Flake density of steam-processed sorghum grain alters performance and sites of digestibility by growing-finishing steers R. S. Swingle, T. P. Eck, C. B.

Theurer, M. De la Llata, M. H. Poore and J. A. Moore J Anim Sci 1999. 77:1055-1065.

The online version of this article, along with updated information and services, is located on the World Wide Web at: http://jas.fass.org

www.asas.org

Downloaded from jas.fass.org by on October 21, 2009.

Flake Density of Steam-Processed Sorghum Grain Alters Performance and Sites of Digestibility by Growing-Finishing Steers1
R. S. Swingle2, T. P. Eck3, C. B. Theurer4, M. De la Llata5, M. H. Poore6, and J. A. Moore5
Department of Animal Sciences, University of Arizona, Tucson 85721-0038

ABSTRACT: The effect of several flake densities ( F D ) of steam-processed sorghum grain on performance, and site and extent of nutrient digestibilities by steers fed growing and finishing diets was determined. The effectiveness of common laboratory methods of starch availability (enzymatic hydrolysis or gelatinization) to provide target specifications for quality control of steam-flaked grains was also measured. In vitro starch availability of the processed grains increased ( P < .05) linearly in response to decreased FD. Flake density was more highly correlated with enzymatic measures than with percentage gelatinization ( R 2 = .87 to .93 vs .36). Using 140 crossbred beef steers (181 kg initial weight), feedlot performance was determined for 112 d with a growing diet (50% grain), followed by 119 d with a finishing diet (78% grain). Each FD treatment (412, 360, 309, and 257 g/L or 32, 28, 24, and 20 lb/bu) was randomly assigned to five pens of seven steers each. Intake of DM by steers decreased linearly ( P < .05) as FD decreased ( 7 and 13%, respectively, for growing and finishing diets). Decreasing FD reduced linearly ( P <

.05) ADG in the finishing phase and for the entire 231-d trial. With the growing diet only, feed efficiency and estimated diet NEm and NEg responses to decreasing FD were curvilinear ( P < .05), with the 360 g/L (28 lb/bu) flake being most efficient. Electrical energy requirements for processing increased linearly ( P < .05) as FD decreased. Using four multicannulated crossbred steers (275 kg), starch digestibility increased linearly ( P < .05) in the rumen (82 to 91%) and total tract (98.2 to 99.2%) as FD decreased. Digestibilities within the small (74%) and large intestines (62%) were not altered by FD. Decreasing FD increased ( P < .05) total CP digestibility, but did not consistently alter fiber digestibility or DE content of the diets. In conclusion, enzymatic laboratory methods to evaluate starch availability in processed grains can be used satisfactorily to establish FD criteria for quality control of the steam-flaking process. The greatest improvements in efficiency, estimated diet NE, and starch and protein digestibilities usually occurred when FD was decreased from 412 to 360 g/L (32 to 28 lb/bu). Based on these measures and processing costs, the optimal FD was 360 g/L (28 lb/bu).

Key Words: Steam-Flaking, Sorghum, Cattle, Performance, Starch Digestion

1999 American Society of Animal Science. All rights reserved.

J. Anim. Sci. 1999. 77:10551065

Introduction
Steam-flaking has method for processing for feedlot cattle since rolling, steam-flaking been the most widely used sorghum grain for use in diets the 1960s. Compared with dryimproves the feeding value of

sorghum grain by 12 to 15%, principally by improving digestibility of starch in the rumen and total tract (Hale, 1973; Swingle, 1992; Theurer et al., 1996; Huntington, 1997). Moisture, temperature, and pressure are important variables affecting the degree of improvement in starch availability (Theurer, 1986).

1Appreciation is expressed to J. Sanders and M. Townsend for feeding and care of animals and grain processing; L. Woo and G. Ghenniwa for help in laboratory analyses; and L. George-Smith for aid in manuscript preparation. 2Present address: Cactus Feeders Inc., 2209 W. 7th Street, Amarillo, TX 79106. 3Present address: Nutrition Service Associates, 1002 N. 4th, Garden City, KS 67846.

whom correspondence should be addressed. address: Alimentos de Sonora, Calle Pequena Industria 2135-A, Parque Industrial Apdo. Postal 699, Ciudad Obregon, Sonora, Mexico. 6Present address: North Carolina State University, Department of Animal Sciences, Box 7621, Raleigh, NC 27695-7621. Received March 31, 1998. Accepted October 5, 1998.
5Present

4To

1055
Downloaded from jas.fass.org by on October 21, 2009.

1056

SWINGLE ET AL.

In the feedlot industry, considerable effort is expended to regulate these variables to achieve the desired degree of processing. Flake density ( FD) and measures of starch availability (i.e., susceptibility to enzymatic hydrolysis or degree of gelatinization) are commonly used as target specifications for quality control in the feedmill. Given the emphasis on quality control in steamflaking of grains, and the time, effort, and expense spent to optimally process grains, there is surprisingly little published data relating the performance of cattle in the feedlot to the extent of steam-processing of corn (Zinn, 1990), and sorghum grain (Xiong et al., 1991; Reinhardt et al., 1997; Theurer et al., 1999). These studies report conflicting effects of varying FD on ADG and efficiency by growing-finishing steers. Zinn (1990) reported that decreasing FD of steamprocessed corn from 360 to 257 g/L (28 to 20 lb/bu) increased total digestibilities of starch, CP, and energy. There are no published data on the effect of flaking steam-processed sorghum grain to a range of densities on site and extent of nutrient digestion. This study was conducted to determine 1 ) the effect of steam-flaking sorghum grain over a range of flake densities (starch availabilities) on performance and nutrient digestibility responses by growing-finishing steers and 2 ) whether different laboratory methods for evaluating starch availability in processed grains can be used satisfactorily to establish FD criteria for quality control of the steam-flaking process.

Samples of each FD were composited over 28-d periods and analyzed for starch availability by quantifying glucose released and percentage of starch hydrolyzed during 30-min incubations with amyloglucosidase (Poore et al., 1989, 1991). Samples were also submitted to commercial laboratories for evaluation of relative starch availability by enzymatic (Nutrition-Laboratory Services, Inc., Tolleson, AZ) and loss of birefringence (High Plains Laboratory, Hereford, TX) methods. Electrical load on the roller mill during production of each FD was measured using an ammeter.

Trial 1. Performance and Nutrient Digestibility by Feedlot Steers

Trials 1 and 2 were conducted with approval and under the supervision of the University of Arizona Institutional Animal Care and Use Committee. In a 231-d feeding trial, 140 crossbred steer calves (about to Brahman, with the remainder a mixture of English and Continental breeds) were fed a growing diet (1.17 Mcal NEg/kg DM, 50% flaked grain, 33% roughage) for the first 112 d and a finishing diet (1.45 Mcal NEg/kg DM, 78% flaked grain, 10% roughage) for the last 119 d (Table 1). Each flake treatment (SF32, SF28, SF24, or SF20) was randomly assigned to five pens (weight replicates) of seven steers each. Roughage in both diets was a mixture (equal NDF basis) of wheat straw and alfalfa hay. The roughages were chopped using a rotor-type mill (alfalfa, 15 cm screen; wheat straw, 2.5 cm screen). Animals were given ad libitum access to feed, which was added to bunks at 0700 and 1400 daily. Steers were weighed (pencil shrunk 4%) individually on two consecutive days to start the experiment (mean initial weight, 181 kg) and on single days at 28-d intervals. Final shrunk live weight was calculated from hot carcass weight using a mean dressing percentage (64.7%) across treatments. Group weights (shrunk 4%) on each pen, taken on the day of shipment, were used to calculate dressing percentage. Steers were implanted with Ralgro (Mallinckrodt Veterinary Inc., Mundelein, IL) at the beginning of the experiment and reimplanted with Synovex-S (Syntex, Des Moines, IA) on d 85 and 169. Performance data were calculated from DMI and shrunk ( 4 % ) live weights. Estimated diet NE concentrations were calculated from performance data using established energy relationships (NRC, 1984). Fecal grab samples were taken from at least four steers in each pen on d 42 of the 112-d growing phase and on d 88 of the 95-d finishing phase. Fecal samples were composited for each pen. Chromic oxide (.05% of DM, growing phase; .1% of DM, finishing phase) was added to diets at least 5 d prior to fecal sampling as a digestibility marker. Diet samples from each pen were dried at 55C in a forced-air oven, and fecal samples were lyophilized

Materials and Methods Grain Processing and Starch Availability

Sorghum grain was steam-processed to FD of 412, 360, 309, and 257 g/L (referred to respectively as SF32, SF28, SF24, and SF20, reflecting bushel weight in pounds, typical of field studies) using a 46- 76-cm roller mill fitted with a 3-m stainless steel steam chest. Steam was from a natural gas-fired steam generator. Grain was flaked at least twice a week and used immediately for preparation of diets. Flow of grain into the rollers was maintained at a constant rate to allow a steaming time of approximately 40 min. Desired FD were achieved by progressively increasing tension on the rollers; processing was always in order from the most dense to least dense in order to maintain the desired intervals in degree of processing. Rollers were preheated by processing about 454 kg of grain prior to preparing experimental flakes. Each time grain was prepared, bulk density was measured on whole dry grain and on flaked grain as it exited the rollers. The DM was determined on the grains at each stage of processing (whole unprocessed, steamed entering the rollers, and on flaked as it exited the rollers and as it was weighed into the diets).

Downloaded from jas.fass.org by on October 21, 2009.

FLAKE DENSITY OF STEAM-PROCESSED SORGHUM

1057

Table 1. Ingredient composition and nutrient analyses of growing and finishing diets (Trials 1 and 2)
Dieta Item Growing Finishing

% of DM Composition (Trials 1 and 2 ) Alfalfa hay Wheat straw Steam-flaked sorghum grainb Cottonseed meal Molasses Animal fat Supplementc Limestone Urea NaCl Mono-, dicalcium phosphate Chromic oxide (Trial 1 ) Nutrient analyses Trial 1 DM, % DM composition, % OM Starch Protein NDF Gross energy, Mcal/kg Trial 2 DM, % Composition of DM, % OM Starch Protein NDF ADF Gross energy, Mcal/kg 20.5 12.2 50.0 8.0 4.0 2.0 1.0 .7 .6 .5 .2 .05 6.3 3.5 78.3 3.0 4.0 2.0 1.0 .9 .5 .5 .1

hart, IN) according to the method described by Karr et al. (1987). Chromium was determined by atomic absorption spectrophotometry after wet ashing (Poore et al., 1991). Data were analyzed using a randomized complete block design, with four treatments in five blocks (weight replicates), with pen as the experimental unit (Steel and Torrie, 1980). Orthogonal polynomials (linear, quadratic, and cubic components) were used to characterize response curves, and orthogonal contrasts were used to compare treatments. Cubic components are not given in the results because linear and quadratic components almost always had lower probability values. Orthogonal comparisons were SF32 vs SF28, SF24, and SF20; SF28 vs SF24 and SF20; and SF24 vs SF20.

Trial 2. Nutrient Digestibilities in the Rumen and Intestines

Four crossbred steers (275 kg) fitted with cannulae in the rumen, duodenum (flange approximately 5 cm from the pyloric sphincter), and ileum (flange approximately 60 cm from the ileocecal junction) were used in a Latin square design. The objective was to evaluate the effects of flaking sorghum grain to the same densities as for the performance trial (SF32, SF28, SF24, and SF20) on site and extent of digestion. Growing and finishing diets were the same as in Trial 1 (Table 1). Digestibilities of the growing diet (first study) and of the finishing diet (second study) were conducted with the same steers. Chromic oxide (.1% of DM) was included in the diets throughout each study as a digestibility marker. Steers were kept in individual pens (2.5 5.0 m ) with concrete floors and partial shade, and had free access to water and trace-mineralized salt blocks (provided 3,500 mg Zn, 2,000 mg Fe, 1,800 mg Mn, 350 mg Cu, 100 mg I, and 60 mg Co per kg of block). Steers were fed twice daily at 0550 and 1750. Feed offered, regardless of treatment, never exceeded the maximum intake established for each steer prior to initiation of each study. In these digestibility studies, DMI averaged 6.8 and 6.4 kg/d for the growing and finishing diets, respectively, and was 91 and 86%, respectively, of ad libitum intakes by feedlot steers fed these same diets in Trial 1. Periods of the Latin square consisted of 10 d of adaptation followed by 4 d of sample collection. Feed and ort samples were taken daily for 4 d. Samples of duodenal ( 300 mL) and ileal digesta ( 150 mL) and feces ( 250 g wet basis) were collected four times daily over 3 d. These samples were immediately placed on ice. Collection times were moved forward 2 h each day. This collection system provides a total of 12 samples, representing each 2 h of a 24-h period. Feed, ort, digesta, and fecal samples were kept frozen at 4C until composited. These samples were pooled by steer within periods and then dried at 55C in a

86.1 92.8 38.9 14.3 27.6 4.18 88.1 92.7 40.4 14.6 28.2 19.2 4.14

88.9 95.5 59.8 12.9 15.4 4.15 84.3 95.6 61.8 12.8 16.0 14.7 4.11

aTrial 1: Growing diet fed d 1 through 112; finishing diet fed d 113 through 231. Calculated NEg for growing and finishing diets were 1.17 and 1.45 Mcal/kg DM, respectively. bSteam-flaked to densities of 412, 360, 309, and 257 g/L (32, 28, 24, and 20 lb/bu). cProvided 3,300 IU vitamin A and 33 mg monensin/kg of diet DM.

prior to laboratory measurements. Diet and fecal samples were ground to pass through a 2-mm screen in a Wiley Mill (Arthur H. Thomas Co., Philadelphia, PA) and then re-ground through a 1-mm screen in a cyclone mill (Tecator Cyclotec Model 1093, Hoganas, Sweden). Final DM was determined by drying at 100C under vacuum. Nitrogen was determined by automated procedures (Technicon, Tarrytown, NY) after Kjeldahl digestion (AOAC, 1984). The NDF was analyzed according to Robertson and Van Soest (1981), using heat stable a-amylase ( Bacillus licheniforms Type XII-A, Sigma Chemical Co., St. Louis, MO). Gross energy was determined by adiabatic bomb calorimetry (Parr Model 1241, Moline, IL). Total starch was determined indirectly as glucose (YSI Model 27, Yellow Springs Instrument Co., Inc., Yellow Springs, OH) following hydrolysis of samples using amyloglucosidase (Diazyme L-200, Miles Inc., Elk-

Downloaded from jas.fass.org by on October 21, 2009.

1058

SWINGLE ET AL.

forced-air oven for 48 h. Sample processing and analytical methods for starch availability, DM, OM, N, NDF, starch, GE, and chromium are described in Trial 1. The ADF was analyzed using the methodology of Goering and Van Soest (1970). Data for each study (growing and finishing) were separately analyzed as a 4 4 Latin square using SAS (1985). Orthogonal polynomials and orthogonal comparisons are described in Trial 1.

Results and Discussion Starch Availability

Characteristics of grain flaked to different densities are presented in Table 2. Moisture content of the grain was increased from 12% to approximately 17% during exposure to steam. Starch availability, as measured by each of the laboratory methods, increased linearly or quadratically ( P < .05) in response to increased extent of processing (decreased FD), with SF32 always being the lowest value and SF20 the highest value. The range of responses was much greater for enzymatic methods than for gelatinization based on loss of birefringence. Table 3 shows correlations among laboratory measures of starch availability. Flake density was more highly correlated with enzymatic measures of starch availability than with percentage gelatinization determined as loss of birefringence ( R 2

= .87 to .92 vs .36, respectively), and the enzymatic methods were highly correlated with each other ( R 2 = .90 to .93). Xiong et al. (1990a) also found that enzymatic methods are more sensitive than gelatinization. Other studies have demonstrated that decreasing FD linearly increases in vitro rate of starch hydrolysis in grains by amylolytic enzymes (Theurer, 1986; Xiong et al., 1990a,b; 1991; Zinn, 1990; Theurer et al., 1999). We emphasize that the data from the present study were determined from grain flaked to different densities in a single mill, under controlled conditions, and over relatively short periods of time. Other factors, including initial density and moisture content of the grain, steaming time, level of moisture addition, and condition of the rollers, affect laboratory measures of starch availability at a given FD (Karr, 1984). We conclude that the enzymatic laboratory methods commonly used to evaluate starch availability in processed grains can be used satisfactorily to establish FD criteria for quality control of the steam-flaking process.

Trial 1. Performance and Nutrient Digestibility by Feedlot Steers

Performance data for growing ( d 1 to 112) and finishing ( d 113 to 231) phases and for the entire feeding period ( d 1 to 231) are in Table 4. The largest and most consistent response to increased extent of

Table 2. Dry matter content and starch availability measurements for sorghum grain steam-flaked to different densitiesa
Flake densityb Item Grain DM, % Exiting rollers Entering diets Starch availability Enzyme method 1c Glucose release, mg/g grain DMyz Percentage starch hydrolysisyz Enzyme method 2d Glucose release, mg/g grain DMy Birefringencee Percentage gelatinizationyz
aEach

SF32 82.6 82.9

SF28 82.9 83.3

SF24 83.5 83.9

SF20 83.9 84.3

SEM .1 .1

531 67.6 283 52.6

614 75.4 321 54.6

684 83.2 376 54.9

730 88.1 413 55.3

7.8 .8 3.9 .4

value is the mean of eight samples (each sample pooled over 28-d periods). Grain DM prior to processing and prior to entering the rollers (mean SD) was 88.0 .39 and 81.8 .50%, respectively. Starch availability measures for unprocessed grain (mean SD) were: method 1, 202 6.3 mg glucose/g DM and 25.3 1.2% starch hydrolysis; method 2, 168 12.2 mg glucose/g DM; birefringence method, 10 2.3 gelatinization. bSF32, SF28, SF24, and SF20 = steam-flaked at densities of 412, 360, 309, and 257 g/L (32, 28, 24, and 20 lb/bu), respectively. cDept. Anim. Sci., Univ. Arizona, Tucson: incubation with amyloglucosidase for 30 min. dNutrition-Laboratory Services, Inc., Tolleson, AZ: incubation with Taka-Diastase (contains a-amylase and other amylolytic, proteolytic, and lipolytic enzymes from Aspergillus oryzae) . eHigh Plains Laboratory, Inc., Hereford, TX: loss of microscopic birefringence (percentage of starch granules gelatinized). yLinear effect of flake density ( P < .05). zQuadratic effect of flake density ( P < .05).
Downloaded from jas.fass.org by on October 21, 2009.

FLAKE DENSITY OF STEAM-PROCESSED SORGHUM

1059

Table 3. Correlations (R ) among starch availability measurements for sorghum grain steam-flaked to different densitiesa
Measurement Flake density, g/L 1.00 .88 .87 .93 .36 Enzyme method 1a, mg/g .88 1.00 .90 .90 .36 Enzyme method 1b, % .87 .90 1.00 .93 .39 Enzyme method 2, mg/g .93 .90 .93 1.00 .46 Percent gelatinization .36 .36 .39 .46 1.00

Measurement Flake density, g/L Enzyme method 1ab, mg glucose/g grain DM Enzyme method 1bb, percentage starch hydrolysis Enzyme method 2c, mg glucose/g grain DM Birefringence methodd percentage gelatinization

and other amylolytic, proteolytic, and lipolytic enzymes from Aspergillus oryzae) . dHigh Plains Laboratory, Inc., Hereford, TX: loss of microscopic birefringence (percentage of starch granules gelatinized).

aNumber of observations per measurement = 32 (four flake densities eight sampling periods). bDept. Anim. Sci., Univ. Arizona, Tucson: incubation with amyloglucosidase for 30 min. cNutrition-Laboratory Services, Inc., Tolleson, AZ: incubation with Taka-Diastase (contains a-amylase

processing (decreasing FD from SF32 to SF20) was a linear ( P < .05) decrease in DMI, expressed either as kilograms per day or as a percentage of BW. Depression in DMI was almost twice as great when steers were fed finishing diets (13%) than when steers were fed growing diets (7%). Xiong et al. (1991), Reinhardt et al. (1997), and Theurer et al. (1999) also reported linear decreases in DMI by feedlot steers as FD of steam-flaked sorghum decreased. In the finishing phase and for the entire trial, ADG decreased linearly ( P < .05) as FD decreased, which resulted in a linear decrease ( P < .05) in final live weights. In the growing phase, the efficiency (DMI/gain) response to decreasing FD was a curvilinear decrease (quadratic, P < .05), with the lowest value (most efficient) for SF28 and the highest value (least efficient) for SF32. The efficiency responses in the finishing phase and overall trial followed a similar pattern ( P > . 10). The major change in efficiencies in the growing and finishing phases and the 231-d trial occurred between SF32 and SF28; efficiencies for SF28, SF24, and SF20 were very similar. The major change in gain/DMI in the study by Xiong et al. (1991) was between SF34 and SF28. The response pattern of ADG and efficiency in other FD studies is not consistent. Xiong et al. (1991) reported that lowering FD (SF34 to SF22) resulted in a linear increase in efficiency by feedlot steers, but no change in ADG. Reinhardt et al. (1997) reported linear decreases in daily gain, but no changes in efficiency, with decreasing sorghum FD (SF28 to SF22). Theurer et al. (1999) noted a numerical decrease in ADG ( 4%) and improvement in efficiency (+4%) by steers as density of steam-flaked sorghum decreased from SF30 to SF20. Estimated NEm and NEg values for the growing and finishing diets (Table 4 ) also showed a tendency for a curvilinear response to FD, although the quadratic component was statistically significant ( P < .05) only for the growing diet. The NEm and NEg concentrations for the growing diet were highest for SF28; for the finishing diet, NEm and

NEg content was lowest for SF32. As with DMI/gain, the largest changes in estimated NE concentrations for both diets were between SF32 and SF28. Ratios of diet NE values estimated from performance to NE values calculated from the NRC (1984) for the growing diet were greater than 1:1 for all FD (average = 1.045), with the highest values for SF28 and SF24 (Table 4). In contrast, estimated/calculated NE ratios for finishing diets were less than 1:1 for all FD (average = .968), with the lowest values for SF32. It seems likely that the feedlot performance responses to increased degree of grain processing are influenced by level of roughage in the diet. Estimated NE intakes were highest for SF28 and lowest for SF20, when steers were fed the growing and finishing diets (data not shown). Mean NEg values for SF32, SF28, SF24, and SF20, respectively, for the growing diet were 6.0, 6.2, 5.8, 5.4 Mcal/d and for the finishing diet were 5.8, 6.0, 5.7, 5.1 Mcal/d. Thus, NEg intakes seemed to decrease in a more curvilinear manner than DMI, which was linear, with decreasing FD. With the exception of carcass weight, which decreased linearly ( P < .05) in response to increased extent of processing, carcass characteristics were not influenced by FD (Table 5). Percentage of pelvic, kidney, and heart fat tended to decrease ( P < .10) with decreasing FD. The percentage of carcasses grading low choice or better (data not shown) was highest (53%) from steers fed SF32, SF28, and SF24, and lowest (31%) for steers fed SF20 ( P = .10). In the present study, this could have been related to differences in carcass weight, but Xiong et al. (1991) also observed a tendency for improved choice percentage when steam-flaked sorghum grain with a heavier FD (SF34 vs SF28 or SF22) was fed. Theurer et al. (1999) reported that dressing percentage and carcass fat thickness decreased linearly as FD decreased from SF30 to SF20. Reinhardt et al. (1997) also found the same dressing percentage response, but reported no differences in carcass merit due to varying sorghum FD.

Downloaded from jas.fass.org by on October 21, 2009.

1060

SWINGLE ET AL.

The greatest changes in nutrient digestibilities by steers in the feedlot occurred as FD was decreased from SF32 to SF28, with both diets (Table 6). Fecal starch concentrations decreased linearly and apparent digestibilities of starch increased linearly in response to increased degree of grain processing in the growing and finishing periods ( P < .05). Fecal starch was twice as high for SF32 when steers were fed the growing compared to the finishing diet, resulting in starch digestibilities of 92 and 98%, respectively. With high roughage diets, SF32 may not be adequately processed to appreciably improve starch digestibility over that of dry-rolled sorghum (90 and 87% for Theurer, 1986, and Huntington, 1997, respectively). Decreasing FD

decreased linearly ( P < .05) the NDF digestibilities in the finishing diet only. Digestibilities of DM, CP, and GE increased linearly ( P < .05) with decreased FD in the growing diet, but not the finishing diet. The DE (Mcal/kg) increased linearly with decreasing FD for the growing diet, but decreased linearly with the finishing diet ( P < .05). The latter was due to the lowest GE digestibility for SF20 and the very low digestibilities of NDF for SF24 and(or) SF20. Characteristically, electrical energy requirements (data not shown) for processing increased linearly ( P < .05) as FD decreased (20.4, 22.0, 26.2, and 30.6 amps for SF32, SF28, SF24, and SF20, respectively). Electrical energy requirements for processing SF20 vs

Table 4. Feedlot performance of crossbred steers fed diets containing sorghum grain steam-flaked to different densities (Trial 1)
Flake densitya Item Pen replicates, n Live weight, kg Initialb 112-db Finalcy DMI, kg/d 1112 dy 113231 dy 1231 dy DMI, % BW 1112 dy 113231 dy 1231 dy Daily gain, kg 1112 d 113231 dy 1231 dy DMI/gain 1112 dz 113231 d 1231 d Diet NEm, Mcal/kg 1112 dz 113231 d Diet NEg, Mcal/kg 1112 dz 113231 d NE estimated/calculatedd NEm 1112 d 113231 d NEg 1112 d 113231 d SF32 5 182 354 486 7.72 7.91 7.82 2.88 1.89 2.34 1.54 1.11 1.31 5.03 7.15 5.95 1.83 1.98 1.21 1.34 SF28 5 181 357 488 7.50 7.59 7.54 2.78 1.80 2.26 1.57 1.10 1.32 4.78 6.95 5.70 1.91 2.09 1.28 1.43 SF24 5 182 353 477 7.36 7.27 7.32 2.76 1.75 2.18 1.52 1.05 1.28 4.84 6.91 5.71 1.89 2.11 1.25 1.45 SF20 5 180 344 463 7.18 6.91 7.04 2.75 1.71 2.19 1.46 1.00 1.22 4.92 6.92 5.75 1.85 2.11 1.21 1.43 SEM 1 4 6 .12 .13 .11 .03 .03 .02 .03 .03 .03 .06 .19 .08 .02 .06 .02 .05

1.01 .93 1.03 .91

1.06 .98 1.09 .98

1.04 .99 1.08 .99

1.02 .98 1.03 .98

aSF32, SF28, SF24, and SF20 = steam-flaked at densities of 412, 360, 309, and 257 g/L (32, 28, 24, and 20 lb/bu), respectively. bLive weights reduced 4% to account for gastrointestinal fill. cFinal live weight (231-d) calculated as hot carcass weight/.647. dRatio of NE estimated from performance to NE calculated from NRC (1984). yLinear effect of flake density ( P < .05). zQuadratic effect of flake density ( P < .05).

Downloaded from jas.fass.org by on October 21, 2009.

FLAKE DENSITY OF STEAM-PROCESSED SORGHUM

1061

Table 5. Carcass measurements of crossbred steers fed diets containing sorghum grain steam-flaked to different densities (Trial 1)
Flake densitya Item Pens (steers), n Hot carcass weight, kgfg Dressing percentageb Fat thickness, cm Ribeye area, cm2 KPHc, %h Marbling scored USDA Quality Gradee USDA Yield Grade SF32 5 314.4 64.7 1.27 78.9 3.0 5.14 9.24 3.08 SF28 5 315.6 64.9 1.34 78.5 2.8 4.94 8.76 3.12 SF24 5 308.8 64.5 1.31 81.4 2.7 5.07 9.28 2.88 SF20 5 299.8 64.7 1.35 78.0 2.8 4.87 8.60 3.06 SEM 3.8 .2 .07 1.3 .08 .16 .29 .08

aSF32, SF28, SF24, and SF20 = steam-flaked at densities of 412, 360, 309, and 257 g/L (32, 28, 24, and 20 lb/bu), respectively. b(Hot carcass weight/live weight) 100. cKidney, pelvic, and heart fat, percentage of carcass weight. dCoded: 4.00 = slight 00, 5.00 = small 00. eCoded: 9.00 = select 00, 10.00 = choice 00. fLinear effect of flake density ( P < .05). gQuadratic effect of flake density ( P < .05). hQuadratic effect of flake density ( P < .10).

SF32 increased by 50%, but these requirements for processing SF28 vs SF32 increased by only 8%. Other studies have demonstrated increased costs as FD is decreased (Karr, 1984; Xiong et al., 1990a; Reinhardt et al., 1997). In the present study, daily gains were not decreased when steers were fed SF28 compared to SF32. Considering increased processing costs, SF28 would be the most efficient FD, because the greatest improvements in feed conversion occurred when FD was reduced from SF32 to SF28. In the present study, starch availability (measured by laboratory methods, and as evidenced by decreased fecal starch concentrations and higher digestibility of starch) increased linearly as FD of sorghum grain

decreased, but electrical energy use for processing also increased linearly. The DMI and ADG decreased linearly as degree of processing increased, resulting in lighter final live weights. Feed efficiency and estimated diet NE responses to increased grain processing tended to follow a curvilinear pattern, being improved most when FD was decreased from SF32 to SF28. Reduced feed efficiency and lower estimated NE values for the growing diet with SF20, as compared with SF28 and SF24, is evidence that sorghum grain can be over processed. Roughage source and level may be major factors that influence the performance response by feedlot cattle to differences in degree of steam-flaking of sorghum grain.

Table 6. Dry matter intake, fecal starch concentrations, and nutrient digestibilities by steers fed diets containing sorghum grain steam-flaked to different densities (feedlot conditions) (Trial 1)
Growing dieta Item DMI, kg/d Fecal starch, % DM Digestibility, % DM Starch CP NDF GE DE, Mcal/kg SF32 7.7 7.6 55.8 91.6 49.1 21.5 53.9 2.24 SF28 7.5 3.6 59.6 96.2 54.2 21.8 58.7 2.43 SF24 7.2 2.7 59.3 97.2 51.0 23.1 59.6 2.52 SF20 7.1 2.3 60.8 97.7 53.5 22.5 60.5 2.54 SEM .1bcy .3yz .9by .4bcyz .8bdy 2.6 1.1by .05by SF32 8.0 3.8 67.9 98.0 43.1 15.1 67.3 2.81 SF28 7.5 2.8 67.7 98.5 46.6 7.2 67.9 2.82 Finishing dieta SF24 7.3 2.4 66.3 98.6 45.1 1.2 66.7 2.77 SF20 6.9 2.1 64.3 98.8 46.1 1.5 65.1 2.69 SEM .3b .3y 1.3 .2by 2.2 4.9by .9 .04y

aSF32, SF28, SF24, and SF20 = steam-flaked bSF32 vs SF28, SF24, and SF20 ( P < .05). cSF28 vs SF24, and SF20 ( P < .05). dSF24 vs SF20 ( P < .05). yLinear effect of flake density ( P < .05). zQuadratic effect of flake density ( P < .05).

at densities of 412, 360, 309, and 257 g/L (32, 28, 24, and 20 lb/bu), respectively.

Downloaded from jas.fass.org by on October 21, 2009.

1062

SWINGLE ET AL.

Trial 2. Nutrient Digestibilities in the Rumen and Intestines

Increasing the extent of steam-processing increased degradability of starch granules in the sorghum grain. Enzymatic starch hydrolysis (as a percentage of total starch) in steam-processed sorghum grain (used in the growing and finishing diets) increased linearly ( P < .05) as FD decreased from SF32 to SF20 (data not shown). Rates of hydrolysis were less (71%; P < .05) for SF32 than for SF28, SF24, and SF20 (79, 85, and 90%, respectively), and values for SF28 were less ( P < .05) than for SF24 and SF20 (SEM = 1.5). The percentage starch hydrolyzed in 30 min for steamflaked grain was about three times that of the unprocessed grain (81 vs 28%). These results are consistent with Trial 1 (Table 2 ) and with other reports (Osman et al., 1970; Frederick et al., 1973; Xiong et al., 1990a,b, 1991; Theurer et al., 1999). Nutrient analyses of growing and finishing diets fed in the digestibility study are shown in Table 1. Daily intakes of DM and starch by steers were not altered by FD with the growing diets; however, these intakes were lower ( P < .05) with the finishing diet for SF32 compared to lighter FD (Table 7). The latter response is in contrast to performance trials (Xiong et al., 1991; Reinhardt et al., 1997; Theurer et al., 1999; Trial 1 ) and one other digestibility study (Theurer et al., 1999), in which decreasing FD depressed DMI, and the highest FD had the greatest DMI. Except for starch flows and digestibilities, most of the digestibility responses to FD were with the finishing diet (Table 7). Digestibilities of DM were not altered by extent of processing in the growing diet; however, ruminal plus small intestinal and total digestibilities (percentage of intake) were increased linearly ( P < .05) as FD was decreased in the finishing diet. Flow of starch to the duodenum ( P < .05), terminal ileum ( P < .08), and feces ( P < .05) decreased linearly as FD decreased in both diets (Table 7). Thus, ruminal, ruminal plus small intestinal, and total starch digestibilities (percentage of intake) for both diets increased linearly ( P < .05) in response to reduced FD. Starch flow for steers fed SF32 was about double that of steers fed SF20. Flows and digestibilities of starch were different ( P < .05) between SF32 and the other FD, with the greatest changes often occurring between SF32 and SF28. Ruminal starch digestion ranged from 82 to 92% (both diets) and seems somewhat greater than average published values of 80% (range = 71 to 89%) for steers fed diets containing 75 to 84% steam-flaked sorghum grain (Theurer, 1986; Theurer et al., 1996; Huntington, 1997). When expressed as a percentage of starch entering into each segment, digestibilities of starch within the small intestine (77 and 71%) and large intestine (70 and 55%) of steers fed growing and finishing diets,

respectively, were not altered by FD (Table 7). Data from the study by Zinn (1991) showed that digestibility of starch as a percentage of entry into the small and large intestines was 85 and 66%, respectively, for steers fed a diet containing 75% steam-flaked sorghum (SF28). Theurer et al. (1996) summarized 12 digestibility trials and concluded that increasing ruminal starch digestibility from 65 to 86% resulted in increased starch digestibility (percentage of entry) within the small intestine from 58 to 72%, and total starch digestibility (percentage of intake) from 91 to 98%. Total tract starch digestibilities were very high in this current study (Table 7; ranges = 98.9 to 99.4% and 97.5 to 98.9% for growing and finishing diets, respectively). Total starch digestibilities were also very high (range = 98.7 to 99.6%) when steers were fed a finishing diet similar to the present study (Theurer et al., 1999). In the summary by Theurer et al. (1996), total tract starch digestibilities were not well correlated ( r 2 = .21) with feed efficiency, but starch digestibility within the rumen was positively correlated ( r 2 = .54) with improved efficiency by feedlot steers. Ruminal plus small intestinal and total tract digestibilities of CP increased linearly ( P < .05) as FD decreased in the growing and finishing diets (Table 7). The SF32 treatment had the lowest ( P < .05) CP digestibilities compared with the mean of the other density treatments (i.e., total digestibility was 59 vs 63% for the growing diet, and 53 vs 59% for the finishing diet). In another study (Theurer et al., 1999), total CP digestibility followed a curvilinear pattern (SF25 was greater than SF30 or SF20) with decreasing FD when steers were fed a finishing diet similar to the present study. In addition, CP digestibility was increased linearly or not altered as FD decreased when steers were fed diets with cottonseed hulls or alfalfa hay, respectively, in separate experiments. Total tract CP digestibilities for steam-flaked sorghum diets in the present study were lower than reported by Rahnema et al. (1987a,b), but comparable to those reported by Husted et al. (1966) and Mehen et al. (1966). For both diets, digestibilities of NDF and ADF in the total tract were not affected by FD; however, fiber digestibilities in the rumen and rumen plus small intestine were often greater ( P < .05) for SF28 than for SF24 and SF20. Fiber digestibilities in the rumen and rumen plus small intestine were quite variable (and sometimes not possible), perhaps because the forage passed through the tract differently than the chromium. It is not known why ruminal NDF and ADF digestibilities by steers fed the finishing diet with the SF24 treatment were so low ( 10%), only about one-half (or less) that of the other FD treatments. Theurer et al. (1999) reported that total tract digestibilities for NDF and ADF decreased as FD decreased (SF30 to SF20) when

Downloaded from jas.fass.org by on October 21, 2009.

FLAKE DENSITY OF STEAM-PROCESSED SORGHUM

1063

Table 7. Nutrient intakes and digestibilities by steers fed diets containing sorghum grain steam-flaked to different densities (Trial 2)
Growing dieta Item DM Intake, g/d Flow, g/d Duodenum Ileum Feces Digestibility, % of intake Rumen Rumen + small intestine (SI) Total tract Starch Intake, g/d Flow, g/d Duodenum Ileum Feces Digestibility, % of intake Rumen Rumen + SI Total tract Digestibility, % of entry into intestine Small intestine Large intestine CP Intake, g/d Flow, g/d Duodenum Ileum Feces Digestibility, % of intake Rumen Rumen + SI Total tract NDF Intake, g/d Flow, g/d Duodenum Ileum Feces Digestibility, % of intake Rumen Rumen + SI Total tract ADF Intake, g/d Flow, g/d Duodenum Ileum Feces Digestibility, % of intake Rumen Rumen + SI Total tract SF32 6,790 4,377 2,686 2,222 35.0 60.4 67.2 2,771 466 115 30 83.0 95.8 98.9 SF28 6,706 4,172 2,607 2,218 37.9 61.1 67.1 2,672 290 52 17 89.4 98.0 99.4 SF24 6,771 4,398 2,889 2,267 34.2 56.7 66.4 2,742 302 64 18 88.8 97.6 99.3 SF20 6,814 4,233 2,862 2,334 37.7 58.1 65.9 2,753 232 56 17 91.6 98.0 99.4 SEM 123 195 151 114 2.5 2.4 1.2 77 66bx 8byz 2.4by 2.2by .3byz .1byz SF32 6,228 3,483 1,995 1,520 43.7 67.7 75.5 3,829 694 221 95 81.6 94.1 97.5 SF28 6,473 3,223 1,775 1,440 49.5 72.3 77.5 4,001 466 140 74 87.9 96.3 98.1 Finishing dieta SF24 6,413 3,410 1,898 1,333 46.0 70.2 79.1 3,934 534 169 46 86.2 95.6 98.8 SF20 6,368 3,215 1,800 1,350 49.3 71.7 78.8 3,984 371 94 44 90.6 97.6 98.9 SEM 67b 121 58b 86 1.4bx .6byz 1.2cx 41b 84by 38cx 15by 2.2by 1.0cx .4by

72.9 73.5 986 1,207 410 400 23.3 58.5 59.3 1,975 1,433 1,383 1,298 27.4 30.1 34.4 1,338 1,069 988 979 20.2 26.4 26.9

79.8 66.6 1,006 1,130 383 381 12.6 61.8 62.3 1,862 1,472 1,368 1,301 21.0 26.4 30.2 1,292 1,086 987 946 15.6 22.7 26.7

78.7 68.2 984 1,167 384 371 19.5 60.8 62.1 1,900 1,463 1,530 1,351 21.4 16.7 28.7 1,299 1,048 965 968 24.7 27.4 27.8

76.1 70.3 996 1,154 353 366 16.3 64.5 63.3 1,882 1,477 1,533 1,357 21.1 18.5 28.1 1,252 1,167 1,062 988 15.7 15.2 21.5

4.0 2.4 15 40 11by 16 3.1 1.2by 1.3x 66.5 98 98 79 5.7 7.0 3.3 51 113 51 61 5.6 4.8 3.0

70.2 55.5 796 1,015 327 375 28.4 58.5 52.8 1,006 775 854 662 21.3 14.1 32.8 616 446 517 425 26.3 15.3 29.5

71.1 46.9 825 975 315 348 20.3 61.1 57.2 1,074 749 767 649 28.9 27.3 38.2 661 478 454 423 26.6 30.4 35.1

69.4 70.1 798 960 309 325 21.8 60.7 59.2 1,039 942 936 613 7.0 8.7 40.4 642 561 523 390 9.8 17.5 38.8

74.5 47.9 845 1,036 279 320 23.2 67.0 62.0 963 775 818 590 18.9 14.5 38.8 620 496 491 393 19.8 20.3 36.5

5.0 6.4 13ex 24z 9by 11by 3.7 1.2bey 1.5by 21dez 56 51 64 4.2d 3.1d 5.3 17 41 23 38 5.4 2.8d 4.6 (continued)

Downloaded from jas.fass.org by on October 21, 2009.

1064

SWINGLE ET AL.

Table 7 (continued). Nutrient intakes and digestibilities by steers fed diets containing sorghum grain steam-flaked to different densities (Trial 2)
GE Intake, Mcal/d Total tract digestibility, % DE, Mcal/kg diet
aSF32, b,cSF32

28.0 65.8 2.71

27.8 65.2 2.70

27.9 64.7 2.67

28.4 63.8 2.65

.6 1.1 .03

25.9 72.9 3.03

26.5 75.1 3.08

26.4 77.5 3.19

26.0 75.9 3.11

.2cz 1.3c .06

SF28, SF24, and SF20 = steam-flaked at densities of 412, 360, 309, and 257 g/L (32, 28, 24, and 20 lb/bu), respectively. vs SF28, and SF24, and SF20 ( P < .05 and P < .09, respectively). dSF28 vs SF24 and SF20 ( P < .05). eSF24 vs SF20 ( P < .05). x,yLinear effect of flake density ( P < .09 and P < .05, respectively). zQuadratic effect of flake density ( P <.05).

steers were fed diets containing alfalfa hay, or wheat straw and alfalfa (same diet as current study). This pattern was reversed when steers were fed the diet containing cottonseed hulls. The effect of FD on NDF and ADF digestibilities may differ with sources of dietary roughages. Digestibilities of GE did not differ with FD when steers were fed the growing diet, but were increased quadratically ( P < .05, SF32 was lowest) with decreasing FD when finishing diets were fed. Dietary concentrations of DE were not affected by FD (2.68 and 3.10 Mcal/kg DM for the growing and finishing diets, respectively). With the finishing diet, DE was lowest for SF32 and greatest for SF24, which agrees with the estimated diet NE in the performance trial. Total tract digestibilities by intact feedlot steers (Trial 1 ) for DM, CP, NDF, and GE were considerably lower than digestibilities determined with cannulated steers (Trial 2). The effect of FD on fecal starch concentrations and total starch digestibilities were the only measurements that were consistent for both diets in Trials 1 and 2. These differences may reflect the 9 to 16% higher DMI for the intact steers compared to the cannulated steers, and the sampling techniques. Twelve fecal samples were obtained in Trial 2 at various hours over a 3-d period, but obtaining fecal samples for only one day under feedlot conditions (Trial 1 ) may not provide accurate digestion coefficients for DM, CP, fiber, and energy. Nutrient digestibilities of starch and protein in the total tract of steers in the more controlled conditions of Trial 2 agree more closely with previous values for steamflaked sorghum grain (Theurer, 1986; Rahnema et al., 1987a,b; Theurer et al., 1996; Huntington, 1997) than do digestibility values obtained under feedlot conditions (Trial 1). In conclusion, decreasing FD of steam-processed grain in the present digestion studies had the greatest effects on starch and protein digestibilities by steers fed the growing and finishing diet. The most consistent responses were for starch digestibilities. The simultaneous increase in digestibilities of starch and

CP at various sites within the digestive tract when sorghum grain is extensively processed is probably due to the disruption of the protein matrix associated with the starch granules within the kernel (Rooney and Pflugfelder, 1986). This disorganization of protein matrix and starch granules allows increased susceptibility of the protein and starch to rumen microbial, pancreatic, and intestinal enzymes. As density of the grain was reduced from SF32 to SF20, values were increased linearly for starch digestibilities in the rumen by 10%, rumen plus small intestine by 3%, and total tract by 1%. Greater ruminal starch digestion can result in improved feed conversions and net absorption of energy (Theurer et al., 1996). Digestibilities of CP in the rumen plus small intestine and total tract were linearly increased by 8 and 15%, respectively, for steers fed both diets. Because greatest improvements in starch and CP digestibilities occurred when density was lowered from SF32 to SF28, optimal sorghum grain FD was SF28.

Implications
The optimal flake (bulk) density for steamprocessed sorghum grain appears to be 360 g/L (28 lb/ bu), based on performance by feedlot steers, grain processing costs, and diet digestibilities of protein and starch. Further decreases in flake density will not appreciably increase net energy value of the grain or improve feed conversions. Flaking to very light densities (i.e., 257 g/L or 20 lb/bu) may result in over processed grain when compared to flake densities of 360 and 309 g/L (28 and 24 lb/bu), resulting in inferior feedlot performance. Decreasing flake density will consistently increase ruminal and total starch digestibilities by steers. Enzymatic laboratory methods commonly used to evaluate starch availability (hydrolysis) are effective in establishing flake density criteria for quality control of the steam-flaking process. Flake density is more closely related to enzymatic methods than to percentage gelatinization.

Downloaded from jas.fass.org by on October 21, 2009.

FLAKE DENSITY OF STEAM-PROCESSED SORGHUM

1065

Literature Cited
AOAC. 1984. Official Methods of Analysis (14th Ed.). Association of Official Analytical Chemists, Washington, DC. Frederick, H. M., B. Theurer, and W. H. Hale. 1973. Effect of moisture, pressure and temperature on enzymatic starch degradation of barley and sorghum grain. J. Dairy Sci. 56: 595601. Goering, H. K., and P. J. Van Soest. 1970. Forage fiber analysis (apparatus, reagents, procedures, and some applications). Agric. Handbook No. 379. ARS, USDA, Washington, DC. Hale, W. H. 1973. Influence of processing on the utilization of grains (starch) by ruminants. J. Anim. Sci. 37:10751080. Huntington, G. B. 1997. Starch utilization by ruminants: From basics to the bunk. J. Anim. Sci. 75:852867. Husted, W. T., W. H. Hale, and B. Theurer. 1966. Digestibility of dry-rolled, steam-processed-flaked, steam-processed-cut and water-soaked milo by steers. In: Ariz. Cattle Feeders Day Rep. pp 4749. Univ. of Arizona, Tucson. Karr, K. J., K. McLeod, and C. R. Richardson. 1987. Enzymatic starch determination utilizing amperometric measurements of glucose. In: Texas Tech. Univ. Agric. Sci. Tech. Rep. No. T5-233. p 80. Lubbock. Karr, M. K. 1984. Steam-flaking techniques and quality control. In: Proc. Feed Grain Utilization Symposium, Texas Tech Univ., Lubbock. pp 99102. Mehen, S., W. H. Hale, B. Theurer, and O. M. Little. 1966. Digestibility of dry-rolled, fine-ground, steam-processed and pressurecooked milo by steers. In: Ariz. Cattle Feeders Day Rep. pp 43. Univ. of Arizona, Tucson. NRC. 1984. Nutrient Requirements of Beef Cattle (6th Revised Ed.). National Academy Press, Washington, DC. Osman, H. F., B. Theurer, W. H. Hale, and S. M. Mehen. 1970. Influence of grain processing on in vitro enzymatic starch digestion of barley and sorghum grain. J. Nutr. 100:11331140. Poore, M. H., T. P. Eck, R. S. Swingle, and C. B. Theurer. 1989. Total starch and relative starch availability of feed grains. Abstr. #35. 20th Bienn. Conf. Rumen Function, Chicago, IL. Poore, M. H., J. A. Moore, T. P. Eck, R. S. Swingle, and W. H. Brown. 1991. Wheat straw or alfalfa hay in diets with 30% neutral detergent fiber for lactating Holstein cows. J. Dairy Sci. 74:31523159. Rahnema, S. H., B. Theurer, J. A. Garcia, W. H. Hale, and M. C. Young. 1987a. Site of protein digestion in steers fed sorghum grain diets. II. Effect of grain processing methods. J. Anim. Sci. 64:15411547.

Rahnema, S. H., B. Theurer, J. A. Garcia, M. C. Young, and R. S. Swingle. 1987b. Site of protein digestion in steers fed sorghum grain diets. 1. Effect of level of feed intake. J. Anim. Sci. 64: 15331540. Reinhardt, C. D., R. T. Brandt, Jr., K. C. Behnke, A. S. Freeman, and T. P. Eck. 1997. Effect of steam-flaked sorghum grain density on performance, mill production rate, and subacute acidosis in feedlot steers. J. Anim. Sci. 75:28522857. Robertson, J. B., and P. J. Van Soest. 1981. The detergent system of analysis and its application to human foods. In: W.P.T. James and O. Theander (Ed.) Analysis of Dietary Fiber in Food, Vol. 3. Marcel Dekker, Inc., New York. Rooney, L. W., and R. L. Pflugfelder. 1986. Factors affecting starch digestibility with special emphasis on sorghum and corn. J. Anim. Sci. 63:16071623. SAS. 1985. Users Guide: Statistics (Version 5 Ed.). SAS Inst. Inc., Cary, NC. Steel, R.G.D., and J. H. Torrie. 1980. Principles and Procedures of Statistics. McGraw-Hill Book Co., New York. Swingle, R. S. 1992. Impact of sorghum grain flake density on feedlot performance. In: Proc. Southwest Nutr. Manage. Conf., Scottsdale, AZ. Univ. of Arizona, Tucson. pp 5563. Theurer, C. B. 1986. Grain processing effects on starch utilization by ruminants. J. Anim. Sci. 63:16491662. Theurer, C. B., O. Lozano, and A. Alio. 1996. Ruminal vs intestinal starch digestion affects nutrient use and efficiency of gain in feedlot cattle. In: Proc. Southwest Nutr. and Manage. Conf., Phoenix, Arizona. Univ. of Arizona, Tucson. pp 97108. Theurer, C. B., R. S. Swingle, R. C. Wanderley, R. M. Kattnig, A. Urias, and G. Ghenniwa. 1999. Sorghum grain flake density and source of roughage in feedlot cattle diets. J. Anim. Sci. 77: 10661073. Xiong, Y., S. J. Bartle, and R. L. Preston. 1990a. Improved enzymatic method to measure processing effects and starch availability in sorghum grain. J. Anim. Sci. 68:38613870. Xiong, Y., S. J. Bartle, R. L. Preston, and Q. Meng. 1990b. Estimating starch availability and protein degradation of steam-flaked and reconstituted sorghum grain through a gas production technique. J. Anim. Sci. 68:38803885. Xiong, Y., S. J. Bartle, and R. L. Preston. 1991. Density of steamflaked sorghum grain, roughage level, and feeding regimen for feedlot steers. J. Anim. Sci. 69:17071718. Zinn, R. A. 1990. Influence of flake density on the comparative feeding value of steam-flaked corn for feedlot cattle. J. Anim. Sci. 68:767775. Zinn, R. A. 1991. Comparative feeding value of steam-flaked corn and sorghum in finishing diets supplemented with or without sodium bicarbonate. J. Anim. Sci. 69:905916.

Downloaded from jas.fass.org by on October 21, 2009.

Citations

This article has been cited by 5 HighWire-hosted articles: http://jas.fass.org#otherarticles

Downloaded from jas.fass.org by on October 21, 2009.