“False heat,” big testes, and the onset of natal dispersal in European pine Martens (Martes martes)

Abstract

Natal dispersal plays a central role in population biology, ecology, and evolution. Therefore, describing dispersal patterns including dispersal rate, onset, duration, and distance and investigating its main ecological and phenotype correlates are of prime importance. Dispersal data are scarce in rare and elusive carnivores such as mustelids and notably in European pine Marten Martes martes. Natal dispersal is mostly thought to occur in late winter either as a consequence of offspring eviction by adult females experiencing a “false heat” period related to the delayed implantation of embryos or as a consequence of physiological changes in offspring related to sexual maturation. Nineteen juvenile and subadult pine martens were monitored by telemetry during their first year of life to describe the dispersal pattern, in particular, to investigate the onset of natal dispersal. Only 13 out of the 19 pine martens monitored gave reliable information, and seven out of them (53.85 %) dispersed. No sex bias either in dispersal rate, distance, or duration was evidenced. As expected, all dispersers whatever their sex left their natal area in a relatively narrow period of 1 month in late winter (February 17th-March 17th). This period did not match with any body mass variation but was clearly synchronous with an increase in testis mass. Our results confirmed that this late winter period is very intense in terms of social interactions and physiological changes in the European pine marten. Last, they also pointed out that human-induced mortality of dispersers in late winter could be high so that late winter might be a focus period for management actions.

Full text

SHORT COMMUNICATION
“False heat,”big testes, and the onset of natal dispersal
in European pine Martens (Martes martes)
Jeremy Larroque &Sandrine Ruette &
Jean-Michel Vandel &Sébastien Devillard
Received: 13 May 2014 /Revised: 9 December 2014 /Accepted: 14 December 2014 /Published online: 25 December 2014
#Springer-Verlag Berlin Heidelberg 2014
Abstract Natal dispersal plays a central role in population
biology, ecology, and evolution. Therefore, describing dis-
persal patterns including dispersal rate, onset, duration, and
distance and investigating its main ecological and phenotype
correlates are of prime importance. Dispersal data are scarce
in rare and elusive carnivores such as mustelids and notably
in European pine Marten Martes martes. Natal dispersal is
mostly thought to occur in late winter either as a conse-
quence of offspring eviction by adult females experiencing a
“false heat”period related to the delayed implantation of
embryos or as a consequence of physiological changes in
offspring related to sexual maturation. Nineteen juvenile and
subadult pine martens were monitored by telemetry during
their first year of life to describe the dispersal pattern, in
particular, to investigate the onset of natal dispersal. Only 13
out of the 19 pine martens monitored gave reliable informa-
tion, and seven out of them (53.85 %) dispersed. No sex
bias either in dispersal rate, distance, or duration was evi-
denced. As expected, all dispersers whatever their sex left
their natal area in a relatively narrow period of 1 month in
late winter (February 17th–March 17th). This period did not
match with any body mass variation but was clearly
synchronous with an increase in testis mass. Our results
confirmed that this late winter period is very intense in terms
of social interactions and physiological changes in the
European pine marten. Last, they also pointed out that
human-induced mortality of dispersers in late winter could
be high so that late winter might be a focus period for
management actions.
Keywords Martes martes .European pine marten .Natal
dispersal .Dispersal onset .Telemetry
Introduction
Natal dispersal, defined as the movement made by an
individual from its birthplace to the place of first repro-
duction (Howard 1960), is a widespread behavioral pro-
cess. Its crucial role in population biology, ecology, and
evolution is now widely recognized albeit natal dispersal
remains poorly understood. Considerable knowledge has
been gained since evolutionary ecologists have started to
sequence natal dispersal behavior in three successive behavioral
stages (Travis et al. 2012): emigration (also called onset, timing
of dispersal), transience (also called transfer), and immigration
(also called settlement). There is now considerable evidence
that the decision to leave or to stay, to make a short or large
movement, and to settle or not in a given place are not influ-
enced by the same ultimate and proximate factors. Both inter-
nal, sensu phenotype-dependent, and environmental, sensu
condition-dependent, factors trigger dispersal and interact in
complex ways (Clobert et al. 2001). It is thus becoming of
prime importance to investigate each stage apart from the others
to better understand the dispersal processes in wild populations
as much as its consequences on individual fitness (Bonte et al.
2012) and on population dynamics and genetics. The basic
dispersal pattern of mammalian solitary carnivores, i.e., sex
Communicated by C. Gortázar
Electronic supplementary material The online version of this article
(doi:10.1007/s10344-014-0889-x) contains supplementary material,
which is available to authorized users.
J. Larroque :S. Ruette :J.<M. Vandel
Office National de la Chasse et de la Faune Sauvage, CNERA-PAD,
Montfort, 01330 Birieux, France
J. Larroque :S. Devillard (*)
Université de Lyon; F-69000, Lyon; CNRS, UMR5558, Laboratoire
de Biométrie et Biologie Evolutive, Université Lyon 1,
F-69622 Villeurbanne, France
e-mail: sebastien.devillard@univ-lyon1.fr
Eur J Wildl Res (2015) 61:333–337
DOI 10.1007/s10344-014-0889-x

and age variations in dispersal rate and distance, has received as
much attention as in other taxonomic groups, but the detailed
process of natal dispersal within the three behavioral stages
framework mentioned above is poorly known, especially in
mustelids (Arthur et al. 1993). The European pine marten
(Martes martes) is one of the species for which almost nothing
is known about natal dispersal. Although it is usually thought
that most of the young individuals disperse either in early fall or
in early spring (Helldin and Lindström 1995 and references
therein), few studies, to our knowledge, have reported natal
dispersal movements (e.g., Helldin and Lindström 1995;
Zalewski 2012) and have given any basic characteristics of
the dispersal pattern. Helldin and Lindström (1995) suggested
likely associations between condition-dependent or phenotype-
dependent factors and the onset of natal dispersal in pine
martens. Indeed they reviewed evidences that a “false heat,”
i.e., an increase in social activity and agonistic intrasexual
interactions, occurred in February and March. One hypothesis
to explain this phenomenon might be related to the yearly
reproductive cycle of adult females. Indeed, this February–
March period corresponds to the restart of gestation, following
a delayed implantation of blastocysts. As a result, adult females
could evict their young, forcing them to disperse during this
period. They also discussed the possibility that this false heat
resulted mostly from the beginning of the dispersal period
simultaneous to a physiological change of young individuals,
namely an increase of sex hormone concentrations. In both
hypotheses, young martens should predominantly disperse in
February–March. Studies in mammals have shown that the
onset of natal dispersal is often correlated with both morpho-
logical and physiological changes (e.g., threshold body mass
hypothesis, in roe deer Capreolus capreolus, Wahlström 1994;
“ontogenic switch hypothesis,”in Belding’s ground squirrels
Spermophilus beldingi, Nunes and Holekamp 1996;butsee
Elliot et al. 2014), making individuals either likely competitors
foradultsorabletobeardispersal costs. Here, we investigated
the onset of natal dispersal in first-year pine martens using
telemetry monitoring and specifically assessed whether the
natal dispersal onset occurred in late winter (mid-February to
March) as suggested by Helldin and Lindström (1995). Using
data from dead recoveries of first-year pine martens of both
sexes, we also evaluated the association between the onset of
dispersal and the phenologies of body mass and testes mass
over the first year of life.
Material and methods
From a larger telemetry dataset (n=44 individuals of all ages),
we selected only those that were captured as juveniles or
subadults between July and February 15th of the year after
their birth year (i.e., between 4 and 11.5 months old assuming
a common birth date of April 1st). Hence, two juvenile ([4–6]
months old, both males) and 17 subadult ([6–11.5] months
old, eight males, nine females) pine martens were retained to
investigate the timing of dispersal. Individuals were trapped in
box cages in the region of Bresse (study area=911 km
2
,
eastern France 5° 13′E, 46° 27′N) over four winters
(electronic supplementary material Table S1), where hunt-
ing and trapping is legally allowed. Animals were anesthe-
tized by intramuscular injection of Domitor® (10 mg kg
−1
,
medetomidine, Pfizer Animal Health, New York, USA) and
revived with Antisédan® (atipamezol, Pfizer Animal
Health, New York, USA) to ensure a quick reversal of
sedation. We marked each individual with a transponder
(Allflex®, Vitré, France) and radio-collared them before
release at their site of capture. Collars were TXH-2 from
Televilt® (Stockholm, Sweden) or TW-5 with a biothane
collar from Biotrack Ltd® (Wareham, Dorset, UK) and
weighed about 32 g. Age was estimated from a set of
morphological variables (body mass, tooth wear, body
length, tail length, neck circumference, posterior foot
length, and baculum length for males, Ruette et al. unpub-
lished data) and confirmed by the number of annual growth
lines visible in the tooth cementum using a standardized
cementum aging model for each species (Matson’s Laboratory,
Milltown, MT, USA, Matson and Matson 1993) when martens
were later recovered dead (11 out of the 19, electronic supple-
mentary material Table S1).
The pine martens were located at least twice a week when
they were resting during the day, with a precision higher than
100 m. For this sample (n=19), the mean duration of moni-
toring was 122±99 days (8–379 days, electronic supplemen-
tary material Table S1)andthemeannumberoflocations
was 39.6±30.56 (6–120 locations). Based on their succes-
sive locations, we assessed whether each individual dis-
persed or not, assuming that the capture location was close
to the birth site. Each individual was classified into
philopatric (“stationary”or “explorer”)ordisperser(“shifter”
or “one-way”) following the initial classification of McShea
and Madison (1992), adapted to pine martens (electronic
supplementary material).
We applied the k-means clustering method (Forgy 1965)on
individual locations of the dispersers only (both shifter and
one-way individuals) to identify two successive spatial clus-
ters over the monitoring, namely, the areas used for resting
before and after the natal dispersal (electronic supplementary
material Fig. S1). We then defined the departure date, i.e., the
timing of dispersal, as the date of the last location in the first
locations’cluster and the settlement date as the date of the first
location in the second cluster. We finally estimated the
natal dispersal distance by computing the Euclidean dis-
tance between the centroids of the two spatial clusters.
Dispersal duration, i.e., the length of the transience phase,
was computed as the difference (in days) between the
settlement date and the departure date.
334 Eur J Wildl Res (2015) 61:333–337

During the study period, 59 male and 32 female pine
martens aged less than 18 months were collected dead on
the study area from trappers and hunters. Age in days was
then estimated by the difference between the recovery date
and April 1st. The dead animals were weighed and aged as
above. For males, the testes were extracted from the corpse
and weighed (n=36 out of the 59). This allowed us to assess
the age variation of both testes mass and body mass in both
sexes during the first year of life.
Results
Captures mostly occurred from early December to late
January (16 out of 19, electronic supplementary material
Tab le S1) which precluded a reliable assessment of dispersal
in early fall. The two juveniles monitored during this early fall
period did not disperse, one was still philopatric at the end of
its monitoring, and the second dispersed during the mid-
February to March period (see below, electronic supplemen-
tary material Table S1). As regards dispersal during the second
focused period of mid-February to March, six individuals
(three males, three females) could not be classified: either
because their monitoring duration was less than 30 days (four
out of the six) or because their monitoring ceased long before
this period (one out of the six) or because it ceased during an
exploration movement (Fig. 1a, one out of the six, electronic
supplementary material Table S1). Among the 13 remaining
individuals, five were classified as philopatric explorer (two
males, three females), i.e., they made exploration movements
but always returned soon after to their birthplace (Fig. 1b), and
one was classified as a philopatric stationary individual.
However, the monitoring of four out of these six individuals
ceased between March 1st and March 17th (electronic
supplementary material Table S1), right in the middle of the
focused period so that we could not rule out that they would
have dispersed later (Fig. 1c).
The seven remaining individuals were classified as dis-
persers (Fig. 2a), either shifter individuals (one male, one
female) or one-way individuals (three males, two females).
The overall natal dispersal rate was 7/13=53.85 %. For these
seven dispersers, the natal dispersal distance ranged from
2256 to 17,257 m and was not significantly different in both
sexes (4968±2057 m in females, 7028±7058 m in males).
Fig. 1 Temporal evolution of the
scaled distance between each
successive location and the
centroid of the natal area from
October 1st of the the year ofbirth
to the end of June the following
year, aunclassified pine martens;
bstationary pine martens;
cexplorer pine martens
Eur J Wildl Res (2015) 61:333–337 335

The natal dispersal duration ranged from 5 to 76 days and did
not differ between sexes (35± 36 days in females, 29±17 days
in males). All the dispersers started their natal dispersal move-
ment 15 days around March 1st the year after their birth
(Fig. 2a, electronic supplementary material Table S1). This
relatively narrow period for the timing of dispersal was clearly
associated with the onset of relative testes mass increase in
subadult males (Fig. 2a) and did not match with any age varia-
tionofbodymassinbothsubadultmalesandfemales(Fig.2b).
Discussion
Emigration movements, sensu departure from the natal area,
were highly synchronous and occurred within a narrow period
of less than one month between February 17th and March
17th. Although we cannot rule out that we could have missed
dispersal movements in early fall (only two individuals were
monitored during this period, most individuals were captured
after December 1st), this finding is in agreement with the idea
that natal dispersal mostly occurs in late winter (Helldin and
Lindström 1995). Both males and females dispersed during
this period without any variation in body mass as subadult
pine martens reach adult size when they are 6 to 8 months
old (Ruette et al. unpublished data). Such synchronous
dispersal is in agreement with the first hypothesis of
Helldin and Lindström (1995) which argued that hormonal
changes accompanying the restart of gestation induce the
adult female’s behavior of driving out its young (sensu
condition-dependent dispersal), independently of off-
springs’phenotypes. Alternatively, the onset of dispersal
of males matched very well with the increase of testis
mass described on our transversal sample in late winter.
Concomitantly with this increase, the level of testosterone
is expected to reach its peak (Audy 1976). This result is
in agreement with phenotype-dependent dispersal and
would support the second hypothesis of Helldin and
Lindström (1995), i.e., the false heat results from the
whole dispersal process. We acknowledge that only longi-
tudinal data on individuals would help to clarify this
process. Unfortunately, we did not have any hormonal
and morpho-anatomical metrics of such a physiological
change in subadult females during this period. To our
knowledge, we provided here the first evaluation of the
Fig. 2 a Coincidence between
the temporal evolution of the
scaled distance between each
successive location and the
centroid of the natal area for
dispersing pine martens and the
phenology of the averaged testis
mass (n=36) from October 1st of
the year of birth to the end of June
the following year. Testis mass
was standardized for body mass
using the residuals of the linear
model between the log-testis mass
and the log-body mass. Testis
mass was scaled to facilitate
graphical display. bTemporal
variation of male (n=58) and
female (n=32) body mass from
October 1st of the year of birth to
the end of June the following year
336 Eur J Wildl Res (2015) 61:333–337

two hypotheses of Helldin and Lindström (1995) about the
false heat in Martes species in a wild population of pine
martens since the paper of Helldin and Lindström (1995)
has been published. Based on our study, it is not possible
to rule out any of them, but data reported here provide the
opportunity for the mustelids community to increase our
common knowledge about this particular false heat pattern.
Finally, highly synchronous natal dispersal in pine martens
might have management implications. Natal dispersal plays
a major role in ensuring connectivity between populations
and/or recolonization of depleted sites (Clobert et al.
2001). In our area, human-induced mortality is mostly
attributed to trapping/hunting and roadkills (Ruette et al.
2015). As trapping is mostly active between January and
March in France, most of the animals caught might be
crossing pine martens, especially since subadult martens
are likely easier to trap than adults. Hence, late winter
would be not only a critical period for pine marten populations
but also the focus period to control human-induced mortality
when necessary.
Acknowledgments We are very grateful to all the hunters and trappers
of Ain and Saône-et-Loire who helped us in field work, especially Willy
Genton, Léon Boully, and Daniel Vivant. We also thank Catherine Carter
who kindly edited the English.
References
Arthur SM, Paragi TF, Krohn WB (1993) Dispersal of juvenile fishers in
Maine. J Wildl Manag 57:868–874
Audy MC (1976) Le cycle sexuel saisonnier du male des mustélidés
Européens. Gen Comp Endocrinol 30:117–127
Bonte D, Van Dyck H, Bullock JM, Coulon A, Delgado M, Gibbs M,
Lehouck V, Matthysen E, Mustin K, Saastamoinen M, Schtickzelle
N, Stevens VM, Vandewoestijne S, Baguette M, Barton K, Benton
TG, Chaput-Bardy A, Clobert J, Dytham C, Hovestadt T, Meier
CM, Palmer S, Turlure C, Travis J (2012) Costs of dispersal. Biol
Rev 87:290–312
Clobert J, Danchin E, Dhondt AA, Nichols JD (2001) Dispersal. Oxford
University Press, New York
Elliot NB, Valeix M, Macdonald DW, Loveridge AJ (2014) Social
relationships affect dispersal timing revealing a delayed infanticide
in African lions. Oikos 9:1049–1056
Forgy EW (1965) Cluster analysis of multivariate data: efficiency vs
interpretability of classifications. Biometrics 21:768–769
Helldin JO, Lindström ER (1995) Late winter social activity in pine
marten (Martes martes)—false heat or dispersal. Ann Zool Fenn
32:145–149
Howard W (1960) Innate and environmental dispersal of individual
vertebrates. Am Midl Nat 63:152–161
Matson GM, Matson J (1993) Progress in cementum aging of martens
and fishers. Special Report no. 931. Matson’sLaboratory,p3
McShea WJ, Madison DM (1992) Alternative approaches to study of
small mammal dispersal: insights from radiotelemetry. In: Stenseth
NC, Lidicker WZJ (eds) Animal dispersal: small mammals as a
model. Chapman and Hall, p 365
Nunes S, Holekamp K (1996) Mass and fat influence the timing of natal
dispersal in Belding’s ground squirrels. J Mammal 77:807–817
Ruette S, Vandel J-M, Albaret M, Devillard S (2015) Comparative
survival pattern of the syntopic pine and stone martens in a rural
trapped area in France. J Zool
Travis JMJ, Mustin K, BartońKA, Benton TG, Clobert J, Delgado MM,
Dytham C, Hovestadt T, Palmer SCF, Van Dyck H, Bonte D (2012)
Modelling dispersal: an eco-evolutionary framework incorporating
emigration, movement, settlement behaviour and the multiple costs
involved. Methods Ecol Evol 3:628–641
Wahlström LK (1994) The significance of male male-aggression for
yearling dispersal in roe deer (Capreolus capreolus). Behav Ecol
Sociobiol 35:409–412
Zalewski A (2012) Alternative strategies in the acquisition of home
ranges by male pine martens in a high-density population. Acta
Theriol 57:371–375
Eur J Wildl Res (2015) 61:333–337 337