Petal Attraction

Petal Attraction

Abstract

In the past few weeks we have been developing an accurate experiment dealing with the alluring characteristics of flowers to their different pollinators. We have created a process in which we categorize the characteristics of flowers into several distinct groups. These groups include diameter of bloom, height, scent, color, number of times visited, and type of pollinator attracted. By studying the collected data, we will attempt to answer the question of what aspects of a flower will attract pollinators. Because flowers will not reproduce without the aid of pollinators, it is important for us to understand pollinators in order to assist to the vitality of natural flower populations. We are currently observing flowers in their natural environment to investigate this process. Our observations thus far have led us to believe that color and odor have the most influence on the number of pollinators. Because of the cool temperatures during the morning and evening, we have found the best time frame for observing flowers is during the afternoon. This results in a more active process of pollination. So far our group has not collected enough data to accurately establish the type of scent, sweet or sour, the pollinators positively react to.

I. Introduction

The purpose of our lab is to determine the most appealing characteristics in flowers to both land and air pollinators. We hypothesize that tall flowers containing bright colors will attract the largest amount of these pollinators. However, we believe that odor will play a significant role to nocturnal pollination. By studying the flowers we will categorize pollinators and their floral interest to determine which species will last longer due to excessive pollination. Taking into consideration the increased rate of extinction among certain species of flowers, we hope to be able to recreate the ideal pollination characteristics of the dying species in order to prolong its life span. We think that it will be interesting to witness firsthand how flowers attract certain kinds of pollinators.

II. Relevance of Question.

Much work has been done this decade on the subject of pollination. The likely explanation is the increasing emphasis on sustaining agriculture on smaller areas of land for larger numbers of population under increasingly challenging conditions, with respect to climate and disasters - either natural or man-made.
A review of five articles provided background to prepare for our own study of this issue, that we might formulate an experimental approach to contribute to the investigation. These articles are quite diverse in their emphasis and in their perspective. Together, however, they provide a clear articulation of several relevant considerations and analysis of research regarding the mechanisms of pollination.
The first of these articles, by Lawrence D. Harder (1), examines the effects of pollen size on pollination of angiosperms (seed-producing, flowering plants) by bees (pollen-collecting or nectar-collecting) and birds. Harder uses statistical methodology to scrutinize the experimental data of many other researchers in this arena, looking at two studies, one of nine species of a particular genera of angiosperms and another using 16 genera (nine families) whose habitats are Australia and North America. From this data he contrasted plants offering an award (the pollinator gets nectar for its efforts) with buzz-pollinated varieties, examining the grooming characteristics of bees and birds with respect to their influence on the evolution of pollen size during pollen transport activity. Harder concluded that the "differences between bee foraging behavior and between bee- and bird-pollination seem not to affect the evolution of pollen size consistently."
Fernando Vega-Redondo (2) took another approach. He analyzed the effects of reward (noted above) in terms of game theory. Vega-Redondo argued that there exist two monomorphic states (where all flowers of a population either provide a reward or they do not) that are evolutionarily stable (the population thrives in their current adaptation). In the case of a relative scarcity of pollinators and one unique monomorphic and evolutionarily stable state when the plants have a relative scarcity with respect to pollinators. His analysis requires the positing of certain assumptions based on the notion that reward (giving up nectar to the pollinator) is costly to the plant (reduces the fertility of the plant) relative to non-rewarding varieties. Vega-Redondo's study particularly looked at orchids, a family of flowers containing 80,000 species, each species of which is monomorphic with respect to reward. His conclusion notes that the single evolutionarily stable state (when pollinators are in abundance and plants in relative scarcity) is that of flowers of both reward categories coexisting, with the non-rewarding type present in smaller frequency. The relative scarcity of pollinators with respect to plants may produce either of two evolutionarily stable states, one monomorphically rewarding and the other monomorphically non-rewarding.
For SWT Batra (3), evolutionary considerations are not evaluated or alluded to. This group simply reviews "the early literature demonstrating the adequacy of pollination by local bee populations before intensive and extensive agricultural practices. Their article is far less imposing in its use of jargon and its intensity of presentation. They consider the economic value of using particular pollinators, especially the European honey bee, as a one-size fits all solution to maintaining ample pollination across the spectrum of agriculture. SWT Batra gives several examples, for instance, where the honey bee is demonstrably unsuited to perform the job adequately, whereas local bee populations, due to different pollinating abilities are more effective (i.e. blueberries, cranberries, and hothouse tomatoes require buzz-pollination, as opposed to nectar-feeding honey bees). He cites size, hairiness, quickness, fidelity, longevity, learning ability, flight range, cold tolerance, season, flower handling ability and other unnamed characteristics as contributing to the particular utility of certain bee populations over a generalized population of honey bees. To provide for a large human population, which requires extensive agriculture, SWT Batra concludes that the effective biodiversity necessary to improve the yields on existing croplands demands the introduction of the most efficient pollinators for any given crop.
Returning to an evolutionary context, John N. Owens, Tokushiro Takaso and C. John Runions (4) discuss the wind-pollination mechanism in conifers (cone-bearing trees) and the (natural) evolution of the process to improve pollination success. These men posit that the diversity in pollen, megastrobili (seed cones) and pollination mechanisms are the result of repeated climactic changes, restricting and isolating species within the taxonomy of conifers (a relatively small taxon - 550 species in 53 genera) for long periods of time. They examine the mechanisms in great detail, suggesting an evolutionary model to explain the current dispersion of those mechanisms in particular habitats. Their conclusion is that five major types of pollination mechanisms have evolved in this group of plants, varying in structure and function, but all achieving the same result - the capture and transport of pollen to the ovule (the female gametophyte of a plant). They further suggest co-evolution of pollen and ovules so that non-saccate (having no wings or sacci) pollen occurs in species that have erect ovules (to receive them more effectively), and saccate pollen occurs in species with inverted ovules (likewise suitably designed for receptiveness).
The final article, written by S. M. Carthew and R. L. Goldinggay, also places its thrust in an evolutionary context, but specifically deals with non-flying mammals as pollinators. They note that of the 40 published studies on this topic, nearly half have been conducted since 1990. As in other studies, their own work centers on three issues: regular and non-destructive visits to flowers by the animals under observation; substantial pickup of pollen and its subsequent transport to flowers of conspecific (of the same species) type; and evidence that pollination of flowers is successful, producing seed. One aspect of their study that applies directly to our own research are the morphological (flower size, shape and color) and physiological (timing of flower opening, pollen presentation and nectar secretion) floral traits as adaptations to pollination by non-flying mammals. Carthew and Goldingay concluded that the evidence of their own and other studies leans very strongly in favor of the attraction of nectar and/or pollen as the determining factor for pollination by non-flying mammals.

III. Materials and Methods

With this research providing background information, we plan to study a variety of flowers in predetermined areas on our campus. We will observe flowers during the day, as well as at night time. Each member of our group will visit their chosen location at seven different times during the course of three weeks, observing the flower for twenty-five minute intervals. Our group will categorize flowers by height, shade of color, scent, and diameter of bloom. Collecting data during several intervals and taking into account five physical aspects of the flower, our group will be able to produce statistically sound results. To collect this data, we will rate the scent of the flower in one of three categories: sweet, none or sour. We will need metric rulers to measure the diameter of the bloom and the height of the flower, cameras to document the flowers that we study, and if available-night vision goggles. We will use the class to derive further develop our own questions about the subject. Using our pictures of flowers, we will propose questions to the class about the specific flowers and the likelihood of their popularity among pollinators. During the week of 10-10, we will begin our three week long study of daily observations. Each day will be split up into three time intervals including morning, afternoon and evening. The morning interval begins at seven am and ends at ten am, the afternoon period will include the ten am through five pm time frame, and the evening observation will cover five pm till seven pm. Group members can choose any time within the specified intervals to do their twenty-five minute observation. The week of 10-17, we will continue observations of flowers to add to our original data. Beginning the week of 10-24, we will start to bring together our analysis of the obtained data from the past two weeks. For the next two weeks we will be working on our lab packet and preparing our presentation for the class. To prepare our final project we will need to analyze our data using statview and Nova. We will finalize our report in the last week of our project and prepare for our final formal presentation by creating a power point slide show.

IV. Results

As our research continued through the semester, our findings gradually changed. In our hypothesis, we guessed that tall, bright, strongly scented flowers with large blooms would be most attractive. At the midpoint of our study we made a more educated assumption that short, yellow flowers with a strong scent and small bloom would attract the greatest number of pollinators. This was determined solely by the observations we made during our experiment and not on any statistically sound evidence. After standardizing all of our data and then entering it into statview, we were able to run different tests and create various graphs of the results. We found that the best way for us to show our findings was through the bar chart method. By creating a dependent and then splitting it by two different variables we were able to discover interesting things about our data. We were mainly concerned about finding which characteristics among the flowers attracted the most pollinators. So based on that, we used number of pollinators as our dependent and then split it by several variables, such as diameter and shade or scent and height. This enabled us to look at how variables effect each other as well as the dependent. We were also able to specify our dependent into either flying or non-flying pollinators. This gave us some more interesting data to consider in our conclusion.
By first creating a table comparing the total number of pollinators to the shade of flower, we discovered that the white flowers had the most visits. After doing this we did the sum of squares and found that the P-value was .0138 and that we should reject the null hypothesis, that there is a significant difference between the results.
In the beginning of our study, we thought that it was important to observe flowers at a number of different locations. After completing our study, we have found that there was no significant difference between the different places that we watched. The pollinators were not drawn specifically to any one location.
We also found that the number of pollinators drawn to a flower was not dependent upon what time of day it was. However, there were more non-flying pollinators counted in the morning and more flying pollinators in the afternoon. We hypothesize that this was because of the temperature difference. When totaled, the daily sum was the same though.
There were some interesting discoveries pertaining to the combination of height and scent. According to our data, sour flowers with tall and short heights attracted the largest amount of pollinators for their height. The medium flowers that contained sweet scents seemed to attract pollinators for their height range, though.
While we could not narrow down a single colored flower to be the most attractive, the white flowers seemed to be the most popular among the pollinators. Yellow was the second most popular. A contrasting piece of data was that although the white flowers were most popular, they had the smallest blooms, while the yellow flowers had the largest.
The white flowers had no non-flying pollinators visit it according to the observations that we made. The white flowers attracted the most flying pollinators out of any of the observed flowers. There could be error in this data though, because we observed more white flowers than any other color. This could have skewed our results.
Our data shows that a large amount of flying pollinators visited tall flowers and non-flying pollinators visited short flowers. This data makes sense because non-flying pollinators would not travel to tall flowers, and flying pollinators would stay nearer to their flying level. Our observations proved to be feasible.
After collecting all of our data and putting it into the computer to by analyzed, we came up with a number of graphs. Many of the graphs were irrelevant, so we had to search through them to find the ones that held importance to our study. The most important graphs are shown previously.

V. Conclusion
After analyzing all of the graphs, we concluded that there actually was no single set of characteristics that was most attractive. In the beginning we had hypothesized that tall flowers with bright colors and strong odors would be the ideal flowers. In reality, there are many factors that make a flower attractive. Different types of pollinators are attracted to different colors and scents. Flying pollinators and non-flying pollinators will be drawn to flowers of different heights. There is no ideal combination of attributes to make a flower attractive to pollinators.
Because of imperfections in our procedure certain errors arouse during the lab process. The major source of error in our data was due to the lack of regulation in color observation. We unknowingly observed a considerably larger amount of white flowers than any other color. This may have skewed our results and made it appear that white flowers were the most attractive. In order to limit the experimental variables we did not take atmospheric conditions into consideration while collecting data. These outdoor conditions may have an impact on pollination habits. Another source of error is the variation in each memberís sense of smell. Some members were able to smell the flowers quite strongly, while others could not. In all experimental studies we could have gathered more data to make our research more substantial. This is our final glitch in our lab structure.

Bibliography

*Harder LD 1998. "Pollen-size comparisons among animal-pollinated angiosperms with different pollination characteristics." Biological Journal of the Linnean Society. 64: 513-525.

*Vega-Redondo F 1996. "Pollination and reward: a game-theoretic approach." Games and Economic Behavior. 12: 127-142.

*WT Batra 1995. "Bees and Pollination in our changing Environment." Apidologie. 26: 361-370.

*Owens JN, Takaso T, Runions CJ 1999. "Pollination in Conifers" Nature. 36: 15-24.

Next Article
Previous Article
Return to the Topic Menu

Here is a list of responses that have been posted to this Study...

Respond to this Submission!

IMPORTANT: For each Response, make sure the title of the response is different than previous titles shown above!

Visit the rest of the site!