelplund T. and Weisbjerg M. (2000) In situ techniques for the estimation of protein degradability. In: Givens D. and Owen E. (eds) Forage evaluation in ruminant nutrition, pp. 233258. Wallingford, UK: CAB International.
  • Kebreab E., France J., Beever D.E. and Castillo A.R. (2001) Nitrogen pollution by dairy cows and its mitigation by dietary manipulation. Nutrient Cycling in Agroecosystems, 60, 275285.
  • Kemble A.R. and Macpherson H.T. (1954) Liberation of amino acids in perennial rye grass during wilting. Biochemical Journal, 58, 4649.
  • Kirchhof S., Eisner I., Gierus M. and Südekum K.-H. (2010) Variation in the contents of crude protein fractions of different forage legumes during the spring growth. Grass and Forage Science, 65, 376382.
  • Kokkonen T., Touri M., Leivonen V. and Syrjala-Qvist L. (2000) Effect of silage dry matter content and rapeseed meal supplementation on dairy cows. 1. Milk production and feed utilization. Animal and Feed Science and Technology, 84, 213228.
  • Krawielitzki K., Schmidt T., Voigt J., Kowalczyk J. and Gabel M. (2006) Dynamics of microbial contamination of protein during ruminal in situ incubation of feedstuffs. Journal of Animal and Feed Sciences, 15, 313328.
  • Lebzien P. and Gädeken D.M. (1996) Untersuchungen zum Einfluss des Trockensubstanzgehaltes von Grassilage auf die Abbaubarkeit des Rohproteins im Pansen. [Investigations on the impact of the dry matter concentration of grass silage on ruminal crude protein degradation]. Landbauforschung Völkenrode, 46, 169173.
  • Licitra G., Hernandez T.M. and Van Soest P.J. (1996) Standardization of procedures for nitrogen fractionation of ruminant feeds. Animal Feed Science and Technology, 57, 347358.
  • Madsen J. and Hvelplund T. (1994) Prediction of in situ protein degradability in the rumen. Results of a European ringtest. Livestock Production Science, 39, 201212.
  • Marsh R. (1979) The effects of wilting on fermentation in the silo and on the nutritive value of silage. Grass and Forage Science, 34, 110.
  • Mass R.A., Lardy G.P., Grant R.J. and Klopfenstein T.J. (1999) In situ neutral detergent insoluble nitrogen as a method for measuring forage protein degradability. Journal of Animal Science, 77, 15651571.
  • McDonald I. (1981) A revised model for the estimation of protein degradability in the rumen. Journal of Agricultural Science, Cambridge, 96, 251252.
  • McDonald P., Henderson A.R. and Heron S.J.E. (ed.) (1991) The biochemistry of silage, 2nd edn. Marlow, UK: Chalcombe Publications.
  • Menke K.H. and Steingass H. (1988) Estimation of the energetic feed value from chemical analysis and in vitro gas production using rumen fluid. Animal Research and Development, 28, 755.
  • Merchen N.R. and Satter L.D. (1983) Changes in nitrogenous compounds and sites of digestion of alfalfa harvested at different moisture contents. Journal of Dairy Science, 66, 789801.
  • Misciattelli L., Madsen J., Møller J., Thøgersen R. and Kjeldsen A.M. (2002) Fodermidlernes indhold af aminosyrer og aminosyrernes andel af aat [Feed materials containing amino acids and the amino acid share of AAT]. Rapport nr. 98. Skejby, Denmark: Landbrugets Radgivningscenter.
  • Muck R.E. (1988) Factors influencing silage quality and their implications for management. Journal of Dairy Science, 71, 29923002.
  • Narasimhalu P., Teller E., Vanbelle M., Foulon M. and Dasnoy F. (1989) Apparent digestibility of nitrogen in rumen and whole tract of Friesian cattle fed direct-cut and wilted grass silage. Journal of Dairy Science, 72, 20552061.
  • Nguyen H.V., Kawai M., Takahashi J. and Matsuoka S. (2005) Changes in nitrogen fractions and ruminal nitrogen degradability of orchardgrass ensiled at various moisture contents, and the subsequent effects on nitrogen utilization by sheep. Asian-Australasian Journal of Animal Sciences, 18, 12671272.
  • Patterson D.C., Yan T., Gordon F.J. and Kilpatrick D.J. (1998) Effects of bacterial inoculation of unwilted and wilted grass silages. 2. Intake, performance and eating behaviour by dairy cattle. Journal of Agricultural Science, Cambridge, 131, 113119.
  • Purwin C., Pysera B., Tokarczyk M., Sederevičius A., Savickis S. and Traidaraite A. (2009) Production results of dairy cows fed grass and alfalfa silage with a different degree of wilting. Veterinarija ir Zootechnika, 46, 6066.
  • Romney D.L., Blunn V., Sanderson R. and Leaver J.D. (2000) Feeding behaviour, food intake and milk production responses of lactating dairy cows to diets based on grass silage of high or low dry-matter content, supplemented with quickly and slowly fermentable energy sources. Animal Science, 71, 349357.
  • SAS (2002) User's guide: Statistics. Cary, NC, USA: SAS Institute, Inc.
  • Sniffen C.J., O'Connor J.D., Van Soest P.J., Fox D.G. and Russell J.B. (1992) A net carbohydrate and protein system for evaluating cattle diets: II. Carbohydrate and protein availability. Journal of Animal Science, 70, 35623577.
  • Spiekers H., Ostertag J., Meyer K., Bauer J. and Richter W.I.F. (2009) Managing and controlling silos to avoid losses by reheating of grass silage. XVth International Silage Conference, Madison, WI. p. 317 (Abstr.).
  • Steingass H., Nibbe D., Südekum K.-H., Lebzien P. and Spiekers H. (2001) Schätzung des nXP-Gehaltes mit Hilfe des modifizierten Hohenheimer Futterwerttests und dessen Anwendung zur Bewertung von Raps- und Sojaextraktionsschroten. [Estimating the uCP concentration by means of the modified Hohenheim gas test and its application in the evaluation of rapeseed and soybean meals]. 113th VDLUFA-Kongreß, Berlin, Kurzfassungen der Vorträge, p. 114 (Abstr.).
  • Storm E. and Ørskov E.R. (1983) The nutritive value of rumen micro-organisms in ruminants. British Journal of Nutrition, 50, 463470.
  • Teller E., Vanbelle M., Foulon M., Collignon G. and Matatu B. (1992) Nitrogen metabolism in rumen and whole digestive tract of lactating dairy cows fed grass silage. Journal of Dairy Science, 75, 12961304.
  • Thaysen J. (2004) Die Produktion von qualitativ hochwertigen Grassilagen [The production of high-quality grass silages]. Übersichten zur Tierernährung, 32, 57102.
  • Universität Hohenheim – Dokumentationsstelle (ed.) (1997) DLG-Futterwerttabellen für Wiederkäuer [DLG feed tables for ruminants], 7th edn. Frankfurt/Main, Germany: DLG-Verlag.
  • Van Straalen W.M., Odinga J.J. and Mostert W. (1997) Digestion of feed amino acids in the rumen and small intestines of dairy cows measured with nylon-bag techniques. British Journal of Nutrition, 77, 8397.
  • Van Vuuren A.M., Tamminga S. and Ketelaar R.S. (1990) Ruminal availability of nitrogen and carbohydrates from fresh and preserved herbage in dairy cows. Netherlands Journal of Agricultural Science, 38, 499512.
  • VDLUFA (2007) Verband Deutscher landwirtschaftlicher Untersuchungs- und Forschungsanstalten – Handbuch der landwirtschaftlichen Versuchs- und Untersuchungsmethodik (VDLUFA-Methodenbuch) Bd. III. Die chemische Untersuchung von Futtermitteln. [Association of German Agricultural and Analytic Research Institutes – Handbook of Agricultural Research and Analytic Methods (VDLUFA Book of Methods) Vol. III. The chemical analysis of feedstuffs]. Darmstadt, Germany: VDLUFA-Verlag
  • Verbič J., Ørskov E.R., Žgajnar J., Chen X.B. and Žnidaršič-Pongrac V. (1999) The effect of method of forage preservation on the protein degradability and microbial protein synthesis in the rumen. Animal Feed Science and Technology, 82, 195212.
  • Volden H., Mydland L.T. and Olaisen V. (2002) Apparent ruminal degradation and rumen escape of soluble nitrogen fractions in grass and grass silage administered intraruminally to lactating dairy cows. Journal of Animal Science, 80, 27042716.
  • Von Keyserlingk M.A.G., Shelford J.A., Puchala R., Swift M.L. and Fisher F.J. (1998) In situ disappearance of amino acids from grass silages in the rumen and intestines of cattle. Journal of Dairy Science, 81, 140149.
  • Weissbach F. (1992) Bestimmung der Pufferkapazität. [Determination of buffering capacity]. Laboratory Manual, Institut für Grünland- und Futterpflanzenforschung. Braunschweig, Germany: Bundesforschungsanstalt für Landwirtschaft.
  • Weissbach F. and Kuhla S. (1995) Stoffverluste bei der Bestimmung des Trockenmassegehaltes von Silagen und Grünfutter: entstehende Fehler und Möglichkeiten der Korrektur. [Substance losses during dry matter determination of silage and fresh herbage: originating errors and possibilities of corrections]. Übersichten zur Tierernährung, 23, 189214.
  • Wilkins R.J. (1984) Summary and conclusions. In: Zimmer E. and Wilkins R.J. (eds) Efficiency of silage systems: a comparison between unwilted and wilted silages, Landbauforschung Völkenrode Sonderheft, 69, 7172.
  • Wilkinson J.M., Hill J. and Leaver J.D. (1999) Effect of swath treatment on water loss during field-wilting and on feeding value of perennial ryegrass silage. Grass and Forage Science, 54, 227236.
  • Wright D.A., Gordon F.J., Steen R.W.J. and Patterson D.C. (2000) Factors influencing the response in intake of silage and animal performance after wilting of grass before ensiling: a review. Grass and Forage Science, 55, 113.
  • Wulf M. and Südekum K.-H. (2005) Effects of chemically treated soybeans and expeller rapeseed meal on in vivo and in situ crude fat and crude protein disappearance from the rumen. Animal Feed Science and Technology, 118, 215227.