Abstract:

The topic of soil seed banks is important because of the impact seed reserves in soil have on current and future vegetation. Seed banks can also reflect evolutionary changes in plant communities as a result of changes in land use. Diversity can be found in different seed banks based on resource availibility and surrounding conditions. The addition of fertilizer, increasing the amount of nitrogen in the soil can lead to a decrease in species richness. This experiment investigates the differences in soil seed banks in two areas of the domain of the University of the South on the Cumberland plateau. One of the seed banks is located in an area used for agricultural cultivation until forty years ago. The other seed bank is located in a historically forested area. Soil samples were collected using a soil coring device and planted on a base of potting soil. These samples were cultivated under greenhouse conditions for three weeks and then exposed to freezing conditions in order to obtain the seed richness and evenness through germination. No seeds germinated to form seedlings and no plant growth or emergence was observed. These results were unexpected and did not support our hypothesis. There were many variables, such as dormancy requirements, amount of sunlight and water, and the depth of seed banks that could have contributed to the lack of results. This experiment could be replicated with attention to these factors in order to obtain more definitive conclusions.

Introduction:

Seed banks are essential to maintaining life and growth in forests. Defined as a "buildup of viable but ungerminated seeds in or on the soil," seed banks exist because of natural selection for plant species that can withstand harsh conditions and germinate in optimal ones (Hyatt, 1999). The presence of seed banks in soil allows a plant species to maximize its chance for survival, creating benefits for that population. Seeds stored in the seed bank can withstand harsh conditions over many years allowing the plant species to be propogated many years after initial seed dispersal. The Seed banks can create evolutionary changes in vegetation as nature selects for or against seeds with different phenotypic characteristics . Many seed properties, such as the dimensions, the shape, and productivity of the seed affect the seed's ease in germinating (Falinska, 1999). Hyatt and Casper (2000) also cite studies which conclude that seed banks are useful for promoting competition between species because different seed species are triggered to germinate by different environmental factors, such as rainfall amounts or sunlight intensity. Seed banks are also helpful to analyze whether changes in vegetation result from the available seeds in the bank (Falinska, 1999). Their ultimate utility, however, comes from maintaining the species that grow in a particular area, especially during wide-scale disturbances where much of the existing vegetation is destroyed (Leckie, 2000). In this study, we will compare seed banks from two different sites in a temperate deciduous forest in Sewanee, TN, predominated by tulip poplar, white oak, and red maple. The first site is a former agricultural area that has developed into woodland; the second site has never been used for agricultural purposes and has remained forest. We will ascertain if there are significant differences in seed bank richness and evenness in the two areas.

Leckie, et al.,(2000) found that in temperate deciduous forests, the seeds of woody species are prevalent in low-sunlight areas with thicker leaf litter and more nutrient-rich soil. They also found that a significant portion of the seed bank contained shade-tolerant species, a finding opposite that of other studies that saw mostly seeds of shade-intolerant species. Other abiotic factors can also have a noticeable effect on the makeup of seed banks (Hyatt and Casper, 2000). Some of these factors affecting seed germination are temperature changes and rainfall amounts (Hyatt, 1999). Remaining fertilizers in the old agricultural tract may also affect richness and seed abundance in the bank. Falinska (1999) found that a post-agricultural land tract showed a decrease in richness over five years. Another possible theory states that the introduction of nitrogen shifts soil chemical composition to an extreme in which less species survive, resulting in lower seed bank diversity. The fertilizers create more uniform soil conditions, decreasing biotic heterogeneity (Haskell, 2000). Lopez-Marino, et al., (2000) supported this theory with results from an experiment in which there was significantly more richness in meadows that had not received organic fertilizers compared to fields that had received fertilizers. Kirkham and Kent (1997) also saw an overall lower density of seeds in fertilized land tracts than unfertilized areas. Bekker, et al., (2000) and Hyatt and Casper (2000), though, found that a restoration of diversity is possible in land once used for farming purposes, mainly depending on the inflowing of seeds from areas near the agricultural site. Some seeds that aided in diversity restoration, however, had been present in the seed bank before the farming period (Bekker, et al., 2000).

We tested the hypothesis that the post-agricultural land of a temperate deciduous forest shows less diversity and richness than the non-agricultural tract by analyzing fifteen soil core samples at each site. We spread the soil samples over potting soil in plastic trays in a greenhouse and watered them daily. Seedlings that sprouted from each soil sample were catalogued so that richness and diversity could be compared at the two sites.

Materials and Methods:

Our experiment was performed on the campus of the University of the South in Sewanee, TN. In the experiment we collected soil from two sites and cultivated them in a controlled environment. We picked two sample sites on the University's domain. Both sites were located off Brakefield Road. The first was located off the right (north) side of the road, behind a recent cut, in an area that was used for agriculture until forty years ago. The second was located one mile further down Brakefield Road on the left (south) side, in an area that has not been used for agriculture. We created transects across each site. Each transect was 100 paces long. We obtained fifteen samples from each arbitrarily-chosen transect on Tuesday, October 2, 2001, from 2-4 PM. A random numbers table was used to determine the location of these samples. Once fifteen random numbers between one and 100 were found we cored the soil at those pace points using a coring device. We brushed away the leaves resting on the ground and dug a three-inch deep soil core. Each soil sample was dumped into a labeled plastic bag. The transect point and site were recorded. These soil bags were then returned to the lab and planted immediately after collection.

We placed potting soil two inches deep in trays. Each tray was then divided in half. Each soil sample was emptied and spread onto half of a tray. These areas were labeled agricultural or forest sites. The trays were spread out in the university greenhouse where they would have access to normal light and dark cycles. All of the soil was watered each day for three weeks. After this period the soil trays were refrigerated for one week. During this week each soil tray was placed in a freezer for a sixteen-hour period. After the week of refrigeration the soil was replaced in the greenhouse. The trays were watered daily for one more week.

To calculate the seed bank abundance and diversity we would have counted the number of species of plants that grew in each soil sample. The mean number of plants and standard deviation would then be found for both the forested and agricultural area. These means would be compared using a t-test to test for a significant difference. The diversity of plants would be quantified using the number of different species that grew from each soil sample. This variety of species would be compared between sites using a t-test comparison for statistically significant differences. A statictically significant p-value, calculated from t-test would indicate that there is a difference in number of species from one site to another. A calculation of evenness would show the relative abundance of each species. A mean and standard deviation of the number of individuals of each species would be calculated. These means could be compared using t-test to determine any statistically significant differences in abundance of plants.

Results:

During five-week observation period, no seeds germinated. We saw no emergence of any plant species; therefore, there is no data to report.

Discussion:

We hypothesized that the seed bank in an area of the domain formerly used for agricultural would be less diverse than the seed bank of a forested area. This question was investigated to examine the results of agricultural practices on the soil seed banks over time. Agricultural practices such as repeated soil depletion and fertilization would favor species that are competitively superior in certain conditions, such as the presence of nitrogen. These species might be numerous. If this was the case a lower species diversity would be present in the seed bank. The results in this experiment do not support the hypothesis. The lack of results was consistent for the soil samples from both sample sites. According to this data the germinating seed banks were equal in a forested area and a formerly cultivated area. At both sites a three inch soil core produced soil samples that did not have any seedling growth over a five week germination period. This should not be the case in an area that does indeed have extensive plant growth and is a sustainable system because seed banks indicate the viable seeds in the soil. Furthermore, following the use of land for agricultural purposes a directional change in which the floristic richness of the seed bank declines is expected (Falinska, 1999).

There are several variables that have affected this experiment and may have contributed to the lack of visible results. The first is the amount of leaf litter and soil collected in the sample. In the method of collection the top layer of dead leaves and sticks were brushed away before the soil core was collected. Many of the seeds not deep in the soil may have been brushed away, resulting in a much lower yield than expected.

A second factor influencing the experiment is greenhouse conditions. The soil trays were grown in a green house exposed to light and dark cycles. However because the seeds were collected in the fall they may have entered dormancy. For the first four weeks of the experiment the seeds were kept at temperatures between 15.6 and 26.7 degrees Celsius. After no germination the seeds were exposed to freezing conditions in order to break a possible dormant stage. Because of lack of space the seeds were only frozen for sixteen hours. This period might not have been long enough to induce germination in these seeds. This factor would also cause variability in the seeds because of the artificial selection. Seed species that require longer periods of freezing in the wild might not appear in the experimental germination, skewing results toward those that require a shorter cold period. A way to modify the experiment to compensate for this factor would be to collect soil samples in the spring when the seed bank had emerged from dormancy. Performing this experiment in the springtime would also allow one to test seed germination in the field as well as the greenhouse. The greenhouse conditions might have contributed to the absence of seed germination. The regulated temperature and chlorinated water are different conditions and resources than are found outside. However, this should not have greatly affected the plants a great amount because other plants in the same greenhouse grew normally. Plants are routinely cultivated using chlorinated water.

An overarching problem that might have caused the lack of results is the duration of the experiment; because of time constraints the germination period over which the seeds were observed was only five weeks. This period of observation may have been too short to allow full germination (Falinska, 1999). All of these considerations are important in their impacts on developing seeds. Seed bank makeup is determined by imput from many sources. These factors include fecundity, seed germination ability, longevity, migration, mode of seed dispersal, life history, and size and shape of seeds. The variation of seeds leads to variation in conditional and resource needs. "Optimal germination conditions will differ from species to species, corresponding to the dominant limiting resource in the environment and from seed to seed, depending on age and condition" (Hyatt, 1999).

This experiment is a good example of the variability of science. It cannot be anticipated what data will be obtained through experimentation. Our experiment was limited by the amount of samples that could be taken and observed in a short period of time with limited space. The small scale of the experiment prompted us to control variation by surveying only two sites. No results were found in this experiment, making it difficult to draw conclusions, however, some conditions that might have influenced the seed germination and growth process were discovered and could be controlled for in future experiments.

Literature Cited:

Bekker, R.M.; Verweij, G.L.; Bakker, J.P.; Fresco, L.F.M. 2000. Soil seed bank dynamics in hayfield succession. Journal of Ecology 88: 594-607.

Falinska, K. 1999. Seed bank dynamics in abandoned meadows during a 20-year period in the Bialowieza National Park. Jounal of Ecology 87: 461-475.

Haskell, D. 11/13/2000. Ecology lecture.

Hyatt, .A.; Casper, B.B. 2000. Seed bank formation during early secondary succession in a temperate deciduous forest. Journal of Ecology 88: 516-527.

Hyatt, L. 1999. Differences between seed bank composition and field recruitment in a temperate zone deciduous forest. The American Midland Naturalist. 142: 31-38.

Kirkham, F.W.; Kent, M. 1997. Soil seed bank composition in relation to the above-ground vegetation in fertilized and unfertilized hay meadows on a Somerset peat moor. British Geological Society: 34: 899-902.

Leckie, S.; Vellend, M.; Bell, G.; Waterway, M.J.; Lechowicz, M.J. 2000. The seed bank in an old-growth, temperate deciduous forest. Canadian Journal of Botany 78: 181-192.

Lopez-Marino, A.; Luis-Calabuig, E.; Fillat, F.; Bermudz, F.F. 2000. Floristic composition of established vegetation and the soil seed bank in pasture communities under different traditional management regimes. Agriculture, Ecosystems, and Environment 78: 273-282.