Molecular Ecology 1997, 6, 21–28

Is the black robin in genetic peril?

S . L . A R D E R N * † and D . M . L A M B E R T * §
* Ecology and Evolution and the Centre for Conservation Biology, School of Biological Sciences, University of Auckland, Private Bag
92019, Auckland, New Zealand

Abstract
In 1980 the entire black robin species Petroica traversi comprised only five birds, and the
current population of P 200 individuals is known to be derived from a single breeding
pair. We show here that levels of minisatellite DNA variation in the black robin are
among the lowest reported for any avian species in the wild. Surprisingly, similarly bot-
tlenecked control populations of a closely related species (P. australis australis) exhibit
significantly higher levels of genetic variation. This suggests that the black robin’s
persistence in a single small population for the last 100 years, rather than the recent bot-
tleneck itself, accounts for the low genetic variation observed. Despite apparent genetic
impoverishment, survival and reproductive performance indicate that the black robin is
viable under existing conditions. This example illustrates that significant levels of genet-
ic variation are not a necessary prerequisite for endangered species’ survival.
Keywords: conservation genetics, minisatellite DNA, population bottlenecks

Received 1 March 1996; revision accepted 14 June 1996

effects of population bottlenecks, founder effects, random

Introduction
drift and inbreeding (Frankel & Soulé 1981).
Understanding the dynamics of genetic variation in small Although the evidence of inbreeding depression is
populations and its importance for population viability compelling, different species clearly show variable
constitutes a long-standing and central focus of research in responses to inbreeding and the loss of genetic variation
evolutionary biology (Fisher 1930; Wright 1931). (Wayne et al. 1991). Because the interpretation of reduced
Population genetic theory predicts that in small popula- performance in captive and managed wild populations is
tions inbreeding is likely to occur with a concomitant loss confounded by the effects of the captive environment and
of genetic variation (Nei et al. 1975; Wright 1978). The zoo practice, studies of wild populations are urgently
potentially negative effects of inbreeding in small popula- needed.
tions have also been of major concern in conservation The cheetah has been widely accepted as a classic
biology (Franklin 1980; Frankel & Soulé 1981). This example illustrating the importance of genetic variation
interest stems from evidence of the deleterious ‘fitness for maintaining population viability (O’Brien et al. 1985;
consequences’ of inbreeding documented in laboratory Yuhki & O’Brien 1990). O’Brien et al. (1983, 1985) argued
and captive populations of typically outbred wild animals, that low levels of genetic variation in cheetahs is correlat-
e.g. Falconer & Roberts (1960) and Ralls et al. (1979). The ed with a range of physiological abnormalities and there-
decreased vigour observed in such inbred populations fore is consistent with a causal relationship between low
may manifest itself in the form of reduced survival, population densities and poor breeding performance in
reduced fecundity and increased susceptibility to disease, captivity. A range of problems plagued cheetah captive-
and has been termed ‘inbreeding depression’. In small breeding programmes such as low fecundity, a low con-
populations inbreeding depression is typically correlated ception rate and high infant mortality rates. In general
with a decrease in genetic variation resulting from the poor sperm quality was recorded, with a high proportion
(71%) of sperm being abnormal in shape. Moreover, the
†Present address: School of Medicine, The University of concentration of sperm in cheetahs was one-tenth of that
Aukland, Private Bag 92019, Aukland, New Zealand. recorded for domestic cats. O’Brien et al. (1986) pointed
§Present address: Department of Ecology, Massey University, out that such levels of sperm abnormalities in other
Private Bag 11-222, Palmerston North, New Zealand. Tel.: +64-6- species is associated with infertility and can be a conse-
356-9099, Fax:+64-6-350-9099. E-mail: [email protected] quence of inbreeding in laboratory animals and livestock.

© 1997 Blackwell Science Ltd

22 S.L.ARDERN AND D.M. LAMBERT

However, it has been suggested that the poor popula- management techniques, including cross-fostering (a man-
tion parameters shown by captive and wild populations of agement tool applied only until 1988), the species has
cheetahs have more to do with ecological factors than made a remarkable recovery. The black robin species cur-
genetic ones (Caro & Laurenson 1994). It is clear that, in rently comprises P 200 birds distributed in two island
captive situations the low breeding success is due to ani- populations (Butler & Merton 1992).
mal husbandry practices for example, inappropriate social Our analysis of genetic variation in the endangered
conditions and a failure to detect oestrus in females. In black robin details current levels of minisatellite DNA
addition, for cheetahs in the wild, the major cause of cub variation. We compare the observed genetic variation in
mortality is predation by lions and hyenas (Caro 1987). black robins with closely related and similarly bottle-
Caro & Laurenson (1994) remarked that: ‘These findings necked New Zealand bush robin P. australis australis pop-
suggest that genetics may have been overemphasized in ulations and examine the association between genetic vari-
relation to the plight of the cheetah’. In fact, in the wild a ation and population viability in the black robin using
large proportion of female cheetahs breed and there is a available data.
high rate of litter production. This implies that the repro-
ductive anatomy and physiology of cheetahs are not
Materials and methods
adversely affected by their genetic constitution.
Nevertheless, phenomena such as founder events and
DNA methods
population bottlenecks are likely to play a role in reducing
genetic variation in cheetahs and other species (Wildt et al. Genetic variation in black robin and bush robin blood
1987; Packer et al. 1991; Hoelzel et al. 1993), the historical samples was assayed by hybridizing human-derived min-
evidence for such events is often lacking. isatellite DNA probes (Jeffreys et al. 1985; Longmire et
We report here the study of a wild avian population al. 1990) which typically produce hypervariable, individ-
which is known to have undergone a recent and severe ual-specific DNA profiles in birds (Burke & Bruford 1987;
population bottleneck. The Chatham Island black robin Wetton et al. 1992). We extracted high molecular weight
Petroica traversi is one of the world’s most endangered and DNA from whole blood using the following technique.
intensively inbred avian species surviving in the wild. Fifteen microlitres of blood was lysed by resuspension in
This small insectivorous passerine is found only on 400 µL of lysing solution (144 mM NH4Cl; 10 mM
Mangere and South East (Rangatira) Islands in the NH4HCO3). The lysate was centrifuged at 7500 r.p.m. for
Chatham Island group, 750 km east of the South Island of P 15 min and the supernatant discarded. The resulting
New Zealand (18 ha) (Fig. 1). Although it once occupied pellet was then resuspended in 400 µl of SET buffer (0.1 M
most of the larger islands in the Chatham group, for most NaCl, 1 mM EDTA, 0.1 M Tris-HCl pH 8.0) to which sodi-
of this century the black robin has been isolated in a single um dodecyl sulphate (20 µL, 10% SDS) and proteinase K
population on Little Mangere Island (Fig. 1). It is unlikely (10 µL, 20 mg/mL) were added, and incubated 65 °C
that this small island would have supported more than overnight. We employed this method in order to remove
20–30 individuals at any one time over the last 100 years the bulk of the haemoglobin before proceeding and higher
(Merton 1990). In the 1970s, the species underwent a rapid quality DNA resulted when this step was included. DNA
and severe population decline primarily as a result of was then extracted and purified using standard
habitat degeneration (Flack 1975). Five individuals phenol/chloroform methods and precipitation and resus-
remained in 1979 and 1980 (Flack 1975, 1978; Merton 1990, pension of DNA was performed in accordance with
1992; Butler & Merton 1992). (Sambrook et al. 1989).
The black robin is highly inbred as a result of this long DNA samples of P 20 µg were digested overnight with
history of isolation, and the extreme population bottleneck the restriction enzyme HaeIII (15 units) in the presence of
which occurred in the late 1970s and early 1980s when a spermidine trihydrochloride and bovine serum albumin
single pair bred successfully. Matings between siblings, (BSA, 2 mg/mL) with the manufacturer’s recommended
and between parents and offspring, have been common- buffer. After this interval we added a further 15 units of
place (Merton 1990) and all existing birds are thought to be HaeIII and incubation continued for approximately 1 h.
descended from a single black robin female (Flack 1975; HaeIII digested DNA was then electrophoresed through
Merton 1990, 1992; Butler & Merton 1992). 0.8% agarose gels in TBE buffer (134 mM Tris, 74.9 mM
The potential effects of intense inbreeding on popula- boric acid, 2.55 mM EDTA pH 8.8) until fragments smaller
tion viability have been of concern throughout the black than 3 kb were run off the gel (P 58 h). After Southern
robin’s recovery programme (Flack 1975; Merton 1983, blotting, probes 33.15, 33.6 and pV47-2 were labelled with
1990; Cemmick & Veitch 1985; Butler & Merton 1992). α32P dCTP by random priming. Hybridization was carried
However, as a result of interisland transfers to better habi- out as described by Westneat (Westneat 1990), except that
tat and enhanced productivity achieved using close-order hybridization temperatures and final wash stringencies

© 1997 Blackwell Science Ltd, Molecular Ecology, 6, 21–28

 CONSERVATION GENETICS OF BLACK ROBIN 23

Fig. 1 The Chatham Islands in relation to

mainland New Zealand showing the loca-
tion of Mangere and South East Islands,
the habitat of the endangered Chatham
Island black robin; Petroica traversi.

used were 61 °C (33.15), or 55 °C (33.6, pV47-2) and scored [33.15 and 33.6 only; proportion of fragments detect-
0.5 u SSC, 0.1% SDS (33.15, 33.6) or 5 u SSC, 0.1% ed by both probes was 0.02 ± 0.03 (SD)]. When they
SDS (pV47-2). DNA profiles were visualized by occurred, overlapping fragments were scored only once.
autoradiography. Percentage differences between pairwise comparisons
The molecular weight regions scored on profiles pro- of DNA profiles of individuals was calculated according
duced with each probe were as follows: > 4 kb (33.15, to Gilbert et al. (1991) and Packer et al. (1991). The average
l HindIII-1 kb); > 5 kb (pV47-2); > 8 kb (33.6). Bands of sim- of all pairwise values, average percentage difference
ilar mobility were considered to be identical if they dif- (APD) was used for each population as an index of within
fered less than twofold in intensity. Bin sizes were population genetic variation (Gilbert et al. 1991). Two ran-
assigned according to curvature observed for internal lane domization techniques were used to investigate the statis-
marker fragments across a gel, being 1 mm and 2 mm for tical significance of differences in APDs of P. traversi and
fragments u > 6 kb, and 6 kb > u > 4 kb, respectively. P. australis australis (Holmes 1994). These two techniques
There were few overlapping DNA fragments among pro- each incorporate different assumptions regarding the
files produced with the different probes in the regions degree of independence between fragments in a DNA

© 1997 Blackwell Science Ltd, Molecular Ecology, 6, 21–28

24 S.L.ARDERN AND D.M. LAMBERT

fingerprint. The first procedure generates bootstrapped Results

pseudosamples with the same number of individuals as in
Minisatellite DNA variation in black robins is among the
the original sample, by sampling individual DNA profiles
lowest reported for any wild avian species. Multilocus
with replacement. In the construction of a pseudosample
DNA profiles produced for a panel of 14 black robin indi-
all bands in each DNA fingerprint are therefore treated as
viduals showed a remarkable degree of similarity
if they are linked, hence this procedure is called the Fixed
(Fig. 2a), being almost identical from one individual to the
Covariance Model (FCM). The second randomization pro-
next. Furthermore, of the small number of variable frag-
cedure generates pseudosamples by sampling each differ-
ments recorded, approximately 30% were attributable to
ent fragment detected in the original DNA fingerprints
sex-specific differences. The APD of 15.8 ± 7.5 (SD)
with replacement across all individuals. The status of each
between black robin individuals (Table 1) contrasts with
unique fragment in the original sample (i.e. presence or
APDs exceeding 70 and 80 recorded for the majority of
absence of the fragment) is sampled with replacement
outbred avian species studied to date, such as Hispaniolan
across all individuals, until all individuals in the pseu-
parrots Amazona ventralis (Brock & White 1992), tropical
dosample have been identified as either possessing or not
wrens Campylorhynchus nuchalis (Rabenold et al. 1990), and
possessing that particular fragment. This method – the
house sparrows Passer domesticus (Burke & Bruford 1987;
Independent Assortment Model (IAM) – assumes that all
Wetton et al. 1992). Black robins exhibit even more simi-
fragments in the DNA fingerprints of the original sampled
larity than inbred New Zealand pukeko Porphyrio por-
individuals are independent of each other, and therefore
phyrio melanotus in which typical mean APDs are P 40
free to assort independently. Unpaired two-tailed t-tests
(Lambert et al. 1994).
were used to determine the statistical significance of dif-
Interpretation of genetic variation in context of the late
ferences in numbers of fragments detected per individual
1970s and early 1980s bottleneck and subsequent inbreed-
in P. traversi compared with populations of P. australis
ing is complicated by the absence of pre-bottleneck
australis.

Fig. 2 Minisatellite DNA profiles of Petroica spp. produced by hybridization of 33.15 with HaeIII digested DNA. (a) P. traversi; (b) P. aus-
tralis australis, B1; (c) P. australis australis, S1. B1 and S1 are a pair of populations related by a translocation event, B1 being the bottle-
necked daughter population and S1 being its source population. P. traversi individuals sampled were second order relatives or more dis-
tantly related. Individuals of B1 and S1 represent a random population sample. The molecular weight range is indicated at left of the
figure.

© 1997 Blackwell Science Ltd, Molecular Ecology, 6, 21–28

 CONSERVATION GENETICS OF BLACK ROBIN 25

Table 1 Summary statistics of minisatellite DNA variation in Petroica populations derived from DNA profiles of HaeIII digests hybridized
with probes 33.15, 33.6, and pV47-2. The notation S1, B1, S2, B2 refers to pairs of source and bottlenecked daughter populations related by
translocation events. Average percentage difference (APD) was calculated as the average of all pairwise difference values within the pop-
ulation sample as described by (Gilbert et al. 1991). Heterozygosity was estimated using the formula derived previously (Stephens et al.
1992). Calculation of the values presented was facilitated by use of the computer programme ‘Thumbprint’ (Marshall 1992)

Total no.
Population Mean novel.
(no. individuals/ number of fragments
no. pairwise Average fragments/ scored in Proportion of
Species comparisons) APD (SD) heterozygosity individual (SD) population fixed fragments

P. traversi (14/91) 15.8 (7.5) 0.20 18.7 (2.0) 24 0.375

P. australis australis S1 (15/105) 55.4 (8.7) 0.65 31.6 (4.0) 95 0.011

B1 (17/136) 39.4 (7.3) 0.52 30.9 (2.8) 65 0.031
S2 (12/66) 49.2 (9.1) 0.59 32.5 (3.7) 86 0.023
B2 (14/91) 43.1 (8.9) 0.54 29.2 (2.9) 65 0.031

estimates of variability in the black robin. However, data the lowest level of genetic variation among P. australis aus-
from several populations of a related species, the New tralis populations studied). These data indicate that the
Zealand bush robin, P. australis australis, were available for extremely low minisatellite DNA variation observed in
comparison. These represent two sets of source and black robins is not a general feature of the genus Petroica
translocated populations, each daughter population and specifically is not a feature of similarly bottlenecked
established from five individuals in 1973 (Flack 1978) – populations of a congeneric species.
essentially control populations for the black robin bottle-
neck event of around 1980.
Discussion
Figure 2(b) and 2(c) together with measures of variabil-
ity presented in Table 1 show markedly lower genetic Comparison of genetic variation among individuals of P.
variation among black robins than among individuals of traversi and P. australis australis suggests that the low level
P. australis australis from both the source and bottlenecked
populations. APDs observed within P. australis australis
populations ranged between 39.4 (± 7.3) and 55.4 (± 8.7)
and were considerably higher than the equivalent measure
for black robins. It appears that the effects of reduced pop-
ulation size may be evident even in the P. australis australis
source populations. Higher APDs of P 80 have been esti-
mated for a population of North Island robins P. a. longipes
(Holmes 1994). None the less, differences in APDs
between P. traversi and both bottlenecked populations of
P. australis australis were each statistically significant using
both the FCM and IAM methods (P < 0.001).
The mean number of fragments scored per individual,
the other parameter amenable to statistical analysis, also
proved to be significantly lower among black robins, in
comparison with the least variable bottlenecked popula-
tion, Motuara (P < 0.001). In addition, the proportion of
fixed fragments detected was over an order of magnitude
higher among P. traversi than P. australis australis. Figure 3
shows the detection of novel fragments with increasing
number of fragments sampled, illustrating the exceptional Fig. 3 Discovery curves for Petroica traversi and P. australis australis
(bottlenecked population B1) showing cumulative numbers of
lack of variation among black robins. Few new fragments
new fragments detected with increasing number of fragments
were discovered in this species after four individuals had sampled. Combined results for HaeIII digests hybridized with
been scored. In contrast, the curve continues to rise until probes 33.15, 33.6 and pV47-2 are given. Squares represent P. aus-
P 10 individuals are sampled for robins of B1 (bottle- tralis australis individuals (n = 17) and dots represent P. traversi
necked daughter population number one which exhibited individuals (n = 15).

© 1997 Blackwell Science Ltd, Molecular Ecology, 6, 21–28

26 S.L.ARDERN AND D.M. LAMBERT

of minisatellite variation currently observed in black robin dence of genetic variation quantified at different genetic
individuals is unlikely to be attributable to the 1980 bottle- loci (Yuhki & O’Brien 1990). Given the probable high level
neck alone. Although a trend towards the loss of genetic of homozygosity at other loci, one might anticipate a
variation through bottleneck events was observed in P. strong association between reduced fitness and low genet-
australis australis, moderate levels were maintained in both ic variation in this species. However, there is little clear
bottlenecked populations, including one in which effective evidence of the deleterious consequences of inbreeding
population size is limited to P 60 birds due to habitat size associated with low genetic variation in the black robin.
(Holmes 1994). Therefore, it seems likely that a long peri- A range of biological indicators, including survival and
od of inbreeding in a single small population, rather than reproductive performance measures suggest that the black
the 1980 bottleneck itself, has been the major determinant robin population was viable under the managed condi-
of the low variation observed in black robins. tions, prior to 1988. For example, egg fertility of P 90%,
Because minisatellite DNA is typically noncoding and hatchability of P 70%, survival to one year of 75% and
therefore selectively neutral, there is not likely to be any adult mortality of less than 30% are typical (Flack 1975;
functional significance of reduced levels of genetic varia- Merton 1990; Butler & Merton 1992). Since intensive man-
tion at minisatellite loci. However, historical evidence of agement ceased in 1988, the species has, however, contin-
long-term isolation and small population size suggests ued to increase and current numbers are P 200.
that genetic variation in other regions of the black robin Furthermore, black robins have not displayed any particu-
genome is also likely to be low (unless maintained by lar susceptibility to disease, although cases of avian pox
overdominance), particularly as minisatellite DNA is and infestations of parasitic mites have been recorded in
amongst the most variable class of genetic elements black robins and their congeners (Butler & Merton 1992;
known. Several authors have presented correlative evi- Flack 1979).

Table 2 Survival and productivity of Chatham Island black robins and New Zealand bush robins. New Zealand bush robin data were
predominantly collected at a mainland location, unless otherwise specified

Black robin Bush robin

(Petroica traversi) References (P. australis australis) References

Number of clutches 1, occasionally 2 (Merton 1990) 1–2, occasionally (Flack 1979)

(Butler & Merton 1992) 3 or 4

Clutch size 2 (Merton 1990) 2 (islands) (Maloney 1991, Flack 1974)

(range 1–3) 2.75 (Flack 1974)
(range 1–4) (Flack 1979)

Percentage of eggs 67.5 (Butler & Merton 1992) 83‡ (Powlesland 1983)
hatched (SD) (14.9)*

Percentage of 70.3 (Butler & Merton 1992) 42‡ (Powlesland 1983)

nestlings fledged (7.6)*
(SD)

Young/pair 1.5✝ (Butler & Merton 1992) 3.7 (Flack 1973)

(SD) (independent juvs) 2.5 (Powlesland 1983)
(0.25) (Fledged young)

Per cent survival 75.6 (Butler & Merton 1992) 9.4z (Flack 1973)
to 1 yr (SD) (14.2)** 16.9‡ (Powlesland 1983)

Per cent annual 29.4 (Flack 1974) 28.2 (range 23–37) (Flack 1973)
adult mortality 21.6** (Butler & Merton 1992) 61.4‡ (Flack 1974)
(SD) (11.5) (Powlesland 1983)

* Data from 12 seasons (1980/81–1991/92).

** Data from 9 seasons (1980/81–1988/89).
✝ Data from unmanaged seasons only (1989/90–1991/92).
‡ Data from 2 seasons (1977/78–1978/79).
z Data from a single season only (1971/1972).

© 1997 Blackwell Science Ltd, Molecular Ecology, 6, 21–28

 CONSERVATION GENETICS OF BLACK ROBIN 27

Table 2 shows available data on survival and produc- range of species require that the genetic consequences of
tivity of black and bush robins. It should be noted that small population size remain a cause of concern for the
bush robins have been most extensively studied on the viability of endangered species. However, the example of
New Zealand mainland rather than on islands, typically the black robin illustrates that, although genetic variation
exhibiting higher reproductive output than island popula- in such populations may indeed be very low, this may not
tions (Flack 1973, 1979; Powlesland 1983; cf. Flack 1974; be associated with low population viability. This contrasts
Maloney 1991). However, survival on the mainland is con- with the findings and interpretation of the cheetah exam-
siderably reduced due to the presence of introduced mam- ple and indicates the need for further studies of this kind.
malian predators such as stoats, cats and rats (Flack 1974). In conservation biology a cautious approach stimulated by
According to Butler and Merton (Butler & Merton 1992) awareness of reported cases of inbreeding depression
the only factor which raises suspicions of inbreeding must be balanced by the knowledge that intense inbreed-
depression in black robins is that P 35% of eggs laid fail to ing and low levels of genetic variation do not necessarily
hatch (Table 2). Published estimates indicate that a lower prohibit the continued survival of endangered species.
proportion of eggs fail to hatch among bush robins (17%)
(Table 2) (Powlesland 1983). The larger proportion of black
Acknowledgements
robin eggs that fail to hatch appears to be attributable pri-
marily to females laying eggs on nest rims, rather than in This research was supported by grants from the Auckland
the nest itself. It has been suggested that the number of University Research Committee and New Zealand Lottery
eggs ‘laid over the rim’ may constitute an abnormality Science to D.M.L. and S.L.A. In addition, the World Wide Fund
for Nature provided support for minisatellite DNA analyses in
among black robin females (Butler & Merton 1992).
the form of a grant to D.M.L. We thank Scott Baker, John Craig,
Interpreting its significance is difficult as similar cases of Ian McLean, Craig Millar, Judith Robins, Allen Rodrigo, Peter
reproductive failure have been recorded among several Ritchie, Euan Young and the New Zealand Department of
bush robin pairs on Tiritiri Matangi Island (recently estab- Conservation, in particular Don Merton and Euan Kennedy, for
lished from a larger more outbred population compared very helpful discussions.
with black robins), resulting from females building nests
in unstable sites (Armstrong, personal communication). It
References
should be pointed out, however, that a variable that was
measured in recent years (young/pair) shows a lower rate Brock MK, White BN (1992) Application of DNA fingerprinting to
in black robins than bush robins (Table 2). the recovery program of the endangered Puerto Rican parrot.
Proceedings of the National Academy of Sciences USA, 89,
At least in the short term, providing there are no natur-
11121–11125.
al catastrophic events, the species appears to be viable and Burke T, Bruford MW (1987) DNA fingerprinting in birds. Nature,
its continued independent survival is anticipated, perhaps 327, 149– 152.
owing to the elimination of deleterious recessive alleles (cf. Butler D, Merton D (1992) The Black Robin: saving the World’s Most
Templeton & Read 1983). Similarly, reduced levels of Endangered Bird. Oxford University Press, Auckland, New
genetic variation have been documented in apparently Zealand.
healthy reintroduced populations of European beavers Caro TM (1987) Cheetah mother’s vigilance: looking out for prey
or for predator’s? Behavioural Ecology and Sociobiology, 20,
which appear to be increasing in numbers (Ellegren et al.
351–361.
1993). Results such as those detailed here for black robins Caro TM, Laurenson MK (1994) Ecological and genetic factors in
may be common for populations that have a long history conservation: a cautionary tale. Science, 263, 485–486.
of small population size (Lambert 1995; Lambert & Millar Cemmick D, Veitch D (1985) Black Robin Country. Hodder and
1995). Interestingly, we have detected similar low levels of Stoughton, Auckland.
minisatellite DNA variation in the Campbell Island teal Ellegren H, Hartmen G, Johansson M, Andersson L (1993) Major
Anas aucklandica nesiotis which is restricted to 50–100 birds histocompatibility complex monomorphism and low levels of
DNA fingerprinting variability in a reintroduced and rapidly
on Dent Island (a 10–15-ha islet off the west coast of
expanding population of beavers. Proceedings of the National
Campbell Island) in the Southern Ocean (D. M. Lambert et Academy of Sciences USA, 90, 8150–8153.
al. unpublished data). This population has been confined Falconer DS, Roberts RC (1960) Effect of inbreeding on
there for at least the past 150 years and the bird was never ovulation rate and foetal mortality in mice. Genetical Research,
recorded on Campbell Island itself by any of the ninteenth 1, 422–430.
century naturalists who visited the island (M. Williams, Fisher RA (1930) The Genetical Theory of Natural Selection. Claredon
personal communication). Press, Oxford.
Flack JAD (1973) Robin research – a progress report. Wildlife – A
Many species of conservation interest subject to phe-
Review, 4, 28–36.
nomena such as population bottlenecks may be undergo- Flack JAD (1974) Chatham Island black robin Wildlife – A Review,
ing reductions in genetic variation. Reported cases of 5, 25–34.
reduced fitness associated with low genetic variation in a Flack JAD (1975) The Chatham Island black robin: extinction or

© 1997 Blackwell Science Ltd, Molecular Ecology, 6, 21–28

28 S.L.ARDERN AND D.M. LAMBERT

survival? Bulletin of The International Council for Bird Packer C, Pusey AE, Rowley H, Gilbert DA, Martenson J, O’Brien
Preservation, 12, 146–150. SJ (1991) Case study of a population bottleneck: lions in the
Flack JAD (1978) Interisland transfers of New Zealand black Ngorongoro Crater. Conservation Biology, 5, 219–230.
robins. In: Endangered Birds: Management Techniques for the Powlesland RG (1983) Breeding and mortality of the South Island
Preservation of Threatened Species (ed Temple SA), pp. 365–372. robin in Kowhai Bush, Kaikoura. Notornis, 30, 265–282.
University of Wisconsin Press, Madison. Rabenold PP, Rabenold KN, Piper WH, Haydock J, Zack
Flack JAD (1979) Biology and ecology of the South Island robin. SW (1990) Shared paternity revealed by genetic
In: Ecology of Kowhai Bush, Kaikoura, Mauri Ora 2, Special analysis in cooperatively breeding tropical wrens. Nature, 348,
Publication (eds Hunt DM, Gill BJ), pp. 22–26. University of 538–540.
Canterbury Biological Society, Christchurch. Ralls K, Brugger K, Ballou J (1979) Inbreeding and juvenile mor-
Frankel OH, Soulé ME (1981) Conservation and Evolution. tality in small populations of ungulates. Science, 206,
Cambridge University Press, Cambridge. 1101–1103.
Franklin IR (1980) Evolutionary changes in small populations. In: Sambrook J, Fritsch EF, Maniatis T (1989) Molecular Cloning: A
Conservation Biology: an Evolutionary–Ecological Perspective (eds Laboratory Manual. Cold Spring Harbor Laboratory Press, Cold
Soulé ME, Wilcox BA), pp.135–149. Sinauer Associates, Spring Harbor, New York.
Sunderland, Massachusetts. Stephens CJ, Gilbert DA, Yuhki N, O’Brien SJ (1992) Estimation of
Gilbert DA, Packer C, Pusey AE, Stephens JC, O’Brien SJ (1991) heterozygosity for single-probe multilocus DNA fingerprints.
Analytical DNA fingerprinting in lions: parentage, genetic Molecular Biology and Evolution, 9, 729–743.
diversity, and kinship. Journal of Heredity, 82, 378–386. Templeton AR, Read B (1983) The elimination of inbreeding
Hoelzel AR, Halley J, O’Brien SJ et al. (1993) Elephant seals, genetic depression in a captive herd of Speke’s gazelle. In: Genetics and
variation and the use of simulation models to investigate Conservation: A Reference for Managing Wild Animal and Plant
historical population bottlenecks. Journal of Heredity, 84, 443–449. Populations (eds Schonewald-Cox CM, Chambers SM,
Holmes SL (1994) Parentage and molecular genetics of New Zealand. MacBryde B, Thomas L), pp. 241–266. Benjamin Cummings,
MSc thesis, University of Auckland. London.
Jeffreys AJ, Wilson V, Thein SL (1985) Hypervariable ‘minisatel- Wayne RK, Lehman N, Girman D et al. (1991) Conservation genet-
lite’ regions in human DNA. Nature, 314, 67–73. ics of the endangered Isle Royale gray wolf. Conservation
Lambert DM (1995) The New Science of Molecular Ecology. New Biology, 5, 41–51.
Zealand Journal of Ecology, 19 (2), 93–96. Westneat DF (1990) Genetic parentage in the Indigo Bunting: a
Lambert DM, Millar CD (1995) DNA Science and conservation. study using DNA fingperprinting. Behavioural Ecology and
Pacific Conservation Biology, 2, 21–38. Sociobiology, 27, 67–76.
Lambert DM, Millar CD, Jack K, Anderson S, Craig JL (1994) Wetton JH, Parkin DT, Carter RE (1992) The use of genetic mark-
Single and multilocus DNA fingerprinting of communally ers for parentage analysis in Passer domesticus (house spar-
breeding pukeko – do copulations or dominance ensure repro- rows). Heredity, 69, 243–254.
ductive success? Proceedings of the National Academy of Sciences Wildt DE, Bush M, Goodrowe C et al. (1987) Reproductive and
USA, 91, 9641–9645. genetic consequences of founding isolated lion populations.
Longmire JL, Kraemer PM, Brown NC, Hardekopf LC, Deaven LL Nature, 329, 328–331.
(1990) A new multi-locus DNA fingerprinting probe: pV47–2 Wright S (1931) Evolution in Mendelian populations. Genetics, 16,
Nucleic Acids Research, 18, 1658. 97–159.
Maloney RF (1991) MSc thesis, University of Canterbury, Wright S (1978) Evolution and the Genetics of Populations, Vol. 4.
Christchurch, New Zealand. University of Chicago Press, Chicago.
Marshall S (1992) Thumbprint v2. School of Biological Sciences, Yuhki N, O’Brien SJ (1990) DNA variation of the mammalian
Victoria University, Wellington, New Zealand. Available from major histocompatibility complex reflects diversity and popu-
‘[email protected]’. lation history. Proceedings of the National Academy of Sciences
Merton D (1983) Chatham Island black robin. Wildlife – A Review, USA, 87, 836–840.
12, 42–46.
Merton D (1990) The Chatham Island black robin. Forest and Bird,
21, 14– 19. This research is part of a continuing programme on the molecular
Merton D (1992) The legacy of Old Blue. New Zealand Journal of ecology of New Zealand wildlife species. In addition to black
Ecology, 16 (2), 65– 68. robins, current studies include: satellite DNA variation in tuatara,
Nei M, Maruyama T, Chakraborty R (1975) The bottleneck effect mitochondrial DNA sequence variation in the kiore rat as a tool to
and genetic variability in popuations. Evolution, 29, 1–10. trace human movements in the Pacific; minisatellite DNA and
O’Brien SJ, Wildt DE, Goldman D, Merril CR, Bush M (1983) The parentage studies of tawaki (Fiordland Crested penguins),
cheetah is depauperate in genetic variation. Science, 221, toutouwai (Robins) and skuas; the effects of population bottle-
459–462. necks on genetic variation in tieke (saddleback); and DNA meth-
O’Brien SJ, Roelke ME, Marker L et al. (1985) Genetic basis for ods for sex assignment in avian species. A number of programs
species vulnerability in the cheetah. Science, 227, 1428–1434. involving Antarctic species such as Adélie penguins and polar
O’Brien SJ, Wildt DE, Bush M (1986) The cheetah in genetic peril. skuas are also ongoing.
Scientific American, 254, 84–92.

© 1997 Blackwell Science Ltd, Molecular Ecology, 6, 21–28