(a) Contemporary introgression among the eastern North American
white oaks has been a case study in the oak syngameon since the mid-1970s (Burger, 1975; Hardin, 1975; Van Valen,
1976) and was the focus of the first chloroplast phylogenetic study
in oaks (Whittemore & Schaal,
1991). Introgression at this scale occurs primarily within the range
of species sympatry and reflects recent generations of gene flow. The figure
illustrates observed gene flow between seven eastern North American white oaks
based on analysis of 80 SNPs that map back to all 12 Quercus
linkage groups, with distances between uniquely mapped loci an average of 7.47
million bp (± 8.74 million bp, s.d.). The figure replicates the
16-species figure of Hardin 1975 (his Fig.
1), including only the subset of species investigated in Hipp et al. (2019b), with
lines indicating hybridizations that Hardin inferred from morphological study.
Thin dashed lines indicate hybridizations identified by Hardin but not by the
SNP genotyping; medium dashed lines were identified by both Hardin and SNP
genotyping, at an admixture level of 0.10 to 0.19 for at least one specimen; and
thick dashed lines indicate SNP admixture levels of 0.20 or higher for at least
one specimen. (Hipp et al.
2019b). (b) As species ranges change in response to
climatic transitions, species that were once sympatric may become allopatric.
Gene flow between them may still be detected by, for example, incongruence
between chloroplast and nuclear topologies (Whittemore & Schaal, 1991; Manos et al., 1999). Here, a RAD-seq (nuclear)
phylogeny for predominantly North American white oaks (left side) shows strong
species cohesion, whereas the phylogeny of whole chloroplasts for the same
samples sorts better by geography than by species (Pham et al., 2017). Such events may also
be detected by phylogenomic incongruence in nuclear loci (McVay et al., 2017a; e.g., Kim et al., 2018).
(c) With enough generations, ancient introgression events that
were sufficiently pervasive may remain detectable. A fascinating case involves
the sister species Q. pontica of the Caucasus, Q.
sadleriana of California, and the Eurasian white oaks, or
Roburoids. Because Q. pontica and Q.
sadleriana are sister species, they are perforce equally closely
related to all other oaks. With careful study, John McVay (2017b) and colleagues found that there are loci shared
among the Roburoid white oaks that ally them much more closely to the Eurasian
Q. pontica than to the American Q.
sadleriana. Panel (c) shows genomic evidence for introgressive
hybridization between wild Q. pontica samples and the Roburoid
oaks: scaled locus similarity on the y-axis between Q. pontica
and the Roburoids is significantly stronger than expected when data are
simulated under incomplete lineage sorting alone (p < 0.002). Observed
rankings of Roburoids to Q. sadleriana sample did not differ
significantly from simulated distribution. The fact that these loci are shared
among the Roburoids in spite of the fact that most are not sympatric with
Q. pontica suggests a history of ancient introgression
between the lineage that contains Q. pontica but not Q.
sadleriana (allowing for extinction, there may have been other
species) and an ancestral Roburoid. This finding is borne out with analysis of a
separate phylogenomic dataset using species tree methods that account for
hybridization. Thus a microevolutionary process, inferred to have preceded the
crown divergence of the Roburoids an estimated 7.6-12.1 Ma (Hipp et al. 2019a),
became frozen in the genome, like a fly in amber.