Figure 1.
Location of the sampling sites.
ITS2 ribotype distribution and ITS2 network including 201 sequences of 481 nucleotides. Pie charts represent ribotype frequency in each population. Ribotype network was constructed using a median-joining algorithm with programme NETWORK [36]. Colours of the network correspond to those mapped with rare ribotypes in each “region” pooled under the same colour for clarity. Circle sizes are proportional to total ribotype frequency and connection lengths correspond to mutation steps (most of the ribotypes are separated by one mutational step but eight mutations separate r4 from r6).
Figure 2.
Genetic population structure of New Zealand and Chilean Gracilaria populations.
(A) based on Bayesian analysis with STRUCTURE software [37], proportions of individual multilocus genotypes were assigned to each of the 5 virtual genetic clusters indicated by the different colours; (B) based on graph analysis with POPULATION GRAPH software [40], differences in node size reflect differences in within population variability whereas edge lengths represent the genetic variation between pairs of samples, there were significantly more edges within New Zealand and Chilean groups than between the two groups (P<0.0001). Node colour corresponds to the 5 clusters as determined by STRUCTURE. For both analyses: N = 7 New Zealand populations, 11 Chilean populations, 567 individuals, 5 codominant loci. Population codes are provided in Table S1.
Figure 3.
Number of alleles observed in each of the 10 allele frequency classes defined in New Zealand and Chilean populations (pooled data from each side of the South Pacific coast).
Reduced number of low frequency alleles is characteristic of a bottleneck [47].
Table 1.
Tests of demographic changes based on ITS2 sequences.