Effects of Diet Particle Size on Pig Growth Performance, Diet Flow Ability, and Mixing Characteristics

Transcription

1 Effects of Diet Particle Size on Pig Growth Performance, Diet Flow Ability, and Mixing Characteristics Introduction Bob Goodband, Crystal Groesbeck, Mike Tokach, Steve Dritz, Joel DeRouchey, and Jim Nelssen Department of Animal Sciences and Industry, Kansas State University, Manhattan Cereal grains are the primary energy source in swine diets. Therefore, not only must producers be concerned about the composition of the grain, but also how it is processed so the pig can fully utilize the nutrients. Because feed represents 65 to 75 percent of overall production costs in a swine operation, improving the efficiency of feed utilization will have a tremendous impact on the cost of production. Not only is proper particle size reduction important for improving profitability, it will also have an effect on nutrient excretion and the impact of swine waste on the environment. Nearly all feed ingredients will be subjected to some type of particle size reduction. Particle size reduction increases the surface area of the grain, thus allowing for greater interaction with digestive enzymes. It also improves the ease of handling and the mixing characteristics. However, fine grinding will increase the energy costs of feed processing and may result in feed bridging, dust problems, and increase the incidence of gastric ulcers in swine. Bridging of bins and feeders has recently become a focus in production management as these out of feed events may be a stressor contributing to finishing pig disease. In addition to particle size reduction, all diets will be mixed to produce a uniform blend ensuring every bite of feed a pig takes will have the identical nutrition composition as the next. Inadequate mixing can result in uneven and reduced growth performance. Therefore the goals of this paper are to evaluate the effects of diet particle size on pig performance and diet flow ability, secondly, to evaluate factors affecting mixer uniformity, and lastly review procedures to measure both particle size and mixer uniformity on the farm. Particle Size and Pig Performance In the past, there has been confusion regarding the optimum particle size in swine diets. This was a result of broad generalizations classifying dietary particle size. In the past, terms like fine, medium, and coarse were used to define particle size. More precise methods for determining particle size have been developed based on the mean geometric diameter of particles (measured in microns) and the geometric mean standard deviation of the particles or their distribution (ASAE 1973). These measurements allow for more precise definition of particle size and allow us to make specific recommendations to optimize swine

2 performance. It is essential that producers understand the procedures used by various analytical labs in particle size analysis and how these slight variances might effect the interpretation of the results. Examples of these variations in particle size analysis include: different number of screens (i.e., 6 vs 13), addition of a flow agent, use of balls and brushes, and calculations used to determine particle size. In addition to different procedures to measure particle size, another factor complicating the data available on the effects of particle size on pig performance are the interactions between particle size and grain source as well as age of the pig. Grain source will be a variable because of different sizes and shapes of kernels. Therefore recommendations for screen sizes in hammer mills or roll corrugations or gap settings will vary with different grains. As for interactions among particle size and age of the pig, it appears the young pig does a better job of chewing its feed than growing-finishing hogs. The greatest potential for fine grinding to improve feed efficiency will be for finishing pigs. However, fine grinding or rolling will improve feed efficiency regardless of age. A study conducted at Kansas State University demonstrates the effects of particle size on starter pig performance. In the study, 192 pigs (initial weight 6.8 to 8.1 kg) were fed either corn- or sorghum-based diets (Table 1). The grains were either processed through a hammer mill equipped with an 3.2 mm (1/8-inch; 539 to 624 microns) or.3 mm (1/4-inch; 722 to 877 microns) screen. Table 1. Effect of Particle Size of Corn and Sorghum Based Diets on Starter Pig Performance a Grain Corn Mill Type Mean Particle size Average Daily gain (microns) (g) (g) Average daily feed Feed/ intake b gain b Hammer mill Sorghum Roller mill , Hammer mill Roller mill , a Ohh et al., Values represent means from 192 weanling pigs initially 6.8 to 8.1 kg with a final weight of approximately 23 kg. b Difference between hammer mill and roller mill (P <.05).

3 Each grain was rolled either fine (822 to 885 microns) or coarse (1,147 to 1,217 microns) by adjusting the roller gap and feeding rate. As expected, fine grinding or rolling reduced mean particle size of the diet. Although the differences in particle size did not affect average daily gain (ADG), feed efficiency (feed/gain) was improved by either fine grinding or rolling (Table 1). By pooling the data across grain type and processing method and classifying it based only on particle size, the improved feed efficiency appears to be a result of improved nutrient digestibility (Table 2). Table 2. Effect of Particle Size of Corn and Sorghum on Apparent Digestibilities a Particle size Digestibility, % (microns) Dry Matter Protein Energy Feed/gain < to 1, >1, a Adapted from Ohh et al., Healy et al. (1994) evaluated growth performance of pigs weaned at 21 days of age and fed starter diets in which the grain (corn and hard or soft endosperm milo) was ground to 900, 700, 500, or 300 microns (Table 3). These results confirm that reducing grain particle size had little improvement on ADG. Average daily feed intake (ADFI) decreased (linear, P < 0.08) as particle size of the diet was reduced. Feed efficiency (F/G) improved then became poorer (quadratic, P < 0.01) as particle size decreased suggesting an optimum particle size between 500 and 700 microns. Pigs fed grain ground to 500 microns had a 6 percent improvement in feed efficiency compared to those pigs fed diets containing grain ground to 900 microns. However, production rate (tons of grain ground per hour) was reduced 43 percent by decreasing particle size from 700 to 500 microns. Also important is the numerical trend for decreased ADG and ADFI and poorer F/G of pigs fed the diets containing grain ground to 300 microns. The decision on optimum diet particle size needs to include assessment of improvements in feed efficiency versus reductions in milling production. These and other data suggest a dietary particle size of approximately 700 microns to optimize both pig performance and milling efficiency.

4 Table 3. Effect of Diet Particle Size on Growth Performance of Starter Pigs a Particle Size, microns Item ADG, g ADFI, g b F/G c Production, rate, t/h a Adapted from Healy et al., Data represent means of pigs fed either corn or hard or soft endosperm milo ground to the respective particle sizes. b Linear effect of particle size (P <.08). c Quadratic effect of decreasing particle size (P <.01). Particle Size and Alternative Grains The type of grain in the diet also will influence the pig s response to particle size reduction. Studies with high-fiber feed ingredients like barley indicate fine grinding of these types of ingredients may greatly improve their feeding value. A study was conducted with finishing pigs fed diets containing barley ground through a hammer mill equipped with either a 3.2, 4.8, or 6.4 mm (1/8-, 3/16-, or 1/4-inch) screen or coarsely rolled (Table 4). Performance of pigs fed these diets was compared to those fed a diet containing milo ground through a 4.8-mm screen. Pigs fed the barley diet ground through the 3.2 mm screen had similar average daily gain and feed efficiency compared to pigs fed the milo diet. Daily gain and feed efficiency became poorer as particle size of the barley diets increased. These data indicate grinding of fibrous feed ingredients to approximately 700 microns improves their feeding value and may make them more attractive as substitutes for corn. Because of its high protein content and propensity to become floury, wheat presents some unique processing problems. If ground too fine, wheat can reduce feed intake. Recommendations for optimum particle size for wheat for use in swine diets should be coarser than corn or milo, between 800 and 900 microns. Roller mills with a differential drive produce a uniform particle size and fewer fines and may be suited for processing wheat in swine diets. Particle size data also may be confounded by the kernel size of the grain and the screen size or roller mill settings by which it is processed. For example, corn ground through a 3/16-inch screen will have a finer particle size than either milo or wheat because of its larger kernel size. Corn kernels must be fragmented before they can pass through a 4.8 mm (3/16-inch) screen opening; however, milo or wheat may fall through the opening intact because of their smaller kernel size.

5 It is difficult to make a specific recommendation for one screen for each type of grain; however, screen size should be adjusted to produce a mean particle size of 700 microns. In addition, other factors such as hammer mill revolutions per minute, tip speed, and the number of hammers also will affect what screen size is necessary to produce 700-micron feed. One of the disadvantages of fine grinding is the increased incidence of gastric ulcers. Finely ground feed is more fluid when mixed with the digestive secretions of the pig s stomach compared to a more coarsely ground feed. As a result, the acids in the stomach have a greater chance of coming into contact with and irritating the esophageal region of the stomach. The frequency of ulceration increases when particle size drops below 500 microns. Other disadvantages of fine grinding include bridging problems in bulk bins and feeders as well as increased dustiness of the feed. Table 4. Effect of Barley Particle Size in Finishing Diets a Grain: Milo Barley Barley Barley Barley Screen Size, mm: Rolled Item Particle Size, µm: ,146 2,200 ADG, kg 0.93 b 0.89 b 0.92 c 0.81 c 0.79 c ADFI, kg 3.14 b 2.93 c 2.81 c 2.94 c 2.94 c Feed/gain 3.39 b 3.32 b 3.58 c 3.65 c 3.72 c a Goodband and Hines, bc Means on the same row with different subscripts differ (P < 0.02). Influence of Particle Size in Lactation Diets Many producers believe that finely ground feeds are unpalatable and will decrease feed consumption by the lactating sow resulting in lower litter weaning weight and increased sow weight loss. In order to determine the appropriate particle size for the lactating sow, Wondra, 1993 compared lactating sow performance when fed diets with corn ground to 1,200, 900, 600, or 400 microns. Litter size, pig survivability, sow weight loss, and sow back fat loss were not influenced by particle size of the diet (Table 5). However, litter weaning weight gain increased linearly (P < 0.05) as particle size decreased from 1,200 to 400 microns. However, the largest response was observed by reducing particle size from 1,200 to 600 microns (9 percent increase in litter weight gain). Surprisingly, feed intake actually increased (linear, P < 0.05) as particle size was reduced suggesting finely ground feed does not decrease palatability of the diet. Digestibility of dry matter, energy, and protein in the diet all improved linearly as particle size decreased. The improvement in digestibility is the reason that litter weaning weights were improved as particle size was reduced.

6 Reduction in particle size is not without a cost. Stomach ulcers and keratinization are always a concern with finely ground diets. Keratinization is an indication of stomach irritation that may lead to ulcers. Stomachs were scored from 0 to 4 (normal to severe ulcer) and from 0 to 3 (none to severe keratinization). Ulceration and keratinization of the stomach was increased as particle size of the diet was reduced. Energy Consumption, kwh/t a Wondra, et al., 1992 Production Rate Energy Consumption Particle Size, microns Production Rate, t/h A previous study by Wondra et al. (1992) demonstrated mill production rates decrease and energy utilization increases as particle size is reduced (Figure 1). Considering these factors, diets for lactating sows should be ground to a similar mean particle size as growing-finishing pigs of approximately 700 microns. Figure 1. Effect of fine grinding on milling production and energy consumption a Table 5. Effects of Lactation Diet Particle Size on Sow and Litter Performance a Particle size, microns Item 1, SE Litter size, d Sow wt loss, kg Litter wt, kg Litter wt gain, kg b Feed intake, kg b DM digestibility, % Ulcer score Keratinization score a Wondra, b Linear effect of particle size (P <.05). Recently the results of Wondra et al. (1993) have been confirmed by Baudon et al. (2003). In this study sows were fed corn-soybean meal-based diets with targeted corn particle sizes of 1,500, 900, and 600 µm. Reducing particle size of the corn in lactation diets from 1,500 to 600 microns resulted in greater

7 ADFI and water usage (linear, P < 0.02), fewer days for return to estrus after weaning (linear, P < 0.04), and less loss of backfat (quadratic, P < 0.03;Table 6). Although, not significant (P = 0.15 or greater) trends for similar improvement were observed in pigs weaned per litter, piglet survivability, litter weaning weight, and litter weight gain. Intakes of DM, N, and GE by the sows were increased by 9, 4, and 7% and apparent digestibilities of DM, N, and GE were increased by 6, 5, and 7%, respectively, as particle size decreased from 1,500 to 600 µm (linear, P < 0.01). Finally, excretion of DM and N in the feces was decreased (linear, P < 0.01) by 178 g/d and 5 g/d, respectively, as particle size decreased. In conclusion, reducing particle size of corn did not affect litter performance but increased feed intake and digestibility of nutrients. Furthermore these data demonstrate the potential for proper feed processing to reduce nutrient excretion and lessen the environmental impact of swine production on the environment. Table 6. Effects of Corn Particle Size on Sow and Litter Performance a Particle size, µm P < Item 1, SE Linear Quadratic No. of observations Lactation BW loss, kg b - Lactation fat loss, mm ADFI, kg Water disappearance, gal/d Initial pigs/litter Pigs weaned/litter Survivability, % Litter weaning wt, kg Litter wt gain, kg Days to estrus a Adapted from Baudon et al., (2003). b Dashes indicate P = 0.15 or greater. Effects of Diet Particle Size of Feed Flow Ability It has been demonstrated that decreasing particle size to between 600 and 700 microns and adding fat to diets can improve pig performance and profitability. Limits to reducing grain particle size and amount of added fat are frequently based on the ability of the feed to flow through feed delivery systems and feeders. Type of grinding may also affect feed flow ability. Grain ground with a roller mill typically has a more uniform particle size and less particle variation

8 than that ground with a hammer mill. Therefore we recently conducted three experiments to evaluate the effects of particle size, added fat, mill type, and particle size standard deviation on flow characteristics of ground corn (Groesbeck at al., 2003). In Exp. 1, corn was ground with either a hammer mill or a roller mill to produce six samples with different particle sizes. The particle size for the corn ground with a roller mill ranged from 1,235 to 502 microns with standard deviation ranging from 1.83 to 2.03 (Table 7). Particle size for corn ground with a hammer mill ranged from 980 to 390 microns with standard deviation ranging from 2.56 to Soy oil was then added at 0, 2, 4, 6, and 8% to each sample. Flow ability was determined by measuring angle of repose (the maximum angle measured in degrees at which a pile of grain retains its slope). A large angle of repose represents a steeper slope and poorer flow ability. There was a threeway interaction (P < 0.05) between particle size, added fat, and mill type. Roller mill ground corn had better flow ability than hammer mill ground corn, and decreasing particle size and increasing added fat, increased angle of repose (Figure 2). However as particle size decreased and added fat increased, the differences between hammer mill and roller mill ground grain became less. Corn ground with a hammer mill without added fat had a similar angle of repose as the corn ground with a roller mill that had 6% added fat. Table 7. Particle Size and Standard Deviation (Exp 1)ª Sample # Roller Mill Hammer Mill (1.98) 984 (2.56) (1.83) 980 (2.52) (1.84) 931 (2.49) (2.03) 665 (2.49) (1.99) 477 (2.25) (1.97) 390 (2.12) ªValues represent the mean of one sample analyzed in duplicate. For both Exp. 2 and 3, batches of roller mill and hammer mill ground corn were sifted with a Ro-Tap tester through a stack of 13 screens. The material on top of each screen was then collected. Soy oil was added at 0, 4, and 8% to each sample. In Exp. 2, 5 roller mill samples were selected from different individual screens with mean particle size ranging from 1,415 to 343 microns and 5 hammer mill samples ranging from 1,382 to 333 microns (Table 8). All samples were selected from the ground corn remaining on top of the individual screens. By selecting samples this way, both roller mill and hammer mill samples had similar particle size standard deviation, ranging from 1.1 to 1.3. There was an interaction (P < 0.05) between particle size, added fat, and mill type. Increasing fat and decreasing particle size, increased angle of repose (Figure 3). However,

9 in fine ground hammer mill ground corn, the differences between amounts of added fat became less as particle size decreased, whereas in roller mill ground samples the differences were maintained. In roller mill ground grain samples, decreasing particle size had less of a negative impact on flow ability than in hammer mill ground grain. In Exp. 3, 4 roller mill and 4 hammer mill samples that were constructed from the previously collected grain (Table 9). All samples were constructed to have a similar mean particle size (641 to 679 microns) with increasing particle size standard deviation (1.62 to 2.27). There was no interaction (P > 0.10) between particle size standard deviation, added fat, and mill type. Increasing fat (P < 0.04) and particle size standard deviation (P < 0.001) decreased flow ability (Figure 4). These data suggest that the greater flow ability of roller mill ground corn compared to hammer mill ground corn appears to be a result of less particle size variation. However, with fine particle sizes other factors, such as particle shape, may also contribute to flow ability. Table 8. Particle Size and Standard Deviation (Exp 2)ª b Sample # Sieve # Roller Mill Hammer Mill (1.09) 1382 (1.12) (1.16) 952 (1.20) (1.17) 686 (1.20) (1.19) 484 (1.19) (1.20) 333 (1.29) ªValues represent the mean of one sample analyzed in duplicate. b Each sample was collected from the material on top of the screen, to minimize particle size standard deviation. Table 9. Particle Size and Standard Deviation (Exp 3) ab Sample # Roller Mill Hammer Mill (1.62) 662 (1.62) (1.89) 641 (1.88) (2.10) 653 (2.12) (2.22) 670 (2.27) ªValues represent the mean of one sample analyzed in duplicate. b Samples were constructed to have similar particle size, from the grain collected on top of the 13 screens.

10 Angle of Repose Particle Size x Mill Type x Added Fat (P<0.05) Particle Size (microns) Figure 2. Effect of Mill Type, Particle Size and Added Fat on the Flow Ability of Ground Corn. Angle of Repose Particle Size x Mill Type x Added Fat (P < 0.05) Particle Size (microns) Figure 3. Effect of Particle Size with Narrow Particle Size Standard Deviation (1.1 to 1.3) on Ground Grain Flow Ability.

11 Angle of Repose Particle size standard deviation effect, < Added fat effect, P < Particle Size Standard Deviation Figure 4. Effect of Particle Size Standard Deviation on Ground Grain Flow Ability. The average particle size equal to 665 microns with increasing particle size standard deviation. Legend for Figures 2, 3, and 4 0% added fat, hammer-milled 0% added fat, roller-milled 3% added fat, hammer-milled 3% added fat, roller-milled 6% added fat, hammer-milled 6% added fat, roller-milled Effects of Diet Mixing Time on Nursery Pig Performance While the importance of thoroughly mixing diets is often emphasized, little data is available to quantify the impact of inadequate mixing on pig growth performance. Therefore, a 28 d trial was conducted to evaluate the effects of mixing time on growth performance of nursery pigs (Groesbeck et al., 2005). Experimental treatments consisted of mixing a diet for 0, 30, 60, 120 or 330 s in a horizontal ribbon mixer. Diets were fed in two phases (d 0 to 14 and 14 to 28) with diets in both phases containing relatively high levels of low inclusion ingredients such as synthetic amino acids, zinc oxide, and phytase. Diets in phase 1 also contained 3.75% fish meal, and 15% dried whey. Eight samples were collected from the mixer at the completion of the respective mixing time for each batch of feed to determine a coefficient of variation (CV). Each bag (23 kg lb) was labeled (first to last) and sampled to determine the degree of mixing that occurred as feed was conveyed from the mixer to the bagger (Table 10). Each pen of pigs was then assigned a bag of feed. Bags were distributed in ordered bagged (1, 2, 3, etc.). As feed was needed, the next chronological bag of feed

12 was then added. Growth performance improved (linear, P < 0.01) in both phases. From d 0 to 28, increasing mix time increased (linear, P< 0.01) ADG and improved F:G (linear, P < 0.01, quadratic P < 0.07; Table 11). Although the greatest improvement in pig performance was observed from mixing 0 to 60 s, ADG and F:G continued to improve through 330 s. The mixer efficiency CV value for the 330 s mixing time was 7 and 12% for phase 1 and 2 respectively; therefore, the linear improvement in ADG and F:G indicates a low CV value is ideal for maximizing pig growth performance. With greater use of low inclusion ingredients such as synthetic amino acids in swine diets, these data demonstrate that inadequate mixing reduces nursery pig performance. Table 10. Coefficient of Variation (CV), % a Mixing Time, s Item D 0 to 14 Mixer b Bag c D 14 to 28 Mixer b Bag d a Quantab chloride titrators were used to determine CV on all samples. b Coefficient of variation was determined from eight samples collected from the mixer for each batch of feed. c The bag CV for each mix time was determined from 12 samples (one sample collected from each bag at the bagger). d The bag CV for each mix time was determined from 22 samples (one sample collected from each bag at the bagger). Particle Size and Mixer Efficiency Testing We recommend that stationary mills and feed mills check ground grain at least weekly with an on-farm 3-screen testing procedure ( and then send a sample to a commercial lab for testing with a full, 10 to 13 screen tester once a month. The three screen procedure has been demonstrated to be more accurate than a single screen test (Benz et al., 2005). Producers with portable grinding feed equipment may test less frequently (every other week and send in a sample every other month) because typically they are grinding less grain than stationary mills. Producers can select from several commercial labs and feed companies who routinely analyze feed for particle size. K-State also analyzes samples at a cost of $10 per sample. A small zip lock bag of ground grain is needed (no more than 0.5 kg or 1 lb) and samples can be sent to KSU Swine Lab, Room 206 Weber Hall, Department of Animal Science, Kansas State University, Manhattan, KS 66505, Attention: Particle Size. If you include your address results can be sent electronically.

13 Table 11. Effects of Inadequate Diet Mixing Time on Nursery Pig Performance ab Mixing Time, s SE Linear Quadratic Day 0 to 14 Initial wt,kg ADG, g ADFI, g F:G Day14 to 28 ADG, g ADFI, g F:G Day 0 to 28 Final wt,kg ADG, g ADFI, g F:G a Groesbeck et al. (2005). A total of 180 weanling pigs (average initial BW of 6.3 ±.8 kg, 21± 3 d of age) with 6 pigs/pen and 6 pens/treatment. b No cubic responses (P < 0.05) observed. All mixers should be tested upon purchase and on an annual schedule to ensure adequate performance of mixing equipment. Salt is an excellent ingredient to use for mixer efficiency testing because it is a relatively inexpensive and simple procedure. Salt can be analyzed with Quantab chloride titrators (Environmental Test System, Elkhart, IN) onsite or sent to a laboratory for analysis (ASAE Standard: S380, ASAE Standard: S303.3). Quantab analysis is comparable to both a salt meter test and laboratory analysis whereas analyses for other nutrients are variable. Results for other ingredient like Ca or P have been shown to have a large variability in analytical values between procedures and laboratories. Collecting samples from a mixer is a potentially dangerous procedure and all necessary precautions and lock-out procedures should be followed. It is recommended that 10 samples are taken from 10 predetermined locations in the mixer to allow for the collection of a representative sample of the feed. Samples should be collected with a grain probe and placed in pre-labeled containers. Because of the potential danger in sampling mixers

14 we recommend the samples be collected directly as the feed is discharged from the mixer or in some instances when it is augured into a bulk bin. The result for mixing efficiency testing is reported as coefficient of variation (CV). A CV of 10% for a mixer is considered excellent, a CV of 10 to 15% is an indicator of good mixing, but may also indicate that mixing time needs to be increased. A CV of 15 to 20% is fair, and a CV of 20% or greater warrants immediate attention. Conclusion While diet formulation is frequently emphasized by nutritionists, how the complete diet is ground and mixed can have equally important implications for pig growth performance. Quality control procedures should include frequent testing for particle size and corrective steps taken if results fall outside a range or 650 to 750 microns. Results suggest that this mean particle size will be optimum for finishing pigs and the breeding herd and will not significantly reduce feed mill capacity. Secondly, mixer efficiency testing should also be included in a quality control program. Results suggest a diet CV of 10% or less should be targeted, especially when diets contain numerous low inclusion ingredients such as, crystalline amino acids, concentrated vitamin and trace mineral premixes, and feed medications. Literature Cited ASAE Method of determining and expressing fineness of feed materials by sieving. ASAE standard S319. In: Agricultural Engineers Yearbook of Standard, ASAE. p 325. Baudon, E.C., J.D. Hancock, and M. D. Tokach Particle Size of Corn in Lactation Diets for Mixed-parity Sows. Kansaas State University Swine Industry Day Report of Progress. Benz, C. K., M. D. Tokach, C. N. Groesbeck, S. S. Dritz, J. L. Nelssen, R. D. Goodband, and J. M. DeRouchey A comparison of bygholm sieve to standard particle size analysis techniques. Kansas Swine Industry Day Report of Progress 964. Goodband, R.D. and R.H. Hines The effect of barley particle size on starter and finishing pig performance. J. Anim. Sci. 65(Suppl. 1):317. Groesbeck, C. N., R. D. Goodband, M. D. Tokach, S. S. Dritz, J. L. Nelssen, J. M. DeRouchey, and C. R. Neill Inadequate diet mixing time reduces nursery pig performance. Kansas Swine Industry Day Report of Progress 964.

15 Groesbeck, C.N., R.D. Goodband, M.D. Tokach, J.L. Nelssen, S.S. Dritz, K.R. Lawrence, and C.W. Hastad Particle size, mill type, and added fat influence flow ability of ground corn. Kansas Swine Industry Day Report of Progress 920. Healy, B.J., J.D. Hancock, G.A. Kennedy, P.J. Bramel-Cox, K.C. Behnke, and R.H. Hines Optimum particle size of corn and hard and soft sorghum for nursery pigs. J. Anim. Sci. 72:2227. Ohh, S.J., G.L. Allee, K.C. Behnke, and C.W.Deyoe Effects of particle size of corn and sorghum grain on performance and digestibility of nutrients for weaned pigs. J. Anim. Sci. 57(Suppl. 1): 260 (Abstr.). Wondra, K.J Effects of particle size, mill type, and diet form on performance of finishing pigs and lactating sows. M.S. Thesis. Kansas State University, Manhattan, KS