Figure 1.
Geographic distribution of both extant and extinct Carya species.
The black areas indicate the geographical distribution of extant Carya showing its disjunction between EA and ENA. Colored circles denote the locations of Carya fruit fossils (detailed information of each fossil in Table S4).
Figure 2.
Phylogeny of Carya based on the combined molecular data set.
ML bootstrap support > 75% and Bayesian posterior probability > 0.80 are indicated at the nodes, respectively. The topologies of strict consensus trees from maximum likelihood and Bayesian inference are same.
Figure 3.
Ancestral area reconstructions for Carya using DIVA.
A: Inference with fossil taxa; B: Inference without fossil taxa. Fossil taxa are indicated by cross symbols after their names. The areas of endemism are defined for both analyses: A, North America; B, Europe; C, Asia.
Figure 4.
Ancestral area reconstructions for Carya using LAGRANGE.
A: Inference with fossil taxa; B: Inference without fossil taxa. Fossil taxa are indicated by cross symbols after their names. A slash in the results of LAGRANGE indicates the split of areas in two daughter lineages. The areas of endemism are defined for both analyses: A, North America; B, Europe; C, Asia.
Figure 5.
BEAST chronogram and LTT plots of Carya.
A: The chronogram of Carya inferred from the combined molecular data set. Gray bars represent the 95% HPD intervals of node ages. Calibration points are indicated with asterisks. B: Lineage-through-time (LTT) plots for Carya. The thin lines indicate the 1000 random trees from the best-fitting model obtained in BEAST (EA clade: pink; ENA clade: light blue); the bold lines correspond to the maximum credibility tree from the BEAST analysis (EA clade: red; ENA clade: blue).
Figure 6.
Hypothetical migratory routes of Carya.
A: Eocene; B: Oligocene; C: Miocene; D: Pliocene and later. The colored arrows and circles indicate the migratory directions and Carya fruit fossil records, respectively.