Date of Award

May 2016

Degree Type


Degree Name

Master of Science


Biological Sciences

First Advisor

Gretchen A. Meyer

Committee Members

Jeffrey D. Karron, James A. Reinartz


Flowering plants must invest energy and resources to produce floral displays that are attractive to pollinators, but these same displays may also attract detrimental insects. How floral traits are shaped by the preferences of both pollinators and herbivores/seed predators is not fully understood. Using Helianthus grosseserratus (sawtooth sunflower) as my study species, I investigated these conflicting selective pressures on floral head traits through a 2-year study in a large, unbroken tract of mesic prairie in Wisconsin. In the first season, I followed individual heads over time and recorded insect visitation patterns and phenological changes to floral head traits. I also dissected seed heads at the end of the flowering period and identified all seed predators to order. In the second year, I measured floral head traits (including disc area, ray area, and UV reflectance patterns) on the day when most florets were presenting pollen. I also performed a hand-pollination experiment to determine if the plants were pollen-limited or resource-limited. I recorded the number and percent developed seeds per head as measures of reproductive success and also counted and identified the seed predators in each head. I also measured the number of flowers surrounding the study head as an additional factor that may affect pollinator and herbivore/seed predator preference. Floral heads were visited by a diverse group of insects: 16 species from 7 orders were recorded. Hymenoptera, Coleoptera, and Diptera were the most common visitors. These 3 orders had highest visitation on the second or third day of pollen presentation. Seed head dissection revealed 6 orders of insect, with Thysanoptera and Diptera being the most common. In year 2, I found that pollinators were required for seed set in this system, as heads that were bagged produced negligible seed. Heads in the hand-pollination treatment had fewer developed seeds and a lower percentage of developed seeds than heads that were open-pollinated, although these differences were not significant. These results suggest that the plants were more likely to be resource-limited than pollen-limited. However, hand-pollinated heads did have significantly more seed predators than open-pollinated heads, which likely reduced seed set. Disc area was the most important trait affecting both the number of developed seeds and the number of seed predators, with larger discs having both greater seed production and more seed predators. Disc area did not influence the percentage of developed seeds, suggesting that the effects on seed number reflect the fact that a larger head has more ovules rather than pollinator attraction. The UV patterning on study heads showed significant polymorphism, where some plants had a strong bulls-eye pattern on rays, while others had no clear demarcation (50% of heads in 2013 had no demarcation; 44% in 2014). My results showed there was no relationship between this patterning and number or percentage of developed seeds, but plants with a stronger bulls-eye pattern (likely because of a reduced amount of UV-absorbing defensive pigments) had more seed predators. These results suggest that UV patterning was important for defense against seed predators. In addition, floral heads with a large ray area had fewer seed insects, while those with a short head height and a large number of flowers in surrounding area had higher number and percentage of developed seeds. Such results highlight the complexities involved in the generalist pollination syndrome and the need to consider a multitude of floral head traits when analyzing plant/insect interactions.