Species level phylogeny of the genus Cornus (Cornaceae) based on molecular and morphological evidence -: implications for taxonomy and Tertiary intercontinental migration

DNA sequences were generated for matK and ITS for 68 and 103 samples of Cornus to reconstruct a species level phylogeny of the genus. The results support the monophyly of most subgenera, except subg. Kraniopsis and suba. Cornus. Subgenus Kraniopsis was suggested to exclude C. peruviana from South America and subg. Afrocrania and subg. Sinocornus were nested within subg. Cornus. Four major clades corresponding to groups also recognizable by morphological differences were revealed: the big-bracted dogwoods (BB) including subg. Cynoxylon, subg. Syncarpea, and subg. Discocrania, the dwarf dogwoods (DW) including subg. Arctocrania, the cornelian cherries (CC) including subg. Cornus, subg. Sinocornus, and subg. Afrocrania, and the blue- or white-fruited dogwoods (BW) including subg. Kraniopsis, subg. Mesomora, and subg. Yinquania. This finding is consistent with previous Studies with more limited sampling. The single South American species C. peruviana was found to be sister to the Asian C oblonga of subg. Yinquania, adding a fourth intercontinental disjunction in the genus that was previously unknown. Species relationships within the subgenera were clearly resolved except for the relatively large subg. Syncarpea and subg. Kraniopsis. Phylogenetic analyses of total evidence combining morphology, matK, ITS, and previously published rbcL and 265 rDNA sequences resolved the relationships among subgenera as (BW(CC(BB,DW))). Biogeographic analyses using DIVA with Or without fossils resulted in different inferences of biogeographic history of the genus, indicating the importance of fossil data in biogeographic analyses. The phylogeny based on the total evidence tree including fossils Supports an origin and early Tertiary diversification of Cornus in Europe and multiple trans-Atlantic migrations between Europe and North America by the early Tertiary. It also supports that distribution of the few species in the southern hemisphere was not ancestral, but a result of migration from the north. This evidence rejects a previous hypothesis of a southern hemispheric origin of the genus.