Seed Structure and Dispersal in Acer
saccharum (Sugar Maple)
The purpose of this lab is to find out if variations in Acer saccharum (sugar maple) seed structure affect dispersal distance. Sipe and Linnerooth (1995) conduct an experiment involving samara morphology and flight behavior in silver maples. Their findings showed a wide variety in seed structure and behavior existing within the silver maple population. This experiment, in conjunction with the findings of Greene and Johnson (1991), relating mass and terminal velocity, formed the basis of our hypothesis. We believe that despite intraspecific variations in seed designs among different Acer saccharum subspecies, located on the lower branches of trees found in a partially open landscape, each seed will uniquely maximize its dispersal distance.
This lab idea has gone through many revisions. It has been narrowed down from general seed dispersal to concentrating on one prominent tree species in the area. We chose the maple tree for their accessible and easily identifiable seeds. The two joined seeds are commonly known as "helicopters" for their wings that propel them into a spiraling motion. Samara is the scientific term for the entire seed structure.
From reading the integrated experiments of Sipe and Linnerooth (1995), Greene and Johnson (1991) and (1992), Guries and Nordheim (1984), and Peroni (1994), we found supportive evidence for our experiment with the Acer saccharum. The topic is a pertinent one as the evolution of seed variation is still debated among scientists as Greene and Johnson (1992). The maple seed structure undoubtedly varies, but why? Did adaptive features evolve? For example, does the samara blade length decrease over time in order for the seed to enter autorotation sooner to maximize its dispersal distance? Or are these
variations hereditary? The objective of this experiment is to show that each seed design has characteristics to help it obtain optimal dispersal.
Our specific questions incorporate seed dispersal variables such as the structure and weight of the seed, the initial falling distance prior to autorotation, and the time and distance of flight. By collecting the seeds from the lower branches of the tree at various locations (utilize a random sampling strategy) and reenacting the moment of seed release, we hope to prove whether or not our hypothesis is correct and coincides with the studies of our predecessors.
Relevance of your research question:
While researching our topic we found a previous Miami University group study with a completed project similar to our own in addition to the studies found in scientific journals. This shows that this issue of maple seed structure and dispersal distance is a prominent topic, not only in the scientific community but also on our campus.
Through previous research already stated, it was also discovered that the maple population is diminishing. Studying the tree’s mechanism of reproduction (the seed) could provide clues as to why this is happening. The maple tree produces an abundance of the seeds to increase the chance of reproduction, but is this enough? Does the maple tree need to adapt its seeds according to its surroundings in order to maximize this chance? We maintain that it does. Through our experiment we hope to discover which structural designs are most advantageous for the reproductive dispersal of the lower branches of trees within a partial open landscape. The maple tree is an example of
how adaptive features in nature attempt to ensure reproductive success, therefore ensuring the survival of the species. In the article by Sipe and Linnerooth (1995), they say, "a significant role in population persistence, and that natural selection can work within a species to shape fruit morphologies that enhance dispersal potential." This implies that maple trees along with other species do in fact alter certain characteristics that prove advantageous to prolonging their species.
Materials and Methods:
To carry out this experiment we will need a fan, permanent marker, tape measure, protractor, ruler, 10-foot ladder, electronic scales, a stopwatch, and numerous samples of seeds representing four different sugar maple trees.
The first part of our experiment will be to observe the samara designs from each of the tested trees. The weight of the seeds, the inner and outer angles of the wings, the length of the seed, and wing shape, will be compared and contrasted to each other. The wing shape will be classified using the diagram, with various shapes from the article by Sipes and Linnerooth (1991). Determine the weight of the samaras to the .01mg using the electronic scale.
The second component of the experiment is a simulation of the seed dispersal due to wind. In preparation, the seeds will be collected and organized according to location. The seed will be suspended from a fixed point 2.4 meters above the ground. The fan will be placed on the ladder at this height as well. Turn the fan on to its medium setting (this will represent an average wind current which is a minimum of 1m/s and a maximum of 4m/s for abscission, according to Greene and Johnson’s 1991 study). The seed will be released and carried by the wind current. Measure the displacement of the dropped seed from the location directly below where the seed was dropped to where it landed. Also observe the distance the seed dropped before autorotation by using a vertical tape measure, attached to the wall. Record the total time from when the seed was released to the moment it touches the ground. Repeat twice for each seed, five seeds from each of the four trees; for a total of 40 trials.
We believe our design is significantly sound because we incorporate a field study* as well as a model. The two tests decrease data discrepancies (human error) by supporting one another. We researched the Internet, proposed our idea to our peers, and consulted faculty for further experimental design assistance. A large sampling pool will ensure unbiased results.
*Due to the adverse weather conditions we were unable to perform the field studies using the tarps. In place of the tarp collection of seeds we collected them ourselves from the lower branches of the trees. We believe, however, that our experiment is still statistically sound because we focused on low discrepancy characteristics (such as weight, angles, and shape).
Material use is as follows: electronic weights will be used to weigh each individual samara, ruler to measure length of wing, protractor to determine inner and outer angles, fan used for wind simulation, and tape-measures to measure distances of flight and distance of fall before autorotation.
Class participation is as follows: The class will be divided into two groups led by two of the experimenters. The class will be asked to observe and record the various components of seed structure: wing length, shape, weight, and the angles of the wings. They will classify the wing shape based on the drawings that they have been furnished with. Reference Appendix A.
Equipped with this newfound knowledge they can deduce which seed design is most advantageous for dispersal. They will next be asked to perform the simulated seed dispersal experiment. Each student will be given one seed and asked to drop it from the top of the ladder, while the rest of the students observe initial fall distance. This will be repeated twice.
A data sheet with three tables have been supplemented, one for each component of the experiment. The first sheet is for the simulated dispersal, the second for the observations, and the third for the averages. Reference Appendix B for raw data and Appendix C for compiled data.
Our experiment will require approximately a week and a half to carry out and may be repeated to increase research accuracy.
*September and October- Worked on research ideas, proposals and Complete
*Oct. 31- Gathered data from the class
*Nov. 1-18- Worked with group and posted small progress reports
*Nov. 19- Posted Thanksgiving break progress report
*Nov. 20-26- Visited respective libraries to scout out new material for our lab as well as observe local maple tree species
*Nov. 27-28- Collated new material
*Nov. 29-30- Analyzed all data and refine data tables
*Dec. 4-6- Reviewed lab
*Dec 7- Lab presentations and final lab packets turned in and posted to web
Results (Methods for Interpretation):
We obtained our empirical results by running StatView t-tests. The p-values derived from these tests indicated that the observed differences in means in the following relationships are significant and the results of our experiment are not products of chance:
q The differences in the outside angle amongst all the seeds types. P-values (.0005, .0013, .0007)
q The different lengths of seed wings amongst all the seed types. P-values (<.0001, .0046, <.0001)
The significant p-values concerning the differences in length encouraged us to take our StatView tests a step further. We performed regression plot summaries, which show the significance of the relationships between variables and not just amongst the seed types. We chose to concentrate on the relationships the seed’s distance from the ladder had with seed weight, length, inside angle, outside angle, total time of flight, and initial fall distance independent of one another. These statistically significant results were obtained:
q As the length of the seed increases, so does the distance of the seed’s flight from the ladder. P-value (.0455)
q As the total time increases, so does the distance of the seed’s flight from the ladder. P-value (.0193)
q The smaller the initial fall distance, the greater the distance the seed travels from the ladder. P-value (.0193)
All of the samara features are interdependent with one another. We were interested in seeing what happens when you look at them separately. We used the stepwise regression summary function of StatView to distinguish which samara feature (weight, length, inside angle, or outside angle) was the most determining factor in dispersal distance. All four variables were correlated with the distance traveled from the ladder. Step zero of the summary indicated that the length of the seed is the most determining feature of the four. The length’s partial correlation was the greatest (.318). When we removed length from the analysis in Step one, inside angle became the most determining with a partial correlation of (-. 324) and these results are significant according to the p-value (.0455). Step two removes inside angle and then weight becomes the most determining factor, with a partial correlation of (-. 320) verified by the p-value (.0079). The last feature left, the outside angle, would be the most determining had all the other features been removed.
The basic gist of these results is that length is the key feature that maximizes dispersal distance. This is why the Quillpen did so well in our study. Also, we are given the most ideal samara seed design from the Regression Summary statistics. In descending order of importance, the seed would be long in length, with a small inner angle, and light in weight. We conclude that these three structural features are the most advantageous in maple seed dispersal. If one feature is lacking, such as short seed length, the tree will equip the seed to be lighter in weight and with a smaller inner angle if it has the reproductive capabilities to do so. Our results show that the Quillpen design is the most advantageous for lower branches of a tree located in a partially open landscape. Quillpen may not, however, prove to be the most advantageous design in other landscapes or on higher branches of the tree. In this experiment, Quillpen had the furthest dispersal distance and therefore is the most advantageous design. Reference Appendix D for all t-tests, regression plot summaries, and bar graphs.
Discussion and Conclusion:
Within our sampling pool of samaras, three types of Acer Saccharinum were identified: Quillpen, Long Straight, and Boomerang. The samaras from tree one, located near Peabody Hall, were all classified as Long Straights, as were those from tree number three, located adjacent to McKee. The Quillpen design was found on the tree near Boyd, designated number four. Both Boomerang and Long Straight samara designs were found on tree two which was located in the regions of Ernst Theater.
Averages were compiled comparing the differences between the three seed designs, these included: seed length, inner and outer angle, weight, distance from the ladder, initial fall distance, and total time of flight. The following chart shows these averages:
Long Straight Boomerang Quillpen
Length (cm) 2.802 2.5 4.94
Angle 54.24°/ 53.88° 47°/ 71.05° 166.2°/ 43.8°
Weight (g) 0.1012 0.06 1.12
Distance from ladder (cm) 13.81 10.5 27.1
Initial fall distance (cm) 138.6 90 120
Total time (s) 1.968 1.95 1.85
Based upon these averages we were able to make a few general observations. Quillpen was found to have the greatest length (4.94 cm) of the three designs. This made a noticeable impact on the seed’s total flight time and distance. Having a greater surface area than both the Boomerang and Long Straight, the Quillpen was able to receive more wind energy, therefore propelling it to greater distances. Previously, we felt that the heavy weight of the Quillpen would hinder long distance flight. Surprisingly though, we discovered that the Quillpen stayed in the air longer than the other, lighter, seed designs.
Boomerang, weighing the least of the three structures as well as having the smallest surface area, proved to be the least successful in flight, traveling a distance of only 10.5 cm. The Long Straight design was slightly longer in length and heavier than Boomerang, therefore having slightly a better advantage in flight.
Flight distances in relation to the angle measurements were also unexpected. We had assumed that the design with a more extreme inside angle would fall into initial autorotation the earliest and travel the greatest distance. As expected, the boomerang, with the smallest inner curve (47°) fell into autorotation the earliest. However, it traveled the shortest distance of all three.
It would stand to reason from these averages, that the Quillpen is the most ideal design for samara dispersal specific to our tree and seed location, traveling the furthest distance. We are hesitant however, to discount the advantages of the other two designs. For example, the samaras collected from the Ernst tree had both Long Straight as well as Boomerang designs. Why would the tree bother producing such inferior designs if they weren’t as effective in reproduction? With this idea in mind, we performed additional tests with the Boomerang design. We decided to do this to increase data for the Boomerang design and to represent both inferior designs. In these later Boomerang tests, performed with the fan simulation of seed dispersal, some interesting new developments emerged. Here is a data table of our results:
Distance from ladder (cm) Initial fall distance Total time (s)
Tree #2 Trial 1 Trial 2 Trial 1 Trial 2 Trial 1 Trial 2
Seed 1 48 47 16 15 0.97 1.5
Seed 2 105 44.5 11 13 2.6 1.5
Seed 3 40.6 73 14 10 2.1 1.88
Tree #2 Weight Angle Length Shape
Seed 1 0.04 72/108 2.0 Boomerang
Seed 2 0.04 60/115 2.2 Boomerang
Seed 3 0.04 55/110 2.5 Boomerang
One must take into account the following conditions when interpreting these results: The tests were performed using seeds from the same sampling pool as the ones used five weeks earlier. This means that the seeds had been in a Ziploc baggy for five weeks, which could have affected i.e. the weight or other characteristics of the seed.
Even though we did not enter the later results into StatView, we now speculate that the inferior designs may have some advantageous characteristics that increase the dispersal distance for different landscapes (such as deep forest or urban settings), and/or higher branch location. These Boomerang tests may also represent results that could have been achieved if we performed additional tests to the Long Straight design. With so many variables that may affect seed dispersal it is impossible to consider all possibilities, therefore we only focus on a few basic design characteristics.
Our specific data and results do not in any way prove that the Boomerang and Long Straight design are advantageous in other landscape settings. We believe that they might be and therefore we have formulated these two theories:
A.) Seed designs vary dependant upon their placement within the tree.
The Boomerang may, in fact, be an inferior design for flight, but requires less energy for the tree to produce it, therefore, enabling the tree to produce additional seeds. For example, the lower region of the tree might produce the Boomerang design because the seed is not required to fly great distances. Whereas the Quillpen would be placed in the higher regions of the tree with the idea that the greater surface area would carry it further distances from the tree, overcoming the spread of the lower branches. The Boomerang would ensure dispersal directly below the tree’s canopy while the Quillpen would spread the seeds at greater distances from the parent tree, maximizing all possibilities for reproduction.
B.) Tree location affects the samara design.
The location of the tree may indeed also have bearing on the structure of the samara. For instance, within a high-density location, such as a forest, less wind will be generated. Maple trees, in turn, will not be required to produce seeds with designs that emphasize the use of wind in dispersal. With this conserved energy the tree is able to produce seeds with other traits desirable for that location. The Boomerang design in this case could be a viable solution.
Based upon the studies of our predecessors and our expected results, we predicted that there is no one samara structure that is advantageous over the others. Our conclusion contradicts that theory, showing that the Quillpen design is the most advantageous of the three designs: Boomerang, Long Straight and Quillpen. Our theories lead us to more discussion regarding how our project relates to our peers’ research as well as raises questions for further investigation.
All of the projects in our class related to landscape ecology. Another group focused on the tree architecture of sugar maples and how the branches are found to be at different angles in relation to their location on the tree. This study may in fact relate to our experiment with sugar maples as well. Perhaps the tree architecture and interior and exterior trees could influence certain seed designs. This is just one of the ways in which the two could labs could be integrated. Further research and investigation may produce significant results relating the seed design and the location and angle of the tree branches.
Another question raised from the lab is how exactly do trees use their resources, such as carbon, to make seeds. Do they use more for each seed and produce a smaller amount of them? Or do they make numerous seeds and only allot a small amount of carbon per seed? This is something that could be investigated with the use of chemicals and other tests to find out the chemical make-up of the seeds.
From this lab, we, as a group, feel that we have gained a great deal of knowledge regarding maple seeds and their dispersal. If possible, further research could be done to enhance our understanding of the samaras!! Leah :) Emily :) Wendi :) Katie :)
Maple Seed Production and Dissemination.”
2.http://www.muohio.edu/dragonfly/itb/maple.htmlx. “Maple Tree Seed Dispersal and
3. Annual Review of Ecology and Systematic. “A Day in the Life of a Seed: Movements
and Fates of Seeds and Their Implications for Natural and Managed Systems.” Vol. 25 (1994) pp.263-292.
4. Snipe, Timothy W. and Amy R. Linnerooth. American Journal of Botany.
“Intraspecific Variation in Samara Morphology and Flight Behavior in Acer Saccharinum. Baltimore.” Vol.82 (1995) pp.1412-1419.
5. Peroni, Patricia A. American Journal of Botany. “Seed Size and Dispersal Potential of Acer Rubrum (Aceraceae) Samaras Produced by Populations in Early and Late Successional Environments.” Durham. Vol. 81 (1994) pp. 1428-1434.
6. Greene, D. F. and E. A. Johnson. “Fruit absicission in Acer saccharinum with
Reference to Seed Dispersal.” Montreal. 1992.
7. The American Nationalist. “Can the Variation in Samara Mass and Terminal
Velocity on an Individual Plant Affect the Distribution of Dispersal Distances.” Vol.139, No. 4. Calgary.
8. Guries, Raymond P. and Erik V. Nordheim. “Flight Characteristics and Dispersal
Potential of Maple Samaras.” Vol. 30, No. 2 (1984).
Return to the Topic Menu
IMPORTANT: For each Response, make sure the title of the response is different than previous titles shown above!
Weather & Earth Science Resources
Tropical Ecosystem Courses
Tools & Other Stuff