You are on page 1of 179



Poultry production is one area of livestock production that contributes significantly

more to human animal protein supply than any other domestic or non-domestic animal.

Consumers have high preference for poultry products (because of the taste and cheapness).

Beside preference, poultry products provide protein of high biological value (Epstein, 1990).

In a developing country like Nigeria, poultry production in both urban and rural areas is of

great importance. It serves as a simple means of generating family income and employment

opportunities (Akinokun, 1971). Compared to other domestic animals, poultry can be raised

with relatively low capital investment and readily available household labour.

Characterization, conservation and use of indigenous animal genetic resources are more

appropriate in the tropics where production is usually characterized by low input. The locally

adapted animals are also more readily available to resource-poor farmers and they can be

productive without high disease control input, yet, lack of information about the genetic

resources present in the indigenous farm animal in developing countries has led to their

under-utilisation, replacement and dilution through crossbreeding

( Therefore, characterisation, utilisation

and conservation of these indigenous animal genetic resources are of paramount importance.

Nigeria is endowed with many poultry species which are local to the country. These have

lived, adapted and produced for several years in Nigerian environment (Momoh, 2005).

FDLPCS (1992) estimated Nigeria's poultry population to be about 33 million. With the ever

growing population and concomitant improvement in the living standard of Nigerians, the

demand for egg and other poultry products will continue to grow, thus outstretching the

supply from chicken. Therefore, there is need to explore other poultry species to supplement

the provision of eggs meat and other poultry products from the chicken. This necessitates the

quest for the development of the ducks which are also widely distributed and are prominent

in the list of available poultry species in Nigeria (FDLPCS,1992).

Ducks are an essential part of many human societies around the world. They are

reported to be domesticated as early as 2500 BC (Farrell,1995). Today,the domestic duck

has come to serve a variety of roles in modern society as a source of food supply (meat and

eggs ), in other societies as a species that is selectively bred for show, feather as

ornament and pillow stuffing, and as a pet. Ducks have several advantages over other

poultry species;These include their disease tolerance and hardiness. They are excellent

foragers and easy to herd, particularly in wet lands since they tend to flock together


Population of ducks

The population of ducks in Nigeria was put at approximately eleven

million(11,000,000) and reported to be distributed all over the agro ecological zones of the

country (FLDPCS, 1992). Domestic ducks were ranked third among various poultry species

in Sokoto state of Nigeria (Hassan and Mohammed, 2003). In spite of this positive score,

very little attention has been paid to the genetic improvement as well as improved husbandry

practices to raise the performance of the domestic duck. There is little or no information

available on phenotypic and genetic parameters of the Nigeria indigenous Muscovy duck.

Ecological Zones

In Nigeria different agro-ecological zones exist and for all species of livestock

variations are reported to exist in both phenotypic and genetic characteristics as a result of

adaptation to these different ecological zones. Numerous comparative studies have provided

insight into the ecological mechanism underlying evolutionary diversification across habitat

gradients. This is evident in the reports of Akinokun (1971), Oluyemi et al. (1982),

Adedokun and Sonaiya (2001) for poultry, Adebambo et al. (1998) for cattle, Epstein (1990)

and Wilson (1991) for sheep and goats. This diversity constitutes a valuable resource for use

in animal breeding programme for improvement of the health and productivity of the animal

species, particularly domestic poultry and also in designing proper conservation strategy. It

also results in classification of the individual species into strains, breeds and ecotypes.

Several varieties of Nigerian local poultry, particularly chickens have been described on the

basis of location ( FLDPCS, 1992). Variations between populations from different locations

arose from adaptation to different micro-environmental conditions and management

practices (Nwosu et al., 1985). Ecotypes are developed through selection and adaptation and

vary from one agro ecological zone to another due to varied climatic characteristics such as

humidity, temperature and rainfall, although in some cases some ecotypes may be similar in

all traits (Adedokun and Sonaiya, 2001).

Genetic and phenotypic variations, especially in quantitative traits occur in a

population of different geographical locations and this may be as a result of genetic drift. The

differences may also have arisen from limited geographical differentiation of the original

population (Girand and Palabort, 1976; Ehiobu and Goddard, 1989; Ehiobu et al., 1990)

which may reflect adaptation to different environmental conditions.


In genetic studies various types of genetic markers such as morphological,

chromosomal, biochemical and molecular are used. Morphological (e.g., pigmentation and

other features) and chromosomal (e.g., structural or numerical variations) makers usually

show low level of polymorphism or differences. Biochemical markers have limitations as

they are often sex-limited, age-dependent and are significantly influenced by the

environment. Phenotypic features are also influenced by dominance effect and provide a

crude estimate of the average of the functional variants of genes carried by a given individual

population. These makers are reported to reflect variability in their coding sequence that

constitutes less than 10% of the total genome (Williams et al., 1990). Molecular makers have

been shown to be an efficient tool in the quantification of genetic diversity of various

populations (Saitbekova et al., 1999; Barker et al., 2001). Quite a number of techniques are

used in molecular maker analysis. The most common and cost effective is the Random

amplified polymorphic DNA technique. This technique has been used extensively in

molecular characterisation of livestock population in detecting polymorphism between

populations and establishing genetic relationship among them (Ali et al., 2003).

Some research findings on ducks by Ksiazkiewic (1995) and Farrell (1995) showed

that they are very susceptible to effective inbreeding and genetic drift if kept in a small

population for few generations. This genetic flexibility will suggest that variation can easily

be found in different populations at different locations .These variations could either be

morphological or genetic or both. Nigerian indigenous Muscovy ducks suffer neglect in the

area of research and improvement leading to remarkable absence of publication relating to

productivity, genetic diversity and phenotypic characterisation .This may not be unconnected

with the fact that ducks are considered as “dirty birds”. The genetic improvement of the

Muscovy ducks in Nigeria will first require investigation into their productivity, genomic

structure and geographical diversity.

1.1 Research Objectives

The overall objective of this study was to evaluate the phenotypic and genetic characteristics

of the Nigerian Muscovy ducks from rainforest and guinea savannah zones with a view to

highlighting their potential to contribute to egg and meat production in Nigeria. This

information will provide a framework for the development of breeding programmes for

conservation and sustainable use of the Muscovy duck genetic resources in Nigeria.

The specific objectives are to:

1. Evaluate the morphological/phenotypic characteristics of local Muscovy ducks from

tropical Rainforest and Guinea Savannah zones;

2. Estimate morphological distance between the two populations;

3. Evaluate performance of the ducks from these zones in basic production traits; of

growth (0-to 20 weeks of age) and egg production;

4. Estimate genetic parameters of growth and egg production traits of the ducks from

the two ecotypes; and

5 Estimate genetic relatedness among the two ecotypes using random

molecular techniques.


2.1 Domestic Ducks: Origin and Domestication

Ducks belong to the order Anseriformes diverged from the chicken Galliformes in about 11

million years ago (Tuinen and Hadley, 2004) According to paleontological data, the main

radiation of the modern duck took place during the Miocene, 5 -23 million years ago

(Olson, 1985). From ancient times domestic ducks have served as a source of food and

income for people in many parts of the world. Ducks are source of meat, eggs and down

feathers. Their meat and eggs are good dietary sources of high quality protein and energy in

addition to several vitamins and minerals. Tuinen and Hadley (2004) classified domestic

ducks into Common,Muscovy and Sterile.

Common Ducks: Most ducks are classified as common ducks and are believed to have

originated from the Mallard (Anas platy rhynchos). Some of the better known breeds of

common domestic ducks include the Pekin, Asylesbury, Rouen, Call, and Indian Runner,

Khaki Campbell, Cayugi, Albio, Maya and Tsaiya. They further added that these ducks can

interbreed and produce fertile offspring.

Muscovy duck

The Muscovy (Cairina moschata) is distinctly different genetically from common

ducks. The breed is reported to have originated differently. Several reports on its origin

have been presented by (Clayton, 1984; Hahn et al ., 1995; Farrell, 1995; Hugue and

Sultana, 2002).

Sterile Hybrid Ducks

These are products of crosses between Muscovy and common ducks. When allowed

to mate naturally their fertility is usually very low. The offspring are often sterile and can not

be used for breeding. These hybrids are usually raised for their meat and in some cases for

their liver (foie gas) and usually referred to as mule or hunny (Olson ,1985)

2.1.1 Domestication of duck

Pottery duck models excavated in the Yan-Shi-Menkou mountain in Fu Jian in

Southern China provided evidence that domestication of duck occurred during the New stone

age between 4000 and 10000 years ago (Wucheng, 1998). Clay models unearthed in

Shanxian country, central Henan province suggested that ducks were domesticated during

the Yangshao culture about 4000 years ago (Jianhua, 1984). Ducks were domesticated in

China at least 3000 years ago. The archaeological evidence along with a favourable

environment and agriculture suggested that ducks were probably domesticated in southern

China at least 1500 years before they were separately domesticated in western Europe.

Watson (1969) indicated the possibility of even an earlier domestication in describing the

finding of pottery models of ducks and geese at Lung Shan site in Hupei about 2500BC in

company with models of sheep , dogs, turtles, and fish. Zeuner (1963) has speculated that

duck domestication may also have occurred in Mesopotamia from Sumerian times beginning

2800 BC , into the Assyrian period (900-600 BC).

There is also indication that ducks were kept as domestic birds in ancient Egypt

(Zeuner, 1963). According to Harper (1972), translation of Egypt papyri confirms that geese,

chickens, and pigeons were being raised from the fifth century BC to the seventh century AD

, but ducks were seldom mentioned and any reference probably was to wild fowl. According

to the early writers, the Greeks and Romans kept ducks, but indications are that the birds

were wild and had to be confined within netting enclosure (Harper, 1972).

2.2 Muscovy Duck

The Muscovy duck is one of the greater wood ducks included in the subfamily

Anatinae tribe cairini (Clayton, 1984). The duck is reported to be the predominant water fowl

in Africa, Asia and Latin America. It thrives well under free range condition in tropics and

Neotropical environment (Haln et al., 1995; Farrell, 1995). Evidences indicate that under

natural conditions the Muscovy duck is a tropical bird and lives in marshy forest. Its

robustness and hardiness enable it to adapt to different climates and habitats. They are

sexually dimorphic, the male being almost twice the weight of the female (Leclercq and De

Cerville, 1985;. Larbier and Leclercq, 1994). They noted that aggression and sexual display

among the birds are simple and not well differentiated.

Parkhurst and Mountney (1988) described Muscovy duck as a species that produces

less fatty carcass. Larbier and Leclercq (1994) added that Muscovy ducks are characterized

by having large pectoral muscles and are sexual dimorphic in favour of the male. Phuoc et

al. (1994) supported this when they reported that French male Muscovy duck can grow to be

quite large weighing 5-7 kg, while the female can attain 2.5 – 3.5 kg.

Townsend (2002) reported that Muscovy ducks are unique because of their bright rest

crest around their eyes and above the beak. He observed that they do not swim much because

their oil gland is under-developed compared to most ducks. It is on this note that Halns et al.

(1995) reasoned that the duck is not generally a duck or goose but suggested closeness to

goose than ducks on the basis that it eats grasses as geese do and has similar long incubation

period of 36 days compared to that of domestic ducks. Despite these observations Muscovy is

still considered duck.

Derrell and Guy (1989) were of the view that despite the variability noted in

Muscovy and its uniqueness, distribution and extensive range it is however the least studied

of all new world water fowls. Most informations on Muscovy duck are historical and sparse

Delacour (1959), Wetmore (1965), and even recent literature discuss the species only on

general terms (Bolen, 1983). Jacquert and Sanveur (1995) made similar observations that, not

withstanding the economic importance of the domesticated Muscovy duck in a number of

countries a few reported its performance.

2.2.1 Origin of Muscovy duck

There are diverse reports on the origin of Muscovy duck; most writers are of the

opinion that South America is the origin of the domesticated Muscovy duck (Johnsguard,

1978; Hahn 1995, ) e.t.c . Mohammed (1996) reported that Muscovy originated from South

America and that it was domesticated by the Columbian and Peruvian Indians and then

introduced to the other world by the Spaniards and the Portuguese in the 16 th century.

Townsend (2002) reported that the Muscovy originated from Brazil and that they are the only

domestic ducks that do not derive from the Mallard stock. He further reported that the

Muscovy duck had probably been domesticated for centuries by South American indigenous

culture at the time of its introduction to European Colonialist. He added that both patrilineal

and matrilineal ancestors of domestic Muscovy are the Wild Muscovy ducks. Encyclopaedia

online explain the origin and domestication of Muscovy as domesticated in Peru and

Columbia by the pre-Columbian Indians and that the Mallard was domesticated in China

about 2000 years ago.

Donkin (1989) described the domestication of Muscovy duck in Africa and that it

was introduced into Africa by the Portuguese in sixteen century and had a wild relative, C.

hartlanbi, which is indigenous to the forest zones of West Africa. Townsend (2002) described

early sources relating to the spread of Muscovy duck in West Africa although he was

reported to draw a wrong conclusion that it was pre Columbian introduction.


2.2.2 Characteristics and potentials of Muscovy duck Morphological characteristics

Reports on morphological characteristics of Muscovy ducks particularly in Africa are scanty

in literature. Few studies Hahn et al. (1995) and Schaef (1995) described plumage colour

pattern in Muscovy duck. Plumage colour among Muscovy ducks have been reported to be a

combination of black and white ranging from mostly black to white (Hahn et al., 1995) .

Hassan and Aliyu (1996) studied the plumage colour pattern of Muscovy duck in North

Western Nigeria and reported that four strains were observed black, white, mixture and

creamy, and that black/white mixture constituted 80% of the population. Hassan and

Mohammed (2003) in their report on same ecotype gave a contrary result in which the black

strain was more in number, representing 41% of the entire population with the black/white

strain significantly heavier than other strains. Various colour patterns were also reported by

other workers, (Schaefer, 2002 and Hahn et al., 1995), who reported additional plumage

colour pattern as Blue, Blue/white and chocolate types and similarly reported that the

muscovy have more distinctive features such as red face with prominent caruncle at the base

of the bill and a low erectile crest of feathers and that their feet are equipped with strong

sharp claws for grabbing trees.

Hassan and Mohammed (2003) further described the Nigeria Muscovy as having bill

colours ranging from black to white, bill length of about 7.03 and 6.02cm for males and

females respectively, and added that significant positive correlation exists between body

weight and bill length. Tegnia et al. (2007) summarized the morphological characteristics of

African Muscovy and reported that mean metatarsals diameter, thoracic perimeter, body

length, length of bill, foot length and wing length in male were 1.3 , 29.5, 57.5, 7.0 27.1 and

27.8 respectively while the average for female were 1.2, 25.8, 51.0, 6.3, 23.3 and 26.9

respectively. Age at sexual maturity

Age at sexual maturity is the time between the date of hatch and the date of first egg.

Age and weight at sexual maturity are influenced by genetic make up of the individual and it

is influenced by feed intake, light, duration of day length and other environmental factors

(Morris and Fox, 1960; Wessels, 1962).

Ola (2000) reported that Muscovy duck attained sexual maturity at 28 weeks of age

(about 6.33 months), further adding that though Muscovy can attain maturity earlier. They

have limited breeding capacity especially during the dry season and may delay egg laying to

beginning of rainy or wet season. Duru et al. (2006) in their findings reported a higher value

of 7.70+.044 months as age at first lay of egg by Muscovy duck in a peri urban management

system under semi-humid condition in Nigeria.

Variability in age at sexual maturity of other duck breeds had also been reported.

Sexual maturity in Khaki Campbell was reported at different ages by different authors, such

as 18 weeks by Klandorf and Harvey (1984), 20 weeks by Hetzel and Gunawan (1984) and

23 weeks by Eswaran et a. (1984). Kalita et al. (2004) in their findings reported age at sexual

maturity in Khaki Campbell, Desi and their crosses as 169.10+9.95, 229+10.06 and 219.89

+10.54 days, respectively.

Age at sexual maturity has been suggested to be an important life history trait (Oli

and Dubson, 1999), with substantial potential for influencing fitness. They noted that within

species a genotype that attains sexual maturity early often has a greater probability of

realizing fitness because its offspring start reproducing earlier than others and thus is

expected to have a higher fitness than in late maturing genotype. Stearns (1992) had a

contrary view on early maturity and feels that late maturity allows for additional growth and

experience that will substantially increase future reproductive output. Tanka and Rosenberg

(1952) earlier reported that selection for early sexual maturity increases the number of eggs

per day of life and contributes to an overall feed efficiency. Morris (1966) reported that age

when the hen is able to reproduce (age at sexual maturity) is not the same as age when the

hen is able to produce (age at first egg). It can be expected therefore that age at first egg can

be earlier or later than age at sexual maturity.

Kalita and Deka (2005) reported that age at sexual maturity for Muscovy duck in a

rural area of Assam in India ranged from 300-365 day. A similar result was also reported

earlier on by Avanzi and Romboli (1979). Mahanta et al. (2001) reported for age at sexual

maturity. However, sexual maturity of both sexes of Muscovy duck was reported to be 20 –

24 weeks (Panda and Kumar, 2004) and 29 -30 weeks (Sauveur and De Carville, 1985). Egg production

Egg production is influenced by several factors such as strain, feed, mortality, health

and management practices, age at point of lay, peak lay and persistency of lay (North, 1984;

Petek, 1999). Sauveur and De Carville (1985) reported that Muscovy duck species have less

expressed seasonal character in egg laying compared to other species of Anas genus. Some

authors noted that the magnitude of egg production in ducks is dependent on the extensive

and intensive management systems, including battery cages as well as on nutrition

programme (Avens et al., 1980; Hetzel and Gunawan,1984; Hetzel, 1984, Scot and

Dean,1991; Farrell, 1995)

Several studies on various egg production characteristics under intensive production

have been reported (Tikk and Vint, 1988; Romboli et al., 1984; Osman, 1997) . These

authors and other researchers conducting studies on Muscovy duck species established that

under intensive breeding conditions, the duck reached laying capacity of 120-150 eggs for

two reproductive seasons. Retailleau (1997) gave higher value of egg production or laying

capacity for Muscovy duck in an intensive breeding, and obtained an average of 210 in two

reproductive cycles. Some Russian authors- Chipchiryuk (1980), Sauveur and De Carville

(1990), Sogomonov and Rahmanov (1988) reported that in a 5- month laying season

Muscovy duck laid about 62-100 eggs. Earlier Romboli et al. (1984) reported that 45 to 100

eggs were laid by Muscovy duck in the first year of reproduction depending on the breeding


Significant differences exist in egg production or laying characteristics of Muscovy

duck in other regions of the world compared to that obtained in African Muscovy duck.

Mopate et al. (1999) reported that local Muscovy duck in N’djamena, Niger Republic

produced 14 to 15 eggs per clutch with an average of 2.6+1.2 clutches per year. Similar

observation was reported earlier by Mourthe (1989) in Chad. Similarly Banga et al. (2007)

reported that Muscovy duck in rural area of Congo Brazzaville lay an average of 14.6 +3.0

eggs per clutch with a maximum of two clutches per year. Mohammed (1996) also reported

similar trend in Mozambique for Muscovy duck. On the contrary Panda and Kumar (2004)

reported average egg production of Muscovy ducks in Assam district in India to be 45-60

eggs per annum. A similar finding was reported earlier by Mahanta et al. (2001). Kalita and

Deka (2005) however reported lower egg numbers from the same zone in Asia. Though this

variability exists, egg production in Nigerian Muscovy ducks tends to follow same pattern

with those reports obtained from sister African countries. For instance Hassan and Aliyu

(1996) and Duru et al. (2006) reported local Muscovy in Northern Nigeria, Ola (2000) in

western Nigeria, Etuk and Abasiekong (2005) from Eastern Nigeria. Clayton (1984) claims

that much of the high egg production of domestic duck is an expression of natural fecundity

given appropriate opportunity and owes little to artificial selection Egg weight

Romboli et al. (1987) reported egg weight in three strains of Muscovy ducks, black,

white and sepia as 81.2, 84.3 and 84.8g, respectively. Another report by Harun et al. (2001)

recorded 84.5g as egg weight of Muscovy duck raised in an intensive management system.

Nickolova and Gerzilow (2000), however reported lower value of 70.6g as egg weight in

Muscovy duck.

Egg weight in the Muscovy duck is reported to be higher than other domestic

ducks. For instance, Sanveur and De Carville (1985) reported egg weight that ranged from

77 to 83g in the Muscovy duck while Karaca et al. (1996) reported egg weight of 69.9g in

Pekin duck. Testik and Sarica (1991) also reported 71.5g as egg weight for the same breed

(Pekin). Isguzar and Testik (1999) noted that there is high variability in egg weight of Pekin

and most water fowl species and added that ducks reared under extensive condition produce

lighter eggs averaging 54.8 to 64.2g when compared to those raised intensively whose egg

weights normally exceed 70g.

Several authors related egg weight to other production parameters, Khurshid et al.

(2004) reported that increase in hatchability increases with egg weight, where as increase in

egg shell weight and thickness result in a decrease in hatchability in birds. Similar

observation was made by Shanaway (1994) for Japanese quail. Fertility of Muscovy duck egg

Fertility refers to the fertile status of group of eggs laid over a period of time by

single bird, by small group of birds or by commercial flock, and is usually expressed as the

percentage of total egg laid (Bodi et al., 1996). It has been widely reported that egg fertility is

affected by age of the breeders, exposure to light, nutrition, management, mating ratio and

semen quality (North,1981: Stahl et al., 1986: Peeble and Brake 1987). Narahari et al. (1988)

added that other factors affecting egg fertility include breeding system, period of lay and

parent live weight. Compared to other domesticated poultry and to the ducks of Anas genus

Muscovy duck species are reported to be distinguished by very high degree of egg fertility

(Nickolova, 2004). Over 92-93% on the average had been reported for the Muscovy duck

(Bagliacca et al., 1989; Bodi et al., 1996; Bogenforst et al., 1996, Bondarenkoa, 1997)

Tikk and Vint (1998) reported very good egg fertility of 96-97% between 3-18

weeks of laying in a year old Muscovy duck in Poland. Meltzer (1988) and Chipchiryuk

(1980) reported that at the beginning of a reproductive season,(April to May), a high egg

fertility of the species 94-98.25% was recorded. Earlier Baglicea et al. (1989) reported that

peak egg fertility for one-year old Muscovy duck was in the second month of the laying

period and was usually 85-95% depending on the variety.

Nickolova (2004) reported that spermatozoa concentration of Muscovy duck was

usually highest in March-April, an indication that fertility of egg was highest at that period.

Though he noted that despite that, egg fertility of Muscovy duck was usually high (above

92%) throughout the reproductive period of March to August. This corroborated Szyniunk

et al. (1991) who reported that average egg fertility for between 3 and 8 weeks of the laying

season in one-year old Muscovy duck was 96.77%. Higher value of fertility of Muscovy duck

eggs was recorded for tropical types, particularly in Nigeria, 90.96%, by Ola (2000) an

indication of high fertility of egg in the breed.

Few reports recorded lower values of percent fertility in Muscovy duck eggs.

Isguzar (2005) reported that percent fertility among three varieties of Muscovy ducks black,

black-white and white strains were 63.4, 33.5 and 34.5%, respectively. Environmental factors

must have played some vital role in this regard. Hatchability

Hatchability is one of the prerequisites for the better propagation of breed (Farooq

et al., 2001). It refers to the proportion of fertile eggs that continue development and produce

viable chicks. Hatchability has been reported to be affected by physiology, age of the breeder,

egg size, nutrition, genotype and possession of major genes (Huntan ,1981; Buss, 1982 ; Peter

et al., 2004). To a large extent it is a function of fertility except when there is presence of

major genes that reduce embryonic livability (Peters et al., 2004). The authors added that

lower hatchability can be observed in a flock if embryonic mortality is high and fertility is

low. Variation in embryonic mortality may be due to poor egg holding period, unbalanced

nutrition, stressful condition, the parent flock was exposed to or any other fault in incubation

and hatching equipment(Peter et al., 2004).

The size of the eggs is also one factor that influences hatchability in most poultry.

Farooq et al. (2001) and Murad et al.(2001) reported significant influence of egg size and

shell weight on hatching performance of Japanese quails. The authors reported lower

hatchability of small-sized eggs and those with poor shell quality. Shanaway (1994) also

reported better hatchability in large-sized eggs. Similar report of better performance based on

egg size was reported earlier by Schader et al.(1985).

In Muscovy duck the process of hatching is reported to span between 28-35 days for

the eggs to change a microscopic germ into a downy chick capable of walking, eating and

expressing its need by its voice and action (Banerjee, 1991). Several authors have reported

the exceptionality of Muscovy duck eggs in hatching. Serbul (1986), Murae (1988), Bonima

and Velez (1992) reported that hatchability of Muscovy duck eggs under artificial incubation

was always found to be lower than under natural incubation, suggesting that heat supply by

conduction normally may result in a better embryo temperature and development than by

convection. This is evident in the report of Isguzar (2005), who recorded 41.1, 35.6 and 4.7%

as hatchability values of fertile eggs of black, black/ white and white strains of Muscovy

ducks .Similar values were earlier reported by Hahn et al. (1995 and Fattouh et al. (2003). In

contrast, higher values were obtained in tropics using natural incubation as reported by

Mopate et al. (1999) who found out that hatchability of Muscovy duck eggs reared in urban

households in Ndjamena Chad was between 80-85% during rainy season (June- November)

and 59-78% during dry season (December to May), with a mean brooding period of 63+25

days. These values were similar to those reported by Romboli (1990) and Gueye et al. (1998),

but lower than those reported by Hassan and Aliyu (1996) for Muscovy duck in Northern

Nigeria. Ola (2000) also recorded high value of 72.57+11.7% for Muscovy duck in South-

western Nigeria. Sanveur and De Carville (1985) reported a hatchability of 80% among rural

Muscovy ducks in Assam, India. These indicate the viability of muscovy in tropical


The importance of water for swimming as an essential component of mating

behaviour and welfare of many domesticated water fowls, as well as a source of moisture for

their eggs in incubation has been reported by several workers (Rauch et al., 1993; Harun et

al., 1998) . However, Narahari et al. (1991) and Harun et al. (2001) reported that absence of

swimming water did not affect hatchability of Muscovy duck eggs. This may explain why

Muscovy ducks are easily kept either in traditional extensive system or in modern harsh dry

18 Hatch weight of ducklings

Weight at day old is a potential attribute of an individual and if this weight is

favourable there is the tendency that better weight will be achieved in the later part of the life

of the individual. Baeza et al. (2001) reported day-old body weight of Muscovy duck as 45g.

They further added that the mean weight of male and female ducklings were similar at hatch,

but became significantly higher(P<0.01) in males as from 4 weeks of age. At 10 weeks of

age, males and females reached 4.57kg and 2.88kg, respectively . Isguzar (2005) reported that

the average weights of day old black, black/white and white ducklings were 41.8 +0.6g ,

46.2+3.3g and 42.0+0.9g, respectively, indicating significant variation between strains.

Similarly, it was further reported that significant variations occurred in weight of day-old in

first and second hatches such as 41.0g and 42.7g in a white variety of Muscovy duck.

Shahin et al. (2000) studied day old weight among water fowl and concluded that

variations occurred across genotypes, reporting 45 to 59g for Peking ducks as earlier reported

by (Knizetora et al., 1992). Karaman and Testik (1995) also reported a range of 47 to 55.8g

for Pekin ducks. Initial weight in water fowl is not a reflection of an adult weight as noticed

in Pekin and Muscovy ducks (Shahin et al., 2000). Growth and productivity characteristics:

Growth is a fundamental property of biological system and it is defined as an

increase in body size per unit time (Schultz et al., 2001; Lawrence and Fowler 2002). Three

processes contribute to growth, which include cell division, assimilation and cell expansion.

Robert (1978) earlier on observed that despite best parameters chosen to estimate growth, it

has been observed that growth showed many irregularities as a result of fluctuation in the

environment or diet in addition to the fact that different parts of the organism would grow at

different rates and stop at different times. Adedeji et al. (2004) reported that Nigerian

indigenous birds are characterized by slow growth and low productivity in terms of meat and

eggs, but possess genes for productive adaptability (Peters, 2000).

Growth performance of African Muscovy ducks has been extensively studied and

reported by many authors (Romboli,1990; Hassan and Aliyu,1990; Mutui and Mbaga ,2001,

Ngapongora, 2004; Etuk et al., 2006). Meulen et al. (1999) stated that Muscovy duck does

not grow very quickly and its final weight depends on the way it is kept and fed. Etuk et al.

(2006) gave 18.34+0.63g, 21.23+2.17g, and 23.75+0.43g as body weight gain for male

Muscovy duck at 5, 6 and 9 weeks, respectively while 19.25+1.53, 17.10+2.65 and

15.09±1.2g were reported as body weight gain for female Muscovy duck at 9, 6 and 5 weeks.

There were marked sexual dimorphism between males and females. There was significant

difference (P<0.05) in average daily weight gains for male (16.11g) and female (10.12g) with

about 37% increase in body weight gain of males over females.

Ngapongora et al. (2001) reported for local ducks in Tanzania that at 9 weeks drakes

reached 2868g while ducks attained 1821g. From Taiwan, Hu et al. (2006) reported 3896g

and 2540g as body weights for male and female Muscovy duck at 10 weeks of age.A similar

weight was reported by Leclerg and DeCarville (1986) for duck in France and in Australia

(Mignon-Grasteau et al.,1998).

Growth in Nigerian local Muscovy duck has not been extensively studied. Reports

by Ola (2000) put the body weight at 5 and 12 weeks as 503 and 1506g for males and 393

and 1015g for females, respectively . He reported body weight gain at 5 and 12 weeks also as

13.35 +2.23 and 15.54+1.38g for males while for females were 10.21+1.25 and 10.02 +0.63g

, respectively
20 Feeding efficiency

The trait with the most impact on profitability is feed efficiency. The conversion of

feed into egg is primarily a function of egg number. It is also influenced by egg size and body

weight. Breeders improved feed conversion throughout the 20th century in poultry (Bochno et

al., 1994). Feed intake and feed conversion efficiency are affected by the rate of growth of

birds, metabolisable energy content of the ration, nutrient adequacy of the ration,

environmental temperature and health condition of the bird (Bochno et al.,1994). Wu and

Han (1983) observed that age and sex of the bird and quality of diet significantly affected the

feed efficiency of bird. They further reported that feed intake was positively correlated to

the coarseness of feed in ducks. Bochno et al. (1994) reported that male ducks grew faster

with more efficiency in feed conversion than female. They added that feed conversion was

more efficient in Muscovy duck than Pekin. Perez (1985) reported earlier that Pekin ducks

had a feed conversion at 8 weeks of 2.83kg feed/kg live weight gain. Duong and Nguyen

(1995) and Luong et al. (1995) reported feed conversion ratio of 2.91 and 2.77kg feed/kg

live weight, respectively for commercial ducks reared intensively. Etuk et al. (2006) observed

that improved management system did not affect final body weight, growth rate, feed intake

and feed conversion ratio of Nigerian indigenous Muscovy duck. Sex ratio on egg fertility of Muscovy duck

Establishing and maintaining the optimal sex ratio in agricultural poultry raising affect

the economic result of poultry breeding in two ways, by achieving maximum egg fertility and

by the number of the drakes raised (Nickolova, 2004). The number of drakes or male ducks

that will produce optimum fertility in a flock forms what is referred to as mating ratio.

Mopate et al. (1997) observed that 3.4 duck per drake was best for effective egg fertility in

scavenging system in Chad. Several authors,( Sauveur and De Carville,1985; Sauveur ,1990;

Sauveur and De Carville, 1990) suggested a ratio of 1:5 as appropriate breeding ratio for

Muscovy duck. The suggestion was made in view of the differences in live weight between

male and female Muscovy ducks (kalita and Deka, 2005). However, Romboli et al. (1987),

Romboli and Battini (1986) and Savitskoy (1989) suggested a lower ratio of 1:4 for optimum

fertility of Muscovy duck eggs. Wang and Xu (1989) reported that a ratio of between 1:4 to

1:10 as recommended for water fowl species in obtaining egg fertility of between 75.9 to

94%. Ola (2000) recommended a ratio of 1:5 for effective fertility in the Nigerian indigenous

Muscovy duck. Incubation

Incubation of eggs in Muscovy duck is quite unique as in most waterfowls. Panda and

Padhi (2004) reported that both male and female Muscovy ducks take part in incubation and

brooding process. Panda and Kumar (2004) also reported that Muscovy ducks have extra

ordinary brooding capacity and are good mothers to raise their own ducklings. They reported

that each duck has the capacity to brood an average of 10 – 15 eggs.

Caldwell and Cornwell (1975) found out that onset of incubation in Muscovy ducks is not

easy to identify as it begin gradually during laying period, that is, it is common for ducks to

lay one or more eggs after the on-set of incubation. Under natural incubation, Nichelmann

(1993) and Nichelmann and Gilsing (1995) had observed that incubation in Muscovy ducks

started before oviposition of the last eggs.

One uniqueness of the Muscovy duck is the elongated period of egg incubation. Un-

like other poultry with an average of 28 days of incubation, muscovy duck has a range of 33-

37 days of incubation with an average of 35 days (Nickolova , 2004).

22 Clutch size and number

Bahr and Palmer (1989) reported that egg production is based on cyclic process with

a period of 24 to 27 hours or more depending on age. These cyclic processes result in

clutches of one or more eggs laid on consecutive days and each clutch is followed by one or

more pause days. The size of a clutch varies little within a hen and is assumed to be highly

heritable (Nalbardov and Opel, 1974). There are two competing theories that attempt to

explain both proximately and ultimately clutch size in water fowl. The egg production

hypothesis developed by Lack (1968) proposed that average clutch size of water fowl species

evolved in relation to average amount of food available to laying female. The second

hypothesis is the egg viability hypothesis advanced by Arnold et al. (1987) who argued that

because water fowl eggs lose viability if they fail to sit on the eggs during incubation and

predation risk is cumulative over that time, the upper limit of about 14 eggs to clutch size of

praive nesting duck is set by the combing effect of these egg mortality factor.

Derrell and Guy. (1989) in Mexico reported that average clutch size in a wild

muscovy population was 13.6 +3.2 with a range of 9-21 eggs. Lower average of 8-10 eggs

were however reported by some authors (Leopold, 1959; Wetmore ,1965; Woodyard and

Bolen 1984). Most studies in tropical Muscovy duck reported clutch size of between 13 to 15

eggs per clutch (Mopate et al., 1999: Gueye, 1999: Ola,2000) with an average of 2.6 clutches

per year. This trend is in line with hypothesis of Arnold et al .(1987).

Season of the year is reported to have significant effect on clutch size. Mopate et al.

(1999) reported egg 1ay per clutch of between 13.9 to 7.3 in December to February and

from 14.5 +2.5 to 14.7 + 3.7 in June to July.


2.3 Sexual Dimorphism in Muscovy Duck

Most animal species exhibit phenotypic differences between males and females (

Darwin, 1871). Sexual size dimorphism in particular has retained the attention of numerous

evolutionary biologists (Maynard-Smith, 1978; Charnov,1982 ; Simons ,2001).

Dimorphism in body weight is observed in almost all domesticated species. However,

the degree of divergence between sexes varies considerably between species (Jean et al.,

2003). The authors believed that sexual dimorphism evolved under the pressure of natural

and sexual selection. A good requirement for sexual size dimorphism to evolve is that the

traits of interest should be controlled by genes differently expressed in the two sexes, since

the genetic correlations between the sexes are usual high for morphometric traits. Theoretical

and empirical analyses have predicted a slow evolutionary rate of sexual size dimorphism

(Lande, 1980; Arnold, 1985; Roff , 1997; Merila et al., 1998).

Observational data (Frith, 1967; Marchant and Higgins, 1990; McCracken, 1999) suggested

that male emancipation for parental care and evolution of a lek mating system led to fixation

of larger body size and other secondary sexual characters in male ducks (Moller, 1990; Zuk et

al., 1990). An alternative hypothesis not related to mating system theory is that niche

divergence perhaps driven by inter sexual competition for food resources is responsible for

observed pattern of sexual dimorphism in Musk ducks (McCraken et al., 2000)

Steven and Sauveur (1985) reported that one of the characteristics of the Muscovy

duck is the obviously expressed sexual dimorphism in the live weight. This was earlier

reported by Holderread(1978), that as early as 10 days of age of the duckling sexual

dimorphism starts to express between male and female. Garzilov (1999) however reported

that at 3-4 weeks of age it begins. He added that at 10-12th weeks live weight of the female

duckling was only about 65% of the male. Similar observations were made by Chipchiryuk (

1980), Sauveur and De Carville(1985) and Sauveur and De Carville (1990). Sauveur (1990)

reiterated this position that selection for increased body weight reinforced sexual

dimorphism at these ages.

Bacon et al. (1993) and Beaza et al. (1998) suggested that Muscovy duck could be an

interesting model for studying endocrine regulation of sexual dimorphism on body weight.

The female displays earlier body and muscular development. Beaza et al. (1999) Meulen-

Vander and Dikken (1999) reported that Muscovy duck does not grow very quickly and its

final weight depends on the way it is kept and fed, but noted variability in growth and body

weight between male and female. Similar observation was made earlier on (Knit et al., 1985;

Payne, 1990). Hassan and Mohammed (2003), in a study of weight differences between male

and female Muscovy duck in North Eastern Nigeria reported a significant difference in

matured body weights between male and female which were 2.32+ 0.03kg and 1.54+0.03kg,


Teguia et al.(2007) observed sexual dimorphism in African Muscovy duck and

reported that it favours the male for all traits with the differences being significant (p<0.05)

only after weeks of age. They reported that at 12 weeks the male weighed 1832.0+ 80.4 9g

while the female reached only 68.2% of that weight of the male.

In general male ducks which have completed their growth usually reached 4.5 to 5kg

while the female usually attain 2.2 to 3kg (Salichou, 1991; Nickolova, 2004). Etuk et al. (

2006) reported significant differences (P < 0.05) in average daily weight gain of males (16.1

9) and female (10.17g) which showed about 32% increase in body weight gain for males

Hu et al. (2006) reported that it will be difficult to modify the sexual dimorphism of

body weight by selection. Sexual dimorphism is at least partly caused by some genes carried

by the sexl chromosomes (Merritt,1966; Bernon and Chambers, 1985)


Kharel and Arboleda (1986) indicated that the sexual dimorphism in body weight of

Muscovy duck was attributed in part to the effect of sex-linked as well as autosomal genes.

Recently, Tai and Rouvier (1998) in a factorial cross breeding experiment between muscovy

and Pekin ducks suggested that beside the effect of sex-linked genes, assumed coding genes

of the non-pseudo-autosomal region of the muscovy W chromosomes compared to those of

Pekin W chromosomes appear to depress growth. Body weight of sexual dimorphic Muscovy

duck increases with age (Hu et al., 2006; Leclercq,1990; Leclercq and De Carville,1986).

2.4. Genetic Parameters in Muscovy Duck :Basis for Variation in Morphological and

Reproductive Traits

In recent years, there have been increased emphases on estimating genetic

parameters not only for production and morphological traits but also for fertility and

survival traits (Kadarmideen, 2004). Estimation of genetic parameters involves partitioning

of observational components, i.e., phenotypic covariance between relatives into casual

components such as variance due to additive genetic effects, dominance, epitasis and

permanent and temporary environmental effects (Falconer, 1981). Parameters of significance

in animal breeding include: heritability, genetic correlation and repeatability.

2.4.1. Heritability

Heritability is defined as the proportion of within population phenotypic variance

(VP) that is additive. It is a parameter that indicates the ability of an individual to pass its

genetic attributes to the offspring. It is not a biological constant, being affected by

management procedure and the method of estimation. It is a property not only of the trait

but also of the population and environmental conditions surrounding the animals (Falconer,

1981) and therefore varies from population to population. Knowledge of the amount of

additive genetic variation and the heritability of traits is essential for understanding how

individual characteristic changes from one generation to another in response to selection

(Falconer and Mackey, 1996). They further observe that heritability is the single most

important conation in determining appropriate animal evaluation methods, selection method

and mating system. They add that it measures that part of the total variability of the trait

caused by genetic differences among the animals on which the measurement were taken.

The numerical value of heritability estimate is given as a percentage or a decimal

and should of course lie between 0 and 1. In some instances the value falls outside this range.

This is a chance occurrence caused by statistical manipulation.

Heritability estimate is a partial description of one trait in one group of animals at some

particular time. It may vary (for each trait) during a time period from herd to herd or it may

vary in the same herd from time to time. This is natural because herds differ in genetic make-

up and because there are many different environmental circumstances from herd to herd or

within a herd from year to year. These genetic and environmental differences influence the

size of the numerical value of the terms (i.e., genetic variance ,VG and total variance VP)

used in the estimation of heritability.

Heritability is estimated from two basic forms of relationship; parent-offspring and

sibling (full sib and half sibs). Parent-offspring relationship involves the regressing of the

mean of offspring performance on mean of parent performance while the other involves

covariance of half sibs or full sibs. Estimation of heritability

In statistical terminology, it is assumed that the second moment statistics (variances)

are less reliable than the first moment statistic (average). Many statistics used in animal

genetics are computed from the variance and covariance- heritability, genetic correlation and

repeatability. These statistics should be estimated with large samples and with appropriate

analytical tools (Statistical algorithms) (Lima Guerra, 2004).

There are currently several methods for estimating variance and covariance

components. Some of the common methods of variance component estimation include the

traditional analysis of variance (ANOVA) methods such as Henderson’s methods 1, 2 , 3 and

4, Minimum-norm-quadratic unbiased estimation (MINQUÉ), Maximum likelihood

estimation (ML), Restricted Maximum Likelihood Estimate (REML) (Harville, 1977;

Henderson, 1984; Meyer, 1989; 1991).

The estimation of variance components in ANOVA is performed with ordinary least squares.

The estimated least squares means from data are set to be equal to derived expression of

their expected values.

ANOVA requires sample data to be balanced, e.g; similar sample size among the

fixed effects. ML and REML estimators do not require the data to be in specific design or

have balanced fixed effects (Harville, 1977). Restricted Maximum Likelihood (REML) and

(DFRML) developed by Thompson( 1962) and expanded by Patterson and Thompson (1971)

has been reported to be marginally sufficient, consistent, efficient and symmetrically normal

(Harville, 1977). Restricted Maximum likelihood do not give negative component of variance

which is usually associated with Maximum likelihood and ANOVA . He further suggested

that when maternal and permanent environmental effect are to be estimated and when

parents are related and selection is going on in a population, animal model procedure of

REML or DFREML is most appropriate.

28 Heritability estimate for morphological traits

The heritability of traits has a major influence on the ability of individuals in a

population to adapt to local environment (Monsseau et al., 2000). Quite a number of studies

on heritability estimates of other parameters have been carried out in Muscovy duck

particularly in France (Mignon-Grasteau et al., 1998; Thompson, 1982), Taiwan (Tai and

Rouvier, 1998; Tai et al., 1989, Merritt, 1966; Hu, 1999; Hu et al., 2004). Studies on

heritability estimates for morphological traits in Muscovy ducks are quite limited in

literature. Several studies however have recorded large heritability estimate of morphological

traits in birds (Merila and Shelden, 2001). Jensen et al. (2003) estimated heritabilities of

three morphological traits in house sparrow as 0.47+0.05, 0.48+0.2 and 0.41+0.06 for tarsus

length, bill and wing length respectively and reported that they ranged from 0.4 to 0.6,

although lower values were reported for body mass and body condition.

Hu et al. (2006) obtained heritability estimates of 0.37 (males) and 0.14 (females) for

feather length in Muscovy ducks . Lindsey (1988) reported that phenotypic traits are highly

variable and often have low heritability, reflecting substantial environmental influences. Heritability estimate for growth traits

Ricard et al. (1983) reported heritability of body weight of muscovy at 10 weeks of

age as dam variance component as 0.47(male) and 0.49 (female) and sire variance

component as 0.24 (male) and 0.62 (female). Pingel (1990) quoted values of heritability

from sire+ dam variance component which ranged from 0.47 to 0.53 for body weight at 70

and 74 days in the male Muscovy duck. Poujardien et al. (1994) also compiled values of

heritability estimates of body weight at 53 to 84 days that ranged from 0.17 to 0.55 for male

Muscovy duck which were force-fed. These estimates were obtained by using the variance

component estimation with model sire and dam nested within sire.As similar finding was

reported by Chamber ( 990) and Adeyinka et al. (2006 ) for broiler chickens. They further

reported that heritability normally increases with age and seems to be little larger in females

than males as found by Chapuis et al. (1996) for turkey and Mignon-Grasteau et al. (1998)

for muscovy. Mignon-Grasteau et al. (1998) found high values of the heritability for juvenile

body weight in Muscovy duck to be 0.40 and 0.51 for body weight at 6 weeks of age,

respectively for male and females. They also reported for weight gain between 6 weeks of

age and slaughtering age as 0.33 and 0.67 respectively for males and females. Ksiazkiewicz

(1995) reported an estimate of heritability of body weight at 7 to 8 weeks ranges between

0.28 and 0.76.

Hu et al. (2006) reported moderate heritability for body weight at 10 weeks of age

as 0.24 and 0.31 respectively for males and females and for 18 weeks of age, as 0.36 and

0.43 for males and females, respectively. The heritability estimates increased with age.

Similar findings were reported by Adeyinka (2006) for chicken. They seem to be larger in

females than in males as found by Chapuis et al. (1996) for turkey and Mignon-Grasteau et

al. (1998) for muscovy. They interpreted their findings to be as a result of more precocious

growth of males which at a given age is more mature than the female and so differ in their

body composition and percentage of fat tissue, which are heritable. Heritability estimates for egg traits and laying performance:

Cheng et al. (1995) estimated heritabilities of 0.09, 0.11, 0.12, 0,17, 019 and 0.20 for

egg shell strength at 40 and 30 weeks of age, egg number at 52, 30, 40 weeks of age, egg

weight at 30 weeks respectively in a Brown Tsaiya laying duck. Stasko (1966) estimates

ranged from 0.29 to 0.46 for annual egg production and 0.46 to 0.49 for egg weight in Pekin


Beaumont (1992) obtained estimate at 33 and 44 weeks of age from layer hens for

fertile eggs (0.23 and 0.22), hatched eggs (0.23 and 0.22) and maximum of fertility (0.20 and

0.21), respectively. Pingel (1990) reported heritability estimate of egg number based upon

sire and sire + dam variance component ranging from 0.17 to 0.42 in Pekin duck.

Brun and Larzul (2003) estimated heritability for fertility in pure bred Anas platyruynchos

Cross with Muscovy duck to be 0.32 for cross-bred and 0.15 for pure bred, while estimates

of hatchabilities were 0.36 and 0.16 for cross- and pure- breds.

Hu et al. (2004) estimated heritabilities for laying performance in a Muscovy duck

of Taiwan as 0.20+0.03, 0.23+0.0.3, 0.27+0.03, 0.20+0.03 and 0.22+0.03 for age at first lay,

number of eggs laid at 40 weeks, 52 weeks and number laid at 15 and 22 weeks of first lay. Heritability values and population divergence

The heritability (h2)of a population provides a measure of its genetic potential to

respond to a generation of selection. The magnitude of h2 only provides information on the

potential over a few generations. As allelic frequencies change so does heritability. A

population showing a high h2 value may have h2 eroded to zero very quickly while another

population with a much smaller heritability value may actually have heritability increase

during selection as rare favourable alleles become more frequent. Hence h2 is a completely

unreliable predictor for long term response although it is generally a good to excellent

predictor of short term response(Hu et al.,2004).

Likewise, measuring heritability value in two populations that show a difference in

their means provides no information on whether the under-lying difference is genetic.

Heritability is only a measure of the current variation in each population; it provides no


information on the past history of either populations. Thus high estimated h2 values in two

divergent populations do not imply that the divergence is genetic (it could be strictly

environmental). Likewise low estimate of h2 does not imply that an observed difference

between two populations is environmental; both populations could have exhausted genetic

variation during selection for their divergence. It can be concluded that variances within

population and means between populations are not comparable (Brun and Larzul, 2003).

2.4..2 Correlation between traits

This is the relationship between two characters and can either be positive or negative

in the individuals of a population. Correlated characters are of interest for several reasons.

Funk et al. (2004) reported that genetic correlation between pairs of traits had an important

effect on the direction and rate of phenotypic evolution. The genetic correlation between two

traits describes the proportion of the phenotypic correlation between the traits that is caused

by genetic variation that affects both traits simultaneously. Genetic correlation between

characters can arise by two mechanisms such as pleitropy or gametic phase disequilibrium

(Lynch and Walsh, 1998). Pleitropy occurs when a single gene affects multiple traits due to

complex biochemical and developmental pathways (Wright, 1968). Gametic phase

disequilibrium is the tendency of genes affecting different traits to be positively or negatively

associated in the same individual. Both pleitropy and gametic phase disequilibrium can cause

positive or negative genetic phase disequilibrium (Funk et al., 2004). Genetic correlation can

constrain or enhance phenotypic evolution depending on whether the correlations are positive

or negative. Genetic correlations are notoriously difficult to estimate accurately because they

require accurate estimates of the genetic variance of the trait and the genetic covariance

between them (Lynch and Ritland, 1999).


Grant and Grant (1995) found large and positive genetic correlation between

morphological traits in the medium ground Flinch (Geospiza fortis). Similarly, positive

genetic correlations have been found between various morphological traits in barnacle goose

(Branta leucosis),(Larsson and Fowlard, 1992). However, negative genetic correlations have

also been documented among morphological traits. Johnson and Johnson (1990) found

negative genetic correlation of -0.55 and 0.89 between bill width and body mass in a study of

feral pigeons (Columba livia)

The cause of negative genetic correlation demonstrated among morphological traits

may either be that genes or environment influences express these traits through different

physiological mechanisms or that the traits have been under directional selection

simultaneously in previous generation (Falconer and Mackey, 1996).

Alexander (2005) reported that when morphological traits are highly correlated it is

appropriate to describe their variation within generation or their change between generation

in terms of size and shape factor.

Many workers have used correlations among body parameters to assess the mutual

dependence and relationship among the parameters and possibly to implicate the genetic

control of certain part on the same genome region. Such knowledge may be crucial to the

constructive manipulation of measurable part in an attempt to alter conformation (Salako,


Braccini-Neto et al. (1997) reported that body weight showed desirable correlations

with sexual maturity, egg weight, laying rate and egg mass, but negatively correlated with

feed efficiency. Negative genetic correlations between some traits prevent fast improvement

of the productive performance of bird flocks, even if the flocks are subjected to constant


Teguia et al. (2007) in their study of live body weight and body characteristics of the

African Muscovy duck reported that body measurements were highly (P<0.01) correlated

with body weight for both male and female and added that the highest correlation coefficients

were obtained between wing length (0.990 and 0.1995) and thoracic perimeter (0.999 and

0.1970) for male and females, respectively. They added that live weight had a linear

relationship with both wing length (R2 = 0.99 and 0.81 ) and thoracic perimeter (R2=0.94 and

0.98) for male and female, respectively.

Adeogun and Adeoye (2007) reported that phenotypic correlation of body weight and

shank length was mostly positive, indicating that an increase in body weight will lead to an

increase in shank length and vice versa. Ibe (1995) earlier reported that selection on shank

length may not be useful in improving overall body growth of an animal unless selection is

designed to improve specific body area of prime economic value.

Hu et al. (2005) in their report on genetic parameter estimates of a closed

experimental strain of Muscovy duck in Taiwan observed that correlations between laying

traits were high, but that genetic correlation between age at first egg and egg number was

negative (- 0.45).

Poivey et al. (2001) reported in an experiment for a cross between Brown Tsaya

duck artificially inseminated with muscovy drake reported high genetic correlation between

number of fertile egg and maximum duration of fertility as 0.96 0.92 between number of

fertile egg and number of hatched males as 0.86 and 91.

In an experiment to estimate heritability and genetic correlation of body weight and

feather length in growing Muscovy duck, Hu et al. (1999) reported positive and high genetic

correlation between the traits at 10 weeks and 18 weeks of age for male and female.

Hu et al. (2006) reported genetic correlation between sexes for body measurements

of Muscovy duck at same age (0.90 at 10 weeks of age and similar 0.89 at 18 weeks of age)

They also reported genetic correlation between sexes for feather length (0.88±0.08). Feather

length in female was also reported to be genetically correlated with body weight of male

(0.8). Mignon-Grasteau et al. (1998) found genetic correlation of 0.88 between sexes in

Muscovy duck body weight at 6 weeks of age. Jenson et al. (2003) reported that there was a

positive genetic correlation between tarsus length and body mass where as tarsus length and

bill length were negatively correlated with body condition index. Chapuis et al. (1996) found

for three strains of turkey average genetic correlation between body weight of both sexes of

0.90 at 12 weeks of age and of 0.88 at 16 weeks of age; only a little inferior to the genetic

correlation between ages (0.95 for females and 0.99 for males). Le Bihan-Dival et al. (1998)

found in chicken very high genetic correlation between traits at the same ages (0.95) for body

weight at six weeks of age.

Poivey et al. (2001) in a work on parameter estimates of reproductive traits in

brown Tsaiya ducks artificially inseminated with semen from Muscovy duck reported that

phenotypic correlation between number of eggs set, total number of dead embryos and

number of hatched mules were positive but small( 33 , 27, 30%) while high and favorable

genetic correlation was found between number of fertile eggs and maximum duration of

fertility (0.96), number of fertile eggs and number of hatched mule (0.90). Positively and

highly correlated characters reveal common genes acting additively (Beaumont, 1992).

2.4.3 Divergence in populations of Muscovy duck

Variation among individuals and populations is a basic fact of biological life.

Biologists refer to this variation as heterogeneity and it takes many forms; from obvious

differences between the populations to subtle yet highly important differences in genetic

make-up, morphological and other features (Mignon-Grasteau et al.,1998). Hartmeyer (2006)

reported that water fowls (ducks ) are a wonderfully-diverse group of species with

remarkable variation in morphology, behaviour, physiology and distribution. They have

strong adaptive ability to specific geographical area.

In a Holoarctic species such as the Muscovy duck considerable polymorphisms (phenotypic

variation within interbreeding populations) exist resulting in separation of the various

morphology, white, black, black/white strains. Hartmeyer (2006) reported that genetically-

similar groups of individuals may be different due to distinct geographic locality and

phenotypic structures, and that high level of this diversity is not an indication of genetic

divergence, but an adaptation strategy of the population. Morphological differences among

population seem to be more closely tied to kinship distance than to genetic distance. This may

result from modification of some morphological features by environment rather than genetic


There is a marked difference in phenotypic traits of Muscovy ducks of tropical and

temperate climate that is evident, particularly in average body weight of male and female in

temperate and subtropical climate (Baeza et al., 1997). Subtropical climate with no controlled

environment could act both to decrease the body weight means and increase the phenotypic

variance(Hu et al., 2006 ). Yahav et al. (1998) found that high relative humidity (RH above

75%) at high ambient temperature deteriorated the growth performance between 10 and 15

weeks of age in turkey, may be due to severe respiratory alkaloses. Random drifts occurring

independently in different subpopulations lead to genetic differentiation between the

subpopulations (Falconer, 1988).

Poultry genetic improvement programmes are based on the genetic variability of

individuals, which may be changed by introducing new genotypes into the flock. Parents

showing high productivity indexes and great genetic diversity may produce a progeny that is

more productive and show great genetic variability (Piassi et al., 1995). Barbosa (2005)

reported that genetic divergence studies may be used to evaluate the behaviour of genotypes

in different environments.

Rafinski and Babik (2000) reported that genetic development may not be

accompanied by similar morphological differentiation. This reflects the hypothesis of

isolation by distance, and that morphological distances were independent of genetic ones.

Gunawan (1989) in a study of native ducks in Indonesia reported that there are some distinct

morphological differences between the native ducks of different regions such as West Java

,Central and East Java, Balil, Lonbok, North Sumatra and Kalimantan.

2.4.4 Genetic distance between populations:

Genetic distance is that difference between two entities that can be described by

allelic variation (Nei, 1973). This definition was later elaborated by Nei (1987) as the extent

of gene differences between population or species that is measured by some numerical

quantity. Hepp et al. (1988) defined genetic distance as a measure of the amount of genetic

divergence between groups. A more comprehensive definition of genetic distance is any

quantitative measure of genetic differences, be it at sequence level or at the allele frequency

level that is calculated between individuals, populations or species (Beaumont et al., 1998).

Studies on the genetic diversity of livestock have increased after the 70' s simultaneously to

the sudden increase in the availability of informatics resources (Sakaguti et al., 1996) .

Genetic divergence studies may be used to evaluate the behaviour of genotypes in different

environments and to evaluate the superiority of some genotypes over others. Similarly, to

identify divergent genotypes that may be used as parents in breeding programs and to relate

genetic divergence with heterosis (Piassi,1994).


Bruchi et al. (2003) reported that the distribution of genetic diversity within and

among populations is a function of the rate of gene flow between populations. The extent of

gene flow in a population depends on the distribution of the habitat it occupies and the size

and degree of isolation. Earlier on , Eding and Laval. (1997) reported that highly inbred

population tends to have an increased genetic distance to other breeds. For poultry,

establishment of measures of genetic distance among breeds and industrial stocks will be

important for identifying unique genetic types not represented (Nother, 1999). Eding and

Laval (1997) reported that genetic distances between breeds or populations are controlled by

mutation, genetic drift, selection and migration. Distance of zero implies no genetic

differences between groups and a maximum value indicates fixation for alternate alleles.

The very first insight into breed diversity can be achieved by examining differences

in phenotypic traits using certain cladistic or phenetic approaches (Moiseyera et al., 1994,

Romanov 1994). In general, phenotypes are divided into discrete traits such as morphological

characters and phenes and continuous traits, e.g. body measurement. These two are used in

assessing genetic variation and phylogenetic relationship between breeds and populations.

Joan et al. (2006) observed that phenotypic traits are highly variable, but often have low

heritability. Morphological distance

Morphological distance between groups or populations is the estimate of

morphological differentiation between the populations or groups and is usually estimated

using discriminant analysis (Thorpe, 1976; Misra and Cascadden, 1987; Hair et al., 1996).

The morphological distance between groups or populations is usually computed as Euclidean

distance. This measurement of differentiation is used as a prelude to estimation of genetic

distance between groups or populations, though several studies (Thorpe, 1982; Hilllis, 1984;

Rafinski and Babik, 2000) reported that morphological differentiation is not an indication of

genetic diversity between groups.

Bradford (2003) reported that morphological traits are generally believed to be

subjected to natural selection. Many birds have evolved divergent morphological structures.

Similarly, significant amount of genetic variation is unlikely to have accumulated as a result

of mutation, since the rate of phenotypic variation brought about by artificial selection in

animal breeding is much greater than the rate of change at the genetic level that is measured

by DNA marker. Estimation of genetic distance using morphological traits:

Several approaches have been developed in the estimation of genetic distance. The

conventional methods are those developed by Nei (1972) using population allele frequencies

for loci in genome. The problem of developing an index for measuring the distance between

populations using attribute data was first developed by Sanghvi (1953) who proposed an

index analogous to the Chi Square statistics for an evolutionary study. Earlier, Czekanowski

(1909) and Pearson (1926) evolved a concept for measuring population differences similar to

this. The method described by these, although, was primarily for classification of population

in terms of quantitative characters.

Fisher (1936) and Mahalanobis (1936) improved on this approach in estimation

using quantitative traits through development of a statistical model widely used today called

Mahalanobis D² statistic. Though morphological distance is likely to grossly under estimate

true levels of genetic variation between populations, it is capable of providing a sound

foundation and reference in the systematic evaluation and characterization of indigenous

species, and will also save cost of experimental material in terms of number of animals which

may be required for evaluation (Salako and Ngere, 2001). Several authors used this approach

in estimation of distance between population (Rodero et al., 1996 and Salako and Ngere ,

2001) in goats, Nikiforov et al. (1998) for chicken; Joan et al. (2006) and Pinheiro et al.

(2005) for fish and Murtha (2000) for horses. In livestock population genetic diversity is

expressed on the phenotypic level as variability in production triats or exterior trait

(Anderson, 2001). These phenotypic differences are the result of genetic diversity and

environmental differences (Oldenbroek, 1999).

Thorpe (1982) reported low genetic distance between frogs from Northern and

Southern Europe. That despite the value the groups are nevertheless unequivocally

genetically-distinct and the clear genetic divergence of the two population was not

accompanied by similar morphological differentiation. A similar finding was reported by

Marques et al. (2006) on Lustanian toadfish that morphological distances were independent

of genetic ones. Multivariate methods

With increase in the sample size of breeding materials and germplasm used in

animal improvement program methods to clarify and other genetic variabilities are

assuming considerable significance. The use of established multivariable statistical

algorithms is an important strategy for classifying germplasm and analyzing genetic

relationship among breeding materials (Mohammadi and Prassana, 2003). Multivariate

analytical techniques, which simultaneously analyze multiple measurements on each

individuals under investigation are widely used in analysis of genetic diversity irrespective

of dataset (morphological, biochemical or molecular marker data). Among these, algorithm

cluster analysis, principal component analysis, principal coordinate analysis and

multidimensional /discriminant analysis are at present most commonly employed and appear

particularly useful (Melchinger, 1993; Brown -Guedira et al., 2000). Barbosa (2005)

reported that techniques of multivariate analysis such as clustering analysis have been

successfully employed as a means to identify developing genotypes and better utilize the

advantages provided by heterosis; he added that the use algorithm produces clusters that

constitute a proposition about the basic and unknown organization of the data.

Cluster analysis refers to a group of multivariate technique whose primary purpose is to

group individuals or objects based on the characteristics they possess, so that individuals with

similar description are mathematically gathered into the same clusters (Hair et al., 1996).

Two types of clustering have been identified- distance -based and model- based method.

Principal component analysis (PCA) can be used as input for clustering rather than directly

applying data from quantitative characters when the correlation among the character are

significant (Goodman, 1972; Everitts, 1980) . Principal component analysis provides variable

independence and balanced weighting of traits, which lead to an effective contribution of

different characters on the basis of respective variation. On the basis of quantitative

morphological traits Van Bueningen and Busch (1997) applied such a procedure for analysis

of genetic diversity among north American spring wheat cultivars. Wiley (1981) defined

principal component analysis (PCA) as a method of data reduction to classify the relationship

between two or more characters and to divide the total variance of the original character into

a limited number of uncorrected new variables. This will allow visualization of the

differences among the individuals and identify possible groups. The reduction is achieved by

linear transformation of the original variable into a new set of uncorrected variables known as

PCs. The first step in PCA is calculating eigen value which defines the amount of total

variation that is displayed on the PC axes. The first PC summaries most of the variability

present in the original data relative to all remaining PCs. The second PC explains most of the

variability not summarized by the first PC and uncorrelated with the first and so on (Wiley,

1981). As PCs are orthogonal and independent of each other, each PC reveals different

properties of the original data and may be interpreted independently. In this way the total

variation in the original data set may be broken down into components that are cumulative.

The proportion of variation accounted for by each PC is expressed as the eigen value divided

by the sum of the eigen values. The eigen vector defines the relation of the PC axes to the

original data axes.

There are few reports about the use of principal component analysis in poultry. Ibe

(1989) analysed the body weight of chickens at different ages together with four body linear

measurements- breast and thigh width, shank and keel length. At all ages the first two

principal components explained at least 85% of the total variation. He further reported that

the principal component could be used in the selection index to simplify them because such

an index would have few PCs in the place of the original traits.

Abreu et al. (1999) used the principal component to evaluate the combining

ability of chicken strains for production traits. In line with his type of data, the methodology

was not efficient because it was not possible to explain high percentage of the total variance

with a reduced number of PCs. However, he worked with egg production and reproductive

data and concluded that principal component analysis was efficient, since it was possible to

explain 98% of the total variation with the first two components and also reported that

selection based on component of high discriminatory values may bring satisfactory result.

Debut et al. (2003) used principal component analysis for evaluating and understanding the

relationship among meat quality indicators such as muscle and meat quality traits and

concluded that it can be used to group traits and also develop selection index.

2.5 Molecular Techniques in the Study of Animal Genetic Diversity

A variety of different molecular techniques are being used for the study of inter-

and intra-specific genetic variation at the deoxyribonucleic acid (DNA) level (Gwakisa et al. ,

1994) . The most widely used techniques are Restricted Fragment Length Polymorphism

(RFLP) of nuclear DNA and Mitochondria DNA, Minisatellite, Microsatellite, Randomly

Amplified Fragment Length Polymorphism (RFLP) and sequencing of mitochondria DNA.

.These techniques differ in the way that they resolve genetic variations and in the taxonomic

levels at which they may be most appropriate. As rich selection of molecular techniques is

available, choice of method to use for genetic diversity is quite often dictated by the power

of the method to generate reproducible polymorphism that can either be tracked in a

mendelian fashion or can segregate a phenotypic trait in a predictable manner (Gwakisa et

al. , 1994).

Chatterjee et al. (2007) reported that several DNA-based techniques such as

polymerase chain reactions RFLP(PCR-RFLP) , random amplified polymorphic DNA (

RAPD) and microsatellite analysis are being adapted to identify breed-specific genetic

markers and to associate these markers with quantitative traits and disease resistance (Welsh

and McClelland, 1990; Williams et al., 1990). The development of molecular techniques has

created new possibilities for selection and genetic improvement of livestock. The discovery

of the PCR had a major impact on the development and application of various DNA markers

(Marle-Koster and Nei, 2003). Holsinger et al. (2002) earlier reported that molecular

markers derived from polymerase chain reaction (PCR) amplification of genomic DNA are

an important part of the tool kit of evolutionary geneticists . By detecting genetic variation,

genetic markers may provide useful information at different levels of population structure,

levels of gene flow, phylogenetic relationship, pattern of historical biogeography and the

analysis of parentage and relatedness (Feral, 2002). Molecular techniques have also proved

useful in the investigation of the origin and domestication of livestock species and their

subsequent migration as well as providing information on evolutionary relationship and

identifying geographical areas of admixtures among populations of different genetic origins.

The advent of molecular techniques has led to an increase in the studies that focus on

the genetic characteristics of domestic breeds using genetic markers (Sharma, 2002),

Genetic markers that provide different estimates of genetic diversity have been described.

The choice of the marker to use for genetic diversity is quite often dictated by the power of

the method used to generate a reproducible polymorphism that can either be tracked in a

mendelian fashion or can be shown to segregate using phenotypic traits in a predictable

manner (Hannotte and Jianlin, 2005) . The choice can also be influenced by the availability of

specialized equipment, operating of equipment and assaying and technical competence.

The first biomarkers to be widely used in livestock characterization were protein

polymorphism known as allozymes (Queller et al., 1993). Several livestock breeds have

been characterized for variation in different proteins (Di Stasio, 1997). Protein

polymorphism, although still used in population studies are of limited value in the

assessment of genetic variations. This is largely because of the relatively low levels of

polymorphism found in protein loci, resulting in a low taxonomic limit for the resolution

power of protein electrophoresis.

Molecular DNA polymorphism is now the tools of choice for the assessment of

genetic diversity among livestock breeds. They have great potentials for discovery of the

fundamental parameters or characteristics important in conservation , including past

effective population size (Ellegren ,1999; Lynch and Ritland , 1999) and sex specific gene

flow. According to Hannotte and Jianlin (2005) , important assumption needed for the use of

genetic markers include, neutrality of the polymorphism and the use of a relative small

number of independent segregating marker loci are a good predictor of the genomic

diversity of a population.

The association between molecular techniques and conventional animal breeding

methods to identify animals with higher genetic potentials tend to maximize the genetic gain

for the traits of interest . Wu et al. (2004) Observed genetic diversities arises from the

consequences of genetic drift and mutation.

2.5.1 Molecular Characterization of genetic diversity in poultry using RAPD;

The effectiveness of RAPD in detecting polymorphism between chicken

populations and their applicability in population studies and establishing genetic relation

among chicken populations had been reported (Sharma et al., 2001; Ali and Ahmad,

2001). It was also used to differentiate strains of chicken in many respect (Ahlarwat et al.,

2004; Chatterjee et al., 2007).

Thirteen highly-inbred chicken lines were genetically characterized by DNA fingerprinting

and polymerase chain reaction using arbitrary primers, the RAPD-PCR band sharing value

ranged from 0.66 to 0.99 for all between-line comparisms. The DNA finger printing (DFP)

banding sharing (BS) values among lines from different breeds ranged from 0.10 to 0.20

(Salem et al., 2005).

Smith et al. (1996) reported that 60 random primers were used in an RAPD analysis

to evaluate genetic polymorphism within and among four chicken breeds and two turkey

populations. Seventy percent of primers tested amplified patterns with at least one

polymorphic fragment in one or more of the populations. Genetic variability through RAPD

markers has been detected within and between strains in White leghorn populations (Singh

and Sharma, 2002), using 50 random decamer primers; only 12 primers detected

polymorphism between the strains. Between strain genetic similarity estimates based on band

sharing (BS) as well as band frequency (BF) ranged from 0.756 to 0.958 and from 0.830 to

0.996, respectively.

Bednarczyk et al. (2002) studied DNA polymorphism among pooled DNA of eight

goose lines by RAPD-PCR. The number of bands amplified by each primer ranged from 1 to

8 with a mean of 2.86. Some bands appeared specific for the live or genetic background.

Huang et al. (2003) found out that random primers were used for RAPD finger

printing in Chinese White Roman and Landaise geese to detect female specific DNA

sequence. They found that one of the primers used in the study produced a 938 bp sex-

specific fragment in all females and I in males of Chinese geese only. Also data showed that

a simple and effective PCR– based sexing technique could be used, suggesting its potentials

for use as population -specific markers (Maciuszonek et al., 2005).

In ducks, RAPD analysis of genetic polymorphism to estimate differences

between breeds and to determine interlineal differences had been reported. The use of

molecular markers generated from RAPD s has been applied in poultry studies and their

usefulness had been demonstrated. A PCR-based DNA finger printing method, termed

Randomly Amplified Polymorphic DNA (RAPD) profiles was developed by Welsh and

McClelland , (1990) and Williams et al. (1990) . In this method a single randomly-

generated primers of 10 bases in length or longer was use to prime and differentiate DNA

from various sources. Williams et al. (1990) reported that, RAPD is a useful technique,

which when used in combination with other tools could produce or generate molecular

element of great value for identifying productive characters. Several studies had used

molecular markers in poultry; it has been used as a tool to identify important productive

characters in native population even though these birds showed a great production capacity

(Welsh and McClelland , 1990)

2.5.2 RAPD –PCR Application in Ducks

Ducks are among the birds that are occasionally consumed and are raised intensively

and commercially. Because of the steady increase in duck meat consumption, duck farming

and production is currently receiving some form of attention particularly in riverine areas.

These raise the present interest in research and development of duck production and thus

duck breeding (El-Gendy, 2005). Breeding needs information on genetic background,

particularly the genetic variations. Very few reports are available about genetic information

on duck population (Pingel, 1990; Knust et al., 1995; Romboli ,1990). With the recent

advances in techniques in molecular genetics, the detail genetic information of animal is

available with high accuracy compared to the information obtained by pedigree relationship

and trait phenotypes. These techniques have been successfully employed to address the

genetic variation and in turn, the genetic diversity among different populations, which help in

breeding programme (El–Gendy, 2005).

Few studies have been recently performed to assess the genetic polymorphism in ducks using

DNA fingerprinting technique. Maak et al. (2000) developed microsatellite marker for the

white Pekin and Muscovy ducks. Also, Dolmatova et al. (2000) studied the possibility of

using RAPD marker for the detection of differences among lines of white Pekin ducks. It was

reported that genetic distance precisely reflected even the slight changes that occurred in the

genetic structure during breeding.

Tiang Fwu et al. (1998), using RAPD procedure showed that nine out of 40 arbitrary

primers amplified clear and reproducible bands. The phylogenetic distance between

muscovy and the domesticated ducks was high. Similarly in their second report the genetic

polymorphism in ducks using RAPD–PCR analysis to differentiates between white Pekin

and Muscovy ducks showed pattern of breed clustering adequately reflecting the actual

genetic background known from the history and genealogy of breeds.

Siripholvat and Teagoonrung (1998) and Tian- Fwu et al. (1998) reported polymorphic

pattern in DNA bands of different duck breeds belonging to both species cairina and anas.

Sharma et al. (2001) demonstrated the presence of monomorphic bands characterising

different breeds. The average heterozygosity estimated was generally less within either

cairina or anas species, revealing the less variation within species compared to that between

species. High levels of band sharing were found between Sudani breeds with an average of

0.74 in Egypt ( El -Gendy et al., 2005 ). They further reported genetic distance indices

between breeds of cairina species (Muscovy and Sudani) showing longer distances to each

other (0.405) compared to the distance between anas species (White Pekin, Damietti and

Khaki Campbell) which ranged from 0.264 to 0.383.

Nagamine and Higuchi (2001) reported the usefulness of genetic distance to classify and

elucidate the evolutionary relationship between populations, levels of genetic variation as

well as history of animals.

Xiao et al. (2004) used RAPD technique to evaluate the genetic diversity of Fugian

local duck population in different ecological zones. They showed that the genetic diversity in

East Fugian (67.97%) was higher than that of West Fugian (59.95%). Genetic differentiation

was estimated to be about 32.03% among populations of East Fugian and about 40.95%

among populations of West Fugian.



3.1 Research approach: Two approaches were adopted in the study; namely, Field
survey ( On-farm) data collection and On-station experiment.

3.1.1 Field survey: A survey on the indigenous local Muscovy duck was carried out at two
agro-ecological zones namely rainforest and guinea savannah of Nigeria to collectinformation
on some morphological characteristics of the ecotypes.

3.2 On-Station Experiment

3.2.1 Experimental site

This study was carried out at the duck unit of the Livestock Complex of the College
of Agriculture, Lafia, located along Doma road, Lafia, Nasarawa State. Lafia falls within the
Guinea Savanna zone of North Central Nigeria and is located between latitude 08. 300 N and
longitude 08. 320E with annual rainfall ranging from 952-1988 mm, and a mean monthly
precipitation of 150 mm. Its minimum and maximum daily temperatures average 20-370C.
Lafia has a mean relative humidity at noon varying between 14 and 74%. It has two distinct
seasons; the wet season covering late April to October and dry season covering November to
early April

The duck house consists of an open sided building divided into 20 pens. Each pen
measures 400 x 250cm2. They are arranged on either side of an alley with each pen having an
extended run for the ducks to bask and feed on green vegetation as suggested by (Nickolova,
2004). A separate building was used for brooding and rearing of the ducklings.

3.2.2 Foundation stock and their management

The base population used for the experiment was made up of local Muscovy ducks
from two agro-ecological zones of Nigeria, the rainforest and the guinea savannah. The
rainforest population or ecotypes were gathered from rural areas of Ikang, Akamkpa and
Calabar in Cross River State, Ekot Ekpene and I’tu in Akwa Ibom State and its adjoining
village of Uturu in Abia state. The guinea savannah ecotypes were collected from rural areas
of Mbaka'an and Mbagwen of Guma and Makurdi Local Government Areas of Benue state,
Assakio, Shabu and Doma in Nasarawa state and Katcha and Bida in Niger state Fig. 1.

The foundation stock consisted of 60 ducks and 10 drakes from each of the agro-
ecological zones (Rainforest and Guinea savannah). They were maintained on the farm as
separate, non pedigreed and unimproved random mating populations. The performance of the
two ecotypes had not been evaluated thus making the study a base study.

3.2.3 Production and multiplication of experimental birds:

For generation multiplication, each ecotype 60 ducks and 10 drakes were placed in a 10
replicate deep litter pen with each containing the ducks in a ratio of 1:6 (i.e., one drake to six
ducks) as foundation stock. The mating arrangement is shown in Appendix 2

Foundation Stock

The ducks and drakes were randomly placed into 10 replicate pens per ecotype in a
drake:duck ratio of 1:6 where they were mated and also to produce fertile eggs. The birds
were fed layers mash containing 18% crude protein at the rate of 170g per bird per day as
recommended by Meulen and Dikken (1999). Six nest boxes were provided in each pen for
laying eggs. Hatching eggs were given sire, dam, batch and ecotype identification.
Incubation of eggs was done naturally by individual laying bird. Six batches of natural
incubation were carried out in all. Ducklings hatched during three consecutive days were
considered as belonging to one batch. A total of 352 ducklings (192 from Guinea savannah
and 160 from Rainforest) was hatched and used for the experiment. Management of experimental birds:

After hatching, the ducklings were transferred from their mother immediately to a
brooding room. Ducklings of each ecotype and their respective batches were wing banded,
weighed and brooded separately on floor pen with wood shavings as litter material. Brooding
for each batch lasted eight weeks. On completion of eight weeks of brooding period the
ducklings were moved to rearing pens. The pens were open-sided, divided into 10 replicate
pens for each ecotype. The ducklings were managed in a biological breeding method in a
semi extensive system as suggested by Nickolova (2004). At 20 weeks of age 60 ducks and
10 drakes from each population were selected for performance testing, until when they
commenced laying. The ducklings were monitored for both first and second egg production /
annual egg production. At 30 weeks of age nest boxes equal to the number of ducks were

placed in each pen for egg laying. Eggs laid were recorded twice daily at 10 am and 5.00 pm
and recorded on the egg chart for each replicate pen and ecotype. Feeding and watering:

The experimental birds were fed on three types of diets according to their growth
phase namely chick mash, grower’s mash and layers mash. The composition and nutrient
analysis of the various diets are shown in Appendix 1. The chick mash was fed to the
ducklings between the ages of 0-8 weeks, grower mash between 9 to 30 weeks and layers
mash from 31 weeks and above. Feed and water were provided ad libitum during brooding,
and rearing, while at laying phase they were fed once daily at 180g per bird per day and
water was supplied ad libitum. Medication:

Incidence of two major viral diseases namely; Newcastle and coccidiosis occur very
often in the area. Therefore, the ducks were vaccinated against Newcastle disease at day old,
28 days and 67 days of age and against infectious bursal disease and fowl pox at the age of 14
days and 56 days, respectively. Coccidiostats and antibiotics were administered occasionally.
Similarly. vitalyte, a vitamin supplement was administered to enhance productivity.

3.3 Data Collection:

3.3.1. Survey data: Data on the morphological (phenotypic) characteristics of the indigenous
Muscovy ducks from the two agro-ecological zones were collected between 2006 and 2007
from 20 focal areas. The samples were taken from 15 households per focal area. Households
which are distantly located from each other were selected in each area. Overall of 680 ducks
were examined for morphological traits during the survey. Data were collected for a period of
10 months starting from October 2006 to July 2007. These birds were reared in village/rural
settings where they were left to freely roam about scavenging for food crumbs, kitchen waste,
insects, earth worms and leafy vegetation. They however return at dusk to the homestead
where minimum shelter and sometimes grain chaffs and supplements were provided. Only
adult birds that were above one year of age were used for data collection and measurements
were carried out in the morning before they were let out for scavenging. Twelve zoometric
traits (body weight, body length, body width, body height, neck length, bill length, bill width,
bill height, shank length, head length and wing length ) were covered using flexible tape

graduated in centimetres for length and a (10) kg capacity kitchen scale for weight as
outlined by (Willin and Erzsebet, 1997).

Body length- measure between the first cervical vertebrae and the pygostye

Bird height – measure from the legs on the ground up to the back of the body.

Body width – distance between the right to the left flank of the body.

Beak length – measure as length of the upper beak rim

Beak width – at the widest part of the beak between the right and the left distance

Beak height - at highest part of the beak.

Shank length – from the knee or knuckle (hock joint)to the region of the tarsus

Wing length – measure as the distance from the caput humeri to the third carpal digit

Head length –as distance between the end of the beak and the end condylus occipitale

Neck length –measure between the first and the last cervical vertebrae

Head width – at the widest part of the head

Head height- at the highest part of the head.

3.3.2 On-Station Experimental Data Growth Traits

a . Duckling hatch weight:- This was obtained when hatching process was over witnin 12
hours using a digital 250g capacity balance (Metler).

b. Body weight- individual live weights were taken at day old, week one, week 3 and 5

weekly interval from hatch to 20 weeks of age using a sensitive 500g digital balance. At 5-

20 weeks the 10 kg capacity kitchen scale was used and at 10 weeks sex

differentiation was done , respectively and weekly measurements were determined

separately for each gender and ecotype.

c Average daily weight gain/growth rate was calculated as

Daily WG = W2 – W1 /28 or 30 or 31


W1 = the initial weight in the month

W2 = the final weight in the month

28, 30 or 31 = the number of days in the month

d Body weight gain: Rate of gain was determined at 5-weekly interval. It was calculated

as weight gain in weeks expressed as a ratio of number of birds present during the 5


e Mortality: Mortality rate from 0 to 20 weeks of age was recorded in two phases 0-8

weeks and 9-20 weeks.

e Degree of sexual dimorphism

Degree of sexual dimorphism between male and female (DSD) in live weight was
calculated by the following formula proposed by Sezer et al. (2006)

Degree of sexual dimorphism (DSD) = MWt –FWt x 100


Where MWt =the mean male live weight at time t

FWt= the mean female live weight at time t

53 Egg production traits

a Age at sexual maturity/first egg. The age ( in days) at which 3 ducks out of the 6 ducks
in a replicate pen laid their first egg

. Thus the average of days in each ecotype was recorded as the average age at first egg

for that ecotype. Alternatively, number of days from hatch to first egg lay in a given


b. Egg production. Egg production was measured in two ways; Percent duck day and

duck-housed production

c. Percent duck day – This is the number of eggs produced expressed as a ratio of

average number of ducks alive during the period , multiplied by 100

d. Percent duck housed – This is measured as total eggs produced divided by the number

of ducks housed at the beginning of laying expressed as percentage.

e. Laying intensity- The number of eggs obtained for a week in a month by each ecotype is

considered as laying intensity according to the procedure outlined by Nickolova (2004)

1 = Ne x 100/ Nd x7 (28 or 30 or 31)


1 = laying intensity

Ne = number of eggs obtained for a week or month)

Nd = number of ducks

7 (30, 28 ,31) = number of days in a week (Month).

f Body weight at first egg- Average body weight at first egg was recorded for each


g Weight of first egg (WFE)- When the first egg of each duck was laid the weight (g) was

immediately taken. The average of all first eggs for each ecotype was computed as mean

for that ecotype.

h Egg weight – All eggs laid by each ecotype were usually weighed on a Friday, using a

digital electronic weighing balance. The mean of all single weights was computed to

form the mean weekly weight for each ecotype.

i Egg length and egg width –These were taken on the long side and breadth of each egg

at both first laying cycle and second laying cycle in the two ecotypes.

j Egg mass – This was calculated as the product of egg number and weight. Egg mass of

each ecotype was calculated on the basis of first egg production cycled and second egg

production cycle.

k Pause length/laying interval- This was calculated (in day) between first and second
laying cycles for each ecotype

3.4 Analytical Procedures For Morphometric and quantitative traits

3.4.1 Multivariate Analysis. Principal component analysis: Pearson’s coefficients of correlation (r) among the
various morphometric traits obtained from the field data were estimated. From the correlation
matrix, data were generated for the principal component factor analysis. Anti-image
correlations, Kaiser-Meyer-Olkin measures of sampling adequacy and Bartlett’s Test of
Sphericity were computed to test the validity of the factor analysis of the data sets. Principal
component analysis according to Everitt et al.( 2001), is a method for transforming the
variables in a multivariate data set, X1, X2, ……., XP, into new variables, y1, y2…, yp, which
are uncorrelated with each other and account for decreasing proportions of the total variance
of the original variables defined as:

y1 = a11 x1 + a12 x2 + .--- + a1pxp

y2 = a21 x1 + a22 x2+ --- +a2pxp

yp = ap1 x1 + ap2 x 2+ --- +app xp

With the coefficients being closed so that y1, y2 ---, yp account for decreasing

proportions of the total variance of the original variables, x1, x2, ---, xp. The factor programme

of SPSS 14.0 (2004) statistical package was used for the principal component analysis. Genetic distance estimation using discriminant analysis

Twelve zoometric traits (body weight, body length, body width, body height, ,neck length,
bill length, bill width, bill height, shank length, head length and wing length ) from the two
ecotypes were used and subjected to step wise discriminatory analysis and canonical
discriminant analysis to assess the discriminatory power of each variable, using the mean
procedure of SAS 1990 package and SPSS 14.0 2004 package. Genetic distance between
ecotypes was obtained using Mahalanobis D2 statistics of SAS (1990) using the mean of
each discriminate variable in each ecotype.

D2 =  Vij (xi - Yi) (Xj - Yj)


D2 = genetic distance between populations in an m-dimensional space

Vij = the element of ith row and jth column of the inverse matrix

Discriminate analysis was used to assess the degree of differences of the ecotypes by
multivariate measurement and to test the impact of individual variable on the discriminant
(Sokal and Rolf, 1995).

3.4.2 Phenotypic evaluation of the morphological and growth traits of the ecotypes:

Morphological traits such as (body weight, body length, body width. body mass,
neck length, beak length, beak width, beak height, shank length and head length ) of the
surveyed data and experimental data at 0, 3, 5, 10, 15, and 20 weeks were subjected to the

generalized linear model (GLM) procedure of SPSS 14.0 (2004) to a two factor (ecotype and
sex) for the surveyed data and three factor (ecotype , sex and batch) factorial analysis of
variance in a complete randomized designed in each case for the experimental data with the
following linear models:

Yijk = µ +Ei+Sj + (ES)ij + eijk


Yijk = Individual phenotypic observation

µ = Population mean.

E I = Effect of the ecotype (i = 1,2)

Sk = Effect of the sex (j= 1, 2 )

(ES)ij = Interaction effect of ecotype and sex

eijk = Residual errors

Yijkl = µ +Ei+Bj + SK + (EB)ij + (ES)ik + (BS)jk + (EBS) ijk + eijkl

Where : Yijkl = Individual phenotypic observation

µ = Population mean.

EI = Effect of the ecotype (i = 1, 2)

Bj = Effect of the batch (j= 1---)

Sk = Effect of the sex (j= 1, 2 )

(EB) ij = Interaction effect of ecotype and batch

(ES) ik = Interaction effect of ecotype and sex

(BS) jk = Interaction effect of batch and sex

(EBS)ijk = Interaction effect of ecotype, batch and sex

eijkl = Residual error


External egg traits (egg weight, egg length, egg width, data were analyzed using the
following model

Yij =µ + Ei + eij


Yij = Single observation

µ = Overall population mean

Ei = Effect of ith ecotype/genetic group (I , 2).

eij = Residual error

Body weights:- Data generated on body weight at hatch 0 week, week1, weeks 3, 5, 10, 15
and 20, were analyzed in two stages. In the first stage body weight from 0-3 weeks were
subjected to a two factor (ecotype and batch) factorial analysis of variance in a completely
randomized design. In the second stage body weight from 5-20 were similarly subjected to a
three factor (ecotype, batch and sex) factorial analysis of variance in a completely
randomized design. The first analysis considered the duckling as mixed sex, since apparent
sexual dimorphism is evident in the 3 weeks of life of the duckling, sex effect was included in
the second model in the analysis of the data (3-20 weeks). The generalized linear model
(GLM) procedure of SPSS version 14. (2004) was used in each with the following models.

Body weight from 0-3 weeks of age:

Yijk = µ + Ei + Bj + (EB)ij + eijk (1)

Body weight from 3 -20 weeks of age:

Yijkl = µ + Ei + Bj + S k + (EB)ij + (ES)ik + (BS)jk +(EBS)ijk+ eijkl (11)


µ = Population mean

Ei = Effect of i th ecotype (i =1, 2)

Bj = Effect of j th batch (j = 1…..)

Sk = Effect of k th sex (k = 1, 2)

(EB)ij = Interaction effect of ecotype and batch

(ES)ik = Interaction effect of ecotype and sex

(BS)jk = Interaction effect of batch and sex.

(EBS)ijk = Interaction effects of ecotype , batch and sex

eij and eijkl = Random errors

Yijk and Yijkl = Individual phenotypic observation

Using the above models body weight at 0, 1, 3 ,5, 10, 15, 20, weeks of age were analyses

Body Weight Gain and Average daily gain:

Body weight gain (BWG) during the periods from 0- 5, 6 -10, 11–15 and 16 -20,
and average daily gain in similar age interval were similarly subjected to the analysis of
variance using GLM procedure of SPSS 14. (2004) The Data were analysed in the following
stage with the appropriate linear models:

BWG and ADG at 0-5, 6- 10, 11-15 and 16-20 weeks

Yij = µ+ Ei+ eij (1)

BWG and ADG at 5-10,11- 15, and 16 -20, weeks

Yijk = µ = Ei + Sj + (ES)ij + eijk (11)


Yij and Yijk = Individual phenotypic observation

µ = Overall population mean

Ei =Effect of ith ecotype (i= 1, 2)

Sj = Fixed effect of jth sex (k= 1, 2 )

(ES)ij= Interactions of fixed effect of ecotype and sex

eij and eijk =Random errors


Egg production traits :

Laying intensity; Laying intensity was calculated on monthly basis for each ecotype using
monthly egg production, for the laying season and were compared using analysis of variance
SPSS 14. 0 (2004) fitting the linear model.

Yij = µ + ai + eij


Yij = Single observation

µ = Overall population mean

ai = Effect of ith genetic group (i = 1, 2)

eij = Residual error

Egg Number, Egg Weight and Egg Mass

The egg production of the ecotypes was divided into part period first laying cycle and
second laying cycle and was analysed using analysis of variance with fitting the model.

Yij = µ + ai + eij


Yij = Single observation

µ = Overall population mean

ai, = Effect of genetic group (i = 1, 2)

eij =Residual error.

Mortality – Mortality were recorded in percentage

3.4.3 Genetic Parameter Estimate Heritability estimates: Growth traits (body weight, body weight gain, and average
daily gain), egg production traits (age at first egg, body weight at first egg, egg number, egg
weight, egg length, egg width and egg mass and pause length) of each of the ecotypes and at

various ages were subjected to genetic analysis using the Mixed Model Least Squares and
Maximum Likelihood Computer Programme, (Harvey, (1990). The reduced Sire Model
(Becker, 1984) was used to fit the data.

Yij = µ + a1 + eij


Yij = observation on the jth bird of the ith sire

µ = Overall population mean

a1 = Random effect of sire (1…10)

eij = Residual error.

Sire variance component were used to estimate heritability, genetic and phenotypic
correlations of all traits using Harvey (1990).

The formulae for heritability used in computation by Harvey programme are given below

h2 = 4δs2



h2 = Heritability estimate from paternal half-sib analysis

δs2 = Cross classified family variance component estimate

otherwise known as sire variance component.

δe2 = Within family variance component otherwise known as

error variance component.

Standard error of heritability was as estimated according ( Swiger et al., 1964).

SE( hs2) = 4 2 (1-t)2 [1 + (k-1) t ]2

K (k-1) (s-1)


t = The intraclass correlation

t = δs2
δ 2
s + δw2

k = Progeny number within sire

s = Number of sires Genetic and Phenotypic correlations between traits

Genetic correlation between any two traits say X and Y was estimate from

rG = Cov (x, y)

δs2 (x)2 δs2 (y)2


rG = Genetic correlation estimate

Covs (xy) = Sire covariance component between trait x and y

δs2 = Sire variance component.

Phenotypic correlation

rp = δe2 (x)2 + 1 – NW δs2 (xi y)2


[δe2 (x)2 + 1 – NW δs2 (x)2] [δe2 (y)2 + 1 – NW δs2 (y)2]



rp = phenotypic correlation estimate

δs2 = sire variance component

δe2 = error variance component

NW = decimal percentage of additively genetic variance in δe2

NRI = decimal percentage of additivity in δs2 (Baker, 1984)

3.5. Molecular characterization

a Protocol

-Blood sample of about 10ml was collected from the brachial vein of 25 individual

birds of either sex from each of the two ecotypes.

-Genomic DNA was extracted by the use of phenol chloroform extraction method

Determination of genomic DNA concentration and DNA purity

The concentration and purity of individual genomic DNA samples were determine by using a

spectrophotometer. The optical density (OD) at 260 and 280 nm were measured. The purity

of DNA was determined by absorbance ratio A260/280. The DNA concentration was

calculated from the absorbance value at 260nm by the following formula:

DNA concentration(μg/μl)=A260 x dilution factor x 50


Agarose gel Electrophoresis

Agarose gel electrophoresis was used to determined the quantity of genomic DNA sample.

Agarose gel (0.8%) was prepared by using 0.8g of agarose powder mixed with 100ml x TBE

buffer and boiling until dissolved, then cooled down at room temperature and poured into

electrophoresis chamber set. The genomic DNA was mixed with 6 x loading dye, 25%

glycerol 60 nm EDTA, 0.25% bromophenol blue and loaded into the gel. Gel electrophoresis

was performed at 80 volt for 1hour

Primers used

Synthetic primers used for the RAPD analysis in this study were purchased from TAG
Copenhagen, Fruebjervel DK-2100 Copenhagen.


Primer Sequence 51 - 31 (G + C)% At 0o

Oligo 1 TCA CGA AGCC 60 28.9

Oligo 2 TGG ACC GGTG 70 33

Oligo 3 GAC CGC TTGT 60 28.9

Oligo 4 AAC GCG TCGG 70 33

Oligo 5 GAA CGG ACTC 60 28.9

Oligo 6 GTG AGG CGTC 70 33

Oligo 7 AAA GCT GCGG 60 28.9


b Random Amplified Polymorphic DNA-Polymerase Chain Reaction Analysis


-RAPD –PCR were carried out with the pooled and the individual genomic DNA


-Seven random primers were used and amplified by polymerase chain reaction

(PCR). Each sample was used for electrophoresis.

-RAPD patterns were visualized on a Ultraviolet (UV) transilluminators and


c Recording of data:

The RAPD bands were scored for their presence (1) or absence (0) .The index for

similarity between ecotypes and within ecotype was calculated using the formula

developed by (Lynch, 1990).

Bab= 2Nab/Na+Nb

Where Nab= number of fragments observed in individuals a and b

Na and Nb = total number of fragments scored in a and b

d Genetic distance based on band sharing:

The genetic distance between the populations was calculated based on band sharing

between the pooled sample profiles. The genetic distance between ecotypes was

calculated as described by Chatterjee et al .(2007) using the POPGENE


program (Population Genetic Analysis) version 1.31 (Yeh et al., 1999)

Dab=1/N.1- (Nab/Na+Nb-Nab)

Where Nab= number of common bands between ecotypes

Na= number of common bands in ecotype a

Nb=number of common bands in ecotype b

N= number of primers



Fig.1. Map of Nigeria showing locations of Muscovy duck source.



4.1 Morphological Differentiation Between Ecotypes

4.1.1 Body weight and body measurements

Mean squares obtained from analysis of variance for the effect of ecotype on body

weight and body measurements (body length, body width, bill length, bill width, bill height,

shank length, body height, head length, neck length, head width and wing length ) from field

data of the indigenous Muscovy duck are presented in Appendix 1. Table 2 presents the least

squares means and standard errors of body weight and body measurements (sexes combined)

for the guinea savannah ecotype and the rainforest ecotype. Ecotype had a very highly

significant (P<0.001) effect on body width, body height, head length, neck length, and wing

length, and also significantly (P<0.05) affected bill height and head width. However,

ecotype had no significant (P>0.05) effect on body weight, body length, bill length, bill

width and shank length. The result showed mean values of body width and head length of

14.58±0.24cm and 5.38±0.07cm, respectively for the rainforest ecotype and corresponding

values of 13.30±0.15cm and 4.80±0.05cm for guinea savannah ecotype. With respect to body

height, neck length and wing length the guinea savannah ecotype significantly had higher

values of 18.59±0.20cm, 14.34± 0.10cm and 25.60± 0.26cm, respectively when compared

with the respective values of 17.22±0.27cm, 13.46±0.14cm and 23.35±0.36cm for the

rainforest ecotype . However for bill height and head width were higher for the rainforest

ecotype (2.41±0.14cm and 2.82±0.06cm, respectively) than the guinea savannah

(1.91±0.01cm and 2.64±0.05cm, respectively).

Appendix 2 presents the mean squares for body measurements of the two Muscovy

duck ecotypes for the effect of sex, ecotype and their interactions. Sex had very highly

significant (P<0.001) effect on all the traits except for bill (P<0.05) and a non

significanteffect (P>0.05) on bill height. The least squares means for the effect of sex and

ecotype are shown in Table 3. In the two ecotypes and for all traits, higher values were

recorded for the male except for bill height in the guinea savannah ecotype, where the male

had less value than the female. In the forest ecotype the females showed greater variability in

all the body measurement characteristics except the wing length. Conversely, the males

demonstrated greater variability in body measurements than females (with the exception of

bill length, shank length and wing length ) in the guinea savannah ecotype.

Table 4 shows the coefficients of correlation between the body measurements of the

adult Muscovy ducks for the two ecotypes. In the male traits, only body length, body width

and wing length had positive (P<0.001) (P<0.05) correlation with body or live weight (0.187,

0.173 and 0.661) respectively. The highest correlation in males (0.698) was recorded between

head length and body width . In the females, all the traits had positive correlation with body

weight except for head width, with the wing length having highest correlation (0.535)

coefficient with body weight.

Kolmogorov-Smirnov test showed that most traits in the two ecotypes were normally

distributed; variances however were not always homogeneous among groupss.



Trait Ecotype Mean  SE Min. Max. CV

BWT (Kg) Guinea 2.400.05 2.31 2.49 32.31

Rainforest 2.420.06 2.30 2.54 32.99

BDL (cm) Guinea 25.220.19 24.86 25.60 11.10

Rainforest 24.900.26 24.40 25.40 15.32

BDD (cm) Guinea 13.300.15b 13.00 13.59 20.10

Rainforest 14.580.20a 14.18 14.97 15.41

BLL (cm) Guinea 5.360.23 4.91 5.81 88.99

Rainforest 5.030.31 4.41 5.64 23.64

BLD (cm) Guinea 2.810.07 2.67 2.94 35.91

Rainforest 2.940.09 2.76 3.12 16.10

BLH (cm) Guinea 1.910.10b 1.71 2.12 95.43

Rainforest 2.410.14a 2.13 2.68 65.59

SHL (cm) Guinea 5.660.16 5.34 5.97 59.10

Rainforest 5.430.22 4.99 5.86 18.53

BH (cm) Guinea 18.590.20a 18.20 18.97 21.07

Rainforest 17.220.27b 16.70 17.74 10.14

HL(cm) Guinea 4.800.05b 4.80 5.00 18.16

Rainforest 5.380.07a 5.25 5.52 15.37

NL (cm) Guinea 14.340.10a 14.14 14.54 13.63

Rainforest 13.460.14b 13.19 13.73 8.84

HD (cm) Guinea 2.640.05b 2.55 2.73 35.41

Rainforest 2.820.06b 2.69 2.94 16.10

WL (cm) Guinea 25.650.26a 25.13 26.17 15.23

Rainforest 23.350.36b 22.64 24.05 22.43

ab = means in the same column within trait sub-group with different superscript differ significantly
(P<0.05) cv = coefficient of variation. BWT= Body weight, BDL= Body length , BDD= Body width, BLL= Bill
Length, BLD =Bill width, BLH =Bill height , SHL= Shank length, BH= Body height, HL= Head length NL= Neck length,
HD =Head width, WL= Wing length



Trait Sex Guinea savannah CV Rainforest CV

Body weight (kg) Male 3.170.04a 19.89 3.280.52a 12.10
Female 1.910.03b 17.40 1.800.44b 12.83
Body Length (cm) Male 27.510.26a 6.46 26.660.33a 13.12
Female 23.780.20b 4.85 23.620.28b 15.03
Body width (cm) Male 15.980.15a 14.05 16.870.18a 6.14
Female 11.610.11b 9.26 12.910.15b 8.56
Bill Length (cm) Male 6.190.36 10.77 6.320.47 5.51
Female 4.840.29 60.77 4.090.40 11.99
Bill Width (cm) Male 2.960.12a 35.23 3.106.14a 9.16
Female 2.210.09b 24.20 2.820.12b 14.92
Bill height (cm) Male 1.760.16b 83.36 2.640.21 20.31
Female 2.010.25a 100.00 2.230.18 69.80
Shank Length (cm) Male 6.270.25a 10.43 6.480.18a 5.43
Female 2.270.20b 79.31 4.660.28b 10.78
Body height (cm) Male 21.110.27a 23.08 17.920.34b 4.56
Female 16.990.21b 11.09 16.720.30b 12.24
Head Length (cm) Male 5.430.07a 17.68 6.230.85a 5.43
Female 4.570.05b 15.39 4.270.07b 8.76
Neck Length (cm) Male 15.400.14a s9.36 14.350.19a 6.13
Female 13.670.11b 14.11 12.810.15b 7.27
Head width (cm) Male 2.840.2a 35.73 3.160.109a 9.16
Female 2.520.06b 34.20 2.570.08b 14.92
Wing Length (cm) Male 29.730.21a 6.61 29.010.28a 7.68
Female 23.070.17b 11.36 19.240.24b 9.57
(320) (308)

ab= means in the same column within the same sex subgroup with different superscripts are
significantly different(P<0.05)

cv=coefficient of variation

( )= number of observations





BWT .353 .074 .070 .141 .118 .114 .402 .019 .152 -.041 .535

BDL .189 .160 .021 .206 055 .094 .290 .092 .192 .030 310

BDD .173 .242 -.079 .157 .120 .017 .086 .462 .140 .156 -.277

BLL -.008 .027 .120 .004 -.017 .004 .000 -.131 -.071 -.015 .080

BLD -.184 .000 .511 .383 .004 .045 .188 .123 .220 .044 .135

BLH -.154 .019 .369 .049 .299 .001 .030 .053 .004 .019 .000

SHL -.050 .073 .635 .077 .623 .291 .189 .034 .254 -.033 .198

BH -.040 .023 .242 .003 .182 -.160 .416 -.084 .575 -.084 .482

HL -.044 .156 .698 .064 .589 .476 .589 -.032 .218 .221 -.079

NL -.119 .242 .466 .055 .450 .199 .429 .275 .444 -.064 .553

HD -.042 .027 .261 .036 .177 .197 .158 -.111 .308 .138 -.111

WL .661 .169 .190 -.027 -.112 -0.49 -.056 -.044 .040 .095 .058

BWT= Body weight, BDL= Body length , BDD= Body width, BLL= Bill Length, BLD

=Bill width, BLH =Bill height , SHL= Shank length, BH= Body height, HL= Head length

NL= Neck length, HD =Head width, WL= Wing length


4.1.2 Results of Principal Component Analysis of the Morphological Traits

Table 5 presents anti image correlation matrix for the body measurements in the male

Muscovy duck. Measures of sampling adequacy ranged from .386 (for bill length) to .835

(for head width). Other images were either negative or positive but not significant.

Anti image correlation computed for male body measurements (Table 5) and female

body measurements (Table 7) show that partial correlations were low, indicating that true

factor existed in both the male and female data. Kaiser-Meyer-Olkin measure of sampling

adequacy studied from the diagonal of partial correlation reveals the proportion of the

variances in the male body measurements caused by the underlying factor (0.709) and for

female (0.584). The overall significance of the correlation matrices tested with Bartlett’s test

of sphericity shows for males the chi square of 769.370 was significant at P<0.001, while

for female chi square of 597.331 was significant at P<0.001. These two tests provided enough

support for the validity of the factor analysis for the male and female data sets.

After Varimax rotation of the component matrix in the male traits four factors with

ratio of variance of 66.839% were extracted as shown in Table 6 . The factor pattern

coefficients were used to assess the relative contributions of the various body measurements

towards determining the numerical value of the corresponding factor, principal component.

Similarly, the variance of the variable was partitioned into a common portion “Communality”

shared with some or all of the corresponding factors or principal components. This showed

that 25.8- 90.0% of the variation in body measurement traits of the male ducks of the two

ecotypes were brought about by the principal components. The first factor was sufficient to

explain 30.458% of the total variance among the 12 traits. The traits associated with the first

factor included body width, bill width, shank length, neck length, head length and body

height. The second factor was associated with body weight and wing length.

Table 7 presents the anti image correlation matrix of body measurements in female

Muscovy ducks ,using measures in female muscovy ducks to measure sampling

adequacy(MSA). Between image correlation ranged between .417 to .884 with the highest

mirror image correlation found between shank length and the least between body width.

Table 8 shows the result of Varimax rotation of the component matrix of the female

traits. Four factors with ratio variance of 56.402 were extracted. The communality values

showed that 26.4 – 76.8% of the variation in the traits were brought about by the principal

components. The first factor in the female traits was sufficient to explain 23.236% of the total

variance among the body measurements. The traits associated with the first factor included

body height, neck length and shank length. The second factor was associated with body

weight and body length while others were associated with the third and fourth factors.



BWT .487a -.133 -.241 -.047 .107 .170 -.085 .084 .069 .255 .102 -.637
BDL -.113 .641 -.132 -.071 .150 .050 .032 .039 -.053 -.228 .034 .024
BDD -.241 -.132 .803a -.081 -.028 -171 -.214 -.238 -.434 -.113 -.121 -048
BLL -.047 -071 -.081 .386 -.463 -.008 .144 .036 .168 .080 -.009 .019
BLD .107 .150 -.028 -.463 .762 .022 -343 .018 -.268 -.203 .011 .043
BLH .170 .050 -.171 -.008 .022 .815 -.081 .237 -.166 .006 -.008 -.028
SHL -.085 .032 -.214 .144 -.343 -.081 .793 -.404 -.217 .000 -.007 .121
BH .084 .039 -.238 .036 .018 .237 -.404 .451 .359 -.192 .127 .013
Hl .069 -.053 -.434 .168 -.268 -.166 -.217 .359 .769 -.110 -.097 -.014
Nl .255 -.228 -.113 080 -.203 .006 .000 -.192 -.110 .784 .000 -.236
Hd .102 .034 -.121 -.009 .011 -.008 .007 .127 -.097 .000 .835 .081
Wl -.637 .024 -.048 .019 .043 -.028 .121 .013 -.014 -.236 -.081 523a

a= measure of sampling adequacy (MSA), BWT= Body weight, BDL= Body length , BDD= Body width,
BLL= Bill Length, BLD = Bill width, BLH = Bill height , SHL= Shank length, BH= Body height, HL= Head
length NL= Neck length, HD = Head width, WL= Wing length





Trait Factor 1 Factor 2 Factor 3 Factor 4 Communality.

Body weight -.099 .881 -090 .060 .798

Body Length .231 .434 .039 -.122 .258

Body width .745 .298 .347 .090 .773

Bill length .051 -.004 -.014 .952 .909

Bill width .668 -.173 .249 .492 .780

Bill height .240 -.128 .706 .004 .573

Shank length .844 -.054 .121 .076 .734

Body height .629 -.053 -.613 -.076 .779

Head length .625 .063 .618 .066 .781

Neck Length .711 .061 .099 -.043 .521

Head width .105 .049 .582 -.003 .358

Wing length -.028 .869 .033 .009 .757

Eigen values 3.655 1.873 1.405 1.087

% Variance 30.458 15.608 11.712 9.662

% Total Va. 30.458 46.066 57.778 66.839





BWT .499a -.110 -.291 -.006 -.045 -.100 -.033 -.272 -.019 .405 .041 -.558

BDL -.110 .797a -.175 .002 -.123 -.010 -.021 -.107 -.007 .082 -.303 -.182

BDD -.291 -.175 .417a -.019 -.071 -.076 .004 -.002 -.328 -.303 -.048 .506

BLL -.006 .002 -.019 .612a -.020 .015 -.011 -.011 .091 .094 -.016 -.085

BLD -.045 -.123 -.071 -.020 .846a .025 .022 -.029 -.021 -.107 -.034 .008

BLH -.100 -.010 -.076 .015 .025 .671a .008 .011 -.001 -.001 -.004 .021

SHL -.033 -.021 .004 -.011 .022 .008 .884a -.035 .005 -.135 .00- -.018

BH -.273 -.103 -.002 -.011 -.029 .011 -.035 .771a .036 -.423 0.42 -.009

HL -.019 -.007 -.328 .091 -.021 -.001 .005 .036 .651a -.173 -.245 .050

NL .405 .082 -.303 .094 *.107 -.001 -.135 -.423 -.173 .509a 089 -.580

HD .041 -.030 -.048 -.016 -.034 -.004 .009 .042 -.245 .089 .619a -.032

WL -.556 -.182 .506 -.085 .006 .021 -.018 -.099 .050 -.580 -.032 .533a

a = measures of sampling adequacy (MSA) BWT= Body weight, BDL= Body length , BDD= Body width, BLL= Bill
Length, BLD = Bill width, BLH =Bill height , SHL= Shank length, BH= Body height, HL= Head length , NL= Neck
length, HD = Head width, WL= Wing length



Trait Factor 1 Factor 2 Factor 3 Factor 4 Commun

Body weight .188 .739 -.049 .228 .636

Body length .120 .642 .204 .076 .474

Body width .027 .018 .772 .231 .650

Bill length -.364 .471 -.164 -.329 .490

Bill width .147 .356 .342 -.234 .323

Bill height -.106 .166 .052 .852 .768

Shank length .512 -.607 -.023 -.033 .264

Body height .661 .430 .040 .040 .625

Head length .180 -.057 .776 .053 .641

Neck length .182 .184 .150 -.063 .736

Head width -.248 .087 .539 -.205 .402

Wing length .504 589 -.299 -.062 .758

Eigen values 2.788 1.812 1.152 1.017

% Variance 23.236 15.097 9.597 8.471

% Total Var. 23.236 38.334 47.931 56.402

Var = Variance

4.1.3 Results of Discriminant Analysis of Morphological Traits

Using the 12 morphological traits from the two ecotypes, test of equality of group

means of the two ecotypes was analysed (Table 9). Using Wilks lambda of the 12 traits,

only seven traits (body width, bill height, body height, head length, neck length, and wing

length ) were significant at (P<0.001) and head width (P<0.05) as shown in Table 11.

Table 10 gives the summary of the canonical discriminant analysis, using the seven

traits that showed significance in test of equality of group means of the two ecotypes

differentiating the two populations with the guinea savannah ecotype centriod as 0.619 while

that of the rainforest ecotype was -1.151 as shown in Table 10.

For the multivariable model a Muscovy duck was classified as guinea savannah if

Disciminant score was greater than zero (D>0) and rainforest ecotype if less than zero (D<0).

The unstandardized canonical discriminant function was used to classidfy individual birds.

Head length, neck length, wing length, body width, body weight, body height, and bill width

were the variables included in the discriminant (D) equation below :

D= -.990BWT- .351BDD -.154BLD +.105BH - .400HL +.212NL+.328WL-3.354

The classification function could be directly used to identify the two populations, since

positive scores indicate guinea savannah and negative scores indicate rainforest ecotype.

Table 11 presents summary of stepwise discriminant analysis showing those variables that

can discriminate between the two ecotypes. When the seven morphometric variables had

been entered, Wilks (λ) dropped to 0.583 with a significant difference between the two

ecotypes such that F = 45.095 was highly significant at P<0.001 . The seven variables

described the morphological differences between the studied populations at the specific level

of each variable accepted into the model. These are the only variables used in the population

differentiation. The discriminant variables were body weight, body width, bill width, body

height, head length, neck length, and wing length . Detail of Fisher's linear discriminant

functions of all the seven characters used in discriminating and considered in the analysis of

sthe two ecotypes are presented in Table 12. The discriminant functions appear to distinguish

the population correctly cent per cent as evident from the sample data. Neck length had the

highest discriminating function of 1.473, which was followed by body width while bill width

had the least as shown in Table 10.

Table 13 shows the classification results of discriminant analysis obtained by the

application of non standardized canonical discriminant function coefficient. The variables

from the rainforest ecotype were most correctly classified as 84.1% while those of the guinea

savannah ecotype was 82.3%. Cross validation and overall success rate (82.9%) of the

original groups (guinea and rainforest) were correctly classified.

The squared Mahalanobis distance between the ecotypes was 2.963 (Table 14). The test of

significance of the squared Mahalanobis distance (F–test) with all the traits and degree of

freedom was presented in Tables15 . The F–ratio was 24.783 at (P<0.001) . Validating the

distance estimation using the variables. A cross validation testing procedure was performed

to assess the ability of the selected variables to predict Muscovy ducks from the two




Trait Wilks F df1 df2 Significant

(Prob. Level)

Body weight 1.000 .085 1 447 .771

Body Length .998 1.044 1 447 .307

Body width .944 26.370 1 447 .000

Bill length .998 .741 1 447 .390

Bill width .997 1.381 1 447 .240

Bill height .982 8.078 1 447 .005

Shank length .998 .702 1 447 .402

Body height .963 17.175 1 447 .000

Head length .935 30.924 1 447 .000

Neck Length .944 26.739 1 447 .000

Head width .989 4.848 1 447 .028

Wing length .943 26.785 1 447 .000




Trait Standardized coeff. Unstandardized Correlation


Body weight -.774 -.990 -.016

Body width -.884 -.351 -.287

Bill width -.176 -.154 -.066

Body height .351 .105 .232

Head length -.352 -.400 -.311

Neck length .363 .212 .289

Wing length 1.473 .328 .289

Constant -3.354

Guinea centroid .619

Rainforest centroid -1.151




Step Variable Wilks Unstandardized Correlation

coefficient coefficient

1 Head length .935 30.924 0.000

2 Neck length .774 65.252 0.000

3 Wing length .723 56.757 0.000

4 Body width .634 64.201 0.000

5 Body weight .609 56.795 0.000

6 Body height .590 51.235 0.000

7 Bill width .583 45.095 0.000




Traits Guinea Rainforest

Body weight -9.181 -7.429

Body width -.159 .464

Bill width -.154 .118

Body length .856 .641

Head length 1.895 2.602

Neck length 3.041 2.666

Wing length 1.619 1.039




Group Predicted Group Corrected


Guinea Guinea Rainforest Total

Count 241 52 293

Rainforest 25 132 157

% Guinea 82.3 17.7 100

Rainforest 15.9 84.1 100

82.9% of original grouped cases correctly.



Class Guinea Rainforest

Guinea 0.00 2.963

Rainforest 2.963 0.00



1 1 2 2

Class F P F P

1 24.78275 0.000

2 24.78275 0.000

4.2 Performance in Growth Traits

4.2.1 Body weight

Mean squares from the analysis of variance showing the effect of ecotype on body

weight from day old to 20 weeks of age are presented in Appendix 3. Table 16 and Fig.2

shows the body weight performance (sex combined) of the guinea savannah and rainforest

ecotypes and the graphical presentation of weight increase. Ecotypes had a very highly

significant (P<0.001) effect on body weight only at day-old and highly significant (P<0.01)

effect at week 3, week 10 and week 15. At week one and week 20, ecotype did not show any

significant (P>0.05) effect on the body weight of the ducks.

At day-old, the body weights were 39.85±0.13 and 36.88±0.15g for the guinea and

rainforest ecotypes, respectively. Higher weight was in favour of the guinea savannah

ecotype. Batch effect on body weight was highly significant (P<0.01) at day-old and very

highly significant (P<0.001) at week 1, but not at the subsequent ages.

At week 20, all the effects( ecotype, batch and their interactions) were not significant

(P>0.05) on the body weight of the Muscovy duck, though the body weight of the guinea

savannah ecotype was consistently higher from day-old to 20 weeks of age (Table 16).

Sexual differences in body weight was obvious (P<0.001) from the third week of age

in both ecotypes up to 20 weeks of age. Mean squares from the analysis of variance showed

the effect of ecotype, sex and their interactions on body weight at 3 and 5 weeks of age . At

week 10 to 20, there was no significant effect(P>0.05) of Ecotype x Sex interaction on body

weight, except at 15 weeks (P<0.05).

Table 17 and Fig.3 presents the least squares means of the body weight of the two

ecotypes by sex and graphical representation of the weight increase. There was significant

difference (P<0.01) between males and females withnin ecotype, with the males

demonstrating heavier weight than the females from the third week up to 20 weeks. Males did

not differ significantly (P>0.05) between ecotypes in body weight as from three weeks to 20

weeks of age. However, significant differences existed in body weight of females among the

ecotypes within these ages with the females of the guinea savannah being heavier than those

of the rainforest.



Age(weeks) Body weight (kg)

Guinea CV Rainforest CV

0 39.85± 0.13 4.51 36.88±0.15 5.28

(192) (160)

1 49.05±0.18 6.07 49.39±0.20 3.83

(181) (146)

3 170.61±2.42a 18.48 160.99±2.73b 18.44

(165) (136)

5 399.19±4.07 14.50 387.09±4.83 11.35

(164) (136)

10 1162.00±17.5 16.66 1152.62±64.35 92.55

(156) (131)

15 1709.54±47.61a 34.64 1510.82±83.85b 38.01

(155) (111)

20 1963.59±52.38 31.58 1958.46±158.96 35.16

(152) (107)

ab = means in the same row within age subgroup with different superscripts' differ

significantly (P<0.05)

2000 Variable

body weight



1 2 3 4 5 6 7





Age Sex Guinea(g) CV Rainforest(g) CV

Week 3 Male 195.70±2.43a 12.51 195.01±2.85a 9.76

(63) (43)

Female 154.52±1.93b 15.66 141.37±2.13b 8.50

(102) (80)

Week 5 Male 464.07±3.72a 7.34 428.36±4.73a 8.88

(62) (41)

Female 359.95±2.85b 6.12 366.53±3.27b 7.85

(102) (95)

Week 10 Male 1308.74±108.25a 11.85 1494.33±116.29a 6.44

(56) (40)

Female 1068.51±87.31b 14.54 956.87±21.72 14.34

(100) (91)

Week 15 Male 2365.53±32.74a 15.92 2209.85±54.89a 18.34

(56) (35)

Female 1276.02±25.88b 7.00 1158.27±47.53b 12.31

(99) (76)

Week 20 Male 2681.39±29.18a 10.15 2676.25±88.89a 13.54

(53) (35)

Female 1491.16±23.70b 12.24 1474.50±70.27 6.26

(99) (72)

( )=number of observations, ab = means in the same row within age subgroup with different

superscripts differ significantly(P<0.05)


3000 Variable
guinea male
guinea female
2500 rainforest male
rainforest female

body weight




1 2 3 4 5



4.2.2 Sexual dimorphism in body weight amongst ecotypes.

Appendix 4 and Table 17 show significant (P<0.001) effect of sex on body weight

from week 3 to the adult age in the Muscovy duck of the two ecotypes. Measures of degree of

sexual dimorphism are presented in Tables 18 and 19 for the guinea and rainforest ecotypes.

In the guinea savannah ecotype, the degree of sexual dimorphism estimates were 20.85,

22.44, 18.36, 46.06 and 44.39 for week 3, 5, 10, 15, and 20. (Table 18). In the rainforest

ecotype estimates of degree of sexual dimorphism were 27.51, 14.43, 35.97, 47.59, and

44.96 at 3, 5, 10, 15, and 20 weeks of age as shown in Table 19. In the two ecotypes similar

trends were noticed in the degree of sexual dimorphism with the highest at 15 week of age,

with tendency to progress with age for the two ecotypes.

Between the two ecotypes sexual dimorphism was exhibited at the earlier stage in the

rainforest ecotype than in the guinea savannah ecotype however at 20 weeks similar degrees

were shown (44.39 and 44.96) for the guinea and rainforest ecotypes respectively.



Age Male CV Female CV DSD

Week 3 195.70±2.43 12.51 154.52±1.93 15.61 20.85

(63) (102)

Week 5 464.07±3.73 7.34 359.95±2.85 6.12 22.44

(62) (102)

Week 10 1308.74±108.25 11.85 1068.51±87.31 14.54 18.36

(56) (100)

Week 15 2365.53±32.74 15.92 1276.02±25.88 7.00 46.06

(56) (99)

Week 20 2681.39±25.18 10.15 1491.16±23.20 12.24 44.39

(53) (99)

DSD= degree of sexual dimorphism

( )= number of observations



Age Male CV Female CV DSD

Week 3 195.01± 2.85 9.76 141.37±2.13 8.50 27.51

(43) (80)

Week 5 428.36±4.73 8.88 366.53±3.27 7.85 14.43

(41) (95)

Week 10 1494.33±116.29 112.28 956.87±81.57 20.35 35.97

(40) (95)

Week 15 2209.85±51.89 18.34 1158.27±47.53 12.31 47.59

(35) (76)

Week 20 2679.25±88.85 13.57 1474.50±70.27 6.25 44.96

(35) (72)

DSD=degree of sexual dimorphism

( )= number of observations

4.2.3 Body weight gain and average daily gain (BWG and ADG)

The mean squares from the analysis of variance of the effect of ecotype and least

square means of the body weight gain and average daily gain (0 to 20 weeks of age) are

presented in Appendix 5 and Table 20. Ecotype significantly (P<0.01) affected body weight

gain between 5 to 10 weeks of age with the guinea savannah ecotype gaining faster than the

rainforest ecotype. Between 15 and 20 weeks of age, ecotype also significantly (P<0.001)

affected both body weight gain and average daily gain with higher values in favour of the

rainforest ecotypes. From the results body weight gain and average daily gain showed gradual

increase in the ecotypes from day- old to 10 weeks of age. After 10 weeks of age there was a

decline in both body weight gain and average daily gain with advancing in age.

Appendix 6 and Table 21 outline the mean square and least squares means for the effect of

ecotype and sex on body weight gain and average daily gain.



Age Guinea Rainforest

0-5 (No of birds) (164) (136)
BWG 358.247±5.033 351.774±5.704
ADG 10.371±0.118 10.064±0.133

6-10 (No of birds) (156) (131)

BWG 755.320±11.748a 647.179±13.316b
ADG 21.788±8.012 32.104±9.081
% mortality (0-10 weeks) 4.88±0.12 3.68±0.16

11-15 (No of birds) (155) (111)

BWG 566.158±31.297 483.821±38.425
ADG 41.245±13.789 31.535±11.829

16-20 (No of birds) (152) (107)

BWG 242.552±11.023b 330.236±12.494a
ADG 6.928±0.389b 9.861±0.441a
% mortality (10-20 weeks) 1.96±0.02 3.60±0.34
BWG=body weight gain, ADG= average daily gain

= means in the same row within an age subgroup with different superscripts differ

significantly (P<0.05)



Parameters/age/(weeks) Guinea Rainforest

Male (No of birds) (56) (40)
Body Weight Gain (g) 896.366±13.174 796.683±15.805
Av. Daily Gain (g) 26.132±13.140 21.943±15.763

Female (No of birds) (100) (91)

Body Weight Gain (g) 671.263±10.171 572.422±11.125
Av. Daily Gain (g) 19.195±10.144 37.182±11.146
Male (No of birds) (48) (35)
Body Weight Gain (g) 1039.407±25.172 920.561±30.202
Av. Daily Gain (g) 76.184±22.479 25.300±26.966

Female (No of birds) (93) (76)

Body Weight Gain (g) 284.121±11.43 265.451±21.356
Av. Daily Gain (g) 20.420±17.352 7.653±19.067
Male (No of birds) (53) (35)
Body Weight Gain (g) 286.966±15.062b 4846.366±0.074a
Av. Daily Gain(g) 7.924±0.591b 13.605±0.710a

Female (No of birds) (99) (72)

Body Weight Gain (g) 216.091±11.631b 253.171±12.780a

Av. Daily Gain (g) 6.374±0.457b 7.990±0.502a

g= gram

4. 3 Genetic evaluation of growth traits

4. 3. 1 Heritability estimate of body weight

Heritability estimates obtained from sire variance component for body weight at

various ages for the genetic groups under consideration (guinea and rainforest) are presented

in Table 22. There are variations in the heritability estimate, (h2) between the genetic groups

in all ages. The heritability estimate for the guinea savannah ecotype ranged from h 2= 0.10 ±

0.17 at 3rd week of age to 1.40 ± 0.50 at day-old which is though outside the normal range of

h2 estimate. In the rainforest ecotype the heritability estimate ranged from h2 = 0.02 ± 0.16 at

3rd week of age to 1.04 ± 0.418 at first week of age. Generally, heritability estimates differ

from one age to another in both ecotypes. Lower heritability estimates were recorded at the

third week and progressively increased in later ages.



Age in weeks Ecotypes
Guinea savanna (h2) Rainforest (h2)
Day old 1.41 ± 0. 50 0.58 ± 0.37
I week 0.89 ± 0.41 1.04 ± 0.48
3 weeks 0.10 ± 0.17 0.02 ± 0.16
5 Weeks 0.17 ± 0.08 0.03 ± 0.18
10 Weeks 0.25 ± 0.22 0.11 ± 6.19
15 Weeks 0.13 ± 0.22 0.10 ± 0.01
20 Weeks 0.24 ± 0.26 0.05 ± 0.20

4.3.2 Heritability for body weight gain

Estimates of heritability from sire variance component for the 5-weekly body

weight gain for the guinea savannah and the rainforest ecotypes are presented in Table 23.

The estimates for the guinea savannah ecotype ranged from h2 = 0.01 ± 0.16 at 5– 10 weeks

to 0.32 ± 0.42 at 15– 20 weeks of age. The estimate of the 1.00 ± - 0.24 is outrageous.

Variation occurred in the heritability estimates in both ecotypes, which tened to increase with

age in the guinea savannah ecotype.




Age (in weeks ) Ecotypes

Guinea savanna (h2 ) Rainforest (h2)
0 –5 1.00 ± 0.24 0.14 ± 0.18
6 – 10 0.01 ± 0.16 0.25 ± 0.26
11 – 15 0.16 ± 0.17 0.01 ± 0.42
16 – 20 0.53 ± 0.37 0.32 ± 0.42

4.3.3 Heritability estimate for average daily gain

Table 24 shows the heritability estimates for 5-weekly average daily gain from

sire variance components for the two genetic groups. For the guinea savannah ecotype it

ranged from h2 0.01 ± 0.16 at 0– 5 weeks of age to 0.45 ± 0.33 at 16– 20 weeks of age, while

in the rainforest ecotype, heritability estimates ranged from h2 0.01 ± 0.16 at 0 – 5 weeks of

age to 0.35 ± 0.30 at 5– 10 weeks of age. In the guinea savannah ecotype there was an

increasing trend from 0 – 5 to 16 – 20 in the estimates.A similar pattern was noticed in the

rainforest ecotype, except that at the 16 – 20 weeks the estimate of 1.23± 0.43 was outside the

normal parametric range.

4.3.4 Phenotypic correlation of body weights at specific ages

Phenotypic correlation between body weights at specific ages for the guinea

savannah ecotype and the rainforest ecotype of indigenous Muscovy ducks are presented in

Table 25. Phenotypic correlations were generally low to high and positive in the guinea

savannah ecotype. Phenotypic correlations between body weight at day old and body weight

at other ages were generally low. A similar trend was observed for weight in week one to

other ages. Higher, positive and significant (P<0.001) phenotypic correlations were observed

for body weight at week three with weights at older ages. A similar pattern was observed in

the rainforest ecotype, except that the phenotypic correlation of body weight at week one

with older ages were low to moderate in magnitude.

4.3.5 Genetic and phenotypic correlations between body weight, body weight gains and

average daily gains at specific age.

Tables 26 present the genetic and phenotypic correlations between body weight,

body weight gains and average daily gains for the guinea savannah and the rainforest

ecotypes of Nigerian indigenous Muscovy duck. In the guinea savannah ecotype genetic

correlations between body weight and body weight gain ranged from rG= -0.063 at 5 weeks to

1.743 at the 10 week. Phenotypic correlation ranged between rp = 0.071 to 0.571 for body

weight and body weight gain. Higher correlations (both genetic and phenotypic) were

recorded between body weight, body weight gain and average daily gain at the 15 weeks of

ages in the guinea savannah ecotype. At weeks 5, 10 and 20 both genetic and phenotypic

correlations (rG and rp) were generally low to moderate and all positive except for genetic

correlation of body weight and body weight gin at 5 weeks, some were also significant


In the rainforest ecotype genetic correlation (r G) between body weight and body

weight gain ranged from 0.145 at 10 week to - 1.122 at 5 weeks. In this genetic group

genetic correlations (rG) between body weight and body weight gain were generally negative

except at 10 weeks of age. Negative genetic correlation coefficients were also recorded

between body weight and average daily gain at 10 and 15 weeks of age, respectively and the

same trend was observed between body weight gain and average daily gain at 20 weeks of


Phenotypic correlations (rp) between the traits in this genetic group were all positive

while some were significant (P<0.001). Phenotypic correlation between body weight gain and

average daily gain in all ages were high and significant 0.805, 0.869, 0.933 and 0.597 at 5,

10, 15 and 20 weeks , respectively (P<0.001).




Age in weeks Guinea savanna (h2) Rainforest (h2)

0 –5 0.01 ± 0.16 0.11 ± 0.16

5 – 10 0.19 ± 0.18 0.35 ± 0.30
10 – 15 0.09 ± 0.12 0.16 ± 0.12
15 – 20 0.45 ± 0.33 1.23 ± 043



Weight in weeks Guinea( rp) Rainforest (rp)

Wto x Wt1 0.559*** 0.141**
Wto x Wt3 0.170 0.131***
Wto x Wt5 0.273 *** 0.300
Wto x Wt10 0.201 ** 0.058
Wto x Wt15 0.196 ** 0.036
Wto x Wt20 0.201 *** 0.068
Wt1 x Wt3 0.280 ** 0.407**
Wt1 x Wt5 0.301 *** 0.401***
Wt1 x Wt10 0.218 ** 0.337***
Wt1 x Wt15 0.306 *** 0.266
Wt1 x Wt20 0.268 0.278
Wt3 x Wt5 0.737 ** 0.822***
Wt3 x Wt10 0.792 *** 0.831***
Wt3 x Wt15 0.880 *** 0.838***
Wt3 x Wt20 0.868 *** 0.844***
Wt5 x Wt10 0.680 *** 0.897***
Wt5 x Wt15 0.689 *** 0.847***
Wt5 x Wt20 0.707 *** 0.853***
Wt10 x Wt15 0.802 *** 0.906***
Wt10 x Wt20 0.809 *** 0.898***
Wt15 x Wt20 0.915 *** 0.941**
BWT=body weight, BWG= body weight gain, ADG= average daily gain



Age(weeks) Parameter Guinea Rainforest

rG rP rG rp

5 BWT x BWG -0.630 0.141 -1.122 0.410

BWT x ADG 0.326 0.113 0.541 0.040

BWG X ADG 0.114 0.931 0.480 0.805

10 BWT x BWG -1.243 0.246 0.245 0.474

BWT x ADG 0.344 1.242 -0.316 0.404

BWG X ADG 0.244 1.857 1.116 0.869

15 BWT x BWG 1.991 0.571 -0.243 0.549

BWT x ADG 0.698 0.516 -1.321 0.514

BWG X ADG 0.783 0.913 0.243 0.933

20 BWT x BWG 0.382 0.071 -0.721 0.447

BWT x ADG 0.467 0.081 1.160 0.204

BWG X ADG 1.029 0.905 -1.116 0.597

BWT= body weight, BWG= body weight gain, ADG= average daily gain

4.4 Egg Production

4.4.1 Body weight at first egg, average age at first egg, number of eggs laid at first cycle

and egg mass

Appendix 7 shows mean squares for body weight at first egg, average age at first

egg, number of egg laid at first egg, average egg weight, and egg mass of the two ecotypes.

The least squares means and standard errors of all the traits (body weight at first egg, age at

first egg number of egg laid at first egg and egg mass) are presented in Table 27. Ecotype

significantly (P<0.01) affected body weight at first egg and age at first egg, with 1496.88±

16.76 gram and 1579.47±17.57g as body weight of the ducks at first lay for the guinea and

rainforest ecotypes respectively. Ages at first egg (in days) were 314.76±3.33 and

331.70±3.49 days for the guinea and rainforest ecotypes, respectively with the rainforest

ecotype taking significantly (P<0.01) more days to attain sexual maturity.

Values for egg mass for the two ecotypes were 317.44± 17.92 and 292.14± 18.80g for the

guinea and rainforest ecotypes, respectively with no significant difference between the



Trait Ecotype Mean± SE

Body weight at first egg (g) Guinea (56) 1496.88±16.76b

Rainforest (54) 1579.47±17.57a
Age at first egg (days) Guinea (56) 314.70±3.33b
Rainforest (54) 331.70±3.49a
Av. No of egg laid at first egg Guinea (142) 6.30±0.36a
Rainforest (102) 5.60±0.38b
Egg mass (g) Guinea (142) 317.44±17.92a
Rainforest (102) 292.14±18.80b

means in the same column having different superscripts are significant different


( )= number of observations

Appendix 8 and Table 28 show the mean squares and the least squares means and standard

errors of egg characteristics (weight of first egg, length, and width of the first egg) and

ecotype significantly (P<0.05) affected width , (P<0.001) egg length (P<0.01) and egg

weight. The mean egg length of the two ecotypes were 5.307± 0.02 and 5.44± 0.02 cm for

the guinea and rainforest ecotype, with the rainforest having longer egg, while the width

means were 4.116± 0.01 and 4.157± 0.01 cm for guinea and rainforest ecotypes, respectively.



Traits Ecotype Mean ±SE

WFE (g) Guinea (142) 50.70±0.88b

Rainforest (102) 52.99±0.92a

LFE (cm) Guinea (142) 5.307±0.018b

Rainforest (102) 5.442. ±0.020a

WDFE (cm) Guinea (142) 4.116±0.011b

Rainforest (102) 4.157±0.012a

means in the same column within the same trait subgroup having different superscripts are
significant different (P<0.05)

WFE = Average weight of egg at first lay

LFE = Average length of egg at first lay

WDFE = Average width of egg at first lay

g=gram, cm= centimetre

( )= number of observations

4.4.2 Second Laying Performance

Mean squares of the second laying cycle characteristics of the two ecotypes are

presented in Appendix 9. Ecotype significant (P<0.001) affected number of eggs laid, egg

mass, pause length and average egg weight, while there was no significant (P>0.05) effect of

ecotype on average egg length and egg width. Table 29 presents least squares means of the

second laying characteristics, (number of eggs laid, egg mass, pause length, average egg

weight, average egg length, and average egg width ). The number of eggs laid at the second

laying cycle were 11.05± 0.55 and 8.27±0.57 for the guinea and rainforest ecotypes,

respectively. Higher (P<0.001) egg mass and longer pause length were recorded for the

guinea savannah ecotype compared with the rainforest ecotype such as 691.27± 37.27

compared with 496.14 ± 38.14 for egg mass and 78.18± 4.32 days compared with 62.28±

3.16 days for pause length.




Traits Ecotype Mean ±SE

Number of eggs laid Guinea (380) 11.05±0.55a

Rainforest (260) 8.27±0.57b

Egg Mass (g) Guinea (380) 691.27±37.27a

Rainforest (260) 496.14±38.14b

Pause length (days) Guinea (56) 78.18±4.32a

Rainforest (52) 62.28±3.16b

Av. Egg weight (g) Guinea (380) 62.41±1.58a

Rainforest (260) 60.86±1.61b

Av. Egg length (cm) Guinea (380) 6.31±0.18

Rainforest (260) 6.02±0.24

Av. Egg width (cm) Guinea (380) 4.57±0.02

Rainforest (260) 4.50±0.02

means in the same column within the same trait subgroup having different superscripts are
significant different (P<0.05)

g= gram, cm =centimetre

( )= number of observations

4.4.3 Percent egg Production/ laying intensity

Table 30 presents the duck- day and laying intensity or duck- day percent for the

guinea savannah and rainforest ecotypes for the annual laying season of the Muscovy duck.

The duck- day and laying intensity were calculated on monthly basis. The duck- day ranged

from 0.8 to 4.04 for the guinea savannah ecotype with an overall mean of 2.25, while for the

rainforest ecotype it ranged from 0.45 to 2.86 with an overall mean of 1.90. There was a

significant (P<0.05) difference between the overall mean of the guinea savannah and the

rainforest ecotype in favour of the guinea savannah ecotype. Highest duck day production for

the guinea savannah ecotype was at the month of August (4.04) while April had the least

(0.8) For the rainforest ecotype similar trend was also recorded with highest in August

(2.86) and least in the month of April.

The duck- day percent production or laying intensity for the two ecotypes

presented in Table 40 showed that for the guinea savannah ecotype laying intensity ranged

from 2.67 to 13.02 with the mean percent production of 7.68. Peak production of this ecotype

was in August with the value of 13.02%. For the rainforest ecotype the percent production

or laying intensity ranged from 3.33% in the month of April to 9.23% in the month of

August, with August recording the peak production for the ecotype, and the mean value for

this ecotype was 6.48%.



Month of lay Guinea Savannah Rainforest ecotype


No of Ducks No.of DD LI No. of No of DD LI

eggs Ducks eggs

April 60 48 0.8 2.67 58 26 0.45 3.33

May 58 78 1.34 4.34 58 60 1.03 3.34

June 56 146 2.61 8.69 54 120 2.22 7.41

July 56 142 2.54 8.18 54 111 2.06 6.63

August 52 210 4.04 13.02 50 143 2.86 9.23

Sept. 52 186 3.58 11.92 50 124 2.48 8.27

Oct. 48 74 1.54 4.97 44 98 2.23 7.18

54.57 884 2.25 7.68 52.57 682 1.90 6.48

DD = duck day

LI = laying intensity

4.5 Genetic and phenotypic parameters of egg production traits:

4.5.1 Heritability estimate of laying characteristic and external egg traits of first laying


Heritability estimates from sire variance components of body weight at first egg,

age at first egg and the external egg characteristics of first laying cycle of the guinea

savannah and rainforest ecotypes of the indigenous Muscovy duck are presented in Table 31.

The estimates were 0.13 ± 0.41, 1.11 ± 0.05, 0.80 ± 0.32, 0.55 ± 0.30 and 0.65 ± 0.33 for

body weight at first egg, age at first egg, egg weight, length and width, respectively for the

guinea savannah ecotype. The corresponding values for the rainforest ecotype were 0.83 ±

0.62, 1.54 ± 0.74, 1.79 ± 0.74, 1.79 ± 0.51, 1.00 ± 0.43 and 1.05 ± 0.51. The estimates of the

rainforest ecotype were high with some above the normal parametric range.

4.5.2 Genetic and phenotypic correlations of external egg quality traits of the first

laying cycle for the two genetic groups.

Table 32 presents the genetic and phenotypic correlations between the external egg

traits of the two genetic groups in the first laying cycle. The genetic correlations between egg

length and width had the highest value of 0.785 and significant (P<0.001) while the least was

between weight and length (0.131) in the guinea savannah ecotype. Similarly, moderate

estimates were recorded for the phenotypic correlations between all the traits in the same


In the rainforest ecotype, genetic and phenotypic correlations were generally high

and significant (P<0.001). Genetic correlation coefficients were 0.831, 0.802 and 0.886 while

the phenotypic correlation coefficients were 0.675, 0.753 and 0.642 for weight and length,

weight and width and length and width combination, respectively.




Parameter Guinea (h2) Rainforest (h2)

Body weight at first egg 0.80 ± 0.41 0.83 ± 0.62

Age at first egg 1.11 ± 0.05 1.54 ± 0.74

Egg weight 0.80 ± 0.33 1.29 ± 0.51

Egg length 0.55 ± 0.30 1.00 ± 0.43

Egg width 0.65 ± 0.33 1.05 ± 0.51




Parameters Guinea Rainforest

rG rp rG rp

WFE x LFE 0.131 0.497*** 0.831*** 0.625***

WFE x WDFE 0.225** 0.462*** 0.802*** 0.753***

LFE x WDFE 0.785*** 0.498*** 0.886*** 0.642***

WFE=weight of first egg, LFE= length of first egg, WDFE= width of first egg

rG=genetic correlation , rp=phenotypic correlation


4.5.3 Heritability estimate of second laying characteristics and external egg traits of the

two ecotypes:

Table 33 shows the heritability estimates for the second laying charateristics and

external egg traits of the guinea and rainforest ecotypes. The (h2) estimate for number of eggs

laid, egg mass, pause length, egg weight, egg length and egg width of the guinea savannah

ecotype were 0.30 ± 0.56, 0.45 ± 0.61, 0.32 ± 0.57, 0.09 ± 0.12, 0.19 ± 0.17 and 0.16 ± 0.15,

respectively. While those of the rainforest ecotype were 1.23 ± 0.52, 0.97 ± 0.98, 0.61 ± 0.86,

0.87 ± 0.49, 1.51 ± 0.64 and 0.86 ± 0.48, respectively. The estimate for the rainforest

ecotypes was generally higher, though the values for the number of eggs laid, and egg length

were outside parametric ranges.




Parameter Guinea Rainforest

(h2) (h2)

Number of eggs laid 0.30 ± 0.56 1.23 ± 0.52

Age mass 1.49 ± 0.61 0.97 ± 0.98

Pause length 0.32 ± 0.57 0.61 ± 0.86

Egg weight 0.09 ± 0.12 0.87 ± 0.49

Egg length 0.19 ± 0.17 1.56 ± 0.64

Egg width 0.16 ± 0.15 0.82 ± 0.48


4.5.4 Genetic and phenotypic correlations of the second laying traits:

Genetic and phenotypic correlations between second laying traits of guinea

savannah and rainforest ecotypes are presented in Tables 34 and 35, respectively. In the

guinea savannah ecotype genetic correlations were all high between the traits, while the

phenotypic correlations were low between number of eggs laid and pause length. Higher and

significant (P<0.001) phenotypic correlation was recorded between number of eggs laid and

egg mass. For the rainforest ecotype, genetic correlations were low (0.241 and 0.102) and

negative between pause length and number of egg laid (-362). The phenotypic correlation

followed, the same pattern as the guinea savannah ecotype, which was low except for the

relationship between egg mass and number of eggs laid (0.845) which was high and

significant (P<0.001).




NEL 0.870*** 0.296

EGM 0.634*** 0.258

PL 1.236 0.884***

NEL= number of egg laid, EGM= egg mass, PL= pause length , ***(P<0.001)




NEL 0.845*** 0.103

EGM 0.241*** 0.207

PL -0.362*** 0.102

NEL= number of egg laid, EGM= egg mass, PL= pause length ***(P<0.001)

4.6 Molecular Characterization of the Muscovy duck ecotypes

To ensure that the amplified DNA bands originated from genomic DNA and not primer

artifacts, negative control was carried out for each primer /ecotype combination. No

amplification was detected in the control reaction. All amplification products were

reproducible when reactions were repeated using the same reaction condition. Five of the

primers (71.4%) were successfully amplified, polymorphic bands among the two ecotypes as

shown in Table 36 and Fig 1. The total number of amplified bands were 59 and 54 for the

guinea and rainforest ecotypes, respectively. There were more number of bands (polymorphic

and monomorphic) in the guinea savannah than the rainforest ecotype. The primer oligo 1

gave the highest bands in both ecotypes. Table 37 gave the genetic variability between the

two ecotypes, and the similarity between the two ecotypes was 0.86 while the genetic

distance between the populations was 0.14



Primer Guinea savannah Rainforest


Oligo 1 15 5 10 14 4 10

Oligo 2 9 3 6 9 3 6

Oligo 4 13 4 9 13 4 9

Oligo 6 10 3 7 8 3 5

Oligo 7 12 4 8 10 2 8

Total 59 19 40 54 16 38

NAB=number of amplified bands, NPB = number of polymorphic bands, NMB = number of

monomorphic bands, RAPD = Random amplified polymorphic DNA



Guinea savannah RainForest

Guinea Savannah 0 0.861

RainForest 0.142 0

1= Genetic similarity

2=Genetic distance




5.1 Morphological Differentiation Between the Rainforest and Guinea Savannah

Muscovy ducks.

Ecological subdivision is said to allow the establishment of population traits

representing suitable model for biogeography and genetic studies. Thus morphological

studies can produce valuable information about phenotypic plasticity of a species and

possible effect of genetic changes on morphological variation (Hauser et al., 1995). The

variation in morphological traits between the two ecotypes of indigenous Muscovy ducks in

body sizes (body width, bill height, body height, head length, neck length, head width and

wing length) is similar to the findings of Scott and Reynolds (1984) in Mexican ducks.

Though scientific reports on indigenous domestic Muscovy duck in Nigeria are lacking to

contrast head to head with this findings. Most investigations were carried out on chickens.

This variation agreed with the submission of Olori (1992) and Horst (1999), who reported

that, in poultry there are marked differences in body morphology between northern and

southern ecotypes. The apparent differences in some morphological traits between the two

ecotypes might be an adaptation. The longer body diameter, shank length, body height, neck

length and wing length in favour of the guinea savannah ecotype might be adaptive features

to the ecosystem.

There was high level of sexual dimorphism when the body measurements of

Muscovy duck for the two ecotypes were analyzed for the effect of sex, ecotype and their

interaction. Sex had significant (P<0. 001) effect on all the traits except for bill height with

male significantly having larger sizes and heavier body weight compared with the female in

the two ecotypes. These findings are in agreement with the report of Tai and Rouvier (1998),

Beaza et al. (2001) and Teguia et al. (2007) on western and African Muscovy duck. One

possible explanation for the appearances of extreme sexual size dimorphism in Muscovy

ducks is that of breeding strong female selection for high quality male or competition among

males for quality female, leading to fixation of larger body sizes and the secondary characters

(Mc Crakens et al., 2000). Badyaev et al. (2001) concluded that major causes of dimorphism

in birds are the strength of selection in different morphological traits and vary between sexes

and that selection always acts more on males than females. This sexual dimorphism is

attributable to the usual action between sex differential hormonal action( Baeza et al., 2001)

which invariably leads to differentiate the growth rate.

The adult body weight obtained in the present study is higher than what Hassan

and Mohammed (2003) reported for adult muscovy ducks from northern Nigeria. Salicon

(1991) and Nickolova (2004) however reported higher values of body weight (4.50 - 5kg)

for males and (2.5 - 3kg) for females.

5.1.1 Principal component analysis

The principal component analysis allowed better understanding of the complex

correlation among the traits and reduced the number of traits studied in the ducks, using only

the first four principal components, without loss of information. Pinto et al. (2006) reported

that principal components detect outlying specimens or measurements but do not

discriminate well between similar populations and with intense sexual dimorphism between

males and females that informed application of the principal component analysis on sexes


The first four principal components (PC) explained together a high percentage of the total

variance in both male and female traits. Therefore, they are interesting for the purpose of

evaluation and comparism of animals. In the male the first component accounted for

30.458% of the total variance and loaded for body width, bill width, shank length, head

length and neck length while in the female the first component accounted for 23.236% of

the total variance and loaded for shank length, body height, neck length and wing length. In

both sexes the first principal component represents the size component of the Muscovy duck.

The results are similar to the findings of Pinto et al. (2006) for chicken , Hammock and

Shrode (1986) for yearling performance in beef cattle and Fumio et al.( 1982) in Japanese

black cattle sire. The four factors (PC) in both males and females could be used to select the

ducks based on a group variable rather than isolated traits. This is accentuated by the findings

of Atiyat(2009), who predicted the effect of the breeding programme using a reduced data set

on morphological traits that are sensitive to correlated response to selection. The principal

components can also be used in development of selection index to simplify them, because

such an index would have few PCs in the place of the original traits ( Ibe, 1989).

The total variance obtained from the four principal components that were greater

than I used in both male and female were 66.839% and 56.404%, respectively as shown in

Table 8 and 10. However, lower than that had been reported by McCraken (2008) for musk

duck. Variation in variance percentage, with higher value for male is in agreement with the

report of McCraken et al.(2000). The reasons for these differences were related to sexual

dimorphism in favour of the male. The major under-lying factor responsible for the observed

cluster in component formation for both sexes may be related to the differences in association

of each measurement with bone, environmental component or time taken to reach maturity,

which is normally time dependent (Salako, 2006).


5.1.2 Discriminant analysis

The use of discriminant analysis in this study showed that the populations were characterized

as not clear distinct clusters. The performance of indigenous Muscovy duck population in

Nigeria is usually low to moderate and often maintained in small populations with no

selection. Though it is very rare to find comparative study stated on basis of morphological or

phenotypic traits as reported here, the results of the multivariate analysis of morphmetric

variables show some levels of differences between the two ecotypes. Similarly, the pair-wise

squared mahalanobis distance and the probability of a significant effect of contrast by the F-

test (P<0.001) between the populations show such diversity between the ecotypes (2.963).

The dissimilarity, according to the squared Mahalanobis distance obtained in this study is

however lower than what Atiyat (2009) reported between layer, broiler and indigenous

chicken populations in Jordan (433.88, 429.87 and 38.31). The low morphological distance

in this study may result from non-selection in the two populations and continuous inbreeding

characteristic of free range management system and possible exchange of genes between the

populations through transportation of the birds from the guinea savannah region to the

rainforest south and vice versa.

Although the univariate analysis revealed differences in body weight and linear type traits of

the two ecotypes of Muscovy duck, the multivariate analysis provided better resolutions as

shown in Table 11, thereby limiting the number of variables contributing to ecological

differentiation. Only a single standardized canonical discriminant function was extracted

similar to the finding of Yakubu et al. (2009) on sex differentiation of African Muscovy

duck. The discriminatory power of body weight and wing length is consistent with the report

of Zenatello and Kiss (2005), where wing length was observed as the most discriminating in

Rose coloured Starlings (Sturnus reseus). Similarly, Yakubu (2009) reported wing length as

having the highest discriminating function in sex differentiation in African Muscovy duck.

The seven variables therefore are sufficiently robust enough to be used in determining the

ecological identification of Nigerian indigenous Muscovy duck. This is an indication that

morphological measurement could be taken into consideration in a selection program toward

improvement of the indigenous Muscovy duck.

5.2 Growth Traits

5.2.1 Body weight: The performance of body weight of the Muscovy duck ecotypes is

similar to those reported by Banga Mboko et al. (2007), Teguia et al. (2007) in African

Muscovy duck, Ola (2000) and Etuk et al. (2006) for indigenous muscovy duck of Nigeria.

The body performance of the guinea savannah ecotype at the early stage is slightly higher

than the values obtained by Teguia et al. (2007). At day-old, a significant difference

(P<0.05) between ecotypes with respect to body weight was detected, with the guinea

savannah ecotype demonstrating heavier body weight than the rainforest ecotype. No

significant difference was detected between sexes for body weight at day-old as reported by

Teguia et al. (2007) in Cameron, Bhuiyan et al. (2005) in Bangladesh and Leclercq (1990) in

France. However, ducklings from Bangladesh and France were heavier than those from

Nigeria, (44g in Bangladesh and 50 -55g in France). Body weights of duckings at day-old

are similar to that reported by Teguia et al. (2007) in Cameroon, Ola (2000) in southern

Nigeria. These differences are probably related to the fact that the birds used in the advanced

countries are of different genetic background and have undergone selection thus, the higher

genetic potential of Muscovy ducks studied by these authors. At three (3) weeks of age

sexual dimorphism appeared in favour of the male being heavier and accentuated with age.

This agreed with the reports of Sauveur (1985), Sauveur and De Carville (1990), Garzilov

(1991), Beaza et al. (2001) and Teguia (2007). However, Holderread (1978) earlier reported

that sexual dimorphism commences at 10 days of age.

Steven (1990), in his weight improvement study of the Muscovy duck reported

that the body weight of Muscovy duck normally increased at about 60g yearly if one is

selecting for improved body weight. Most reports (Cavalchini et al., 1979; Pilla and Quilici

,1993; Baeza, 1998; Ngapongora et al., 2001) pertaining to growth of Muscovy duck

indicated higher body weight at all ages than the present findings. The reason might be due to

the fact that Muscovy duck of most countries like France, and other advanced countries are

managed under intensive system of rearing with high stocking rate, balanced feeding and

management and also have undergone selection for improved body weight.

5.2.2 Body weight gain and average daily gain:

Body weight gain and average daily gain of the two ecotypes (guinea and rainforest)

showed similar patterns of increment with increase in age from day-old to the 10th week,

with a decline from then to the 20th week of age as shown in Table 25. Variation occurred in

sexes in both body weight gain and average daily gain as shown in Table 27, and

significantly in favour of the male in the two ecotypes. In males the highest body weight gain

was recorded between 10-15 weeks of age and gradually declined with age while in females

the highest body weight gain was at 0-5 weeks.A similar pattern was recorded for average

daily gain. This pattern is similar to the finding of Banga-Mboko et al. (2007), who recorded

maximum daily weight gain in female Muscovy ducks at 5 weeks whereas in male it was at

four weeks. The values obtained in this study are slightly higher than what Ola (2000) earlier

reported on body weight gain at 5 and 12 weeks which were (13.35±2.3 and 15.54±1.38g for

males and 10.21±1.25 and 10.07±0.63g for the females). The variation in weight gain and

average daily gain between the ecotypes used in this study may have to do with inheritance

as the trait can be influenced by both genetic and non-genetic factors. A similar observation

was reported by some authors such as Ola (2000), Yue (2006), Etuk et al .(2006) and

Banga-Mboko et al. (2007) for both European and Africa Muscovy ducks. This supports the

earlier sexual dimorphism displayed by the male by attaining higher body weight gain and

average daily gain than the female to support muscle deposited in the male.

5.3 Parameter Estimates for Growth Traits

5.3.1 Heritability of growth traits:

Body weight: Heritability of day-old body weight of the rainforest ecotype is

similar to the finding of Mignon-Grasteau et al. (1999), who reported h2 as 0.40. At 10

weeks of age the estimate of 0.25±0.22 is similar to what Ricard et al. (1989) and Hu et al.

(2006) obtained in Muscovy duck, also from sire variance components. Higher values were

however obtained by Pingel (1990) at the same age from sire and dam variance components.

In the two ecotypes heritability tends to increase with age. A similar observation was earlier

reported by Mignon Grasteau et al. (1998) for Muscovy duck, Chapius et al. (1996) reported

for turkey, Chamber (1990) and Adeyinka (2006) for chickens.

There was marked variation in the heritability estimate in the two populations with

the guinea savannah ecotype generally having higher values. Mignon–Grasteau (1998)

interpreted such scenario to be due to the fact that heritability is only a measure of the current

variation in each population thus providing information on part history of either population.

Similarly, Poujardieu et al. (1994) compiled values of heritability of body weight at 53 days

up to 84 days of age ranging from 0.17 to 0.55 for male Muscovy ducks which were forced


The differences in the values obtained from this study with other workers reports could be

because the estimate here was obtained by using the variance component estimation with the

model sire and dam nested within sire. In this case the full relationship matrix between all

individuals was not taken into account and result may be biased, especially in selected

populations (Poujardieu et al., 1994). The lower value of heritability estimate obtained from

this study, compared to others in literature might agree with the proposition by Ricard et al.

(1983) that in subtropical climate due to variability of response to climatic effect a higher

residual variance could be expected and so lower values of the heritability. This lower

heritability implies that mass selection can be the likely options for such trait.

Diversification of heritability estimate of body weight was reported for layers and

broilers of other species of poultry by many authors (Danbaro et al., 1995; Le Bihan Duval et

al., 1997, Mignon Grasteau et al., 1998; Tufvensson et al., 1999). Therefore generalization of

result across population seems difficult. Since genetic variability of a given strain is

determined both by selection pressure and specific gene pools. There are several single genes

that affect the body weight of birds (Merat, 1990). From the economic perspective their

identification in particular population is not always well founded, On the other hand

segregation of genes with relative large effects can lead to biased estimate of genetic

parameter. From the physiological perspective body weight and its changes over time are

affected by many processes and determined by various effects of genes depending on the age

of the individual (Szwaczkowski et al., 2007).

5.3.2 Phenotypic correlation of body weight

There was similarity in magnitude and direction in the phenotypic correlation

between weights at various ages in both genetic groups. The estimate between weights at

each stage and other ages were low and increased as the birds got older. This similar trend

was reported by Hu et al. ( 2006) in Muscovy duck in Taiwan, where they obtained

phenotypic correlation between body weight at 10 weeks and at 18 weeks as (rp) 0.21.

Significant (P<0.001) positive phenotypic correlations were found between body weight at 3

weeks and week 5, 10, and 20 and similar trend at between weeks 5 and 10, 15, and 20. This

shows that selection for body weight at week 3 in both genetic groups will also lead to

increase in weight at older ages.

5.3.3 Heritability estimates of body weight gain and average daily gain.

The estimates of h2 of body weight gain for the two ecotypes tend to follow same

pattern as they increases with increase in age. The estimate of 0.32±0.4 for body weight

gain in the rainforest ecotype at 16 to 20 weeks is consistent with what Mignon-Grasteau et

al.(1998) obtained for Muscovy duck in France. Low to moderate values of heritability for

average daily gain were estimated for the two genetic groups. In the rainforest ecotype the

heritability estimate for average daily gain at 6-10 weeks (0.35±0.30) is greater than those for

all other ages. Therefore, the detection of genetic variability for average daily gain at 6-10

weeks in this genetic group seems to be more difficult compared to other ages.

5.3.4. Genetic and phenotypic correlation between body weight, body weight gain and

average daily gain:

Positive and significant (P<0.001) genetic and phenotypic correlations were

observed between body weight and average daily gain in the guinea savannah ecotype, and

this is similar to the finding of Mignon-Grasteau et al. (1998). The positive genetic

correlation between these traits is an indication that selection for body weight will equally

lead to increase in weight gain. On the other hand, some genetic correlation coefficients were

over-estimated; these unexpected estimates of genetic correlation between the studied traits

may be due to sampling errors and some missing observation (El-Deen et al., 2008).With

possibility of happening due to small data set.

5.4 Egg Production Traits

5.4.1 First cycle laying performance: Body weight at first egg: The mean body weight at first egg of the rainforest ecotype

(1496.88±16.76g) is higher (P<0.05) than that of the guinea savannah ecotype

1579.47±1757g. These values are in line with the report of Ola (2000) and Teguia et al.

(2007) for the African Muscovy duck, but lower than what Mahanta et al. (2001) obtained

for Indian Muscovy duck(2266g) and Nickolova (2004) for European Muscovy ducks

(2588.129). The differences between the African Muscovy duck and other regions of the

world in body weight at first egg might not be unconnected with the fact that African

Muscovy duck population has not been subjected to body weight selection leading to high

variability and low performance in weight. Age at first egg.

The ages at first egg for the two ecotypes were 314.7±3.33 days and

331.70±3.49 days for the guinea and rainforest ecotypes representing 44.9 weeks for the

guinea savannah ecotype and 47.3 weeks for the rainforest ecotype. From the results the

guinea savannah ecotype became matured or laid eggs earlier than the rainforest ecotype. The

values obtained from this study are higher than what Ola (2000) reported for Muscovy duck

in southern Nigeria (28 weeks) Duru et al. (2006)for Muscovy duck in northern Nigeria

(7.7±0.04 months). Similarly, lower values had been reported in other parts of the world for

Muscovy duck such as Kalita and Deka (2005), who reported 300-305 days in Asam India,

Panda and Kumar (2004) reported 24-25 weeks. Avanzi and Romboli (1979), Mahanta et al.

(2001) Islam et al . (2009) reported 285.52 days. Variation in age at sexual maturity noticed

in the present study might be due to the limited breeding capacity reported in Muscovy duck.

Ola (2000) reported that though Muscovy duck can attain maturity earlier they have limited

breeding capacity, especially during the dry season and may delay egg laying to the

beginning of rainy or wet season. Though Morris and Fox (1960) earlier reported that age

and weight at sexual maturity are influenced by genetic make-up of the individual and are

further influenced by feed intake, daily light duration of day length and other environmental

factors, this could be the reason for the variations observed. Egg number and egg mass: The average egg number of the two ecotypes laid at first

laying cycle were 6.30± 0.36 and 5.60±0.38 eggs with the guinea savannah ecotype showing

higher value of egg number. The value obtained from this study is lower than what Mopate

(1999) reported (14-15 eggs) Mouthe (1989) (13.01) in Niger Republic and Chad,

Mohammed (1996) and Banga-Mboko et al.(2007) in Mozambique and Congo Brazzaville

reported 14.6±3.0 eggs for the African duck. Similarly, Ola (2000) reported higher value for

Nigeria Muscovy duck in southern Nigeria. The values reported by these authors are higher

than the one from present study probably due to the fact that their values might be over-

estimations by the respondents giving higher values in survey as opposed to on-station

results. Another reason might be poor performance of the bird in confinement in this study as

most of the results obtained on African Muscovy ducks are under free-range management


Egg mass reported here for first laying cycle favoured the guinea savannah

ecotype which had 317.44±17.92g compared to the rainforest ecotype which had

292.14±18.80g. The variability in both egg number and egg mass among the ecotypes might

be a reflection of the genotype-environmental interaction between the ecotypes in favour of

the guinea savannah where the two populations are managed.

5.4.2 Performance at second laying cycle:

The egg production performance of the two ecotypes at the second laying cycle

with appreciable variation between ecotypes represents mature laying performance of the

Muscovy duck from the two ecotypes. The performance in the second cycle is close to what

Mopate et al. (1999) reported for local ducks in N’djamena in Niger Republic, Mourthe

(1998) in Chad, Banga-Mboko et al. (2007) from Congo Brazzaville, Mohammed (1996)

from Mozambique and Ola (2000), Etuk and Abasiekong (2005) and Duru et al. (2006) in

Nigeria. There is a significant difference in egg production between the African Muscovy

ducks and Muscovy ducks from other parts of the world, though Clayton (1974) claims that

much of the high egg production of domesticated ducks is an expression of natural fecundity,

given appropriate opportunity and owes little to artificial selection, and the interplay between

inheritance and environmental factor might be responsible for the low production of African

Muscovy duck.

According to Sauveur and De Carville (1985) the first and the second

reproductive periods continued from three to five months and, the pause between them is

about six months. The pause length obtained from this study is shorter and similar to that

reported by Sauveur and De Carville (1985). Though there is a difference in pause length

between the two ecotypes (78.18±4.32 days and 62.28±3.16 days for guinea savannah and

rainforest respectively), this variation could be due to physiological response.


5.4.3 Laying Intensity:

The highest peak in egg production of the two ecotypes of 13.02% for the

guinea savannah and 9.33% for the rainforest is similar to what Andrew et al. (1984)

obtained for indigenous ducks in India reared under semi intensive system. The values

obtained from this study are, however lower than what Sauveur and De Carville (1985;1990)

reported for Muscovy ducks in France under intensive management systems, and Tian-Fwu

et al. (1998), Jayart et al. (2009) and Ravi et al. (2009) reported for indigenous ducks of

Kerala, India. The smaller peak percent egg production or laying intensity in the two ecotypes

in this study might be due to the non-selection and inbreeding depression characteristic of

indigenous stocks.

In the present study two peaks of lay were obtained for the guinea and rainforest

ecotype, similar to the report of Sauveur and De Carville (1985) in France. This might be a

breed characteristic of the Muscovy duck. The months of peak lay in the two ecotypes,

though similar (June to September) and coincide with when the sun is at the equator and

Northern hemisphere, differ from what Nickolova (2004) obtained, his peak periods were in

the month of April to July, which coincide with spring and summer in the northern

hemisphere. The differences in the months of peak between the present study and that of

Nickolova (2004) might have to do with the weather difference in the two countries, and the

fact that the ducks are sensitive to light and weather change, and laying is season-dependent.

5.4.4 Genetic and Phenotypic parameters of Laying traits: Heritability of laying performance and external egg traits:

The low and high heritabilities of body weight at first egg reported here for the

guinea savannah and rainforest ecotypes have been similarly reported by Tai et al. (1989) in

Taiwan. This implies that body weight could easily be improved by predicting breeding

values and selecting on a within-line basis in the rainforest ecotype with high results while in

the guinea savannah ecotype with low estimates, selection for body weight at first egg cannot

be carried out using individual selection procedure but mass selection. The estimate for age at

first egg for both genetic groups was both high and though outside the theoretical range of

parameter estimation, was higher than what Hu et al. (2004) 0.20, Cheng et al. (1995). 20

and, Ricard et al. ( 1983) 0.16 obtained from Muscovy duck strain and brown Tsaya duck at

first laying cycle.

The estimates for egg weight, egg length and width in the first laying cycle in both

genetic groups were moderate though in the rainforest ecotype the values are outside normal

parameter range. The values for egg weight are consistent with the findings of Stasko (1966),

Pingel (1999), and Besbes et al. (1992) for the domestic duck. Estimate for heritability of the

second laying characteristics are lower than the estimates of the first laying cycle, and this is

in agreement with the finding of Sabri et al. (1991), who reported that estimates are always

higher when measured early compared with later in laying year. Hagger and Abplanalp

(1978) and Sabri et al. (1991) found the same trend when they estimated heritability of

residual feed consumption, which is a measure of feed efficiency for egg production. They

suggested that the genetic potentials of egg production, and related traits are well expressed at

peak egg production which minimize effect of environment factors compared to later times

of the laying year. Higher heritabilities at early stage of the laying year are advantageous

because early selection decision can be made. Parameter estimates in the second laying cycle:

Heritabilities of all the traits (number of egg laid, egg mass, pause length, egg weight,

egg length, and egg width) were low to moderate in the guinea savannah ecotype while in the

rainforest ecotype all the estimates were high. The low heritability estimates obtained for the

guinea savannah ecotype show that environmental factors have more effect on the second

laying performance of this genetic group than the additive genetic mak-eup. The estimate of

heritability of 0.30± 0.06 for number of egg laid in the second cycle for guinea savannah

ecotype is consistent with those of Tai et al. (1989), indicating that the additive genetic

variation is low for number of eggs laid at that age. Tai et al. (1989) explained that it is a

component of maternal inheritance and higher estimate can only be obtained using dam

variance component. The standard errors of the estimate in both genetic groups were high and

this question the reliability of the estimates.

Positive and significant genetic and phenotypic correlations were observed

between egg weight, egg length, and egg width in both genetic groups in the second laying

cycle. This similar finding was reported by Zchang et al. (2005). Although the relationships

were low in the guinea savannah and high in the rainforest population, they indicate that in

guinea savannah selection for egg weight may not necessarily translate to longer and wider

egg while in the rainforest the high genetic correlation is an indication that increment in any

of the external traits would translate to improvement in all others. Genotypic correlations

were higher than phenotypic correlations in the rainforest population. These are similar to

the findings of Brah et al. (1992) and Saatci et al. (2003). This shows the degree of genetic

influence in the relationship between the traits compared with environmental influence.

Significant and positive genetic correlations were recorded between number of eggs laid, egg

mass and pause length. Though genetic correlation between pause length number of eggs

was negative and outside parameter estimate range, similar finding was reported by Akbas


5.5 Molecular evaluation:

The similarity or band sharing between ecotypes obtained in this study was 0.86 is similar to

the findings of El-Gendy et al. (2005), who reported 0.83 for Muscovy duck populations in

Egypt. These values are however higher than what was obtained for other breeds like White

Pekin, Khaki Campbell and Dametti ducks with 0.68, 0.74 and 0.68. respectively as reported

by same author in Egypt. This is an indication of closeness of the populations.

Genetic distance between the ecotypes was as low as 0.14. This aids in understanding the

genetic relationship and divergence between the ecotypes. The value obtained here is

however lower than what El-Gendy et al.(2005) obtained for cairina species (0.40) when

they used five primers. Similarly, higher values were obtained by Xiao et al. (2005) for

Fugian local duck populations in different ecological zones (67.97%). The higher similarity

obtained in this study for the ecotypes of Nigerian indigenous Muscovy duck supports the

low distance between them, suggesting possibility of little genetic divergence between the

populations, thus indicating that they have common ancestors and have little genetic

differentiation despite ecological isolation.



Indigenous Muscovy duck (Cairina moschata) accounted for 20% of the total

poutry population in Nigeria and can provide opportunity for increasing protein supply and

income for smallholders, because they require low capital investment, have short generation

interval and a high rate of productivity. They also play a complementary role in relation to

other poultry like chickens particularly in rural settings. Characterization, utilization and

conservation of these poultry genetic resources is highly important.

In the guinea savannah and rainforest agro-ecological zones of Nigeria indigenous Muscovy

ducks are managed under village-based extensive system with no cross breeding, because of

the uniqueness of the bird, thus under continuous inbreeding maintaining consistent breeding

value. There is little or no literature available on the breeding performance of these birds in

Nigeria, and this necessitates the bases for this study. Hence the present study was carried

out to identify, characterize and evaluate phenotypic variations of the two ecotypes

(rainforest and guinea savannah) of the indigenous Muscovy duck.

Further objective was the generation of preliminary data on performance characteristics

under improved management and genetic variation using random molecular technique.

6.1 Conclusion

From the results obtained in this study the following conclusions can be reached

6.1.1 Morphological differentiation

i Significant morphological variations exist between the rainforest and the

guinea savannah ecotype

ii The rainforest ecotype was found to be heavier than the guinea savannah though not


iii. There was marked sexual dimorphism between sexes in all the ecotypes.

iv. The principal component analysis of the morphostructural traits of the two sexes tend

to display variation in traits associated with possible differences in index

development as explained by the dimorphism that exists between the sexes.

v. Seven out of the 12 morphological traits had discriminatory power to distinguish

between the two ecotypes, providing bases for use of morphological traits in a selection

programme towards improvement of the two ecotypes.

vi. Mahalanobis D2 statistics clearly distinguish the two ecotype on the basis of

morphological traits, providing sufficient evidence for morphological differentiation between

the two ecotypes.

6.1.2 Productivity of the ecotypes

Performance characteristics of the two ecotypes reveals the following:

i Body weight was significantly higher in the guinea savannah ecotype between 0 to

10 weeks, but not significant at adult age.

ii Sex significantly affected body weight from 3 to 20 weeks of age in both ecotypes,

indicating sexual dimorphism in the bird.

6.1.3 Egg production characteristics

i There were significant differences between the guinea and the rainforest ecotypes with the

rainforest ecotype having heavier body weight at first egg than the guinea savannah ecotype.

ii. There were significant difference in egg production parameters at the first laying and the

second laying cycles. The first laying favoured the rainforest ecotype while the second

laying cycle favoured the guinea savannah. This was supported by the frequency of lay,

which was more in the guinea savannah ecotype.


6.2. Heritability estimates of growth traits (from 0 to 20 weeks)

i Heritabilities of growth traits in the two ecotypes did not differ significantly, and were

all low to moderate

ii Body weight gain in the two ecotypes had their heritability increasing with age, low to

high, while heritability for average daily gain were low to moderate in both cases.

iii The high estimates of heritability of the body weight at early age in the two ecotypes

indicate that selection for body weight can start as early as first week of life to result in

high body weight at later age; this will allow for early selection as those birds that show

good weight at early part of their life can be selected for increased body weight,

particularly if one needs to select for meat type, while selection for body weight at older

age will require indirect or mass selection.

The low heritability of body weight of the two genetic groups at older age indicates that they

are genetically-unimproved, showing environmental factors playing more important role in

influencing the growth performance at that age .

The high heritability of body weight gain and average daily gain at weeks 15 to 20 is an

indication that selection for feed conversion efficiently can start at this age.

b. Heritability estimates of egg production traits

Heritability estimates of egg weight, length and width at the first laying cycle were all high

indicating that selection for egg characteristic at early lay is possible.

Moderate to high estimates of heritability were recorded for number of eggs laid, egg mass,

and pause length in the second laying cycle, suggesting that selection for egg number, egg

mass and pause length may be effective when the birds are at the second laying cycle.

It is obvious that the heritabilities for egg traits are higher for the guinea savannah ecotype at

early stage suggesting selection for the traits in this ecotype is best at first laying cycle while

for the rainforest at second laying cycle.


6.2.2 Genetic and phenotypic correlations of growth traits (from 0 to 20 weeks)

In both genetic groups phenotypic correlation coefficient were high between weight at 3

weeks and subsequent ages. This high correlation shows that body weight at 3 weeks in both

populations is an indicator of weight at subsequent ages.

Positive and significant genetic and phenotypic correlations between body weight and body

weight gain, body weight and average daily gain in the guinea savannah ecotype show that

increase in weight gain and average daily gain will lead to increase in body weight for the


Genetic and phenotypic correlations of egg production traits

i The high, significant and positive genetic and phenotypic correlations between egg

weight, length and width in both ecotypes suggest that high genetic relationships

exist between the traits in both populations and correlated response will be obtained if

physical egg characteristics are to be considered in Muscovy duck.

ii Positive relationships (genetic and phenotypic) were found between number of eggs laid,

egg mass and pause length. This is understandable as they are dependent traits, and so

selection for egg number will automatically leads to increase in egg mass

6.3 Molecular characterization

This is the first time the indigenous Muscovy duck has been studied at molecular level. This

study provides important information for future conservation of the Nigerian indigenous

Muscovy duck resources. Therefore, it is a powerful tool for breeding improvement, because

it will help in the preservation of the local breed and control of breeding in future restocking


The random amplified polymorphic DNA used in this study was found to be useful and

informative for studying genetic diversity in the Nigerian Muscovy duck. The result reveals

that there is no significant genetic divergence between populations, which is an indication of

common origin and lack of genetic flexibility of the indigenous ducks in the study areas.

6.4 Recommendations

1 Because of significant differences in performances of the two ecotypes, the rainforest ecotype

can be selected for increased body weight as meat producing type, while the guinea savannah

ecotype can be selected for egg production as layer type.

2 This study happens to be a base study on characterization of indigenous Muscovy duck, the

sample size used and the population covered might not provide sufficient information on

the diversity that exists in the Nigerian indigenous Muscovy duck, and so further study is

suggested along this line for additional information.

3 On farm evaluation of indigenous genetic resource is recommended, so as to effectively

monitor and evaluate performance of birds, particularly those that remain consistently on

free-range since their domestication. This study provided a platform for that and continuous

approach in this direction will provide adequate information on productivity and potentials.

Since there are varied ecological zones in Nigeria and Muscovy ducks have been reported to

have the tendency to evolve adaptive measures to acclimatize with their environment, it is

recommended that Muscovy duck from each of the agro-ecological zone be collected and

studied for both quantitative, qualitative and molecular to further ascertain the diversity that

occurs in the population either at the phenotypic or at DNA level.

4 Superior molecular techniques such as RFLP, AFLP and Microsatelitte should be used for

further characterization of the indigenous Muscovy duck



Abreu, V.M.N., Silver, M. A., Cruz, C.D., Figueiredo, E. A.P. and Abreu, P.
G. (1999). Capacidale de combinacao de caracteristicas productivas de linhagen
dematrizes de frangos de corte. Usando atechnica de components princiais.
Rev.Bras. Zootec. 28:250-25

Adebambo, O., Williams, J.L., Blot, S. and Urquhart, B. (1998). Genetic variation in
Nigeria cattle breeds using 26 micro satellite markers .In: Proceedings of the
Silver Jubilee of Annual conference West Africa Society for Animal
Production. March 24-26pp.

Adedeji, T., Olufunmilayo, A. Adebambo, Ozoje, M.O., Morenike, A. and Dipeolu, A.

(2004). Early growth performance of crossbred chickens resulting from different sire
strains. Proceedings of the 29th Annual Conference of the Genet. Soc. of Nig.,2004,
UNAAB, Abeokuta, pp: 126-129.

Adedokun, S. A. and Sonaiya E. B. (2001). Comparism of the performance of

Nigeria Indigenous chicken from three agro-ecological zones of Nigeria. 13/2/aded132htm

Adeogun, I.O. and Adeoye, A.A. (2004). Heritabilities and phenotypic correlations of growth
performance traits in Japanese quails. Livestock Research for Rural
Dev. 16: 12-16.

Adeyinka, I. A., Oni, O. O., Nwagu, B. I. and Adeyinka, F. D. (2006). Genetic

parameters estimates of body weights of Naked neck broiler chickens . Intern. J.
Poult. Sc. 5(60 589-592.

Ahlawat, S.P.S., Sunder J., Kundu A., Chatterjee, R.N., Rai, R.B,. Kumar, B., Senani, S.,
Saha, S.K. and. Yadav. S.P. (2004). Use of RAPD-PCR for genetic analysis of
Nicobari fowl of Andamans. Br. Poult. Sci., 45: 1-7.

Ajala, M. K., Nwagu, B. I. and Otchere, E.O. (1988). Socio-economic of free range poultry
production among agro pastoral Fulani women in Kaduna state Nigeria. In:
Proceeding of the Silver Anniversary Conference of the NSAP and the Inaugural
Conference of the WASAP 21 -26 March 1998 Abeokuta, Nigeria.

Akbas, Y., Akiek, A., Onenc, A. and Gungor, M. (2006). Growth curve analysis for body
weight and dry matter intake in Friesian limousine x Friesian pietmontese x
Friesian cattle. Arch. Tiez Dummerstorf 49 (2006) 4; 329-339.

Akinokun, O. (1971). The problem of poultry production in Nigeria. Paper presented

at the 7th Annual Conference of Agricult. Soc. of Nig. Kano.pg35.

Al – Atiyat, R. (2009). Diversity of chicken Populations in Jordan determined using

discriminate analysis of performance traits. Int. Journ. of Agric. and Biology 1814 –

Alexender, V. B. (2005) Maternal inheritance and rapid evolution of sexual size

dimorphism. Passive effect or active strategies. The Amer. Naturalist 166:17-30

Ali, B.A., Ahmed, M.M.M. and Aly, O. M .(2003). Relationship between genetic
similarity and some Production traits in local chicken strains. Africa J. of Biotech. 2
(2) 46 -47

Ali, B.A. and Ahmed, M.M. (2001).Random Amplified Polymorphic DNA in some chicken strains.
First conf. Of Role of Biochemist in environmental and agriculture Cairo University Egypt

Anderson, L. (2001). Genetic section of phenotypic diversity in farm animals. Nature

Genetics 2:130-138

Andrew, C.V.,Venogopalan, C.K.,Unni, A.K.K. and Ramakrishnan, A.(1984). comparative

performance of desi duck reared under two system of housing.Kerala J.Vet

Arnold, S. J. (1985). Quantitative genetic models for sexual selection. Evolution


Arnold, T. W. , Rohwer, E .F. and Armstrong, M.(1987). Egg viability, nest prediction
and the adaptive significance of clutch size in praire ducks. Amer. Naturalist

Atiyat, R. (2009). Diversity of chicken Populations in Jordan determined using discriminate

Analysis of performance traits. Int. Journal of Agric. And Biology 1814 – 9896

Avanzi, C. F. and Romboli, I. (1979). Environmental factors and reproductive traits in

Muscovy duck. Aviculture 48(9) 41-46

Avens, J .S. , Athearn, R.M. , and McNeat, J. H. (1980). Egg production and efficiency
of food conversion of Khaki Campbell duck under different management
system. Br. Poultry Sc. 21(1980) 333-337.

Bacon, W.L., Nestor, K.E., Emmerson, D.A., Vasilatos-Younken, R., Long D.W. (1993).
Circulating IGF-I in plasma of growing male and female turkeys of medium and
heavy weight lines, Domest. Anim. Endocrinol. 10 267–277

Badyaev, A. V., Hill G. E. and Whittingham, L. A. ,(2001). The evolution of sexual size
dimorphism in House finch. In population divergence in ontogeny. Evolution

Baeza, E., De Carville, H., Salichon, M.R., Marche, G., Leclercq, B. (1997). Effects of
selection, over three and four generations, on meat yield and fatness in Muscovy
ducks. British Poult. Sc. 38 (4): 359 - 365.

Baeza, E., Salichon, M.R., Marche, G., Juin, H. (1998). Effect of sex on growth,

technological and organoleptic characteristics of the Muscovy duck breast muscle.

British Poult.Sc. 39: 398 - 403.

Baeza, E., Marche, G., Wacrenier, N. (1999). Effect of sex on muscular development
of Muscovy ducks. Reproduction, Nutrition, Dev. 39: 675-682

Baeza, E., Williams, J., Guemene, D. and Duclos, M. J.,(2001). Sexual dimorphism of
growth in Muscovy ducks and changes in insulin-like growth factor 1 (IGF-1) growth
hormone (GH) Reprod.Nutri.Dev. 41 :173-179.

Bagliacca, M., Pacci, G., Marzoni, M.and Avanzi, C.F. (1989 ). Technoglia di incubazione
nell anatra muschiata;effetto della posizone e del raffreddamento delleuora durante1
incubazione zoot.Nutri. Anim. 15;23-430.In Nicklova M studies on egg laying
characteristics of Muscovy duck dependeing on the breeding method 2004.
Balogun, O. Y. (2005). Senior secondary atlas .2nd edition. Longman.

Banerjee, C. C. (1991). A Textbook of Animal Husbandry 7th edition . Oxford IBH

Publishing co.pvt ltd New Delhi- Bombay Calcuta.

Banga- Mboko, H., Lelou, B., Maes, D. and Leroy, P.L. (2007). Indigenous Muscovy duck
in Congo Brazzaville. Preliminary observation on indigenous Muscovy duck reared
under moderate input in Congolese condition .Tropical Anim. Health and Prod. 39

Barbosa, L. (2005). Genetic divergence among five lines of laying hens .Revista da
sociedade Brasileira de Zootecnia. 2005. 56(3) 642-646.

Bahr, J. M. and Palmer, S.S. (1989). The influence of aging on ovarian function CRC Critical
Review in Poultry. 2 (2) 103-110.

Barker, J.S.F., Tan, S. G., Moore, S.S., Mukhere, T.K, Matheron, J. L., Selvaraj, O.S.
(2001). Genetic variation within and Relationship among population of Asian
goats (Capra hircus) J. of Anim. Breed. and Genetics 118 :213 –
Beaumont, C. (1992). Genetic parameters of the duration of fertility in hens. Can. J.
Anim. Sci. 72:193–201.

Becker, W. A. (1984) Manual of Quantitative Genetics Fifth Edition 45-52.

Bednarczyk, M., M. Siwek, A. Mazanowski and Czekalskid, P. ( 2002). DNA polymorphism

in various goose lines by RAPD-PCR. Folia Biol., (Krakow) 50:45-48.

Bernon, D. E. and Cambers, J .R. (1985) Maternal and sex linked effect in broiler
parent stock. Poult.Sc.64:29-38

Besbes, B., Ducroco, V., Fouley, J. L., Proteins, M., Tavernier, A., Tixier-Boschard, M. and
Beaumont, C. (1992). Estimation of Genetic Parameter of Egg Production Traits of

Laying Hens by Restricted Livelihood Applied to a Multiple Trait Reduced Animal

Model. Genetic Sel. Evol. 24, 539 – 552.

Besbes, B., Ducroco, V., Fouley, J. L., Proteins, M., Tavernier, A., Tixier-Boschard, M. and
Beaumont, C. (1992). Estimation of Genetic Parameter of Egg Production Traits of
Laying Hens by Restricted Livelihood Applied to a Multiple Trait Reduced Animal
Model. Genetic Sel. Evol. 24, 539 – 552.
Bhuiyan, M. M., Khan, M.H., Khan, M.A.H., Das, B.C., Lucky, N.S. and Uddin, M.B.
(2005). A study on the comparative performance of different breeds of broiler ducks,
under farmer’s condition at Farming System Research and Development (FSRD) Site
Syihef Bangladesh. Int. J. of Poult. Sc. 4 (8) 596 – 599.

Bochno, R., lwozuk, A. and.Wawro, E. (1994). A comparative study of the growth and
feed conversion efficiency of muscovy and Pekin duck .Poult. Abstr.20-81.

Bodi, I., Meszavos, E., I.A.C. S., Kozak, J., Karsaine, K. (1996). Prolificacy performance
of Hungarian upgraded and grey landes goose breed.1 st paper.Spring laying
cycle.Allattenyesztes-es Takarmanyozas(Hungary)45(5) 473-480.

Bolen, E.G. (1983). Cairina moschata. Pp 554-556. In Costa Rican natural history (DH
Jensen ed) Uni. Chicago Press Chicago Illinoise.

Bondarenkoa ,Y.V. (1997) Polymorphism of proteins of egg of chickens in different

breeds and lines Ptaklivnitstoo17:32-37.

Bogenfürst, F., Szász, S., Szécsényi, R. (1995). Evaluation of growth rate and carcass
characteristics of mule ducks according to the sex and feeding. Proc. of the 11th European
Symposium on the Quality of Poultry Meat, Tours, France. p. 240-245

Bonima, E. and Velez, A. M. (1992). Influence of natural and artificial incubation on

the hatchability in the Argentine Muscovy duck eggs 115-117. In proc. Of the 1st
intern.Symposium on Waterfowl. Italy.

Braccini Neto, J., Dunello, N.J.L., Silvera, J.L., Bongalhardo, D.C., Estivalet, J.C.N.
(1997). Estimation de parmetros geneticos e fenotipicos da cuiva de crescimento de
galiritias poedeiras.Revista da sociadade Brasileira de Zootecnia 1997 ;26(5):894-904.

Bradford, M. W. (2003). Conservation genetics of UK livestock from molecules to

management British societies of Anim. Sc..special issue on conservation of genetic
resources. Fouthhonrning Univ.of Nottingham press, Nottingham ,UK.

Brah, G. S., Chaudhary, M. L. and Sandhu, J.S. (1992). Heritabilities and correlations for egg
shell crack frequency, body checking, egg number and egg weight in laying hens. Br.
Poult. Sc. 37 : 947 – 951.
Brown , C. J., Brown, J. E. and Butts, W. T. (1973). Evaluating relationship among
Immatured measure of size, shape and performance of beef bulls. 11 .The
relationships between immature measures of size shape and feedlot traits in young
bulls.J. Anim.Sci., 36;1021.

Brown-Guedira, G. L., Thomas, J. A., Nelson, R. I .andWarburton, M. L. (2000).

Evaluation of genetic diversity of soyabeans .Introduction and North American
ancestors using RAPD and SSR markers .Crop Sc. 40:815-823.

Bruchi, P., Vendramin, G.E., Bussoh, F. and Grossoni, P. (2003) Morphological and
molecular diversity among Italian population of Quercus petraea Fagaceae.Ann
Botany 91;707-716.

Brun, J. M. and Larzul, C. (2003). Inheritance of reproductive traits of female common ducks (Anas
platyrynchos) in pur –breeding and in inter-generic crossbreeding with Muscovy
duck(Cairina moschata) Brit. Poult.Sc. 44, 1-6.

Buss, E. G. (1982). Genetic difference in avian egg shell formation . Poultry Sc. 61(3) 2048-2055.

Caldwell, P. J. and Cornwell, G.W. (1975). Incubation behaviour and temperature of the mallard duck
Auk 92: 706-731.

Carpenter, Jr.J A., Fitzhugh, Jr. H. A., Cartwright, T.C., Meltton, A. A.and Thomas, R. C.
(1971). Principal component for size of Herford cows J. Anim.Sci. 33:197

Castellini, C. and Panella, F. (1988). Heritability of pre and post weaning weight in rabbit.
Proceedings of the 4th World Rabbit Congress PP. 112 – 119.

Cavalchini, L., Dell-orto, G. and Roncoroni, C. (1979). Body weght gain, food consumption
and dressing percentage in two different strain of Muscovy duck Rivista- di-
Zootechnica-Veterinaria 5: 329 – 340

Chapius, H.M., Tixiev-Boichard, Y., Delabrose, A. and Ducroq, C.(1996). Multivariate

restricted maximum likelihood estimation of genetic parameters for production traits
in three selected turkey strain .Gen. Sel. Evol.28:197-215

Charnov, E. L. (1982). The theory of sex allocation. Princeton University press.


Chatterjee, R.N., Sharmer, R.P. ,Reddy, B. L.N, Niranja, M., Shivaprasad, and Mishra,
S. K. (2007). Genetic Analysis of highly inbred chicken using RAPD-PCR and
Immunocompetence. Intern. J. of Poultry Sc. 6(12) 967-972.

Chineke, C. A. and Adeyemi, A. (2001). Estimates of heritability of weaning and post

weaning body measurement in domestic rabbit. In proceeding of 26 th Annual
Conference of Nig. Soc. of Animal Production 26: 23 – 24.

Choi, J.H., Kang, W.J., Bark, D.H and Park, H.J. (1983). A study on some characteristics
of fractions and shell quality of the chicken eggs. Korea J. Anim. Sci. 25: 651-655.

Choprakarn, K., Salangam, I. and Tanaka, K. (1998). laying performance egg

characteristic and egg composition in the indigenous hen. J. Natl. Res.

Council Thailand 30: 1-17 .

Chamber, J.R. (1990). Genetics of growth and meat production in chickens.In Crawford
RD(ed) Poultry Breeding and Genetics Amsterdam,The Netherlands Elsevier 612-

Cheng, Y.S., Rouvier, R., Poivey, J. P., Tai, C. (1995). Genetic parameters of body
weight, egg production and shell quality traits in the Brown Tsaiya laying ducks.
Genet. Sel. Evol., 27, 459-472.

Chipchiryuk, G. (1980) .Ptitsevodstvo, 7:19- 20 /in Russian/

Clayton, G. A. (1984). Muscovy duck .In:Evolution of domestic animals .Mason and

Longman London Uk.

Coltman, D. M., Pillington, J., Krunk, L.E.B., Wilson, K.and Penberton, J. M. (2001)
Positive genetic correlation between parasite resistance and body size in a free living
ungulate population. Evolution 555:2116-2125.

Czekanowski, J. (1909) Zur differential diagnose der nean der tal grupe. Korresponden Zbratt dent
sch. Ges. Antrop. Ethnol. Urgesch 40:40-47.

Dan baro, G.K. , Oyame, P., Mukai, F., Tsuji, S., Tataishi, T.and Mae, M. (1995). Heritability
and genetic correlation from a selection experiment in brooder breeders using
Restricted maximum likelihood Japanese. Poult Sc. 32: 257 – 266.

Darwin, C. (1871) The descent of man and selection in relation to sex. John Murray London.

Debut, M., Berri, C., Baéza, E., Sellier, N.,Arnould, C., Guémené, D., Jehl N., Boutten,B.,
Jego,Y., Beaumont, C., Le Bihan-Duval, B. (2003). Variation of chicken technological
meat quality in relation to genotype and pre-slaughter stress conditions, Poultry Sci.

Delacour, J.T. (1959). The waterfowl of the world. vol 3 ,Country life London, England.

Derrel, E.M and Guy, A.B. (1989). Breeding biology of Muscovy duck using net boxes
in Mexico. Wilson Bull 101 (4) : 621-626.

Di Stasio, L. (1997). Biochemical genetics. In Genetics of sheep. Piper, L. & Runvisky, A. (Eds),
Oxon, CAB International, 1997..133-48 pp.

Dolmatova, I.Yu., Saitbatalov, T.F. andGareev, F.T. (2000). Study of geneticpolymorphism

using randomly amplified polymorphic DNA in ducks: differences
between breeds. Genetika Moskva, 36: 682-687.

Dolmatova, I.Yu., Saitbatalov, T.F. andGareev, F.T. (2000). Study of geneticpolymorphism

using randomly amplifiedpolymorphic DNA in ducks: differences
between breeds. Genetika Moskva, 36: 682-687.

Donkin, R. A. (1989). The Muscovy duck (cairina moschata domestica) . Origin

dispensal and associated aspect of the geography of domestication A A Balkema

Duong, X. T. and Nguyen, M. (1995). Studies on the performance of CV super M duck breed in the
southern province of Vietnam.Poult. Sc. Vol 23 (5) 140.

Duru, S., Akpa, G. N., Saidu, L., Olugbemi, T. S. and. Jokatan, G. E. (2006). A Preliminary
Study on Duck Management Under Peri-Urban System.
Eding, J. H. and Laval, G. (1997). Measuring genetic uniqueness in livestock. In JK
oldenbroek (ed) Gene bank and the conservation of farm animal genetics Resource.
DLO Institute for Animal Science and health .Lelystad. The Netherlands 33-58

Ehiobu, N. G., and Goddard, M. E. (1989). Heterosis in crosses between lines of

Drosophila melanogaster selected for adaptation to different environments.
Theor. Appl. Genet. 77:253 -259.

Ehiobu, N. G ., Goddard, M. E. and Taylor, J .F (1990). Heterosis in crosses

between geographical Separated populations of Drosophila melanogaster. Theor.
Appl. Genet. 80(4):569-575.

El-Deen, M. B, Kosha, M. A. and Soliman, A. S. A. (2009). Studies of some performance

and blood constituent traits in Japanese Quails. Egypt. Poult. Sc. Vol. (29) (iv): 1187
– 1208.
Ellegren, H. (1999). Inbreeding and relatedness in scadinavian grey wolves canis lupus
Hereditas 130: 239-244.
El-Gendy, E.A.,Helal M.A., Goher, N.H. and Mostageer, A.(2005).Molecular
characterization of genetic biodiversity in duck using RAPD-PCR analysis.Arab J.
Biotech.vol . 8 (2):253-264.

Emara, M. G. and Kim, H. (2003).Genetic markers and their application in poultry

breeding.Poult. Sc. 82:952-957.

Epstein, H.(1990). The context and prospect for development of smallholder poultry
production in Africa. In small holder rural poultry production .proc CTA seminar G13
Thessalonica Greece 35-50.

Eswaram, K. R., Ramakrishnam, A.,Venugopalan, C. K. and Nair, G. R. (1984). Comparative

performance of Khaki Campbell and desi duck. Indian j. of Poultry Sc. 19(1) 12-45.

Etuk, I.F. and Abasiekong, S.F. (2005). Duck production in South Eastern Nigeria. J. Agric.
Food and Envi. Vol. 2: (1) and 2.

Etuk, I. F., Abasiekong ,S.F., Ojewola, G.S. and Akomas, S.C. (2006). Carcass and
organ characteristics of Muscovy ducks reared under three management systems in
South Eastern Nigeria. Intern. J. of Poult. Sc.5 (5): 474–476

Everrit, B. (1980). Cluster Analysis 2nd edition Halstead press New York.

Everitt, B.S.,Landau, S.and Leese, M. (2001). Cluster Analysis. 4th Edition, London, Arnold
Publisher, Breed. Genet. 106 : 461–469.

Falconer, D.S. (1981). Introduction to quantitative genetics 2nd edition London New
York, Longman.34l pp

Falconer, D. S. (1989) .Introduction to Quantitative Genetics. 3rd Edition, Longman,

London . 203 – 206 pp

Falconer, D.S. and. Mackey, T.F.C (1996). Introduction to Quantitative Genetics 4th
ed Longman Essex.

Farrell, D. J. (1995). Table egg laying ducks: Nutritional requirement and current
husbandry System. J. of Avian Biol. Rev.6:55-69

Farooq, M , Mian, M. A , Faisal S , Durani F R , Arshad, M. and Khurshid, A.( 2001). Status of
broiler breeder stock in Abbotabah and Mansahra Sarhad. J. Agric 17(4):489-495

Fattouh, M. H., Tag El Din, T. H.; Gad, H. A. M.and Khalifah, M. M. (2003). Effect of
force moulting method on some post moulting traits in ducks 2- Fertility and
hatchability. 2 World Waterfowl Conference, 7- 9 October, Alexandria/Egypt .

F DLPCS (1991) Federal Department of Livestock and Pest Control services. Federal
Government of Nigeria 287- 288 pp

F L D PCS, (1992). Federal Livestock Department and Pest Control services Annual
report 1992.128pp

Feral, J.P.,( 2002). How useful is the genetic markers in Kulikova,Attempts to understand and
manage marine biodiversity. J. Exp. Mar. Biol. Ecol., 268: 121-145.

Fisher, R. A. (1936). The genetical theory of Natural selection. New York dover

Frith, H. J. (1967). Waterfowl in Australia Sydney, Angus and Robertson.

Funk, W. C. , Tyburezy, J. A., Knudsen, K.L., Lindner, K. R and. Allendorf, F. W. (2004 ).

Genetic basis of variation in morphological and life history traits of a world
population of pink salmon. J. of Heridatary 96(1) :24-33

Fumio, M., Hideaki, N. and Toyokazu, F. (1982). Application of principal component

analysis for defining size and shape in Japanese black cattle sires. The Science reports
of Faculty of Agriculture, Kobe University, Vol. 15 : 169-176,

Garcia, F., Baselga, M., Blasco, A.and Deitoro, J. (1982). Genetic analysis of some
productive traits in meat rabbits in numeric traits: In Proceeding of 2 nd World
Congress of Genetic Applied to Livestock Production. 557 – 562 pp.

Garzilov,V. (1991). Study on the growing characteristics and slaughtering content of

meat from Muscovy duckling to 70 days of age at mixed and separated sexual
breeding .Anim.Sc. xxxvi book 3-4 :27-32 pp.

Girard, P. and Palabost, L. (1976). Etude du polymorphisme enzymatique de 15

population natureles de drosophila melanogaster Archives de zoologies
experimentale et genetics117:41-55.

Grant, P. R. and Grant, B. R.,(1995). Predicting micro-evolutionary response to

directional selection on heritable variation. Evolution 49: 241-251.

Goodman, M. N. (1972). Distance analysis in biology System Zool.21:174-186.

Gueye, E. F. (1998) .Village egg and fowl meat production in Africa. World Poultry Sci.
54 73-86.

Guéye, E.F., Ndiaye, A. and Branckaert R.D.S., (1998). Prediction of body weight
on the basis of body measurement in mature indigenous chickens in Senegal.
Livestock Research for Rural Development 10(3)

Gunawan, B. (1989). Heritability estimate for egg production traits in Indonesian layer
duck.Majalah ilmu dan peternakan 41-48 pp

Gwakisa, P.S., Kemp, S.J. and Teale, A.J. (1994).Characterization of Zebu cattle
breeds in Tanzania using random amplified polymorphic DNA markers. Anim.
Genet., 25: 89- 94.

Hagger, C. and Abplanalph, H. (1978). Food consumption record for the genetic
improvement of income over food cost in laying flock of white leghorn. Br.
Poult. Sc. 19:651-667

Hair, Jr J. F., Anderson, R., Tatham, R. and Black, W. (1996). Multivariate data analysis
with reading prentice incorporated New Jersey.56pp

Hammock, S.P. and Shrode, R.R. (1986). Calf hoodweights,body measurements and
measuresof fatness versus criteria of overall size and shape for predicting yearling
performance in beef cattle. J. Anim. Sci. 63:447-452,.

Hanotte, O. and Jianlin, H. (2005). Genetic characteristic of livestock population and its uses in
conservation decision making. The role of biotech. Villi. Gualino Turin Italy 5-7 march

Hahn, D.T.; Hai, L.T.H.; Hung, D.S.; Tinh, N.H. (1995). Improving the Productivity
Of Local Muscovy duck and Initial Production Experiment of Mulard in Vietnam.
10 European Symposium on Waterfowl Proceeding, 31pp

Harper, J. (1972). The tardy domestication of the duck.Agric.History 46:385-389

Harun ,M. A .S ., Veenekans, R. J. ,Visser, H., Kampen, M. V. A.N., Vankampen, M .
(2001). Artificial incubation of Muscovy duck egg, why some eggs hatch and others
do not. Poultry Sc. 80(2) :219-224.

Harun , M .A. S., Veenaklass, R.J., Vankampen, M. and Mabasso, M. (1998). Breeding
biology of Muscovy duck (cairina moschata) in natural incubation; The effect of
nesting behaviour on hatchability. Poult.Sc.77:1280-1286.

Hassan, W .A. and Aliyu, A. T. ( 1996). Reproductive performance of duck in an arid

northern Nigeria. ANRPD newsletter 6 (1) 4-6.

Hassan, W. A. and Mohammed, W.S., (2003). Ecotype of the Muscovy duck in the north
west of Nigeria: variation in body weight and bill length. In: Proceedings. of 8th
Annual Conf. of Anim. Sc. Association of Nigeria.

Harun, M.A.S., Veeneklaas, R. J., Visser, G.H. and VanKampen, M ,

(2001). Artificial incubation of Muscovy duck eggs: why some eggs hatch and others
do not. Poult. Sc. 80 2, 219-224.

Hauser, L., Carvalho, G.R. and Pitcher, T. J., (1995). Morphological and Genetic
differentiation of the Africa clupeid Limnothrisssa miodun 34 years after introduction
to Lake Kivu ,J. of Biogeography 28.831-841.

Harville, D.A. (1977). Maximum likelihood approaches to variance component

estimation and to related problem. J. Anim. Sc. Assoc. 72:320.

Harvey, W.R. (1990). Mixed model least squares and maximum likelihood computer
programme .pc- 2 Mimeographs. Ohio State University press Ohio.

Heartmeyer, M. (2006). Super hens (part1) Differential among ducks .Mad duck. org.
The sportsman’s voice for waterfowl conservation 1-7.

Henderson, C. R. (1953). Estimation of variance and covariance components

Biometrics 9:226 – 253

Hesfer, M. (1997). Targeted mapping of chicken genome PhD. Theses submitted to

University of Hertfordshire 245pp

Hepp, G.R., Novak, J.M., Seriibner, K.T., Stangel, P.W. (1988). Genetic distance and
hybridization of black duck and mallards. A morph. Of a different colour AUK vol.
105 :804-807
Henderson, C. R. (1984). Application of linear models in Animal breeding .Univ. of
Guelph, Canada.

Hetzel, D. T .S. and Gunawan, B. (1984). Egg production of Indonesian native and

crossbred ducks under intensive and extensive conditions. Poult. Abstract. 13;2100.

Hetzel, D. J. S. (1984). Comparative performance of intensively managed Khaki

capbell and native Indonesian duck. Tropical Anim. Health and prod. 16:49-55.

Hillis, D.M. (1984). Misuse and modification of Nei genetic distance system. Zoo.

Holderread, D. (1978).The home of duck flock. The hen House Corvallis Oregon

Holsinger, K.E.,. Lewis, P.O. and Dey, D.K. ( 2002). A Bayesian approach to inferring
population structure from dominant markers. Mol. Ecol., 11: 1157-1164

Hunton, P. (1981). Genetic factor affecting egg shell quality. World Poult. Sc. J. 38 :

Huang, M.C., Lin, W.C., Horng, Y.M.,Rouvier, R. and Huang, C.W. (2003). Female-specific
DNA sequence in geese.British Poult. Sc., 44: 359-364

Hugue, K. S. and Sultana, N. (2002). Study on the existing duck production system in
Bangladesh. A report of Bangladesh livestock research institute, Bangladish.

Hu, Y. H., Rouvier R., Poivey, J. P., Lui, H. E. and Tai, C. (2006 ). Selection studies for
15 generation of Muscovy duck (cairina moschata) in duck Research centre 2006
.Symposium COA/INRA scientific corporation in agriculture. Taiwan.132-138.

Hu, Y. H., Poivey, J. P., Rouvier, R., Lui R. S. C. and Tai, C. (2004). Heritabilities
and genetic correlation of laying performance in muscovy selected in Taiwan .British
Poult. Sc. 45:180- 185.

Hu ,Y. H. (1999) .Variabilte genetique des performances de croissance et deponte

dans une lignee de canard de barbarie selectione a Taiwan. Viabilite embryononaire
precoce dan la croisement intergenrique descarned these de doctorate inst. Polytechn.
deToulous france.

Horst, P. (1999). Evaluation of local poultry resources for creating genetic stock with
improved adaptability, productivity and disease resistance in tropical environment
CTA 1999.78-83.

Ibe, S.N.(1982 ). Repeatability of growth traits of Nigeria local chickens using early
records. Nig. J. Anim. Prod. 22 (1&2)5-8.

Ibe, S.N. (1989). Measurement of size and confirmation in commercial broilers. J. Anim.
Breed.and Ge net. 106:461-469.

Islam, M.A., and Nishibori, M. (2009). Indigenous naked neck chicken, a valuable genetic
resource for Bangladesh. World Poult. Sc. J. 65: 125 – 138.

Islam, M., Sanpkota, J.D., Barna, N., Kalita, K.P. and Zaman, G. (2009). Certain
performance traits of Cina Hanh (Cairina moschata) Duck of Assam under range
condition. In. Proc. Of 4th World Waterfowl Conference. Thrissur India 108 – 112.

Isguzar, E. and Testik, A. (1999). An investigation on local genotypes of waterfowl in

Isparta province of Turkey. 12 European Symp. on Waterfowl, 51-56

Isguzar, E. ( 2005). Some egg and hatching traits of local ducks , Turkish Pekin and
Muscovy duckling Isperta Turkey.Arch Tierz Dummerstorf 48 (1) 94-100

Iraqi, M. M. (2003). Estimation and evaluation of genetic parameter for body weight
traits of New Zealand White rabbit in Egypt using different animal models

Jacquert, J. M. and Sanveur, B. (1995). Photoperiodic control of sexual maturation in

muscovy drakes. Domest. Anim. Endocrin. 12:191-195

Jayart, G., Peethambaran, P.A, Jalaludeen, A., Nara Yanan Kutty, K., and Ally, K. (2009).
Utilization of dried cuttlefish (Sepia officialis) waste silage in layer duck ration. In
Proc. Of 4th World Waterfowl Conference kerala, India 167 – 170.

Jean, R. D., Patricia, G., Sandin, M., Helene, L., George, P., Catherine, B. and
Bridgette, M.(2003). Genetic variability of sexual dimorphism in natural
population of drosophila melanogaster. An iso femaleline approach to genetic
Gen. Sel. Evol. 82(3):79-88.

Jensen, H., Seather, B.E., Ringsby, T.H., Tuffo J., Griffin, S. C. and Ellegren, H. (2003).
Sexual variation in heritability and genetic correlation of morphological traits in
house sparrow(passer domesticus) Gen. Sel .Evol 102:234-238.

Jianhua, W. (1984). The tri coloured duck poteries of the Tang Dynasty.China and World
Cultural Exchange.69.1.

Johnsguard, P. A. (1978). Duck geese and swans of the world . Univ. Nebraska press
Lincoln Nebraska.

Johnson ,S.G. and Johnson R.F. (1990). Environmental variation and quantitative
genetic parameter in the feral pigeon, columba livia .Biol. J. Linn. Soc. Land.40;321-

Joan, F., Marques A., Rego A.L., Costa, J. L. ,Costa M. J. and Cabral, H. N.( 2006).
Genetic and morphological differnciation of the Lutianian toad fish
(halobatrachus didactylus) between estuarine and coastal areas. In Portgal Sc. 749-758

Kadarmideen, H.N. (2004). Genetic correlations among body conditionscore, somatic cell
score, milk production, fertility and conformation traits in dairy cows.
Anim. Sci., 2004; 79: 191-201
Kalita, N., Serne D., Talikdar, J.K., Barua, N. and Ahmed, N. (2004). Comparative

performance of Khaki Campbell duck, dessi duck and their reciprocal crosses for
certain economic traits in rural condition. World Poult. Sc. J . 60:349-556

Kalita, N. and Deka, R. (2005). International Network for family poultry dev. Newletter 15:
Kaiser, M.G, Yonash, N., Cananer, A. and Lamont, S.J. (2000). Microsatelite polymorphism
between and within broiler population. Poult. Sc. 79: 626-628

Karaca, O.; Okut, H. and Altin, T. (1996). The estimation on some parameters of Pekin
duck’s egg productions. I. National Congrees of Zootecnical, Antalya/Turkey

Karaman, M. and Testik, A. (1995). The study of effect on hatching and growth
performance of Pekin ducks different egg weight.Intern.Congress of Poult.
Husbandry .Istanbul 1995. 485-492

Khalil, M. H., Owen, J. B. and Affifi, E. A. (1986). A review of phenotype and genetic
parameters associated with meat production traits in rabbits. Animal Breed.
Abst. 54: 726 – 749.

Khalil, M. H., Afifi, E. A. and Emara, M. E. (1987). Doe litter performance at weaning
for two breeds of rabbit with special emphasis on sire and doe effect. J. of Applied
Rabbit Res. 10:12 – 19.

Khurshid, A., Faruoq, M., Durani, F. R., Sarbiland, K. and Mansoor, A. ( 2004).
Hatching performance of Japanese quails LRRD 16(1)2000.

Kharel, M. and Arboleda, C. R. (1986). Some genetic factors responsible for body
growth in Muscovy duck. J. of the Inst of Agric. and Animal Sci. Nepal 7:101-110.

Klandorf, H. and Harvey, S. (1984) .Changes in thyroid function and gonadotropin

activity during sexual development in duck. Journ. of Zool. 203: 103-113.

Knit, H. G., Traore, A. and Wilson, R. T. (1985). A profile of small holder poultry
production in central Mali .In proceeding of of 2 nd International DLG-Symposium
,poultry production in hot climate of the middle East and Far East 16-19 June 1985
,Sporthel” Harzstern” Goslar-Hahnenklee,Germany 199-219 .

Knizetova, H., Hyanek, J., and Cerveny, J. (1992).Egg size and post-hatch growth of
Pekin duck. Arch. Geflügelkunde 56 3, 128-131.

Knudsen, P. B. and. Sohael, A. S (1970). The Vom herd: A study of performance in a

mixed Fresian – Zebu herd in a tropical environment, Tropical Agricult. (Trinidad)

Knust, U., Pingel, H., Lengerken, G.V.and Wicke, M. (1995). Effect of stress on

duck meat quality. Proc. 10th EuropeanSymposium on Waterfowl. World

Poult.Sc. Ass. 307-310pp.

Kul, S. and Seker, I. (2004) .Phenotypic correlations between some external and
internal egg quality traits in the Japanese quail (Coturnix japonica) Intern. J. of
Poult. Sci. 3: 400-405.

Ksiazkiewicz, J. M. (1995). Duck gene pool. International Symposium on conservation

measures for rare farm Animal breeds .Balice –Poland (1995)289-292.

Lack, D. (1967). The significant of clutch size in waterfowl. Wild fowl Trust 18 th .Ann Rept

Lack, D. (1968). Ecological and adaptation for breeding birds , Mthuen and co,
London, UK.56pp.

Lande, R. (1980). Sexual dimorphism , sexual selection and adaptation in phylogenic

character. Evol. 34:293-305.

Larson, K. and Fowlard, P. (1992). Genetics and social inheritance of body and egg
size in Barnacle goose (Branta leucopsi) Evolution 46:235-244.

Lawrence, T.L.J. and Fowler, V. R. (2002). Growth of Farm Animals 2nd ed CAB
inter. Wallingford UK 347.

Leclercq, B. (1990). Croissance et composition corporelle du canard de Barbarie. In

:Le Canard de Barbarie (Sauveur B. et de Carville H., Ed.). Institut National de la
Recherche Agronomique, Paris : 23–29

Leclercq, B. and DeCarville, H. (1985). Growth and body composition of muscovy duck.
Duck production science .Proceedings of workshop 18-22 November 1085 .Indonesia.

Leopold, A.S. (1959). Wildlife of Mexico University California Press Berkeley


Larbier, M. and Leclercq, B. (1994). Żywienie drobiu. In: Pastuszewskiej B., Smulikowskiej
S. (eds.): Tłumaczenie zbiorowe pod red. Wyd. Nauk. PWN, Warszawa.

Le Bihan – Duval, E., Beaumont, C. and Colleau, J.J. (1998). Estimation of the genetic
correlation between twisted leg and growth or conformations traits as brooder. J.
Breed. Genet. 112: - 239 – 259.
Lima Guerra, L.J. (2004).Statistical model and genetic evolution of binomial traits.PhD
Dessertation Luisiana state Univ. USA138pp.

Lindsey, C.C. (1988). Factors controlling meristic variation .J Physiol. 11 :197-274.

Lukefahr, S. D. (1982). Evaluation of rabbit breeds and crosses for overall

commercial productivity PhD Thesses Oregon State University Corrattis, USA.89pp.

Lukefahr, S. D., Hohenboken, W. D., Cheeke, P. R and Patton, N.M. (1984). Genetic
effect on maternal performance and litter preweaning and post- weaning traits
in rabbits. Animal Sc.J. 34 231-238.

Loung, T.N., Hong, L., Le X.T., Doan, V.X. and Nguyen, D.T. (1995). Growth and meat
performance of Cv supper M. Duck under management system in the red river Delta
Poultry. Abst. 23(5)140.

Lynch, M. (1990). The similarity index and DNA fingerprinting. Mol. Biol 7:478-484
Lynch, M. and Ritland, B. (1999). Estimation of pairwise relatedness with molecular markers
J. Gen. Soc. Of America 4:127-130.

Lynch, M. and Walsh, B. (1998). Genetics and analysis of quantitative traits. Sinauer
Associates. Sunderland. M A. 45-53.

Maak, S.,Neumann, K.,Vonlengerken, N.and Gattermann, L.(2000). First seven microsatellite

developed for the Pekin duck. Anim. Genet. 31:233.

Maciuszonek, A., Grajewski, B.and Bednarczyk, M. (2005). RAPD-PCR analysis of various

goose populations. Folia Biol. (Krakow) 53: 83-85.

Mahanta, J.D., Sapcota, D., Milli, D.C., and Chakraborty, A. ( 2001). A survey on duck
farming Lakhimpur and Dhemaji district of Assam .India Vet. J. 78(6) 531-532.

Mahalanobis, P. C. (1936 ) .On the generalised distance in statistics . Proceedings of

Natural Institute of Science 1936:2:49-55.

Marchant, S. and Higgin, P .J. (1990) . Handbook of Australian , New Zealand and
Antartic birds. Oxford University Press , Oxford.87pp.

Marle-Koster, E.V. and. Nel, L. H. (2003). Genetic markersand their application in livestock
breeding in south Africa: A Rreview. S. Afr. J. Anim. Sci., 33: 1-10.

Marques, J. F., Rago, A. L., Costa, J. L.,Costal, M. J. and Cabral, H. N. (2006).

Genetic and morphological differentiation of the Lutianian toadfish (halobachus
didactylus) between estuarine and coastal area in Purtugal .Scientia Merina 70(4) 749-
758 .
MaynardSmith, J. (1978). The evolution of sex .Cambridge University. Press Cambridge.

Mcainsh, C.V, Kusina, I., Madsen, J. and Nyoni, O. (2004). Traditional chicken
production in Zimbabwe .World Poult. Sc. 60: 233-244.

McClleland, W.J. (1990). Fingerprinting genomes using PCR with arbitrary primers .Nucl.
Acid Res. 18:7213-7218.

Mc Cracken, K. G. (1999). Systematic ecology and social biology of the musk duck
(Bizuira lobata) of Australia PhD Dessert. Louisiana State Univ. Baton ronge.

Mc Cracken, K.G, Paton, D .C. and Afton, A. D., (2000). Sexual size dimorphism of
the musk duck. Wilson Bull.112 (4) 200 457-466.

Melchinger, A. E. (1993). Use of RFLP markers for analysis of genetic relationships

among breeding materials and prediction of hybrid performance 621-628. In
DR Burtor (ed) Proc. Of the Internal Crop Sc. Congress 1st Amer. 1992 USA
Madison WI.
Merat, P. (1990). Pleiotropic and associated effects of major genes , egg production in
chicken, In crowford RD Poultry breeding and genetics Amsterdam Elsevier Science
Publishers. 1990 : 429 – 467.
Merat, P. (1986). Potential usefulness of the Na (naked neck) gene in poultry production.
World poult. Sc. J. 42: 120 – 142.

Merrit, E.S. (1966). Estimate by sex of genetic parameter for body weight and skeletal
dimension in a random bred strain of meat type fowl. Poult. Sci. 45:118-125

Merila, J. and Shelden, B.C. (2001). Avian quantitative genetics .In Nolan V.ed
current ornithology, Kluwer Academic Plenum Publisher.82pp.

Merila, J., Shelden, B. C. and Ellegren H. I. (1998). Quantitative genetics of sexual

size dimorphism in the collared fly catcher Ficedula albicollis. Evol. 52:870-876

Metzer, A. (1988). Incubation of Muscovy duck egg. Proc. Of Intern.symposium on

waterfowl production. The satellite conf. for xviii world poult. Congress Beiging
China 11-18 Sept 1988

Meulen-Vander, S.J. and Dikken, L. G. (1999). Duck keeping in the tropics CTA
Agrodokseries33 ,56-64

Meyer, K. (1991). Estimating variance covariance from multivariate animal models by

restricted maximum likelihood .Genet.Sel. Evol. 23:65

Meyer, K. (1989). Restricted maximum likehood to estimate variance component for

animal models with several random effects using a derivative free algorithm.
Genet. Sel. Evol. 21:31

Migno-Grasteau, S. ,Beaumont, C., Poivey, J. P. and Rochambeau, H. (1998). Estimation

of the genetic and morphological differentiation of the lustanian toad fish
(Halobatrachus didectylus)between estuarine and coastal area in Putugal.Scentia

Mignon- Grasteau S., Beaumont, C., Le Bilan-Duval, E.,. Poivey, J. P, Rochambeau, H. and
Ricard, F. H. (1999) .Genetic Parameters of Growth Curve. Parameters in male and
female chickens. Poult. Sc. 40:44-51.

Minitab, (2004). Statistical Software for Windows Versions 14.0.

Misra, R. K. and Carscaden, J. E. (1987). A multivariate morphometrics to detect


difference in population(Mallotus villosus) J. Cons. Int.Explor. 43

Mohammed, A.S. (1996). Hatchability of Muscovy duck (Cairina moschata) eggs and energy
metabolism of duckling in a tropical climate.In Proc. South Africa

Mohammadi, S.A. and Prasanna, B.M. (2003) ..Analysis of genetic diversity in crop plant,
salient statistical tools and consideration, Crop Sc. 43; 1235-1248.

Momoh, O. M. (2005). Genetic and Phenotypic Evaluation of the Nigerian Heavy

Chicken Ecotype and Its Cross Bred with the Light Ecotype PhD Thesses University
of Agriculture Makurdi.2005.

Morris, I.R. and Fox, S. (1960). Use of light to delay sexual maturity in pullet. Anim.
Breeding Abst. 28:458.
Morrison, D. F. (1976). Multivariate statistical methods. McGraw-Hill Company, New

Morris, T. R. (1966). The effect of light on sexual maturity in the female domestic fowl.PhD
dissertation. Univ. of Readings UK.

Mogese, H. H. (2007). Phenotypic and genetic characteristic of indigenous chicken

population in Northern Ethiopia Ph D. dissertation submitted to department of animal
and wildlife and grassland service University of Free State Bloemfontein, South

Moller, A. P. (1990). Effect of haematophagous mite on the Barn swallow(Hiv undo

rustica) a test of the Hamilton and zuk hypotesis. Evol. 44:721-784.

Monseau, T.A., Senrro, B., and Endler, J. (2000). Adaptive genetic variation in the
wild .Oxford University Press Oxford.

Moiseyera, I. G., Semyenova, S. K ., Benhicora, I. V. and Filipora, N.D.(1994). Genetic

structure and origin of an old Russian orlf chicken breed.Genetica. 30:651-69.

Mopate, L.Y., Balndoh, G., Zeuh, V.and Gonguet, G. P. (1999). Muscovy duck
(Cairina moschata) rearing in urban households of NDjamena Chad .ANRPD
Newsletter 8 (1):42.48.

Mourthe, M. (1989). Lles oles et canards. Collection la vie en vert dargaud editeur 8 e

Murad, A., Farouq, M., Mian, M. A. and Muqarrab, A. K. (2001). Hatching performance
of fayoumi egg.Sirhad J.of Agric. 17(1) 1-6

Morae, R. D.(1988). Natural and artificial hatchability studies in Muscovy duck

(cairina moschata)pages 643-646.In proceedings of the xviii world poultry congress
Nagoya, Japan.

Murtha, A.G. ( 2000). Morphological variation of Horse mackerels (Trachurus trachurus) in the
Iberian and North African Atlantic: implications for stock identification. J. Mar. Sci., 57:

Mtui, D. and Mbaya, S.H. (2001). Performance characteristics of Muscovy duck

under improve management . http// retrieved on

Nagamine, Y., and Higuchi, M. (2001). Genetic distance and classification of domestic
animals using genetic markers. J. of Anim. Breed. and Genet. 118: 101-

Narahari, D,K. Abdul-Mujev, A., Thangavel-Ramamurthy, N., Visunathan, S.Mohan, B.,

Murgananda, B. and Sunrararosu, V. (1988). Traits influencing the hatching
performance of Japanese quail eggs. Br. Poult. Sc. 29: 101-112.

Narahari, D., Mujeer, K.A., Ahmed, M., Rajini, R. A. and Sundararasu, V. (1991). Factors
influencing the hatching performance of duck egg .Br Poult. Sc. 32:313-318.

Nalbardov, A. V. and Opel, H. (1974). Egg laying 394-408.Animal agric. H .Cole

(ed) Freeman and Co NY.72pp.

Nei, M. (1987) .Molecular evolutionary genetics. New York: Columbia University


Nogales, M., Lopez, M., Jimenez Asensio, J., Larruga, J. M., Hernandez, M.and Gonzalez,
P. (1998). Evolution and biogeography of the genus Tarentola(Sauria Gekkonidae) in
the canary Island , inferred from mitochondria DNA sequence. J.Evol.Biol.11:481-

Nei, M. (1973). Analysis of gene diversity in subdivided population. Proc. Natl. Acad.
Sci USA 70:3321- 3323.

Nei, M. (1972). Genetic distances between populations. American Naturalist 106:


Ngapongora, J.M.N., Mbaga, S.H. and. Mutayoba, S.K. (2004). Study on the
performance of growing native Muscovy ducks under semi and fully confined
rearing systems. www.husdyr. on 4/8/2009.

Nguyen, T.K. D., Ogle, B. and Preston, T.R. (1997). Effect of level of local supplements for
fattening Muscovy duck by poor farmers in remote villages in Mekong Delta of
Vietnam. hHp: //www. 91.htm.Retrieved on 6/11/2009.

Nichelman, M. (1993 ). Early experience and adaptation. 8-18 .In proc. Of the 4th
European symposium on poultry welfare . C J Savory and B O Hughes ed Univ.
Federation for Animal welfare .Edingburg UK. 234-242.

Nichelman, M. and Gilsing, A. (1995). Influence of stressor on physiology and

behavioural reactions in domesticated Muscovy ducks, In proc. Of 10th
European Symposium on Water Fowl Halle Scabe Germany 124-133

Nickolova, M.and Gerzilov, V.(2000). Study on egg productivity of Muscovy ducks

(Cairina moschata). II. Morphological characteristic. Zhivotnov"dni Nauki 37 4, 13-
16; Rec. 173 of 758 in CAB Abstracts 2000/08-2002/07.

Nickolova, M. (2003). Study on some major reproduction factors of Muscovy ducks

(Cairina moschata) with element of the incubation technology PhD thesis Agricultural
University of Plovdiv. Bulgaria 2003.146pp.

Nickolva, M. (2004). Effect of the sex ratio on the fertility of Muscovy duck( Cariana
moschata).Central European J. of Agric. 5 : 4 (367- 372).

Nikiforov, A. A, Moiseyeva, I.G., and Zakharrov, I.A .(1998). Position of Russian chicken breed in
thediversity Eurasian breeds. Genetika. 34:850-851.

Njenga, S. K. (2005) . Productivity and socio –cultural aspect of local poultry

phenotypes in coastal Kenya M Sc Thesis The Royal Veterinary and Agricultural
University. Copenhagen, Denmark.55pp.

North, M. O. (1981). Egg quality as a major hatchability factor .Indian

Poult..Review.VOl.xiii 1 :32- 35.

North, M. O. (1984). Breeder management .In commercial chicken production.

Manual. The Avi publishing Connecticut 298-321.

Notter, D. R. (1999). The importance of genetic diversity in livestock population of

the future. J. of Anim. Sc. 77:61-69.

Nwosu, C .C., Gowen, F. A.,Obioha, F.C., Akpan, I. A.and Onuora, I.(1985). A biometrical
study of the Conformation of the native chicken .Nig. J. Anim .Prod. 12 (1&2) 141-

Odubote, I. K. and Somade, B. (1992). Genetic analysis of rabbit litter traits at birth
and weaning. Nig. J. of Anim. Prod. 19:64-69.

Ogah, D. M., Musa, I. S., Yakubu, A., Momoh, M.O. and Dim, N.I. (2009). Variation in
morphological traits of geographical separated population of indigenous muscovy
duck (Cairina moschata)in Nigeria. In: Proceeding of 5th Inter. Poult. Conf. Taba
Egypt. 46-52.

Ogundu, U. E. (2001). Repeatability estimates and prediction of milk suckled per kit
for performance trait in Dutch rabbit. J. of Sustainable Agric. and Environ. 3:149 –

Ola, S.I .(2000). Vital reproduction and production characteristics of the Nigerian
Muscovy duck. In proc. of Nig.Soc.of Animal Production Conf. 19-23 march 2000

Oli, M. K. and Dubson, F. S. (1999). Population cycle in small mammals :The role of
age at sexual maturity. Oikos 86:557-561

Olson ,S. L. (1985). The fossil record of birds.76-238.In avian boil.Acad. Press New York.
Oldenbroeck , J.K. (1999) .Gene bank and the conservation of farm animal genetic
resources DLO Institute for Animal Science and Health. Lelystad

Olori, V. E. ,(1992). An evaluation of two ecotypes of the Nigerian Indigenous

chicken. Msc Thesis submitted to dept of Animal Sc. OAU Ile Ife.142pp

Oluyemi, J. A, Longe, G. O. and Songu, T.,(1982). Requirement of the Nigeria

Indigenous fowl for protein and amino acid. Ife J. of Agric 4: 105-110

Ohiosimuan, O. O., Somade, B.and Odubote, K. (1994). Production characteristics of

rabbit breeds in the humid tropics of Nigeria.In: Proceeding of Nig. Soc. of Animal
Production 1994. 212-214

Osman, A. (1997). Effect of diet containing dried poultry manure and nivginamicun in
the reproductive performance of muscovy duck. Egypt Poult. Sc. J. 17 (2) 171-191

Ozimba, C. E. and Lukefahr, S. D. (1990). Evaluation of pure bred and cross-bred

rabbit for carcass merit. Alabama AAM University Huntsville 35762.

Panda, B. K. and Kumar, R. (2004). Muscovy duck and geese are best choice for
backyard farming. Poultry Planner Year 5 issue 12 :11-12.

Panda, B. K. and Padlin, M. K. (2004). The past present and future scope of duck
farming in India. Poultry Planner Year 6 issue 2 :19-22.

Parkhurst, C. R .and.Mountney, G. J. (1988). Poultry meat and egg production.

Published by Van Nostrand Rainhold comp.New York USA.

Pascal, J. (1981). Editorial, Poultry Science. 2167-2171 pp.

Patterson, H. D., and Thompson, R., ( 1971). Recovery of interblock information

when block sizes are unequal. Biometrika 58:545–554.

Payne, W. J .A. (1990). Longman Scientific and Technical 881pp.

Pearson, K. (1926) . On the coefficient of racial likeness. Biometrica 18:337-433


Peeble, L. and Brake, J. (1987) Egg shell quality and hatchability in broiler eggs .
Poult. Sc. 66:596-604

Peebles, E.D., and Marks, H.L .(1991). Effect of selection for growth and selection
diet on egg shell quality and embryonic development in Japanese quail. Poult. Sc. 20:

Pederson, C.V. (2002). Production of semi-scrounging chicken in Zimbabwe PhD Thesis

Royal Vet. And Agricultural Univ. Copenhagen 2002.
Perez, R.( 1985). Rearing manual . University of west India 57 Augustine Trinidad.

Petek , M. (1999).Production traits and economic efficiencies of different genotypes of layers

reared by enterprise in Bursa province and its near vicinity. Veteriner Fakultesi
Dergisi uludag Univ.18(1-2) 15-77.

Peters, S. O. (2000). Genetic variation in reproductive performance of indigenous

chicken and the growth rate of its pure and half bred progeny .M.Agric Dissertation
Univ. of Agric. Abeokuta 2000.

Peter, S. O., Omediji, E. A. , Ikeobi, C.O.N., Ozoje, M.O. and Adebambo, O. A.

(2004). Effect of naked neck , and frizzle genes on egg traits, fertility and hatchability
in local chicken. In self sufficient in animal protein in Nigeria.In Proc. Of 9th Ann.
Conf.of ASAN Abakiliki 2004, 262-264.

Piassi, M. (1994). Avillacao do desempenho de luna gen de postura mantidas na

universidade federal de vicosa, em competicae commarcas commciaisi vicosa (mg)
Universidade Federal de Vicosa 1994.

Piassi, M., Silva, M. A. ,Regazza, A.J. Torres, R.A. Soares, P.R. Torres,and Junior,
A.A. (1995). Estudo da divergencia genetica entre oito grupos de aves de posturn por
melo de teenicas de analise multivariada.Revista da Sociedade Brasileira de Zootecnia
1995 ,24(5)715-727.

Pilla, A.M. and Quilici, R. (1973). .Influence of different diets and some slaughter
characteristics of Muscovy ducks and other annali-dell-instituto-sperimental er-la-
Zootehnica 6 (1) 25 – 95.

Piles, M .O., Rafel, P., Ramon, J. and Varona, L. (2005). Genetic parameters of fertility
in two lines of rabbits with different reproductive potentials J. of Anim. Sc. 83:340
– 343

Pingel, H, (1990). Genetic of Growth and meat production in waterfowl in poultry breeding
and genetic, R-D. Crawford Ed. Elsevier Amsterdam, Netherlands 191 – 704.
Pinheiro, A. C.M., Teixeira, A.L., Rego Marques, J.F. and Cabral, H.N. ( 2005). Genetic and
morphological variation of Solea lascaris (Risso, 1810) along the Portuguese coast. Fish. Res.,
73: 67-78.

Pinto, L.F.B., Parker, I.U., Melo, C. M.R., Ledur, M.C. and Coutinho, L.L. (2006). Principal
component analysis applies to performance and carcass trait in chicken. Anim. Res. 55
419 – 428.

Phuoc, T. H., Mein B. X. and Nguyen, T. K. D. (1994). A study of performance of local

Muscovy duck in Mekong delta of Vietnam science and technology university of
Cantho.Mekang Delta of Vietnam no2, 22-24.

Philip, J .C. (1922). A natural history of the duck. Vol.1 Houghton Mifflin co.Boston

Poivey, J.P., Cheng, Y.S, R., Rouvier, C., Tai, C.T. Wang, Liu H.L (2001). Genetic
parameters of reproductive traits in Brown Tsaiya duck artificially inseminated with
semen from Muscovy ducks. Poult. Sc. 80, 703 – 709.
Poujardieu, B., Guichard, F.and Laventure P. (1994). Parametres genetiques de croissance et
de garage de la cane commune. Genet. Sel. Evol. 26, 467 – 472.
Prado-Gonzales, E.A., Ramirez-Avila, L. and. Segura, J.C. (2003). Genetic
parameters for body weights of Creole chickens from Southeastern mexico
usingan animal model. Livest. Res. Rural Dev., 15.

Queller, D.C., Strassmann, J.E. and Hughes, C.R. (1993). Microsatellites and Kinship. Tree,
8: 285-288.

Rafinski, J. and Babik, W. (2000). Genetic differentiation among northern and southern
populations of the moor frog Rana Arvalis Nilsson in central Europe.Hereditary vol.
84. (5) 610-618.

Rauch, H. W., Pingel, H. and Bilsing, A. (1993). Welfare of waterfowl pages 139-147 . In
Proc. Of the 4th European symposium on poultry on poultry welfare C J savory and B
O Hughes ed. Anim. Welfare Edingbough UK. 242-248.

Rastogi, R. K., Lukefahr, S. D. and. Lauckar, F. B. (2000). Maternal heritability and

repeatability for letter traits in rabbit in a humid tropical environment. Livestock Prod.
Sc. 67:123 – 128.

Retailleau, B. (1997). Performances of roasting mule duck testing of various genotypes

and of various slaughter ages. Proc. 11 th European Symp. on Waterfowl, Nantes,
France 633-638.
Ravi, S., Peethambaran, P.A., Jalaludeen, A. and Leo, J. (2009). Effect of calcium and
phosphorus in some carcass traits of indigenous layer ducks in Kerala In: Proc. of 4th
World WaterFowl Conference, Thrissur, India. 267 -268.

Ravi, S.,. Peethambaran, P.A, Jalaludeen, A., leo J.and Gangadevi, P. (2009). Influence of
dietery calcium and phosphorus level on production performance in indigenous
layers duck in cages. In: Proc. of 4th World Water Fowl Conference, Thrissur, India.

Ricard, F. H., Lerclercq, B. and De Carville, H. (1983). Mise au piot sur les caracteristiques
genetiques et les methods delevage du canard de barbarie.Ann. de zootechnic32;189-
Riva, J., Rizzi, R., Marelli, S. and Cavalchini, L.G., (2004). Body measurements in
Bargamasca sheep. Small Rum. Res. 55: 221-227.

Robert, M. B. V.( 1978). Biology, A functional Approach by St Paris Press Ltd Malta.

Rochambeau de, H.( 1988 ). Genetics of rabbit and their crosses. J. of Applied
Rabbit Res. 12 231 – 212.

Roff, D. A. (1997). Evolutionary quantitative genetics. Chapman and hills London.

Rodero, E., deLa Haba, N. R., Rodero, A. Y., Harera, M. (1996) . Genetic and phenotypic
profile of endergered Andulusian Sheep and goat breeds FAO Health paper vol 8.

Romboli, I., Mori, B., Salami, A.,and. Avanzi, K. F. (1984). Storage condition and
hatchability in Muscovy duck egg 17th congress. Mord. Aviculture,Helsinki
Romboli, I. and Battini, T. (1986). Del guscio eschivsa nelle uova di auatra.Avicoltura3: 65.

Romboli, I., Giulotti, L., Avanzi, C.F. (1987). Reproductive activity of some varieties
of Muscovy duck in successive egg laying cycles. Rivista di Avicoltura, 56 12, 73-
75, Rec. 201 of 408-CAB Abst. (1987-1989).

Romboli, I. (1990). Quelques donnees sur lelevage extensive du canard de Barbarie

. In: Actes du Colloque de Belgrade du 5 au 7 novembre 1987

Romanov, M. N. (1994). Using phonetic approaches for studying poultry population

under preservation and breeding. Proc.of 5th World Conf. on Genetic Applied to
Livestock Production Guelph Canada vol21 :551-559.

Romboli, I. (1995). Production factors and meat quality in waterfowl. Proc. 10 th

European Symposium on Waterfowl. World Poult. Sc. Ass. 249-259.

Ruene, J.(1993). Documenting the world domestic animal resources. Animal Genet.
Resources Information.11:13-22.

Saatci, M., I. Dewi, A .P. and Aksoy, A.R. (2003). Application of REML procedure to
estimate the genetic parameters of weekly live weight in one to one sire and dam
pedigree recorded Japanese quail. J. Anim Breeding and Genet. 120 : 23 – 28.

Sabri, H. M., Henry, R.W.,Robert, H.H. and Wilcox, C. J. ( 1990). Genetic parameters for
egg and related characteristic of white leghorn hens in a subtropical environment.
Gen. and Molecular Biol. 22(2) 183-186.

Sabri, H. M.,Wilcox, C. J.,Wilson, H. R. and.Harms, R.H. (1991). Measurement of genetic

variation in residual metabolizable energy intake in laying hen . Poult. Sc.

Salicon, I. (1991). Elevage of gavage des Palmipede Paris 1991.231-238.

Sakaguti, E., Silva, S., Regazzi, A. J. Torres, R. A., Martin, L.N. (1996). Analise de
divergencia genetica entre nore grupos geneticos de coelhos. Revista da sociedade
Brasileira de zootecnia 1996 25(3)647-660.

Salako, A. E. and Ngere, L.O. (2001). Genetic distance between West African Dwarf
(WAD) and Yankasa Sheep in South West Nigeria. Nig. J. of Genetics 16(2001) 37-

Salako, A. E. (2006). Principal component factor analysis of the morphostructure of

immature Uda sheep . Inter. J. Morph.24(4) 571-774.

Salem, H. H., Ali, B. A. , Huang, T. H. and Quin, D. N.(2005). Use of Random amplified
polymorphic DNA marker in poultry research. Inter. J. of Poult. Sc.

Sanghiv, L. D. (1953). Comparism of genetic and morphological methods for a study

of biological differences. Amer. J. Phys.Anthrop. 11:385-404.

Sauveur, B. (1990). Origines et performances comparées du canard de Barbarie et

du canard commun de race Pékin. In Le Canard de Barbarie (B Sauveur & H. de
Carville, éds). Institut National de la Recherche Agronomique, Paris, 3–11.

Sauveur, B. and De Carville, H. (1990). Effect of day length on early puberty on female
Muscovy duck, control of fertility in domestic birds Tours, France 2-4 July Colloques
–de- 1 INFA 1990 5l: 192 – 203.
Sauveur, B. and De Carville, H. (1985). Duck production Science and world practice.
Proceedings of a workshop 18 – 27 Nov. 1985 Indonesia.

Saitbekova, N., Gaillard, C., Obexer-Ruff, G., Dolf, G. (1990). Genetics based on
microsatelite analysis. Anim. Genet. 30:36-41.

Savistzky, B. A. C. (1989). Aquatic foraging in the two independently evolved species of

snake Ner dicrhombifer (colubridae) and Agkistrodon piscivirus (viperidae) PhD
thesis, Univ. of Tennesse Knoxville.

SAS, (1990). SAS user guide statistics SAS Inc Cary N C 633.

Scherf, B.D., (1995). World Watchlist for domestic animal diversity.3 rd ed. Part1.9 database.Rome Italy FAO/UNDP.

Schearf, B. (2003). World Watch list for domestic animal diversity FAO 2003.423-432.

Schader, A. K. , Ahuja, S. D., Thomas, P .C. and Agrawal, S.K. (1985). Effect of egg
weight and duration storage on the weight loss,fertility and hatchability traits in
Japanese Quail. Indian J..Poult.Sc.20:19-22.

Scott, N. D. and Reynolds R. P. (1984). Phenotype variation of the Mexican duck in


Scot, M. L. and Dean, W. F. (1991). Nutrition and management of ducks.M L Scot. Ed

1991 1-17.

Schultze, V. R. , Rohe, H., Loft, R.and kalm, E.(2001). Genetic analysis of the course of
individual growth and feed intake of group-pened performance tested boar. Arch
Tierz Dummerstorf 44:139-156.

Serbul , V.P. (1986). Improving Muscovy duck egg incubation technique pages 819-
822.In proc. of the 7th European Poult. Conf.D M Lerbier Tour, Paris, France.

Serbul, V. (1983). Characteristic of Muscovy duck egg incubation, Ptitsevodstvo 6:17-

Sezer, M., Berberoglo, E. and Ulutas, Z. (2006).Genetic association between sexual maturity
and weekly live weights in Japanese quail. South African J. of Animal Sc.36(2)142-

Siripholvat, V. and Tragoonrung, S. (1998).Comparative studies the indigenous domestic

duck and muscovy genomic DNA using thechicken specific microsatellite
primers. Proc.6th World Congress on Genetics Applied toLivestock Production.
Armidale, Australia, 26: 308 (Abstract).

Shanaway, M. M. (1994) .Quail production system. A review. Animal production and

health FAO Rome , Italy.76-79.

Shahin, K.A., Soliman, A.M. and Moukhtar, A.E. (1993).Sources of shared variability
for the Egyptian buffalo body shape (conformation). Livest. Prod. Sci, 36: 323- 334.

Shahin, K. A., Shemes, A. R., Abdallah, O. Y., Saleh, K. (2000) . Effects of genetic
control of subcutaneous fat deposition via using restricted selection indexes on live
performance and carcass characteristics of Pekin ducks. Arch. Tierz., Dummerstorf.
43 1, 69-77.

Sharma, A.K.,. Singh, R.K, Kumar, S.and Sharma, D. (2001.) Genetic relatedness among

chicken breeds using randomly amplified polymorphic DNA (RAPD) markers.

Indian J. Anim. Sci., 71: 941-945.

Simons, L. W. (2001). Sperm competition and its evolutionary consequence in the insects
.Princeton Univ.Press.Princeton.

Singh, R.V. and Sharma, D. (2002). Within and between strain genetic variability in White
Leghorn population detected through RAPD markers. Br. Poult. Sci., 43:33-37.

Smith, E.J., Jones, C.P., Bartlett, J. and Nestor, K.E. (1996).Use of randomly amplified
polymorphic DNAmarkers for the genetic analysis of relatedness
and diversity in chickens and turkeys. Poult. Sci., 75: 579-584.

Snedecor, G.W and Cochran, W.G. (1980). Statistical method seventh edition. The Iowa
State University Press Ames lOWA. USA.

Sogomonov, A. and Rahmanov, L. (1988) The production of Muscovy duck in hot


Sokal, R.R. and , Rohlf, F.J. (1981). Biometry. 2nd ed. New York: W.H. Freeman and

SPSS, Statistical Package for Social Sciences.( 2004)SPSS Inc.,(14.0 )444 Michigan
Avenue, Chicago, IL60611.2004.

Stasko, J. (1966 ).The heritability of the live weight of duck at 28 day of age .Ved.Pr
Vysk ust Chov. Hyd invake pri Dunaji 3 :35-42.

Stadelman, W. J. (1986). The preservation of quality in shell eggs. In W.J Stadelman.

O.J cotteni (eds) Egg science and technology. Avi publishing con. Inc. West
port Connecticut USA.

Stahl, J.L., Cook, M. E. and Sunde, M.L. (1986). Zinc supplementation its effect on egg
production feed conversion, fertility and hatchability. Poult. Sc. 65:2104-2109

Stearn, S.C. (1992) .The evolution of Life history.Oxford Univ.Press.

Stephen, P. and Sauveur B. (1985). Duck production and management in France.

Proc. Of Workshop at CIPANAS, Bogov, Indonesia nov. 18-22.

Stephen, P. (1990). Selection de Canarc de Barbarie. Inshe Canard de Barbarie (B. Sauveur
and H. de Caville eds) Institute National de la Recherche Agronomique, Paris

Swiger, L.A., Harvey, W.A.,Everson, D.O. and Gregory, K. E. (1964). The variance of
intraclass correlation involving groups with one observation.Biometrics 20:818-826
Szyniuk, B., Holszwska, Z. and Palnszak, H.(1991) Zootechinca 22:55-60

Szwaczkowski .T. , O. Wezyk, E.S Barezak .J. Badowski , H. Balinska and A. Wole (2007)
Genetic venerability of body weight in two goose strain under long term selection. J.
Appl. Genet. 48 (3) 2007 . 235 – 260

Takahashi, H., Nirasawa, Y., Nagamine, Y., Tsudzuki, M. and Yamamoto, Y. (1998).
Genetic relationships among Japanese Native breeds of Chicken based on
microsatellite DNA polymorphisms. J. Her. 89, 543-546.

Tanka, T. and Rosenberg, M. M. (1952). Interrelationship of certain economic

characters in four strains of New Hampshire.Poult.Sc. 31:56-64

Tai, C. and Rouvier, R. (1998). Crossbreeding effect on sexual dimorphism of body

weight in intergeneric hybrids obtained between Muscovy and Pekin duck. Genet. Sel.
Evol. 30:163–170.

Tai, C., Rouvier, R.and Poivey J.P. (1989). Genetic parameters of some growth and egg
production traits in laying Brown Tsaiya (Anas Platyrhynchos).
Gen. Sel. Evol. 21. 377-384.

Teguia, A, Ngondjou, H. M., Defang, H. and Tchoumbone, J. (2007). Studies of the

live body weight and body characteristics of the African Muscovy duck.
Trop. Animal Health and Prod. 40:5-10.

Testik, A., Sarica, M. (1991). An investigation on the egg quality traits of Pekin
ducks. Intern. Congress of Poultry Husbandry, Istambul proceeding, 85-

Tikk, H. and Viht, I. (1988). Egg laying incubation properties of egg from different
Muscovy duck population means of increasing intensification and development of
energy saving technologies of egg and poultry production. 97-89.

Tinsley, H.E.A. and Brown, S. D. (2000). Hand book of applied multivariate statistics
and mathematical modelling. New York Academic Press 720.

Thompson, W. A. (1962). The problem of negative estimates of variance components

. Ann. Math.Stat.33:272.

Thompson, R. (1982). Estimation of realised heritability in a selected population using

mixed model methods. Genetic Sel. Evol. 18:475-484.

Thorpe, R. S. (1976). Biometric analysis of geographical variation and racial affinities

.Biol. Rev 51:407-452.

Thorpe, J .P .(1982). The molecular clock hypothesis: biochemical evolution , genetic

differentiation and systematic .Ann. Rev. Eco. System 13:139-168.

Tian-Fwu C., Yu-Tsai L. and Chin-Moo, P. (1998). Calcium requirements of cage feeding
laying Tsaiya duck. J. Taiwan Livestock Res. Sc. (1) : 27:33.
Townsend, C. W. (2002).The Muscovy duck. Auk 33 9-14

Tufvesson, M., Tufvesson, B., Schantz, T., Johansson, K., and Wilhelonson, M. (1999).
Selection for sexual male characters and their effect on other fitness-related traits in
white leghorn chicken. J. Anim Breed. Genet. 116 : 127 – 138.

Tuinen, M. V. and Hadley, E. A. (2004). Error in estimation of rate and time inferred from
the early amniote fossilsrecordand avian molecular clock. J. Mol.Evol.(55)267-276.

Tyler, C. (1961). Shell strength, its measurement and its relationship to other factors.
Br. Poult. Sc. 2: 3-19.

Van Bueningen, L.T.and Busch, R. H.( 1997). Genetic diversity among nort Americanspring
wheatcultivers.1 Analysis of the coefficient of percentage matrix .Crop Sc.
Wang, G. and Xu, L. (1989). Reproductive performance of white Muscovy duck .J .of
Fujian Agric. College 16(4) 320-324.

Waston, W. (1969). Early animal domestication in China. In Ucko. PJ and GW.52-59

Welsh, J. and McClelland, M. (1990). Fingerprinting genomes using PCR with

arbitrary primers. Nucleic Acids Res., 18: 7213-7218. Retrieved on 25 /6/2009.

Wessel, J P. H. (1962). Relationship between weight increase feed intake age at

sexual maturity and early egg production In New Hampshire pullets. Anim. Breeding

Wetmore, A. (1965). The birds of the republic of panama Smithsonian Misc. Coll vol.
Wiley, E. O. (1981) phylogenetics.The theory and practice of phylogenetic systematics.
Wiley New York 567pp.

Wilhan, J.G.K., Kubelik, A.R., Livak, K.J., Rafalski, J.A. , Tingey, S.V. (1990). DNA
Polymorphism Amplified by arbitrary primers are used as genetic markers
.Nucleic Acid Res. 18:6531-6535.

Williams, J.G.K., Kubelik, A.R., Livak, K.J., Rafalski, J.A. and. Tingey, S.V, (1990).
DNA polymorphisms amplified by arbitrary primers are useful as genetic markers.
Nucleic Acids Res. 18: 6531-6535.

Williams, J. G.,. Kubelik, A. R, Livak, K. J., Jarafalsk, I. and Tingey, S.V. (1990). DNA
polymorphisms amplified by arbitrary primers are useful as genetic markers. Nucleic
Acids Res. 18: 6,531-6,535.

Willin, S .E.and Erzsebet, M. (1997). Changing of body measurements and the correlation of
these with body weight from 0 to 16 weeks of age in geese. Allattenyesztes es
Takarmanyozas. 1997; 46: 409-414.

Wilson, R. T. (1991). Small Ruminant Production and the small ruminant genetic
resource in tropical Africa FAO Animal Production and Health Paper no 88 23pp.

Wimmers, K., Ponsuksili, S., Hardge, T., Walle – Zarate, A., Mathur, P.K. and Horst, P.
(2000). Genetic distinctness of African, Asian and South American Local chickens.
Anim. Genetic. 31: 159 – 165).

Woodyard, E. R. and Bolen, E.G. ( 1984). Ecological studies of Muscovy duck in

Mexico southwest Nat.29:453-465.

Wright, S. (1968). Evolution and the genetics of population.In Genetic and biometric
foundations .Chicago.Univ.of Chicago press.

Wu, C. K. and Han, Y. (1983). Feed efficiency and utilization of energy and protein of
broiler and different diet. Poult. Abst.10(7)165.

Wu, P., Jiang,T.X.,Suksaweang, S.,Widelitz,R.B.and Chung, C.M.(2004). Molecular shaping

of the beak. Science. 305: 1465-6.
Wucheng, B. (1988). The reach on the origin of house duck in China. In proceeding of the
Inter. Symposium on waterfowl production. The satellite conference. World Poultry
Cong.China.Pergahon press Oxford 125-129.

Xiao, T.F., Zheng, X.C.and Wu, Q.E. (2004). Genetic diversity of Fujian local duck
populations indifferent ecological type. Chinese J. of Applied Ecology, 15, 879–882.

Yahav, S., Luger, D., Cahaner, A. and Dotan, M. (1998).Thermoregulation inages:

implications for meat quality. Poultry Science naked-neck chickens subjected to
different ambient 80:418425. temperatures. British Poult. Sc. 39:133138.

Yakubu, A. (2009). An Assessment of Sexual Dimorphism in African Muscovy ducks

(Cairina moshata) using Morphological Measurements and Discriminant Analysis. In
proc. of 4th World Water Fowl Conf. Thrissur India.

Yakubu,. A.,. Ogah, D. M. and Idahor, K. O. ( 2009). Principal component analysis

of the morphostructural indices of white Fulani cattle.Trakia J .Sc.7(2)67-73.

Yeh, F.C., Yang, R.C, Boyle, T.B.J., Ye, Z.H and Mao, J.X. (1999). POPGENE, the
user- friendly shareware for population genetic analysis. Molecular Biology
and Biotechnology Centre, University of Alberta, Canada.
endix-9.pdf Retrieved on 23/5/2009.

Zaitoun, I. S., Tabbaa, M .J. and Bdour, S. (2005). Differentiation of native goat breeds of
Jordan on the basis of morphostructural characteristics. Small. Rum. Res. 56: 173 –
182.00 .

Zenatello, M. and Kiss, J. B. (2005). Biometrics and sex Identification of the Rose-coloured
starlig sturnus roseus. Ringing and Migration 22: 163-166.

Zeuner, F. E. (1963). The dog in a history of domesticated animal. Haper and Row New York
1963 7111.
Zhang, L.C, Wing, Z. H., XU, G.Y., Hou, Z.C. and Yang, N. (2005). Heritabilities and
genetics and phenotypic correlations of egg quality traits in brown egg dwarf layers.
Poult . Sc. 84: 1209 – 1213.

Zuk, K. M., Thornhill, R., Hyon, J.D., Johnson, K., Austad, S., Ligon, S.H., Thornhill,
W. and Costin, C. (1990). The role of male ornament and courtship behaviour in
female mate choice of red jungle fowl.Amer. Nat.136. 459-473.