Contents:
San Francisco Church and its imposing yellow facade, is one the most representative, not only because the magnificent building is so well preserved, but also because it reflects the passage of Catholic missionaries across the region.
You can walk along its main trail, suitable for absolutely everyone — even for people with reduced mobility — and continue to trek along the beautiful and mysterious Pier of Souls. A magical wooden pier built on a cliff where, according to legend, you can hear the wails and pleas of the sorrowful souls waiting for the boatman Tempilkahue to ferry them away towards eternal rest. On this tour you will see that the Valdivian jungle predominates with a dense forest formed by evergreen trees, shrubs and climbing plants.
On the way through all these destinations, we recommend taking small breaks at the food stalls that pop up along the way. Here, there are several different small towns, among them Chullec stands out for its incomparable silence; Achao, has the largest wooden church on the islands, and Curaco de Velez is where the oldest church is located.
End the day in Dalcahue, which was inhabited by the Huilliches and Chonos peoples before the arrival of the Spaniards. Located just 25 kilometers from Castro, this small town is charming and travelers are advised to visit its small Plaza de Armas main square and walk along the coast to see its most emblematic places. This approach allows a better visualization of the heterogeneity of the genetic divergence patterns across a landscape, and has proven useful in phylogeographic studies [60].
The residual genetic distances are derived from the linear regression of all pairwise genetic distances on geographical distances, allowing removal of the effect of geographical distance. Historical changes in Ne for each of the main clades were evaluated using graphic reconstructions of population sizes through time estimated with a lineage-through-time LTT, Bayesian skyline plot analysis plot using BEAST 1.
Because the cyt-b rate of evolution for L. In order to estimate the age of the main nodes on the cyt-b tree and analyze the association between the geographic distributions of the haplogroups and their ages, we conducted an analysis with a Yule process of speciation using the software BEAST 1. To place the mean priors in the tree we used two fossils from the subgenus Eulaemus , both from Argentina.
One fossil correspond to We obtained a substitution rate of 6. The magnitude and spatial arrangement of gene flow was assessed with the coalescence approach implemented in Migrate 3. These analyses were based on the population clusters defined by Geneland. Within each of the two main clades see below we estimated directional gene flow between clusters and, when a cluster included widely distributed localities, we further delimited geographic areas in order to estimate directional gene flow between physiographic regions e. Coastal mainland vs. For all runs the first 20, genealogies were discarded as burn-in.
The criterion of Gelman was used to verify that sampling was made from the stationary distribution. We conducted an ecological niche modelling analysis in order to identify potential refugia for L. Ecological niches and potential geographic distributions were modeled using the maximum entropy method [67]. Climatic envelopes were estimated from 19 environmental variables that likely summarize niche dimensions that are particularly relevant to determining species distributions [68].
For the LGM and present climate conditions, we used two sets of monthly climate data precipitation and temperature. For current conditions means — , we used WorldClim, a global climate database with a spatial resolution of 2. Maxent generates Ecological Niche Models ENM using presence-only records, contrasting them with pseudo-absence data sampled from the remainder of the study area.
The present-day ENM were developed based on the confirmed occurrence points for L. As suggested by Waltari et al. This threshold identified smaller areas than a lowest-presence threshold that yielded zero omission error, thus resulting in a more restricted picture of potential distribution. We obtained an alignment of base pairs of the cyt-b gene from sequences gathered from specimens of L. We found a total of haplotypes and segregating sites. Pairwise genetic distance between localities ranged from 0.
The zero values correspond to comparisons between localities from the southernmost mainland range close to the Andes i. A list of the haplotypes per sampling site is given in Table S2. Of the unique haplotypes found, only 13 were shared among localities. Twenty of 46 localities included at least one shared haplotype.
Only Verified Singles. Take Control of Your Dating. Safe & Secure. Find Matches! 4 Million+ Single Women on the #1 International Dating Site. Advanced Matching. Start Now.
Twelve localities presented a single shared haplotype and 2 sites shared 3 haplotypes. Most of the shared haplotypes were detected in the central and southern range of the distribution of the species.
Phylogenetic reconstruction of the unique cyt-b sequences recovered two well-supported reciprocally monophyletic clades of L. Within the former, haplotypes from subclade N1A were all from the region of Las Trancas in the northern Andean Cordillera, which corresponds to the distributional range of the subspecies L.
Subclade N1B was widely distributed, weakly resolved, and included haplotypes from the Coastal Cordillera and the Andean Cordillera Fig. Branch colors as codes in the map of Fig. Numbers on tips of branches correspond to haplotypes detailed in Table S2. Numbers in parentheses are localities, grey and black bars identify the two main clades, and colored bars represent the subspecies of L. The larger clade S2 was composed of two subclades S2A and S2B having a moderate probabilistic support.
Subclade S2A was well supported and included haplotypes from the Western Andes e. These localities are within the Chilean distributional range of the subspecies L. The larger subclade S2B was a weakly supported polytomy of five subclades. The smallest of these, S2B1, was composed of one haplotype shared by individuals from the Andean locality of Malalcahuello Western Andes , an area geographically very close to some localities harboring haplotypes from the Northern clade.
Subclade S2B2 included five haplotypes with the longest branch lengths of the whole sample, most of them from a small region of the Coastal range corresponding to localities 16 MN Alerce Costero and 20 Close to Heicolla. The two subclades S2B3 and S2B4 were composed of haplotypes broadly distributed between western Argentina, the southern Chilean Andes and the Coastal range.
Only two haplotypes collected at localities 40 and 41 in the north of the island were not in this clade; both were part of clade S2B4. The number of segregating sites and the haplotype richness were lower in the Northern clade than in the Southern clade vs and 30 vs , respectively; Table 1 , and the respective values of haplotype richness standardized for sample sizes were 0. The average genetic distance between the Northern and Southern clades was 0. In general, all pairwise p-distance values involving comparisons between Northern and Southern clades are higher than values from comparisons among the Southern subclades.
In addition, the among-Northern subclade values are higher than among-Southern subclade values Table S4. Basal divergence within L. Clade N is estimated to have diverged approximately , ya ,—, , and clade S about , ya ,—, , during the second half of Pleistocene Table 2. The cyt-b haplotype networks are highly congruent with the Bayesian genealogy. The Northern and Southern haploclades are recovered as separate networks Fig. The root of the cyt-b network should be between the highly divergent N and S clades, and this allowed us to infer the ancestral haplotypes within each haplogroup Fig.
The inferred ancestral haplotypes from the mtDNA network showed the shortest branch lengths and the smallest number of changes from the root in the Bayes tree Fig. Colors are the same as in Fig. Overall, from the 10 clades revealed in the cyt-b network, four suggested ancestral haplotypes were from the Coastal range, five from localities in the western Chilean Andes, and one from the Eastern Argentinean Andes. Among the six Andean ancestral haplotypes, two were distributed in areas outside of the LGM ice shield boundaries, whereas the other four fell within this border.
For detailed distribution of ancestral haplotypes see Table S2. The nuclear gene networks Fig. S1 were partially congruent with the cyt-b network; 18 haplotypes were detected for each gene. Although the LDB5B alleles did not form reciprocally monophyletic geographic groups, contrary to what observed on the cyt-b gene tree, there was a clear tendency for the Northern haplotypes to form a single group. The position of the outgroups in this network suggests that haplotype B5, broadly distributed, is ancestral. The EXPH5 network did not recover discrete clades, rather all haplotypes are organized in a linear array of similar distances with Northern haplotypes at one end Fig.
The position of the outgroups in this network suggests that haplotypes E5 and E6 from the Western Andes are ancestral. In general terms, the samples of the Northern mtDNA clade clade N tend to also form a unique cluster in both nuclear gene networks, although they were not exclusive. The haplotypes in both nuclear genes showed much less structure relative to the mitochondrial network.
Grey haplotypes correspond to the outgroups L. Color codes as in the map of Fig. The general distributions of these clusters are detailed in the Table 3. Despite the strong correlation for the total pairwise comparisons, three clouds of spots could be detected Fig. One of these clouds shows high genetic distances both for geographically close and distant comparisons, and included the between N-S comparisons. This pattern was confirmed by the genetic landscape shape interpolation analyses for which no strong discontinuity could be observed from South to North Fig. This graphical representation shows more genetic homogeneity among southern localities than among those of the northern part of the range.
Demographic analyses showed signals of recent population expansions as expected for the Southern group but also for the Northern group, and Fu's neutrality tests gave significant negative Fs values for both Table 1. Mismatch distribution analyses provided values of the raggedness index for each group below 0. Bayesian skyline plot analyses showed a reduction of N e in both the Northern and the Southern group; although as expected, oscillations are more pronounced for the latter Fig. The N e reduction in both clades started about 50,—60, ya, and the lowest N e values were estimated around 16,—17, ya.
The Southern group showed a greater magnitude in both reduction and recovery of population size, although both groups show rapid and pronounced recoveries of N e with no signal of a slow down in this trend in the Southern group.
X and Y axes correspond to geographic locations within the overall physical landscape examined in this study Fig. Surface plot heights reflect genetic distances. Bayesian skyline plots and mismatch analysis for the two major mtDNA clades of L. The black line is the temporal trend of the mean of effective population size.
Vertical light blue band corresponds to the LGM. The assessment of gene flow among the groups within each clade revealed a mixed pattern Table 3. In the clade N, there is greater evidence of historical gene flow from Andean to Coastal populations. Globally, for all 15 comparisons the estimates of gene flow could be considered as non-zero values.
Among these comparisons, five revealed predominant gene flow from Andean localities lying within the LGM ice shield boundaries towards Coastal localities, but as predicted we find a clearly higher value of gene flow from Coastal to Andean populations only for the Southern cluster A. Ecological niche modeling predicts several fragmented refugia for L. The Maxent model predicts a current distribution close to the one known for L.
The habitat suitability landscape shows current high habitat probabilities for L. The LGM distribution predicted for L. During the LGM increased habitat suitability is predicted for the lowlands in the central distribution of L. Some small east-west areas following valleys in the west flank of the Andean Cordillera, and within the presumed boundaries of the large ice shield, may also have remained suitable.
According to the predicted sea level during the LGM, the potential distribution of L. The central and northern parts of the current distribution of L. The color scale corresponds to the probability for suitable habitat; red and light blue indicate highest and lowest suitability values respectively. The heavy blue line delimits the maximum extension of ice shield during LGM according to Heusser [11]. The analysis of the mitochondrial and nuclear genetic variation of Liolaemus pictus across its distribution revealed a complex phylogeographic pattern and demographic history.
Results indicate a mix of divergence due to isolated refugia followed by dispersal from them. Our results show the effect of recent glaciations on L. In fact, evidence of dispersal both from glaciated and non glaciated source areas was found, along ancestral haplotypes distributed both in the Andean and Coastal ranges. Our analyses show a complex phylogeographic history, considerable genetic structure, and signals of complex patterns of gene flow in L.
Both clades include Coastal and Andean localities and overlap in a small area close to Malalcahuello in the western Andean piedmont localities 8 and 9 in Fig. The northern group is heterogeneous, with well-differentiated and deeply divergent internal subgroups, whereas the southern group is more homogeneous see p distances in Table S4. In general, genetic diversity was higher at southern localities, and spatial analyses showed that this group, despite having higher haplotypic diversity, had a weaker geographic structure than the northern clade.
These analyses, along with the IBD analyses, suggest a rather continuous distribution of the genetic divergence throughout the range of L. However, these results must be interpreted with caution because they may be reflecting the differences between the highly differentiated clades rather than a specific pattern within these clades, as indicated by the three groups in the IBD scatterplot. The deepest splits between subclades were observed in Clade N; the MRCA of the phylogroup N1A from the northernmost locality in Las Trancas was 72, ya 15,—, ya , but the split from its sister clade N1B was , ya ,—, ya.
These divergence levels suggest strong fragmentation, low connectivity and reductions of Ne, especially in clades N1A and N2. Although the effects of the glacial cycles were less prominent in the North, Nothofagus forests are more fragmented and exposed to changes in size because those areas are close to the limits of their distribution [4] , [23].
Thus, while the effects of the glaciers in these regions may have been minimal, the marginal conditions of the habitat may have promoted fragmentation, differentiation, and unstable population sizes in L. In contrast, genetic distances within clade S are lower and the clade is characterized by shorter branch lengths, suggesting a greater level of dispersal throughout this area. Some of these haplotypes are shared by more than one locality and with a broad distribution e. Our results disagree with Vidal et al. This suggests that immigration from the continent occurred prior to the LGM, in response to the formation of land bridges linking the island with the mainland during previous glacial periods.
Although such recurrent events should facilitate the migration of organisms, L. Retreat of this ice shield may have also produced abundant meltwater, and such geologically recent fluvioglacial barriers may have prevented island-mainland gene flow. This highlights the importance of considering the temporal dynamics throughout each glacial cycle if one wants to understand the effects of the glaciations on gene flow. Although L. The split for both deer clades was estimated to approximately , ya, which is similar to our estimate for L.
Independent of the presence of small or medium sized fluvial barriers during glacial periods, this species should have been able to move between the island and the continent if the forest had colonized continuous formations over the exposed shelf, thereby erasing the pattern of reciprocal monophyly detected in our study. Similarly, there is no evidence of shared haplotypes between these two areas in the frog Eupsophus calcaratus [14].
For samples from clade S, recent gene flow is inferred from the occurrence of shared haplotypes among several localities, all of which include Southern clade and Andean sites within the maximum limits of the LGM ice sheet.