Diana Pilson University of Nebraska dpilson1@unl.edu

     I am an evolutionary ecologist -- I study both the ecological factors that cause natural selection and the genetic basis of ecologically important characters.  Using manipulative field experiments and quantitative genetic methods, I hope to understand constraints on response to natural selection.  My research focuses in two general areas: 1) the evolution of plant response to herbivory by insects, and 2) the evolution of plant mating systems.  Together with my graduate students and collaborators at other universities I am involved in three ongoing projects. These are the evolution of response to herbivory in wild sunflower, Helianthus annuus, the evolutionary and ecological consequences of transgene escape into wild populations, and the evolution of the sex ratio and mating system in Croton texensis, an annual in the Euphorbiaceae. Descriptions of these projects are found below.

Evolution of Plant Response to Herbivory

     Herbivory is generally detrimental to plant fitness.  In addition, genetic variation for resistance to herbivory is nearly ubiquitous in natural plant populations.  Together these data suggest that insects should frequently impose selection for plant resistance, and further, that such selection should proceed until either plant populations are maximally resistant or genetic variation is exhausted.  That selection has apparently not acted in this way has stimulated considerable theoretical interest.

Using Helianthus annuus (Asteraceae), a wild sunflower, and its associated herbivores as a model system we are examining the evolution of plant response to herbivory.  My students and I are examining several possible explanations for the abundance of genetic variation for resistance typically found in natural plant populations.  For example, characters that confer resistance, such as secondary chemicals or trichomes, may be costly for plants to produce.  If this is the case then highly resistant plants would have high relative fitness in the presence of herbivores but low relative fitness in the absence of herbivores.  Such a trade-off, together with spatially or temporally variable herbivore populations, could result in intermediate levels of resistance being maintained.

     Another possibility is that plants may evolve tolerance of damage rather than resistance to damage.  Tolerance and resistance are distinct characters: a plant is tolerant if damage causes little reduction in fitness, while it is resistant if it incurs little or no damage.  Although genetic variation for tolerance has been found in several plant populations, characters conferring tolerance are poorly understood.  Nonetheless, theoretical work suggests that tolerance and resistance do not evolve independently, and thus, that intermediate levels of resistance and tolerance could be the result of selection acting on these characters simultaneously.  In ongoing work we are examining natural selection on resistance and tolerance in Helianthus annuus, as well as examining possible mechanisms underlying tolerance.

     Finally, the evolution of resistance may be constrained by trade-offs between resistance and other characters, such as competitive ability.  For example, plants experiencing high levels of intraspecific competition flower later than plants growing in less crowded conditions.  Moreover, resistance to several herbivores is determined, in part, by flowering phenology.  For example, plants flowering relatively early in the flowering season are more susceptible to damage by the sunflower bud moth, Suleima helianthana (Lepidoptera: Tortricidae).  The flowering phenology of plants that are relatively poor competitors may be strongly affected by the amount of inter- or intra-specific competition experienced, and thus, resistance in these plants could change dramatically between competitive environments.  These observations suggest that resistance could be correlated with competitive ability, and thus, the evolution of resistance might be constrained by selection on competitive ability. Eric Sundvall, one of my current graduate students, is in the process of testing five specific hypotheses concerning the ways in which variation in the competitive environment might constrain the evolution of resistance.

Ecological and Evolutionary Consequences of Transgene Escape

     In collaboration with Allison Snow (Ohio State University), Loren Rieseberg (Indiana University), and Helen Alexander (University of Kansas) we are investigating the ecological and evolutionary consequences of transgene escape into wild populations.  Although several transgenic crops have been released commercially (e.g. Roundup-Ready soybeans and Bt-corn) the ecological consequences of the escape of transgenes into wild populations have been little investigated.  The risk of transgene escape is remote for some crops, such as corn, which are grown (at least in the US) outside the native range of their wild progenitors.  However this is not the case for sunflower.  Not only is commercial sunflower grown within the natural range of its wild progenitor, it is fully interfertile with its wild relatives.  Recent work indicates that crop genes escape into and sometimes persist in wild populations, suggesting that any transgenes released in commercial varieties of sunflower will escape as well.  Because commercial sunflower is susceptible to several herbivores that feed on developing seeds, Bt-sunflower is currently being developed by a number of seed companies.  Our work will first evaluate the fitness consequences of a Bt gene in wild populations of sunflower.  If a Bt gene is likely to increase plant fitness, presumably because it reduces herbivory by lepidopterans, then an escaped Bt-transgene is likely to spread in wild populations. 

     If a Bt gene is established in wild populations there are several possible ecological effects.  For example, release from lepidopteran herbivory may allow wild sunflower to become a better competitor or persist longer in succession.  Wild sunflower is already an important weed in corn and soybean in Nebraska and acquisition of a Bt gene may make eradicating weedy populations more difficult.  However, the effect of a Bt gene on the population dynamics of wild sunflower depends on what processes currently limit sunflower population size. For this reason we are investigating the effect of insect herbivory on the local and metapopulation dynamics of wild sunflower.

     Another possible ecological consequence of the escape of a Bt transgene into wild populations is that the guild structure of herbivores feeding on sunflower inflorescences could be altered. However, the effect of "removing" some herbivores (through the effects of a Bt transgene) on the population sizes of the remaining herbivores depends on competitive interactions among these species. Matt Paulsen, a graduate student in my lab, and Nick Pleskac, a former undergraduate honors student, are investigating competitive interactions among herbivores feeding in sunflower inflorescences. Their work will provide more definitive information on the potential effects of a Bt gene on herbivore community structure in sunflower, as well as inform the ongoing debate concerning the importance of competition among herbivorous insects.
 

Evolution of the sex ratio and mating system in Croton texensis (Euphorbiaceae)

Sex ratio evolution -- In populations of sexually reproducing organisms negative frequency dependent selection generally acts to maintain a sex ratio which reflects equal investment in males and females.  There are exceptions to this explanation, as for example, when local mate competition or local resource competition are present.  Another exception, often invoked for unequal sex ratios in plant populations, is predicted by sex allocation theory when the environment is heterogeneous and sex is environmentally determined.  In the case of environmental sex determination (ESD) the sex ratio is expected to be 1:1 in the population as a whole, but locally biased, from predominantly male to predominantly female, along an environmental gradient.

     Although spatial segregation of the sexes along an environmental gradient has frequently been reported for plants, the existence of ESD as a mechanism to explain this distribution has not been explicitly tested.  Croton texensis, a dioecious annual in the Euphorbiaceae, exhibits sex ratios which fluctuate among years from significantly female biased to significantly male biased.  Karin Decker, a former M.S. student, and I tested the hypothesis that these fluctuations could be caused by ESD together with year to year variation in environmental conditions.  Our data suggest that sex determination is genetic, but that environmental variation leads to sex-biased variation in germination and/or early mortality.  We are planning further experiments to verify or refute this suggestion.

Mating system evolution --There are two general explanations for the evolution of dioecy (separate sexes) vs. monoecy (or hermaphroditism) in plants.  One explanation is that separate sexes are favored by natural selection because no selfing can occur, and thus, inbreeding depression is avoided.  Models that focus on inbreeding find that the trade-off between the benefits of selfing and the costs of inbreeding depression determine whether monoecy or dioecy will evolve.  The second explanation for the evolution of dioecy is that resources can be efficiently allocated to either male function or female function, but not both.  Thus, resource allocation models find that the shape of the curve plotting fitness gained through female function against fitness gained through male function determines whether dioecy or monoecy will be stable.  However, these models have been difficult to test because most species are either strictly dioecious or strictly monoecious.  As a result there has been some controversy in the literature about which explanation more frequently accounts for the observed distribution of plant mating systems.

     Although Croton texensis is primarily dioecious, approximately 1 in 200 individuals in our study population is monoecious, bearing both male and female flowers.  This variation has allowed us to test the inbreeding depression and resource allocation models of the evolution of dioecy.  Results to date indicate that males and females efficiently share resources, suggesting that monoecy should be favored by natural selection.  However, seeds produced by hermaphrodites have a lower probability of survival following germination.   Thus, it appears that in C. texensis inbreeding depression, rather than constraints on resource allocation, favors the maintenance of dioecy.