J. Anim Sci. 2007. 85:545-555. doi:10.2527/jas.2006-292
© 2007 American Society of Animal Science
Chemical composition and in situ dry matter and fiber disappearance of sorghum x Sudangrass hybrids1
P. A. Beck2,
S. Hutchison3,
S. A. Gunter,
T. C. Losi4,
C. B. Stewart,
P. K. Capps and
J. M. Phillips
University of Arkansas Division of Agriculture, Southwest Research and Extension Center, Hope 71801
 |
Abstract
|
|---|
Three sorghum x Sudangrass hybrids were planted in twelve 0.2-ha plots to test the effect of date of harvest and hybrid on plant maturity, DM yield, chemical composition, and in situ DM and fiber disappearance. Sweet Sunny Sue (a non-brown midrib (BMR) hybrid; nonBMR), NutriPlus BMR (a BMR hybrid; NP-BMR), and Dry Stalk BMR (a BMR hybrid; DS-BMR) were planted on 26 June 2003 at 22.4 kg of seed/ha. Beginning 34 d after planting, plant height and phenological growth stage were assessed weekly in 10 random, 0.5-m2 quadrats per plot. Plants were clipped to 2.5 cm in height and analyzed for CP, NDF, and ADF using near-infrared spectroscopy. Composite samples harvested from each plot on d 34, 48, and 63 were incubated in the rumen of 3 steers to determine the in situ disappearance of DM and NDF in a 3 x 3 Latin square. Forage yield was greater (P 
0.02) for nonBMR than NP-BMR on d 41 and 55 and tended (P = 0.08) to be greater on d 48. The DS-BMR hybrid produced more (P = 0.04) forage DM than the NP-BMR on d 48. When DM yield was regressed on growth stage at harvest, BMR hybrids were predicted to produce 265 kg/ha more DM (P < 0.01) than nonBMR, at the late-boot stage. At all harvest dates, NDF concentrations were less (P 0.02) for BMR than nonBMR. The DS-BMR had greater (P 0.02) NDF concentrations than NP-BMR on d 41, 48, 55, and 63. Detergent fiber concentrations were predicted to be greater (P < 0.01) in nonBMR than BMR when regressed on growth stage at harvest, but the magnitude of the differences in fiber concentration diminished with growth stage. The A fractions of DM and NDF were greater (P < 0.01) and the C fraction was less (P < 0.01) for BMR hybrids than nonBMR. The B fraction of DM was not affected (P = 0.15) by hybrid type. The B fraction of NDF was not different (P = 0.28) on d 34 but was greater (P < 0.01) on d 48 and 63 for BMR than nonBMR. Effective degradability of NDF and DM was greater (P < 0.02) for BMR than nonBMR on all harvest dates. The A fraction of DM was less for DS-BMR than NP-BMR (P < 0.01), but the B and C fractions of NDF and DM did not differ (P > 0.13) between BMR hybrids. This research indicates that forage chemical composition and ruminal in situ disappearance are improved in the BMR sorghum x Sudangrass hybrids tested compared with the nonBMR. Yield reductions are commonly reported for BMR hybrids, but predicted DM yields in the current study were not reduced if harvested at a similar phenological growth stage.
Key Words: brown midrib • digestibility • sorghum sudangrass • yield
|
INTRODUCTION
|
|---|
Voluntary DMI is critical to animal performance, and cell-wall concentration of forages is negatively associated with intake of forage diets because of contributions to ruminal fill (Jung and Allen, 1995
). In a review, Allen (2000) stated that when in situ or in vitro NDF digestibility increased by 1% in corn silage diets, DMI increased by 0.17 kg and milk yield increased by 0.25 kg. Differences among in vivo digestibility of corn silage diets were only 40% of the differences measured in vitro; this disagreement is because differences in NDF digestibility were depressed as ruminal retention time decreased and DMI increased (Oba and Allen, 1999).
Warm-season annual grasses, such as forage sorghums, Sudangrasses, and sorghum x Sudangrass hybrids [Sorghum bicolor (L.) Moench.] contribute to the possibility of developing year-round forage systems that produce high-quality forages (Fribourg, 1985). Brown-midrib (BMR) mutations, when homozygous (recessive), result in reduced lignification and cell wall concentration and increased DM digestibility and intake (Cherney et al., 1991). When forage sorghums were fed in total mixed diets to lactating dairy cows, diets containing sorghum silage with the BMR-6 mutation produced similar milk yields to those of corn silage-based diets; milk yield was least for cows fed diets containing conventional sorghum silage and intermediate for diets containing sorghum silage with the BMR-18 mutation (Oliver et al., 2004). When feedlot diets contained corn silage (10% DM basis), 10% BMR sorghum silage, or 7.5% sorghum silage (to contribute NDF equal to that of the corn silage diet), McCollum et al. (2005) reported no differences in DMI, ADG, feed efficiency, or carcass traits of steers.
This research was designed to determine the effect of harvest date on DM yield, chemical composition, and in situ DM and NDF disappearance of 3 sorghum x Sudangrass hybrids.
|
MATERIALS AND METHODS
|
|---|
All procedures involving animals in the following experiment were conducted in accordance with the recommendations of Consortium (1988) and were approved by the University of Arkansas Institutional Animal Care and Use Committee.
Small-Plot Study
A nonBMR sorghum-Sudangrass hybrid (Sweet Sunny Sue, Production Plus Quality Seed, Plainview, TX; nonBMR), a BMR sorghum-Sudangrass hybrid (NutriPlus BMR, Production Plus Quality Seed; NP-BMR), and a BMR sorghum-Sudangrass hybrid selected for reduced moisture content of the stalk (Dry-Stalk BMR, Production Plus Quality Seed; DS-BMR) were planted in 18-cm rows at a seeding rate of 22.4 kg/ha using a no-till drill (Sukup Mfg., Jonesboro, AR) in twelve 0.20-ha plots (4 plots/hybrid) on 26 June 2003 to test effects of harvest date, phenological growth stage at harvest, and hybrid on forage yield, chemical composition, and kinetics of in situ DM and NDF disappearance. These plots were located at the University of Arkansas Division of Agriculture, Southwest Research and Extension Center (33° 42' N, 93° 31' W) near Hope, AR. The soil type was Una, silty clay loam, which are deep, poorly drained, and nearly level (slopes, 0 to 1%) soils located on a flood plain (Hoelscher and Laurent, 1979). At planting, the plots were fertilized with 57 kg of N, P, and K (17-17-17) fertilizer per ha.
On d 34 after planting and at weekly intervals thereafter, plant standing height and phenological growth stage (Stauss, 1994) were assessed within each plot on plants in 10 random 0.5-m2 quadrats before clipping to a 2.5-cm height with garden shears. No area within each plot was clipped more than once; therefore, all forages collected were original growth. Samples were dried to a constant weight at 50°C in a forced-air oven and weighed to determine plant DM. Dried samples were ground to pass a 2-mm screen (Thomas A. Wiley Laboratory Mill, Model 4, Thomas Scientific, Swedesboro, NJ).
Forage CP, NDF, and ADF were estimated using near-infrared reflectance spectroscopy (Feed & Forage Analyzer, Model 6500, Foss North America, Eden Prairie, MN). Samples were selected for wet-lab analysis for equation calibration using WinISI software (version 1.04, Foss North America). Selected samples were analyzed for NDF and ADF by the batch procedures outlined by Ankom Technology Corp. (Fairport, NY). Concentration of N was determined by rapid combustion (FP-528, Leco Corp., St. Joseph, MI). Crude protein was calculated as the percentage of N in the sample x 6.25. The CP calibration equation had a SE of calibration of 0.92, a SE of cross validation of 0.93, and an R2 of 0.96. The NDF calibration equation had a SE of calibration of 2.63, a SE of cross validation of 2.73, and an R2 of 0.95. The ADF calibration equation had a SE of calibration of 1.66, a SE of cross validation of 1.70, and an R2 of 0.93.
Statistical Analysis.
The effect of hybrid, harvest date, and the hybrid by harvest date interaction on plant height, maturity, DM yield, and chemical composition were analyzed as a completely randomized design using the MIXED procedure (SAS Inst. Inc., Cary, NC); plot was considered the experimental unit, and sample within plot for each harvest date was the sampling unit. Plot within hybrid was included in the random statement, and the variance components option was used as the covariance structure. In the presence of a significant hybrid x harvest date interaction (P < 0.01), the effect of hybrid was analyzed within harvest date. Least-squares means were separated using predicted differences.
Effects of hybrid and phenological growth stage at harvest on plant height, DM yield, and chemical composition were analyzed by regression using PROC REG of SAS. These data were fitted to the following model for 1 independent variable: 
I = ß0 + ß1
I1 + ß2 I2 + ß3 I3 + ß12 I1 I2 + ß13 I1 I3 + 
I, for which ß0 is the intercept of the model, I1 equals the phenological growth stage at harvest, I2 represents the BMR nonBMR comparison, I3 is the comparison between BMR hybrids, I1 I2 and I1 I3 are the hybrid x growth stage interactions, and I is residual error. The nonBMR, NP-BMR, and DS-BMR hybrids were assigned the indicator variables –2, 1, 1 and 0, –1, 1 for the nonBMR vs. BMR comparison and the comparison between BMR hybrids, respectively. Indicator variables were used to test: i) BMR vs. nonBMR hybrids, ii) NP-BMR vs. DS-BMR, and iii) the hybrid – growth stage interaction. Accuracy and bias of the resultant reduced models were determined, as described by Gunter and Galyean (2000). Accuracy was measured using simple correlation (PROC CORR of SAS) as a measure of the association between the predicted DM yield and chemical composition and the observed DM yield and chemical composition at the same growth stage. Model bias was measured by the ß1 coefficient of a simple regression (PROC REG of SAS) with a 0 intercept resulting from the regression of observed values on predicted values.
In Situ DM and Fiber Disappearance
To determine the effect of harvest date and hybrid on in situ DM and NDF disappearance, the 10 forage subsamples from each plot collected on the 34-, 48-, and 63-d harvests were composited by harvest date and plot, resulting in 4 samples per hybrid for each harvest date. These harvest dates were chosen to evaluate the broadest possible range of plant maturities. One gram aliquots were weighed into polyester bags (7 x 7 cm; 25-um pore size, Ankom Technology Corp.) and heat-sealed at the top using an impulse sealer (Model CD-200, National Instrument Co. Inc., Baltimore, MD).
Samples were placed sequentially in the ventral rumen of 3 ruminally cannulated steers (BW = 584 ± 10.4 kg) in triplicate. Samples from each harvest date were incubated in 1 steer during each experimental period in a 3 x 3 Latin square design. Samples were soaked in water for 20 min before ruminal incubation for 6, 12, 24, 36, 48, 72, 96, and 120 h in a mesh bag. Upon removal, samples were washed in a hand-operated washer (Wonder Clean, Wonder Wash Corp., Bala Cynwyd, PA) 10 times for 2 min each until the rinse water remained clear and then dried to a constant weight at 60°C for 48 h. Composite samples for each plot and in situ samples from each steer were analyzed for NDF by the batch procedures described previously (Ankom Technology Corp.) in the bags used for ruminal incubation. Steers were maintained on mixed grass hay, which was predominantly dallisgrass (Paspalum dilatutum Poir.) and bermudagrass (Cynodon dactylon [L.] Pers.). The hay (10% CP, 62% NDF, and 34% ADF; DM basis) and fresh water was offered in quantities sufficient for ad libitum intake.
Statistical Analysis.
Disappearance curves for each steer were analyzed by nonlinear regression using the NLIN procedure of SAS. Fractions of DM and NDF were partitioned based on relative susceptibility to ruminal degradation (Ørskov and McDonald, 1979). The A fraction was defined as the immediately soluble fraction, fraction B was composed of DM and NDF that disappeared at a measurable rate, and fraction C was considered undegradable in the rumen (NRC, 1996). For the B and C fractions, disappearance rate (Kd), and lag time were determined by using the nonlinear regression model. The A fraction was calculated as 100 – (B + C). The effective degradability of OM and was calculated as A + {B x [Kd/(Kd + Kp)]} (Ørskov and McDonald, 1979), where Kp was the ruminal particulate passage rate, which was assumed to be 3.5%/h, as observed for grass hay by Ovenell et al. (1991).
The in situ DM and NDF disappearance characteristics for each hybrid were analyzed as a Latin square design using the MIXED procedure of SAS. Hybrid, harvest date, and their interaction were included in the model; steer and period were placed in the random statement; and the variance components option was used as the covariance structure. Least-squares means were separated using the following contrasts: i) BMR vs. nonBMR hybrids; ii) NP-BMR vs. DS-BMR; and iii) the hybrid x harvest date interaction; as well as the iv) linear and v) quadratic effects of harvest date.
|
RESULTS AND DISCUSSION
|
|---|
Yield and Chemical Composition
For all forage growth characteristics and measures of chemical composition, there was a harvest date x hybrid interaction (P < 0.01); therefore, DM yield, phenological growth stage, and chemical composition of the sorghum-Sudangrass hybrids are shown by harvest date (Table 1). The nonBMR hybrid yielded 1,158 kg/ ha more DM (P = 0.02) on d 41 than BMR hybrids. On d 48, DS-BMR yielded 1,095 kg/ha more (P = 0.04) DM yield than NP-BMR, whereas yield of the nonBMR hybrid tended (P = 0.08) to be greater than NP-BMR. On d 55, nonBMR yielded 1,602 kg/ha more (P < 0.01) than NP-BMR and tended (P = 0.07) to yield 924 kg/ha more than DS-BMR. On d 63, DM yield was 1,033 greater (P = 0.04) for nonBMR than DS-BMR. Yield of NP-BMR was intermediate and did not differ (P > 0.24) from nonBMR or DS-BMR. Reduced DM yield is a common phenomenon reported when the BMR gene is included in forage sorghum, Sudangrass, and sorghum-Sudangrass hybrids compared with nonBMR. Day et al. (2005) reported DM yields of BMR sorghum-Sudangrass hybrids that were approximately 19% less than nonBMR. When harvested at the soft dough stage (a phenological growth stage of 85), BMR forage sorghums had an average of 12% reduction in DM yield compared with nonBMR over 4 growing seasons (McCollum et al., 2005). Sudangrass hybrids containing the BMR gene yielded 15% less forage DM yield than related nonBMR Sudangrass hybrids (Casler et al., 2003). The nonBMR hybrid was taller (P 0.05, Table 1) than BMR hybrids at all harvest dates, and DS-BMR was taller (P = 0.02) than NP-BMR on d 48. Casler et al. (2003) reported that reduced DM yields of BMR Sudangrass hybrids harvested at the full-head stage were associated with shorter height (13 cm), reduced tillering, and reduced ground cover compared with nonBMR hybrids.
View this table:
[in this window]
[in a new window]
|
Table 1. Dry matter yield, phenological growth stage, and nutrient characteristics of sorghum-Sudangrass hybrids with progressing harvest date1
|
|
Phenological growth stage (Table 1
) was more advanced (P < 0.01) for nonBMR and DS-BMR than NP-BMR on d 48, 55, and 63. Growth stage of nonBMR was more advanced (P < 0.01) than DS-BMR on d 63. On d 48, nonBMR and DS-BMR had reached the early boot stage, while NP-BMR was at the 7-node stage. On d 55, the nonBMR and DS-BMR hybrids were in the late boot stage of development, and NP-BMR was in the early boot stage. By the d 63 harvest, the nonBMR had reached the hard dough stage, while the BMR hybrids had only reached 6 and 58% head emergence for NP-BMR and DS-BMR, respectively. A common recommendation for hay harvest is for cutting at the late-boot stage (phenological growth stage of 49; Coblentz and Phillips, 2005), whereas many variety evaluations are made at a common harvest date. The regression equation developed for the effect of phenological growth stage on DM yield explained 59% of the variation (P < 0.01, Table 2). The accuracy of the reduced models was 73, 80, and 77% for nonBMR, NP-BMR, and DS-BMR, respectively, and the bias was 1.0% for all reduced models. When DM yield was regressed on phenological growth stage (Figure 1), there were effects (P < 0.01) of growth stage, hybrid, and growth stage x hybrid. The models for BMR hybrids differed (P < 0.01) from the model for nonBMR, and the models for the BMR hybrids also differed (P = 0.01). From this regression analysis, it appears that when harvested at a similar maturity, instead of a common harvest date, DM yield reductions commonly reported for BMR hybrids disappear for the BMR hybrids tested in the current study. Predicted DM yield was greater for nonBMR than NP-BMR up to early-boot stage (phenological growth stage = 43). When the grasses reached the late boot stage (phenological growth stage = 49), NP-BMR was predicted to produce 767 kg/ha more than nonBMR and 1,003 kg/ha more than DS-BMR.
View this table:
[in this window]
[in a new window]
|
Table 2. Predicted plant height, DM yield, and nutrient characteristics of brown mid rib (BMR) and nonbrown mid rib sorghum x Sudangrass hybrids with progressing phenological growth stage1
|
|
View larger version (15K):
[in this window]
[in a new window]
|
Figure 1. Reduced model of DM yield of sorghum-Sudangrass hybrids with increasing phenological growth stage. Hybrids included nonbrown midrib (nonBMR; Sweet Sunny Sue), NP-BMR (Nutri Plus BMR), and DS-BMR (Dry Stalk BMR). All varieties were from Production Plus Quality Seed, Plainview, TX. The relationship between DM yield and phenological growth stage was explained (R2 = 0.59, P < 0.01, SE = 1,789.3) by the following model for 1 independent variable: I = 221.1I1 – 1,201.9 I2 + 1,213.8 I3 + 26.4 I1 I2 + 34.7 I1 I3 – 4,941.6, where I1 = the phenological growth stage at harvest, I2 = the BMR vs. nonBMR comparison, I3 = the comparison between BMR hybrids, and I1 I2 and I1 I3 = the hybrid x growth stage interactions. The model was reduced to = 168 –2,538 for nonBMR (—• •—),= 282-7,357 for NP-BMR (- - - -), and = 212-4,930 for DS-BMR (——). Growth stage at harvest, BMR gene, comparison of BMR hybrids, and interactions (all P < 0.01).
|
|
Dry matter (Table 1) at harvest was greater (P < 0.01) for DS-BMR than nonBMR and NP-BMR on d 34, whereas DM of NP-BMR was greater (P = 0.02) than nonBMR. On d 55, DM was greater (P < 0.01) for nonBMR and DS-BMR than NP-BMR. Dry matter at harvest was greater (P < 0.01) for nonBMR than NP-BMR, and DS-BMR and DS-BMR contained more (P < 0.01) DM than NP-BMR on d 63. Although there were significant differences in DM content of these hybrids, the practical importance of these differences should be questioned because the average difference in whole plant DM were <3 percentage units.
The nonBMR hybrid contained more CP (P < 0.01, Table 1) than DS-BMR on d 34. Crude protein concentration of nonBMR and NP-BMR were greater (P 0.03) than DS-BMR on d 41. On d 55 and 63, CP concentrations of NP-BMR were greater (P 0.01) than nonBMR and DS-BMR. On 63, nonBMR contained more CP (P = 0.02) than DS-BMR. McCollum et al. (2005) reported that BMR hybrids were on the average 0.6 percentage units greater in CP concentration than nonBMR hybrids when harvested at the soft dough stage. The regression equation developed for CP by phenological growth stage at harvest explained 43% of the variation (P < 0.01, Table 2), and the effects of growth stage, hybrid, and growth stage x hybrid were significant (P < 0.01, Figure 2). The accuracy of the reduced models was 66, 64, and 62% for nonBMR, NP-BMR, and DS-BMR, respectively, and the biases were 1.0, 1.0, and 0.9%, respectively. The relationship for BMR hybrids differed (P < 0.01) from the nonBMR hybrid, but the equations were not different (P = 0.97) for the BMR hybrids. Concentrations of CP decreased with increasing phenological growth stage. The average predicted CP concentration of the BMR hybrids was less than the nonBMR hybrid at all growth stages. Cherney et al. (1990) found no difference in CP concentration for BMR or nonBMR pearl millet. Crude protein of all sorghum-Sudangrass hybrids decreased with advancing harvest date.
View larger version (16K):
[in this window]
[in a new window]
|
Figure 2. Reduced model of CP concentration in sorghum-Sudangrass hybrids with increasing phenological growth stage. Hybrids included nonbrown midrib (non-BMR; Sweet Sunny Sue), NP-BMR (Nutri Plus BMR), and DS-BMR (Dry Stalk BMR). All varieties were from Production Plus Quality Seed, Plainview, TX. The relationship between CP concentration and phenological growth stage was explained (R2 = 0.43, P < 0.01, SE = 2.57) by the following model for 1 independent variable: I = 18.35 – 0.23 I1 + 0.79 I2 – 0.03 I3 – 0.02 I1 I2 – 0.02 I1 I3, where I1 = the phenological growth stage at harvest, I2 = the BMR vs. nonBMR comparison, I3 = the comparison between BMR hybrids, and I1 I2 and I1 I3 = the hybrid x growth stage interactions. The model was reduced to = –0.19+ 16.77 for nonBMR (—••—),= –0.27+ 19.11 for NP-BMR (- - - -), and = –0.23+ 19.77 for DS-BMR (——). Growth stage at harvest, BMR gene, and interaction (all P 0.01); comparison of BMR hybrids and interaction with growth stage at harvest (P > 0.24).
|
|
Neutral detergent fiber concentrations (Table 1) of NP-BMR and DS-BMR were 7 to 10% (4.4 to 7.0 percentage units) and 2 to 6% (1.6 to 6.0 percentage units) less (P 0.02), respectively, than nonBMR across harvest dates. The DS-BMR contained more (P 0.02) NDF than NP-BMR on d 41, 48, 55, and 63. The regression equation developed for NDF concentration by phenological growth stage at harvest explained 49% of the variation (P < 0.01, Table 2). The accuracy of the reduced models was 50, 55, and 64% for nonBMR, NP-BMR, and DS-BMR, respectively, and the bias was 1.0% for all reduced models. The effects of growth stage, hybrid, and growth stage x hybrid were significant (P < 0.01, Figure 3). The models for BMR differed (P < 0.01) from nonBMR, but the models for the BMR hybrids did not differ (P = 0.64) in predicted NDF concentration. At all phenological growth stages, BMR hybrids were predicted to contain less (P < 0.01) NDF than nonBMR. Casler et al. (2003) reported that NDF concentrations of nonBMR Sudangrass were 3.7% (2.3 percentage units) greater than BMR Sudangrasses, and Cherney et al. (1990) found NDF concentrations were 2.7% (1.8 percentage units) greater in nonBMR pearl millet than BMR pearl millet. McCollum et al. (2005) reported no differences in NDF concentrations between BMR and nonBMR forage sorghums. Fritz et al. (1988) reported that BMR sorghum-Sudangrass hybrids had 2% less NDF concentration than nonBMR when second growth was harvested at the late-boot to early-head stage.
View larger version (16K):
[in this window]
[in a new window]
|
Figure 3. Reduced model of NDF concentration of sorghum-Sudangrass hybrids with increasing phenological growth stage. Hybrids included nonbrown midrib (nonBMR; Sweet Sunny Sue), NP-BMR (Nutri Plus BMR), and DS-BMR (Dry Stalk BMR). All varieties were from Production Plus Quality Seed, Plainview, TX. The relationship between NDF concentration and phenological growth stage was explained (R2 = 0.49, P < 0.01, SE = 3.32) by the following model for 1 independent variable: I = 55.56 + 0.24I1 –2.81I2 + 0.42I3 + 0.03I1I2– 0.02I1I3, where I1 = the phenological growth stage at harvest, I2 = the BMR vs. nonBMR comparison, I3= the comparison between BMR hybrids, and I1I2 and I1I3 = the hybrid x growth stage interactions. The model was reduced to = 0.18 + 61.18 for nonBMR (—••—), ß = 0.25 + 53.16 for NP-BMR (- - - -), and = 0.29 + 52.34 for DS-BMR (——). Growth stage at harvest, BMR gene, and interaction (all P < 0.01); comparison of BMR hybrids (P > 0.46).
|
|
Across hybrids, concentrations of ADF increased with harvest date, these increases in fiber concentrations with harvest date observed in the current trial were expected (Table 1). Acid detergent fiber concentrations of NP-BMR were from 6 to 10% (2 to 4 percentage units) less (P < 0.01) than nonBMR hybrids and 4 to 10% (1 to 4 percentage units) less (P 0.05) than DS-BMR across harvest dates. Acid detergent fiber concentrations of nonBMR were greater (P < 0.01) than DS-BMR on d 48 and 55. The regression equation developed for ADF concentration by phenological growth stage at harvest explained 36% of the variation (P < 0.01, Table 2). The accuracy of the reduced models was 46, 41, and 55% for nonBMR, NP-BMR, and DS-BMR, respectively, and the bias was 1.0% for all reduced models. Predicted ADF concentrations were greater (P < 0.01, Figure 4) for nonBMR than BMR, but as the phenological growth stage increased the differences diminished (BMR gene x growth stage interaction, P = 0.01). The slope coefficients for the BMR hybrids did not differ (P = 0.16) for ADF concentration. Fritz et al. (1988) found that presence of the BMR gene decreased ADF concentration by 8 to 10% in sorghum x Sudangrass hybrids. Casler et al. (2003) reported that BMR Sudangrass hybrids contained 3% (1.2 percentage units) less ADF than normal hybrids when harvested at heading. McCollum et al. (2005) reported no difference in ADF concentration between nonBMR and BMR forage sorghum varieties. Cherney et al. (1990) reported that BMR pearl millet varieties contained 4.4% (1.5 percentage units) less ADF than nonBMR pearl millet varieties.
View larger version (15K):
[in this window]
[in a new window]
|
Figure 4. Reduced model of ADF concentration of sorghum-Sudangrass hybrids with increasing phenological growth stage. Hybrids included nonbrown midrib (nonBMR; Sweet Sunny Sue), NP-BMR (Nutri Plus BMR), and DS-BMR (Dry Stalk BMR). All varieties were from Production Plus Quality Seed, Plainview, TX. The relationship between ADF concentration and phenological growth stage was explained (R2 = 0.36, P < 0.01, SE = 2.98) by the following model for 1 independent variable: I = 31.28 + 0.17 I1 – 1.71I2 + 1.11I3 + 0.02I1I2 + 0.02I1I3, where I1 = the phenological growth stage at harvest, I2 = the BMR vs. nonBMR comparison, I3 = the comparison between BMR hybrids, and I1I2 and I1I3 = the hybrid x growth stage interactions. The model was reduced to = 0.13 + 34.70 for nonBMR (—••—), = 0.17 + 30.68 for NP-BMR (- - - -), and = 0.21 + 28.46 for DS-BMR (——). Growth stage at harvest, BMR gene, and interaction (all P < 0.01); comparison of BMR hybrids and interaction with growth stage at harvest (P > 0.16).
|
|
In Situ DM and NDF Disappearance
The fractionation and kinetics of DM disappearance are shown by hybrid for each harvest date in Table 3. The A fraction of DM was affected by main effects of hybrid and harvest date (P < 0.01); the associated interaction was not significant (P > 0.15). The A fraction was greater (P < 0.01) for BMR than nonBMR for all harvest dates (2.9, 3.3, and 4.1 percentage units greater for the 34-, 48-, and 63-d harvests, respectively). The A fraction of NP-BMR was, on average, 1.9 percentage units greater (P < 0.01) than DS-BMR for all harvest dates. Hybrid had no affect (P = 0.13) on the B fraction, but there were linear and quadratic affects for harvest date (P 0.02). Across hybrids, the B fraction decreased from 45.9 to 45.5% between the 34- and 48-d harvests. Between the 48- and 63-d harvests, the average B fraction decreased from 45.0 to 41.4%. The C fraction of DM was greater (P < 0.01) for nonBMR than BMR on all harvest dates, but hybrid x harvest date interaction was significant (P < 0.01) because the magnitude of this difference changed with harvest date. When forage was collected 34-d following planting, the C fraction of nonBMR was 1.9 percentage units greater. On the 48-and 63-d harvests, nonBMR was 4.8 and 6.2 percentage units greater than BMR, respectively. The C fraction of BMR hybrids did not differ (P 
0.17) on any harvest date.
Lag time between insertion in the rumen and disappearance of DM from the bags was not affected by hybrid (P = 0.46). However, lag time decreased (linear, P = 0.01) from 12 to 9.5 h with increasing harvest date. Rate of DM disappearance was more (P < 0.01) rapid for BMR hybrids than nonBMR, averaging 3.57 and 4.04%/h for nonBMR and BMR hybrids, respectively. Rate was not different (P = 0.31) for the BMR hybrids. Across hybrids, the rate of DM disappearance decreased (linear, P < 0.01; quadratic, P = 0.04) with harvest date (4.28, 3.68, and 3.69 %/h, for the 34-, 48-, and 63-d harvests, respectively). The effective degradability of DM from nonBMR was less (P < 0.02) than BMR on all harvest dates. The hybrid x harvest date interaction was significant (P < 0.01) because the magnitude of the differences increased with harvest date. On d 34, the effective degradability of BMR was 2.5 percentage units greater (P = 0.02) than nonBMR; whereas on d 48 and 63, effective degradability of BMR was 5.3 and 7.6 percentage units greater (P < 0.01), respectively. Effective degradability was not different (P 0.17) for the BMR hybrids on any harvest date. Fritz et al. (1988) reported that apparent DM digestibility of BMR sorghum-Sudangrass hybrids was 5 percentage units greater than nonBMR hybrids. Cherney et al. (1990) reported that IVDMD and apparent dry matter digestibility of nonBMR pearl millet was less than for BMR pearl millet. McCollum et al. (2005) reported that forage sorghums containing the BMR gene had 4.5 percentage units greater in vitro true digestibility than nonBMR hybrids. When the effective DM degradability of the sorghum-Sudangrass forages is coupled with the DM yield data from Table 1, it appears that on d 34, nonBMR produced 627 kg of degradable DM/ha compared with 879 and 992 kg/ha for NP-BMR and DS-BMR, respectively. On d 48, degradable DM production/ hectare was 2,101, 2,234, and 2,796 for nonBMR, NP-BMR, and DS-BMR, respectively, and on d 63, degradable DM production/hectare was 3,375, 3,799, and 3,315 for nonBMR, NP-BMR, and DS-BMR, respectively.
The fractionation and kinetics of NDF disappearance are shown in Table 4 for each harvest date. The A fraction of NDF was 1.9 percentage units less (P < 0.01) for nonBMR than BMR across harvest dates. There was no difference (P = 0.90) between the BMR hybrids on any harvest date. Harvest date did not affect (P = 0.46) the A fraction of NDF. Hybrid x harvest date was significant (P < 0.01) for the B fraction of NDF. The B fraction of NDF did not differ (P > 0.28) among hybrids on the 34-d harvest, whereas the B fraction of the nonBMR was 3.9 and 4.6 percentage units less (P 0.02) than the BMR hybrids between the 48 and 63-d harvest, respectively. Across hybrids, the B fraction decreased (linear, P < 0.01) from 62.6 to 54.7 percentage units of NDF between the 34 and 63-d harvests. The C fraction of NDF was greater (P < 0.01) for nonBMR than BMR on all harvest dates. Hybrid x harvest date was significant (P < 0.01) because the magnitude of this difference increased with harvest date. On d 34, the C fraction of nonBMR was 1.4 percentage units greater (P = 0.01); whereas on d 48 and 63, the nonBMR was 4.9 and 6.5 percentage units greater (P < 0.01) than BMR, respectively. The C fraction of NDF did not differ between BMR hybrids (P = 0.37) on any harvest date. Lag time between insertion in the rumen and disappearance of NDF was not affected by hybrid (P > 0.51), but there were linear (P = 0.04) and quadratic (P = 0.02) changes in lag time with increasing harvest date (12.3, 9.4, and 10.4 h for the 34-, 48-, and 63-d harvests, respectively). Rate of NDF disappearance of BMR hybrids was more rapid (P < 0.01) than nonBMR, averaging 3.54 and 3.98%/h for nonBMR and BMR hybrids, respectively. No differences (P = 0.72) in NDF disappearance rate were observed between the BMR hybrids. Across hybrids, the rate of NDF disappearance changed (linear, P < 0.01; quadratic, P = 0.02) with harvest date (4.23, 3.61, and 3.67 %/h, for the 34-, 48-, and 63-d harvests, respectively). The effective degradability of NDF from nonBMR was 3 percentage units less (P < 0.01) than BMR. The NDF degradability was not different (P = 0.93) between BMR hybrids. Effective degradability declined (linear, P < 0.01; quadratic, P = 0.03) with harvest date (44.8, 40.1, and 38.6% for 34-, 48-, and 63-d harvests, respectively).
The increased in situ detergent fiber disappearance of BMR hybrids agrees with research reported by Fritz et al. (1988) for sorghum-Sudangrass hybrids, Cherney et al. (1990) for pearl millet, and Casler et al. (2003) for Sudangrass. Fritz et al. (1988) found that the apparent NDF digestibility of BMR sorghum-Sudangrass hybrids was 5 percentage units greater than nonBMR hybrids when fed to nonlactating dairy cows, whereas the 72-h in situ extent of NDF disappearance was 4.3 to 11.6 percentage units greater with the presence of the BMR gene depending on sorghum Sudangrass hybrid. Apparent digestibility of BMR pearl millet by lambs was greater than normal pearl millet (Cherney et al., 1990). Neutral detergent fiber digestibility predicted by near-infared spectroscopy was greater for BMR than nonBMR Sudangrasses (Casler et al., 2003).
|
Footnotes
|
|---|
1 This project was conducted with funding from the Univ. of Arkansas Agric. Exp. Sta., Hatch Project No. AR001735 and supported by gifts from Kaufman Seed Co. (Ashdown, AR) and Production Plus Quality Seed (Plainview, TX). 
3 Current address: Kansas State University; 216 Weber Hall; Manhattan, KS 66505.
4 Current address: Rua 02, Quadra 06, Lote 08, apto 102; Bairro Mundinho; Cep: 75830-000; Mineiros – GOIS; Brazil.
2 Corresponding author: pbeck{at}uaex.edu
Received for publication April 4, 2006.
Accepted for publication September 3, 2006.
|
LITERATURE CITED
|
|---|
Allen, M. S. 2000. Effects of diet on short-term regulation of feed intake by lactating dairy cattle. J. Dairy Sci. 83:1598–1624.[Abstract]
Casler, M. D., J. F. Pederson, and D. J. Undersander. 2003. Forage yield and economic losses associated with the brown-midrib trait in sudangrass. Crop Sci. 43:782–789.[Abstract/Free Full Text]
Cherney, J. H., D. J. R. Cherney, D. E. Akin, and J. D. Axtell. 1991. Potential of brown-midrib, low-lignin mutants for improving forage quality. Adv. Agron. 46:157–198.
Cherney, D. J. R., J. A. Patterson, and K. D. Johnson. 1990. Digestibility and feeding value of pearl millet as influenced by the brown-midrib, low-lignin trait. J. Anim. Sci. 68:4345–4351.[Abstract]
Coblentz, W. K., and J. M. Phillips. 2005. Summer annual grasses. Univ. Arkansas Coop. Ext. Ser. FSA-2032, Little Rock, AR.
Consortium. 1988. Guide for the Care and Use of Agricultural Animals in Agricultural Research and Teaching. Consortium for Developing a Guide for the Care and Use of Agricultural Animals in Agricultural Research and Training, Champaign, IL.
Day, J. L., A. E. Coy, and P. A. Rose. 2005. Georgia 2005 soybean, sorghum grain and silage, and summer annual forages performance tests. Georgia Agric. Exp. Sta. Res. Rep. 702. http://www.griffin.uga.edu/swvt/2005/sysr05/RR702-contents.pdf Accessed June 28, 2006.
Fribourg, H. A. 1985. Summer annual grasses. Pages 278–287 in Forages. The Science of Grassland Agriculture. 4th ed. M. E. Heath, R. F. Barnes, and D. S. Metcalfe, ed. Iowa State University Press, Ames.
Fritz, J. O., K. J. Moore, and E. H. Jaster. 1988. In situ digestion kinetics and ruminal turnover rates of normal and brown midrib mutant sorghum x sudangrass hays fed to nonlactating Holstein cows. J. Dairy Sci. 71:3345–3351.[Abstract/Free Full Text]
Gunter, S. A., and M. L. Galyean. 2000. Using a segmented model to describe in situ nutrient disappearance. J. Appl. Anim. Res. 18:1–14.
Hoelscher, J. E., and G. D. Laurent. 1979. Soil survey of Hempstead County, Arkansas. USDA, SCS. Arkansas Agric. Exp. Sta., Fayetteville, AR.
Jung, H. G., and M. S. Allen. 1995. Characteristics of plant cell walls affecting intake and digestibility of forages by ruminants. J. Anim. Sci. 73:2774–2790.[Abstract]
McCollum, F. T., III, K. McCuistion, and B. Bean. 2005. Brown Midrib and photoperiod-sensitive forage sorghums. Pages 36–46 in Proc. 2005 Plains Nutr. Counc. Spring Conf. Pub. No. AREC 05–20. Texas A&M Univ. Agric. Res. Ext. Center, Amarillo, TX.
NRC. 1996. Nutrient Requirements of Beef Cattle. 7th ed. Natl. Acad. Press, Washington, DC.
Oba, M., and M. S. Allen. 1999. Evaluation of the importance of the digestibility of neutral detergent fiber from forage: Effects on dry matter intake and milk yield of dairy cows. J. Dairy Sci. 82:589–596.[Abstract]
Oliver, A. L., R. J. Grant, J. F. Pederson, and J. O’Rear. 2004. Comparison of brown midrib-6 and -18 forage sorghum with conventional sorghum and corn silage in diets of lactating dairy cows. J. Dairy Sci. 87:637–644.[Abstract/Free Full Text]
Ørskov, E. R., and I. McDonald. 1979. The estimation of protein degradability in the rumen from incubation measurements weighted according to rate of passage. J. Agric. Sci. 92:499–503.
Ovenell, K. H., K. S. Lusby, G. W. Horn, and R. W. McNew. 1991. Effects of lactational status on forage intake, digestibility, and particulate passage rate of beef cows supplemented with soybean meal, wheat middlings, and corn and soybean meal. J. Anim. Sci. 69:2617–2623.[Abstract]
Stauss, R. 1994. Compendium of growth stage identification keys for mono and dicotyledonous plants. Extended BBCH scale. Compiled by Reinhold Stauss. Ciba Geigy Ag, Basel, Switzerland.
Top
Abstract
INTRODUCTION
