Genetic differences between wild and artificial populations of Metasequoia glyptostroboides:: Implications for species recovery

Information on population genetics is fundamental to developing in situ or ex situ conservation strategies. Few researchers, however, have compared the genetic structure of restored and natural populations of threatened plant species. Metasequoia glyptostroboides Hu & Cheng (dawn redwood) (Taxodiaceae), a living fossil endemic to China, may be the most successfully recovered threatened species, with many more individuals and a much wider distribution than fossil records indicate. We used random amplification of polymorphic DNA markers to compare the genetic structure of artificial populations with that of wild ones and to determine whether the genetic structure of M. glyptostroboides has been recovered as has its distribution. The genetic variation of wild populations of M. glyptostroboides was lower than the average of gymnosperms, indicating the effects of glaciations and recent habitat loss and fragmentation. Genetic variation in artificial populations was less, but not substantially, compared with wild populations. The unweighted pair group method with arithmetic mean revealed that the wild and the artificial populations formed two distinct groups. Artificial populations were more similar to each other (mean Nei's genetic distance = 0.0924) than to wild populations (mean distance = 0.2054). This might be the result of biased seed collection, vegetative propagation, or a mixture of propagules from different populations and an ultimate propagule source. These results suggest that although the quantity and distribution range have been successfully restored, the genetic structure of M. glyptostroboides has not recovered appropriately, given the loss of genetic variation and biased genetic composition in artificial populations. Therefore, in addition to protecting the wild populations, additional ex situ genetic reserves should be established based on genetic knowledge and via appropriate approaches. We suggest that population genetic and demographic indices should be considered when downlisting or delisting threatened species.