Grass Height Effects on Insect Pollination Relating to Increased Fruit Production in Daucus carota
INTRODUCTION:
The wild carrot species, Daucus carota, or Queen Anne’s Lace, is an annual or biennial carrot species that is distinguished by umbels that contain thousands of small, white, bisexual florets (Westmoreland et al. 1996). At anthesis, a typical plant produces 5 to 10 umbels (Wilkinson et al. 1991). This species originally came from Europe, but today is quite pervasive in the middle to eastern United States. One finds this species in patches along roadsides and in open fields. Daucus carota is an adaptable species that is able to grow in poor soil, but is most successful in loose, fertile, and well drained sites such as the tall grass prairies of Illinois. The second year growth habit of Daucus carota reaches heights of 1 meter under optimal growing conditions.
The ecological pattern studied for this research proposal considers insect visitation and fruit production in Daucus carota as it relates to surrounding heights of prairie grass species that may affect insect pollination. The number of insect visits depends on the abundance of pollen and nectar found on individuals of Daucus carota.
Illinois lies within the prairie peninsula, which constitutes part of the tallgrass prairie region. Tallgrass prairies are dominated by species, such as Andropogon gerardii (Big bluestem) and Sorghastrum nutans (Indian grass). Such grasses sometimes reach heights of 3 meters or more. Daucus carota is host to numerous insect pollinators such as flies (Drosophilidae), ants (Formicidae), beetles (Coleoptera), and bees (Apidae) (Wilkinson et al. 1991). Given that Daucus carota has a relatively long, flowering period and an abundance of nectar and pollen on its mature inflorescences, all of these insect families are successful plant pollinators for Daucus carota (Elliott et al. 1991). Insects usually become dusted with pollen as they feed from nectaries among individuals of Daucus carota, and thus, the insects act as plant pollinators (Pollard et al. 1995).
AIMS OF THE PROPOSED STUDY:
The purpose of this proposed study is to determine whether individuals of Daucus carota surrounded by nearby grasses of lower height receive more visits from insect pollinators than those individuals of Daucus carota surrounded by taller grasses. Thus, individuals of Daucus carota surrounded by shorter grasses should have increased fruit production, due to more visits by insects. The study will consist of counting the number of insect visits made to the small, white bisexual florets of the umbels of Daucus carota and measuring subsequent fruit productivity in these plants.
MATERIALS AND METHODS:
This experiment will be conducted in a tallgrass prairie in Illinois. First, a rectangular grid of 1 m by 60 m will be mapped out within in a prairie of tall grass. Then a quadrat of 1 m by 1 m each, with each quadrat containing an area of 1 m2, will be considered the experimental unit for this study. Ten quadrats will be placed in alternating positions along the 1m by 60m rectangular grid with 5 meters of untouched prairie grass separating the test sites. The separation is necessary because plant density greatly influences growth and fruit yield within Daucus carota, and wider spacing avoids spillover effects from different plots (Reid et al. 2000). Each quadrat will contain two to three individuals (grown in clumps or bunches) of Daucus carota, with heights of 1m each, surrounded by grass species of Andropogon gerardii (Big bluestem) and Sorghastrum nutans (Indian grass) of varying heights. The approximate number of Daucus carota individuals will be maintained by either digging up excess individuals or planting extra individuals in each plot. The grass in the first quadrat will be cut to the base of the individual of Daucus carota. The other nine quadrats will contain Daucus carota individuals of the same height of 1m, but the surrounding grass height will vary according to the specific quadrat under study. The second quadrat will contain grasses that are 0.33 meters tall. Each ensuing, alternate quadrat will have grass height that is 0.33 meters taller than the previous quadrat. In the last quadrat, the final grass height will be 3 meters. The location of each quadrat will be chosen randomly, and the plots will be placed across a linear block of 1 meter by 60 meters. If grasses grow above their predetermined heights, they will be cut and the clippings will be removed from the plot. The intervening quadrats, between the ten quadrats being studied, will not be altered, and no counts will be taken from them. Their purpose is to maintain a minimal separation between test quadrats in order to demarcate a clear delineation among grass heights.
FIGURE 1: Ten quadrats will be used to test whether there is a correlation between surrounding prairie grass height and Insect visitation to Daucus carota. Each quadrat is 1 m by 1 m with an individual area of 1 m2. The actual quadrats will be set up in a 1 m by 60 m grid separated into twenty 1 m by 1m quadrats. Each quadrat will contain two to three individuals of Daucus carota that are 1m tall. Each quadrat will be separated by a distance of 5 m.In addition, the grass species Andropogon gerardii (Big bluestem) and Sorghastrum nutans (Indian grass) will be in the ten quadrats, with each quadrat having a grass of varying height. Quadrat 1 contains grasses that have been cut to the base of Daucus carota of about 5 cm. Quadrat 2 will contain grass heights of 0.33 m. Each ensuing quadrat will contain grass heights increasing in height by 0.33 m from the previous quadrat. Thus, the last quadrat will 3m. The location of the quadrats are chosen at random along the prairie. The number in the quadrat refers to a specific plot with a particular grass height as shown on chart 1. Note, the length of the gray, intervening quadrats, although not to scale, represent rectangles with dimensions of 1m by 5m.
Picking a region to set up quadrats requires consideration of soil moisture content and nutrient availability. A region with a downward incline will insure proper drainage for the ten quadrats. Also, soil nutrients would be assumed to be quite similar between the two sites if they were in relatively close proximity to each other. Essentially, I will use a prairie of uniform appearance and topography to carry out the experiment.
The number of individuals of Daucus carota in each of the ten quadrats will be maintained at a constant two to three clumps in each plot. Then, the number of visits by different insect families to Daucus carota flower heads will be recorded for each quadrat. I will set up a timetable for the observation of the number of visits made by different insect families on the umbels of the flower head. The observations will occur for one month at the height of flowering, which occurs during the summer months of June through September. The ten quadrats will be observed at three different times of the day to ensure temporal replication of experimental units over time. These ten quadrats will be observed for five-minute intervals in succession at 10:00a.m. for evidence of insect visitations. The counts of insects will be taken at the family level. The ten quadrats will be observed again at 3:00p.m. and then at 7:00p.m., with insect visitation being observed in the same manner as in the morning. This procedure will be carried out for 30 days, with the order in which different plots are observed being alternated at random each day. Average insect visitations found by calculating the mean number of insect insect visits over the course of one day will be collected for each of the ten quadrats, so there should be 10 means collected every day. When completed, the experiment should yield 300 mean values of insect visitation, with 30 means per quadrat.
I will measure fruit productivity in Daucus carota by counting the number of fruits greater than 3 mm long (Westmoreland et al. 1996). Thus, if plots with lower grass height received more insect visits, there should be increased fruit production found on individuals of Daucus carota.
A statistical test examining the data will be carried out on SPSS. If the data are homogeneous, a parametric test such as the Pearson correlation analysis will be carried out to test whether two continuous variables, grass height and mean number of insect visitations, are associated in a systematic way. A Pearson correlation analysis will also be carried out to test whether there is a systematic relationship between grass height and fruit production. Another Pearson correlation analysis will be carried out to test whether a systematic relationship exists between insect visits and fruit production. A scatterplot of the data will be constructed to give a visual representation of a correlation between the two continuous variables. For each of the three cases, if the two continuous variables are associated in a systematic way, a regression analysis will be carried out on the data, where, for example, grass height is the dependent variable and the number of insect visits (at the family level) is the independent variable. If the data are not homogeneous, a nonparametric test such as the Spearman rank correlation or Kendall’s tau-b correlation will be carried out using, for example, the continuous variables of grass height and mean insect visitations
Literature Cited
Elliott, N.B., and W.M. Elliott. 1991. Effect of an Ambush Predator, Phymata americana Melin, on Behavior of Insects Visiting Daucus carota. 126: 198-202..
Pollard, S.D., M.W. Beck, and G.N. Dodson. 1995. Why do male crab spiders drink nectar. 49: 1443-1448.
Reid, J.B., and J.M. English. 2000. Potential Yield in Carrots (Daucus carota L.): Theory, Test and an Application. 85: 593-605.
Westmoreland, D., and C. Mutan. 1996. The Influence of Dark Central Florets on Insect Attraction and Fruit Production in Queen Anne’s Lace (Daucus carota L.). American Midland Naturalist. 135(1): 122-129.
Wilkinson, K., and D. Westmoreland. 1991. Effects of Spider Predation on Insect Visitation and Pollination of Queen Anne’s Lace. American Midland Naturalist. 125(2): 364-367.