doi:10.1111/j.1365-2052.2011.02296.

Characterization of the genetic diversity, structure and admixture of

British chicken breeds
S. Wilkinson*, P. Wiener*, D. Teverson†, C. S. Haley*‡ and P. M. Hocking*
*The Roslin Institute and Royal (Dick) School of Veterinary Studies, University of Edinburgh, Easter Bush, Midlothian EH25 9RG, UK. †17,
Old Road, BishopÕs Itchington, Leamington Spa, Warwickshire CV47 2RR, UK. ‡MRC Human Genetics Unit, Western General Hospital,
Crewe Road, Edinburgh EH4 2XU, UK

Summary The characterization of livestock genetic diversity can inform breed conservation initiatives.
The genetic diversity and genetic structure were assessed in 685 individual genotypes
sampled from 24 British chicken breeds. A total of 239 alleles were found across 30
microsatellite loci with a mean number of 7.97 alleles per locus. The breeds were highly
differentiated, with an average FST of 0.25, similar to that of European chicken breeds. The
genetic diversity in British chicken breeds was comparable to that found in European
chicken breeds, with an average number of alleles per locus of 3.59, ranging from 2.00 in
Spanish to 4.40 in Maran, and an average expected heterozygosity of 0.49, ranging from
0.20 in Spanish to 0.62 in Araucana. However, the majority of breeds were not in Hardy-
Weinberg Equilibrium, as indicated by heterozygote deficiency in the majority of breeds
(average FIS of 0.20), with an average observed heterozygote frequency of 0.39, ranging
from 0.15 in Spanish to 0.49 in Cochin. Individual-based clustering analyses revealed that
most individuals clustered to breed origin. However, genetic subdivisions occurred in several
breeds, and this was predominantly associated with flock supplier and occasionally by
morphological type. The deficit of heterozygotes was likely owing to a Wahlund effect
caused by sampling from different flocks, implying structure within breeds. It is proposed
that gene flow amongst flocks within breeds should be enhanced to maintain the current
levels of genetic diversity. Additionally, certain breeds had low levels of both genetic
diversity and uniqueness. Consideration is required for the conservation and preservation of
these potentially vulnerable breeds.
Keywords biodiversity, conservation, livestock, microsatellites, substructure.

(FAnGR) in cattle (Europe Cattle Genetic Diversity

Introduction
Livestock biodiversity is threatened by the marginalization
of breeds owing to agricultural intensification (FAO 2007),
and the less commercial breeds are frequently maintained
and managed by very few holders. Knowledge of the genetic
diversity and structure of breeds can inform sustainable
management initiatives and decision-making processes for
conservation of breed biodiversity. In the past decade there
have been several collaborative European-wide initiatives
on the characterization of farm animal genetic resources
Address for correspondence
S. Wilkinson, The Roslin Institute and Royal (Dick) School of Veterinary
Studies, University of Edinburgh, Easter Bush, Midlothian EH25 9RG,
UK.
E-mail: samantha.wilkinson@roslin.ed.ac.uk
Accepted for publication 24 August 2011
Consortium; http://ec.europa.eu/agriculture/publi/genres/
prog94_99_en.pdf), chickens (AVIANDIV; http://aviandiv.
tzv.fal.de/), goats (ECONOGENE; http://www.econogene.eu/),
pigs (PIGBIODIV; http://www.projects.roslin.ac.uk/pigbiodiv/)
and sheep (ECONOGENE; http://www.econogene.eu/).
At a national level, Britain has assessed its own farm
animal biodiversity by contributing samples to European
initiatives (e.g. pigs, sheep, cattle) (SanCristobal et al. 2006;
Lawson Handley et al. 2007; Laloë et al. 2010) and con-
ducting genetic studies on local populations (e.g. cattle,
pigs) (Wiener et al. 2004; Wilkinson et al. 2011). However,
in comparison with other British farm animal species, there
has been a noticeable lack of attention to poultry biodiver-
sity, with only one British chicken breed genotyped as part
of the European-wide initiative AVIANDIV (Hillel et al.
2003; DEFRA 2009).

552  2011 The Authors, Animal Genetics  2011 Stichting International Foundation for Animal Genetics, 43, 552–563
 Genetic structure of British chicken breeds 553

Britain has a large number of chicken breeds owing to a
long tradition of breed development for competitive showing
of fancy breeds (Roberts 1997; Hams 2004). This reservoir
of variability is a combination of old indigenous breeds,
foreign introduced breeds and more recently developed
breeds (Table 1). The immense diversity of British chicken
breeds can be observed through the size, shape and colour of
many morphological characteristics including body, comb,
feathers, skin and feet (Roberts 1997). The large pool of
British chicken breeds is potentially a major source of
diversity and, considering the absence of allelic diversity
amongst commercial chickens owing to bottlenecking early
in the development of the lines (Muir et al. 2008), it is
imperative that an assessment of British poultry genetic
resources is undertaken.
The objective of this study was to determine the levels of
genetic diversity within British chicken breeds, the genetic
structure of breeds and the levels of admixture in these
breeds, using the microsatellite marker panel recommended
by the Food and Agriculture Organization (FAO). The
characterization of the state of the genetic resources of
British chicken breeds will both inform management
initiatives and help to set priorities for conservation, as well
as contribute to the European perspective on FAnGR.

Table 1 Summary details of 24 chicken breeds.

Breed N1 Flocks2 Types3 Breed origin4 Introduction4 Development in Britain4 Status5

1 Appenzeller 30 9 3 Switzerland 1970s Endangered

2 Araucana 29 7 3 Chile 1930s
3 Brahma 26 6 3 China via USA 1800s At risk
4 Buff Orpington 25 10 1 Britain Buff Cochin, Dark Dorking, At risk
Hamburgh, Langshan (1800s)
5 Cochin 29 6 2 China 1800s At risk
6 Croad Langshan 28 8 1 China 1800s
(late)
7 Derbyshire Redcap 30 9 1 Britain Old indigenous
8 Dorking 26 7 3 Britain Old indigenous
9 Hamburgh 30 10 3 Britain Old indigenous
10 Indian Game 26 9 4 Britain Asil, Old English Endangered
Game, Malay (1800s)
11 Ixworth 27 6 1 Britain Jubilee Indian Game, Critical
White Sussex, White
Wyandotte (1930s)
12 Leghorn (coloured) 29 7 4 Italy via USA 1800s
13 Lincolnshire Buff 30 6 1 Britain Orpington Buff, Dorking, Critical
Cochin (1980s)
14 Maran 30 6 2 France 1900s (early)
15 Marsh Daisy 28 6 1 Britain Old English Game, Sicilian Endangered
Buttercup, Malay,
Hamburgh (early 1900s)
16 Norfolk Grey 19 8 1 Britain Birchen Game, Silver Critical
Duckwing, Leghorn (1930s)
17 Old English 28 4 1 Britain Old indigenous Endangered
Pheasant Fowl
18 Rhode Island Red 30 9 1 USA 1900s
19 Scots Dumpy 28 7 1 Britain Old indigenous
20 Scots Grey 29 6 1 Britain Old indigenous At risk
21 Silkie 29 6 3 Asia 1600s
22 Spanish 28 7 1 Mediterranean 1700s Critical
23 Light Sussex 30 6 1 Britain Southern England (1800s)
24 Sussex 30 6 3 Britain Southern England (1800s)

1
Sample size.
2
Number of flocks that contributed to the total sample size of a breed.
3
Number of morphological types/varieties present in the total sample size of a breed.
4
Information taken from the following sources (Roberts 1994; Roberts 1997; Vorwald Dohner 2001; Hams 2004).
5
Population size status measured by the number of breeding females as follows: Critical = 100, Endangered = 200, Vulnerable = 300, At risk = 500
(DEFRA 2010) (The adopted quantification was previously used by RBST to categorize breed status).

 2011 The Authors, Animal Genetics  2011 Stichting International Foundation for Animal Genetics, 43, 552–563
554 Wilkinson et al.
Materials and methods
Data sampling and DNA extraction
Owners of flocks of known provenance were approached
and asked to provide up to 12 fertile hatching eggs. A total
of 1416 samples that were described by 54 breed types
(different colours or strains) were collected from 92 flock
owners. These primary samples were grouped into 27 rec-
ognized breeds. The number of samples per flock was limited
to a maximum of six and the number of flocks to a mini-
mum of five to obtain a similar sample size across breeds for
genetic analysis. Thus, a total of 24 breeds with 19–30
individuals from 4 to 10 flocks were chosen to represent
British poultry genetic resources (Table 1). Eggs were
incubated for 7 days, at which time the embryos were
harvested. The embryo was divided and the head and body
were stored separately at )80 C. Data collection took place
over 3 years (2007–2009). The DNA was extracted from
thawed heads using the Maxell 16 system.
Microsatellite genotyping
Individual samples were genotyped at 30 microsatellite loci.
Further information on the microsatellite loci can be found
at http://aviandiv.tzv.fal.de/primer_table.html. Genotyping
was conducted using standard laboratory procedures for
PCR and the resulting products were visualized on an
Applied Biosystems 3730xl genetic analyser (Applied Bio-
systems/Hitachi, Applera). GENEMAPPER v3.5 software
(Applied Biosystems, Applera) was used to estimate frag-
ment sizes by comparing them to an internal size standard
(LIZ500; Applied Biosystems).
Individuals with >20% missing data (n = 20) were dis-
carded. Preliminary individual-based clustering analysis
revealed that a number of individuals did not cluster to
breed origin. Individuals were traced back to their supplier.
If the supplier also provided samples of the breed that these
individuals did cluster with, then these individuals were
removed from the dataset (n = 4). After data cleaning the
total sample size included 674 individuals (Table 1).
Marker polymorphism, within and amongst
population diversity
The total number of alleles, allele frequencies, observed het-
erozygosity (HO) and expected heterozygosity (HE) were esti-
mated using FSTAT 2.9.3 (http://www.unil.ch/izea/softwares/
fstat.html) (Goudet 1995). The extent of genotypic linkage
disequilibrium (LD) between pairs of loci in each breed was
tested by performing probability tests using GENEPOP 4.0.7
(http://kimura.univmontp2.fr/~rousset/Genepop.htm) (Rousset
2008). Tests for deviations from Hardy-Weinberg equilibrium
(HWE) across all loci for each population were performed in
GENEPOP 4.0.7, applying the Ôexact testÕ and using the Markov
 2011 The Authors, Animal Genetics  2011 Stichting International Foundation for Animal Genetics, 43, 552–563
chain algorithm with default settings to calculate P-values
(Guo & Thompson, 1992). Weir & CockerhamÕs (1984)
extension of WrightÕs within-population inbreeding coeffi-
cient FIS was calculated in FSTAT 2.9.3. The significance of FIS
was tested by 10 000 permutations for each population.
Population genetic differentiation was examined using Weir &
CockerhamÕs unbiased estimator of WrightÕs fixation index
(FST) (Weir & Cockerham 1984) computed using FSTAT 2.9.3
(Goudet 1995). ReynoldsÕ genetic distance (Reynolds et al.
1983) was calculated between pairs of breeds using allele
frequencies, and consensus statistical support was calculated
from 1000 bootstrap replicates using PHYLIP 3.67 (http://
evolution.genetics.washington.edu/phylip.html) (Felsenstein
1989). An unrooted neighbour-joining cladogram was con-
structed from the genetic distance matrix of values for all pairs
of breeds using the R package APE (Paradis et al. 2004).
Clustering of individuals to populations
Population structure and admixture were investigated using
the Bayesian genotypic clustering method implemented in
BAPS 5.2 (Corander et al. 2008). BAPS 5.2 uses a greedy sto-
chastic optimization algorithm to first assign individuals to a
population at a given K value and then to estimate the level
of admixture in each individual (the membership probabili-
ties for each individual being assigned to one or more clus-
ters, measured by q) (Corander et al. 2008). It operates by
maximizing HWE and linkage equilibrium in the inferred
clusters. BAPS 5.2 was implemented for 2 £ K £ 35 with 10
runs at each K value. Genetic clustering solutions were
visualized in the statistical package R (Team, 2010).
Individual pairwise shared allele genetic distances (DSA)
were estimated using the proportion of shared alleles
(Bowcock et al. 1994), implemented in MICROSAT 1.5b
(http://hpgl.stanford.edu/projects/microsat/) (Minch et al.
1997). Consensus statistical support was calculated from
1000 bootstrap replicates in PHYLIP 3.67 (http://
evolution.genetics.washington.edu/phylip.html) (Felsen-
stein 1989). An unrooted neighbour-joining cladogram was
constructed from the genetic distance matrix of values for
all pairs of individuals using the R package APE (Paradis
et al. 2004).
Breed genetic contributions
A breedÕs contribution to diversity was quantified following
Ollivier & Foulley (2005), a commonly implemented method
in livestock breed conservation analyses (e.g. pig breeds)
(Laval et al. 2000). The contribution to between-breed
diversity (CB), using the Weitzmann method (Weitzman
1992), was estimated from the ReynoldsÕ pairwise genetic
distance matrix. CB was estimated as follows: CB = 1 – V(S/k)/
V(S) where V(S) is the total genetic distance of the whole set
of breeds considered, and V(S/k) is the total genetic distance
of the set excluding breed k. The relative decrease in genetic
 Genetic structure of British chicken breeds 555

distance with the removal of a breed (Vk) is the contribution
of breed k to between-breed diversity and can be seen as a
measure of the genetic uniqueness of a breed. The contri-
bution to within-breed diversity (CW) was estimated using
average expected heterozygosity (HE) values of breeds as
CW = 1 – HE(S/k)/HE(S), where HE(S) is the average
expected heterozygosity of all the breeds, and HE(S/k) is the
average expected heterozygosity excluding breed k. The
relative increase or decrease in average expected heterozy-
gosity with the removal of a breed [HE(k)] is the contribution
of breed k to total within-breed diversity. Aggregate diversity
(D) was estimated as D = FSTCB + (1)FST)CW.
Results
Genetic diversity within and amongst breeds
Every breed had at least one individual that failed to amplify
at one or more loci and, in total, 202 individuals failed to
amplify at one or more loci. Of the loci, MCW0123 had the
most missing data at 6.1%, followed closely by MCW0020
with 5.7% missing genotypes. A total of 239 alleles were
identified across the 30 microsatellite loci with a mean
number of 7.97 alleles per locus. The number of alleles per
locus ranged from 2 (MCW0103, MCW0284, MCW0098)
to 24 (LEI092).
The average number of alleles per locus ranged from 2.00
in Spanish to 4.40 in Maran, with an average across all the
breeds of 3.66 alleles per locus (Table 2). Not all markers
were found to be polymorphic in each of the 24 breeds; the
number of monomorphic loci per breed ranged from 0 to 10
(Table 2). A total of 33 breed-specific (private) alleles were
detected in 15 breeds (Table 2). Twenty-two of the private
alleles possessed frequencies <0.1%. The remaining 11
breed-specific private alleles were found in Brahma (1), Buff
Orpington (1), Cochin (1), Croad Langshan (1), Indian
Game (1), Leghorn (1), Lincolnshire Buff (1), Marsh Daisy
(2), Silkie (1) and Spanish (1). The average expected het-
erozygosity (HE) over all loci ranged from 0.20 in Spanish to
0.62 in Araucana, while observed heterozygosity (HO)

Table 2 Genetic diversity measures for 24 chicken breeds.

Breed N1 n2 HO3 HE4 Monomorphic5 Private alleles6 LD7

1 Appenzeller 94 3.13 0.31 0.43 2 0 0

2 Araucana 129 4.30 0.47 0.62 0 3 7
3 Brahma 124 4.13 0.43 0.53 0 4 4
4 Buff Orpington 118 3.93 0.51 0.55 0 1 1
5 Cochin 124 4.13 0.49 0.56 0 3 3
6 Croad Langshan 107 3.57 0.42 0.49 1 3 1 (1)
7 Derbyshire Redcap 94 3.13 0.35 0.42 1 0 1 (1)
8 Dorking 107 3.57 0.36 0.47 2 0 4
9 Hamburgh 98 2.97 0.22 0.42 4 0 5
10 Indian Game 100 3.33 0.44 0.49 0 1 0
11 Ixworth 90 3.00 0.41 0.43 0 0 1
12 Leghorn 119 3.97 0.37 0.54 0 2 43 (2)
13 Lincolnshire Buff 105 3.50 0.41 0.48 1 2 1
14 Maran 132 4.40 0.47 0.58 0 3 3
15 Marsh Daisy 96 3.20 0.37 0.49 1 3 2 (1)
16 Norfolk Grey 89 2.97 0.40 0.50 1 2 1 (1)
17 Old English Pheasant Fowl 105 3.50 0.40 0.48 1 0 1
18 Rhode Island Red 121 4.03 0.41 0.57 0 0 13
19 Scots Dumpy 126 4.20 0.44 0.51 0 1 2
20 Scots Grey 95 3.17 0.32 0.40 2 0 2
21 Silkie 115 3.83 0.45 0.56 0 1 14 (1)
22 Spanish 60 2.00 0.15 0.20 10 1 0
23 Light Sussex 121 4.03 0.48 0.54 0 0 5
24 Sussex 117 3.90 0.38 0.57 0 3 34 (4)
Average 3.59 0.39 0.49

1
Total number of alleles.
2
Average number of alleles.
3
Observed heterozygote frequency.
4
Expected heterozygote frequency.
5
Number of monomorphic loci.
6
Number of breed-specific private alleles.
7
Number of significant tests of linkage disequilibrium (LD) out of 435 possible pairs of loci [after Bonferroni correction, P < 0.000005, Rice (1989)],
number in brackets indicate tests that occurred between loci pairs found on the same chromosome.

 2011 The Authors, Animal Genetics  2011 Stichting International Foundation for Animal Genetics, 43, 552–563
556 Wilkinson et al.

(a) (b)

0.5

0.1 0.2 0.3 0.4 0.5

0.3
FIS

FIS
0.1
–0.1

Norfolk Grey
Old English Pheasant Fowl

Scots Dumpy
Derbyshire Redcap

Scots Grey
Appenzeller
Araucana
Brahma

Maran
Cochin
Croad Langshans

Indian Game

Lincolnshire Buff

Marsh Daisy

Rhode Island Red

Silkie
Spanish
Dorking

Light Sussex
Leghorn
Buff Orpington

Ixworth
Hamburgh

Sussex

–0.1

Sussex (Speckled)
Hamburgh

Maran
Rhode Island Red
Leghorn
Figure 1 Within-population inbreeding coefficients (FIS): (a) breed average and 95% confidence intervals and (b) average and 95% confidence
intervals for the largest subpopulations of breeds exhibiting genetic subdivision.

varied from 0.15 in Spanish to 0.49 in Cochin (Table 2).

Average estimates of HE and HO over all loci and breeds
were 0.49 and 0.39 respectively.
There was a large deficiency of heterozygotes compared
with HWE expectations across the multilocus dataset. Of the
24 British chicken breeds, only two breeds, Buff Orpington
and Ixworth, were in HWE (P > 0.05). This was reflected in
the within-population FIS estimates where, with the excep-
tion of these two breeds, all other breeds had significant
positive FIS estimates, ranging from 0.12 in Light Sussex to
0.48 in Hamburgh (Fig. 1a).
After Bonferroni correction, significant LD was found in
148 out of 10 440 pairs of loci. The number of locus-pairs
with significant LD ranged from none in three breeds (Ap-
penzeller, Indian Game and Spanish) to 43 occurrences in
Leghorn (Table 2). Of the significant cases, 137 involved
pairs of loci situated on different chromosomes. The 11
significant cases of pairs of loci on the same chromosome
occurred in Croad Langshan (1), Derbyshire Redcap (1),
Leghorn (2), Marsh Daisy (1) and Norfolk Grey (1), Silkie (1)
and Sussex (4).
The genetic differentiation (FST) between pairs of breeds
ranged from 0.1 (Rhode Island Red vs. Sussex and Brahma
vs. Cochin) to 0.52 (Ixworth vs. Spanish), with average FST
across breeds ranging from 0.18 for Araucana to 0.42 for
Spanish. Overall breed genetic differentiation (FST) was 0.25
(Table 3). ReynoldsÕ pairwise genetic distance ranged from
0.34 between Brahma and Cochin to 0.72 between Ixworth
or Lincolnshire Buff and Spanish. Average pairwise genetic
distance across all breeds ranged from 0.44 for Araucana to
0.66 for Spanish and was positively correlated with average
breed FST (SpearmanÕs rank correlation = 0.99, P £ 0.05)
(Table 3). The phylogenetic reconstruction of breed rela-
tionships is shown in Fig. 2 (bootstrap support >50% indi-
cated). There was overall low bootstrap support of genetic
relationships between the chicken breeds, with two highly
supported breed pairs: Cochin and Brahma, Lincolnshire
Buff and Buff Orpington.
Clustering of individuals to populations
The BAPS results showed that the likelihood of the data was
best described by 30–35 genetic clusters as can be seen in
Fig. 3, where the log-likelihood values start to plateau from
K = 30. The BAPS clustering solutions are presented in
Fig. 4. Populations that split to form independent clusters at
lower K values can be interpreted as being relatively
genetically distinct (Rosenberg et al. 2001). Breeds split to
form their own distinct genetic clusters in the following
order: Silkie, Indian Game, Scots Grey, Buff Orpington and
Lincolnshire Buff, Croad Langshan, Appenzeller, Spanish,
Marsh Daisy, Ixworth and Scots Dumpy at K = 6, 7, 10,
11, 12, 13, 14, 15, 16 and 18 respectively. A number of
breeds clustered together, indicating genetic affinities. Bra-
hma and Cochin formed a distinct cluster at K = 8 and
remained together until K = 22. From K = 5–10, Buff
Orpington and Lincolnshire Buff formed their own cluster,
after which they split into two clusters. Derbyshire Redcap,
Dorking, Hamburgh and Old English Pheasant Fowl clus-
tered together until K = 19, whereupon Derbyshire Redcap
split to form an independent cluster, followed by Old English
Pheasant Fowl at K = 25 and Dorking and Hamburgh
at K = 28. Notable genetic admixture within individuals
occurred in Araucana, Dorking, Leghorn and Light Sus-
sex breeds, as evidenced by the proportion of genomes
assigned to two or more clusters (0.2 < q < 0.8) at various
K values.
The phylogenetic reconstruction of 674 individuals using
the proportion of shared allele distance measure is shown in
Fig. 5. There was no bootstrap support for genetic rela-
tionships amongst the chicken breeds. There was high
bootstrap support (>50%) for the Lincolnshire Buff, Scots
Grey and Spanish clades. Eleven chicken breeds formed
single distinct genetic units where all individuals clustered
to breed origin, although the bootstrap support was weak:
Buff Orpington, Cochin, Croad Langshan, Derbyshire Red-
cap, Indian Game, Ixworth, Marsh Daisy, Norfolk Grey,

 2011 The Authors, Animal Genetics  2011 Stichting International Foundation for Animal Genetics, 43, 552–563
 Table 3 Population genetic differentiation amongst 24 chicken breeds.

Breed 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 Reynolds FST

1 Appenzeller 0.44 0.60 0.58 0.58 0.60 0.52 0.46 0.48 0.57 0.61 0.43 0.62 0.56 0.52 0.53 0.47 0.53 0.53 0.55 0.55 0.69 0.57 0.52 0.54 0.28
2 Araucana 0.17 0.43 0.46 0.42 0.46 0.48 0.41 0.45 0.45 0.47 0.35 0.51 0.40 0.38 0.41 0.43 0.38 0.41 0.46 0.46 0.61 0.42 0.36 0.44 0.18
3 Brahma 0.35 0.17 0.45 0.34 0.46 0.59 0.55 0.60 0.56 0.56 0.52 0.52 0.45 0.53 0.51 0.57 0.44 0.49 0.60 0.46 0.70 0.48 0.46 0.52 0.26
4 Buff Orpington 0.33 0.19 0.18 0.41 0.48 0.57 0.52 0.57 0.55 0.55 0.50 0.45 0.43 0.52 0.52 0.55 0.42 0.52 0.59 0.50 0.70 0.46 0.43 0.51 0.25
5 Cochin 0.32 0.16 0.10 0.15 0.42 0.58 0.53 0.57 0.54 0.54 0.48 0.48 0.39 0.47 0.50 0.56 0.41 0.46 0.57 0.46 0.68 0.42 0.43 0.49 0.23
6 Croad Langshan 0.35 0.19 0.20 0.21 0.16 0.58 0.51 0.57 0.53 0.49 0.52 0.56 0.40 0.47 0.52 0.55 0.43 0.43 0.59 0.48 0.66 0.40 0.45 0.50 0.24
7 Derbyshire Redcap 0.26 0.21 0.34 0.31 0.32 0.33 0.44 0.46 0.57 0.60 0.48 0.61 0.55 0.53 0.51 0.46 0.56 0.53 0.56 0.54 0.63 0.53 0.53 0.54 0.28
8 Dorking 0.19 0.15 0.29 0.25 0.26 0.25 0.17 0.38 0.47 0.55 0.42 0.56 0.48 0.46 0.46 0.40 0.49 0.47 0.53 0.51 0.62 0.47 0.48 0.49 0.22
9 Hamburgh 0.21 0.18 0.34 0.31 0.31 0.30 0.20 0.12 0.54 0.62 0.43 0.61 0.54 0.47 0.47 0.40 0.54 0.49 0.53 0.51 0.57 0.54 0.52 0.52 0.25
10 Indian Game 0.31 0.18 0.29 0.28 0.28 0.27 0.31 0.20 0.28 0.54 0.50 0.59 0.49 0.52 0.51 0.51 0.51 0.50 0.57 0.53 0.69 0.48 0.50 0.53 0.27
11 Ixworth 0.36 0.21 0.30 0.29 0.28 0.23 0.35 0.29 0.36 0.28 0.53 0.60 0.51 0.53 0.60 0.58 0.54 0.50 0.59 0.55 0.72 0.45 0.50 0.55 0.30
12 Leghorn 0.17 0.10 0.26 0.23 0.22 0.26 0.21 0.15 0.16 0.23 0.27 0.55 0.46 0.45 0.44 0.46 0.47 0.44 0.46 0.50 0.60 0.46 0.40 0.47 0.21
13 Lincolnshire Buff 0.37 0.25 0.25 0.18 0.21 0.30 0.36 0.30 0.36 0.33 0.35 0.29 0.51 0.55 0.57 0.59 0.49 0.57 0.63 0.55 0.72 0.54 0.51 0.56 0.30
14 Maran 0.30 0.14 0.19 0.16 0.13 0.14 0.28 0.21 0.27 0.22 0.24 0.20 0.24 0.45 0.45 0.52 0.38 0.45 0.55 0.47 0.65 0.37 0.38 0.47 0.21
15 Marsh Daisy 0.25 0.13 0.27 0.26 0.21 0.20 0.27 0.19 0.21 0.25 0.27 0.18 0.29 0.18 0.47 0.49 0.44 0.44 0.50 0.49 0.63 0.45 0.43 0.49 0.22
16 Norfolk Grey 0.26 0.14 0.24 0.25 0.23 0.26 0.24 0.19 0.20 0.24 0.35 0.17 0.31 0.18 0.20 0.47 0.45 0.47 0.50 0.48 0.63 0.46 0.45 0.50 0.23
17 OEFP 0.21 0.17 0.31 0.29 0.30 0.29 0.20 0.14 0.14 0.24 0.32 0.20 0.34 0.25 0.22 0.20 0.52 0.50 0.53 0.50 0.59 0.52 0.50 0.51 0.24
18 Rhode Island Red 0.27 0.13 0.18 0.16 0.15 0.16 0.30 0.22 0.27 0.24 0.27 0.20 0.22 0.13 0.17 0.18 0.25 0.46 0.55 0.49 0.67 0.40 0.34 0.47 0.21
19 Scots Dumpy 0.27 0.15 0.23 0.25 0.20 0.17 0.27 0.20 0.21 0.23 0.24 0.17 0.31 0.18 0.18 0.20 0.23 0.19 0.51 0.47 0.62 0.40 0.43 0.48 0.22
20 Scots Grey 0.28 0.19 0.34 0.34 0.31 0.33 0.30 0.26 0.26 0.31 0.35 0.19 0.38 0.29 0.24 0.23 0.26 0.29 0.24 0.54 0.69 0.53 0.51 0.55 0.29
21 Silkie 0.29 0.19 0.20 0.23 0.19 0.21 0.28 0.24 0.24 0.26 0.29 0.23 0.28 0.21 0.23 0.21 0.24 0.22 0.20 0.28 0.67 0.46 0.47 0.50 0.24
22 Spanish 0.47 0.36 0.48 0.48 0.45 0.43 0.38 0.38 0.31 0.47 0.52 0.35 0.51 0.41 0.38 0.40 0.34 0.43 0.37 0.46 0.44 0.67 0.66 0.66 0.42
23 Light Sussex 0.32 0.16 0.21 0.19 0.16 0.14 0.26 0.20 0.28 0.21 0.18 0.20 0.28 0.12 0.19 0.19 0.26 0.15 0.14 0.26 0.19 0.43 0.36 0.47 0.21
24 Sussex 0.25 0.11 0.19 0.17 0.17 0.18 0.26 0.21 0.25 0.24 0.23 0.14 0.24 0.13 0.16 0.17 0.23 0.10 0.16 0.25 0.21 0.42 0.11 0.46 0.20

 2011 The Authors, Animal Genetics  2011 Stichting International Foundation for Animal Genetics, 43, 552–563
The upper diagonal contains the estimates of pairwise genetic distance between breeds estimated by ReynoldsÕ genetic distance. The lower diagonal contains the pairwise genetic differentiation between
breeds estimated using Weir & CockerhamÕs FST (Weir & Cockerham 1984). The two columns to the far right of the table represent average breed ReynoldsÕ genetic distance and average breed FST.
Genetic structure of British chicken breeds
557
558 Wilkinson et al.

Spanish Norfolk Grey) and Scots Grey (2 clustered with Old English
Pheasant Fowl) (Fig. 5). These misassignments were
mirrored by the BAPS results, and the individuals appeared
admixed for the following: Appenzeller (1), Brahma (1) and
Derbyshire Redcap
Indian Game Old English Pheasant Fowl (2) (Fig. 4). The misassignments
Norfolk Grey
Old English that were not admixed could be the result of mislabelling
Pheasant Fowl
Ixworth Scots Dumpy Hamburgh
during the data collection process.
The remaining breeds exhibited extensive genetic sub-
Dorking
division, with four groups each detected in Dorking and
Light Light Sussex, three groups each detected in Araucana,
Sussex Scots
Croad Grey Hamburgh, Rhode Island Red and Sussex, and two groups
Langshan
Leghorn each detected in Leghorn and Maran. BAPS concurred that
903 664
960
there was extensive genetic subdivision in Araucana, Leg-
Buff
Orpington horn, Maran, Rhode Island Red and Sussex. In addition, the
Appenzeller
BAPS split both Cochin and Marsh Daisy into two clusters
Lincolnshire Araucana Marsh Daisy
Buff
and Silkie into three clusters (Fig. 4).
Brahma
Cochin Maran
Sussex
Rhode Island Red
Silkie Breed genetic contributions

Figure 2 Phylogenetic reconstructions of the chicken breeds using Estimates of the contribution of breeds to genetic variation
ReynoldsÕ genetic distance. Bootstrap support values >50% are are presented in Table 4. Contribution to between-breed
indicated. diversity ranged from 3.17 in Cochin and Sussex to 6.71 in
Spanish and was negatively correlated with the genetic CW
(SpearmanÕs rank correlation = )0.70, P £ 0.05), with CW
values ranging from )2.67 in Spanish to 1.13 in Araucana.
–50 000 –48 000 –46 000 –44 000 –42 000 –40 000

Aggregate diversity ranged from )0.33 in Spanish to 1.67

in Araucana. As the calculation of D was weighted by the
FST estimate, greater weight was applied to CW than CB
and, as a result, the breed estimates of D essentially mirrored
those of CW (SpearmanÕs rank correlation = 0.94, P £
Mean likelihood

0.05).

Discussion

Genetic diversity within breeds

0 5
Figure 3 Plot of the likelihood output of BAPS with increasing K value.
Points indicate the average likelihood estimated across 10 runs for
each K value.
Scots Dumpy, Silkie and Spanish (Fig. 5, all designated as
yellow in various shapes).
In the phylogenetic reconstruction of individuals, five
other breeds formed single distinct genetic units, with the
exception of a couple of individuals: Appenzeller (1 indi-
vidual clustered with a mixed clade), Brahma (1 clustered
with mixed clade), Lincolnshire Buff (1 clustered with Croad
Langshan), Old English Pheasant Fowl (2 clustered with
The microsatellite markers revealed that 25% of the genetic
variation across individuals in this dataset could be ascribed
to between-breed differences. This was evidenced by the
estimated average number of alleles and expected hetero-
10 15 20 25 30 35 zygosity within the breeds (Table 2), with levels comparable
K to those reported in other European chicken breeds (Gran-
evitze et al. 2007; Berthouly et al. 2008; Bodzsar et al.
2009). Extensive genetic variation within breeds is a stan-
dard observation in livestock species (Bruford et al. 2003).
However, an exception, in this study at least, was the
Spanish breed, which exhibited a considerably lower genetic
diversity than other breeds (Table 2). A well-established
breed in Britain (Table 1), the Spanish has an extensive
white face (Roberts 1997; Hams 2004). Very limited ge-
netic introgression to maintain this physically distinctive
characteristic, as well as genetic drift, enhanced by the
small population size (Table 1), may have led to the fix-
ation of numerous loci and low genetic diversity in this
breed.

 2011 The Authors, Animal Genetics  2011 Stichting International Foundation for Animal Genetics, 43, 552–563
 Genetic structure of British chicken breeds 559

20

25

30

35
Appenzeller
Araucana
Brahma
Buff Orpington
Cochin
Croad Langshans
Derbyshire Redcap
Dorking

Indian Game
Ixworth
Leghorn
Lincolnshire Buff
Maran
Marsh Daisy
Norfolk Grey
Old English Pheasant Fowl
Rhode Island Red
Scots Dumpy
Scots Grey
Silkie
Spanish
Light Sussex
Sussex
Hamburgh

Figure 4 Individual assignment based on BAPS analysis at various values of K. Histograms demonstrate the proportion of each individualÕs genome that
originated from each of 24 populations. Each individual is represented by a vertical line corresponding to its membership coefficient (q).

Island Red and Sussex) were used to investigate whether the

Heterozygote deficiency within breeds
Wahlund effect was a cause for the observed heterozygote
The observed heterozygosity within breeds was lower than deficiencies and high-positive within-population FIS esti-
expected, assuming random union of gametes, leading to mates. These breeds were chosen because both individual
significant deviations from HWE in the majority of the clustering methods confirmed genetic substructure (Figs 4 & 5)
British chicken breeds. Consequently, the estimates of and there was extensive LD within each breed (Table 2).
within-population inbreeding coefficients (Fig. 1a) were The genetic subpopulations defined by BAPS were used and
considerably higher in the British chicken breeds than those the largest genetic subpopulation from each breed was
reported in other European chicken breeds (Granevitze et al. extracted for further analysis (K = 30, Fig. 4). Re-estima-
2007; Berthouly et al. 2008; Bodzsar et al. 2009; Zanetti tion of HE and HO showed a decrease in positive FIS and no
et al. 2010). deviations from HWE proportions within each of the largest
Several non-mutually exclusive hypotheses could subgroups of the following breeds: Sussex (Speckled mor-
account for the observed heterozygote deficiency: null phological variety), Maran and Rhode Island Red (Fig. 1b),
alleles, Wahlund effect (pooling of distinct genetic popula- indicating that the Wahlund effect was a likely cause of
tions) and mating between close relatives. The presence of observed heterozygote deficit in those breeds.
null alleles would cause a locus-specific heterozygote deficit It was not possible to assess the influence of mating of
across all breeds, which was not observed in any loci. close relatives owing to the absence of a pedigree. Although
Moreover, the microsatellites used in this study are a com- this practice may be a contributing factor, the principal
monly adopted FAO-recommended panel, and the influence cause of heterozygote deficit is likely to be genetic sub-
of null alleles has not previously been reported (Granevitze structuring within breeds, as this was evident in many
et al. 2007, 2009; Bodzsar et al. 2009). breeds (Figs 4 & 5).
The microsatellite markers revealed genetic substructure In contrast to the results reported here, Lawson Handley
in many of the breeds (Figs 4 & 5). Indeed, the extensive et al. (2007) concluded that livestock breeds are primarily
genotypic LD observed in certain breeds (Table 2) could be homogenous genetic populations that rarely deviate from
owing to differences in allele frequencies amongst geneti- within-breed HWE proportions. Indeed, no evidence for
cally differentiated populations within breeds (Nei & Li substructure was found within Hungarian chicken breeds
1973). Five breeds (Hamburgh, Leghorn, Maran, Rhode (Bodzsar et al. 2009). A broad study on European chicken

 2011 The Authors, Animal Genetics  2011 Stichting International Foundation for Animal Genetics, 43, 552–563
560 Wilkinson et al.
Derbyshire Redcap
Leghorn Scots Grey
Marsh
Daisy
Norfolk Grey
Rhode Island
Red
* *
Sussex
Ixworth
*
Light
Sussex
Buff Orpington
Lincolnshire Buff
Maran
Brahma
Figure 5 Phylogenetic reconstructions of individual samples. Bootstrap support values >50% are indicated with an asterisk. Breeds where all
individuals clustered to origin are designated as yellow (various symbols). All other breeds had a specific colour such that individuals that did not
cluster to origin can be traced on the phylogenetic tree.
breeds reported substantial heterozygote deficits, but only in
German fancy breeds (Granevitze et al. 2007). In European
cattle and pig breeds, deviations were limited to a quarter of
the sampled populations (SanCristobal et al. 2006; Li et al.
2007). However, there are exceptions to this pattern. In
Iberian pig and European sheep breeds, extensive observed
heterozygote deficit was observed and primarily attributed
to the Wahlund effect (Fabuel et al. 2004; Lawson Handley
et al. 2007).
Genetic substructure within breeds
In general, European chicken breeds have been observed to
be distinct homogenous genetic populations with little evi-
dence of substructure within breeds (Bodzsar et al. 2009;
Zanetti et al. 2010). In contrast, as mentioned above, many
of the British chicken breeds exhibited extensive genetic
substructure (Figs 4 & 5). The patterns of genetic subdivi-
sion within the Leghorn and Sussex breeds were associated
with morphological type. However, genetic subdivision
could not be explained by morphological type in the
Araucana, Buff Orpington, Dorking, Hamburgh, Maran,
Rhode Island Red and Light Sussex. Instead, the observed
genetic substructuring in these breeds reflected suppliers,
 2011 The Authors, Animal Genetics  2011 Stichting International Foundation for Animal Genetics, 43, 552–563
Appenzeller
Dorking
Old English
Pheasant Fowl
Spanish
Hamburgh
Silkie
Scots
Dumpy
Indian
Game
Croad Araucana
Langshan
Cochin
whereby all individuals from a particular flock were genet-
ically separated from the rest of the same chicken breed.
Such genetic substructure associated with flock supplier has
been observed in a few other chicken breeds (Rosenberg
et al. 2001; Zanetti et al. 2010).
The division of breeds into genetic subpopulations (as
detected by individual clustering methods) appears to be the
exception rather than the rule in livestock species. The
extent of genetic substructure is generally limited to one or
two exceptional breeds [e.g. horse (Glowatzki-Mullis et al.
2006), chicken (Zanetti et al. 2010) and pig (Wilkinson
et al. 2011)]. Certain management practices within these
breeds (e.g. restricted gene flow between farms through
geographical isolation or line breeding) could have pro-
duced subtle genetic heterogeneity amongst flocks within
breeds (Rosenberg et al. 2001).
Genetic differentiation amongst breeds
The genetic differentiation amongst the British chicken
breeds was high (Table 3, Fig. 2), with levels comparable to
those of other European local chicken breeds (Berthouly et al.
2008; Bodzsar et al. 2009; Zanetti et al. 2010). By compari-
son, genetic differentiation amongst breeds is relatively low in
 Genetic structure of British chicken breeds 561

Table 4 Genetic contributions of the 24 chicken breeds to diversity. indigenous breeds (Table 1). These breeds are claimed to be
1 2 3 closely related and may have contributed to the breed
Breed CB CW D
improvement of each other (Hams 2004).
1 Appenzeller 4.81 )0.55 0.79 Regarding the second group, it has been suggested that
2 Araucana 3.28 1.13 1.67 Asian breeds may have been used to improve the Sussex
3 Brahma 3.61 0.25 1.09 breeds (Vorwald Dohner 2001). The Asian breeds also
4 Buff Orpington 4.12 0.51 1.41
influenced the development of Buff Orpington and Lincoln-
5 Cochin 3.17 0.60 1.24
shire Buff (Table 1) (Hams 2004). In turn, Sussex contri-
6 Croad Langshan 4.23 )0.02 1.04
buted to the development of the white-feathered Ixworth
7 Derbyshire Redcap 4.91 )0.64 0.75
8 Dorking 3.51 )0.20 0.73
(Table 1) (Roberts 1994). Cross-breeding of the Light Sussex
9 Hamburgh 3.51 )0.64 0.40 and Rhode Island Red in the 1940s became the basis of
10 Indian Game 4.91 )0.02 1.21 commercial poultry enterprises in Britain (Hams 2004) and
11 Ixworth 5.07 )0.55 0.86 could have had an effect on non-commercial stocks of these
12 Leghorn 3.75 0.42 1.25 breeds. Before the importation of Maran to Britain, the breed
13 Lincolnshire Buff 5.12 )0.12 1.19 was improved using, amongst other breeds, Croad Lang-
14 Maran 3.41 0.78 1.44 shan stock (A. Sheppy, personal communication).
15 Marsh Daisy 4.37 )0.02 1.08 It was not possible to explain the genetic relationships
16 Norfolk Grey 4.37 0.07 1.16
amongst the remaining breeds (Appenzeller, Araucana,
17 Old English Pheasant Fowl 3.84 )0.11 0.88
Indian Game, Leghorn, Marsh Daisy, Norfolk Grey, Silkie,
18 Rhode Island Red 3.59 0.69 1.42
Scots Dumpy and Scots Grey). The predominantly long
19 Scots Dumpy 4.02 0.16 1.13
20 Scots Grey 5.15 )0.81 0.68
branches and unresolved branching indicate that these
21 Silkie 4.35 0.60 1.54 breeds are genetically distinct and do not have shared
22 Spanish 6.71 )2.67 )0.33 recent histories (Fig. 2).
23 Light Sussex 3.60 0.42 1.22
24 Sussex 3.17 0.69 1.31
Genetics for breed conservation
1
Contribution to between-breed diversity.
2 The application of population genetics to breed conservation
Contribution to within-breed diversity.
3 essentially encompasses two different components: identi-
Aggregate diversity.
fying high breed genetic diversity (measured by average
other livestock species [e.g. European cattle: FST = 0.11 number of alleles, HE or CW) and high breed genetic
(MacHugh et al. 1998) and European sheep: FST = 0.13 uniqueness (measured by FST, genetic distance or CB). These
(Lawson Handley et al. 2007)]. In general, chicken breeds tools can be used to identify genetically robust breeds as well
appear to be genetically distinct populations with limited gene as those of potential conservation concern.
flow amongst the phenotypically diverse breeds. Given the The most genetically diverse breeds were generally those
small population sizes for many of the British chicken breeds represented by more than one morphological colour type.
(Table 1) and the short generation interval, random genetic For instance, Leghorn had four sampled types; Araucana,
drift has likely contributed to the elevated levels of observed Brahma, Silkie and Sussex each had three; and Cochin and
genetic differentiation. The flocks sampled in this study were Maran each had one (Tables 1, 2 & 4). However, there were
primarily from smallholders not involved in competitive also several genetically diverse breeds with only one sam-
poultry showing. It was expected that such breeders would pled morphological type (Buff Orpington, Rhode Island Red,
employ a rigorous pedigree breeding program (i.e. limi- Scots Dumpy and Light Sussex).
ted cross-breeding), and the high genetic differentiation The most genetically distinctive breeds were also amongst
observed between breeds suggests that the sampling strategy the most morphologically distinctive breeds (Tables 3 & 4,
used here was generally successful. Fig. 2). For example, Indian Game (short legs, squat and
Although the British chicken breeds were highly differ- wide), Silkie (degenerate feathers, dark skin) and Spanish
entiated (Table 3), there were, nonetheless, genetic rela- (white face) possess traits not found in other breeds sampled in
tionships between certain breeds, which can be supported this study (Roberts 1997). Furthermore, the most genetically
by historical information on breed development. Phyloge- distinctive breeds also tended to have low population sizes
netic reconstruction revealed two principal clades, one (e.g. Appenzeller, Buff Orpington, Hamburgh, Indian
consisting of Derbyshire Redcap, Dorking, Hamburgh, Old Game, Ixworth, Lincolnshire Buff, Scots Grey and Spanish)
English Pheasant Fowl and Spanish and the other clade (Table 1).
consisting of Buffs (Buff Orpington and Lincolnshire Buff), Although the Spanish is the highest contributor to
Asian (Cochin, Brahma and Croad Langshan), Sussex, between-breed diversity (CB, Table 4) owing to high genetic
Ixworth, Maran and Rhode Island Red breeds (Figs 2 & 5). differentiation (Table 3), it also possesses very low genetic
The first group (excluding the Spanish) consisted of old variation (Table 2). Indeed, generally the genetically unique

 2011 The Authors, Animal Genetics  2011 Stichting International Foundation for Animal Genetics, 43, 552–563
562 Wilkinson et al.
breeds had lower genetic diversity than other breeds (mea-
sured by HE and FST: SpearmanÕs rank correlation between
these two statistics was )0.78, P £ 0.05) and these mea-
sures are therefore not independent (Tapio et al. 2005). The
ranking of breeds based on the levels of CB has been the
subject of some criticism (Toro et al. 2009), because there is
greater emphasis on highly differentiated populations,
which tend to be more monomorphic than less differentiated
populations. This is of particular relevance for livestock
breeds, because of the extensive genetic diversity observed
within breeds (Bruford et al. 2003). Therefore, defining
conservation priorities based solely on genetic uniqueness
will ignore this genetic variation. Instead, a method that
incorporates both between- and within-breed contributions
is preferable to produce a more objective inference of the
genetic state of a breed (Toro et al. 2009). By adopting one
such approach (Ollivier & Foulley 2005), the Spanish was
the lowest contributor to overall genetic diversity (measured
by D, Table 4). Conversely, the highest contributors to
overall diversity were breeds with high levels of genetic
diversity such as Araucana, Buff Orpington, Maran, Rhode
Island Red and Sussex.
Two breeds, Buff Orpington and Silkie, were not only
genetically unique (FST > 0.24, Table 3), but also possessed
high levels of genetic diversity (HE > 0.50, Table 2), and
thus these breeds were high contributors to overall British
chicken breed biodiversity (Table 4). The reverse perspective
can be adopted to identify genetically vulnerable breeds (i.e.
those that possess both low genetic diversity and unique-
ness). Three breeds satisfied these criteria: Dorking, Ham-
burgh and Old English Pheasant Fowl (Tables 2 & 3), and
this was confirmed by the low contribution of these breeds
to overall aggregate diversity (Table 4). In addition, these
breeds possessed no private alleles (Table 2). The genetic
results (coupled with the low population size of Old English
Pheasant Fowl, Table 1) highlight that the future viability
of these breeds is of potential concern.
Acknowledgements
We are indebted to Graeme Robertson for collating and
archiving the samples and to Karen Troup and David Morrice
from the ArkGenomics group for DNA extraction and geno-
typing. The authors thank Andrew Sheppy for invaluable
comments that improved the manuscript. S. Wilkinson would
like to thank the UK Biotechnology and Biological Sciences
Research Council (BBSRC) and Rare Breeds Survival Trust
(RBST) for studentship funding. The Roslin Institute is sup-
ported by a core strategic grant from the BBSRC.
References
Berthouly C., BedÕHom B., Tixier-Boichard M., Chen C.F., Lee Y.P.,
Laloe D., Legros H., Verrier E. & Rognon X. (2008) Using
molecular markers and multivariate methods to study the genetic
 2011 The Authors, Animal Genetics  2011 Stichting International Foundation for Animal Genetics, 43, 552–563
diversity of local European and Asian chicken breeds. Animal
Genetics 39, 121–9.
Bodzsar N., Eding H., Revay T., Hidas A. & Weigend S. (2009)
Genetic diversity of Hungarian indigenous chicken breeds based
on microsatellite markers. Animal Genetics 40, 516–23.
Bowcock A.M., Ruiz-Linares A., Tomfohrde J., Minch E., Kidd J.R.
& Cavalli-Sforza L.L. (1994) High resolution of human evolu-
tionary trees with polymorphic microsatellites. Nature 368,
455–7.
Bruford M.W., Bradley D.G. & Luikart G. (2003) DNA markers
reveal the complexity of livestock domestication. Nature Reviews.
Genetics 4, 900–10.
Corander J., Marttinen P., Siren J. & Tang J. (2008) Enhanced
Bayesian modelling in BAPS software for learning genetic
structures of populations. BMC Bioinformatics 9, 539–52.
DEFRA (2009) Review of molecular characterization studies relat-
ing to UK Farm Animal Genetic Resources. Available at: http://
www.defra.gov.uk/fangr/randd.htm. Accessed 1 July 2010.
DEFRA (2010) Poultry in the United Kingdom. DEFRA, London, UK.
Fabuel E., Barragan C., Silio L., Rodriguez M.C. & Toro M.A. (2004)
Analysis of genetic diversity and conservation priorities in Iberian
pigs based on microsatellite markers. Heredity 93, 104–13.
FAO (2007) The State of the WorldÕs Animal Genetic Resources for
Food and Agriculture – In Brief. Commission on Genetic Resources
for Food and Agriculture. FAO, Rome, Italy.
Felsenstein J. (1989) PHYLIP – Phylogeny Inference Package (Version
3.2). Cladistics 5, 164–6.
Glowatzki-Mullis M.L., Muntwyler J., Pfister W., Marti E., Rieder S.,
Poncet P.A. & Gaillard C. (2006) Genetic diversity among horse
populations with a special focus on the Franches–Montagnes
breed. Animal Genetics 37, 33–9.
Goudet J. (1995) FSTAT (Version 1.2): a computer program to cal-
culate F-statistics. Journal of Heredity 86, 485–6.
Granevitze Z., Hillel J., Chen G.H., Cuc N.T.K., Feldman M., Eding H.
& Weigend S. (2007) Genetic diversity within chicken popula-
tions from different continents and management histories. Animal
Genetics 38, 576–83.
Granevitze Z., Hillel J., Feldman M., Six A., Eding H. & Weigend S.
(2009) Genetic structure of a wide-spectrum chicken gene pool.
Animal Genetics 40, 686–93.
Guo S.W. & Thompson E.A. (1992) Performing the exact test of
Hardy-Weinberg proportion for multiple alleles. Biometrics 48,
361–72.
Hams F. (2004) Old Poultry Breeds. Shire Publications Ltd, Buck-
inghamshire, UK.
Hillel J., Groenen M.A.M., Tixier-Boichard M. et al. (2003) Biodi-
versity of 52 chicken populations assessed by microsatellite typ-
ing of DNA pools. Genetics, Selection, Evolution 35, 533–57.
Laloë D., Moazami-Goudarzi K., Lenstra J.A. et al. (2010) Spatial
trends of genetic variation of domestic ruminants in Europe.
Diversity 2, 932–45.
Laval G., Iannuccelli N., Legault C. et al. (2000) Genetic diversity of
eleven European pig breeds. Genetics, Selection, Evolution 32,
187–203.
Lawson Handley L.J., Byrne K., Santucci F., Townsend S., Taylor
M., Bruford M.W. & Hewitt G.M. (2007) Genetic structure of
European sheep breeds. Heredity 99, 620–31.
Li M.H., Tapio I., Vilkki J. et al. (2007) The genetic structure of
cattle populations (Bos taurus) in northern Eurasia and the

neighbouring Near Eastern regions: implications for breeding
strategies and conservation. Molecular Ecology 16, 3839–53.
MacHugh D.E., Loftus R.T., Cunningham P. & Bradley D.G. (1998)
Genetic structure of seven European cattle breeds assessed using
20 microsatellite markers. Animal Genetics 29, 333–40.
Minch E., Ruiz-Linares A., Goldstein D., Feldman M. & Cavalli-
Sforza L. (1997) MICROSAT v.1.5d: a computer program for cal-
culating various statistics on microsatellite allele data. Available
at: http://lotka.stanford.edu/microsat/microsat.html. Accessed
13 May 2008.
Muir W.M., Wong G.K.-S., Zhang Y. et al. (2008) Genome-wide
assessment of worldwide chicken SNP genetic diversity indicates
significant absence of rare alleles in commercial breeds. Proceed-
ings of the National Academy of Sciences 105, 17312–7.
Nei M. & Li W.H. (1973) Linkage disequilibrium in subdivided
populations. Genetics 75, 213–9.
Ollivier L. & Foulley J.L. (2005) Aggregate diversity: new approach
combining within- and between-breed genetic diversity. Livestock
Production Science 95, 247–54.
Paradis E., Claude J. & Strimmer K. (2004) APE: analyses of phy-
logenetics and evolution in R language. Bioinformatics 20, 289–
90.
R Development Core Team (2010) R: A language and environment
for statistical computing. R Foundation for Statistical Computing,
Vienna, Austria. Avaialbe at: http://www.R-project.org.
Reynolds J., Weir B.S. & Cockerham C.C. (1983) Estimation of the
co-ancestry coefficient – basis for a short-term genetic-distance.
Genetics 105, 767–79.
Rice W.R. (1989) Analyzing tables of statistical tests. Evolution 43,
223–5.
Roberts M. (1994) British Large Fowl. Domestic Fowl Research,
Devon, UK.
Roberts V. (1997) British Poultry Standards. Wiley-Blackwell,
Oxford, UK.
 2011 The Authors, Animal Genetics  2011 Stichting International Foundation for Animal Genetics, 43, 552–563
Genetic structure of British chicken breeds 563
Rosenberg N.A., Burke T., Elo K. et al. (2001) Empirical evaluation
of genetic clustering methods using multilocus genotypes from
20 chicken breeds. Genetics 159, 699–713.
Rousset F. (2008) GENEPOP Ô007: a complete re-implementation of
the GENEPOP software for Windows and Linux. Molecular Ecology
Resources 8, 103–6.
SanCristobal M., Chevalet C., Haley C.S. et al. (2006) Genetic
diversity within and between European pig breeds using micro-
satellite markers. Animal Genetics 37, 189–98.
Tapio M., Tapio I., Grislis Z., Holm L.E., Jeppsson S., Kantanen J.,
Miceikiene I., Olsaker I., Viinalass H. & Eythorsdottir E. (2005)
Native breeds demonstrate high contributions to the molecular
variation in northern European sheep. Molecular Ecology 14,
3951–63.
Toro M.A., Fernandez J. & Caballero A. (2009) Molecular charac-
terization of breeds and its use in conservation. Livestock Science
120, 174–95.
Vorwald Dohner J. (2001) The Encyclopedia of Historic and Endan-
gered Livestock and Poultry Breeds. Yale University Press, Yale,
USA.
Weir B.S. & Cockerham C.C. (1984) Estimating F-statistics for the
analysis of population-structure. Evolution 38, 1358–70.
Weitzman M.L. (1992) On diversity. Quarterly Journal of Economics
CVII, 363–405.
Wiener P., Burton D. & Williams J.L. (2004) Breed relationships and
definition in British cattle: a genetic analysis. Heredity 93, 597–
602.
Wilkinson S., Haley C., Alderson L. & Wiener P. (2011) An
empirical assessment of individual-based population genetic
statistical techniques: application to British pig breeds. Heredity
106, 261–9.
Zanetti E., De Marchi M., Dalvit C. & Cassandro M. (2010) Genetic
characterization of local Italian breeds of chickens undergoing in
situ conservation. Poultry Science 89, 420–7.