Cottonwood and the River of Time: On Trees, Evolution, and Society

13. Migrant Trees

13 / miGRant tReeS t he geographical variation in phenotypic traits we now encounter as we sample natural populations carries the footprint of past history, and no other historical episode has left a larger footprint on temperate-zone vegetation than the last ice age. Glacial and interglacial climates have been major forces in shrinking and expanding the natural distribution ranges of species. withdrawal into temporary refuges during glacial periods, the experience of population shrinkage, the subsequent rate of migration during interglacials, the role played by the founding of new populations through distant dispersal , and the merging of migratory paths have all shaped the current pattern of genetic variation found in the landscape. moreover, while it is impossible to accurately reconstruct the detailed history of any species, paleoecologists have been busy in applying a variety of independent 115 116 migranT Trees tools, from the analysis of pollen records and macrofossils to the use of computer simulation, to come up with plausible approximations. add genetic data, and a new synergy arises. Lodgepole Pine The lodgepole pine data from wheeler and Guries1 provided an attractive material on which les Cwynar and Glen macdonald, two paleoecologists , could test several hypotheses about the likely postglacial history of that species and its genetic consequences. They hypothesized that the range expansion of lodgepole pine happened through successive founding events, in which long-distance seed dispersal established new, small populations ahead of the main body; such a population would then expand and serve as the source for another pioneering group, and so on. But each founding event, owing to the small numbers involved, would also result in some loss in allele diversity because of chance alone, a process termed genetic drift (as we saw in chapter 8). if so, one would expect genetic diversity to decrease in the direction of migration. They also hypothesized that since every seed-dispersal event offered a test for dispersibility, repetitive selection at successive founding events would favor traits that improved such dispersal. if so, the most recently established populations should have the most dispersible seeds.2 one of the possible, if indirect, tests for these hypotheses was to put the northward expansion of lodgepole pine into a temporal framework. Collections of fossil pollen, combined with carbon dating, made it possible to date the sequential appearance of pine at successive locations along its 2,200 km migratory trek from the south to the Yukon during the 12,000 years of its postglacial history. fifteen of these locations were close to lodgepole sample populations of the latifolia subspecies in the wheeler and Guries (or W&g) study. when Cwynar and macdonald plotted data from these populations against the time since their founding , several trends emerged. one such trend showed the mean number of alleles per locus had increased with time since founding; or, conversely , that genetic diversity was lower in the more recently founded northern populations, as predicted. as to the dispersibility of seed, of the thirteen morphological traits of seed and cones measured by W&g, migranT Trees 117 seed mass and wing length were the only ones showing a relationship to time since founding, and clearly so; the most recently established populations had the smallest seed with comparatively the largest wings, that is, seed that could be transported farthest by lateral winds.3 here, too, the data supported the prediction. But how reliable was the assumption that the northernmost populations were really the most recently established? Remember that the allozyme data by W&g were highly suggestive of the persistence of some lodgepole pine in the north through the last glacial maximum. which data are more informative? Perhaps it may be helpful to briefly familiarize ourselves here with paleoecological methods and their strengths and weaknesses before showing how combining these methods with population-genetic data can strengthen the accuracy of our reconstruction of the past. Such reconstruction of the past need not be viewed merely as an academic exercise; it may give us some clues to the ways biota will respond to future shifts in climate. fossils are keys to the past. Just as we trace the evolution of Homo sapiens by examining fossil evidence from preserved skulls, teeth, or other skeletal parts, so do paleobotanists benefit from the study of fossilized plant fragments, such as leaves, twigs, wood, and seed. These macrofossils can be found within buried layers in bogs and lake bottoms and often can be reliably linked to a specific genus or even species...