G Model

ECOIND-2745; No. of Pages 5 ARTICLE IN PRESS

Ecological Indicators xxx (2015) xxx–xxx

Contents lists available at ScienceDirect

Ecological Indicators
journal homepage: www.elsevier.com/locate/ecolind

A new method for ageing wild boar using dental measures

Pilar Gonçalves a,∗ , David Risco a,b , Pedro Fernández-Llario c , Oscar Barquero-Pérez d ,
Emmanuel Serrano b,e , Javier Hermoso-de-Mendoza a , Concha Mateos a
a
Red de Grupos de Investigación Recursos Faunísticos, Facultad de Veterinaria, Universidad de Extremadura, 10003 Cáceres, Spain
b
CESAM, Departamento de Biologia, Universidade de Aveiro, 3810-193 Aveiro, Portugal
c
Innovación en Gestión y Conservación de Ungulados S.L. (INGULADOS), 10002 Cáceres, Spain
d
Departament of Signal Theory and Communications, Universidad Rey Juan Carlos, 28943 Fuenlabrada (Madrid), Spain
e
Servei d’ Ecopatología de Fauna Salvatge (SEFaS), Wildlife Diseases Research Group, Departament de Medicina i Cirugia Animals, Facultad de Veterinaria,
Universitat Autònoma (UAB), E-08193 Bellaterra, Barcelona, Spain

a r t i c l e i n f o a b s t r a c t

Article history: Over the last decades, wild boar (Sus scrofa) populations in Europe have increased with impacts on agricul-
Received 10 March 2015 ture, livestock and biodiversity. Thus, changes in population management for wild boar are increasingly
Received in revised form 4 November 2015 important. Knowledge of the age structure of the population is crucial to designing effective management
Accepted 9 November 2015
plans. However, the costs and efforts required to estimate the age of wild boar, primarily in the oldest
Available online xxx
animals, is problematic for managers and researchers. Here, we describe a new method to estimate wild
boar age based on simple dental measures (the external aperture of the pulp cavity, root length and
Keywords:
crown length from primary and secondary incisors). Our study was based on data from 93 wild boar
Wild boar
Age estimation
of known age belonging to two different populations in central and south-eastern Spain. We propose
Dental measures a model based on Boosted Regression Trees (BRT). Our results show a final age estimation model that
Management included all the explanatory variables proposed (dental measures) and showed a high percentage of esti-
mated deviance (61%), obtained by cross-validation. Thus, at least in Iberian wild boar populations, our
method constitutes a low-cost and reliable method for wild boar age estimation.
© 2015 Elsevier Ltd. All rights reserved.

1. Introduction populations (Bieber and Ruf, 2005; Keuling et al., 2008a; Risco et al.,
2013). Thus, developing simple and accurate age estimation meth-
Approaches and policies for wild boar (Sus scrofa) population ods remains a central challenge.
management are changing throughout Europe. This is largely due Age estimation methods in suids are usually based on the
to the rise in populations across the range of distribution as well as study of tooth development. Traditionally, age has been esti-
increasing density (Apollonio et al., 2010; Massei et al., 2015). This mated using patterns of tooth eruption and replacement but this
increase has led to problems in some regions, since wild boar can method is only accurate in animals younger than 26–30 months
cause important crop damage and can facilitate the spread of some of age (Anezaki, 2009; Choquenot and Saunders, 1993; Matschke,
pathological conditions (García-Jiménez et al., 2013; Herrero et al., 1967; Oroian et al., 2010). Other methods have been developed
2006; Onida et al., 2014; Schley et al., 2008). to improve age estimations. These techniques, such as root apex
Many studies have aimed to develop methods and approaches to closure (Habermehl, 1985), pulp cavity ratio (Sáez-Royuela et al.,
designing and integrating management plans in wild boar popula- 1989), counting the incremental lines of tooth cementum (Clarke
tions (Fernández-Llario and Mateos-Quesada, 2003; Herrero et al., et al., 1992; Choquenot and Saunders, 1993; Quere and Pascal,
2008; Keuling et al., 2008b). Knowledge of the age structure can be 1984; Sáez-Royuela et al., 1989) or tooth wear (Choquenot and
crucial to the study of many different aspects of these populations Saunders, 1993; Magnell, 2006; Oroian et al., 2010), seem to work
(seasonal and annual space use, incidence of different pathogens) properly in older wild boar. However, most of these methods are
and to developing hunting strategies as a measure to control expensive or require laboratory procedures, which are often not
available to managers or researchers.
In the present study, we recorded different measures from pri-
∗ Corresponding author at: Departamento de Biología y Zoología, Facultad de Vet-
mary and secondary incisors of wild boar of known age with the
aim to develop a simple and economical age estimation method for
erinaria, Universidad de Extremadura, 10003 Cáceres, Spain. Tel.: +34 927257150.
E-mail address: [email protected] (P. Gonçalves). this species.

http://dx.doi.org/10.1016/j.ecolind.2015.11.013
1470-160X/© 2015 Elsevier Ltd. All rights reserved.

Please cite this article in press as: Gonçalves, P., et al., A new method for ageing wild boar using dental measures. Ecol. Indicat. (2015),
http://dx.doi.org/10.1016/j.ecolind.2015.11.013
G Model
ECOIND-2745; No. of Pages 5 ARTICLE IN PRESS
2 P. Gonçalves et al. / Ecological Indicators xxx (2015) xxx–xxx

2. Materials and methods

2.1. Study site

The study was carried out on two different estates. One estate
is 700 ha and located in south-eastern Spain. The climate is
Mediterranean, with annual precipitation around 810 mm/year and
average temperatures of 6 ◦ C in the winter months and 23 ◦ C in the
summer months. The habitat is Mediterranean forest, where the
dominant tree species is the Holm oak (Quercus Ilex), interspersed
with well-developed thickets with species such as Cistus ladanifer,
Erica spp and Genista anglica. Wild boar density in this area was
about 93 wild boars/km2 .
The other estate is located in central Spain spanning an area
of 2000 ha. The climate is Mediterranean with continental influ-
ences, with annual precipitation around 550 mm/year and average
temperatures of 3 ◦ C in the winter months and 20 ◦ C in the sum-
mer months. The dominant tree species are Pinus sylvestris, Quercus Fig. 1. Incisor measures; 1 minimum root length, 2 maximum crown length and 3
pyrenaica and Quercus faginea, interspersed with species such as pulp cavity.
Cistus spp and Erica spp. Wild boar density in this area was about
40 wild boar/km2 . We included two areas in this study to control
for the possibility of different growth rates in different population, correlations among the variables themselves, resulting in problems
which could affect our methodology. of multicollinearity (Fig. 2).
The game estates are completely fenced and supplementary Collinearity refers to the non-independence of predictor vari-
food (cereal crops and special wild boar fodder) is regularly pro- ables and can be problematic for parameter estimations as it inflates
vided to wild boar. Both estates are divided into two different the variance of regression parameters and, hence, potentially leads
areas: a free ranging area, where wild boar live in natural conditions to the misidentification of relevant predictors in a statistical model
and where different game activities are carried out, mainly in the (Dormann et al., 2013). One way to better assess multicollinear-
autumn–winter; and a 25 Ha breeding area (farm) where wild boar ity than the assessment of the correlation matrix is to compute
are bred in captivity and later released into the free ranging areas the variance inflation factor (VIF). The VIF quantifies the amount of
for hunting. All of the animals that are born in the farm areas are multicollinearity in simple linear regression models. VIF is the ratio
identified with a subcutaneous microchip (Friendchip, AVID Iden- of the variance of each parameter in the linear regression model,
tification Systems, Spain) at the age of two or three months. These when fitting the full model, divided by the variance of individual
microchips are injected at the base of the left ear. Subsequently, parameters if fit on its own (James et al., 2013). The smallest pos-
at the age of 6 months, identified animals are released into the sible value for VIF is 1, which indicates the complete absence of
free-ranging areas where they live until they are hunted. collinearity. Often in the literature, a VIF value exceeding 5 to 10
is taken to indicate a problematic degree of collinearity. Table 1
2.2. Animals

During the years 2012, 2013 and 2014, a total of 93 marked

wild boar males were hunted in the two study areas (38 from the
south-eastern estate and 55 from the central estate). The identifi-
cation numbers of these animals were checked using a microchip
reader and their ages were determined on the basis of the farms’
data, ranging between 2 and 8.5 years old. The jaws from these 93
animals were collected and transported to the laboratory.

2.3. Dental measures

Some dental measures were used as possible estimators of wild

boar age. Two lower incisors, the primary incisor (I1) and the sec-
ondary incisor (I2), were removed from each wild boar jaw. For
each incisor we measured different parameters: external aperture
of pulp cavity (the maximum diameter of the root aperture), min-
imum root length and maximum crown length (taken the most
cervical position of the enamel-cementum junction at the labial
crown surface as the landmark for measuring both maximum
crown length and minimum root length) (Fig. 1). All the measures
were taken with a digital calliper (±0.1 mm).

2.4. Development of the age estimation model

Fig. 2. Correlation analysis among different parameters measured in this study.
In order to explore the relationship between the dental Results show multicollinearity among dental measurements and a strong corre-
lation between age and pulp cavity I1 and I2. The areas of the circles represent the
parameters measured and age, the correlations between possible
absolute value of the correlation coefficient, while the colour represents the sign of
estimators and the actual age of the wild boars were studied. Some the correlation. (For interpretation of the references to colour in this figure legend,
of the variables showed a good correlation with age but also showed the reader is referred to the web version of this article.)

Please cite this article in press as: Gonçalves, P., et al., A new method for ageing wild boar using dental measures. Ecol. Indicat. (2015),
http://dx.doi.org/10.1016/j.ecolind.2015.11.013
G Model
ECOIND-2745; No. of Pages 5 ARTICLE IN PRESS
P. Gonçalves et al. / Ecological Indicators xxx (2015) xxx–xxx 3

Table 1 Table 2
VIF factor for the parameters of each variable. Values greater than 10 are assumed Relative importance of variables estimated by the BRT model. The most important
to indicate a problematic amount of collinearity. variable was pulp cavity I1 by one order of magnitude.

Variable VIF Variable Relative influence

Pulp cavity I1 9.38 Pulp cavity I1 69.61

Minimum root length I2 10.80 Minimum root length I2 8.80
Minimum root length I2 94.65 Minimum root length I1 7.62
Maximum crown length I2 70.81 Maximum crown length I2 6.64
Maximum crown length I1 60.70 Maximum crown length I1 4.27
Pulp cavity I2 79.81 Pulp cavity I2 3.10

shows the VIF computed for all the explanatory variables indicating width of each incisive by drawing on the radiograph perpendicular
a problem of multicollinearity in these data. lines to the concave surface of each tooth (at a distance of 10 mm
To avoid multicollinearity, we propose a model to estimate wild from the tip of the root). We also estimated the ratio between pulp
boar age based on Boosted Regression Trees (BRT), a state-of-the- cavity diameter and the total tooth width, as this ratio has been
art regression method. BRTs are based on the idea of adaptively proven to correlate well with the age of wild boar. We assessed the
combining large numbers of relatively simple tree models, to correlation between this ratio and the known age of each wild boar,
optimize predictive performance (Elith et al., 2008; Hastie et al., and examined the agreement with the results obtained by our BRT
2009). This compromise between model complexity and predictive model.
performance provides a strategy to cope with problems of multi-
collinearity (Dormann et al., 2013). This method has been widely
3. Results
used to generate predictive models in ecological and biological
studies (Elith et al., 2008).
3.1. Age estimation with the BRT model
Unlike simple regression trees, BRT models can be more difficult
to interpret. However, they can provide some summary statistics
After carrying out the BRT modelling process, we obtained a final
that allow a better understanding of the final model. The “relative
age estimation model that included all the explanatory variables
importance” and the “partial dependence” plots are two such statis-
proposed (dental measures) and showed an explained deviance
tics. The “relative importance” measures the contribution of each
percentage of 61% that was estimated using cross-validation.
variable to improving the final model, scaled such that the total
Table 2 shows the relative influence of each explanatory variable,
sum reaches 100, with higher numbers for more important vari-
i.e. its contribution to explaining the final model. The external aper-
ables (Elith et al., 2008; Hastie et al., 2009). “Partial dependence”
ture of pulp cavity of the primary incisor was the most important
plots are graphical tools to quantify the effect of one variable on the
variable, by one order of magnitude. The effect of each explanatory
response after accounting for the average effects of the remaining
variable on the response variable is graphically showed in Fig. 3.
variables in the model (Elith et al., 2008).
To fit the BRT model, we ran an initial model using all the
explanatory variables proposed. Subsequently, two parameters 3.2. Influence of sampling area
required fixing: “learning rate” and “tree complexity”. The num-
ber of trees and the estimated deviance in the final BRT model was To assess the influence of the origin of the animal, the BRT model
estimated using a 10-fold cross-validation (CV) technique (Hastie was rebuilt using “Zone” as an explanatory variable. The estimated
et al., 2009). In this technique, the set is randomly divided into 10 explained deviance does not change statistically when including
subsets. In each run, 9 of the subsets are used as training data and this variable, yielding a result of 0.66. This indicates that the inclu-
the remaining subset as a testing set, with different combinations in sion of “Zone” has no effect on the BRT model estimation of age.
subsequent runs. This allows for error measures for the model using
data different to those used to train the model (Hastie et al., 2009; 3.2.1. Comparison with other age estimation methods
James et al., 2013). The number of explanatory variables included The proportion between pulp cavity width and tooth width was
in the final model was also chosen using CV methods. statistically correlated with wild boar age when this ratio was cal-
culated using both the incisor 1 measures and incisor 2 measures
2.5. Assessing the influence of the sampling area (p < 0.05). The percentages of variance explained using these ratios
were 41% and 33%, respectively. These values were lower than the
To assess whether the variables proposed as predictors have a percentage of variance explained by the BRT method (61%).
similar influence in animals from the two different areas, all data
were divided into two populations depending on where the animals 4. Discussion
came from, and a binary variable, called “Zone”, was included in the
BRT analysis. The relevance of the “Zone” variable was assessed by This study provides a new method for wild boar age estimation.
the relative importance provided by the final BRT model. We provide a predictive model based on three simple measures:
external aperture of pulp cavity, maximum root length and maxi-
2.6. Comparison to other age estimation methods mum superficial length from primary and secondary lower incisors.
Our model explains a high percentage of estimated deviance (61%),
The results obtained in our BRT model were compared with obtained by cross-validation. This validation technique provides a
another contrasting age estimation method designed for wild boar: fair assessment of the model results, and allows us to characterize
the pulp cavity ratio. This technique is also based on dental mea- the generalization ability of the model.
sures from wild boar incisors (Sáez-Royuela et al., 1989). More The predictive model proposed appeared to work for both sam-
specifically, it consists of measuring incisor pulp cavity thickness pling areas. When we include “Zone” in our model, this variable
using a medical X-ray machine. We placed the extracted teeth on only explains a 5% of the final explained variance, and this percent-
the X-ray machine table with the left side surface towards the table. age may be due to the fact that the mean age of the animals differed
We standardized measurements of pulp cavity diameter and tooth between the two areas (Fig. 4).

Please cite this article in press as: Gonçalves, P., et al., A new method for ageing wild boar using dental measures. Ecol. Indicat. (2015),
http://dx.doi.org/10.1016/j.ecolind.2015.11.013
G Model
ECOIND-2745; No. of Pages 5 ARTICLE IN PRESS
4 P. Gonçalves et al. / Ecological Indicators xxx (2015) xxx–xxx

Fig. 3. Partial dependence plots showing the marginal effect of each explanatory variable on the response variable, considering the average effect of other independent
variables.

management. Age estimations in wild boar older than three years

would allow the study of the growth of canines, which, in turn,
could provide managers of wild boar populations the tools neces-
sary to improve the quality of this hunting trophy. Thus, testing this
method with different subspecies and management systems would
be an interesting next step.
Several age estimation methods have been developed for older
animals, e.g. pulp cavity ratio method (Sáez-Royuela et al., 1989)
and a method based on counting the incremental lines of tooth
cementum (Clarke et al., 1992; Choquenot and Saunders, 1993).
We compared our method to the pulp cavity ratio method. The
pulp cavity method only explained a small percentage of variance
when we used this technique with our teeth of known age (41%
for I1 and 33% for I2) and thus results indicate a low explana-
Fig. 4. Boxplot showing the distribution of response variable “Age” and the origin
zone of the animal. tory power for this method. Sáez-Royuela et al. (1989) outlined
a method that showed a high correlation between age and their
Our findings show that the external aperture of pulp cavity of tooth measures, but they only used cementum layers to validate
I1 was the most important variable, by one order of magnitude their method. Cementum layers in mammals are associated with
(Table 1). Results suggest that this aperture is only present in the the annual period of active growth and the annual season of food
teeth of wild boar around 5 years old, which is consistent with a restriction (Spinage, 1973). Therefore, the age determination from
previous work in wild boar reporting that the closure of the root growth layers requires annual cement growth to be recognized
apex (except for a small permanent canal) in the lower permanent and counted (Choquenot and Saunders, 1993; Sáez-Royuela et al.,
first incisor occurs around 4–6 years of age (Habermehl, 1985). We 1989). In Spain, supplementary feeding is a standard management
included all the measures, which enabled us to calculate the age in practice for wild boar populations, especially in the south and
older animals, even when the external aperture was not present. south-west (Navarro-Gonzalez et al., 2013). This practice may affect
Thus, unlike other relative age methods such as the patterns of the annual cementum deposition and could thus decrease the pre-
tooth eruption and replacement or eye-lens weight, which only cision of age estimation.
allow the estimation of age up to 2 or 3 years (Choquenot and A boosted tree model is a statistical tool that has been used
Saunders, 1993; Matschke, 1967), our method works with older widely in ecology and biology (Elith et al., 2010; Elith and
wild boar males. We are currently working to test the validity of Leathwick, 2009; Juggins et al., 2015). Despite its complexity, our
our method with females. method is compatible with the design of software to allow wild
Developing new methods to improve age estimations, especially boar managers, taxidermists and others to easily estimate the age
in older animals, may be important to development of effective of wild boar.
management strategies. It has been demonstrated that age is cor-
related with different parameters such as space use, mortality and Acknowledgements
the incidence of some pathological processes (Gaillard et al., 1992;
Keuling et al., 2008a; Massei et al., 1996; Risco et al., 2013). More- This study was supported by Gobierno de Extremadura
over, accurate age estimates may be crucial to effective population (GRU10142, GRU10134) and the Collaboration Agreement between

Please cite this article in press as: Gonçalves, P., et al., A new method for ageing wild boar using dental measures. Ecol. Indicat. (2015),
http://dx.doi.org/10.1016/j.ecolind.2015.11.013
G Model
ECOIND-2745; No. of Pages 5 ARTICLE IN PRESS
P. Gonçalves et al. / Ecological Indicators xxx (2015) xxx–xxx 5

Consejería de Agricultura, Desarrollo Rural, Medio Ambiente y Hastie, T., Tibshirani, R., Friedman, J., 2009. The Elements of Statistical Learning.
Energía, UEX and FEDER. The authors acknowledge the Vet- Springer, New York.
Herrero, J., García-Serrano, A., Couto, S., Ortuño, V.M., García-González, R., 2006. Diet
erinarian Clinic Galan (Cáceres) for their technical support. P. of wild boar Sus scrofa L. and crop damage in an intensive agroecosystem. Eur. J.
Gonçalves acknowledges the University of Extremadura for her fel- Wildl. Res. 52, 245–250.
lowship. E. Serrano was supported by the postdoctoral program Herrero, J., García-Serrano, A., García-González, R., 2008. Reproductive and demo-
graphic parameters in two Iberian wild boar Sus scrofa populations. Acta Theriol.
(SFRH/BPD/96637/2013) of the Fundação para a Ciência ea Tecnolo- (Warsz.) 53, 355–364.
gia, Portugal. James, G., Witten, D., Hastie, T., Tibshirani, R., 2013. An Introduction to Statistical
Learning. Springer.
Juggins, S., Simpson, G.L., Telford, R.J., 2015. Taxon selection using statistical learning
References techniques to improve transfer function prediction. Holocene 25, 130–136.
Keuling, O., Stier, N., Roth, M., 2008a. Annual and seasonal space use of different age
Anezaki, T., 2009. Estimating age at death in Jomon Japanese wild boar (Sus scrofa leu- classes of female wild boar Sus scrofa L. Eur. J. Wildl. Res. 54, 403–412.
comystax) based on the timing of molar eruption in recent comparative samples. Keuling, O., Stier, N., Roth, M., 2008b. How does hunting influence activity and spatial
Mamm. Study 34, 53–63. usage in wild boar Sus scrofa L.? Eur. J. Wildl. Res. 54, 729–737.
Apollonio, M., Andersen, R., Putman, R., 2010. European Ungulates and Their Man- Magnell, O.,2006. Tooth wear in wild boar (Sus scrofa). In: 9th ICAZ Conference.
agement in the 21st Century. Cambridge University Press. Oxbow Books, pp. 188–202.
Bieber, C., Ruf, T., 2005. Population dynamics in wild boar Sus scrofa: ecology, elas- Massei, G., Genov, P.V., Staines, B.W., 1996. Diet, food availability and reproduction
ticity of growth rate and implications for the management of pulsed resource of wild boar in a Mediterranean coastal area. Acta Theriol. (Warsz.) 41, 307–320.
consumers. J. Appl. Ecol. 42, 1203–1213. Massei, G., Kindberg, J., Licoppe, A., Gačić, D., Šprem, N., Kamler, J., Baubet, E.,
Clarke, C., Dzieciolowski, R., Batcheler, D., Frampton, C., 1992. A comparison of tooth Hohmann, U., Monaco, A., Ozolinš, J., Podgórski, T., Nickolay, M., Boštjan, P.,
eruption and wear and dental cementum techniques in age determination of Rosell, C., Náhlik, A., 2015. Wild boar populations up, numbers of hunters down?
New Zealand feral pigs. Wlildl. Res. 19, 769–777. A review of trends and implications for Europe. Pest Manag. Sci. 71, 492–500.
Choquenot, D., Saunders, G., 1993. A comparison of three ageing techniques for feral Matschke, G.H., 1967. Aging European wild hogs by dentition. J. Wildl. Manag.,
pigs from subalpine and semi-arid habitats. Wlildl. Res. 20, 163–171. 109–113.
Dormann, C.F., Elith, J., Bacher, S., Buchmann, C., Carl, G., Carré, G., Marquéz, J.R.G., Navarro-Gonzalez, N., Fernández-Llario, P., Pérez-Martín, J.E., Mentaberre, G.,
Gruber, B., Lafourcade, B., Leitão, P.J., 2013. Collinearity: a review of methods to López-Martín, J.M., Lavín, S., Serrano, E., 2013. Supplemental feeding drives
deal with it and a simulation study evaluating their performance. Ecography 36, endoparasite infection in wild boar in Western Spain. Vet. Parasitol. 196,
27–46. 114–123.
Elith, J., Kearney, M., Phillips, S., 2010. The art of modelling range-shifting species. Onida, P., Garau, F., Cossu, S., 2014. Damages caused to crops by wild boars (S. scrofa
Methods Ecol. Evol. 1, 330–342. meridionalis) in Sardinia (Italy). J. Mt. Ecol., 3.
Elith, J., Leathwick, J.R., 2009. Species distribution models: ecological explanation Oroian, T.E., Ioan, P., Oroian, E., Covrig, I., 2010. Methods of Age Estimation by Den-
and prediction across space and time. Ann. Rev. Ecol. Evol. Syst. 40, 677. tition in Sus scrofa ferus sp. Bull. UASVM Anim. Sci. Biotechnol., 67.
Elith, J., Leathwick, J.R., Hastie, T., 2008. A working guide to boosted regression trees. Quere, J., Pascal, M.,1984. Données préli-minaires sur l’utilisation de la squelet-
J. Anim. Ecol. 77, 802–813. tochronologie dans la détermination de l’âge individuel chez le sanglier (Sus
Fernández-Llario, P., Mateos-Quesada, P., 2003. Population structure of the wild boar scrofa scrofa). In: Symposium international sur le sanglier. INRA, Paris, France,
(Sus scrofa) in two Mediterranean habitats in the western Iberian Peninsula. Folia pp. 78–91.
Zool. 52, 143–148. Risco, D., Fernández-Llario, P., García-Jiménez, W., Gonçalves, P., Cuesta, J., Martínez,
Gaillard, J.-M., Sempéré, A.J., Boutin, J.-M., Laere, G.V., Boisaubert, B., 1992. Effects of R., Sanz, C., Sequeda, M., Gómez, L., Carranza, J., Hermoso-de-Mendoza, J., 2013.
age and body weight on the proportion of females breeding in a population of Influence of porcine circovirus type 2 infections on bovine tuberculosis in wild
roe deer (Capreolus capreolus). Can. J. Zool. 70, 1541–1545. boar populations. Transbound. Emerg. Dis. 60, 121–127.
García-Jiménez, W., Fernández-Llario, P., Benítez-Medina, J., Cerrato, R., Cuesta, J., Sáez-Royuela, C., Gomariz, R.P., Tellería, J.L., 1989. Age determination of European
García-Sánchez, A., Gonçalves, P., Martínez, R., Risco, D., Salguero, F., Serrano, wild boar. Wildl. Soc. Bull., 326–329.
E., Gomez, L., Hermoso-de-Mendoza, J., 2013. Reducing Eurasian wild boar (Sus Schley, L., Dufrêne, M., Krier, A., Frantz, A.C., 2008. Patterns of crop damage by wild
scrofa) population density as a measure for bovine tuberculosis control: Effects boar (Sus scrofa) in Luxembourg over a 10-year period. Eur. J. Wildl. Res. 54,
in wild boar and a sympatric fallow deer (Dama dama) population in Central 589–599.
Spain. Prev. Vet. Med. 110, 435–446. Spinage, C., 1973. A review of the age determination of mammals by means of teeth,
Habermehl, K.H., 1985. Altersbestimmung bei Wild-und Pelztieren. Paul Parey, with especial reference to Africa. Afr. J. Ecol. 11, 165–187.
Berlin.

Please cite this article in press as: Gonçalves, P., et al., A new method for ageing wild boar using dental measures. Ecol. Indicat. (2015),
http://dx.doi.org/10.1016/j.ecolind.2015.11.013