| Livestock Research for Rural Development 17 (4) 2005 | Guidelines to authors | LRRD News | Citation of this paper |
Poultry can be managed under different feeding systems, depending on the husbandry skills and the feed available. Under free-choice feeding, birds are usually offered a choice between three types of feedstuffs: an energy source, a protein source, and in the case of laying hens, a calcium source. Protein is an expensive dietary constituent and there is more interest in optimizing its dietary concentration in commercial practice. This review therefore concentrates on diet selection for protein, although it must be acknowledged that in most cases it is only an assumption that protein is the target of the bird's selection, as details on the content of minerals and vitamins are not always published. When whole grain is used, the intake of which is accompanied by increased efficiency of feed utilization, the costs of grinding, mixing and many of handling procedures associated with mash and pellet preparation are eliminated.
The realization that it is necessary for animals to be able to differentiate between foods with different nutrient compositions by colour, taste and/or position, and that they need to be taught to associate the sensory properties of foods with their yields of nutrients, has made it possible to envisage a learned appetite for each of the essential nutrients. Birds can improve the balance between their nutrient requirements and their nutrient intake if they can be taught to select an appropriate diet.
Key words: choice-feeding, characteristics, chickens, ducks, turkeys
The ability of chickens to select a balanced diet if offered a choice of various feed ingredients has been demonstrated by a number of workers (Graham 1934; Holcombe et al 1975; Emmans 1977; Summers and Leeson 1978). The principle underlying choice feeding of poultry is that individual birds reared in a flock are able to select between various feed ingredients according to individual needs and production capacities. This is not possible when a single conventional food is given. Cumming (1992a) and Ciszuk et al (1998) reported that choice feeding (of whole grains plus a protein concentrate and calcium) of layers has financial advantages for rural small-scale poultry production.
This review concentrates on choice feeding of poultry, with the emphasis on factors in the feed, which determine its intake. The influence of the feeds' appearance, odour and amino acid balance are also discussed.
According to Emmans (1991) the three key parameters in the feed are protein, energy, and minerals. The birds are selecting foods with a nutritional purpose. This is strengthened if their food intake over long periods is compared with the amount which is selected against when choice is given. Flavours can initially influence intake and preference but soon lose this ability as the birds learn that there is no nutritional implication in the different flavours (Balog and Millard 1989). Some flavour preferences are very strong, however, and chickens will not drink solutions of saccharin, salt or quinine but take readily sucrose (Jacobs and Scott 1957), glucose (Engku and Forbes 1989) and citric acid (Balog and Millar 1989). In general, visual cues are probably more useful than oral cues in the chicken.
It has been observed that when a new diet formulation is introduced, or a new type of food is presented, animals will often refuse to eat the food for a period of time or intake is reduced (Jones 1986; Launchbaugh et al 1997). Furthermore, it has been noted that this may also occur when a new batch of the same diet formulation is presented to broiler chickens or turkeys. This has been studied by Haskell et al (2001), who found that food consumption was significantly lower for experimental groups compared to control groups on all days from the day of change onward in broiler chickens.
Bennet (2003) reported that if vitamin and mineral premixes are placed in separate feeders, some birds might not eat them because they will not like the taste, while other individuals may over-consume them and suffer toxic side effects. The same author concluded that when mixing whole grain, supplement and limestone together into a traditional laying hen diet that is fed in one feeder, not more than 50% of the diet should be as whole grain. The rest of the grain in the feed should be ground because at higher levels of inclusion of whole grains the hens sometimes have trouble finding the supplement among all the grains.
Up to about 14 weeks of age growing pullets require protein only for feather development and a relatively slow rate of muscle deposition. However, around fifteen weeks of age, there is rapid development of the ovaries and oviduct, which would be expected to increase the protein demand, and thus protein intake, in a choice feeding situation. This has been observed by Scott and Balnave (1989) as a marked increase in the protein content of their diet about 2 weeks before the onset of lay in choice-fed pullets. Once in full lay, the birds are capable of choosing a balanced diet when given the opportunity.
Laying hens are normally fed a mash diet. The use of whole grain would not only save the energy cost of grinding and mixing but is also accompanied by increased efficiency of food utilisation. McIntosh et al (1962) showed that grinding or pelleting wheat, barley, oats or maize did not result in a consistent increase in the metabolizable energy of poultry feed. Moreover, they found that whole wheat yielded more metabolisable energy than ground or pelleted wheat in two out of three experiments, probably because the intake of whole grain is accompanied by increased efficiency of feed utilisation (Henuk and Dingle 2002). Emmans (1977) offered hens either a complete layer mash or a choice between the complete diet and ground barley, and found no differences in performance between the two treatments. Karunajeewa (1978) fed laying hens complete mash diets, either barley or wheat based, or a choice between the whole grains and a concentrate mixture. Hens receiving wheat laid better than those fed barley, but the hens receiving a choice laid heavier eggs and consumed 11% less food than those given the complete diets. Mongin and Sauveur (1974) reported that hens increased their calcium intake when ovulation was about to occur during the laying period. Hughes and Wood-Gush (1971) found that calcium appetite appeared to be a learned response and that the laying hen has the ability to selectively consume calcium to meet maintenance and production requirements.
An experiment carried by Olver and Malan (2000) showed that between16-80 weeks of age, the hens offered a choice did not lay more eggs, nor did they consume more food than those fed a complete mash diet (Table 1). However, the hens offered the choice fed diet laid heavier eggs and eggs with thicker shells (p<0.01) than those fed the complete diet. The heavier egg size of the choice-fed hens accounted for the better (p<0.01) food conversion ratio obtained by these hens compared to those fed the complete diet. These results are at variance with the finding that increased protein intake results in heavier eggs (Scott et al 1982). However, both Blair et al (1973) and Karunajeewa (1978) reported heavier eggs from choice fed-hens, that also consumed less energy and protein. Olver and Malan (2000) also reported a higher calcium intake by choice fed-hens than those fed a mash diet. This was coupled with differences (p<0.01) in eggshell thickness, which indicated that the increased egg mass was due to increased calcium intake. The Haugh unit scores of eggs did not differ (p>0.05) between diets, but the yolk colour of eggs laid by choice-fed hens was darker (p<0.01) than that of eggs laid by hens fed the complete mash diet.
|
Table 1. The effect of diet on the daily food consumption in laying hens from 16-80 weeks of age, g/day |
|||
|
|
Maize, calcium and pelleted concentrate |
Maize, calcium and |
Mash diet |
|
Maize |
66.6±3.04a |
65.6±2.68b |
68.9±1.30a |
|
Protein concentrate |
34.4±2.25a |
36.2±2.10a |
39.9±0.75b |
|
Limestone powder |
15.5±0.86b |
15.0±1.12b |
9.57±0.10a |
|
ME intake, MJ |
1.26±0.04a |
1.26±0.04a |
1.36±0.03b |
|
Protein intake |
17.5±0.78a |
18.1±0.51a |
19.6±0.37b |
|
Calcium intake |
6.38±0.34b |
6.24±0.44b |
4.03±0.08a |
|
Olver and Malan 2000 |
|||
Hens are in many cases able to adapt their feed intake to their requirement in a quantitative (amount of feed ingested) and also qualitative (level of protein, energy, minerals etc.) way (Burel et al 1999). However, Dana and Ogle (2002) evaluated the effect of scavenging on diet selection and the comparative performance of exotic and local chickens and found that despite the provision of a choice of supplements providing a source of energy and protein, both the scavenging and the confined birds on choice feeding failed to eat sufficient amounts of the supplement to meet their protein requirements. On the other hand, they noticed that the energy intake of both scavenging and confined hens under choice feeding was in excess of requirement (by 22-29%), although the ME requirement of scavenging birds is slightly underestimated, since the additional energy cost for scavenging was not considered in the calculation. They concluded that the amount of energy contributed by the scavenging feed resource seems to be relatively little. Under choice feeding, these authors recorded smaller eggs laid by scavenging hens compared to the confined groups.
Farrell et al (1981) carried out a feeding experiment comparing a commercial layer diet and three free-choice diets, consisting of shell grit, a protein source and one of three energy sources: maize (100%); maize (60%) plus cassava (40%); and maize (94%) plus palm oil (6%). They found that feed and energy intakes were significantly higher on the commercial diet and lower on the maize and cassava mixture than the other treatments. There was no significant difference for protein intake. Egg production and egg weight were similar on the four treatments but gross efficiency of feed conversion to egg mass and cost of production were much higher on the commercial diet.
Olver and Malan (2000) reported that during the rearing period (7-16 weeks) the pullets offered the choice-fed diets gained more weight (p<0.05) than those fed the commercial grower diet. These pullets consumed approximately 7 g maize per day more than the control pullets, but had a lower total food intake. This suggests that the pullets given whole maize were able to utilize dietary energy more efficiently than those given ground maize. This substantiates the finding of McIntosh et al (1962) that whole grains tend to yield more ME than pelleted or ground grains. Olver and Malan (2000) also found that the pullets fed the free-choice diets consumed more than twice the amount of limestone than those offered the complete diet
From the evidence available, laying hens seem less able than broilers to balance their protein intake when offered high-protein and low protein foods, but it may be explained by the fact that mature birds learn more slowly than rapidly growing ones (Forbes and Shariatmadari 1994). Broiler chickens make a sensible choice when given the option of choosing between diets differing in protein quality. Gous and Swatson (2000) showed that broiler chickens, when provided with 2 or 3 foodstuffs containing just one protein source, on a free-choice basis, which in some proportion will meet their requirements, effectively select a combination which maximizes their biological performance.
Under conditions of heat stress, with day-time temperatures rising to 33°C, Cumming (1992a) demonstrated that broiler chickens will reduce their grain (energy) intake by 34% but their protein intake by only 7%, compared to similar birds in a cooler (20°C) environment. Remarkably, choice-fed birds have been found to have a "protein memory" and consume early the next day, before the temperature rises, the protein they did not eat during the previous hot day. Thus, the performance of choice-fed broilers was very significantly better than that of the same birds fed the most sophisticated complete diets (Mastika and Cumming 1985) in hot environments.
Gous and Dupreez (1975) gave layer-strain cockerels, in alternating periods of 6 or12 hours, two foods, which were individually poorly balanced but complementary in their amino acid composition. There were no significant differences in food intake or weight gain, either between the two alternating treatments or compared with controls given the two foods mixed together. Thus, the growing birds appear to have the ability to compensate for short periods on amino acid imbalanced foods. Elkin et al (1985) suggested that regulation of protein intake in choice feeding situations is only possible when the diet offered has adequate sulphur amino acids.
Newman and Sands (1983) offered newly hatched layer chicks a choice between a low-lysine food and one with an excess of lysine. Although the birds ate some of the supplemented food, it was not enough to maintain a growth rate as high as those in a control group, which were given a single adequate feed. The two feed containers were kept in the same place throughout the 21-day experiment, but no colour cues were given and no separate training period was provided. Because of the small amounts of lysine involved and its relative lack of colour and taste it would be necessary to provide a strong cue such as food colour. In a second experiment, a low-lysine food was given in choice with L-lysine HCI, but although the birds ate some lysine, it was not enough to support normal growth (Newman and Sands 1983). In this case it would be expected that there was sufficient contrast between the appearance and taste of the two foods on offer for adequate intake to occur. When D-lysine HCI was offered as a choice with a low-lysine food, the birds ate some, suggesting that D-lysine triggers a receptor mechanism even though it is unavailable for metabolism. Given a choice between L- and D-lysine, birds ate more of the L form. There is therefore some evidence of nutritional wisdom, but not sufficient to give a properly balanced diet.
In layers, a methionine-deficient diet depressed egg production from 0.85 to 0.67 eggs per bird per day, and a choice between deficient and adequate diets resulted in a 0.58 choice of the adequate diet and egg production of 0.80. Selection for methionine therefore occurred, but not quite enough to prevent a decline in egg production (Hughes 1979).
Evidence produced in a laboratory suggests that an active functioning gizzard may well play a role in the chickens' resistance against coccidiosis (Cumming 1992b). Further work has shown that chickens free-choice fed on a high (42% CP) protein concentrate and whole wheat were even more resistant to coccidiosis than those on complete high fibre diets (Cumming 1987, 1989). In male broiler chickens, occyst output has been shown to be negatively correlated to relative gizzard size, whether fed on conventional complete diets or free-choice fed (Cumming,1992b). Finally, the feeding of insoluble grit has been shown to reduce oocyst output from chickens fed either complete diets or free-choice fed whole grains (Cumming 1992b).
Work in a laboratory (Mastika 1981,1987) has highlighted the importance of experience and group learning in free-choice broiler chickens. Broiler chickens take about ten days to learn to accurately balance their protein concentrate and whole grain intakes. They need to be in groups of at least eight birds and be offered the protein concentrate (in mash or crumble form) and whole grain in identical, adjacent troughs, or even better, in the same feed trough. Forbes and Shariatmadari (1994) and Forbes and Covasa (1995) reviewed the effects of choice feeding on the performance of laying hens and reported that both growing and laying birds needed a period of learning before becoming proficient in the selection of feedstuffs
Cumming (1992a) demonstrated quite marked genetical differences in the ability of different strains to adapt to free-choice. Egg-type stock adapt quicker than broilers. The same author found considerable differences between adults of layer strains in adapting to free-choice feeding. Brown egg layers seem to adapt more readily than white or tinted egg layers. However, all Australian strains of commercial layers and broilers can learn in ten to fourteen days to very accurately balance their energy and protein intakes to maximize production and optimise economic returns.
Emmerson et al (1990) studied diet selection by turkey hens. Controls were given a complete feed containing 181g/kg CP and 11.23 MJ/kg ME. Putting high-protein ingredients into one food (351g/kg CP and 8.12 MJ/kg ME), they made the high-protein feed and then the low-protein feed by including low-protein ingredients. The choice-fed turkeys consumed 10% less food, 44% less protein and the same amount of energy yet laid a similar number of eggs as those fed conventionally. In a further experiment of similar design (Emmerson et al 1991) there was no difference in egg production over a 20-week period. Broodiness tended to be reduced by choice feeding but fertility and hatchability were lower. There was no significant difference in food intake but choice-fed birds had a higher energy intake as they selected a higher proportion of the low protein food than was included in the control feed. Energy intake fell as the experiment progressed, presumably because of the rise in environmental temperature. Protein intake was about 35% lower for choice-fed birds, and the overall protein concentration of 110g/kg diet was considerably below NRC (1984) recommendations. The fact that the protein content of the chosen diet increased throughout the experiment suggested that turkey hens fed single diets should be given feed with stepwise increases in protein content. The energy/protein ratio was 104.6 MJ/kg protein compared with 62.1 MJ/kg for the control group.
An experiment was carried out in Vietnam by Bui Xuan Men et al (2001), who offered commercial diets with different levels of crude protein to growing broiler ducklings ad-libitum. They found that the ducklings preferred high protein to low protein feeds, resulting in excess protein intake, and higher protein conversion ratios. From this study it was concluded that choice feeding is not an economically viable system for growing meat ducks.
Choice-feeding can be an easy way for small flock owners to feed their laying hens (NSWA 2002). According to Bennett (2003) when hens are choice-fed, grinding the grain or mixing it accurately with a supplement are not necessary. Instead, separate feeders for the grain and supplement should be provided and the birds allowed to choose how much of each they want to eat. If the feed mill making the supplement recommends extra limestone or oyster shell, a third feeder full of one of these calcium sources can be provided. Laying hens can make good nutritional choices and pick out the amounts of grain, supplement and limestone or oyster shell that they need to be healthy and productive. According to Bennett (2003), hens fed this way will lay as many eggs as hens fed a traditional laying hen diet as long as some basic rules are followed:
1. Do not give the hens too many choices. Hens can handle up to three choices quite well (grain, supplement and limestone or oyster shell). If more than one grain is used, such as wheat and barley, they should be mixed together in the same feeder.
2. Give the hens choices that are nutritionally distinct. For example, grain is high in starch and energy, the supplement is high in protein and vitamins and limestone is high in calcium. When provided such clear choices, the hens learn which feeders to go to and how much to eat to meet their basic nutritional needs. Some choices may not be clear enough for the hens. For example, wheat and peas both are high in starch and have moderate levels of protein, and having separate feeders of wheat and peas may not provide a distinct enough nutritional difference for the birds.
3. Introduce the whole grain and choice-feeding a month before the onset of lay (at about 15 weeks of age for commercial layers and 24 weeks for village hens). This adjustment period will allow the birds time to learn how to choice-feed themselves before they are exposed to the nutritional demands of egg production. It will also allow the pullets the opportunity to increase their calcium consumption and build up the calcium reserves in their bones before they start to lay eggs. Finally, it takes the gizzard three weeks to build muscle mass to enable the hen to be able to efficiently grind the grain once egg production begins.
4.Vitamins or micro-minerals (e.g. copper, zinc etc.) should not be provided in a separate feeder, rather mixed with the supplement. If vitamins or micro-minerals are placed in a separate feeder, some birds may not eat them because they do not like the taste while other birds may overconsume them and suffer toxic side effects.
5. Birds should be given adequate feeder space. For large backyard flocks, several feeders for each ingredient may be needed. For a one hundred-hen flock, two hanging feeders each of grain, supplement and limestone are suggested.
6. A supplement designed to be mixed with grains or grains and limestone to provide a complete laying hen diet should be used. A supplement formulated in this manner will contain a range of 25% to 40% CP. A grower supplement may be used prior to the start of egg production but a laying hen supplement used once the birds begin laying eggs.
7. It is not necessary to grind grains when choice-feeding hens. The birds will readily eat whole wheat, whole oats or whole barley (but they can have difficulty eating whole maize). After about three weeks of eating whole grains, the hens' gizzards will increase in muscle mass and will grind the grain as efficiently as a hammer mill. Hens can successfully consume 70% of their diet as whole grain when it is choice-fed. It is important to note that if the grains, supplements and limestone are provided in different feeders, these separation problems are avoided.
Balog J M and R J Millard 1989 Influence of the sense of taste on broiler chick feed consumption. Poultry Science. 68:1519-1526.
Bennett C 2003 Choice-feeding of small laying hen flocks. Manitoba Agriculture and Food, University Crescent Winnipeg.
Blair R, Dewar W A and Dowine J N 1973 Egg production responses of hens given a complete mash or ungrounded grain together with concentrate pellets. British Poultry Science 14:373.
Bui Xuan Men, Ogle B and Lindberg J E 2001 Effect of choice feeding on the nutrient intake and performance of broiler ducks; Asian-Australian Journal of Animal Science 14:1728-1733.
Burel C, Ciszuk P, Brännäs E, Wiklund B, Kiessling A and Liljedahl L 1999 Study of the individual feed choice in a group of hens using an automatic registration syste. Proceedings from NJF Seminar No 303. Ecological Animal Husbandry in the Nordic Countries, Harsens, Denmark 303.
Ciszuk P, Charpentier L and Hult E 1998 Free-choice of feed for ecological hens. FAKTA. Jordbruk. Sammanfattar aktuell forskning vid SLU. Nr.
Cumming R B 1987 The effect of dietary fibre and choice-feeding on coccidiosis in chickens. 4th edition. Asian Australasian Association of Animal Production Societies Congress, Hamilton 216.
Cumming R B 1989 Further studies on the dietary manipulation of coccidiosis. Australian Poultry Science Symposium, University of Sydney 96.
Cumming R B 1992a The advantages of free-choice feeding for village chickens.19th World's Poultry Congress, 627.
Cumming R B 1992bThe biological control of coccidiosis by choice-feeding. In Proceedings of XIX World's poultry Congress, Amsterdam. P525-P427.
Dana N and Ogle B 2002 Effects of scavenging on diet selection and performance of Rhode Island Red and Fayoumi breeds of chicken offered a choice of energy and protein feeds. Tropical Animal Health and Production. 34:417-429.
Elkin R G, Ndife L I and Rogler J C 1985 Dietary self-selection and the regulation of protein and energy intake in chicks; Physiology and Behavior. 34:743-749.
Emmans G C 1977 The nutrient intake of laying hens given a choice of diets in relation to their production requirements. British Poultry Science.18:227.
Emmans G C 1991 Diet selection by animals:Theory and experimental design. Proceedings of the Nutrition Society. 59-64.
Emmerson D E, Denbow D M and Hulet R M 1990 Protein and energy self-selection of turkey breeder hens: Reproductive performance. British Poultry Science. 31:283-292.
Emmerson D E, Denbow D M, Hulet R M, Potter L M and Van K P 1991 Self-selection of dietary protein and energy by turkey breeder hens. British Poultry Science. 32:555-564.
Engku A E A and Forbes J M 1989 Growth, food intake and energy balance of layer and broiler chickens offered glucose in the drinking water and the effect of dietary protein content. British Poultry Science. 30:907-917.
Farrell D J, Hamid H and Hutagalung R I 1981 Free-choice feeding of laying hens in the humid tropics. Tropical Animal Health and Production. 6:1.
Forbes J M and Covasa M 1995 Application of diet selection by poultry with particular reference to whole cereals. World's Poultry Science Journal. 51:149.
Forbes J M and Shariatmadari F S 1994 Diet selection for protein by poultry. World's Poultry Science Journal. 50:7-24.
Gous R M and Dupreez J J 1975 The sequencial feeding of growing chickens. British Journal of Nutrition. 34:113-117.
Gous R M and Swatson H K 2000 Mixture experiments: a severe test of the ability of a broiler chicken to make the right choice; British Poultry Science. 41:136-140.
Graham J C 1934 Individuality of pullets in balancing the ration. Poultry Science. 13:34-39.
Haskell M J, Vilarino M, Savina M, Atamna J and Picard M 2001 Do broiler chicks have a cognitive representation of food quality? Appetitive, behavioural and ingestive responses to a change in diet quality. Animal Behaviour Science. (72) 1(2):63-77.
Henuk Y L and Dingle J G 2002 Practical and economical advantages of choice feeding systems for laying poultry; World's Poultry Science Journal. 58: 199-208.
Holcombe D J, Roland D A and Harms R H 1975 The ability of hens to adjust calcium intake when given a choice of diet containing two levels of calcium. Poultry Science. 54:552-261.
Hughes B O 1979 Appetites for specific nutrients in food intake regulation in poultry. British poultry Science 141-169.
Hughes B O and Wood-Gush D G M 1971 A specific appetitive for calcium in domestic chickens. Animal Behaviour 19: 490-499.
Jacobs H L and Scott S 1957 Factors mediating food and liquid intake in chickens. Study on the preference of sucrose and saccharin solutions. Poultry Science. 8-15.
Jones R B 1986 Responses of domestic chicks to novel food as a function of sex, strain and previous Experience. Behaviour Processes. 12:261-271.
Karunajeewa H 1978 The performance of crossbred hens given free choice feeding of whole grains and a concentrate mixture and the influence of source of xanthophylls on yolk color. British Poultry Science. 19:699.
Launchbaugh K L, Provenza F D and Werkmeister M J 1997 Overcoming food neophobia in domestic ruminants through addition of a familiar flavor and repeated exposure to novel foods. Applied Animal Behaviour Science. 54: 327-334.
Mastika I M 1981 Studies on free-choice feeding of growing chickens. M.Sc. Ag. Thesis, University of New England, Armidale, Australia.
Mastika I M 1987 Some basic principle underlying free-choice feeding of growing chickens. Ph.D. Thesis, University of New England, Armidale, Australia.
Mastika I M and Cumming R B 1985 Effect of nutrition and environmental variations on choice feeding of broilers. Recent Advances in Animal Nutrition in Australia.University of New England Publishing Unit, New England 19.
McIntosh J I, Slinger S J, Sibbald I R and Ashton G C 1962 Factors affecting the metabolizable energy content of poultry feed: The effect of grinding, pelleting and grit feeding on the availability of the energy of wheat, corn, oats and barley. A study on the effect of dietary balance. Poultry Science 41:445.
Mongin P and Sauveur B 1974 Voluntary food and calcium intake by the laying hens. British Poultry Science 15:349-359.
Newman R K and Sands D C1983 Dietary selection for lysine by the chicks. Physiology and Behaviour 31:13-20.
NRC 1984 Nutrient requirements of poultry. 8th revised edition, National Academy Press: 8.
NSWA 2002 Small-scale poultry keeping-feeding. Agfact A5, 25.
Olver M D and Malan D D 2000 The effect of choice feeding from 7 weeks of age on the production characteristics of laying hens. South African Journal of Animal Science. 2:30.
Scott T A and Balnave D 1989 Responses of sexually-maturing pullets to self-selection feeding under different temperatures and lighting regimes. British Poultry Science 30:135-150.
Scott M L, Nesheim M C and Young R G 1982 Nutrition of the chicken. Scott ML, Associate editors, Ithaca USA.
Summers J D and Leeson S 1978 Dietary selection of protein and energy by broilers. British Poultry Science 19:425-430
Received 8 January 2005; Accepted 22 January 2005; Published 1 April 2005