Selection Experiments in Flowering Plants

In my laboratory, we use naturally occurring genetic variation within species found in various environments to study how plants have adapted to environmental stressors of many kinds. In this particular study, we ask whether similar traits evolve in response to the stresses of low water, low nutrients, and low light in annual plant species. Although dominant growth forms may change from one environment to the next, comparative studies of many species have shown that many alternative growth forms are equally capable of survival in a given environment. Generalizations about the process of adaptation to environmental stress have therefore been elusive.

Our research is an attempt for such generalities. In particular, we hypothesize that efficient use of resources will be selectively favored in resource-poor environments. Moreover, this efficiency will evolve at a cost to the ability of plants to acquire resources rapidly and hence at a cost to growth. We hypothesize that the net result of selection in impoverished sites is evolution of a population containing slow-growing, efficient plants. To test these hypotheses, two sets of studies are being done: (1) a multigeneration selection experiment using Wisconsin Fast plants (Brassica rapa) as a model system, complemented by (2) field experiments with three annual plant species that have populations naturally found in low and high water (Impatiens pallida, touch-me-not); light (Abutilon theophrasti, velvetleaf); or nutrients (Poa annua, annual bluegrass).

Patricia Gardner, as a West Virginia University undergraduate student, completed a biology honors project using fast plants that yielded enticing preliminary results and helped define the conditions we eventually chose to use in the long-term selection experiment. She found that after only two generations, populations subjected to selection under low and high resource levels had evolved differences in developmental rate. Plants that had experienced a history of selection in high resource conditions reached the flowering stage earlier, regardless of the environment in which they were grown. Patricia has taken the lead on managing the expanded version of this experiment, which will be carried through for 10 generations. Measurements of growth and resource use efficiency will be made on the products of selection.

In the field component, Patricia, along with undergraduates Trisha Kalbaugh and Justine Swaney, helped transplant more than 1500 velvetleaf plants of known parentage into plots of corn (low light) and soybean (high light). Careful measurements of survival, growth, and reproduction were made throughout the 1994 growing season. In the previous season, nearly 3000 individuals of touch-me-not from genetically distinct source populations were planted on floodplain (moist) and hillside (dry) sites, with similar measurements made on all plants. By examining how the association between growth rate and fitness changes across such environmental gradients, we will partially test our original hypothesis. Detailed studies of resource use efficiency and acquisition rate in the field will be carried out in future work.