Crop Science Journal of Natural Resources and Life Sciences Education
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (7)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Coblentz, W.K.
Right arrow Articles by Weyers, J.S.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Coblentz, W.K.
Right arrow Articles by Weyers, J.S.
Agricola
Right arrow Articles by Coblentz, W.K.
Right arrow Articles by Weyers, J.S.
Crop Science 40:1375-1383 (2000)
© 2000 Crop Science Society of America

CROP QUALITY & UTILIZATION

Storage Characteristics and Nutritive Value Changes in Bermudagrass Hay as Affected by Moisture Content and Density of Rectangular Bales

W.K. Coblentz, J.E. Turner, D.A. Scarbrough, K.E. Lesmeister, Z.B. Johnson, D.W. Kellogg, K.P. Coffey, L.J. McBeth and J.S. Weyers

Dep. of Animal Sci., Univ. of Arkansas, Fayetteville, AR 72701 USA

coblentz{at}comp.uark.edu


   ABSTRACT
TOP
 NOTES
 ABSTRACT
INTRODUCTION
 Materials and methods
 Results and discussion
 Conclusions
 REFERENCES
 
Conserving hay at moisture concentrations >200 g kg-1 is known to cause spontaneous heating and negative effects on forage nutritive value. While these relationships have been evaluated extensively for alfalfa (Medicago sativa L.), less research has evaluated these factors in warm-season grasses, specifically bermudagrass [Cynodon dactylon (L.) Pers.]. In this study, `Greenfield' bermudagrass was grown on a Pickwick silt loam soil (fine-silty, mixed, semiactive, thermic Typic Paleudult) and packaged in conventional rectangular bales at five concentrations of moisture (178, 208, 248, 287, and 325 g kg-1) and at high and medium bale densities (overall means = 208 and 186 kg m-3, respectively). Bale density had little effect on forage nutritive value after storage. Furthermore, bale density had no effect (P > 0.05) on indices of spontaneous heating (heating degree days >35°C, maximum temperature, and 30-d average temperature), visual evaluation of mold, and dry matter (DM) recovery. Positive linear relationships occurred between concentrations of fibrous forage components and indices of spontaneous heating; coefficients of determination (r2) statistics were similar for each index of spontaneous heating that was used as the independent variable in these linear regressions. Concentrations of acid detergent insoluble N and neutral detergent insoluble N expressed on a DM basis were positively related to measures of spontaneous heating in close linear relationships (r2 >= 0.80). Regressions of other N fractions on measures of spontaneous heating exhibited lower coefficients of determination, but generally had significant slopes. Results of this study indicate that N in bermudagrass is very susceptible to reductions in bioavailability through heat damage during bale storage, and that this damage is increased with increases in initial bale moisture.

Abbreviations: ADF, acid detergent fiber • ADIN-DM, acid detergent insoluble N expressed on a total DM basis • ADIN-N, acid detergent insoluble N expressed on a total N basis • DM, dry matter • HDD, heating degree days >35°C • IVDMD, in vitro DM disappearance • NDF, neutral detergent fiber • NDIN-DM, neutral detergent insoluble N expressed on a total DM basis • NDIN-N, neutral detergent insoluble N expressed on a total N basis • NDSN-DM, neutral detergent soluble N expressed on a total DM basis • NDSN-N, neutral detergent soluble N expressed on a total N basis


   INTRODUCTION
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
Materials and methods
 Results and discussion
 Conclusions
 REFERENCES
 
THE HARVEST, STORAGE, AND CASH SALE of improved varieties of bermudagrass hay are large components of the cattle (Bos taurus) and horse (Equus caballus) industries in the southern USA. Producers who package their hay in small rectangular bales routinely receive $154 Mg-1 for this product. Prevailing weather conditions throughout the southern USA include high relative humidity and a relatively high probability of regular rainfall events during considerable portions of the time bermudagrass is actively growing, which clearly puts this valuable cash crop at risk. Producers in the USA can be faced with the choice of baling before adequate desiccation has occurred or subjecting their crop to rain damage.

Undesirable storage characteristics and changes in nutritive value that occur when alfalfa hay is baled at moisture concentrations >200 g kg-1 are well documented (Collins et al., 1987; Coblentz et al., 1996). Considerably less information is available concerning storage characteristics and changes in nutritive value that occur in grass hays, particularly in warm-season grass hays. Reduction of forage nutritive value, primarily induced by microbial activity and the subsequent generation of heat, includes oxidation of nonstructural carbohydrates (Coblentz et al., 1997), mold growth and associated production of toxins (Roberts, 1995), and increased concentrations of fiber components and heat-damaged N (Rotz and Muck, 1994). These changes can reduce the nutritional value of the forage and result in decreased animal performance. Therefore, a better understanding of the mechanisms governing negative changes in nutritive value is important to maximize both the quality of stored grass hays and the productivity of livestock consuming these forages. The objectives of this research were to examine the effects of initial bale moisture and bale density on spontaneous heating, storage characteristics, and negative changes in the nutritive value of bermudagrass hay. A secondary objective was to relate the nutritive value of these hays after storage to indices of spontaneous heating using linear regression.


   Materials and methods
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 Materials and methods
Results and discussion
 Conclusions
 REFERENCES
 
Generation of Sample Set
An {approx}15-yr-old stand of Greenfield bermudagrass was harvested with a John Deere Model 1219 (John Deere Corp., Moline, IL) mower-conditioner equipped with metal conditioning rollers on 15 June 1998 at the University of Arkansas Forage Research Area in Fayetteville, AR. The bermudagrass was mowed at 1000 h in three blocks of 10 swaths each and allowed to dry, undisturbed, until 0930 h on 16 June. At this time, two swaths were raked together with a New Holland Model 258 side-delivery rake (Ford New Holland, New Holland, PA). Therefore, a total of five double rows per block remained after raking. Double-rows in each block were allocated randomly to one of five whole plots, based solely on moisture concentration at the time of baling. Double-rows were inverted an additional time at 1300 h with the side-delivery rake to enhance drying and allow the inclusion of bales packaged at moisture concentrations <200 g kg-1 within the treatment structure. A moisture concentration of 200 g kg-1 has been suggested as the threshold level for acceptable storage characteristics in conventional rectangular bales (Collins, 1995; Collins et al., 1987).

The subplot treatment factor for this study was bale density. Two density treatments were established within each moisture concentration; these could be considered high or medium, relative to the range of densities produced by commercial hay producers. A New Holland Model 320 baler (Ford New Holland) with hydraulic density control was used to produce the bales. The medium density treatment was created by rotating the hydraulic valve that controls tension on the bale chamber by one-half revolution in a counter-clockwise direction from the setting used for the high density bales.

Four bales (average size = 0.48 by 0.38 by 0.98 m) were made per field block for each combination of moisture concentration and bale density. The protocol for stacking baling treatments was similar to that reported previously (Coblentz et al., 1996). All bales entering storage were weighed and, due to time constraints, measured for length only. Predetermined respective averages for height and width measurements were 0.38 ± 0.01 and 0.48 ± 0.01 m, and these values were assumed to be consistent throughout the trial. Bale volume was used to calculate bale density.

For each stack, bale temperatures were monitored by inserting single thermocouples into the center of the two bales in each stack that were not core sampled prior to stacking. Bale temperatures were recorded at 0700 and 1600 h for the first 10 d after baling and once daily (at 1600 h) thereafter, until the end of the 60-d storage period. All temperature data were obtained with an Omega 450 AKT Type K thermocouple thermometer (Omega Engineering, Stamford, CT). For purposes of analysis, the mean internal bale temperature for a given day was considered the same as the observed temperature, except during the first 10 d, when the mean of the two observations was used. Degree days greater than 35°C (HDD) were computed as the summations of the daily increment by which the mean internal bale temperature was >35°C. Therefore, HDD accumulated during storage can be viewed as a single numerical value that represents and combines both the intensity and duration of spontaneous heating in hay. Previous studies (Coblentz et al., 1994b, 1996) have used 30°C as the threshold temperature level for calculating HDD; however, the 60-d storage period lasted from mid June until mid August and coincided with prolonged, excessively hot weather. Ambient temperatures approached 43°C on numerous occasions during this time period; therefore, a higher threshold was used to calculate HDD, thereby limiting the effects of elevated ambient temperatures on the number of HDD accumulated during bale storage. Indices that were evaluated as response variables for monitoring spontaneous heating included maximum temperature, minimum temperature, 30-d average temperature, 60-d average temperature, and HDD.

Prior to creating treatment stacks, two of each set of four bales were core sampled (Star Quality Samplers, Edmonton, AB, Canada) for determining both the initial concentration of moisture and nutritive value of all treatment combinations. At least two cores (35-cm depth) were taken from the ends of each bale. One cored bale was placed on the top and bottom layer of each bale stack. Core samples were dried under forced air at 50°C. After 60 d of storage, all bales that were monitored daily for internal bale temperature were core sampled (two from each stack) in a manner identical to that described previously and then visually appraised for mold growth by the method described by Roberts et al. (1987). Recoveries of DM for all stacked bales were determined from calculated DM weights of each bale before and after storage.

Chemical Analysis of Forage
Dry forage samples were ground through a Wiley mill (Arthur H. Thomas, Philadelphia, PA) equipped with a 1-mm screen and subsequently analyzed for N, neutral detergent fiber (NDF), acid detergent fiber (ADF), neutral detergent insoluble N (NDIN), acid detergent insoluble N (ADIN), acid detergent lignin, and in vitro DM disappearance (IVDMD). Respective concentrations of NDIN and ADIN were calculated and reported on the basis of total DM (NDIN-DM and ADIN-DM) and total N (NDIN-N and ADIN-N). Neutral detergent soluble N was calculated on a DM (NDSN-DM) and total N (NDSN-N) basis, where NDSN-DM = total N - NDIN-DM and NDSN-N = 1000 - NDIN-N. Total plant N and the concentration of N in NDF and ADF residues were determined using a macro-Kjeldahl procedure (Kjeltec Auto 1030 Analyzer, Tecator, Inc., Herndon, VA). Neutral detergent fiber, ADF, acid detergent lignin, hemicellulose, and IVDMD were determined by batch procedures outlined by ANKOM Technology Corp. (Fairport, NY). Rumen fluid was obtained from a ruminally cannulated crossbred steer that was offered a diet of 80% bermudagrass hay and 20% concentrate at a maintenance level of intake. The steer was adapted to the diet for 10 d prior to collecting the rumen fluid. Concentrations of hemicellulose were calculated mathematically as the difference between NDF and ADF.

Statistical Analysis
All bale characteristics and prestorage measures of nutritive value were analyzed as a split-plot design with five moisture concentrations as whole plots and two bale densities as the subplot treatment factor. Initial bale moisture was tested for significance using the mean square for the bale moisture x block interaction as the error term; bale density and the bale moisture x bale density interaction were tested with the residual error mean square as the error term. The PROC ANOVA procedure of SAS (SAS Institute, 1985) was used in all cases. Actual treatment means for bale characteristics were compared using Fisher's protected least significant difference test.

Initially, all indices of spontaneous heating, storage characteristics, and poststorage measures of nutritive value were analyzed as a split-plot design identical to that described previously. However, in order to identify trends in the data and to simplify the interpretation of results, these data were subjected to a trend analysis (PROC GLM; SAS Institute, 1985) that partitioned the sum of squares for bale moisture into linear, quadratic, cubic, and quartic effects. The mean square for the bale moisture x block interaction was used as an error term to test these effects for significance. Bale density and the associated interactions of bale density with the linear, quadratic, cubic, and quartic effects of bale moisture were tested for significance with the residual error mean square.

The relationships between poststorage nutritive value of the treatment bales and the associated measures of spontaneous heating (HDD, maximum temperature, and 30-d average temperature) were determined by the PROC REG procedures of SAS (SAS Institute, 1985). Prior to conducting this analysis, a test of homogeneity (PROC GLM; SAS Institute, 1985) was used to evaluate the effects of bale density on these relationships.


   Results and discussion
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 Materials and methods
 Results and discussion
Conclusions
 REFERENCES
 
Bale Characteristics
Bale densities were generally comparable with those reported for alfalfa hay made with similar equipment within similar moisture ranges (Coblentz et al., 1996; Buckmaster and Rotz, 1986); however, the upper limit of bale density for bermudagrass hay did not exceed 250 kg m-3 (Table 1) , which was somewhat less than the maximum reported by Coblentz et al. (1996) for rectangular bales of alfalfa hay (313 kg m-3). On both a dry and wet basis, bale moisture and density influenced (P <= 0.002) the weight and density of our treatment bales, and there was no interaction between these factors (P >= 0.196). The mean difference between high and medium density treatments for initial wet bale weights averaged across all moisture treatments was 3.75 kg (37.24 vs. 33.49 kg), which contributed to a difference of 22 kg m-3 (208 vs. 186 kg m-3) for wet bale densities of high and medium density baling treatments.


View this table:
[in this window]
[in a new window]
  		Table 1 Bale characteristics of bermudagrass hay made at five concentrations of moisture and at high (H) and medium (M) bale densities

 
Temperature Responses
Temperature vs. time curves for selected moisture treatments (Fig. 1) indicate that the pattern of spontaneous heating for bermudagrass hay was similar to that reported for alfalfa (Coblentz et al., 1996). Spontaneous heating began immediately after baling in the moist hay treatments, but subsided after 2 d of storage. An initial heating spike has been associated with respiratory processes of plant enzymes and microflora associated with the hay at the time of baling (Roberts, 1995; Hlodversson and Kaspersson, 1986; Wood and Parker, 1971). The initial temperature spike was less intense than the heating that occurred subsequently, which was consistent with observations made for alfalfa stored in a similar manner (Coblentz et al., 1996); however, the time interval associated with the initial temperature spike in bermudagrass appeared {approx}2 d shorter than those associated with this peak in alfalfa hay studies (Coblentz et al., 1996, 1994a,b). A second, prolonged period of heating was initiated on Day 2, continued for {approx}2 wk, and has been associated with the respiratory processes of storage microorganisms (Roberts, 1995). Little evidence of spontaneous heating was observed after 20 d in storage, which was a shorter heating interval than was observed in several studies with alfalfa hay that maintained elevated bale temperatures for 25 to 35 d (Coblentz et al., 1994a,b, 1996).



View larger version (12K):
[in this window]
[in a new window]
  		Fig. 1 Temperature vs. time curves for selected moisture treatments of bermudagrass hay. Concentrations of initial bale moisture are represented as follows: heavy solid line, 325 g kg-1; dashed line, 248 g kg-1; and light solid line, 178 g kg-1. Bale density had no effect on measured indices of spontaneous heating (P > 0.05); therefore, temperature vs. time curves represent the mean of high and medium density baling treatments

 
Hays baled with the greatest concentrations of moisture exhibited a more intense and prolonged period of heating than did the drier hays. Maximum internal bale temperatures exceeded 60°C in many moist bales; however, the bales made at the lowest moisture concentration also exhibited a measurable increase in the maximum internal bale temperature and accumulated HDD. These indications of respiration and spontaneous heating in dry hay were relatively minor. The extreme ambient temperatures, which approached 43°C during the study, may have exacerbated evidence of heating, particularly in bales made at the lowest moisture concentration.

Although elevated bale densities have been shown to promote spontaneous heating in hay bales (Rotz and Muck, 1994; Buckmaster and Rotz, 1986), bale density and all interaction terms including bale density had no effect (P >= 0.05) on any index of spontaneous heating in this study; therefore, only moisture means are presented in Table 2 . The mean difference in bale density between the high and medium densities (22 kg m-3) may not have been sufficient to affect spontaneous heating characteristics. All indices of spontaneous heating increased linearly (P < 0.01) with initial bale moisture. This was expected because the threshold level of moisture for satisfactory storage of rectangular bales is {approx}200 g kg-1 (Collins, 1995) and the moisture concentrations of most of our treatment bales exceeded this threshold. Quadratic, cubic, and quartic terms did not explain (P > 0.05) any characteristic of spontaneous heating.


View this table:
[in this window]
[in a new window]
  		Table 2 Heating and storage characteristics of bermudagrass hay made at five concentrations of moisture*

 
Dry Matter Recovery
Recovery of DM after the 60-d storage period decreased linearly (P < 0.001) in an inverse relationship with initial bale moisture; bale density and the associated interaction terms had no effect (P > 0.05) on DM recovery. Recoveries of DM (Table 2) increased from 941 g kg-1 in hay packaged at the 325 g kg-1 moisture level to near total recovery in the bales made at 178 g kg-1 of moisture. These effects were expected since the major factor increasing DM loss during hay storage is the moisture content of the hay as it enters storage (Rotz and Muck, 1994; Collins, 1995).

Visual Mold Appraisals
Visual appraisals of mold increased (P < 0.001) linearly with initial bale moisture (Table 2). Increases in the visual appraisal of mold in response to increases in initial bale moisture have been observed commonly in other studies with alfalfa hay (Coblentz et al., 1998a, 1996, 1994a,b). This reflects the more favorable environment for microbial growth that exists within hay bales made at high concentrations of moisture (Roberts, 1995). Bale density and the associated interaction terms had no effect (P > 0.05) on mold development. Hay made at the 325 g kg-1 moisture concentration exhibited discoloration, dustiness, obvious musty odor, and the presence of white mold between some bale flakes. In contrast, hay baled at the lowest moisture concentration exhibited little evidence of undesirable microbial activity.

Prestorage Nutritive Value of Forages
On a prestorage basis, baling treatments had little effect on forage nutritive value. The split-plot model was not significant (P >= 0.125) for all N fractions (total N, NDSN, NDIN, and ADIN); similar results were observed for IVDMD. For fiber components (ADF, NDF, hemicellulose, and lignin), only ADF and lignin exhibited moisture effects (P < 0.05). Concentrations of ADF decreased from 323 g kg-1 for hay packaged at the highest concentration of moisture to 309 g kg-1 for bermudagrass made at the lowest moisture concentration (data not shown). However, these data represent a very small range (14 g kg-1) and suggest that little variability existed in our treatment forages when they entered storage. Initial bale moisture and the bale moisture x bale density interaction affected initial concentrations of lignin; however, these data also represented a relatively small range (20.8 to 34.2 g kg-1; data not shown) and exhibited no clear pattern in response to treatment. Because baling treatments had little effect on the nutritive value of the hay on a prestorage basis, these data were combined and presented as single overall means (Tables 3 and 4) . Generally, the bermudagrass used in this study was of relatively high nutritive value and was comparable with that described for sun-cured, late vegetative `Coastal' bermudagrass (National Research Council, 1989).


View this table:
[in this window]
[in a new window]
  		Table 3 Fiber characteristics and in vitro dry matter disappearance (IVDMD) of bermudagrass hay made at five concentrations of moisture and stored in small stacks for 60 d.{dagger}

 

View this table:
[in this window]
[in a new window]
  		Table 4 Nitrogen characteristics of bermudagrass hay made at five concentrations of moisture and stored in small stacks for 60 d.{dagger}

 
Poststorage Forage Nutritive Value
For measures of nutritive value in forages sampled after 60 d in bale storage, bale density and the interaction of linear, quadratic, cubic, and quartic moisture terms with bale density were generally not significant (P > 0.05). Therefore, to simplify discussion, only moisture means are presented in Tables 3 and 4, and the discussion in the text is limited to the associated trend analysis based on initial bale moisture.

All indices of fiber composition (ADF, NDF, hemicellulose, and lignin; Table 3) increased in response to increased moisture content at baling. For each of these fiber components, the relationship with initial bale moisture was linear (P < 0.001). Higher-order terms were not generally effective (P > 0.05) in explaining the relationship between concentrations of fibrous components and initial bale moisture. A lone exception was the cubic effect observed for concentrations of hemicellulose. The maximum difference between initial and final ADF concentrations (43 g kg-1) occurred in hay packaged at the highest moisture concentration. This differential was smaller than that reported for alfalfa hay (78 g kg-1) made in conventional bales at similar concentrations of moisture (Coblentz et al., 1996). The differences between the initial and final concentrations of NDF, hemicellulose, and lignin for bales packaged at 325 g kg-1 of moisture were 71, 28, and 29.7 g kg-1, respectively. For NDF, this increase was less than half that reported for alfalfa hay bales (158 g kg-1) made at similar concentrations of moisture (Coblentz et al., 1996). Of these fiber components, the concentration of lignin exhibited the greatest increase on a percentage basis; the final concentration increased by 113% over the storage period. Typically, fiber components are not lost during hay storage or in response to the associated spontaneous heating that may occur; however, their concentrations increase indirectly due to the preferential oxidation of non-fiber components, particularly nonstructural carbohydrates (Rotz and Muck, 1994). Clearly, these data support this premise.

Concentrations of IVDMD declined with increases in initial bale moisture (Table 3). The negative relationship between concentrations of IVDMD and initial bale moisture was explained by both linear (P < 0.001) and quadratic effects. There was essentially no change in IVDMD in bales made at 178 and 208 g kg-1, relative to prestorage concentrations. The quadratic effect can probably be explained on this basis, which suggests that concentrations of IVDMD are relatively stable when hay is baled within the 200 g kg-1 moisture threshold for acceptable storage described by Collins (1987). In bales made at 325 g kg-1 of moisture, concentrations of IVDMD decreased dramatically relative to the prestorage concentration (137 g kg-1), thereby illustrating a profound effect of spontaneous heating on the digestibility of bermudagrass. Bales made at the 248 g kg-1 moisture concentration exhibited a depression in IVDMD concentration of 28 g kg-1, which agrees closely with depressions in IVDMD (30 g kg-1) reported by Rotz and Muck (1994) for hay packaged at 250 g kg-1 of moisture.

Concentrations of N increased linearly with moisture concentration at baling (Table 4); however, higher order terms had no effect (P > 0.05) on concentrations of N after storage. Increases in the concentration of N have been observed previously (Coblentz et al., 1996; Rotz and Abrams, 1988) in hays sampled after relatively short storage periods (30–60 d) and may be the indirect result of preferential oxidation of nonstructural carbohydrates early in the storage period (Rotz and Muck, 1994).

Recent efforts (Mass et al., 1999; Coblentz et al., 1999) to improve the evaluation of forage protein quality and degradation characteristics of forage N have focused attention on the fraction of forage N that is insoluble in neutral detergent (NDIN). This fraction is believed to offer more resistance to ruminal degradation than N found within the cell solubles (Sniffen et al., 1992). Empirically, grasses that fix carbon by the C4 pathway, which includes bermudagrass (Ball et al., 1996), have large concentrations of NDIN that may account for >50% of the entire N pool (Brown and Pittman, 1991; Coblentz et al., 1998b; Juarez et al., 1997). To our knowledge, the effects of bale moisture and bale density on poststorage concentrations of NDIN have not been evaluated, although Goering et al. (1973) has reported increases in the concentration of this fraction in rehydrated forages that were heated artificially.

Concentrations of NDIN increased with initial bale moisture when this fraction was expressed on both a DM (NDIN-DM) and total N (NDIN-N) basis; these relationships were explained by linear (P <= 0.001) and quadratic (P <= 0.01) effects (Table 4). Generally, concentrations of NDIN-DM and NDIN-N reached a maximum when the initial bale moisture reached 248 g kg-1. Prestorage concentrations of NDIN constituted about 50% of the total plant N, which is consistent with reports for other warm-season forages (Brown and Pittman, 1991; Coblentz et al., 1998b). On a poststorage basis, NDIN-N exceeded 50% for all bales made at moisture concentrations >178 g kg-1.

The fraction of N that is soluble in neutral detergent (NDSN) is degraded primarily in the rumen, although some of this fraction may escape to the lower gut depending on relative rates of digestion and passage (Sniffen et al., 1992). In theory, this fraction should decrease in response to spontaneous heating because heating is known to irreversibly bind forage N within the ADF matrix (Goering et al., 1973). In this study, NDSN expressed on a DM basis (NDSN-DM) declined in an inverse relationship with initial bale moisture; there were linear and quadratic effects in this relationship (Table 4). This fraction reached a minimum in bales made at 287 g kg-1 of moisture; however, a sharp increase was observed in the wettest hay. Neutral detergent soluble N expressed on a total N basis (NDSN-N) also declined in an inverse relationship with initial bale moisture, and exhibited linear and quadratic effects.

Quantification of N that is insoluble in acid detergent (ADIN) is used to evaluate heat damage to forage N via the nonenzymatic (Maillard) browning reaction (Licitra et al., 1996; Van Soest, 1982; Goering et al., 1973). Increased concentrations of ADIN are normally assumed to be the product of nonenzymatic browning and are frequently associated with spontaneous heating in hay and silage (Goering et al., 1973, 1972; Yu and Thomas, 1976). The ADIN fraction may also increase in response to the commercial dehydration of alfalfa (Goering, 1976). Generally, bermudagrass hays that have heated spontaneously have not been evaluated for ADIN. In this study, ADIN increased linearly (P < 0.001) with moisture content at baling; this was true when ADIN was expressed on both a DM (ADIN-DM) and N (ADIN-N) basis. The maximum proportion of N bound within the ADF matrix constituted {approx}14% of the total plant N in bales packaged with 325 g kg-1 of moisture.

Regression of Forage Nutritive Value on Indices of Spontaneous Heating
A test of homogeneity indicated that bale density had no effect (P > 0.05) on the relationship between final concentrations of individual measures of nutritive value and indices of spontaneous heating; therefore a common regression line that combined bales made at high and medium densities is discussed for all cases.

Fiber Components
Poststorage concentrations of ADF, NDF, hemicellulose, and lignin were regressed linearly on three indices of spontaneous heating; these included HDD, maximum temperature, and 30-d average temperature (Table 5) . Although HDD is a single number that incorporates both the magnitude and duration of spontaneous heating during the storage period, there was no obvious advantage to using this calculation as an independent predictor variable. Maximum temperature and 30-d average temperature both yielded regression equations with comparable r2 statistics. Generally, r2 statistics were high for all fiber components (r2 >= 0.681), regardless of which index of heating is used as the predictor variable. Indices of spontaneous heating were most closely related to the concentration of NDF (r2 >= 0.940). Close relationships were also observed for ADF (r2 >= 0.813) and lignin (r2 >= 0.844). Similar relationships have been reported previously for alfalfa hay made at moisture concentrations of 202 and 297 g kg-1 that was sampled over time in storage (Coblentz et al., 1996).


View this table:
[in this window]
[in a new window]
  		Table 5 Regressions of fiber components and in vitro dry matter disappearance (IVDMD) on heating degree days >35°C (HDD), maximum temperature (Max.), and 30-d average temperature (Avg.) for bermudagrass hay made at five moisture concentrations and high and low bale densities (n = 10 treatments) and stored in small stacks for 60 d

 
Although the hays baled with the highest concentrations of moisture in this study accumulated substantial HDD (>300), the overall increases in concentrations of NDF from prestorage levels were comparatively small (71 g kg-1) relative to those observed for alfalfa hay made with similar concentrations of moisture and in similar bale packages (158 g kg-1; Coblentz et al., 1996). Rotz and Muck (1994) suggested that fiber components (ADF, NDF, crude fiber, lignin, and ash) are relatively stable during storage, and that their concentrations increase indirectly during this period because of the oxidation of nonfiber components (particularly nonstructural carbohydrates) to CO2, water, and heat. The disproportionately large increases in concentrations of ADF and NDF observed previously in alfalfa hay bales (Coblentz et al., 1996) could be explained on the basis of higher concentrations of cell solubles generally, and nonstructural carbohydrates particularly, in this cool season legume when the hay enters storage (Coblentz et al., 1997). Warm season grasses typically possess limited quantities of cell solubles, including nonstructural carbohydrate (Van Soest, 1982), which may limit the preferred substrate for respiration and the potential indirect increase in concentrations of fiber components.

In Vitro Dry Matter Disappearance
Concentrations of IVDMD declined linearly (r2 >= 0.698) with all measures of spontaneous heating (Table 5). This negative relationship with heating occurs largely by the preferential oxidization of highly digestible plant sugars during respiratory processes within the bale during the storage period (Rotz and Muck, 1994). Reductions in digestibility in response to unsatisfactory storage have been reported previously (Collins et al., 1987; Rotz and Abrams, 1988). Linear regressions have been reported for alfalfa hay that negatively relate IVDMD to initial bale moisture (Collins et al., 1987) and positively relate in vitro DM indigestibility to heating degree days > 30°C (Coblentz et al., 1996).

Nitrogen Components
In contrast to the fibrous plant components, the concentration of some N fractions can increase by direct mechanisms. Goering et al. (1973) suggested that the effective heating period, temperature, moisture content, and forage species all contribute to the binding of protein within the ADF matrix by nonenzymatic browning. The primary reaction in this pathway involves the chemical polymerization of sugars and other carbohydrates with amino acids (Rotz and Muck, 1994); principal carbohydrates include sucrose and hemicellulose, the latter of which occurs in much higher concentrations in grasses than in legumes (Van Soest, 1982). These factors were corroborated by the results of this study in that ADIN-DM and ADIN-N were closely related linearly (r2 >= 0.850) to indices of spontaneous heating (Table 6) . Close linear relationships (r2 >= 0.850) were observed for all measures of spontaneous heating (HDD, maximum temperature, and average temperature). Similar positive, linear relationships between ADIN and indices of spontaneous heating have been reported previously for alfalfa hay (Coblentz et al., 1996, 1998a).


View this table:
[in this window]
[in a new window]
  		Table 6 Regressions of N components on heating degree days >35°C (HDD), maximum temperature (Max.), and average temperature for the first 30 d of storage (Avg.) for bermudagrass hay made at five moisture concentrations and high and low bale densities (n = 10 treatments) and stored in small stacks for 60 d

 
Concentrations of NDIN-DM were also closely related (r2 >= 0.799), in a positive manner, to indices of spontaneous heating, but regressions for NDIN-N were characterized by poorer r2 statistics (r2 >= 0.542). In theory, these fractions should also include products of the Maillard reaction described previously. Regressions of NDSN-DM on indices of spontaneous heating were not significant (P > 0.05). Concentrations of total plant N generally increased in response to heating, probably due to preferential oxidation of nonstructural carbohydrates (Rotz and Muck, 1994); however, significant regressions (P <= 0.05) were exhibited only when maximum temperature and 30-d average temperature were used as independent variables.


   Conclusions
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 Materials and methods
 Results and discussion
 Conclusions
REFERENCES
 
Storage characteristics and changes in nutritive value for bermudagrass hays that have heated spontaneously followed patterns similar to those reported for alfalfa and other forage crops. In this study, the initial moisture concentration of bermudagrass hay bales clearly increased the subsequent spontaneous heating that occurred during storage. Increased concentrations of fiber components and cell wall–bound N were also observed, especially when initial concentrations of bale moisture exceeded 200 g kg-1. Bale density had no effect on most response variables. Positive linear relationships were found between concentrations of fibrous forage components and characteristics of spontaneous heating (HDD, maximum temperature, and 30-d average temperature). Concentrations of artifact N and NDIN-DM were positively related to indices of spontaneous heating in close linear relationships (r2 >= 0.799). Other N fractions exhibited poorer linear relationships with indices of spontaneous heating. The relationships between measures of nutritive value and heating indices were not affected by bale density. Based on the heating increments measured in these treatment bales and the proportions of N bound within the ADF matrix, N in bermudagrass appears to be very susceptible to Maillard reaction damage.SAS Institute 5


   NOTES
TOP
 NOTES
ABSTRACT
 INTRODUCTION
 Materials and methods
 Results and discussion
 Conclusions
 REFERENCES
 
Contribution no. 99109 of the Arkansas Agric. Exp. Stn.

Received for publication October 26, 1999.


   REFERENCES
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 Materials and methods
 Results and discussion
 Conclusions
 REFERENCES
 




This article has been cited by other articles:


Home page
 J ANIM SCIHome page
R. K. Ogden, W. K. Coblentz, K. P. Coffey, J. E. Turner, D. A. Scarbrough, J. A. Jennings, and M. D. Richardson
Ruminal in situ disappearance kinetics of nitrogen and neutral detergent insoluble nitrogen from common crabgrass forages sampled on seven dates in northern Arkansas.
J Anim Sci, March 1, 2006; 84(3): 669 - 677.
[Abstract] [Full Text] [PDF]

Home page
 J DAIRY SCIHome page
R. T. Rhein, W. K. Coblentz, J. E. Turner, C. F. Rosenkrans Jr., R. K. Ogden, and D. W. Kellogg
Aerobic Stability of Wheat and Orchardgrass Round-Bale Silages During Winter
J Dairy Sci, May 1, 2005; 88(5): 1815 - 1826.
[Abstract] [Full Text] [PDF]

Home page
 Agron. J.Home page
J. E. Turner, W. K. Coblentz, D. A. Scarbrough, K. P. Coffey, D. W. Kellogg, L. J. McBeth, and R. T. Rhein
Changes in Nutritive Value of Bermudagrass Hay during Storage
Agron. J., January 1, 2002; 94(1): 109 - 117.
[Abstract] [Full Text] [PDF]

Home page
 Crop Sci.Home page
W. K. Coblentz, J. E. Turner, D. A. Scarbrough, K. P. Coffey, D. W. Kellogg, and L. J. McBeth
Effects of Spontaneous Heating on Estimates of Ruminal Nitrogen Degradation in Bermudagrass Hays from Two Harvests
Crop Sci., September 1, 2001; 41(5): 1572 - 1578.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (7)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Coblentz, W.K.
Right arrow Articles by Weyers, J.S.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Coblentz, W.K.
Right arrow Articles by Weyers, J.S.
Agricola
Right arrow Articles by Coblentz, W.K.
Right arrow Articles by Weyers, J.S.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
The SCI Journals Agronomy Journal Vadose Zone Journal
Journal of Plant Registrations Soil Science Society of America Journal
Journal of Natural Resources
and Life Sciences Education		 Journal of
Environmental Quality