EFFECTS OF ANTHROPOGENIC SELECTION PRESSURE ON THE EVOLUTION OF A COMMON AGRICULTURAL WEED: DOES ADAPTATION COME WITH A DETECTABLE COST? A STUDY OF GLYPHOSATE-RESISTANT (GR) AND -SUSCEPTIBLE (GS) BIOTYPES OF Conyza canadensis

Author

Karla Leigh Gage, Southern Illinois University CarbondaleFollow

Date of Award

5-1-2013

Degree Name

Doctor of Philosophy

Department

Plant Biology

First Advisor

Gibson, David

Second Advisor

Young, Bryan

Abstract

As the acreage of glyphosate-resistant (GR) cropping systems increases, so does the occurrence of herbicide-resistant weeds in the landscape. Biotypes of GR Conyza canadensis were first observed in Delaware, USA, in 2000. Since the first documentation of GR C. canadensis, there have been many instances of independent evolution of GR C. canadensis biotypes. The ecology of GR C. canadensis and any potential fitness consequences of GR trait are still unknown. If there is no fitness cost or a fitness increase associated with the GR trait, GR C. canadensis may increase in prevalence in the landscape even in the absence of glyphosate application. With the consideration that fitness is a difficult parameter to measure, other variables may be used as surrogate measures of fitness, such as reproduction, growth rate, phenology, survivorship, etc. This research seeks to determine if differences exist in surrogate measures of fitness - patterns of growth, reproductive allocation, and competitive ability - and if so, how differences may apply to future population changes, for glyphosate-resistant and -susceptible C. canadensis. The first two experiments presented test two populations of GR plants and two populations of GS plants, first in a greenhouse study of shading effects and intraspecific competition (Chapter 2), and second in a field survivorship study within two habitat types (Chapter 3). The third study tests for the prevalence of GR C. canadensis in field margins of GR cropping systems and seeks correlations in the occurrence of resistance and factors related to geography and field management (Chapter 4). The greenhouse study (Experiment 1, Chapter 2) of surrogate measures of fitness (growth, reproduction) in response to intraspecific competitive ability and shading showed that, while all populations had the same response to shading, there were differences between populations in growth and reproduction. While all measures were greater for individual plants with no competition, plants competing with individuals from one of the GR populations (R2) often had measures that were not different from the no competition treatment or were greater than the other competition treatments (R1, S1, and S2). The field survivorship study compared growth, survival, and reproduction of the previously studied four populations of C. canadensis is two habitats, a old-field in the second year of succession (ruderal) and a soybean field planted with a GR cultivar (agrestal) (Experiment 2, Chapter 3). One of the GR populations (R1) had the highest survivorship of the four populations in the ruderal habitat, while there was no difference in survivorship of the four populations in the agrestal habitat. While there was little difference in growth between ruderal populations, in the agrestal habitat, one GS population (S1) consistently had the smallest diameter, height, and leaf number. One of the agrestal GR populations (R2) was larger than the others, though not always different from R1. R1 was the first population to senesce regardless of habitat. Both agrestal GR populations produced more capitulae (seed heads) than the GS populations, and R1 also showed high reproductive success in the ruderal habitat. A test for the occurrence of GR C. canadensis in field margins of GR cropping systems (Experiment 3, Chapter 4) utilized a discriminating spray test on seed collected from the margins of 17 agricultural fields in 2008 from Illinois, Indiana, and Nebraska. The relationships between the occurrence of GR C. canadensis in 2008 field margins and 1) geography, 2) field management practices, 3) herbicide diversity, 4) weed community diversity, 5) field-interior changes in C. canadensis populations (λ), and 6) knowledge or suspicion of difficult-to-control C. canadensis populations, were tested. Geography was the clearest relationship in these data and was related to geography, with greater, more variable occurrence of GR C. canadensis in 2008 field margins in the southern and eastern sites than the northern and western sites. Management practices in field-interiors from year 2006 to 2008 had little relationship to occurrence of GR C. canadensis in 2008 field margins, except for the increased use of tillage and the increased use of non-glyphosate herbicide modes of action in 2007 field-interiors of fields with the highest occurrence of GR C. canadensis in the field margins in 2008. Additionally, the greatest occurrence of GR C. canadensis was associated with high diversity in the weed community in 2007 field-interiors. These results suggest that GR C. canadensis persisted in 2008 field margins even though there was a perceivable increase in management intensity the previous year. In conclusion, although there were no differences in fitness clearly associated with the GR trait in C. canadensis, GR populations may possess equal or greater vigor in growth, reproduction, and competition as GS populations. Based on these results, GR C. canadensis may persist in agricultural field-margins or other ruderal, unmanaged habitats and act as a seed source for future field-interior infestations. Depending on the characteristics of the GR biotype, GR C. canadensis may increase in frequency in the landscape.