converting website visitors - news, features, articles and disease information for the poultry industry

Featured Articles

Recent Advances in the Production, Management and Nutrition of Intensively-farmed Domestic Ducks

13 May 2015

A review of the global duck sector, looking important areas of the production of meat- and egg-type as well as Muscovy ducks by David J. Farrell, Research Consultant to the School of Agriculture and Food Sciences at the University of Brisbane, Australia. There is a special focus on duck nutrition, including trial work that has not previously been published.


  • Per-capita duck meat consumption is 600g per year and is increasing at 3.4 per cent per year. Most ducks (83.5 per cent) are in Asia; of these 79 per cent are grown in China. Growth rate, feed efficiency, reduced carcass fat and increased breast meat yield are improving at an impressive rate.
  • Per-capita duck egg production is about 7.6 per cent of total world egg production. Most are consumed in China. Annual consumption in Thailand is 4.7kg per person per year.
  • Ducklings grow very rapidly. At 14 days, they are twice as heavy as broiler chickens. They drink copious amounts of water and have a much higher capacity to digest oil and probably fibre than chickens at all ages. Throughout growth, dietary energy concentration remains high but crude protein can be gradually reduced, giving opportunity for a specific starter, grower and finisher diet.
  • Most requirements for total amino acids are not determined but are taken from chicken data. To convert total to digestible amino acid requirement using a single factor is not recommended. Here, the author presents values for individual ileal amino acid digestibility based on coefficients determined with chickens for each essential amino acid. However, new data with ducks are also presented here on amino acids and the apparent metabolisable energy of some common feed ingredients.
  • Problem areas in the management of ducks particularly when they are young are: toxins in feedstuffs to which ducklings are particularly sensitive; dietary requirements especially for available phosphorus; floor type and litter. Inability to stand is not just related to nutrition but may also be genetic. Although hardy, there are a few specific diseases; ducks are silent carriers of the highly pathogenic avian influenza virus. This is of major concern and outbreaks are on the rise in poultry.
  • As ducklings age, feather-pecking can become a problem. Low lighting and adequate floor space are important preventive factors. Diet may also be an issue.
  • Independent comparisons are provided in two separate studies of two popular strains of Pekin ducks and their crosses and showing significant differences for most parameters measured in summer and winter. There was no advantage in keeping the sexes separate.
  • Nutrient specifications from various sources are updated as far as possible for meat-type and egg laying ducks. Amino acid and profile in duck meat and amino acid pattern relative to lysine are given. Ducks much prefer pelleted to mash diets.
  • Muscovy ducks are not closely related to domestic ducks and the male is often grown to 4kg to produce lean, red meat not unlike beef. Females make good mothers and in some systems are often used to incubate eggs. Nutrient requirements are given, although they are mostly similar to Pekin ducks.
  • Results from two choice feeding experiments using a combination of any two of four diets that differed in crude protein and apparent metabolisable energy (AME) gave interesting results. Those combinations with a high energy diet were consistently chosen and protein became less important for ducklings. For growing ducks the selections were less clear. A second experiment on a commercial farm again showed the importance of diets with a high AME.
  • The final section focused on research into feed enzymes, especially phytase in duck diets high in byproducts and particularly in rice bran used extensively in Asian duck feeding systems. There were additional benefits in addition to the release of phytin phosphorus.
  • It is emphasised that feed enzymes are not only beneficial to livestock but are also friendly to the environment; a factor that is of increasing importance today.


David Farrell
Dr David Farrell

Wild ducks have been part of man’s diet for many centuries. Australian Aboriginal people captured them for their meat and eggs. As far back as 1550BC, the Egyptian tombs had images of ducks that suggested that they were then hunted for food and the Romans raised wild ducks for festive occasions. The Chinese domesticated ducks some 3,000 to 4,000 years ago.

In Europe and Great Britain, domestication probably did not occur until around 1150 AD.1 It is generally agreed that all domestic ducks are derived from the green-headed wild Mallard (Anas platyrhynchos platyrhynchos) and should not be confused with Muscovy or Barbary ducks (Cairina moschata). These belong to the genus Cairina and originate in South America where they lived in trees. This breed, which has dark, red meat, has changed little as a result of domestication other than in body size and growth rate. We will deal with the Muscovy duck later.

Ducks are difficult to sex but at about four weeks of age, the tail feathers of the male only are curled. In commercial practice for meat production, the sexes are not normally separated, although at about 3kg live weight, the males are about 200g heavier than females. This is in line with expectations of the popular Cherry Valley Super M3 breed.

As a result of domestication, numerous breeds have evolved for their meat and eggs or both. Domestic ducks are fully feathered at about 60 days of age and feathers are an important by-product yielding about 100g per bird; of this about 30g is down.

Production of liver pate (foie gras) achieved by force-feeding ducks and geese is now a welfare issue, and will not be dealt with here.2. France produces almost 20,000 tonnes of foie gras per year and mainly from Muscovy ducks.

Duck-rice production systems are common in many Asian countries when the herded ducks forage around the seedlings, controlling weeds, adding manure and effectively controlling insects.3 After the rice is harvested, they feed on the fallen grains. The ducks are usually egg layers that have been selected with special foraging characteristics but meat-type ducks are sometimes raised under these conditions. In this system, ducks are confined overnight. During this time, they lay their eggs.

As a result of selection for meat and egg production, most breeds have lost the ability to incubate their eggs; as a result there have been some ingenious methods, especially in Asia, to replace the broody bird. However, the Muscovy duck is an excellent brooder and a good mother. They are widely used in these small-scale production systems. Because of changes in growing rice and new disease outbreaks, this age-old system is in decline.

The author will focus here only on hybrid ducks raised for their meat and eggs and kept mainly under intensive management systems.

David Farrell University of Queensland

Although only four per cent of world meat consumption, the global commercial meat duck consumption (Table 1) is growing at a faster annual rate (3.4 per cent per year) than the chicken meat industry, in part due to increased prosperity, particularly in Asia where 83.5 per cent of the industry occurs.

World per-capita consumption is 600g per year but for China, which produces 79 per cent of Asian duck meat, it is 2kg.

Special Characteristics of Ducks

Digestive physiology

The digestive system of the duck has been described, and unlike the chicken it does not have a diverticulum and the crop is merely a widening of the oesophagus, as opposed to a storage organ in fowl. In the duck, this controls movement of feed to the proventriculus and may be responsible for the rapid rate of digesta passing through their digestive tract.

Gastric digestion occurs in the gizzard.

There is not much information available on fibre digestion (fermentation) but ducks appear to cope well with fibre in their diet. One experiment5 compared the digestibility of fibrous components in rye grass, and found that adult Pekin ducks digested 28.3 per cent of the crude fibre, which was similar to adult geese and mule ducks.

David Farrell University of Queensland

Shown in Table 2 are Malaysian data on Muscovy ducks fed diets containing palm kernel cake and indicating a significant ability to digest the various fibrous components of the cake even at inclusions of 35 per cent.6

Growth rate

Ducks have a more rapid growth rate than chickens. Typical body weights are shown in Table 3.

David Farrell University of Queensland

The rapid growth of ducks means that they need a high-quality diet, particularly during the early stages of growth, and leg problems are not uncommon. However ducks at slaughter age have more fat than chickens and less breast meat. Chinese Roasted Peking Duck depends on significant fat under the skin, when air is pumped between the skin and the fat before cooking. Demand is now for more lean meat and less fat, and this is the main focus of most breeding programs. Performance of male Pekin ducks is improving and male ducks may reach 2.4kg in 35 days with FCR under 2.0.

Brooding and drinking

Duck eggs hatch out in 28 days. They do not need to be brooded for as long as chickens, sometimes for only 10 to 14 days, and often not at all in the tropics.

They also drink copious amounts of water, probably to propel food along their digestive tract and consequently produce watery excreta. This has implications both in quality of litter and its management. Litter is usually chopped straw, rice hulls, sawdust or wood shavings, although these have been shown to cause foot ailments and breast blisters. As ducks tend to drink water frequently and spill some, location of drinkers is important. Where possible, out-of-doors drinkers are best or, if indoors, on bare floor when the spilled water can drain into the under-floor pond. Litter should be turned or renewed on a regular basis.

Ducks tend to drink more in the summer months than in winter. One study7 showed on average a water to feed ratio of 3.3:1 in summer months and 2.7:1 in winter, and only small differences between breeds and the sexes. Here, nipple drinkers delivered the water although ducks seem to favour bell drinkers over nipple drinkers because they can become involved in such behavioural activities as dabbling and head dipping.

The author found that ducklings aged 5 to 22 days had a much higher water to feed ratio of 4.1:1 compared to 2.3:1 for meat chickens aged 11 to 23 days.8


Ducks are hardier and more resistant to disease than chickens, consequently they have fewer disease outbreaks.

Newcastle disease, endemic in poultry, is rarely seen in ducks. However, highly pathogenic avian influenza, which has decimated poultry flocks, especially in Asia, does not seem to affect ducks. They are, however, silent carriers of the often virulent virus, so ducks can infect other poultry. This has led to major changes in management of village ducks under traditional especially herded ducks in the warm, humid tropics.9

Escherichia coli is by far the most common disease in ducks. Poor farm hygiene is a common cause.

Infections with Pasteurella multocida are usual, especially in Asia and also Rimerella (Pasteurella) anatipestifer.

Hygiene and biosecurity are early precautionary measures.10

Ducks are highly susceptible to mycotoxins, especially aflatoxin and particularly when young.11 This is a continuing problem and not always recognised, as ducks may not always die but performance may be reduced. A toxin binder should be routinely incorporated in duck diets.

Bathing water

There has been much discussion about allowing ducks access to bathing water. Although the consensus is, ‘not essential’, ducks are commonly given access to water on most Asian duck farms. This allows the ducks to remain cool and probably helps to maintain feed intake.

A duck loses heat mainly through its bill and feet. These ponds can also solve the problem of waste disposal when the droppings are flushed into the pond where fish may be part of the production system. With open ponds, biosecurity is compromised.

Housing and floors

Today, ducks are sometimes housed in fully-enclosed houses where climate is controlled. But these houses are rarely found in the warm tropics where open-sided houses are common and the ducks are on slatted, mesh-wire or synthetic weave floors located over pits which can be drained. A mixture of slatted (30 per cent) and straw (70 per cent) from 14 days to slaughter is not uncommon. An all-slatted floor is not permitted under EU regulations.

Foot lesions can occur on mesh wire floors and mesh wire size will change with body size. All-wire floors are not recommended for developing duck breeders or laying ducks by Scott and Dean11, still the gurus of duck management and nutrition. Good ventilation is important, as ammonia production is high and this can cause metal corrosion. It should be kept below 15ppm. Curtains are sometimes used to separate ducks of different ages in the one house or to isolate them for incubation purposes.

Comfort zone of mature ducks is about 8 to 26°C. In commercial practice after 21 days of age, 21°C is recommended. Fan ventilation may be necessary to reduce shed temperature and ammonia levels in enclosed housing systems.


Feather-pulling can be a serious problem, especially in Muscovy ducks.

There are many factors involved: too high a stocking density, floor type (slatted floors), too high light intensities, diet form and diet composition. Low-protein diets, or diets with insufficient of the essential amino acids (especially methionine) or other nutrient deficiencies, too high dietary energy or too low a fibre level can all contribute to cannibalism.

There is no one definitive answer to preventing cannibalism.11 Bill-trimming at about seven days of age or heat treatment at day-old is a last resort. Reducing light intensity and introducing red or blue light bulbs may be beneficial.12

Duck breeding

It is outside the scope of this review to discuss in detail, as this is a highly specialised area.

Today, commercial lines have been selected for growth rate and carcass composition; for egg production and there is the ‘mule duck’ which is usually an intergeneric cross between the Muscovy (Barbary) duck and the common duck. Although the offspring are sterile, they grow much faster than their parents due to hybrid vigour.

Today, two companies tend to dominate the global duck breeding market: Cherry Valley Farms Ltd in the UK (now Thai-owned) and Groupe Grimaud Freres (GF) in France. That is not to say that there are not numerous other duck breeding companies. Most breeding companies keep information in-house but there is more information on these two companies than most others. Cherry Valley (CV) recently claimed that 60 per cent of ducks consumed in China came from Cherry Valley stock. It is suggested that GF stock grow to a larger size than CV but mature later and the meat may be of lower eating quality.7

A comparison was made of these two commercial strains without identification by Hay and Scott (2007)13 who recorded averages of 3.3kg and 3.5kg at 47 and 49 days, respectively, with a FCR of about 2.35:1 and 2.40:1. Dressed carcass was about 2kg. Both commercial companies select from strains of white Pekin ducks. Breast meat yield is usually an important parameter and was 12.4 and 13.6 per cent for the two breeds but this depends on whether the skin is left on or removed, giving different yields.

Typical body composition is: meat 23 to 24 per cent of the carcass, fat and skin 33 to 34 per cent and bone 31 to 32 per cent but this will depend on the market requirements. As mentioned, the Chinese sometimes like a fat carcass, although Cherry Valley maintains that their meat ducks contain only 27 per cent fat. This is an important component of the flavour of the meat.

More recent results from the same research organisation (The University of Sydney), looked at these same breeds and their crosses (Cherry Valley and Grimaud Freres) but without identifying breeds.7 Dietary allowances were generous, and breeds were either mixed or raised separately in floor pens (1.5 × 3.0 metres) and each treatment was replicated four times.

David Farrell University of Queensland

Results in Table 4 show no differences in performance between keeping the sexes separately and those in mixed pens. Data are therefore combined. Males were consistently heavier than females. Because of customer requirements, ducks were grown only to about 2.8kg live weight.

Interestingly, there was a marked difference in growth rate between the two strains both in winter (415g) and summer (258g) but FCR was the same. FCR was different among the groups but not in winter. Strain B was 300g heavier in winter than in summer while for strain A, the difference was under 150g. FCR was about 2.00 for all strains and crosses in summer but consistently well over 2.5 in winter.

The A×B cross showed a consistent response in live weight compared to strain A but when strain B was crossed with strain A, only FCR decreased in the summer and was significantly lower than other groups.

Floor space

Adequate floor space, particularly after 17 days of age, is extremely important, not only for optimum growth but for feed intake, economic performance, welfare, health and hygiene and litter quality.

Cherry and Morris (2008)14 provided detailed research in a temperate and cool climate (the UK) into 17-day-old ducks on slats and litter. When birds were given a floor space above 4.1 square metres per bird, they showed a gradual decline in live weight to 46 days as floor space was decreased to 8.5 birds per square metre; growth rate had declined by 10 per cent. However, in practice, floor space is not always constant and can vary with season. Space may be increased as ducks grow by using partitions that are moved as they age. Also, it may be increased as ducks grow by using partitions that are moved as they age.

Grimaud Freres’ management manual15 differentiates between duckboards and litter for stocking densities when ducks (mixed sexes) are over 3kg live weight (Table 5).

David Farrell University of Queensland

Cherry and Morris14 found that live weight gain was maximum at about 135g per day when ducks are 3.5 weeks of age and grown to 12 weeks. It declines rapidly after that age. After 21 days, growth was best at a mean 10 to 12°C as was feed intake highest (230g per day). Ducks do best in very low humidity.

Thus duck performance is unusually high, and under commercial conditions, mixed sexes growth rate may reach 100g per day at 35 days.15

Some useful general equations have been generated by Cherry and Morris14 for ducks between 28 and 42 days of age:

  • Daily feed intake (g) = 295.2 – 4.752 x (x, mean daily maximum and minimum temperature °C)
  • Daily rate of gain (g) = 110.52 + 1.481x - 0.002 X2 (x, feed intake in g)
  • Live weight (kg) = 3.56 + 0.007x - 0.0016x² (x, °C)
  • Feed intake (kg) = 8.481 - 0.021x- 0.025 x² (x, °C)

Plucking and cooking yield

Defeathering of ducks is difficult and in three stages: scalding, plucking and finally waxing to remove pin feathers. This was difficult but today, plucking and processing of ducks is now completely automated.

Several years ago, the author looked at the composition of duck carcasses given various treatments16. It was found that, when starved for six hours before slaughter, the birds lost 200g and FCR was 2.43 compared to 2.32 on full feed. Dressing out was 65 per cent, cooking yield 63 per cent and lean yield is 60 per cent of cooking yield, giving 737g of edible meat from a 3-kg duck.

Egg Laying Ducks

Ducks are prolific layers. Substantial progress has been made over the past 25 years.18 After 46 weeks of lay, modern hybrids produce about 250 eggs with a fertility of over 92 per cent. In western countries few duck eggs are eaten but not so for some Asian countries, particularly in China where large numbers are consumed.

Shown in Table 6 are data for ‘other eggs’ of which almost all are duck eggs. Most (95 per cent) are consumed in Asia. About one-third of all eggs eaten in Thailand are duck eggs. In the Philippines, ‘balut’ eggs are 90 per cent of all duck eggs produced. These are fertile duck eggs incubated to 19 days then cooked and the embryo consumed with various condiments. Other forms of processed eggs are ‘salted’, ‘1000 year eggs’ and ‘alkalized eggs’.

David Farrell University of Queensland

Nutrition of Meat-type Ducks

Compared to chickens, little research has been undertaken with ducks and it is usually assumed that there is little difference in the apparent metabolisable energy (AME) of feedstuffs between the two species.19

This may not always be true. Siregar et al (1986)20, in three experiments, showed that unlike chickens, the AME of conventional duck diets declined as they aged.

On the other hand, we have shown that ducks can digest 70 per cent of the oil in rice bran when only three to seven days of age21. The AME value of oil in rice bran increased from 15.6MJ to 17.9MJ per kg at 17 to 21 days of age. Corresponding values for chickens at similar ages were 10.5 and 13.0MJ per kg.

Historically, many feed formulators relied on the US National Research Council’s Nutrient Requirements of Poultry, last revised in 1994, but these are mostly not now relevant to the modern genotypes.22

Dietary energy

A high-energy diet is important particularly for the newly-hatched duckling. Keeping dietary energy consistently high throughout growth is critical for best economic returns. The substantial research by Cherry and Morris14 found that over 12MJ AME per kg feed was best to maximise growth, feed efficiency and carcass characteristics. This figure is in line with the recommendations for minimum feeding standards for Cherry Valley’s SM line and for Grimaud Freres’ Star 53 medium body weight birds.

There follow the values for starter, grower and developer ducks.23 This is in line with the nutrient recommendations by Creswell24 who has published nutrient recommendations for meat-type ducks in these three categories: starter to two weeks, grower from two to four weeks and finisher to market weight.

Although his values for energy are a minimum of 12MJ AME per kg, they do increase slightly as ducks age.25 In commercial practice, recommendations are from a minimum of 12MJ per kg usually increasing to 13MJ for grower/finisher stock. Although not said, it is assumed that values here and elsewhere are on an ‘as-fed’ basis.

Unlike dietary energy, dietary protein can be reduced markedly after about 14 days. For the first 14 days, a 21 per cent crude protein diet, reduced thereafter to 15.7 per cent, gave better final live weight and FCR at 42 days than when the low-protein diet was fed throughout.14

A consensus of crude protein values in the literature are: 20 to 21 per cent, 16 to 18 per cent and 14 to 15 per cent for starter, grower and finisher ducks, respectively.

Although the early days are very important, when ducklings were given diets with 12.6MJ AME per kg and only 16 per cent crude protein (CP) to six, eight or 10 days and then starter diets all with 12.9MJ AME per kg but with 21.5 per cent CP, grower with 17.2 per cent CP and finisher of 16 per cent CP, those ducklings on the lowest CP diet to six days had the highest body weight at 48 days but the worst FCR of 3.46kg and 2.277, respectively.14 Those on a standard programme (controls) values were 3.54kg and FCR of 2.34. It is probably, therefore, better not to tinker with the dietary crude protein for very young ducklings.

Recent AME values for feedstuffs26 determined with 21-day-old Pekin ducklings using the total collection method, and the coefficient for digestible crude protein are given in Table 7.

David Farrell University of Queensland

For comparison, recent US AME data22 are: maize 13.0 to 14.1; rice 14.3; and soybean meal 12.6. The differences may be that the US data are on an ‘as fed’ basis and not dry matter (DM) shown here.

Amino acids

Lysine and methionine are usually the most critical amino acids and are expressed as a percentage of the diet although there are other more meaningful ways (g per day). Creswell25 has expressed his requirement in ‘digestible’ amino acids using a constant factor of 0.85. This has been used here to convert back amino acid data to ‘total’ as the factor varies according to ingredient composition.27 Lysine values ( per cent) become 1.11, 0.84 and 0.73 for his three categories for feeding growing ducks.

Cherry and Morris14 have also used available amino acids but have not given their conversion factor to calculate the digestible value. These authors showed a small increase in daily gain with increasing dietary lysine up to 35 days but no further response thereafter. All data are for Pekin ducks. However, they did show that when they increased the dietary energy level to 13.2MJ per kg in a diet of 18.6 per cent crude protein, the age at slaughter was reduced to meet a bodyweight of 3.2kg, although this may not be an economical option.

Although not new information, Scott and Dean11 reported for Pekin ducks a lysine content of 1.10 and 0.77 per cent for starter and grower diets, respectively, and both with 12.5MJ AME perkg. They demonstrated what is widely known, and that is, as dietary AME is increased, feed efficiency improves. Information is required on amino acids needed to improve breast meat yield, to reduce carcass fat and not just growth rate and FCR.

David Farrell University of Queensland

Using a constant value to convert total to digestible amino acids is fraught with dangers. Shown in Table 8 is a comparison of average dietary digestible values and for lysine and methionine in chickens, ducks and geese in two different experiments but on the same diet. Differences between species is substantial and also between the two experiments. It is for these reasons that total values will be used in Table 11. These were calculated from the original digestible amino acid values in parentheses by applying a constant factor.

David Farrell University of Queensland

Shown in Table 10 are recently determined apparent ideal digestibility amino acids for some feedstuffs measured in growing Pekin ducks.28 Interestingly, values for threonine and tryptophan are the lowest and this should be taken into consideration when formulating duck diets. However measurements on ducks from Purdue University did not find these low values for threonine and tryptophan.22

David Farrell University of Queensland
David Farrell University of Queensland

The data in Table 12 are of interest not only because duck meat tends to have a higher amino acid profile (concentration) than chicken meat, but the amino acid pattern varies considerably between the two species This pattern is a good indication of net amino acid dietary requirements.

David Farrell University of Queensland

Minerals and vitamins

Values used for ducks are mainly taken from chicken data. However in one review,14 some differences were found. Phosphorus (P) is particularly critical in early growth as leg problems and rubbery beaks can easily occur.11 Calcium (Ca) levels of 1.0 per cent are high and recommendations of 6 to 8g per kg for Ca, and 6.5g per kg for total P have been made with available P at 4.3g per kg.¹¹

Selenium at 0.24mg per kg and manganese and zinc at 50 and 60mg per kg, respectively, are suggested. Research in Taiwan indicates that copper should be increased from 4mg to 8mg per kg. Ducks have a higher requirement than chickens for vitamin A and nicotinic acid.30

Nutrition of Egg-laying Ducks

Ducks are prolific layers. However there is a dearth of information on this topic although the composition of the duck egg is different from the hen’s egg. It is higher in dry matter (39 versus 26 per cent), fat (14 versus 11 per cent), protein (14 versus 13 per cent) and usually much heavier. One would expect nutrient specifications to differ and indeed they do.

Shown in Table 13 are some nutrient specification for laying/breeder ducks. After four weeks of age, these generally decline until just prior to lay.

David Farrell University of Queensland

In Table 14 are the nutrient requirements of ducks in lay from three different sources. There is reasonable agreement amongst authors although several values are higher than for today’s hybrid brown egg layers.

David Farrell University of Queensland

Feed form

Ducks prefer pelleted or crumbled feed to mash which tends to form a sticky cake on their bill. They then make frequent trips to watering points. A firm pellet size is important.33

Up to two to three weeks of age, 3mm in diameter and 10mm in length and after this age, 4mm in diameter are recommended and 15mm long with fines restricted to about five per cent of the total feed.

Nutrition of Muscovy (Barbary) Ducks

As mentioned, the Muscovy duck is not closely related to the domestic duck and is a different genus. It is very popular, especially in France2 where much of the research has been undertaken.34,35 The meat has less fat and is dark red. Fillet yield is about 700g. This is about 70 per cent of the duck’s value while the thighs and shanks are 27 per cent.2 It is for this reason that male ducks are grown to 4.5 to 5.5kg in about 84 days with a FCR of about 2.75 and fillet weight of 16 per cent of live weight.18 If killed at 51 days, bodyweight would be 3.3kg and FCR 2.0.

The red meat from the very heavy male duck competes with the meat from the beef industry to produce duck breast steaks. The female Muscovy is much smaller and reaches only 2.4 to 3.0kg in 68 days. She lays up to 250 eggs in 46 weeks and fertility is over 90 per cent. The eggs take 35 days to hatch compared to 28 days for other duck breeds. Muscovies are sometimes used to incubate eggs in village duck production systems and are good mothers.

Genetic progress has been rapid over the past 42 years.2 Today, bodyweight at slaughter is about 5.5kg (52.8 per cent); breast muscle plus skin, 1007g (18.3 per cent); thigh + shanks, 803g (14.6 per cent) and abdominal fat, 83g (1.5 per cent) .

Muscovy ducks are sometimes crossed with Pekin ducks to produce a sterile offspring (Mule or Moulard) with more meat and less fat. Muscovies are more adaptable to hot weather than most other breeds.11

The Muscovy duck has a slightly curved upper bill or hook to allow it to harvest grass seeds. It can digest protein to a greater extent than the chicken when given a variety of diets. They do not need high-energy diets as they showed no difference in growth rate up to 10 weeks on diets that varied in AME from 10.4 to 13.3MJ per kg.34

They also found that protein requirements declined rapidly from 21 to 15 per cent of the diet between zero to three and six to 10 weeks of age. Although this research was carried out almost 30 years ago, the general observation still holds true.

In a recent report, lower carcass fat and higher amounts of edible parts were found in the Muscovy and Mule duck when compared with the Pekin duck.36

Nutrient requirements

Again the published data are scarce. Scott and Dean11 have published the most detailed requirements for Muscovy ducks for meat production.

The author has provided information on nutrient requirements of meat ducks in Tables 8, 9 and 11. These do not change markedly11 for Muscovy and Mule ducks, other than dietary protein for the starter and grower diets are 20 per cent and 18 per cent, respectively. Energy levels are 11.7 and 12.1MJ per kg for starter and grower diets,respectively. In practice, these are only small differences between requirements for Pekin and Muscovy ducks.

Apart from dietary protein (16 versus 17 per cent) and lysine (0.70 versus 0.75 per cent) for Pekin and Muscovy layers, respectively, other differences in nutrient requirements are negligible for amino acids.

However calcium and phosphorus requirements are about 20 per cent lower than for other breeds of laying ducks. There is insufficient information to recommend any differences in vitamin requirements between the two breeds.

Choice Feeding

Choice feeding or self-selection has been used in poultry as a guide to their nutritional needs especially for protein and energy.37 Unpublished38 experiments with ducks have also been shown to be a useful indicator of their nutritional needs and are shown here for the first time.

There were two experiments. The first was undertaken in 18 pens each holding 20 one-day-old meat-type ducklings.

The four diets were:

  1. high protein (100 per cent amino acid requirements)-low energy (7.5MJ AME per kg)
  2. low protein (60 per cent amino acid requirements)-low energy
  3. high protein-high energy (12.5MJ AME per kg) and
  4. low protein-high energy.

These gave a total of six diets when fed in all combinations to each of three pens of ducklings.

Experiment 1

The highlight of this experiment was that only during weeks 0 to 2 were there any significant (p<0.05) differences between diets.

Here, those combinations of diet 2 (low protein-high energy) with either 3 (high protein-high energy) or 4 (low protein-high energy) gave the highest growth rate.

The interesting result is that the ducklings selected 83 to 84 per cent of the high-energy and only 16 to 18 per cent of the high-protein diet. The unequivocal conclusion is that energy is by far the most important dietary component in diets for ducklings up to two weeks of age.

There were differences in feed intake and FCR which, as would be expected, was lowest (best) on diets containing the high energy choice (1+3, 2+3).

Although the final bodyweight at six weeks was different due solely to differences in growth rates up to two weeks of age, FCR (0-6 weeks) did not differ.

The interesting outcome was that from two to six weeks, the combination of diets 3 and 4 of the two choices was quite different from weeks 0 to 2. Except for the treatment combination 1+2 containing high- and low-protein with low energy, ducks were unable to select between choices offered.

The conclusion is that dietary energy is more important than dietary protein up to two weeks of age. After that, statistically, protein and energy become equally important but the trends suggested that high-energy diets were still important.

Experiment 2

The second experiment was undertaken on a commercial duck farm (Luv-a-Duck, Nhill, Victoria) in open-sided houses with curtain facilities, as needed. Because brooding was in a single, communal group the experiment of 100 White Pekin ducks (Cherry Valley, UK) per pen (3.7 metres × 9.3 metres) ran from weeks 2 to 6.

Diets were as in Experiment 1 except that the farm’s standard formulation was included as a control. Diets were replicated thrice.

Although mortality was six per cent, this was distributed more or less evenly across treatments.

Liveweight at six weeks was measured on ducks starved overnight and all groups exceeded 2.9kg per duck and a carcass weight of over 2.1kg. Statistical differences between groups were therefore very few, indicating that ducks have an unusual ability to select the best combination given any two choices.

When data were combined, ducks on the experimental diets grew faster (p<0.05) than those on the company’s diet (control). The group on diet 4 with the two high-energy combinations grew the fastest, reaching over 3kg at slaughter and with a FCR of 1.91, as would be expected. Results were almost identical to the diet with the low protein-low energy and the high protein-high energy combination, although FCR was 2.06. Even ducks on the diet combination with the two low-energy diets managed to exceed 2.9kg.

Intake on the control diet was similar for both choices as would be expected since each contained the same formulated feed. Only diets D (low protein-low energy/high protein-high energy) and F (high protein-high energy/ low protein-high energy) showed that the ducks made the same choice for each of the four weeks.

Ducks on choice D consistently selected the high protein-high energy feed, while those on choice F also consistently favoured the same diet (high protein-high energy). This again highlights the importance of high-energy diets in duck production.

David Farrell University of Queensland

One of the interesting outcomes of this experiment was the consistent overall intake of energy (MJ of AME) amongst the six groups: Group A, 1,300MJ; Group B 1,388; Group C 1,346; Group D 1,361; Group E 1,380 and Group F 1,258.

There is little doubt that, if allowed, most livestock eat to meet an energy requirement first and foremost.

Lowest calculated lysine intake was observed on choice E, which was 0.81kg; for all other treatments, intake was over 1kg.

The determined carcass fat content on this treatment was the highest of all treatments (55.9 per cent on a dry weight basis) followed by the control group (52.5 per cent) due to insufficient lysine.

The results show that the commercial, standard diet formulated for Luv-a-Duck was consistently inferior to those combinations selected in several choices and raises the question as to whether ducks should not be allowed to select diets to meet their continuing changing nutrient requirements as they age.

The evidence from these two experiments suggests that there may be significant economic potential.

Feed Enzymes in Duck Diets Containing Byproducts

In the early 1990s, interest gathered rapidly in the incorporation of feed enzymes in poultry diets; companies sprang up and papers were read at conferences on recent developments. Today, feed enzymes are routinely included in poultry feedstuffs to improve the digestibility and utilisation of certain feed substrates.

Often there are other unexpected benefits. Although, compared to chickens, research with enzymes in duck diets is relatively sparse, ducks do have a greater capacity to consume byproducts than chickens and therefore feed enzymes have significant potential in this area.

Early work focussed on the non-starch polysaccharides found in grains, such as wheat, barley and rye when viscosity was increased in the gut of fowl and especially in young broilers; this reduced the digestibility of grains, hence their energy value. But it was the early work of Hew et al.(1995)39 that alerted the feed industry to additional benefits when it was shown that when feed enzymes were added to a wheat-based diet, amino acid digestibility was improved right across the spectrum.

The response of layers and broilers to enzymes in corn-soya diets greatly increased their potential inclusion because of the widespread use of this combination of feedstuffs. Starch in corn is only 85 per cent digestible at the terminal ileum of chickens.40

Creswell (2013)41 has compiled some of the studies on the use of feed enzymes in diets of meat and layer ducks with significant amounts of by-products (palm kernel cake, milling grain by-products, broken rice, distillers dried grains and solubles) undertaken mainly in Asia. This review demonstrated that ducks are often more responsive to feed enzymes than chickens, particularly as they can utilise byproducts better than chickens, allowing their higher inclusion in formulated feed.

The author's early work42 on feed enzymes was in duck diets with high amounts of hull-less (naked) oats that contain high levels of beta glucans. These increase viscosity in the gut and reduce nutrient absorption. Results showed a consistent improvement in AME, fat digestibility, growth rate and FCR - even in diets with as much as 73 per cent of oats. Surprisingly, when we added a feed phytase alone, at the highest oat inclusion, feed intake improved and liveweight gain increased by 18g per day over the corresponding, unsupplemented diet and were the same as the control group without oats.

The importance of phosphorus (P) in the diets, especially of ducklings and the author's previous work on rice bran identified this as an important area for research. About 50 million tonnes of rice bran are produced globally each year but not all is fed to livestock. It varies greatly in chemical composition. High-quality rice bran contains 16.2 to 18.1g phosphorus per kg but much of this is in the form of phytic acid P and therefore unavailable. Phosphorus is like fossil fuel, a diminishing and a non-renewable resource that is becoming an expensive component of poultry diets.

Early work by the author showed just how well ducklings, grown from two to 19 days, can utilise rice bran even at inclusions as high as 400g per kg diet.43 With only 1g of added inorganic P per kg, there was a consistent response in growth rate and often in feed intake to the phytase (1000 units per kg) but less so when 3g inorganic P was added, indicating that at this level, the diet was adequate in P.

Rice bran contains substantial amounts of non-starch polysaccharides, predominantly arabinose and xylose but when the appropriate enzymes were added to duck diets, responses in ducklings and growing ducks were not significant.44

Interestingly, those on the 300g rice bran per kg diets performed better and grew faster (p<0.05) with a lower FCR than ducks on the sorghum-based control diet. Growth of those ducks on 600g rice bran per kg had the same growth rate and FCR as those on the control diet. The fact that relative viscosity declined with increased inclusions of rice bran indicates that these NSPs are probably not water-soluble and, therefore, not an impediment to duck performance.

One of the most detailed experiments undertaken on the effects of a phytase supplement (1g or 3g inorganic P per kg ) in ducklings and growing ducks (19 to 40 days) was by Martin and Farrell.45 As previously, up to 600g rice bran per kg replaced sorghum in the control diet and all diets contained 12 to 13MJ AME per kg and 192 to 209g crude protein per kg. Lysine was higher in the older ducks (10.1 versus 8.7 to 9.7g per kg).

In the duckling experiment, all diets, including the control but not the diet with 3g inorganic P gave a significant positive response to the enzyme addition to diets containing 200g and 400g rice bran per kg for growth rate and food intake but not FCR.

Digestibility measurements showed a significant improvement in nitrogen, calcium and phosphorus digestibility with enzyme addition when measured in ducks at two to 10 days and again at 10 to 19 days.

In the second experiment,45 ducks (19 to 40 days of age) again showed a response to the phytase addition in diets containing 0, 300 and 600g rice bran per kg. Growth rate and feed intake declined but FCR remained unchanged with increasing intake of rice bran. The addition of phytase improved growth rate substantially and FCR but not intake.

Digestibility of dry matter declined with increasing rice bran measured at 33 to 40 days of age but phytase had a positive effect on this, P and some other minerals. AME increased from 13.2 to 13.6MJ AME per kg. Nitrogen retention increased from 46.6 to 51.4 per cent but this was almost significant at P<0.1.

The significance of this research is not only the improvement in growth rate, feed intake and FCR but the improved digestibility of several dietary components that is not often recognised in assessing the additional economic benefits of a feed phytase.

Finally, it should be emphasised that diets used here were vegetable-based. The additions of only a small amount of animal protein tends to negate the benefits of the enzyme in rice bran-based diets. However, it was shown46 that the addition of a feed phytase to an all-vegetable rice bran-based diet increased the apparent digestibility of lysine significantly from 76 to 81 per cent and similar to that of the diets with added fishmeal.

An interesting study carried out in Germany,47 in which a feed phytase (Ronozyme P5000) was added in six incremental amounts from 0 to 1,500U per kg (Trial 1) and seven incremental amounts 0 to 2,000U per kg (Trial 2) to a mainly maize-soybean based diets, significantly increased calcium (Ca) and phosphorus (P) retention. In Trial 1, P utilisation increased from 30.4 to 55.0 per cent and Ca was from 39.4 to 54.8 per cent.

In Trial 2, retention increased more than previously but plateaued at the highest phytase inclusion of 2,000U per kg.

A growth trial (Table 16) showed significant responses to increasing the phytase inclusion in both male and female ducks grown to 35 days. Diets contained 0, 1000 and 10,000U per kg of feed phytase. FCR however did not differ.

David Farrell University of Queensland

In summary, these experiments demonstrate the positive benefits of adding a feed phytase to plant-based diets. Response is not just the release of bound phosphorus but the increase in the apparent digestibility of several important nutrients, improved duck performance and increased calcium utilisation. This should be taken into account in diet formulations as excessive dietary calcium may be detrimental to duck performance.

Asia in particular uses rice bran routinely in duck diet formulations where amounts above 200g per kg can be added without any depression in growth rate when a feed phytase is included, although FCR may be compromised slightly.

However, it is not always appreciated that feed enzymes are friendly to the environment at a time when global pollution is now of major concern. Feed phytase releases phosphorus from plant feedstuffs that would otherwise be expelled into waterways causing a blue-green algae bloom (a dangerous toxin) and eutrophication of the oceans. Other feed enzymes will increase feed digestibility and therefore reduce the amount of excreta voided. Any increase in nitrogen utilisation by fine tuning amino acid needs will again reduce nitrification of effluent which can contaminate underground drinking water with nitrates and is known to be a contributing factor to the rapidly declining well-eing of Australia’s Great Barrier Reef.48


A recent review of waterfowl production over the past 100 years49 showed considerable progress but there is substantial room for improvement, particularly in carcass characteristics of ducks.

Unfortunately these reviews are often orientated to the European rather than to the Asian production systems where almost 80 per cent of all ducks are farmed. In the EU, there is little interest in duck eggs. There is the need to clarify just what nutrient requirements should be determined with ducks and just what requirements for chickens can be used satisfactorily for both layer and meat-type ducks.

This review provides considerable new data and will be important to those involved in the many aspects of duck production.

Because the main interest in Muscovy duck production is in Europe, it is often difficult to access recent developments of this breed.

Much of the research by the big duckling producers has shown just how much improvement has been made, particularly over the past 10 to 15 years.

Useful research by the University of Sydney has brought some information into the public arena about the two most popular commercial strains of Pekin ducks and the crossing of these two strains.

The author's own published research and that of Creswell has identified just what feed enzymes have potential in improving performance in ducklings and ducks.

Acknowledgement: This review was commissioned by DSM Nutritional Products (Singapore), which recognised the importance of commercial duck production, particularly in the Asian region where over 80 per cent of all ducks are farmed. Much of the sparse information on many aspects of duck production is in the private sector, so that this review is particularly important because it updates commercial practice and takes the information into the public domain. For further information, contact Rider Perez at DSM.


  1. Hetzel D.J.S. (1986) Domestic ducks: an historical perspective In: Duck Production Science and World Practice. Eds.DJ Farrell and P Stapleton. The University of New England, Armidale, NSW pp 1-5
  2. Guy G. (2013) French waterfowl production. Proceedings 5th Waterfowl Conference of the Asian Pacific Federation of the World’s Poultry Science Association, November 6-12, Hanoi, Vietnam
  3. Petheram R.J. and Thahar A. (1983) Duck egg production systems in West Java. Agricultural Systems 10, 87-97.
  4. Global Poultry Trends 2013: Record World Meat Production in 2013. The Poultry Site. Accessed 1/12/14
  5. Jeroch H., Timmler R. and Guy G. (1997) Comparison between geese, Pekin and mule ducks in order to determine their ability to digest common rye grass. Proceedings 11th Symposium on Waterfowl, Nantes, France. September 8-10, 1997.
  6. Mohamed M. and Alimon R. (2003) Feeding value of palm kernel cake for ducks. Unpublished data.
  7. Downing J.A. and Taylor W. (2010) The effect of strain and season on the performance of commercial ducks under Australian conditions. Proceedings of the Australian Poultry Science Symposium 10, 182-185.
  8. Siregar A.P. and Farrell D.J. (1980) A comparison of the energy and nitrogen metabolism of fed ducklings and chickens. British Poultry Science 61, 213-227.
  9. Farrell D.J. (2014) Small-scale duck production: the way ahead. Journal of Animal Husbandry, Science and Technology (IAHST) No.8, 2014 p73-80.
  10. Johnston A. (2007) Current diseases of ducks and their control. Poultry International p24-28 January 2007.
  11. Scott M.L. and Dean W.F. (1991) Nutrition and Management of Ducks. M.L. Scott of Ithaca, Publisher, New York.
  12. Rodenburg T.B., Bracke M.B.M., Berk J. et al. (2005) Welfare of ducks in European duck husbandry systems. World’s Poultry Science Journal 61, 633-646.
  13. Hay G.C. and Scott T.A. (2007) Growth performance and its prediction in two commercial strains of meat duck, p45-48. Proceedings of the Australian Poultry Science Symposium 19, 45-48.
  14. Cherry P. and Morris T. (2008) Domestic Duck Production Science and Practice. CAB International, Cambridge, Mass, USA.
  15. Grimaud Freres (2010) Rearing Guide Roasting Peking Ducks 2010.
  16. Farrell D.J. (1999) Nutrition and Carcass Quality in Ducks. In eds J. Wiseman and P.C. Garnsworthy, Recent Developments in Poultry Nutrition 2, p203-226. Nottingham University Press, UK.
  17. FAOSTATS, .
  18. Grimaud F. (2008) Past, present and future duck breeding. International Poultry Practice. 22(7)21-25.
  19. Farrell D.J. (1986) Energy expenditure of laying ducks: confined and herded. In eds Farrell DJ and Stapleton P: Duck Production Science and World Practice, p70-82. The University of New England, Armidale, NSW.
  20. Siregar A..P, Farrell, D.J. and Cumming R.B. (1982) The nutrition of meat-type ducks. III. The effects of fibre on biological performance and carcass characteristics. Australian Journal of Agricultural Research. 33, 877-866.
  21. Martin E.A. and Farrell D.J. (1995) Proceedings of the Australian Poultry Science Symposium 7. University of Sydney, NSW.
  22. Adeola O (2006) Review of research in duck nutrient utilization. International Journal of Poultry Science. 5(3):201-218.
  23. Feeding Standards For Ducks (undated in Chinese). Cherry Valley
  24. Creswell D. (2002) Nutrition of intensively raised ducks - Part 2. Proceedings 10th Annual ASA Southeast Feed Technology Workshop, May 27-31, Merlin Beach Resort, Phuket, Thailand.
  25. Creswell D. (2013) Feeding ducks for maximum performance. Part 1. Asian Poultry. April 2013 p10-12
  26. Hoai H.T., Kinh L.V., Viet T.Q. et al. (2013) Determination of the metabolizable energy content of common feedstuffs in meat-type growing ducks. Proceedings 5th Waterfowl Conference of the Asian Pacific Federation of the World Poultry Science Association November 6-8, 2013 Hanoi, Vietnam.
  27. Helmbrecht A. (2012) Amino acid knowledge lacking in duck meat production nutrition. World Poultry (9), 28, 17-19.
  28. Kinh L.V., Hoai H.T., Viet T.Q. et al. (2013) Determination of faecal and ileal protein and amino acid digestibility of common feeds for growing ducks. Proceedings 5th World Waterfowl Conference of the Asian Pacific Federation of the World’s Poultry Science Association. November 6-8, 2013, Hanoi, Vietnam.
  29. Helmbrecht A. and Hou S.S. (2013) Lysine requirement of Pekin ducks 35-49 days of age. Proceedings 5th World Waterfowl Conference of the Asian Pacific Federation of the World Poultry science Association, November 6-8 Hanoi Vietnam.
  30. Farrell D.J. (1999) 14 Nutrition and carcass quality in ducks. In eds Wiseman J. and P.C. Garnsworthy. Recent Developments in Poultry Nutrition 2 p203-206. Nottingham University Press, Nottingham, UK.
  31. Farrell D.J. (1995) Table egg laying ducks: nutritional requirements and current husbandry systems in Asia. Poultry Avian and Biology Reviews. 6 (1) 55-69.
  32. Shen T.F. (1988) Duck Nutrient Requirements Handbook. National Taiwan University, Taipei, p48.
  33. Hall A.D. (2008) Improvement in duck performance. International Hatchery Practice. 22 (6) 7-9.
  34. Leclerq B. and de Carville H. (1986) Dietary energy, protein and phosphorus requirements of Muscovy ducks. In eds Farrell D.J. and Stapleton P. Duck Production Science and World Practice, p58-69. The University of New England, Armidale, NSW.
  35. Leclerq B and de Carville H (1986) Growth and body composition of Muscovy Ducks. In: eds Farrell D.J. and Stapleton P. Duck Production Science and World Practice, p102-109. The University of New England, Armidale. NSW.
  36. Grimaud F. (2008) Past, present and future duck breeding. International Poultry Practice 22,(7) 21-25.
  37. Rose S.P, Burnett A. and Elmogeid R.A. (1986) Factors affecting the diet selection of choice-fed broilers. British Poultry Science. 27 215-224.
  38. Farrell D.J. (2000) Meat Type Ducks - Self Selection of Diets. Rural Industries Research and Development Corporation. Publication No 00/156.
  39. Hew L.I., Ravindran V., Mullah Y. et al. (1995) Enzyme supplementation improves ileal amino acid digestibility values of wheat for broilers Proceedings of the Australian Poultry Science Symposium 7, 189.
  40. Barrat B. (2001) Ducks perform with enzymes. World’s Poultry. 17(4) 16-17.
  41. Creswell D. (2013) Feeding ducks for maximum performance. Part 2. Asian Poultry. May 2013, p6-9.
  42. Farrell D.J. and Martin E.A. (1993) Feed enzymes in poultry nutrition: recent findings. Ed D.J. Farrell. Proceedings Recent Advances in Animal Nutrition in Australia, p266-276, Armidale, NSW.
  43. Farrell D.J. (1994) Utilization of rice bran in domestic fowl and ducklings. World’s Poultry Science Journal. 50, (2) 115-131.
  44. Farrell D.J. and Martin E.A. (1998) Strategies to improve the nutritive value of rice bran in poultry diets. 1. The addition of food enzymes to target the non-starch polysaccharide fractions in diets of chickens and ducks gave no response. British Poultry Science. 39, 549-554.
  45. Farrell D.J. and Martin E.A. (1998) Strategies to improve the nutritive value of rice bran in poultry diets111. The addition of inorganic phosphorus and phytase to duck diets. British Poultry Science. 39, 601-611.
  46. Martin E.A., Nolan J.V., Nitsan Z. and Farrell D.J. (1998) Strategies to improve the nutritive value of rice bran in poultry diets. IV. Effects of addition of fish meal and microbial phytase to duckling diets on bird performance and amino acid digestibility. British Poultry Science. 39 612-621.
  47. Rodehutscord M., Hempel R. and Wendt P. (2006) Phytase effects on the efficiency of utilisation and blood concentrations in Pekin ducks. British Poultry Science. 47, 311-321.
  48. Jackson J.C.B. (2008) Evolution and extinction in the brave new ocean. Proceedings National Academy of Science 105 (supplement 1) 11, 458-465.
  49. Hueng J.F., Pingel H., Guy G. et al. (2012) A century of progress in waterfowl production, and a history of the WPSA working group. Journal of the World’s Poultry Science Association. 68, 551-563.

May 2015 

Related Articles

Most Read Articles

Our Sponsors


Seasonal Picks

Poultry Breeds and Management