METABOLISM AND NUTRITION. Pearl Millet in Diets of White Pekin Ducks 1

Transcription

1 METABOLISM AND NUTRITION Pearl Millet in s of White Pekin Ducks 1 O. ADEOLA 2 and J. C. ROGLER Department of Animal Science, Purdue University, West Lafayette, Indiana 7907 T. W. SULLIVAN Department of Animal Sciences, University of Nebraska, Lincoln, Nebraska ABSTRACT In two 21-d experiments, the performance, nutrient utilization, and carcass composition of ducks fed diets containing pearl millet were compared with those fed diets containing corn. Corn, in diets at two protein levels (22 vs 18% in Experiment 1 and 22 vs 16% in Experiment 2), was replaced by pearl millet either on an equal-weight or isonitrogenous basis. During the first 2 wk of the first experiment, corn diets compared on an isonitrogenous basis were superior (P <.05) to the millet diets in promoting weight gain (666 vs 633 g per duck) and feed consumption (965 vs 91 g per duck); but when compared on an equal-weight basis, the millet diets were similar to the corn diets in promoting growth. No significant differences were observed for weight gain, feed consumption, or feed efficiency at the end of 3 wk. Over the 21-d study, the higher protein diets were more efficiently (P <.05) converted to weight gain than the lower protein diets (.63 vs g gain:g feed). Carcass protein in ducks fed the corn diets was greater (P <.05) than in those fed the pearl millet diets. In Experiment 2, growth performance differences across dietary protein levels were similar to those in Experiment 1. However in contrast to Experiment 1, feed consumption was similar for corn and pearl millet diets (776 vs 786 g per duck, respectively), and ducks fed the corn diets gained less (P <.05) than those fed the pearl millet diets (590 vs 622 g per duck) during the first 2 wk of Experiment 2. Results of the nutrient utilization trial conducted at the end of Experiment 2 revealed that diets containing pearl millet were higher (P <.05) in AME than those containing corn (3,300 vs 3,100 kcal/ kg). These results indicate that pearl millet has a higher energy value and appears to be at least equal to corn in promoting weight gains in ducks. (Key words: pearl millet, duck, growth performance, carcass composition, metabolizable energy) INTRODUCTION Pearl millet (Pennisetum glaucum) is a cereal grain with good drought tolerance Received for publication June 21, Accepted for publication October 8, Purdue University Agricultural Experiment Station Journal Paper Number To whom correspondence should be addressed. O. Adeola, Department of Animal Sciences, Purdue University, 1151 Lilly Hall, West Lafayette, IN Poultry Science 73:25-35 and hardiness commonly grown in the semiarid regions of Africa and Asia on an estimated 26 million ha (Andrews and Kumar, 1992). The drought tolerant and hardy properties make pearl millet a cereal grain that could withstand some of the late summer droughts of the midwestern U.S. In addition to these properties, pearl millet has a relatively short growing season, which makes it a potential crop that can be double-cropped aft^r wheat has been harvested so that valuable

2 26 ADEOLA ET AL. soil is not allowed to lay idle prior to the following cropping season. Feed accounts for greater than 60% of production costs of market ducks. The availability of alternative energy feeds will ensure a feed supply at competitive rates that will minimize increases in the costs of production. Pearl millet could fill the niche during periods when corn production is affected by unseasonable weather. The limited information on the nutrient composition of pearl millet indicates that it compares favorably with other cereals (Ejeta et al, 1987). The energy density of pearl millet is relatively high, arising from the higher oil content of this grain relative to corn, wheat, or sorghum (Hill and Hanna, 1990; Sullivan et al, 1990; Haydon and Hobbs, 1991). Sullivan et al. (1990) also reported that pearl millet usually has a superior amino acid profile and a higher protein content than corn. Research on the use of pearl millet in livestock diets is limited. In broiler diets, pearl millet could replace corn without adversely affecting weight gains or feed efficiency (Singh and Barsaul, 1976; Sharma et al, 1979; Smith et al, 1989; Sullivan et al, 1990). Pearl millet also has the potential for use in swine (Haydon and Hobbs, 1991) and beef cattle (Cherney et al, 1990; Hill and Hanna, 1990) diets. The objectives of the experiments reported in this communication were to compare the performance, nutrient utilization, and carcass composition of ducks fed diets containing pearl millet vs corn. Corn, in diets at two protein levels, was replaced by pearl millet either on an equal-weight or isonitrogenous basis and fed to ducks from 1 to 21 d of age. A second objective was to compare effects of prior pearl millet diets fed from 1 to 21 d of age on subsequent performance and carcass characteristics when the ducks received the same diet from Day 21 to 9. MATERIALS AND METHODS Day-old White Pekin ducklings were housed in pens (1.05 x 3.66 m) with raised plastic-coated wire floors. Each pen was equipped with two hanging cylindrical feeders, one flat plastic feed tray, two satellite waterers, and a multinozzle drip nipple waterer. On Day 5, the satellite waterers were removed from the pens after the ducks had trained to drink from the drip nipples; the plastic feed tray was also removed at this time. Temperature in the room was controlled by ventilation fans and thermostatically controlled gas brooders. Incandescent light bulbs were used to provide ducks with 23 h light/d. Ducks had free access to feed and water at all times. The proximate and amino acid analyses of the major nutrients used in feed formulation for the two experiments are presented in Table 1. s used in both experiments were pelleted and crumbles were made by passing pellets through a double roller device attached to the mill. Experiment 1 Corn and pearl millet were compared at higher (s 1 to 3, with crude protein contents of 22, 22, and 2.6%, respectively) and lower (s to 6, with crude protein contents of 18, 18, and 21%, respectively) protein levels (Table 2). Within a protein level, corn and pearl millet were compared on an isonitrogenous basis (s 1 vs 2, and vs 5) where the protein remained constant by adjusting the amounts of grain and soybean meal. Also within a protein level, corn and pearl millet were compared on an equal-weight basis (s 1 vs 3, and vs 6); the protein was allowed to vary and the amounts of grain and soybean meal remained constant. Lysine-HCl and DLmethionine were added to all diets at and %, respectively, to ensure a minimum 1.08% lysine and.81% total sulfur amino acids in the lower protein diets. Four hundred and eighty day-old White Pekin ducks were wing-banded, weighed, and divided into 2 groups of 20 ducks each. The 2 groups of ducks were blocked on the basis of weight with six groups per block. Each group within a block of similar weight was then randomly allotted to each of the six diets. Thus, there were four blocks of 20 ducks each per treatment so that the initial weights of all treatments were identical. Each block of ducks was assigned to six pens at random. Feed and water were provided to the ducks for ad libitum consumption for a 3-wk experimental period. Ducks and feed were weighed on a weekly basis.

3 PEARL MILLET FOR DUCKS 27 On Day 21, two ducks from each pen were killed following electrical stunning, exsanguinated, scalded, picked, and eviscerated. Eviscerated carcasses were frozen, ground, and lyophilized. Lyophilized carcasses were stored at C until subsequent analyses. The remaining ducks were fed the same high-performance nutrientdense diet (18% protein,.95% lysine, % methionine plus cystine, and 3,100 kcal ME/kg) through the finishing period to Day 9. Feed intake and weight gains were monitored every other week. Ducks were shipped to Maple Leaf Farms, Franksville, WI, 53126, for processing, during which the following records were taken: dressing percentage, weights of breast meat, breast skin and fat, body fat, and keel length. Experiment 2 The second experiment was similar to Experiment 1 except that the protein level was dropped by two percentage units in the Analyses lower protein diets (Table 3). Within a protein level, com and pearl millet were compared on an isonitrogenous basis (s 7 vs 8, and 10 vs 11); that is, the protein remained constant by adjusting the amounts of grain and soybean meal. Also within a protein level, corn and pearl millet were compared on an equal-weight basis (s 7 vs 9, and 10 vs 12), where the protein was allowed to vary and the amounts of grain and soybean meal remained constant. Lysine-HCl and DLmethionine were added to all diets at and %, respectively, to ensure a minimum of 1% lysine and.75% total sulfur amino acids in the lower protein diets. A total of 288 day-old White Pekin ducks, divided into 2 groups of 12 ducks each, were used in the experiment. On Day 21, two ducks from each pen were transferred into cages for a digestibility study (see below) and the remaining ducks were fed the same high-performance, nutrientdense diet through the finishing period to TABLE 1. Proximate and amino acid composition of corn, pearl millet, and soybean meal Proximate analyses Moisture Protein Ether extract Fiber Ash Indispensable amino acids Arginine Histidine Isoleucine Leucine Lysine Methionine Phenylalanine Threonine Tryptophan Valine Dispensable amino acids Alanine Aspartic acid Cystine Glutamic acid Glycine Proline Serine Tyrosine a Not determined. Corn Experiment Pearl millet 1 Experiment 2 C" \ ND ND ND Soybean meal

4 28 ADEOLA ET AL. Day 9 as in Experiment 1. Feed intake and weight gains were monitored every other week. Digestibility Study On Day 21 of Experiment 2, two ducks from each pen were transferred into cages for nutrient digestibility studies. The ducks had free access to water and the diets they were receiving prior to being transferred to cages. The diets contained chromic oxide at %, which was used as an index of indigestibility. There were four cages for each of the six diets. Trays lined with waxed paper were placed under the cages for collecting excreta. Excreta were collected during the subsequent 8-h period and dried at 65 C. Dried excreta were ground to pass through a 1-mm screen and stored at C until subsequent analysis. Analyses Samples of feed ingredients, diets, carcass, and excreta were analyzed for moisture, crude protein (N x 6), ether extract, and ash (Association of Official Analytical Chemists, 1980). s and excreta were analyzed for chromic oxide by atomic absorption spectrophotometry (Williams et ol., 1962), and the energy values determined in an adiabatic bomb calorimeter. Phenyl thiocarbamyl derivatives of amino acids of 6 N HC1 hydrolyzates were separated by reverse-phase HPLC as described previously (Adeola et rt., 1988). Data were subjected to analysis of variance procedures (Steele and Torrie, 1980). Means were compared by three structured contrasts: higher vs lower protein, corn vs pearl millet on an equal-weight basis, and corn vs pearl millet at similar dietary protein level. RESULTS AND DISCUSSION The proximate and amino acid analyses of the corn, millet, and soybean meal used in the experiments are presented in Table 1. The pearl millet used in the experiments had 25 to 5% more protein than the corn. TABLE 2. Composition of diets comparing corn and pearl millet for ducks, Experiment 1 Ingredients and analyses Pearl millet Corn Soybean meal Corn oil Dicalcium phosphate Vitamin and trace mineral mix 1 Limestone Salt Lysine HC1 DL-methionine Pellet binder 2 Total Calculated analyses Protein Lysine Total sulfur amino acids ME, kcal/kg Calcium Total phosphorus Available phosphorus , , , s 1 ') , , ,531.5 Provided the following per kilogram diet: vitamin A, 11,000 IU (retinyl acetate); cholecalciferol,,00 ICU; vitamin E, 11 ICU (dl-a-tocopherol acetate); choline chloride, 800 mg; niacin, 77 mg; calcium pantothenate, 11.9 mg; riboflavin, 6.6 mg; menadione sodium bisulfite, 3.3 mg; folacin, 660 \i%; D-biotin, 110 ^g; vitamin B 12,11 ^g; manganese, 100 mg; zinc, 100 mg; iron, 100 mg; copper, 20 mg; iodine, 3 mg; selenium, 200 \>%. 2 Carboxymethylcellulose, a pellet binder (Hercules Cellulose Gum; Hercules, Inc., Wilmington, DE 1989).

5 PEARL MILLET FOR DUCKS 29 Also, the amino acid concentrations of the pearl millet (on an as is basis) were higher than corn (Table 1). More importantly, the concentrations of the important amino acids in animal nutrition (lysine, methionine, threonine, and tryptophan) were higher in pearl millet as compared with corn. Amino acid concentrations in the 11% protein pearl millet are similar to those reported for the Kordofani variety of pearl millet by Ejeta et al. (1987) and Singh et al. (1987). The pearl millet used in the first experiment had more protein than that used in the second experiment (13.3 vs 11.2%, respectively). Soil moisture content, soil nitrogen level, and the time of nitrogen fertilizer application are some of the factors that influence the protein content of wheat (Wood and Fox, 1965). It is possible that these factors, in addition to the fact that the pearl millet was grown in different locations, may have contributed to the difference in the protein content. The higher ether extract content of pearl millet compared with corn (Table 1) is in agreement with a previous report (Sullivan et al, 1990). In earlier reports, Sharma et al. (1979) and Osagie and Kates (198) observed the ether extract in pearl millet to vary from to 7%, which was found to be higher than in corn. Live Performance The results of the first 3-wk experiment are presented in Table. At the end of Week 2, weight gain of ducks on the higher protein diets were, as expected, significantly greater than those on the lower protein diet (s 1, 2, and 3 vs,5, and 6). Increased dietary protein concentration has been reported to increase live weight gains during the finishing stage of production in Muscovy ducks (Auvergne et al, 1991). Also at the end of Week 2, the corn diets, compared on an isonitrogenous basis across protein levels (s 1 and vs 2 and 5), were significantly superior to the millet diets in promoting weight gain. When compared on an equal weight or percentage TABLE 3. Composition of diets comparing corn and pearl millet for ducks, Experiment 2 Ingredients and analyses Pearl millet Corn Soybean meal Corn oil Dicalcium phosphate Vitamin and trace mineral mix 1 Limestone Salt Lysine HC1 DL-methionine Pellet binder 2 Chromic oxide Total Calculated analyses Protein Lysine Total sulfur amino acids ME, kcal/kg Calcium Total phosphorus Available phosphorus , , , s 10 (%) , , , Provided the following per kilogram diet: vitamin A, 11,000 IU (retinyl acetate); cholecalciferol,,00 ICU; vitamin E, 11 ICU (dl-a-tocopherol acetate); choline chloride, 800 mg; niacin, 77 mg; calcium pantothenate, 11.9 mg; riboflavin, 6.6 mg; menadione sodium bisulfite, 3.3 mg; folacin, 660 fig; D-biotin, 110 /jg; vitamin B 12,11 /ig; manganese, 100 mg; zinc, 100 mg; iron, 100 mg; copper, 20 mg; iodine, 3 mg; selenium, 200 ng. 2 Carboxymethylcellulose, a pellet binder (Hercules Cellulose Gum; Hercules, Inc., Wilmington, DE 1989).

6 30 ADEOLA ET AL. basis (s 1 and vs 3 and 6), the pearl millet diets were equal to the corn diets in promoting growth during the first 2 wk of the experiment, particularly at the lower protein level. Substitution of millet for corn on an equal-weight basis produced a similar performance to corn diets in broilers (Abate and Gomez, 1983). The contrast of higher vs lower protein diets was, however, not significant over the 3-wk period, indicating a combination of adaptive response and changing requirement with age. Furthermore, there were no significant differences in weight gain over the 3-wk period between the corn and the pearl millet diets. Feed intake during the 3-wk period of the experiment was higher (P <.05) in ducks that received corn s 1 and, than those that received pearl millet s 2 and 5 (Table ). It was reasoned that a further reduction in the protein content of diets would better test the quality of protein in pearl millet. Thus, 22% protein diets were compared with 16% protein diets in the second experiment. The difference in weight gain between the higher and lower protein groups in the first 2 wk of the second experiment (Table 5) followed a similar pattern as was observed in Experiment 1. The higher protein diets promoted greater Variable Initial weight, g Gain, g per duck Day 0 to 1 2^ Day 0 to 21 Feed consumption, g per duck Day 0 to 1 3 Day 0 to 213 Gain:feed, g:g Day 0 to 1 2 Day 0 to 21 2 n" (P <.05) weight gains than the lower protein diets (s 7,8, and 9 vs 10,11, and 12). In contrast to the first 2 wk of Experiment 1, in which corn diets promoted greater weight gains than pearl millet diets, weight gains were lower (P <.05) in ducks fed the corn diets than those fed pearl millet when compared on an isonitrogenous basis (s 7 and 10 vs 8 and 11). This difference between the first 2 wk of Experiments 1 and 2 is mostly attributable to feed intake. During the first 2 wk in Experiment 1 (Table ), feed intake was higher (P <.05) in ducks that were fed the diets containing corn (s 1 and ) than those fed isonitrogenous diets containing pearl millet (s 2 and 5); but in Experiment 2 (Table 5), feed intake was similar across diet comparisons. Smith et al. (1989) also observed that when pearl millet replaced corn in chick diets, weight gains increased during the first 2 wk but not over the 3 wk of the experiment. Over the 3-wk period of Experiment 2, feeding the higher protein diets or pearl millet diets resulted in greater (P <.05) weight gains than lower protein diets or corn diets, respectively (Table 5). As for feed efficiency, the higher protein diets were superior (P <.05) to the lower protein diets when compared across grains in both experiments (Tables and 5). The TABLE. Performance of White Pekin ducks fed on diets containing corn and pearl millet at higher and lower dietary protein levels, Experiment , , , , , , s , , , , , ,997 J 1 = control 22% CP corn-soybean meal; 2 = pearl millet-soybean meal diet isonitrogenous to 1; 3 = pearl millet-soybean meal diet with weight-for-weight replacement of corn in 1; = 18% CP corn-soybean meal; 5 = pearl millet-soybean meal diet isonitrogenous to ; 6 = pearl millet-soybean meal diet with weight-for-weight replacement of corn in. 2 Contrast of s 1, 2, and 3 vs, 5, and 6 (higher vs lower protein) (P <.05). 3 Contrast of s 1 and vs 2 and 5 (corn vs pearl millet at similar dietary protein level) (P <.05). Represents data from 20 ducks in each of four pens per diet..59 SD

7 PEARL MILLET FOR DUCKS 31 ducks that received the diets containing pearl millet on an isonitrogenous level in Experiment 2 (s 8 or 11) were more efficient (P <.05) at converting feed to body weight gains than those that received isonitrogenous s 7 or 10 containing corn (Table 5). Over the 3-wk period of Experiment 2, ducks that received the pearl millet s 8, 9,11, and 12 were more efficient in converting feed to body weight gains than those that received the corn s 7 and 10. This is similar to the observations reported by Smith et al. (1989) in broiler experiments where replacement of corn with pearl millet improved the efficiency of feed conversion, an improvement that was attributed to an increase in the ME contents of diets that contained pearl millet. It appears that the summer period during which Experiment 2 was conducted reduced the feed intake of ducks. This led to lower weight gains compared to Experiment 1, which was conducted in the spring. The study also compared the effects of prior pearl millet diets fed from 1 to 21 d of age on subsequent weight gain, feed intake, and feed efficiency, when the ducks received the same diets from Day 21 to 9 in both experiments. Subsequent performance was Variable Initial weight, g Gain, g per duck Day 0 to 1*A* Day 0 to Feed consumption, g per duck Day 0 to 1 Day 0 to 21 Gain:feed, g:g Day 0 to 1 Day 0 to 21^ n= not affected by feeding pearl millet between 1 and 21 d (data not shown). Nutrient Utilization At the conclusion of 21-d feeding in Experiment 2, two ducks from each pen were transferred to cages for a nutrient utilization study. Dry matter utilization (Table 6) was lower (P <.05) in the higher protein diets than it was in the lower protein diets (s 7,8, and 9 vs 10,11, and 12). Dry matter in diets containing corn (s 7 and 10) was less well utilized (P <.05) than that in the diets containing pearl millet (s 8, 9, 11, and 12). Although there was a tendency for a higher apparent nitrogen retention when ducks received the higher protein diets or diets containing pearl millet compared with lower protein diets or diets containing corn, respectively, the difference was not significant. Energy retention (percentage of energy ingested) was lower (P <.05) in the higher than lower protein diets. The diets containing corn had lower (P <.05) energy retention values than those containing pearl millet (s 7 and 10 versus 8,9,11, and 12). Thus, higher (P <.05) AME values (Table 6) TABLE 5. Performance of White Pekin ducks fed on diets containing corn and pearl millet at higher and lower dietary protein levels, Experiment , , , , , , s , , , , , , SD ! 7 = control 22% CP corn-soybean meal; 8 = pearl millet-soybean meal diet isonitrogenous to 7; 9 = pearl millet-soybean meal diet with weight-for-weight replacement of corn in 7; 10 = 18% CP corn-soybean meal; 11 = pearl millet-soybean meal diet isonitrogenous to 10; 12 = pearl milletsoybean meal diet with weight-for-weight replacement of corn in Contrast of s 7, 8, and 9 vs 10, 11, and 12 (higher vs lower protein) (P <.05). 3 Contrast of s 7 and 10 vs 8, 9, 11, and 12 (corn vs pearl millet) (P <.05). Contrast of s 7 and 10 vs 8 and 11 (corn vs pearl millet at similar dietary protein level) (P <.05). 5 Represents data from 12 ducks in each of four pens per diet

8 32 ADEOLA ET AL. were observed for diets containing pearl millet than those containing corn (3,300 vs 3,100 kcal/kg). The higher ME value of the diets containing pearl millet is in conformity with its higher ether extract content compared with corn (Table 1). Feed consumption was similar across diet comparisons in Experiment 2, but the caloric value of diets containing pearl millet was higher than those containing corn. This observation is an indication that a compensatory increase in feed intake to overcome a decrease in energy content of the diet does not always occur. The AME content of the diets containing corn (s 7 and 10) determined in the nutrient utilization study (Table 6) are similar to values calculated (Table 3) using the NRC (198) composition of feedstuffs. Using NRC (198) values for pearl millet underestimated the AME contents of diets containing pearl millet by kcal/kg (Tables 3 and 6). This observation is consistent with the data reported by Fancher et al. (1987), suggesting that previously reported ME n values for pearl millet maybe underestimated by up to 21%. In an experiment on feed grain utilization in broiler diets, Sharma et al. (1979) reported that pearl millet had lower ME than corn, an observation that is contrary to the results of the present experiment on nutrient utilization. The fact that broilers were used in their experiments, that the ether extract content of corn (3.7%) and pearl millet (.3%) were similar, and that the level of protein meal used in their experiments were higher than that used in the present experiments may be responsible for the disparity. The CP content of pearl millet and corn used in their experiments were 11. and 9.9%, respectively. Carcass Composition The protein and ash contents of carcasses at the end of 21-d feeding in Experiment 1 was affected by dietary protein level (Figure 1). Ducks that received the higher protein s 1, 2, and 3 had more (P <.05) carcass protein (Figure 1A) and ash (Figure IB) than those that received the lower protein s, 5, and 6 (13.9 and 2.78% vs 13.1 and 2.8%, respectively. The reverse was the case with carcass ether extract (Figure 1C) content; ducks that received the higher protein diets had less (P <.05) ether extract than those that received the lower protein diets. The higher proportion of ether extract in carcasses of ducks that received the lower protein s, 5, and 6 during the 21-d feeding experiment resulted in the ducks having more (P <.05) carcass dry matter (Figure ID) than those that received the higher protein diets (36.6 TABLE 6. Dry matter, nitrogen, and energy utilization in White Pekin ducks fed on diets containing corn and pearl millet at higher and lower dietary protein levels on Day 21, Experiment 2 1 Item 7 CP, % Gross energy, kcal/kg Dry matter utilization, 3 - % Nitrogen retention, % Energy retention, 3 ' % ME,* kcal/kg n , , , , , ,31 s , , , , , ,377 Expressed on a dry matter basis. 2 7 = control 22% CP corn-soybean meal; 8 = pearl millet-soybean meal diet isonitrogenous to 7; 9 = pearl millet-soybean meal diet with weight-for-weight replacement of corn in 7; 10 = 18% CP corn-soybean meal; 11 = pearl millet-soybean meal diet isonitrogenous to 10; 12 = pearl milletsoybean meal diet with weight-for-weight replacement of corn in Contrast of s 7, 8, and 9 vs 10, 11, and 12 (higher vs lower protein) (P <.05). ^Contrast of s 7 and 10 vs 8, 9, 11, and 12 (corn vs pearl millet) (P <.05). Represents data from two ducks in each of four cages per diet. SD

9 PEARL MILLET FOR DUCKS 33 ducks that received the diets containing corn (s 1 and ) was greater (P <.05) than in ducks that received the diets containing pearl millet (s 2, 3, 5, and 6). The corn-pearl millet comparison for other carcass components was not statistically significant (Figure 1). The effects of corn and pearl millet at higher and lower dietary protein levels on rates of deposition of protein and ether extract in duck carcasses are shown in Panels E and F of Figure 1. The calculation of deposition rates of protein and ether extract was based on 5% dressing percent- FIGURE 1. Chemical composition (percentage), protein, and fat deposition rates (grams per day) in White Pekin ducks fed on diets containing corn and pearl millet at higher and lower dietary protein levels from Day 0 to 21 in Experiment 1. Each bar represents a mean of eight ducks with the following standard deviations:.6,1.59,.33, and 1.9% for carcass protein, ether extract, ash, and dry matter, respectively. Deposition rates of protein and ether extract were based on 5% dressing percentage, 15% CP, and 13.3% ether extract for day-old White Pekin ducks (Summers and Leeson, 1985); standard deviations were.8 and.8 g/d for carcass protein and ether extract deposition rates, respectively. Significant (P <.05) differences between high- and low-protein diets (s 1,2, and 3 vs, 5, and 6) occurred for all variables. Significant (P <.05) differences between corn and pearl millet diets (s 1 and vs 2, 3, 5, and 6) occurred for percentage carcass protein.

10 3 ADEOLA ET At. age, 15% CP, and 13.3% ether extract for day-old White Pekin ducks (Summers and Leeson, 1985). Protein accretion in grams per day (Figure IE) was greater (P <.05) in ducks that were fed the higher protein s 1,2, and 3 than those that were fed the lower protein s, 5, and 6 (.8 vs. g/d). Ducks that were fed the higher protein diets had lower (P <.05) ether extract accretion rate (Figure IF) than those that were fed the lower protein diets (5.7 vs 6.7 g/d). The corn-pearl millet comparison for protein or ether extract accretion (grams per day) was not statistically significant. ary protein level during the first 3 wk did not affect 9-d carcass weight (Figure 2A) or dressing 3! 2,00 I 2,300 2, > n 0) 350 % (0 CD 325 L 150 E g ^T 12 I 22 2,331 VA I X c «a R3 68 * ) s 08 ~ 175 I CO g 550 >«a o CO O 67.8 v\ V i l ' : S FIGURE 2. Carcass characteristics of 9-d-old White Pekin ducks fed on diets containing corn and pearl millet at higher and lower dietary protein levels from 1 to 21 d of age but fed the same high-performance nutrient-dense diet through the finishing period to Day 9 in Experiment 1. Each bar represents a mean of eight ducks with the following standard deviations: 188, 2.7, 28, 2, 6, and 59 for carcass weight, dressing percentage, breast meat, breast skin and fat, keel, and body fat, respectively. 1 = control 22% CP corn-soybean meal; 2 = pearl millet-soybean meal diet isonitrogenous to 1; 3 = pearl millet-soybean meal diet with weight-for-weight replacement of corn in 1; = 18% CP corn-soybean meal; 5 = pearl millet-soybean meal diet isonitrogenous to ; 6 = pearl millet-soybean meal diet with weight-for-weight replacement of corn in.

11 PEARL MILLET FOR DUCKS 35 percentage (Figure 2B). Neither were any of the remaining carcass characteristics presented in Figure 2 affected by preceding dietary treatments. Auvergne et al. (1991) reported that 12-wk-old Muscovy ducks fed higher or lower protein diets had similar weights of breast meat, abdominal fat, and breast skin and fat during the finishing period. The conservative view is that pearl millet appears to have a similar feeding value to corn when substituted for corn on an equalweight basis. s containing pearl millet have higher ME than those containing corn, but data is lacking on the digestibility of nutrients in pearl millet. ACKNOWLEDGMENTS The authors are grateful for the efforts of Ken Wobler and the staff of Baker- Purdue Poultry Research Center and Marisue Freed in caring for the ducks. The excellent technical assistance of Ray A. Sweet is also thankfully acknowledged. This study was supported by Purdue University New Crops Center and by Maple Leaf Farms, Milford, IN 652, which donated the White Pekin ducklings. REFERENCES Abate, A. N., and M. Gomez, Substitution of finger and bulbrush millet for maize broiler feeds. Anim. Feed Sci. Technol. 10: Adeola, O., J. G. Buchanan-Smith, and R. J. Early, Gas and high performance liquid chromatographic analysis of amino acids in soybean meal, corn and triticale. J. Food Biochem. 12: Andrews, D. J., and K. A. Kumar, Pearl millet for food, feed, and forage. Adv. Agron. 8: Association of Official Analytical Chemists, Official Methods of Analysis. 13th ed. Association of Official Analytical Chemists, Washington, DC. Auvergne, A., C. Baudonnet, and R. Babile, Influence of protein and methionine concentrations and body size on the growth and carcase of Muscovy ducks in the finishing stage of production. Br. Poult. Sci. 32: Cherney, D.J.R., J. A. Patterson, and K. D. Johnson, Digestibility and feeding value of pearl millet as influenced by the brown-midrib, lowlignin trait. J. Anim. Sci. 68: Ejeta, G., M. M. Hassen, and E. T. Mertz, In vitro digestibility and amino acid composition of pearl millet (Pennisetum typhoides) and other cereals. Proc. Natl. Acad. Sci. USA 8: Fancher, B. I., L. S. Jensen, R. L. Smith, and W. W. Hanna, Metabolizable energy content of pearl millet [Pennisetum americanum (L.) Leeke]. Poultry Sci. 66: Haydon, K. D., and S. E. Hobbs, Nutrient digestibilities of soft winter wheat, improved triticale cultivars, and pearl millet for finishing pigs. J. Anim. Sci. 69: Hill, G. M., and W. W. Hanna, Nutritive characteristics of pearl millet grain in beef cattle diets. J. Anim. Sci. 68: National Research Council, 198. Nutrient Requirements of Poultry. 8th rev. ed. National Academy Press, Washington, DC. Osagie, A. U., and M. Kates, 198. Lipid composition of millet (Pennisetum americanum) seeds. Lipids 19: Sharma, B. D., V. R. Sadagopan, and V. R. Reddy, Utilization of different cereals in broiler diets. Br. Poult. Sci. 20: Singh, S. D., and C. S. Barsaul, Replacement of maize by coarse grains for growth and production in White Leghorn and Rhode Island Red birds. Ind. J. Anim. Sci. 6: Singh, P., U. Singh, B. O. Eggum, K. A. Kumar, and D. J. Andrews, Nutritional evaluation of high protein genotypes of pearl millet. J. Sci. Food Agric. 38:1-18. Smith, R. L., L. S. Jensen, C. S. Hoveland, and W. W. Hanna, Use of pearl millet, sorghum, and triticale grain in broiler diets. J. Prod. Agric. 2: Steel, R.G.D., and J. H. Torrie, Principles and Procedures of Statistics: A Biometrical Approach. 2nd ed. McGraw-Hill, New York, NY. Sullivan, T. W., J. H. Douglas, D. J. Andrews, P. L. Bowland, J. D. Hancock, P. J. Bramel-Cox, W. D. Stegmeier, and J. R. Brethour, Nutritional value of pearl millet for food and feed. Pages 83-9 in: Proceedings of the International Conference on Sorghum Nutritional Quality. Purdue University, West Lafayette, IN. Summers, J. D., and S. Leeson, Poultry Nutrition Handbook. University of Guelph, Guelph, ON, Canada. Williams, C. H., D. J. David, and O. Iismaa, The determination of chromic oxide in faeces samples by atomic absorption spectrophotometry. J. Agric. Sci. 59: Wood, H. L., and W. E. Fox, The interaction of nitrogen and water on yield, protein, and mottling in wheat grown on black earth in Queensland. Exp. Agric. 1: