183
of adult individuals of the Eastern populations could have induced genetic drift
processes and, therefore, an increase of population homozygosity. A simple sampling design was used to indirectly test this hypothesis. The isozymic genetic diversity of the set of 14 West-East oriented pairs of populations was compared. Each
two compared populations were at a distance between 3 and 15 km. In all the cases,
the seeds of the fragmented population located at the East had a higher genetic
diversity than the seeds from the West, in some pairs even up to 1.5 times higher.
The level of historical gene flow was also estimated through the number of migrants
(Ne m) and varied between 3.04 and 27.55. This was interpreted as the result of an
extensive gene flow which, through the unidirectional West-East winds characteristic of the region, moved genetic information through pollen (Gallo et al. 2004a),
similarly to the movement of seeds and vegetative propagules downstream rivers
(Schleuning et al. 2011). Levels of genetic diversity and differentiation were moderate and comparable with other conifers (Gallo et al. 2004a). The genetic analysis
defined three clusters: one cluster consisting of populations with moderate use
located in mixed Nothofagus pumilio-Araucaria araucana forests, a second cluster
composed of eastern and severely affected populations where past and present
human activities affected natural regeneration and a third cluster consisting of continuous forests under moderate use.
The development of new molecular tools for studying A. araucaria (e.g. SSRs,
simple sequence repeats) allowed a deeper understanding of the distribution of the
genetic diversity. Ten nuclear SSR markers (Marconi et al. 2011; Martín et al. 2012)
were used to genotyped 254 individuals from 15 populations, located either on glaciated or non-glaciated sites from the eastern range. In general, high levels of diversity were found, and the genetic differentiation among populations was moderate
and significant (Gst″ = 0.349, p = 0.001). A general trend of increasing allelic richness with latitude (i.e. north to south) was detected (R = 0.71 p = 0.003). Interestingly,
a significantly higher allelic richness as well as more private alleles was observed in
eastern populations, i.e. those located outside the limits of the Last Glacial
Maximum, which coincides with the most fragmented and disturbed populations
(Table 7.1). Bayesian clustering (BAPS; Corander et al. 2008) revealed the existence of five clusters, three of them composed of a single population (Fig. 7.3b).
Similarly, discriminant analysis (DAPC; Jombart et al. 2010), after running the find.
clusters routine, encountered five clusters with the two first discriminant functions
separating three groups of clusters: cluster 2, clusters 1 and 3 and clusters 4 and 5
(Fig. 7.4). Clusters 4 and 5 predominated in populations located at eastern and more
fragmented populations outside the limits of the LGM (Fig. 7.4).
To deepen the knowledge on the levels of effective gene flow, a parentage analysis
was performed using highly polymorphic microsatellite markers (Moreno et al. 2009,
2011; Moreno 2012). Two areas with different characteristics were chosen: the area
of Tromen (39° 37′ 02″ S, 71° 20′ 23″ W) has a continuous forest towards the west
and extends to more fragmented locations to the east, while the other area, named
7 Araucaria araucana and Salix humboldtiana…
of adult individuals of the Eastern populations could have induced genetic drift
processes and, therefore, an increase of population homozygosity. A simple sampling design was used to indirectly test this hypothesis. The isozymic genetic diversity of the set of 14 West-East oriented pairs of populations was compared. Each
two compared populations were at a distance between 3 and 15 km. In all the cases,
the seeds of the fragmented population located at the East had a higher genetic
diversity than the seeds from the West, in some pairs even up to 1.5 times higher.
The level of historical gene flow was also estimated through the number of migrants
(Ne m) and varied between 3.04 and 27.55. This was interpreted as the result of an
extensive gene flow which, through the unidirectional West-East winds characteristic of the region, moved genetic information through pollen (Gallo et al. 2004a),
similarly to the movement of seeds and vegetative propagules downstream rivers
(Schleuning et al. 2011). Levels of genetic diversity and differentiation were moderate and comparable with other conifers (Gallo et al. 2004a). The genetic analysis
defined three clusters: one cluster consisting of populations with moderate use
located in mixed Nothofagus pumilio-Araucaria araucana forests, a second cluster
composed of eastern and severely affected populations where past and present
human activities affected natural regeneration and a third cluster consisting of continuous forests under moderate use.
The development of new molecular tools for studying A. araucaria (e.g. SSRs,
simple sequence repeats) allowed a deeper understanding of the distribution of the
genetic diversity. Ten nuclear SSR markers (Marconi et al. 2011; Martín et al. 2012)
were used to genotyped 254 individuals from 15 populations, located either on glaciated or non-glaciated sites from the eastern range. In general, high levels of diversity were found, and the genetic differentiation among populations was moderate
and significant (Gst″ = 0.349, p = 0.001). A general trend of increasing allelic richness with latitude (i.e. north to south) was detected (R = 0.71 p = 0.003). Interestingly,
a significantly higher allelic richness as well as more private alleles was observed in
eastern populations, i.e. those located outside the limits of the Last Glacial
Maximum, which coincides with the most fragmented and disturbed populations
(Table 7.1). Bayesian clustering (BAPS; Corander et al. 2008) revealed the existence of five clusters, three of them composed of a single population (Fig. 7.3b).
Similarly, discriminant analysis (DAPC; Jombart et al. 2010), after running the find.
clusters routine, encountered five clusters with the two first discriminant functions
separating three groups of clusters: cluster 2, clusters 1 and 3 and clusters 4 and 5
(Fig. 7.4). Clusters 4 and 5 predominated in populations located at eastern and more
fragmented populations outside the limits of the LGM (Fig. 7.4).
To deepen the knowledge on the levels of effective gene flow, a parentage analysis
was performed using highly polymorphic microsatellite markers (Moreno et al. 2009,
2011; Moreno 2012). Two areas with different characteristics were chosen: the area
of Tromen (39° 37′ 02″ S, 71° 20′ 23″ W) has a continuous forest towards the west
and extends to more fragmented locations to the east, while the other area, named
7 Araucaria araucana and Salix humboldtiana…