Journal of Heredity Advance Access published November 8, 2010

Journal of Heredity Ó The American Genetic Association. 2010. All rights reserved.
doi:10.1093/jhered/esq106 For permissions, please email: [email protected].

Insights into Genetic Diversity,

Parentage, and Group Composition
of Atlantic White-Sided Dolphins
(Lagenorhynchus acutus) off the West
of Ireland Based on Nuclear
and Mitochondrial Genetic Markers

Downloaded from jhered.oxfordjournals.org by guest on November 30, 2010

LUCA MIRIMIN, EULALIA BANGUERA-HINESTROZA, EILEEN DILLANE, ALAN R. HOELZEL, TOM F. CROSS, AND
EMER ROGAN
From the Department of Zoology, Ecology and Plant Sciences, University College Cork, North Mall, Distillery Fields, Cork,
Ireland (Mirimin, Dillane, Cross, and Rogan); the School of Biological and Biomedical Sciences, Durham University, Durham,
DH1 3LE, UK (Banguera-Hinestroza and Hoelzel). Luca Mirimin is now at the Martin Ryan Institute, Carna, National
University of Ireland, Galway, Muigh-inis, Carna, County Galway, Ireland.

Address correspondence to L. Mirimin at the address above, or e-mail: [email protected].

Abstract
The analysis of stranding events and the application of molecular markers can be powerful tools to study cryptic biological
aspects of delphinid species that occur mainly in open ocean habitat. In the present study, we investigated nuclear and
mitochondrial genetic variability of Atlantic white-sided dolphins that stranded from 1990 to 2006 (n 5 42) along the west
coast of Ireland, using 8 microsatellite loci and 599 bp of the mitochondrial DNA control region. Results from both classes
of markers are concordant with the hypothesis of a large random-mating population of white-sided dolphins along the west
coast of Ireland. In addition, the analyses of 2 live mass stranding events (19 and 5 individuals, respectively) revealed that
dolphins within each group were mainly unrelated to each other, suggesting dispersal of both sexes from the natal group
(i.e., no natal phylopatry). Parentage analyses allowed the identification of mother–offspring pairs but ruled out all adult
males as possible fathers. In combination with data on age of individuals, these results confirmed previous knowledge on
life-history parameters, with sexually mature females ranging between 11 and 15 years of age and an interbirth interval of at
least 2 years. The present study provides novel information on population and group composition of Atlantic white-sided
dolphins along the west coast of Ireland, where population and social structure of the species are still poorly understood.
Key words: Atlantic white-sided dolphin, genetic markers, group composition, Lagenorhynchus acutus, life history, parentage analyses

In recent times, the application of molecular markers in pole system or biopsy darting of free-ranging individuals)
combination with increasing sample availability has allowed (Krützen et al. 2002; Bilgmann et al. 2007; Green et al.
the study of cryptic biological aspects of many marine 2007). An increasing number of studies which combine
pelagic species, especially cetaceans (baleen and toothed molecular, observational, and photo-ID data have revealed
whales) (Hoelzel 2002). Following the advent of polymerase a wide range of grouping patterns in cetacean species at
chain reaction (PCR), the potential to carry out molecular both inter- and intraspecific levels (Mann et al. 2000;
studies from very small starting amounts of target DNA Michaud 2005). Differing habitats and ecological conditions
fragments has provided the opportunity to work on can play a significant role in shaping social structure of
degraded tissue (e.g., from museum specimens or beach- cetacean species (Gowans et al. 2007), where groups of
cast carcasses) (Rosel 2003; Pimper et al. 2009), and it has different composition and size may form as a consequence
led to the development of novel sampling techniques (e.g., of resource distribution, predator avoidance, and parental

1
Journal of Heredity

care. Ranging from the fluid fission–fusion societies described Levels of genetic diversity and within-group pedigree
in small delphinid species (e.g., Norris and Dohl 1980; Würsig relationships have been little studied in white-sided dolphins
et al. 1994) to matrilineal groups of larger toothed whales (Amaral 2005; Banguera-Hinestroza 2008), and it is currently
(e.g., Hoelzel and Osborne 1986; Bigg et al. 1990; Baird and unknown whether group composition resembles more the
Whitehead 2000), individuals of both or either sex can form fluid fission–fusion pattern found in some bottlenose dolphin
temporary or permanent associations, ranging from few to (Tursiops truncatus) populations or the matrilineal social orga-
hundreds of individuals including single or multiple gen- nization of killer whales (Orcinus orca), among other social
erations. However, although most data on cetacean social structures. In the present study, we investigated genetic
structure have been recorded from species occurring in variability at nuclear (8 microsatellites) and mitochondrial
coastal and hence more accessible habitat, very little is known (599 bp of the control region) markers from Atlantic white-
about oceanic delphinids, for which the main source of sided dolphins that stranded along the west coast of Ireland,
information is often provided by opportunistic sampling, in the eastern North Atlantic, from 1990 to 2006, including 2
such as the examination of stranding or bycatch events. live mass stranding events. The aims of the present study are
Atlantic white-sided dolphins (Lagenorhynchus acutus Gray, to present an evaluation of nuclear and mitochondrial genetic
1828) (referred to as white-sided dolphins hereafter) are diversity of white-sided dolphins in the studied area, to test
oceanic delphinids endemic to temperate and subpolar waters whether dolphins stranded in the 2 live strandings are
of the North Atlantic Ocean (Kinze et al. 1997; Northridge more related to each other than expected by chance, to
et al. 1997; Palka et al. 1997; Weinrich et al. 2001; MacLeod elucidate pedigree relationship within each stranding event

Downloaded from jhered.oxfordjournals.org by guest on November 30, 2010

et al. 2005; Wall et al. 2006). The species tend to be more (parentage and kinship analyses), and to combine these results
abundant in the western than eastern Atlantic (Northridge with age data to infer possible life-history parameters
et al. 1997; MacLeod 2004), it is rarely found in inshore waters (e.g., maturity at age of mother–offspring pairs and interbirth
and seems to occur predominantly along the continental shelf interval).
edge and offshore habitats related to high bottom relief (Reid
et al. 2003; Wall et al. 2006). In the 1970s, Atlantic white-sided
dolphins in US waters were reported to have shifted their Materials and Methods
distribution from offshore to a more inshore (continental Sample Collection
shelf) habitat, possibly related to a change in habitat of
Skin or muscle tissue was obtained from all available samples
preferred prey species (Katona et al. 1993; Kenney et al. 1996).
of Atlantic white-sided dolphins that stranded along the Irish
Although population dynamics of the species are still little
coasts from 1990 to 2006, for a total of 42 individuals. These
understood (Banguera-Hinestroza 2008), white-sided dolphins
comprised 19 beach-cast carcasses along the west coast of
are currently regarded as abundant and of least concern by the
Ireland, including 2 dolphins that were found dead in close
International Union for Conservation of Nature red list of
proximity on the same day (Mullet peninsula, County Mayo);
endangered species (Hammond et al. 2008). However, the
one individual that was bycaught in a pelagic drift-net fisheries
species has been affected to some extent by both direct
along the continental shelf; and individuals from 2 live mass
(Jefferson et al. 1993; Reeves et al. 1999) and indirect catches
strandings that had occurred in September 1994 at Ross point
(Morizur et al. 1999; Waring et al. 2006), which has raised
(Killala Bay, County Mayo) and in October 1998 at Claggan
concern about its conservation status (Evans 2009).
Strand (Clew Bay, County Mayo) (18 and 5 individuals,
White-sided dolphins live in groups and numerous live
respectively) (see details in Supplementary Material). Previous
mass stranding events, where 2 or more individuals strand
examination of maturity status and age in the larger group of
ashore alive, have been reported throughout the species’
mass stranded dolphins (referred to as Ross hereafter)
distribution range (Gresson 1969; Geraci et al. 1978; Sergeant
revealed the presence of adults of both sexes (between 8
et al. 1980; Rogan et al. 1997). Most information on the
and 17 years old) and 4 calves (i.e., up to 2 years old), as
species has come from the investigation of such events
determined by inspection of gonads and growth layer counts
(Sergeant et al. 1980; Addink et al. 1997; Kinze et al. 1997;
in tooth sections (Rogan et al. 1997). Similarly, 3 adults and 2
Rogan et al. 1997) and some observational data, where groups
calves (based on total body length) were present in the smaller
generally include individuals of both sexes and may vary in
mass stranding event (referred to as Claggan hereafter)
size from few up to a hundred individuals (Kinze et al. 1997;
(Rogan et al. 2002). Unfortunately, due to unavailability of
Weinrich et al. 2001). The finding of gender-biased groups
tissue samples, the present study does not include one adult
and the lack of immature individuals (3–7 years of age) in
male and fetuses of 3 pregnant females from the original Ross
mass strandings suggest some degree of sex and age
stranding.
segregation (Sergeant et al. 1980; Rogan et al. 1997, 2002).
In both the western and eastern North Atlantic, single and
Data Collection and Analyses
mass stranding events tend to occur mainly during winter–
spring months (September–May), as a possible consequence Genomic DNA was isolated from tissue samples by means
of inshore–offshore movements, which appear to be related of a proteinase-K digestion and phenol–chloroform extrac-
to a number of oceanographic, environmental, and ecological tion method (modified from Sambrook et al. 1989),
conditions (Palka et al. 1997; Amaral 2005; Vanmann et al. followed by ethanol precipitation. Multilocus genotypes
2005). were obtained from all 42 individuals using 8 microsatellite

2
 Mirimin et al.  Genetic Diversity, Parentage, and Group Composition of Atlantic White-Sided Dolphins

loci that were previously isolated from other cetacean from Gerber et al. (2000)). Levels of genetic relatedness
species: Delphinus delphis (Dde65, Dde66, Dde69, Dde70, between pairs of individuals and among different groups were
Dde72; Coughlan et al. 2006), Stenella coeruleaoalba (Sco11, evaluated using the relatedness estimator (R) of Queller and
Sco28; Mirimin et al. 2006) and Megaptera novaeangliae Goodnight (1989). In order to ensure reliability and
(GGAT416; Palsboll et al. 1997). PCR thermocycling robustness of the R estimator, we performed a rarefaction
conditions followed those described in the original analysis (10 000 simulations), where the difference between
publications. Size of amplified products was resolved on consecutive overall average R estimates is expressed as
6% polyacrylamide gels on an LI-COR 4300 automated a function of the total number of loci, using the web-based
DNA sequencer by comparison with a standard size ladder program RERAT (Schwacke et al. 2005). The same program
(LI-COR). was used to calculate pairwise R estimates, corrected by
In order to eliminate possible bias due to the presence of excluding all calves from allele frequency calculation.
closely related individuals, the 6 calves were removed from Furthermore, to test whether individuals grouped by stranding
the following tests, except from the parentage analysis and event and/or by sex class were more related to each other
pairwise relatedness estimation. Possible presence of null than expected by chance, the observed within-group average
alleles, large allele drop out, or genotyping errors due to R estimates were compared with values obtained from 1000
stuttering were tested using MICRO-CHECKER 2.2.3 (van randomizations, using the Microsoft Excel Macro GROUP-
Oosterhout et al. 2004). Linkage disequilibrium between all RELATE (Valsecchi et al. 2002). Each randomization was
pairs of loci (10 000 permutations and 10 initial conditions for generated by replacing the original genotypes with alleles

Downloaded from jhered.oxfordjournals.org by guest on November 30, 2010

the Expectation-Mazimization algorithm, Slatkin and Excoff- drawn randomly from the observed allele frequencies but
ier 1996) and exact test for departure from Hardy–Weinberg retaining information on the compared group size and any
expected proportions (1 000 000 chain length and 100 000 missing data to correct for potential biases due to the structure
dememorization steps, Guo and Thompson 1992) were of the original data set. In order to identify possible groups
performed using ARLEQUIN 3.1 (Excoffier et al. 2005). The (partitions) of related individuals, we used a pedigree re-
same program was used to calculate the number of alleles and construction algorithm as implemented in PEDIGREE 2.2
observed and expected heterozygosities within each sample (available at http://herbinger.biology.dal.ca:5080/Pedigree/).
and overall. The inbreeding coefficient FIS and its departure Based on multilocus genotypic data and following Mendelian
from zero (number of randomizations over a total of 8000 inheritance, PEDIGREE first computes the pairwise likeli-
that gave a larger FIS than the observed value) were calculated hood ratios of being related (full- or half sibs) versus being
using FSTAT 2.9.3.2 (Goudet 2001). Bonferroni correction unrelated for every pair of individuals in the data set and then
was applied to multiple tests. A test for heterozygosity excess uses these pairwise ratios to calculate the score associated with
(Luikart et al. 1998) was carried out to detect possible recent a given partition (following formulas detailed in PEDIGREE
reduction in effective population size, using the program online manual). Following a number of trial runs, parameters
BOTTLENECK (Cornuet and Luikart 1996; Piry et al. 1999). for the Markov Chain Monte Carlo engine were set to
For this test, observed gene diversity was compared with 5 000 000 iterations, no full-sib constraint, temperature of 10,
expected equilibrium gene diversity (10 000 simulations) by weight of 2 and a random seed number. Significance of
means of a one-tailed Wilcoxon test, following the 2-phase a partition was then evaluated by comparing the observed
mutation model (95% single-step and 5% multiple-step score value with scores obtained from 1000 randomly
mutations, with a variance among multiple steps set to 12) generated data sets which used the same parameters and
(DiRienzo et al. 1994). The probability that 2 individuals share random seed of the original run.
the same genotype by chance (probability of identity, PI) A portion of the mitochondrial DNA (mtDNA) control
(Waits et al. 2001), the probability to exclude a putative parent region was amplified using universal primers: MTCRF
given the markers considered (probability of exclusion, PE) (5#-TTC CCC GGT CTT GTA AAC C-3#) and MTCR-R
(Jamieson and Taylor 1997), and the presence of identical (5#-ATT TTC AGT GTC TTG CTT T-3#) (Hoelzel and
genotypes in the data set were calculated using FAMOZ. Green 1998), with the following conditions: 20–50 ng of
Parentage analyses were carried out using a categorical DNA, 10 PCR buffer, 1.5 mM MgCl2, 50–100 ng of
allocation method (Gerber et al. 2000), as implemented in primers, 2.5 mM dNTP, and 1 U of Taq polymerase.
FAMOZ (Gerber et al. 2003). The ratio of the logarithm of Amplifications were conducted on an MJ Research thermo-
odds (LOD) scores between the 2 most likely parents (or cycler with the following cycle conditions: 94 °C 2 min
parent pairs) is calculated according to formulas given in followed by 35 cycles of 94 °C 30 s, 54 °C 30 s, and 72 °C
Gerber et al. (2000) and then compared with distributions of 30 s. Amplified products were purified using a QUIAGEN
values obtained from simulated offspring (100 000 simulated purification kit (Qiagen Inc.) and were sequenced from both
individuals), assuming a 0.1% simulation and calculation error ends using an ABI 377 automated sequencer. To ensure
rate and no heterozygote deficit. All adults from both single- reliability of sequencing products, homologous sequences of
and mass strandings (n 5 36) were considered as potential 5 individuals (12% of the total number of individuals) were
parents, and significance threshold values were determined by processed independently twice (i.e., from DNA extraction to
the intersection of the distributions of LOD scores from sequencing). Sequences were aligned using the program
simulated offspring from genotyped parents and offspring CLUSTAL X 1.83 (Thompson et al. 1997) and inspected by
generated according to allele frequencies (following guidelines eye. Haplotype and nucleotide diversity were estimated using

3
Journal of Heredity

ARLEQUIN 3.1. To increase power of parentage (maternal) The 3 most common haplotypes (Haplo 1, Haplo 2, and
analyses, mtDNA haplotypes were also taken into account in Haplo 3) were found in the single strandings and in either
comparisons between calves and putative mothers. Ross or Claggan live mass stranding groups, whereas 10 of 19
haplotypes were represented only once in the whole sample
set (Table 2). Similar levels of haplotype diversity were found
in the sample of single-stranded dolphins (0.941 ± 0.039)
Results (Table 2) and in the large Ross live mass stranding (0.934 ±
Gene Diversity 0.045) (calculated without the 6 calves), whereas 2 different
haplotypes were found among the 3 adults in the Claggan
All 8 microsatellite loci were successfully amplified in Atlantic
stranding (see details in Table 2 and Supplementary Material).
white-sided dolphins and showed degrees of polymorphism.
Over the whole sample set (excluding the 6 calves), the
Parentage and Relationship Analyses
number of alleles per locus ranged between 5 and 10 and the
average observed and expected heterozygosities were 0.699 No identical genotypes were found in the data set when at
(standard deviation [SD] 0.092) and 0.697 (SD 0.102), least 4 loci were considered. When considering all 8 loci, the
respectively (Table 1). Estimates of possible null alleles, large cumulative PI was 0.02%, and the PE was 94.5, 99.5, and 99.9
allele drop out, and genotyping errors were negligible for single parent, paternity, and parent pair, respectively.
(,0.1%), and no evidence of physical linkage between any Inspection of LOD score distributions from simulated

Downloaded from jhered.oxfordjournals.org by guest on November 30, 2010

pair of loci was found. The whole sample, as well as the Ross offspring from genotyped parents and from the overall allele
stranding (excluding calves), showed no deviation from frequencies (calculated without calves) indicated a significance
Hardy–Weinberg expected proportions and no evidence of LOD score threshold of 1.9 and 1.14 for parent–offspring
inbreeding (i.e., nonsignificant inbreeding coefficient FIS) and parent pair–offspring relationships, respectively. Table 3
(Table 1). Observed gene diversity did not differ significantly shows the results of the categorical allocation approach
from diversity expected in populations at equilibrium (one- (FAMOZ) taking into account all noncalves (n 5 36) as
tailed Wilcoxon heterozygosity excess test, P . 0.5). A 599 potential parents and allowing for a 0.1% error rate. Based on
bp portion of the mtDNA control region was successfully simple genotypic/haplotypic exclusion (i.e., no mismatching
obtained from 41 of the 42 sampled individuals (one single- alleles and haplotypes), 5 of 6 calves were associated to
stranded individual repetitively failed to yield clear sequence a single parent, whereas the sixth calf (Ross 8) was associated
traces, probably due to a highly degraded tissue sample) to 2 putative parents (Ross 4 and WSD 1/98). The
(EMBL Nucleotide Sequence Database accession numbers: incorporation of a statistical LOD score threshold (as defined
FR668237–FR668246 and FR682916–FR682924). Most se- above) supported all the nonexcluded parents (i.e., all pairs
quences used in this analysis had a Phred quality score of 30 showed an LOD score above threshold) (Table 3). Although
or higher, and all 5 individuals that were sequenced twice no father–mother–offspring triad was identified based on
yielded consistent results (100% matches), indicating re- genotypic exclusion, the triad WSD1/98/Ross8/Ross4
liability of sequencing. Alignment and analysis of reliable showed an LOD score (5.65) higher than the predefined
sequences revealed 23 polymorphic sites (21 transitions, 2 threshold (1.14), but allowing for one mismatching locus
transversions, and no indels), 19 different mtDNA haplo- (possible genotyping error). If we do not allow for genotyping
types, and an overall nucleotide and haplotype diversity of errors, given an error rate of less than 0.01% and knowing
0.006 (±0.004) and 0.942 (±0.020), respectively (Table 2). that WSD 1/98 was not part of the group of live stranded

Table 1 Genetic variability at 8 microsatellite loci in the single stranded sample (Strandings, n 5 19), Ross live stranding sample
without calves (n 5 14) and Overall, including the 3 adults from the Claggan sample (n 5 36)

Strandings Ross Overall

Locus NA HO HE FIS NA HO HE FIS NA HO HE FIS P
Dde65 5 0.684 0.741 0.079 5 0.714 0.786 0.094 5 0.667 0.747 0.109 0.168
Dde66 8 0.765 0.815 0.063 7 0.929 0.802 0.166 9 0.823 0.846 0.027 0.419
Dde69 4 0.737 0.681 0.084 5 0.714 0.709 0.008 5 0.722 0.694 0.041 0.715
Dde70 6 0.643 0.685 0.064 7 0.857 0.735 0.173 10 0.742 0.716 0.036 0.733
Dde72 6 0.737 0.740 0.004 6 0.786 0.677 0.167 7 0.778 0.716 0.088 0.867
Sco28 5 0.579 0.597 0.032 4 0.629 0.668 0.038 5 0.629 0.618 0.018 0.648
Sco11 5 0.579 0.509 0.141 5 0.357 0.434 0.182 7 0.528 0.498 0.061 0.814
GGAT416 8 0.684 0.784 0.130 6 0.750 0.732 0.026 9 0.706 0.740 0.046 0.382
Mean 5.9 0.676 0.694 0.027 5.6 0.725 0.693 0.048 7.1 0.699 0.697 0.003 0.535
(SD) (1.5) (0.071) (0.100) (1.1) (0.169) (0.115) (2.0) (0.092) (0.102)

NA, number of alleles; HO, observed heterozygosity; HE, expected heterozygosity; FIS, inbreeding coefficient; P, proportion of randomizations that gave
a larger FIS than the observed value.

4
 Mirimin et al.  Genetic Diversity, Parentage, and Group Composition of Atlantic White-Sided Dolphins

Table 2 Genetic diversity and haplotypic frequency of for all groups and/or sex classes, where adult males among
a 599 bp portion of the mtDNA control region in the single- the Ross live stranding were the only group that showed
stranded (Strandings) and live mass stranded groups (Ross and a marginal but significantly higher average R than expected by
Claggan) (see details in Supplementary Material for individual
haplotypic information)
chance (P , 0.05). Using the PEDIGREE approach, the
most likely partition of the data included 11 groups of
Strandings Ross Claggan Overall potentially related individuals (full- or half siblings) (score 5
n 18 18 5 41
3534.39), ranging between 4 and 2 individuals per group (data
NH 12 9 2 19 not shown). However, 950 of 1000 randomized data sets
H 0.941 0.934 0.667 0.946 produced the same or higher partition scores, indicating that
(±0.039) (±0.045) (±0.314) (±0.020) the observed groups are most likely artifacts and consequently
p 0.006 0.006 0.004 0.007 composed of mainly unrelated individuals. When including
(±0.004) (±0.004) (±0.004) (±0.004) the calves, these grouped with each respective mother (as
Frequencies expected in first-degree relatives) but did not cluster together,
Haplo 1 4 3a 7
Haplo 2 2 3a 5 indicating that they were not related to each other (i.e., did
Haplo 3 1 4a 5 not share the same father).
Haplo 4 3a 3
Haplo 5 3a 3
Haplo 6 2a 2
Discussion

Downloaded from jhered.oxfordjournals.org by guest on November 30, 2010

Haplo 7 2 2
Haplo 8 2 2 Gene Diversity
Haplo 9 1 1
Haplo 10 1 1 White-sided dolphins that stranded along the west coast of
Haplo 11 1 1 Ireland showed relatively high nuclear and mtDNA diversity.
Haplo 12 1 1 2 This is in agreement with ongoing work on the same species
Haplo 13 1 1
Haplo 14 1 1
(Banguera-Hinestroza personal communication) and with
Haplo 15 1 1 levels found in other congeneric oceanic delphinid species
Haplo 16 1 1 characterized by large population sizes (Hayano et al. 2004;
Haplo 17 1 1 Cassens et al. 2005). The observed levels of nuclear and
Haplo 18 1 1 mitochondrial diversity, Hardy–Weinberg Equilibrium, and
Haplo 19 1 1 no inbreeding among the 36 adult dolphins are consistent
with a scenario of a large random-mating population of
EMBL Accession numbers: FR668237-FR668246 and FR682916-
Atlantic white-sided dolphins along the west coast of Ireland.
FR682924. n, number of individuals; NH, number of different haplotypes;
H, haplotype diversity; p, nucleotide diversity. Analyses of the observed levels of nuclear gene diversity
a
Including one calf.
(heterozygosity excess test) showed no evidence of a recent
bottleneck in the studied sample, suggesting that the
population could be at mutation-drift equilibrium. Large
individuals, it is likely that this triad is a false positive (i.e., numbers of white-sided dolphins have been reported in
with individual WSD 1/98 sharing alleles with calf Ross 8 by waters off the northwest of Scotland to the west of the Outer
chance and not by descent). Rarefaction analysis indicated Hebrides (21 371, coefficient of variation [CV] 5 0.54,
stable estimates of R when using the 8 loci, with less than 6% during summer 1998) and in the Faroe Shetland Channel
change in average R between addition of the seventh and (74 626, CV 5 0.72) (MacLeod 2004), and North Sea and
eighth locus. Table 4 shows average relatedness (R) estimates Celtic Sea areas (11 760 individuals, CV 5 0.26; although this

Table 3 Parentage analyses showing all putative parent–offspring pairs, as identified by genotypic/haplotypic exclusion
(8 microsatellite loci and 599 bp of the mtDNA control region) and by categorical allocation (LOD score) (threshold for parent–pair
relationship 5 1.9)

Genotypic exclusion
Putative number of loci compared/ Haplotypic exclusion
Calf parent (sex) number of mismatches mtDNA offspring/parent LOD score Pairwise R
Claggan 1 Claggan 4 (F) 8/0 Haplo 2/Haplo 2 4.61 0.350
Claggan 3 Claggan 2 (F) 8/0 Haplo 6/Haplo 6 5.40 0.648
Ross 1 Ross 3 (F) 7/0 Haplo 5/Haplo 5 2.74 0.508
Ross 8 Ross 4 (F) 8/0 Haplo 1/Haplo 1 6.44 0.539
Ross 8 WSD 1/98 (M) 8/0 Haplo 1/Haplo 1 2.14 0.444
Ross 13 Ross14 (F) 7/0 Haplo 3/Haplo 3 4.92 0.487
Ross 17 Ross 5 (F) 8/0 Haplo 4/Haplo 4 4.78 0.573

Pairwise R estimates are also given for each pair.

5
Journal of Heredity

Table 4 Average relatedness (R) estimates in the 2 large groups related individuals (e.g., matrilineal groups of killer whales and
of live mass stranded (Ross) and single-randomly stranded pilot whales (Globicephala melas) (Bigg et al. 1990; Amos et al.
(Strandings) and Overall (including the 3 adults from the Claggan 1993) but may adopt a social organization pattern similar to
stranding)
the fission–fusion societies reported for other oceanic
n Ross n Strandings n Overall delphinid species, such as spinner dolphins (Stenella longirostris)
(Karczmarski et al. 2005) and common dolphins (D. delphis)
All adults 14 0.012 19 0.027 36 0.019
Adult females 7 0.039 6 0.116 15 0.062
(Viricel et al. 2008). In particular, results presented here are
Adult males 7 0.167* 13 0.015 21 0.060 very similar to those reported in a genetic analysis of live mass
stranded common dolphins, where high gene diversity was
Calves were excluded from these calculations. n, number of individuals; found in both mass and single-stranded groups, although no
*
Significantly positive (P , 0.05). mother–offspring pairs were detected (probably due to
incomplete sampling of the group (Viricel et al. 2008).
Age and sex segregation have been suggested in white-
estimate also included white-beaked dolphins (L. albirostris)
sided dolphins in both the west and east North Atlantic,
(Hammond et al. 2002). However, although a more thorough
based on the lack of juveniles in multiple strandings between
analysis of population genetic structure across the species’
3 and 6 years of age (e.g., Sergeant et al. 1980; Rogan et al.
range (including the Irish sample) is currently underway
1997) and gender-related differences in bioaccumulation of
(Banguera-Hinestroza personal communication), the current
contaminants in tissues (McKenzie et al. 1997; Weisbrod

Downloaded from jhered.oxfordjournals.org by guest on November 30, 2010

lack of understanding of population structure hampers the
et al. 2001). Sex segregation has been reported to varying
estimation of population size. Despite the fact that the test
degrees in a number of cetacean species (e.g., Miyazaki and
used in the present study is regarded as robust for as little as 4
Nishiwaki 1978; McKinnon 1994; Smith et al. 1994;
microsatellite loci, higher numbers of microsatellite loci
Whitehead and Weilgart 2000), where it may be related to
would increase the power of detecting changes in population
energy requirements due to differences in body size and/or
size (e.g., 20 loci, Cornuet and Luikart 1996).
adaptation to food resources and suitable habitat (e.g., safe
environment for mother–offspring pairs) (reviewed in
Genetic Relatedness and Group Composition
Michaud 2005). However, considering that no closely
Both Ross and Claggan live mass stranding events included related adult individuals were found in the present study,
adults of both sexes and calves (1–2 years of age) but no given the unusual circumstances characterizing stranding
juveniles between 3 and 7 years of age. The present genetic events and no supporting genetic data from other groups
analysis revealed that both groups consisted of mostly (only 2 groups were analyzed), it is not possible to establish
unrelated adults and several mother–offspring pairs. The whether the Claggan and Ross strandings consisted of stable
high nuclear diversity and presence of multiple maternal social units or simply a temporal aggregation of individuals.
lineages in both mass stranded groups suggest that group Previous age estimates (based on growth layers in tooth
composition was not driven by genetic relationships. In- sections) revealed that adults from the Ross stranding
terestingly, adult males within the Ross stranding showed ranged between 8 and 17 years (Rogan et al. 1997) (see also
marginal but significantly positive average relatedness Supplementary Material for details). In the present study, the
estimates (R 5 0.167, P , 0.05). However, pedigree analyses application of genetic markers allowed the identification of
did not confirm the presence of close relatives, with the mother–offspring pairs in the 2 live mass strandings, with all
exception of mother–offspring pairs, indicating that such putative mothers ranging from 11 to 14 years of age and
a finding may reflect some degree of relationship but not as offspring being young individuals between 1 and 2 years of
close kin (e.g., half- or full sibs). Although the 2 mass age. These findings are in agreement with life-history
stranding events analyzed here may not be representative parameters, which indicate that females attain sexual maturity
social units of the species, as they may have been part of at approximately 7–8 years of age (Sergeant et al. 1980) and
larger unsampled aggregations, our findings are in line with confirm that calves spend at least the first 2 years of their lives
predictions for species occurring in open ocean habitat, in proximity to their mothers. Two of the identified mothers
where resource distribution and availability may be patchy (Claggan 4 and Ross 3) were also pregnant. Taking into
and formation of groups of unrelated individuals provides account a gestation period of ;11 months (Sergeant et al.
benefits such as increased predator avoidance (Aviles et al. 1980), the presence of 2 adult females that were both mothers
2004). Furthermore, these results are in agreement with of 1- to 2-year-old calves and pregnant at the same time
previous data of live mass stranded groups of white-sided indicates that the period between consecutive births could
dolphins in the western North Atlantic, in which low average be as short as 2 years and suggests that adult females may be
relatedness and no father–offspring were found when tested involved in mating behavior even while still in close
against calves or fetuses within the same stranding (Amaral association with their last born. In contrast, no father–
2005). Thus, findings from the present study provide no offspring pairs could be detected, indicating that adult males
indication of natal group phylopatry for either sex and from both strandings were not responsible for fathering any
indicate that group formation may form regardless of the of the calves, but may have been responsible for fathering the
genetic relationship among individuals. This also indicates fetuses, which unfortunately were not included in the present
that white-sided dolphins do not form groups of closely study due to unavailability of samples. Furthermore, calves

6
 Mirimin et al.  Genetic Diversity, Parentage, and Group Composition of Atlantic White-Sided Dolphins

within each stranding event were not closely related to each animals bycaught in the Northeastern Atlantic. Rep Int Whal Comm.
other (e.g., half sibs), indicating that females within each 47:637–639.
group mated with different males (which are no longer Amaral K. 2005. Atlantic white-sided dolphins (Lagenorhynchus acutus) social
present in the group), suggesting a promiscuous mating structure based on stranding trends and genetics. Dartmouth (MA):
University of Massachusetts Dartmouth.
system.
Amos B, Schlotterer C, Tautz D. 1993. Social structure of pilot whales
revealed by analytical DNA profiling. Science. 260:670–672.
Aviles L, Fletcher JA, Cutter AD. 2004. The kin composition of
Conclusions social groups: trading group size for degree of altruism. Am Nat. 164:
In conclusion, the present study characterized nuclear and 132–144.
mitochondrial genetic diversity of Atlantic white-sided Baird RW, Whitehead H. 2000. Social organization of mammal-eating
dolphins found along the west coast of Ireland, where they killer whales: group stability and dispersal patterns. Can J Zool. 78:
2096–2105.
appear to be part of a discrete population at mutation-drift
equilibrium. Group composition analyses indicates no natal Banguera-Hinestroza E. 2008. Phylogeography of Lagenorhynchus acutus and
Lagenorhynchus albirostris and phylogeny of the genus Lagenorhynchus. In:
phylopatry for either sex and that genetic relationships among
School of biological and biomedical sciences. Durham (NC): Durham
individuals did not play a significant role in group formation University. p. 194.
of the 2 mass stranding events under study. Furthermore, the
Bigg MA, Olesuik PF, Ellis GM, Ford JKB, Balcomb KC. 1990. Social
combination of parentage and age data provided important organization and genealogy of resident killer whales (Orcinus orca) in the

Downloaded from jhered.oxfordjournals.org by guest on November 30, 2010

information on life-history parameters, such as confirming age coastal waters of British Columbia and Washington State. Rep Int Whal
at maturity of females and an interbirth interval of a minimum Comm. 12:383–405.
of 2 years. Although data presented here offer a first insight Bilgmann K, Griffiths OJ, Allen SJ, Möller LM. 2007. A biopsy pole system
on possible group composition and social organization of for bow-riding dolphins: sampling success, behavioral responses, and test
white-sided dolphins in the eastern North Atlantic, further for sampling bias. Mar Mamm Sci. 23:218–225.
information from additional aggregations of white-sided Cassens I, van Waerebeek K, Best PB, Tzika A, van Helden AL, Crespo
dolphins is required in order to test hypotheses raised in the EA, Milinkovitch MC. 2005. Evidence for male dispersal along the coasts
present study. but no migration in pelagic waters in dusky dolphins (Lagenorhynchus
obscurus). Mol Ecol. 14:107–121.
Cornuet J-M, Luikart G. 1996. Description and power analysis of two tests
for detecting recent population bottlenecks from allele frequency data.
Supplementary Material Genetics. 144:2001–2014.
Supplementary material can be found at http://www.jhered Coughlan J, Mirimin L, Dillane E, Rogan E, Cross T. 2006. Isolation and
.oxfordjournals.org/. characterization of novel microsatellite loci for the short-beaked common
dolphin (Delphinus delphis) and cross-amplification in other cetacean species.
Mol Ecol Notes. 6:490–492.
DiRienzo A, Peterson AC, Garza JC, Valdes AM, Slatkin M, Freimer NB.
Funding 1994. Mutational processes of simple-sequence repeat loci in human
This work was supported by the Irish Higher Education populations. Proc Natl Acad Sci U S A. 91:3166–3170.
Authority (Republic of Ireland), the Programme Alban, Evans PGH. 2009. Status and distribution of white-beaked and Atlantic
European Union Programme of high level Scholarship for white-sided dolphins in the Eastern North Atlantic. In: Evans PGH,
Latin America (identification number EO3D17203CO) and Teilmann J, editors. ASCOBANS/HELCOM small cetacean population
structure workshop. Bonn (Germany): ASCOBANS.
United Nations Environmental Program/Agreement on the
Conservation of Small Cetaceans of the Baltic, North East Excoffier L, Laval G, Schneider S. 2005. Arlequin ver. 3.0: an integrated
software package for population genetics data analysis. Evol Bioinform
Atlantic, Irish and North Seas. Samples were collected with
Online. 1:47–50.
funding from the Heritage Council of Ireland, National
Geraci JR, Testaverde SA, St Aubin DJ, Hoop TH. 1978. A mass stranding
Parks and Wildlife Service, INTERREG Ireland-Wales and
of the Atlantic white-sided dolphin, Lagenorhynchus acutus : a study into
the EC-funded BIOCET programme. pathobiology and life history. Washington (DC): Marine Mammal
Commission.
Gerber S, Chabrier P, Kremer A. 2003. FAMOZ: a software for parentage
Acknowledgments analysis using dominant, codominant and uniparentally inherited markers.
The authors would like to thank all the people who assisted with data Mol Ecol Notes. 3:479–481.
collection. Dr Christophe Herbinger helped with the use of the PEDIGREE Gerber S, Mariette S, Streiff R, Bodénès C, Kremer A. 2000. Comparison of
software and anonymous reviewers, and Dr Patricia Rosel and Dr Jens microsatellites and AFLP markers for parentage analysis. Mol Ecol.
Carlsson provided useful comments on previous versions of the manuscript. 9:1037–1048.
Goudet J. 2001. FSTAT, a program to estimate and test gene diversities and
fixation indices (version 2.9.3). Switzerland: University of Lausanne.
References Available from: http://www2.unil.ch/popgen/softwares/fstat.htm.
Gowans S, Würsig B, Karczmarski L, David WS. 2007. The social structure
Addink M, Hartmann MG, Couperus B. 1997. A note on life-history
and strategies of Delphinids: predictions based on an ecological framework.
parameters of the Atlantic white-sided dolphin (Lagenorhynchus acutus) from Adv Mar Biol. 53:195–294.

7
Journal of Heredity

Green ML, Herzing DL, Baldwin JD. 2007. Non invasive methodology for McKenzie C, Rogan E, Reid RJ, Wells DE. 1997. Concentrations and
sampling and extraction of DNA from free-ranging Atlantic spotted patterns of organic contaminants in Atlantic white-sided dolphins (Lageno-
dolphins (Stenella frontalis). Mol Ecol Notes. 7:1287–1292. rhynchus acutus) from Irish and Scottish coastal waters. Environ Pollut.
Gresson RAR. 1969. White-sided dolphins, Lagenorhynchus acutus (Gray) 98:15–27.
stranded at Cloghane, Co. Kerry. Ir Nat J. 16:228. McKinnon J. 1994. Feeding habits of the dusky dolphin, Lagenorhynchus
Guo S, Thompson E. 1992. Performing the exact test of Hardy–Weinberg obscurus, in the coastal waters of central Peru. Fish Bull. 92:569–578.
proportion for multiple alleles. Biometrics. 48:361–372. Michaud R. 2005. Sociality and ecology of the odontocetes. Sexual
Hammond PS, Bearzi G, Bjørge A, Forney K, Karczmarski L, Kasuya T, segregation in vertebrates: ecology of the two sexes. Cambridge (UK):
Perrin WF, Scott MD, Wang JY, Wells RS, et al. 2008. Lagenorhynchus acutus. Cambridge University Press. p. 303–326.
IUCN 2009. Published by the International Union for Conservation of Mirimin L, Coughlan J, Rogan E, Cross T. 2006. Tetranucleotide
Nature Red List of Threatened Species. microsatellite loci from the striped dolphin (Stenella coeruleoalba Meyen,
Hammond PS, Berggren P, Benke H, Borchers DL, Collet A, Heide- 1833). Mol Ecol Notes. 6:493–495.
Jørgensen MP, Heimlich S, Hiby AR, Leopold MF, Øien N. 2002. Miyazaki N, Nishiwaki M. 1978. School structure of the striped dolphins off
Abundance of harbour porpoise and other cetaceans in the North Sea and the Pacific coast of Japan. Sci Rep Whale Res Inst. 30:65–115.
adjacent waters. J Appl Ecol. 39:361–367. Morizur Y, Berrow SD, Tregenza NJ, Couperus AS, Pouvreau S. 1999.
Hayano A, Yoshioka M, Tanaka M, Amano M. 2004. Differentiation in the Incidental catches of marine mammals in pelagic trawl fisheries of the
Pacific white-sided dolphin Lagenorhynchus obliquidens inferred from northeast Atlantic. Fish Res. 41:297–307.
mitochondrial DNA and microsatellite analyses. Zool Sci. 21:989–999. Norris KS, Dohl TP. 1980. Behavior of the Hawaiian spinner dolphin,
Stenella longirostris. Fish Bull. 77:821–849.

Downloaded from jhered.oxfordjournals.org by guest on November 30, 2010

Hoelzel AR. 2002. Marine mammal biology: an evolutionary approach.
Malden (MA): Blackwell Publishing. Northridge S, Tasker S, Webb A, Camphuysen K, Leopold M. 1997. White-
Hoelzel AR, Green A. 1998. PCR protocols and population analysis by direct beaked Lagenorhynchus albirostris and Atlantic white-sided dolphin L. acutus
DNA sequencing and PCR-based fingerprinting. Molecular genetic analysis distributions in Northwest European and US North Atlantic waters. Rep
of population, a practical approach. Oxford: Oxford University Press. Int Whal Comm. 47:797–805.
Hoelzel AR, Osborne RW. 1986. Call characteristics of killer whales, Palka D, Read A, Potter C. 1997. Summary of knowledge of white-sided
implications for cooperative foraging. Behavioural ecology of the killer dolphins (Lagenorhynchus acutus) from US and Canadian Atlantic waters. Rep
whale. New York: Liss & Co. Int Whal Comm. 47:729–734.
Jamieson A, Taylor SS. 1997. Comparisons of three probability formulae for Palsboll PJ, Berube M, Larsen AH, Jorgensen H. 1997. Primers for the
parentage exclusion. Anim Genet. 28:397–400. amplification of tri- and tetramer microsatellite loci in baleen whales. Mol
Ecol. 6:893–895.
Jefferson TA, Leatherwood S, Webber MA. 1993. Marine mammals of the
world: FAO species identification guide. Rome (Italy): Food and Pimper LE, Remis MI, Natalie R, Goodall P, Baker CS. 2009. Teeth and
Agriculture Organization of the United Nations. bones as sources of DNA for genomic diversity and sex identification of
Commerson’s dolphins (Cephalorhynchus commersonii) from Tierra del Fuego,
Karczmarski L, Wursig B, Gailey G, Larson KW, Vanderli C. 2005. Spinner
Argentina. Aquat Mamm. 35:330–333.
dolphins in a remote Hawaiian atoll: social grouping and population
structure. Behav Ecol. 16:675–685. Piry S, Luikart G, Cornuet J-M. 1999. BOTTLENECK: a computer
program for detecting recent reductions in the effective population size
Katona SK, Rough V, Richardson DT. 1993. A field guide to whales,
using allele frequency data. J Hered. 90:502–503.
porpoises and seals from Cape Cod to Newfoundland. Washington (DC):
Smithsonian Institution Press. Queller DC, Goodnight KF. 1989. Estimation of genetic relatedness using
allozyme data. Evol. 43:258–275.
Kenney RD, Payne PM, Heinemann DW, Winn HE. 1996. Shifts in
Northeast shelf cetacean distributions relative to trends in Gulf of Maine/ Reeves RR, Smeenk C, Brownell RL Jr., Kinze CC. 1999. Atlantic white-
Georges Bank finfish abundance. The northeast shelf ecosystem: sided dolphin Lagenorhynchus acutus (Gray, 1828). Handbook of marine
assessment, sustainability, and management. Cambridge (MA): Blackwell mammals, volume 6: the second book of dolphins and the porpoises.
Science. p. 169–196. San Diego (CA): Academic Press. p. 31–56.
Kinze CC, Addink M, Smeenk C, Hartmann MG, Richards HW, Sonntag Reid JB, Evans PGH, Northridge S. 2003. Atlas of cetaceans distribution in
RP, Benke H. 1997. The white-beaked dolphin (Lagenorhynchus albirostris) and north-west European waters. Peterborough (UK): Joint Nature Conserva-
the white-sided dolphin (Lagenorhynchus acutus) in the North and Baltic seas: tion Committee (JNCC).
review of available information. Rep Int Whal Comm. 47:675–682. Rogan E, Baker JR, Jepson PD, Berrow S, Kiely O. 1997. A mass stranding
Krützen M, Barré LM, Möller LM, Heithaus MR, Simms C, Sherwin WB. of white-sided dolphins (Lagenorhynchus acutus) in Ireland: biological and
2002. A biopsy system for small cetaceans: darting success and wound pathological studies. J Zool. 242:217–227.
healing in Tursiops spp. Mar Mamm Sci. 18:863–878. Rogan E, Gassner I, Mackey M, Fennelly S, McDonnell M. 2002. White-
Luikart G, Sherwin WB, Steele BM, Allendorf FW. 1998. Usefulness of sided dolphin Lagenorhynchus acutus (Gray). Ir Nat J. 27:171–172.
molecular markers for detecting population bottlenecks via monitoring Rosel PE. 2003. PCR-based sex determination in Odontocete cetaceans.
genetic change. Mol Ecol. 7:963–974. Conserv Genet. 4:647–649.
MacLeod CD, Bannon SM, Pierce GJ, Schweder C, Learmonth JA, Sambrook J, Fritsch EF, Maniatis T. 1989. Molecular cloning: a laboratory
Herman JS, Reid RJ. 2005. Climate change and the cetacean community of manual. Cold Spring Harbor (NY): Cold Spring Harbor Laboratory Press.
north-west Scotland. Biol Conserv. 124:477–483. Schwacke L, Schwacke J, Rosel P. 2005. RERAT: relatedness estimation
MacLeod K. 2004. Abundance of Atlantic white-sided dolphin (Lageno- and rarefaction analysis tool. Available from: http://people.musc.edu
rhynchus acutus) during summer off northwest Scotland. J Cetacean Res /;schwaclh/.
Manag. 6:33–40. Sergeant DE, St Aubin DJ, Geraci JR. 1980. Life history and Northwest
Mann J, Connor RC, Tyack PL, Whitehead H. 2000. Cetacean societies: field Atlantic status of the Atlantic white-sided dolphin, Lagenorhynchus acutus
studies of dolphins and whales. Chicago: The Univeristy of Chicago Press. (Gray). Cetology. 37:1–12.

8
 Mirimin et al.  Genetic Diversity, Parentage, and Group Composition of Atlantic White-Sided Dolphins

Slatkin M, Excoffier L. 1996. Testing for linkage disequilibrium in genotypic Wall D, O’Brien J, Meade J, Allen BM. 2006. Summer distribution and
data using the EM algorithm. Heredity. 76:377–383. relative abundance of cetaceans off the west coast of Ireland. Biol Environ
Smith TG, Hammill MO, Martin AR. 1994. Herd composition and Proc R Ir Acad. 106B:135–142.
behaviour of white whales (Delphinapterus leucas) in two Canadian arctic Waring GT, Josephson E, Fairfield CP, Maze-Foley K. 2006. U.S. Atlantic
estuaries. Bioscience. 39:175–184. and Gulf of Mexico marine mammal stock assessments—2005. Woods
Thompson JD, Gibson TJ, Plewniak F, Jeanmougin F, Higgins DG. 1997. Hole (MA): NOAA Technical Memorandum NMFS-NE. p. 346.
The ClustalX windows interface: flexible strategies for multiple sequence Weinrich M, Belt C, Morin D. 2001. Behavior and ecology of the Atlantic
alignment aided by quality analysis tools. Nucl Acids Res. 24:4876–4882. white-sided dolphin (Lagenorhynchus acutus) in coastal New England waters.
Valsecchi E, Hale P, Corkeron P, Amos W. 2002. Social structure in Mar Mamm Sci. 17:231–248.
migrating humpback whales. Mol Ecol. 11:507–518. Weisbrod AV, Shea D, Moore MJ, Stegeman JJ. 2001. Species, tissue and
van Oosterhout C, Hutchinson WF, Willis DPM, Shipe P. 2004. MICRO- gender-related organochlorine bioaccumulation in white-sided dolphins,
CHECKER: software for identifying and correcting genotyping errors in pilot whales and their common prey in the northwest Atlantic. Mar Environ
microsatellite data. Mol Ecol Notes. 4:536–538. Res. 51:29–50.
Vanmann C, Bogomolni A, Moore MJ, Touhey K. 2005. The spatio-temporal Whitehead H, Weilgart L. 2000. The sperm whale: social females and roving
distribution and oceanographic correlations of Atlantic white-sided (Lageno- males. Cetacean societies: field studies of dolphins and whales. Chicago:
rhynchus acutus) and common dolphin (Delphinus delphis) strandings in Cape University of Chicago Press. p. 154–173.
Cod waters, Massachusetts, USA. In: 16th Biennal Conference on the Biology Würsig B, Wells RS, Norris KS, Würsig M. 1994. The Hawaiian spinner
of Marine Mammals. San Diego (CA): The Society for Marine Mammalogy. dolphin. Berkeley (CA): University of California Press.
Viricel A, Strand A, Rosel P, Ridoux V, Garcia P. 2008. Insights on

Downloaded from jhered.oxfordjournals.org by guest on November 30, 2010

common dolphin (Delphinus delphis) social organization from genetic analysis Received April 6, 2010; Revised August 18, 2010;
of a mass-stranded pod. Behav Ecol Sociobiol. 63:173–185. Accepted August 26, 2010
Waits LP, Luikart G, Taberlet P. 2001. Estimating the probability of identity
among genotypes in natural populations: cautions and guidelines. Mol Ecol.
10:249–256. Corresponding Editor: Scott Baker