Part III: Conservation and Management384
been recently reintroduced in the UK, Ireland (McDevitt et al.
2013), Sweden, Denmark (the population is recovering from
Germany; Jordt et al. 2015), and Norway (wild boar is spreading
from the Swedish population; Rosvold & Andersen 2008). The
magnitude of this population decline is exemplified by the case
of the British Isles: while in the Mesolithic there were around
one million wild boar (Yalden 1999), they became extinct around
the thirteenth century due to over-hunting and habitat destruc-
tion. It was not until the 1980s that free-living populations of
wild boar, originating from individuals escaped from farms,
were established in the Weald in Kent/Sussex and the Forest of
Dean (Goulding 2011). Similarly, Italian wild boar underwent a
severe demographic reduction during the eighteenth and nine-
teenth centuries, although they were not completely extermi-
nated (Ghigi 1911). At the beginning of the twentieth century,
in Italy, wild boar could be only found in Sardinia and some cen-
tral–southern regions of the country (Ghigi 1911). Nevertheless,
a considerable demographic expansion of this species followed
World War II, and now it is spread throughout the whole Italian
Peninsula (Vernesi et al. 2003).
The main causes of the demographic recession of wild boar
have been habitat loss and fragmentation, due to deforestation
and changes in land use, and, above all, over-hunting. Indeed,
94 per cent of wild boar deaths in Sweden are caused by hunt-
ing, 4 per cent by car accidents, and 2 per cent by other causes
(Rosvold & Andersen 2008). Historically, wild boar hunting,
trapping, and poisoning were routinely practised by farmers to
prevent the heavy damage that this species can inflict to crops.
Hunting was also a means of subsistence and recreation. In cer-
tain cultures, this antagonistic relationship between humans
and wild boar had a religious dimension. In Japan, the word
tatari designates the curse of the hunter by the wild boar. Fear
to this supernatural threat prevented excessive wild boar cull-
ing and promoted the celebration of ritual requests to pacify the
souls of the killed wild boar (Knight 2003).
Climatic factors can also modulate wild boar demography.
In northern Europe, freezing and snowy winters may decimate
wild boar populations, because piglets have a poor control of
thermoregulation and, in adults, rooting becomes unfeasible
when the ground is frozen and covered by a thick layer of snow
(Rosvold & Andersen 2008). In Mediterranean countries, dry
and hot summers can also increase wild boar mortality due to the
unavailability of food and water and the hardening of the ground
(Cahill et al. 2003). Epidemics can be another cause of wild boar
decimation. Classical swine fever outbreaks have plagued wild
boar populations from Austria, Germany, Ukraine, France, Italy,
Slovakia, and Belgium (Artois et al. 2002), being characterized by
a sudden mortality peak followed by a long phase of decrease in
the infection rate. Pasteurellosis (Risco et al. 2013), salmonellosis
(Conedera et al. 2014), and pseudorabies (Gortázar et al. 2002)
can also be associated with significant mortality rates in wild boar.
Sharp demographic reductions can modify the patterns
of genomic variation substantially and rapidly. A sudden
decrease in population size involves a loss of genetic diversity
that mostly affects low-frequency alleles, because they are more
prone to extinction than those at medium and high frequencies(Gattepaille et al. 2013). This deficit of low-frequency alleles
can be detected, e.g. with Tajima (Tajima 1989) and Fu and Li
D-statistics (Fu & Li 1993). Another potential signature left by
population bottlenecks is a slower decay in linkage disequi-
librium motivated by the loss of haplotypic variation and the
stronger impact of genetic drift on small-sized populations
(Slatkin 2008). A bottlenecked population may also show an
increase in the frequency and length of genomic segments that
are fully homozygous (Gattepaille et al. 2013), the so-called runs
of homozygosity (ROH). In pigs, this genomic feature has been
shown to be strongly associated with population demography
(Bosse et al. 2012). It is not straightforward, however, to dis-
criminate the genetic signatures produced by bottlenecks from
those caused by inbreeding, low recombination, and population
structure (Gattepaille et al. 2013).
The analysis of wild boar populations from Portugal with
a set of microsatellites provided evidence that they had gone
through recent bottlenecks (Ferreira et al. 2009). This study was
based on the calculation of an M-ratio statistic comparing the
number of observed alleles with the range in allele size (Garza
& Williamson 2001). When there is a sudden demographic
decline, the former parameter is expected to decrease faster
than the latter one. The demographic history of wild boar in
Portugal is consistent with these findings, i.e. in the 1960s, the
census of this species was so low that hunting was prohibited
(Fonseca 2004). In contrast, other studies focusing on Tunisian
(Hajji & Zakos 2011) and European wild boar (Vernesi et al.
2003; Scandura et al. 2008) did not provide conclusive evidence
of past population bottlenecks. Discrepancies among studies
might have methodological and biological causes. Indeed, the
successful identification of past bottlenecks is heavily influ-
enced by their magnitude and the time elapsed since they hap-
pened. For instance, the emergence of new mutations some
time after the ending of a bottleneck would lead to an excess
of low-frequency variants, thus masking the initial bottleneck
footprint (Gattepaille et al. 2013). Moreover, the demographic
history of wild boar seems to have been very complex, with a
high differentiation of populations inhabiting neighbouring
locations (thus evidencing that drift has been strong) and, para-
doxically, a weak divergence between distant populations as a
consequence of restocking.
High-throughput genotyping and sequencing methods have
made it possible to detect genomic segments that are homozy-
gous. These ROH emerge when identical haplotypes are inher-
ited from each parent, and their length and abundance reflect
past demographic changes as well as the amount of inbreeding
and local recombination (Bosse et al. 2012). Herrero-Medrano
et al. (2013) analysed Iberian wild boar with the Porcine 60K
SNP BeadChip and observed a high frequency of short ROH that
covered less than 20 per cent of the total genome (Figure 34.8).
This finding is consistent with the occurrence of historical bot-
tlenecks and low inbreeding in recent times (Herrero-Medrano
et al. 2013). We have observed a similar pattern in Romanian
wild boar (Manunza et al. 2016), where the majority of ROHs
were in the range of 5–20 Mb and their total genome coverage
was below 10 per cent. In contrast, wild boar from Sardinia.03612:55:56