Animal Conservation (2005) 8, 321–328 C 2005 The Zoological Society of London. Printed in the United Kingdom doi:10.1017/S1367943005002283
Environment influences morphology and development
for in situ and ex situ populations of the black-footed
ferret (Mustela nigripes)
Samantha M. Wisely1,†, Rachel M. Santymire2, Travis M. Livieri3, Paul E. Marinari4, Julie S. Kreeger4, David E. Wildt2
and JoGayle Howard5
1 Division of Biology, Kansas State University, Manhattan, Kansas 66506, USA
2 Department of Reproductive Sciences, Conservation & Research Center, Smithsonian’s National Zoological Park, Front Royal, Virginia 22630, USA
3 Prairie Wildlife Research, Wall, South Dakota 57790, USA
4 National Black-Footed Ferret Conservation Center, U.S. Fish & Wildlife Service, Laramie, Wyoming 82070, USA
5 Department of Reproductive Sciences, Conservation & Research Center, Smithsonian’s National Zoological Park, Washington,
District of Columbia 20008, USA
(Received 4 October 2004; accepted 4 January 2005)
For selected species, conservation breeding has become integrated into recovery plans, most often through
the production of offspring for reintroduction into nature. As these programs increase in size and scope, it is
imperative that conservation managers retain the biological integrity of the species. This study investigated the
causes ofmorphological changes that are known to occur in black-footed ferrets (Mustela nigripes)maintained
ex situ. In a previous study, ferrets maintained in captivity were 5–10% smaller in body size than pre-captive,
in situ animals. In the present study, the authors compared nine morphological characters among ex situ
animals and their in situ descendants. Within the ex situ population, cage types were compared to determine
whether housing influenced morphometry. Black-footed ferrets born to reintroduced individuals quickly
returned to their pre-captive size suggesting that a diminutive morphology ex situ did not have a genetic basis.
Furthermore, cage type affected overall body size and shape; ulnas and tibias were as much as 9% shorter
for ex situ animals. The authors hypothesise that small cage size and environmental homogeneity inhibit
the mechanical stimuli necessary for long bone development. These findings have ramifications for ex situ
managers who need to create artificial captive settings that promote natural physical development. In the
absence of such an environment, ‘unnatural’ morphologies can result that may contribute to poor fitness or
perhaps even domestication.
Conservation breeding is a powerful tool for preserving
endangered species (Ebenhard, 1995). Candidates for
this drastic conservation measure are taxa whose in situ
populations are in eminent risk of extinction due to
anthropogenic disturbance (Snyder et al., 1996). Typically,
some or all of the remaining individuals of a species are
taken into captivity with the goal of reinforcing small
isolated in situ populations or reintroducing animals to
previously extirpated regions once a sufficient number of
animals have been produced in captivity (Soul´e et al.,
1986). Because healthy, self-sustaining reintroduced
populations are the ultimate goal of conservation breeding
(Ebenhard, 1995), it is imperative to understand how
†All correspondence to: Samantha M. Wisely, 232 Ackert Hall,
Division of Biology, Kansas State University, Manhattan, Kansas
66506, USA. Tel: 785.532.0978; Fax: 785.532.6653; E-mail:
the captive environment affects reintroduced individuals.
Very little scientific information is available on the
influence of captivity on the behavioural, physiological
and morphological fitness of individuals destined for
Conservation breeding is not a panacea; serious risks
to species’ viability exist both in situ and ex situ (Snyder
et al., 1996). For example, some taxa such as the giant
panda (Ailuropoda melanoleuca) are difficult to breed in
captivity and self-sustaining ex situ populations have, consequently,
been difficult to establish (Wildt et al., 2003),
which creates the potential to diminish the global population
of a species (Conway, 1986). Because animals in
the ex situ environment are often at higher densities than in
situ, the risk of disease epidemics (e.g. cheetahs (Acinonyx
jubatus) and feline infectious peritonitis: Evermann
et al., 1983) and other catastrophic events are also a
significant threat to ex situ populations (Jacobson, 1993).
Other species, such as California condors (Gymnogyps
californianus) are difficult to reintroduce back in the wild,
322 S. M. WISELY ET AL.
because ex situ animals are na¨ıve to potential threats
(Meretsky et al., 2000).
Some risks have a genetic basis. All species are subject
to the loss of genetic diversity as successive generations
are produced in captivity (Lande & Barrowclough, 1987;
Lacy, 1994). Although conservation-breeding programs
focus on retaining the maximal genetic variation,
incremental loss each generation is inevitable (Ballou &
Foose, 1996). The loss of rare alleles and increase in
homozygosity may lead to a decrease in fitness (inbreeding
depression), further inhibiting breeding efforts in captivity
(Ralls & Ballou, 1983) or decreasing the chance for
success in the wild.
Another genetic risk associated with captive breeding
is the domestication of wildlife due to unintentional
selection. Heritable changes to morphology, behaviour
and physiology have all been documented in captive
wildlife, yet these adaptations to the captive environment
are seldom beneficial in the wild (Frankham et al., 1986).
Morphological domestication includes shortened facial
features (Zeuner, 1963), decreased brain size (Kruska,
1996) and smaller body size (Frankham et al., 1986).
These changes can occur in a few generations in spite
of efforts to minimise such forces, because the intensity
of unintentional selection can be very strong. For example,
estimated selection intensity was 10–20 times greater
for captive tigers (Panthera tigris altaica) than for an
idealised free-ranging population (Flesness & Cronquist-
Jones, 1987), which would create ample opportunity for
rapid morphological change.
Few studies have verified phenotypic change as a result
of conservation breeding. A study of museum specimens,
however, documented morphological differences between
historical populations of in situ black-footed ferrets
(Mustela nigripes) collected prior to captive breeding
and black-footed ferrets maintained in captivity (Wisely,
Ososky & Buskirk, 2002a). The study suggested that
skull size decreased and skull shape changed once this
species was brought into captivity, but whether the causes
were environmental or geneticwas undeterminable. In this
paper, we compared morphology between captive blackfooted
ferrets and their in situ descendants to determine
the basis of these changes. If morphometric changes were
heritable, we predict that in situ individuals would remain
the same size and shape as ex situ individuals; however,
if morphometeric changes were due to environmental
factors, in situ individuals would return to the same
size and shape as pre-captive ancestors. Because ex situ
individuals were raised in two types of enclosures, we
examined whether enclosure size affected morphology of
ex situ animals.
The history and status of the black-footed ferret
has been reviewed elsewhere (Seal et al., 1989; Clark,
1994; Dobson & Lyles, 2000). Briefly, all individuals
comprising the black-footed ferret recovery program are
descendants of 8–11 founders (the unknown paternity of
some wild-caught litters created some uncertainty as to
the exact number) and have been bred in captivity for
18 years (7.4 generations with 2.3 years generation time:
Wisely, McDonald & Buskirk, 2003). Genetic variability
is limited (Wisely et al., 2002b) and inbreeding has increased
over time (Wisely et al., 2003), but the rate of
loss has remained below the goal set by the American
Zoo and Aquarium Association’s Species Survival Plan
(SSP) due, in part, to the mean kinship strategy used for
genetic management (Wisely et al., 2003). Medium-sized
carnivores comprise 13% of all SSP species subject to conservation
breeding (see http://www.aza.org/ConScience/
ConScienceSSPList/) and many of these captive populations
are small with vulnerably low amounts of genetic
diversity. The black-footed ferret provides a model for
understanding the implications of ex situ conservation and
the possible ramifications to in situ recovery.
We measured nine morphological characters (defined
in Table 1, below) on 211 black-footed ferrets (n=77
juveniles, n=134 adults) that were living in captivity
(n=129) or in the wild (n=82) between September
2002 and October 2003. For the health and safety of
anaesthetised animals, we chose morphological traits that
minimised their manipulation. To increase measurement
accuracy, we chose morphological traits that minimised
the amount of soft tissue measured. Each animal
was handled to induce and maintain anaesthesia by
inhalation of isoflurane gas. Morphological characters
were measured on each animal’s right side by RMS
using digital calipers (±0.02 mm, Mitutuyo Corporation,
Japan). Because we wanted to minimise the duration of
animal handling, measurements were not repeated. We
excluded individuals from analysis when morphological
structures (exclusively dental traits) were missing or
Ex situ animals were measured during routine health
examinations; these animals were part of a captivebreeding/
reintroduction program. Data on juvenile blackfooted
ferretswere exclusively collected from the National
Zoological Park’s Conservation & Research Center
(CRC, Front Royal, Virginia). Data on adults were
collected from six captive-breeding facilities (National
Black-Footed Ferret Conservation Center, Wheatland,
Wyoming; Cheyenne Mountain Zoo, Colorado Springs,
Colorado; Phoenix Zoo, Phoenix, Arizona; Louisville
Zoo, Louisville, Kentucky; Toronto Zoo, Toronto,
Ontario; and CRC, Front Royal, Virginia). Animals were
raised in two types of enclosures. At CRC, enclosureswere
6×3×3.6m or 6×6×3.6m pens that included a nest
box on a mulch covered floor. Some pens were outdoors
and others were indoors with skylights supplemented
with artificial light. All other facilities held ferrets in
smaller enclosures (1.2×2.4×0.9m)with two nest boxes
connected by a flexible tube.
We evaluated in situ black-footed ferrets during
autumn or spring monitoring surveys at the Conata
Basin/Badlands reintroduction site in the Buffalo Gap
National Grasslands, South Dakota. All in situ animals
measured for this study were born in the wild. We
Morphological change in black-footed ferrets 323
chose to compare this in situ population to the ex
situ population because it was a large, self-sustaining
population (n>200) that had genetic diversity equivalent
to that of the ex situ population (Wisely et al., 2003).
Because genetic diversity was equivalent between in situ
and ex situ populations, we could exclude differences
in genetic diversity, overdominance, inbreeding or
inbreeding depression as a possible cause for differences
in body size. Founders of the Conata Basin population
were representative of the captive-breeding population
(Wisely et al., 2003).
The Conata Basin/Badlands population was established
in 1996, then supplemented with ex situ individuals
from 1997 to 1999 and exponential population growth
was documented through 2003 (Conservation Breeding
Specialist Group, 2003). Trapping and immobilisation
protocols for the black-footed ferret are well established
(Sheets, 1972; Kreeger et al., 1998). Animals were cagetrapped
at night and returned to the same trap location
following examination and recovery from anaesthesia. All
trapping was authorised, coordinated and conducted by
the U.S. Fish and Wildlife Service’s Black-Footed Ferret
Recovery Team as part of routine population monitoring
for reintroduction sites.
We conducted separate analyses for juvenile and adult
animals. All captive-reared juveniles used in this study
came from one location, CRC, and were measured
between September 2002 and October 2003. At the time
of the autumn in situ survey (16–23 September, 2002)
we assumed the age of wild-born juveniles to be 75–
135 (mean=105) days, based on an average estimated
June 1st date of birth (T. Livieri, unpublished data).
We compared these wild-born individuals to captive
individuals that were 104.0±5.7 (mean±SE) days old
for females and 109.3±4.5 days old for males. Although
Vargas & Anderson (1996) reported that females reached
95% of their adult body mass by 105 days of age and
males reached 95% of their adult body mass by 126 days
of age, our pilot studies revealed that morphologically,
individuals<135 days of age were smaller than the
average adult size. Thus, ex situ animals were considered
to be adults when>135 days old. Extensive monitoring
since the year of the first release included marking animals,
which helped us confirm the juvenile status of small,
unmarked animals. We assumed all in situ animals to be
adults at the time of the spring in situ survey (24–30 March
2003). Two individuals that were captured in the autumn
survey were recaptured in the spring; autumn and spring
measurements were used in separate analyses of juvenile
and adult body size.
We tested for normality of the distribution of data for
all nine morphological characters using the Kolmogorov–
Schmirnov test; we corrected for experiment-wise error
due to multiple univariate testing with a sequential
Bonferroni adjustment (Rice, 1989). We excluded the
distance from the second to the fourth mandibular
premolars (P2 to P4) from adult analyses because many
wild adults had missing or damaged P2 teeth. Overall
differences in morphological characters between ex situ
and in situ animals were compared using a multivariate
analysis of variance (MANOVA) with sex and location as
fixed factors. For adults, location was divided into three
treatments: large ex situ enclosures at CRC, small ex situ
enclosures at all other breeding facilities and South Dakota
where animalswere born in situ.We usedTukey’sHonestly
Significant Difference to test for post-hoc differences
 To test the hypothesis that ex situ animals were
a different shape and size than in situ animals, we
reduced the variables to two principal components.
Other morphological studies (e.g. Fleischer & Murphy,
1992; Komers & Komers, 1992) including one for this
species (Wisely et al., 2002a) inferred the first principal
component to represent body size and the second principal
component to be consistent with body shape.
the first two principal components using a MANOVA with
sex and location as fixed factors. We conducted separate
analyses for juveniles and adults. All tests were conducted
at α =0.05.
 Of the nine morphological characters only tibia length
was not normally distributed
. Visual inspection of the
distribution of values revealed a bimodal distribution
influenced by sexual dimorphism. When each sex was
tested for normality separately, the distribution was not
significantly different from normal (male: Kolmogorov–
Schmirnov Z=0.84, P=0.5; female: Z=0.61, P=0.9).
This measure was included, untransformed, in further
In all multivariate analyses, males were significantly
larger than females; test results are not presented because
black-footed ferrets are known to be sexually dimorphic
in body size (Anderson et al., 1986). In juveniles, four
characters were significantly different for animals born at
CRC and wild-born individuals (Table 1). Tail length, ulna
and tibia length were significantly smaller in the ex situ
group, however the distance between mandibular canines
was significantly larger in this group
 In adults, we found no difference in tibia length
between animals born at CRC or in the wild, but animals
from captive-breeding facilities other than CRC had
significantly shorter tibias (Table 2). Compared to the
in situ group, ulna length was shorter for ferrets at all
captive locations, including CRC, and upper canine width
was smaller in CRC-housed animals compared to captive
animals at other facilities or in the wild-born group
(Table 2).We found a significant interaction effect between
sex and location for long bone measurements (Fig. 1); the
difference in size between males in the wild and males in
captivity was greater than between females.
For the juvenile data set, PC1 had an eigenvalue of
5.4 and explained 61% of the variance. PC2 had an
eigenvalue of 1.0 and explained 11% of the variance.
Trends in eigenvector character loadings for PC1 and
324 S. M. WISELY ET AL.
Table 1. Mean±SE (in mm) for nine morphometric measurements from 77 juvenile black-footed ferrets (Mustela nigripes)
UCC LCCa IW NP P24 TAb UCW RLc FAd
Captive (n =14) 10.2±0.08 7.1±0.16 6.8±0.05 10.5±0.11 8.9±0.11 124.8±3.77 3.1±0.08 59.1±0.45 50.2±0.42
Wild (n =24) 10.1±0.10 6.9±0.08 6.8±0.04 10.4±0.08 8.8±0.08 128.4±1.63 3.1±0.04 60.5±0.39 51.8±0.31
Captive (n =18) 11.2±0.02 8.0±0.12 7.3±0.07 11.7±0.12 9.6±0.12 128.5±3.39 3.4±0.07 64.2±0.73 54.7±0.59
Wild (n =21) 11.1±0.09 7.7±0.11 7.3±0.04 11.4±0.11 9.6±0.13 136.9±1.50 3.4±0.04 66.3±0.46 55.9±0.44
UCC, distance from medial edges of maxillary canines; LCC, distance from medial edges of mandibular canines; IW, distance from lateral
edges of I3s; NP, greatest distance between lateral edges of the nose pad; P24, distance from anterior edge of P2 to posterior edge of
P4; TA, distance from the base of the tail to the end of the last vertebrae; UCW, distance from the anterior to the posterior edge of the
maxillary canine; RL, tibia length; FA, ulna length. AMANOVA test for the effect of location (wild versus captive) was significant (Wilks’
Lambda=0.6, F9,73 =4.7, P <0.001).
a For effect of location: F1,73 =5.6, P =0.02.
b For effect of location (captive or wild): F1,73 =5.6, P <0.001.
c For effect of location: F1,73 =10.8, P =0.002.
d For effect of location: F1,74 =9.6, P =0.03.
Table 2. Mean±SD (in mm) for eight morphological measurements from 134 adult black-footed ferrets
UCC LCCa IW NP TA UCWb RLc FAd
CRC captive (n =8) 10.3±0.07 7.2±0.17 6.7±0.07 10.6±0.10 132.9±2.69 3.4±0.03 60.5±0.49 51.5±0.49
Other captive (n =14) 10.1±0.13 6.7±0.16 6.8±0.08 10.5±0.08 125.8±2.30 3.5±0.05 59.4±0.53 49.8±0.35
Wild (n =20) 10.0±0.07 7.1±0.11 6.8±0.04 10.4±0.11 129.4±1.39 3.7±0.04 61.0±0.36 52.3±0.38
CRC captive (n =22) 11.2±0.11 7.6±0.19 7.3±0.06 11.7±0.15 132.4±2.15 3.8±0.06 65.6±0.47 54.7±0.75
Other captive (n =53) 11.0±0.08 7.3±0.08 7.2±0.03 11.7±0.08 132.7±1.13 4.0±0.04 65.1±0.30 53.7±0.23
Wild (n =17) 10.9±0.12 7.5±0.12 7.3±0.05 11.9±0.14 134.6±1.9 4.2±0.07 69.5±0.44 59.3±0.34
For description of morphometric measurements, see Table 1. A two-way MANOVA test with sex and location (wild versus captive)
was significant (Wilks’ Lambda=0.4, F16,242 =10.4, P <0.001). Results of post-hoc univariate tests of the effect of location and the
interaction effect of sex and location are given when significant (α<0.05).
a Test of location: F2,128 =3.8, P =0.03.
b Test of location: F2,128 =13.0, P <0.001; Tukey’s HSD: CRC versus OTHER + WILD.
c Test of location: F2,128 =24.2, P <0.001; Tukey’s HSD: CRC+WILD versus OTHER; interaction of sex and location F2,128 =7.7,
d Test of location: F2,128 =41.3, P <0.001; Tukey’s HSD: CRC+OTHER versus WILD; interaction of sex and location F2,128 =8.3,
PC2 for both adult and juvenile data sets were similar
to those found by Wisely et al. (2002a) and thus we
used the same logic in interpreting our results. For both
adult and juvenile data, character loadings for all variables
were positive and≥0.5 (Fig. 2) for PC1, suggesting that
PC1 was correlated with overall body size. Character
loadings in the eigenvector of PC2 were both positive and
negative; we interpreted this component to be indicative
of body shape (Somers, 1986). Results of the MANOVA
using PC scores were significant for the test of ex situ
versus in situ juveniles (Wilks’ Lambda=0.8, F2,72 =7.3,
P=0.001); post-hoc univariate tests indicated that only
PC2 was significantly different between ex situ and
in situ juveniles (F1,73=14.9, P<0.001). For the adult
data, eigenvalues of PC1 and PC2 were 4.2 and 1.0
explaining 53% and 13% of the variation in the variables,
respectively. For adults, we found significant differences
in PC scores between location treatments (Wilks’
Lambda=0.73, F2, 127 =11.1, P<0.001). Post-hoc tests
revealed a difference in values of PC1 (F2,128 =12.5,
P<0.001) and in PC2 (F2,128 =7.6, P=0.001: Fig. 3)
Environment or genetics?
Wisely et al. (2002a) previously reported that pre-captive
populations of black-footed ferrets were larger and were
differently shaped than ex situ animals based on skull
morphology. They found no correlation between pedigreebased
inbreeding and skull size suggesting that inbreeding
depression was not the cause for small ex situ animals,
but they could not further distinguish the cause of these
morphological differences. The present study compared
individuals raised in captivity to those born in the wild to
Morphological change in black-footed ferrets 325
Forearm length (mm)
Rear leg length (mm)
CRC Other Wild
Fig. 1. Mean length (mm) of ulna and tibia of adult anaesthetised
black-footed ferrets born and raised in two captive environments,
large enclosures at the National Zoological Park’s Conservation
and Research Center (CRC) or small enclosures at other breeding
facilities (Other) and animals born and raised in the wild (Wild).
We found a significant interaction between sex (continuous line,
male; broken line, female) and location treatment groups. Males
were disproportionately smaller than females in captivity.
determine if these morphological changes were the result
of the developmental environment and, therefore, not
heritable or were the product of domestication resulting
from unintentional selection. Our results confirm that
environment induced the diminutive body size of ex situ
black-footed ferrets. Principal component analysis suggested
that overall body size was smaller and body shape
differed for ex situ versus in situ adults
(Fig. 3). While
our measurements of live, anaesthetised animals were
not directly comparable to those taken by Wisely et al.
(2002a) on skulls of pre-captive in situ animals, we did
find similarity in the magnitude of observed change in
body size. Wisely et al. (2002a) concluded that ex situ
animals were 5–10% smaller than pre-captive in situ
individuals; the present study found that ex situ individuals
ranged from 2–9% smaller than in situ descendants of
the captive population.  These findings suggest that wildborn
members of reintroduced populations returned to
pre-captive body size in the 3 years that elapsed between
the final reintroduction and our study.
It is unlikely that these rapid morphological changes in
the in situ population were a result of natural selection
for increased body size. High survival and fecundity rates
TA RL FA NP UCW IW UCC LCC P24
TA RL FA NP UCW IW UCC LCC
Fig. 2. Character loadings from a principal component analysis
of nine morphometric measures of (a) juvenile and (b) adult blackfooted
ferrets. Character loadings of principal component one (PC1)
are filled bars and loadings of principal component two (PC2) are
open bars. We interpreted PC1 to represent overall body size and
PC2 to represent body shape.
and concomitant exponential growth experienced by the
Conata Basin/Badlands population would not promote
a ‘large body size genotype’ differentially contributing
to the next generation. In addition, captive animals that
were reintroduced between 1996 and 1999 would have
overwhelmed any differential success of these genotypes
due to genetic swamping with ex situ individuals. Finally,
the rapid return to pre-captive body size (3 years or
1–2 generations) suggests that the observed changes
were not due to natural selection.  Our results confirm
that environmental factors, not heritable ones, influenced
the observed difference in these characters between
Adult black-footed ferrets had equivalent tail length
regardless of source location (Table 2; TA), yet ex situ
juvenile males had tails that were 6% shorter than agematched
juveniles in situ, suggesting delayed tail growth
in the former group (Table 1; TA). The mechanism for
this difference is poorly understood, but may be due to
slower vertebral development for young ferrets raised in
captivity. It appears, however, that tail vertebrae reach
their full growth potential ex situ; adult tail lengths were
equivalent among in situ and ex situ groups.
326 S. M. WISELY ET AL.
-2.5 -2 -1.5 -1 -0.5 -0
-2 -1 0 1 2 3
Principal Component 1
Principal Component 2
Fig. 3. Polygons encircling principal components one and two
out of eight morphometric variables measured from adult blackfooted
ferrets raised at National Zoological Park’s Conservation
& Research Center (long broken line), all other captive breeding
facilities (short broken line) and in situ (continuous line). Ovals
represent the centroid of values for each location.
 The most striking feature of our comparative assessment
was the significantly shorter ulna and tibia in black-footed
ferrets maintained ex situ, for both juveniles and adults.
Limb length was 9% smaller for ex situ adult males
than for in situ males
(Fig. 1). This difference was less
profound in animals raised in the larger enclosures of
CRC. We interpret these results to mean that long bone
development and growth is curtailed in ex situ animals and
that enclosure size plays a role in the difference
we found differences in juvenile morphometry, we
hypothesise that differential elongation begins prenatally
or early (<75 days) in postnatal development.
Many factors affect long-bone development including
growth hormones that are regulated by the pituitary gland
(Williams & Hughes, 1977) and induction of growth
by mechanical stimuli (Lovejoy, Cohen & White, 1999).
Differences in relative long bone length among domestic
breeds of dogs are believed to be controlled by polygenic
regulation of the concentration of growth hormones during
development (Wayne, 1986). For black-footed ferrets it
is possible that unknown environmental stimuli inhibit a
growth factor responsible for prenatal or early postnatal
long bone growth. A more parsimonious explanation
involves differential mechanical stimuli in juvenile blackfooted
ferrets raised in different captive environments
or born in the wild. Long bones develop from the
epiphyseal growth plates at the proximate and distal ends
(Hinchliffe & Johnson, 1980); cartilage cells at these
growth plates transduce mechanical stimuli into extracellular
matrix production (Lovejoy et al., 1999). These
stimuli are believed to simultaneously build bone, yet
prevent the epiphyses from ossifying prematurely (Marzke
et al., 1996). The result is that bone architecture is
modelled in response to the environment such that bone
is built in areas subjected to stress and reabsorbed in
areas without stress, a phenomenon known asWolff’s Law
(Wolff, 1891; Salter, 1970). Experimental studies confirm
these assertions; for example, immature rats whose legs
were immobilised with casts, had shorter and thinner
long bones (Steinberg & Trueta, 1981). For the blackfooted
ferret, there appears to be a direct relationship
between bone length and enclosure size. Ulna length
was the longest in wild-born animals, shorter in those
raised in large enclosures and shortest in animals kept
in small enclosures.
Insufficient mechanical stimulation
of the long bones of young ferrets raised in a small
homogeneous space could explain the shorter limb length.
In situ, juveniles were exposed to a more heterogeneous
environment that allowed for more uninhibited movement
and stimulated a greater degree of bone lengthening than
in the ex situ environment. We hypothesise that these
opportunities to experience multi-variant stimuli in an
expansive space may be a necessary prerequisite in early
postnatal development for proper limb development.
We found sexual dimorphic body size differences
to be more pronounced in the wild; developmental
differences between captive-reared and wild-born animals
were greater for males (Fig. 1). A similar pattern of
reduced dimorphism was found in ranch-raised American
mink (M. vison) when compared to feral mink in Norway,
Ireland and Britain (Lynch & Hayden, 1995). These
authors attributed their findings to genetic mechanisms
associated with selection. For ranch mink, an animal under
extreme artificial selection, this is the most parsimonious
explanation. For the reasons outlined above, we believe
that size differences between in situ and ex situ blackfooted
ferrets were due to environmental factors, which
probably affected the sexes differentially.
Implications for in situ and ex situ conservation
We have documented morphological distinctions between
ex situ and in situ conspecifics caused by influences of
the ex situ environment. Over the next several decades as
more organisations use conservation breeding as a tool to
save an increasing number of threatened taxa, it will be
imperative that biologists understand the consequences of
this conservation action. One of the biggest concerns for
ex situ populations is that they remain representatives of
their taxa (Snyder et al., 1996). As revealed by the present
work, this concern is not a trivial mandate as animals
Morphological change in black-footed ferrets 327
held ex situ are vulnerable to an array of perturbations.
To date, most studies that have documented changes to
ex situ populations have reported heritable changes that
are the result of adaptation to the captive environment
(e.g. Lewis&Thomas, 2001; Kruska&Sidorovich, 2003).
Clearly, heritable changes to the phenotype that result in
unintentional selection for morphological domestication
should be a primary concern to conservation curators, yet
non-heritable changes, that are the result of environmental
factors, should not be overlooked. Phenotypically altered
individuals may have lower survival or fecundity upon
reintroduction than more representative individuals,which
lowers the growth potential of reintroduced populations.
Survival and reproduction of black-footed ferrets
with an altered phenotype demonstrates how easily
development is altered to produce variant phenotypes in
the captive environment. For this species, shorter limbs in
captivity may or may not confer a fitness advantage, but
genes that control the growth and ossification of cartilage
cells are no longer under stabilising selection to maintain
growth for a set time, which increases the opportunity
for unintentional selection. To avoid possible genotypic
change, we reiterate the importance of recommendations
by Ballou & Foose (1996) that conservation breeding
be maintained for as few generations and for as long a
generation time as possible. For some species, such as
the black-footed ferret whose habitat has nearly vanished,
terminating the breeding program is not yet possible
without incurring a serious risk of extinction. When
conservation breeding is necessary for many generations,
the effects of unintentional selection due to high density
or small enclosures must be balanced with the positive
effects of high juvenile production for reintroduction, as
is the case for the black-footed ferret recovery program.
 Morphological changes have occurred to black-footed
ferrets ex situ; animals born in captivity are smaller and
shaped differently than animals in situ. Most notably
the fore- and hind-limbs of ex situ animals are shorter
than animals in situ. These changes are environmentally
induced and not heritable; nevertheless, selection for
‘wild-type’ phenotypes has been relaxed, setting the stage
for the heritable morphological alteration of unintentional
selection.Wolff’s Lawsuggests that changes to limb length
could be due to the lack of mechanical stimuli in the
ex situ environment
. We recommend that other conservation-
breeding programs initiate comparative studies
of long bone growth rates in situ and ex situ to determine
if this phenomenon occurs in other endangered, mediumsized
carnivores. Should morphological changes be found,
recovery planners will need to balance the trade-off of
increased enclosure size with the decreased numbers
of animals available for reintroduction (assuming space
limitations in captivity). For the black-footed ferret, a new
breeding facility to be completed in 2005 will have large
enclosures similar to CRC that will help to ameliorate this
environmentally caused change in morphology.
The authors thank Mike Lockhart, Bill Perry, Doug
Sargent, Brian Kenner, Doug Albertson and Greg
Schroeder for assistance and logistical support in the
field.We thank LinwoodWilliamson, Dr Mitch Bush, Lisa
Ware and the animal care staff of CRC and the National
Black-Footed Ferret Conservation Center for assistance
with ex situ animals. Funding was provided by a grant
from the National Fish and Wildlife Foundation to SMW
and JGH and by the Smithsonian’s National Zoological
Park, Friends of the National Zoo, US Forest Service,
National Park Service, US Fish and Wildlife Service and
Prairie Wildlife Research. All methods were approved by
Kansas State University Institutional Animal Care and
Use Committee (Protocol 2310).
Anderson, E., Forrest, S. C., Clark, T. W. & Richardson, L. (1986).
Paleobiology, biogeography, and systematics of the black-footed
ferret, Mustela nigripes (Audubon and Bachman), 1851. Great Basin
Nat. Mem. 8: 11–62.
Ballou, J. D. & Foose, T. J. (1996). Demographic and genetic
management of captive populations. In Wild mammals in captivity:
263–283. Kleiman, D. G., Allen, M. E., Thompson, K. V. &
Lumpkin, S. (Eds). Chicago: University of Chicago Press.
Clark, T. W. (1994). Restoration of the endangered black-footed ferret:
a 20-year overview. In Restoration of endangered species: conceptual
issues, planning and implementation: 272–297. Cambridge:
Cambridge University Press.
Conservation Breeding Specialist Group. (2003). Black-footed ferret
population management workshop. Apple Valley, Minnesota: World
Conway, W. G. (1986). The practical difficulties and financial
implications of endangered species breeding programmes. Int. Zoo
Yearbk. 24/25: 210–219.
Dobson, A. & Lyles, A. (2000). Black-footed ferret recovery. Science
Ebenhard, T. (1995). Conservation breeding as a tool for saving animal
species from extinction. Trends Ecol. Evol. 10: 438–443.
Evermann, J. F., Burns, G., Roekle, M. E., McKeirnan, A. J.,
Greenlee, A., Ward, A. C. & M. L. Pfeifer. (1983). Diagnostic
features of an epizootic of feline infectious peritonitis in captive
cheetahs. Am. Ass. Vet. Lab. Diagn. 26: 365–382.
Fleischer, R. C.&Murphy, M. T. (1992). Relationships among allozyme
heterozygosity,morphology and lipid levels in house sparrows during
winter. J. Zool. (Lond.) 226: 409–419.
Flesness, N. R. & Cronquist-Jones, K. G. (1987). Possible selection
in captive Panthera tigris altaica. In Tigers of the world: 363–370.
Tilson, R. L. & Seal, U. S. (Eds). Park Ridge, NJ: Noyes.
Frankham, R., Hemmer, H., Ryder, O. A., Cothran, E. G., Soul´e,
M. E., Murray, N. D. & Snyder, M. (1986). Selection in captive
environments. Zoo Biol 5: 127–138.
Hinchliffe, J. R. & Johnson, D. R. (1980). The development of the
vertebrate limb: an approach through experiment, genetics, and
evolution. Oxford: Clarendon Press.
Jacobson, E. R. (1993). Implications of infectious diseases for captive
propagation and introduction programs of threatened and endangered
species. J. Zoo Wildl. Med. 24: 245–255.
Komers, P. E. & Komers, E. J. (1992). Juvenile male magpies dominate
adults irrespective of size differences. Can. J. Zool. 70: 815–
Kreeger, T. J., Vargis, A., Plumb, G. E. & Thorne, T. E. (1998).
Ketamine-medetomidine or isolflurane immobilization of blackfooted
ferrets. J. Wildl. Mgmt. 62: 654–662.
328 S. M. WISELY ET AL.
Kruska, D. (1996). The effect of domestication on brain size and composition
in the mink (Mustela vison). J. Zool., Lond. 239: 645–661.
Kruska, D. C. T. & Sidorovich, V. E. (2003). Comparative allometric
skull morphometrics in mink (Mustela vison Schreber, 1777) of
Canadian and Belarus origin: taxonomic status. Mamm. Biol. 68:
Lacy, R. C. (1994). Managing genetic diversity in captive populations
of animals. In Restoration of endangered species: conceptual issues,
planning and implementation: 63–89. Bowles, M. E. & Whelan,
C. J. (Eds). Cambridge: Cambridge University Press.
Lande, R. & Barrowclough, G. (1987). Effective population size,
genetic variation, and their use in population management. In
Viable populations for conservation: 87–124. Soul´e, M. E. (Ed.).
Cambridge: Cambridge University Press.
Lewis, O. T. & Thomas, C. D. (2001). Adaptations to captivity in
the butterfly Pieris brassicae (L.) and the implications for ex situ
conservation. J. Insect Conserv. 5: 55–63.
Lovejoy, C. O., Cohn, M. J. & White, T. D. (1999). Morphological
analysis of the mammalian postcranium: a developmental
perspective. Proc. Natl. Acad. Sci., U.S.A. 96: 13247–13252.
Lynch, J. M. & Hayden, T. J. (1995). Genetic influences on cranial
form: variation among ranch and feral American mink Mustela vison
(Mammalia: Mustelidae). Biol. J. Linn. Soc. 55: 293–307.
Marzke, M. W., Young, D. L., Hawkey, D. E., Su, S. M., Fritz, J. &
Alford, P. L. (1996). Comparative analysis of weight gain, hand/wrist
maturation, and dental emergence rates in chimpanzees aged
0–24 months from varying captive environments. Am. J. Phys.
Anthropol. 99: 175–190.
Meretsky, V. J., Snyder, N. F. R., Beissinger, S. R., Clendenen,
D. A. & Wiley, J. W. (2000). Demography of the California condor:
implications for reestablishment. Conserv. Biol. 14: 957–967.
Ralls, K. & Ballou, J. D. (1983). Extinction: lessons from zoos.
In Genetics and conservation: 164–184. Shonewald-Cox, C. M.,
Chambers, S. M., MacBride, B. & Thomas, I. (Eds). Menlo Park,
Rice, W. R. (1989). Analyzing tables of statistical tests. Evolution 43:
Salter, R. B. (1970). Textbook of disorders and injuries of the
musculoskeletal system. Baltimore: Williams and Wilkins.
Seal,U. S., Thorne, E. T., Bogan, M. A.&Anderson, S. H. (Eds). (1989).
Conservation biology and the black-footed ferret. New Haven, CT:
Yale University Press.
Sheets, R. G. (1972). A trap for capturing black-footed ferrets. Am.Mid.
Nat. 88: 461–462.
Snyder,N. F. R.,Derrickson, S. R., Beissinger, S. R.,Wiley, J.W., Smith,
T. B., Toone, W. D. & Miller, B. (1996). Limitations of captive
breeding in endangered species recovery. Conserv. Biol. 10: 338–
Somers, K. M. (1986). Multivariate allometry and removal of size with
principal components analysis. Syst. Zool. 35: 359–368.
Soul´e, M., Gilpin,M., Conway,W. & Foose, T. (1986). The millennium
ark: how long a voyage, how many staterooms, how many
passengers? Zoo Biol. 5: 101–113.
Steinberg, M. E. & Trueta, J. (1981). Effects of activity on bone growth
and development in the rat. Clin. Orthop. 156: 52–60.
Vargas, A.&Anderson, S. H. (1996). Growth and physical development
of captive-raised black-footed ferrets (Mustela nigripes). Am. Midl.
Nat. 135: 43–52.
Wayne, R. K. (1986). Limb morphology of domestic and wild canids:
the influence of development on morphologic change. J. Morphol.
Wildt, D. E., Ellis, S., Janssen, D. & Buff, J. (2003). Toward a more
effective reproductive science for conservation. In Reproductive
science and integrated conservation: 2–20. Holt, W. V., Pickard,
A. R., Rodger, J. C. & Wildt, D. E. (Eds). Cambridge: Cambridge
Williams, J. P. G. & Hughes, P. C. R. (1977). Hormonal regulation and
post-natal limb growth in mammals. In Vertebrate limb and somite
morphogenesis: 281–292. Ede, D. A., Hinchliffe, J. R. & Balls, M.
(Eds). Cambridge: Cambridge University Press.
Wisely, S. M., Ososky, J. J. & Buskirk, S. W. (2002a). Morphological
changes to black-footed ferrets (Mustela nigripes) resulting from
captivity. Can. J. Zool. 80: 1562–1568.
Wisely, S. M., McDonald, D. B. & Buskirk, S. W. (2003). Evaluation
of the species survival plan and captive breeding program for the
black-footed ferret. Zoo Biol. 22: 287–298.
Wisely, S. M., Fleming, M. A., McDonald, D. B., Buskirk, S. W. &
Ostrander, E. A. (2002b). Genetic diversity and fitness in blackfooted
ferrets before and during a bottleneck. J. Hered. 93: 231–
Wolff, J. (1891). Das gesetz der transformation der knochen. Berlin: A
Zeuner, F. E. (1963). A history of domesticated animals. London: