際際滷

際際滷Share a Scribd company logo

Plant Competition Experiment

Download as DOCX, PDF
Jordan Sedlock
Jordan Sedlock

This experiment investigated the effects of plant density and nutrient levels on competition between dwarf marigold plants. Plants were grown at densities of 3, 5, or 10 seeds per pot and received 0, 1, or 4 mL of potassium solution per week. Higher plant densities resulted in significantly lower biomass per plant, indicating intraspecific competition for resources increased with density. Nutrient levels did not significantly affect biomass or flower production. The highest densities showed the strongest effects of competition through reduced growth.

Convert to study guideBETA

Transform any presentation into a summarized study guide, highlighting the most important points and key insights.

1 of 11
Downloaded 24 times
Plant Competition Experiment General Ecology Dr. Gresens Jordan Sedlock December 5, 2014 Partners: Shelby Montgomery & Kayla Haile
2 Introduction: One of the key processes in plant communities and populations is competition (Berger 2008). Competition is the process that arises, when organisms, whether from the same species (intra) or of different species from each other (inter) share common, essential resources and as a result, experience pressure when it comes to growth, survival, and reproduction. Resources are defined as something necessary for survival, in the case of plants that is generally things such as water, light, nutrients, and space. In this manipulative experiment, nutrients, specifically potassium, and space were manipulated, based on a full factorial design, in order to determine if there were interactions occurring to suggest intraspecific exploitation competition or intraspecific resource competition. It is hypothesized that low density and therefore maximum amount of space will yield little evidence of intraspecific competition and conversely, that high density and therefore minimal amount of space will yield substantial evidence of intraspecific competition. It is also hypothesized that little to no added nutrients will hinder growth but fail to indicate intraspecific competition in lower density experimental units and conversely, higher amounts of added nutrients will more strongly influence growth but also more strongly indicate intraspecific competition in high density experimental units. Methods: To perform this experiment, dwarf marigolds were selected to be the plant of choice for manipulation. A full factorial design was used when planning this experiment. The following table (Table 1) provides a general schematic of the physical set-up of this experiment.
3 Table 1. General, physical set-up of full factorial experiment including factors, factor levels, species, and replicates FACTORB: Nutrients(K) = 1 pot FACTOR A: Density 3 Marigold Seeds 5 Marigold Seeds 10 Marigold seeds No Added Potassium ≒ ≒ ≒ ≒ ≒ ≒ Low Potassium Added (1 mL) ≒ ≒ ≒ ≒ ≒ ≒ High Potassium Added (4 mL) ≒ ≒ ≒ ≒ ≒ ≒ There were three factor levels: low, medium, and high as well as two different factors that were being manipulated, which included seed density per experimental unit as well as amount of a particular nutrient per experimental unit. This experiments nutrient was potassium, chosen because potassium is known to increase crop yield as well as aid in many physical processes essential for plant life in its presence and in its absence is known to stunt plant growth (SMART 2013). Experimental units consisted of one flowerpot filled with potting soil. There were a total of 36 replicate, experimental units. Each experimental unit contained a different density of seeds, ranging from 3 10 seeds with 5 seeds being the intermediate between the highest density (10 seeds) and the lowest density (3 seeds). Twelve experimental units contained 3 seeds, a second set of twelve experimental units contained 5 seeds, and the final set of twelve experimental units contained 10 seeds. Pots were assigned random numbers using a random number table, and then labeled. Once labeled, pots were arranged in ascending order based on random number assignment to ensure randomization of experimental units and reduce any bias or confounding that may have occurred due to grouping all of the same replicates per each manipulated set of factors together. Pots were watered three times per week and the amount of water ranged from 10 50 mL depending upon soil conditions as well as physical plant conditions and taking into consideration the temperature in the greenhouse where plants remained for the entire experiment.
4 Once the experiment began, different levels of nutrients were then manipulated to determine if there were interactions between the density of dwarf marigolds and the amount of nutrients they received. To maintain stoichiometry and prevent other nutrients from becoming limiting, other than potassium, other essential nutrients were added to all experimental units. A solution of disodium phosphate (0.0136 g/L) was created and 2 mL of this was added to each experimental unit, once per week. A solution of sodium nitrate (0.085 g/L) was created and 2 mL of this was also added to each experimental unit, once per week. A final solution of potassium chloride was created (0.0748 g/L) and used as the second factor to be manipulated in this experiment. Twelve replicates of varying densities received 0 mL of KCl solution, once per week, which was representative of the control group. Twelve replicates of varying densities received 1 mL of KCl solution, once per week, which was representative of the medium factor level and the final twelve replicates received 4 mL of KCl solution, once per week, which was representative of the high factor level. Total duration of the experiment was eight weeks. Week one of the experiment was designated to allow initial germination and growth of plants. Beginning week two, addition and therefore, manipulation, of nutrients including Na2HPO4, NaNO3, and the KCl solutions, in manner and amounts previously mentioned, began and continued to the end of the experiment. Following the eight-week experiment, plants were harvested. All above ground biomass was cut at the dirt level, placed in an envelope and weighed on a top-loading balance. Total number of surviving plants was summed as well as flowers produced by each plant and buds produced by each plant. Response variables that were examined include per capita biomass of each pot, and the proportion of flowers per total number of final plants.
5 Results: Following the harvesting of all experimental units, data analysis was then performed on the per capita biomass of marigolds. Analysis of variance using the ANOVA technique was conducted to determine whether or not the two factors that were manipulated had significant effects on the per capita biomass of marigolds. Table 2 contains the output from the ANOVA analysis for this response variable. Table 2. ANOVA of Per capita biomass of marigolds Source of Variation SS df MS F P-value Density 22.82866314 2 11.41433157 18.72819942 7.9126E-06 KCl added (mL) 0.238713718 2 0.119356859 0.195836177 0.82330436 Interaction 5.049209952 4 1.262302488 2.071137725 0.112492507 Within 16.45577055 27 0.609472983 Total 44.57235736 35 Due to a large value of F for the density factor (Table 2) p < 0.05 (p = 7.913E-6) (Table 2), and therefore indicates that density had a highly significant effect on the per capita biomass of marigolds. Due to a small value of F for the nutrient level factor, (Table 2) p > 0.05 (p = 0.8233) (Table 2), and therefore indicates that nutrient level did not have a significant effect on the per capita biomass of marigolds. Following the statistical analysis, graphical analysis (Figure 1) was performed in order to depict the data from Table 2.
6 Figure 1. Interaction plot of per capita biomass of marigolds Figure 1 depicts interactions of factors as a function of the response variable. Figure 1 indicates that density had the strongest effect on per capita biomass in experimental units with 3 seeds. The amount of nutrients added to experimental units with 3 seeds produced a negative effect on the per capita biomass. Experimental units with 5 seeds yielded lower per capita biomass than experimental units with 3 seeds and experimental units with 10 seeds yielded the lowest per capita biomass. Overall, amount of nutrients added, more strongly influenced experimental units with 5 seeds in them and produced per capita biomass that is nearly comparable to that produced in experimental units with 3 seeds. Amount of nutrients added produced an overall positive effect on experimental units with 5 seeds in them. Amount of nutrients added produced little positive affect on experimental units with 10 seeds in them. Error bars in Figure 1 indicate high variation amongst replicates and could therefore contribute to the higher p value (Table 2) for amount of nutrients added to plants. Analysis of variance using the ANOVA technique was then conducted to determine whether or not the two factors that were manipulated had significant effects on the proportion of 0 1 2 3 4 5 6 None (0 mL) Low (1 mL) High (4 mL) PercapitaBiomass(g) Potassium addition level Mean (+ S. E.) of per capita biomass of marigolds 3 seeds 5 seeds 10 seeds
7 flowers per marigold plant. Table 3 contains the output from the ANOVA analysis for this response variable. Table 3. ANOVA of proportion of flowers per plant Source of Variation SS df MS F P-value Density 0.480873331 2 0.240436666 2.241704469 0.125691869 KCl added (mL) 0.190476505 2 0.095238253 0.887951162 0.42317498 Interaction 0.435833963 4 0.108958491 1.015871415 0.416754327 Within 2.895916951 27 0.107256183 Total 4.00310075 35 Due to a small value of F for the density factor (Table 3) p > 0.05 (p = 0.125) (Table 3), and therefore indicates that density had no significant effect on the proportion of flowers per plant. Due to a small value of F for the nutrient level factor, (Table 3) p > 0.05 (p = 0.423) (Table 3), and therefore indicates that nutrient level had no significant effect on the proportion of flowers per plant. Following the statistical analysis, graphical analysis (Figure 2) was performed in order to depict the data from Table 3. Figure 2. Interaction plot of proportion of flowers per final number of plants 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 None (0 mL) Low (1 mL) High (4 mL) ProportionFlowers:FinalPlants Potassium addition level Mean (+ S. E.) of flowersto plants 3 seeds 5 seeds 10 seeds
8 Figure 2 depicts interactions of factors as a function of the response variable. When considering experimental units with 3 seeds, it appears, overall, that these units yielded the most flowers per plant. When considering experimental units with 5 seeds, it appears, overall, that these units yielded the least flowers per plant. When considering experimental units with 10 seeds, it appears, overall, that these units yielded the second most flowers per plant compared to units with 3 seeds. Considering amount of nutrients added to the experimental units, units with 3 seeds appear to have been negatively affected overall, units with 5 seeds appear to have been positively affected overall, and units with 10 seeds appear to have been negatively affected overall. Error bars in Figure 2 indicate very high variation within replicate groups. Due to substantial overlap of error bars in the density levels, it is not possible to discern with certainty that there was any interaction of factors on the response variable measured in this analysis. Discussion: Before beginning this experiment, it was hypothesized that low density and therefore maximum amount of space would yield little evidence of intraspecific competition and conversely, that high density and therefore minimal amount of space would yield substantial evidence of intraspecific competition. This assumption was based off of a previous experiment that found that as density of a plant population increases, above ground biomass decreases (Wang 2005). It was also hypothesized that little to no added nutrients would hinder growth but fail to indicate intraspecific competition in lower density experimental units and conversely, higher amounts of added nutrients would more strongly influence growth but also more strongly indicate intraspecific competition in high density experimental units. After completion of the experiment, it has been concluded that units with the lowest density yielded the highest per capita biomass (Figure 1) and the highest proportion of flowers to plants (Figure 2) and is
9 therefore indicative of little intraspecific competition. It has also been concluded that units with the highest density yielded the lowest per capita biomass (Figure 1) and is therefore indicative of prevalent intraspecific competition. It cannot be stated with certainty that density strongly affected the proportion of flowers to plants (Figure 2) with units of highest density because the intermediate factor level (5 seeds) yielded a lower proportion of flowers to plants than the highest density factor level (10 seeds), however, the data still suggest, overall, that with increased density, intraspecific competition also increased. These results support the first part of the aforementioned hypothesis. Upon further investigation of experimental results and examining effects of added nutrients, it has been concluded that the second part of the aforementioned hypothesis is not supported because there is high variation within replicates and very little evidence of apparent trends in the data to state with certainty that amount of nutrients added had any effect on intraspecific competition. When considering Figure 1, added nutrients decreased per capita biomass in units with 3 seeds and 10 seeds, however, units with 5 seeds experienced, overall, more growth as a result of an increase in nutrients and therefore experienced lower intraspecific competition than units with 3 seeds and 10 seeds. When considering Figure 2, increased amounts of nutrients greatly lowered the proportion of flowers to plants in units with 3 seeds and 10 seeds, however, units with 5 seeds experienced, overall, more production of flowers per plant and therefore experienced lower intraspecific competition than units with 3 and 10 seeds. To design an experiment that better aims at supporting the second part of the previously stated hypothesis, many things could be done. To decrease amount of variance within replicate groups, more replicates could be added, possibly increasing sample size from 36 to 56, for example. Experimental units with 5 seeds appeared to be the outliers in the data sets and
10 experienced more overall effects of manipulation of factors. When studying proportion of flowers to plants, decreasing the amount of seeds from 3 to 2 in the low density units as well as increasing the amount of seeds from 10 to 12, perhaps, and leaving 5 seeds in the intermediate- level density units could allow for results that support the thought that highest density would yield lowest proportion of flowers to plants, intermediate density would yield the second lowest proportion of flowers to plants and lowest density would yield the highest proportion of flowers to plants. Along with increasing replicates to decrease variation, adding more factor levels for amount of nutrients added might yield better results with apparent trends that more strongly indicate what type of effect amount of nutrients has on marigolds as well as suggest an interaction between density and nutrient levels. It is possible that there was so much variation among replicates due to variation in amounts of water each plant received. Rather than watering plants every other day with different amounts of water based on soil conditions and temperature of the environment, plants could be kept in an environment with a more stable temperature and more access to sunlight as well as being watered everyday with lesser amounts of water that would remain consistent within all replicates throughout the duration of the experiment. Variation could also have been introduced when administering nutrients once per week. Cross contamination of pipettes used for administering nutrients could have occurred if another group borrowed pipettes for their solutions. Lack of precision and accuracy when delivering nutrients to plants could be another cause of variation within replicates. To correct for these causes of variation, more care can be taken when delivering nutrients to plants to ensure better accuracy and precision. Also, pipettes can be labeled and/or placed away so as not to allow use by other groups.
11 Literature Cited Berger, U. et. al. 2008. Competition among plants: concepts, individual-based modeling approaches, and a proposal for a future research strategy. Perspectives in Plant Ecology, Evolution, and Systematics 9: 121-135. SMART. December 3, 2014. http://www.smart-fertilizer.com/articles/potassium-in-plants. Wang, L. et. al. 2005. Effects of intraspecific competition on growth and photosynthesis of Atriplex prostrate. Aquatic Botany 83: 187-192.

More Related Content

Plant Competition Experiment

  • 1. Plant Competition Experiment General Ecology Dr. Gresens Jordan Sedlock December 5, 2014 Partners: Shelby Montgomery & Kayla Haile
  • 2. 2 Introduction: One of the key processes in plant communities and populations is competition (Berger 2008). Competition is the process that arises, when organisms, whether from the same species (intra) or of different species from each other (inter) share common, essential resources and as a result, experience pressure when it comes to growth, survival, and reproduction. Resources are defined as something necessary for survival, in the case of plants that is generally things such as water, light, nutrients, and space. In this manipulative experiment, nutrients, specifically potassium, and space were manipulated, based on a full factorial design, in order to determine if there were interactions occurring to suggest intraspecific exploitation competition or intraspecific resource competition. It is hypothesized that low density and therefore maximum amount of space will yield little evidence of intraspecific competition and conversely, that high density and therefore minimal amount of space will yield substantial evidence of intraspecific competition. It is also hypothesized that little to no added nutrients will hinder growth but fail to indicate intraspecific competition in lower density experimental units and conversely, higher amounts of added nutrients will more strongly influence growth but also more strongly indicate intraspecific competition in high density experimental units. Methods: To perform this experiment, dwarf marigolds were selected to be the plant of choice for manipulation. A full factorial design was used when planning this experiment. The following table (Table 1) provides a general schematic of the physical set-up of this experiment.
  • 3. 3 Table 1. General, physical set-up of full factorial experiment including factors, factor levels, species, and replicates FACTORB: Nutrients(K) = 1 pot FACTOR A: Density 3 Marigold Seeds 5 Marigold Seeds 10 Marigold seeds No Added Potassium ≒ ≒ ≒ ≒ ≒ ≒ Low Potassium Added (1 mL) ≒ ≒ ≒ ≒ ≒ ≒ High Potassium Added (4 mL) ≒ ≒ ≒ ≒ ≒ ≒ There were three factor levels: low, medium, and high as well as two different factors that were being manipulated, which included seed density per experimental unit as well as amount of a particular nutrient per experimental unit. This experiments nutrient was potassium, chosen because potassium is known to increase crop yield as well as aid in many physical processes essential for plant life in its presence and in its absence is known to stunt plant growth (SMART 2013). Experimental units consisted of one flowerpot filled with potting soil. There were a total of 36 replicate, experimental units. Each experimental unit contained a different density of seeds, ranging from 3 10 seeds with 5 seeds being the intermediate between the highest density (10 seeds) and the lowest density (3 seeds). Twelve experimental units contained 3 seeds, a second set of twelve experimental units contained 5 seeds, and the final set of twelve experimental units contained 10 seeds. Pots were assigned random numbers using a random number table, and then labeled. Once labeled, pots were arranged in ascending order based on random number assignment to ensure randomization of experimental units and reduce any bias or confounding that may have occurred due to grouping all of the same replicates per each manipulated set of factors together. Pots were watered three times per week and the amount of water ranged from 10 50 mL depending upon soil conditions as well as physical plant conditions and taking into consideration the temperature in the greenhouse where plants remained for the entire experiment.
  • 4. 4 Once the experiment began, different levels of nutrients were then manipulated to determine if there were interactions between the density of dwarf marigolds and the amount of nutrients they received. To maintain stoichiometry and prevent other nutrients from becoming limiting, other than potassium, other essential nutrients were added to all experimental units. A solution of disodium phosphate (0.0136 g/L) was created and 2 mL of this was added to each experimental unit, once per week. A solution of sodium nitrate (0.085 g/L) was created and 2 mL of this was also added to each experimental unit, once per week. A final solution of potassium chloride was created (0.0748 g/L) and used as the second factor to be manipulated in this experiment. Twelve replicates of varying densities received 0 mL of KCl solution, once per week, which was representative of the control group. Twelve replicates of varying densities received 1 mL of KCl solution, once per week, which was representative of the medium factor level and the final twelve replicates received 4 mL of KCl solution, once per week, which was representative of the high factor level. Total duration of the experiment was eight weeks. Week one of the experiment was designated to allow initial germination and growth of plants. Beginning week two, addition and therefore, manipulation, of nutrients including Na2HPO4, NaNO3, and the KCl solutions, in manner and amounts previously mentioned, began and continued to the end of the experiment. Following the eight-week experiment, plants were harvested. All above ground biomass was cut at the dirt level, placed in an envelope and weighed on a top-loading balance. Total number of surviving plants was summed as well as flowers produced by each plant and buds produced by each plant. Response variables that were examined include per capita biomass of each pot, and the proportion of flowers per total number of final plants.
  • 5. 5 Results: Following the harvesting of all experimental units, data analysis was then performed on the per capita biomass of marigolds. Analysis of variance using the ANOVA technique was conducted to determine whether or not the two factors that were manipulated had significant effects on the per capita biomass of marigolds. Table 2 contains the output from the ANOVA analysis for this response variable. Table 2. ANOVA of Per capita biomass of marigolds Source of Variation SS df MS F P-value Density 22.82866314 2 11.41433157 18.72819942 7.9126E-06 KCl added (mL) 0.238713718 2 0.119356859 0.195836177 0.82330436 Interaction 5.049209952 4 1.262302488 2.071137725 0.112492507 Within 16.45577055 27 0.609472983 Total 44.57235736 35 Due to a large value of F for the density factor (Table 2) p < 0.05 (p = 7.913E-6) (Table 2), and therefore indicates that density had a highly significant effect on the per capita biomass of marigolds. Due to a small value of F for the nutrient level factor, (Table 2) p > 0.05 (p = 0.8233) (Table 2), and therefore indicates that nutrient level did not have a significant effect on the per capita biomass of marigolds. Following the statistical analysis, graphical analysis (Figure 1) was performed in order to depict the data from Table 2.
  • 6. 6 Figure 1. Interaction plot of per capita biomass of marigolds Figure 1 depicts interactions of factors as a function of the response variable. Figure 1 indicates that density had the strongest effect on per capita biomass in experimental units with 3 seeds. The amount of nutrients added to experimental units with 3 seeds produced a negative effect on the per capita biomass. Experimental units with 5 seeds yielded lower per capita biomass than experimental units with 3 seeds and experimental units with 10 seeds yielded the lowest per capita biomass. Overall, amount of nutrients added, more strongly influenced experimental units with 5 seeds in them and produced per capita biomass that is nearly comparable to that produced in experimental units with 3 seeds. Amount of nutrients added produced an overall positive effect on experimental units with 5 seeds in them. Amount of nutrients added produced little positive affect on experimental units with 10 seeds in them. Error bars in Figure 1 indicate high variation amongst replicates and could therefore contribute to the higher p value (Table 2) for amount of nutrients added to plants. Analysis of variance using the ANOVA technique was then conducted to determine whether or not the two factors that were manipulated had significant effects on the proportion of 0 1 2 3 4 5 6 None (0 mL) Low (1 mL) High (4 mL) PercapitaBiomass(g) Potassium addition level Mean (+ S. E.) of per capita biomass of marigolds 3 seeds 5 seeds 10 seeds
  • 7. 7 flowers per marigold plant. Table 3 contains the output from the ANOVA analysis for this response variable. Table 3. ANOVA of proportion of flowers per plant Source of Variation SS df MS F P-value Density 0.480873331 2 0.240436666 2.241704469 0.125691869 KCl added (mL) 0.190476505 2 0.095238253 0.887951162 0.42317498 Interaction 0.435833963 4 0.108958491 1.015871415 0.416754327 Within 2.895916951 27 0.107256183 Total 4.00310075 35 Due to a small value of F for the density factor (Table 3) p > 0.05 (p = 0.125) (Table 3), and therefore indicates that density had no significant effect on the proportion of flowers per plant. Due to a small value of F for the nutrient level factor, (Table 3) p > 0.05 (p = 0.423) (Table 3), and therefore indicates that nutrient level had no significant effect on the proportion of flowers per plant. Following the statistical analysis, graphical analysis (Figure 2) was performed in order to depict the data from Table 3. Figure 2. Interaction plot of proportion of flowers per final number of plants 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 None (0 mL) Low (1 mL) High (4 mL) ProportionFlowers:FinalPlants Potassium addition level Mean (+ S. E.) of flowersto plants 3 seeds 5 seeds 10 seeds
  • 8. 8 Figure 2 depicts interactions of factors as a function of the response variable. When considering experimental units with 3 seeds, it appears, overall, that these units yielded the most flowers per plant. When considering experimental units with 5 seeds, it appears, overall, that these units yielded the least flowers per plant. When considering experimental units with 10 seeds, it appears, overall, that these units yielded the second most flowers per plant compared to units with 3 seeds. Considering amount of nutrients added to the experimental units, units with 3 seeds appear to have been negatively affected overall, units with 5 seeds appear to have been positively affected overall, and units with 10 seeds appear to have been negatively affected overall. Error bars in Figure 2 indicate very high variation within replicate groups. Due to substantial overlap of error bars in the density levels, it is not possible to discern with certainty that there was any interaction of factors on the response variable measured in this analysis. Discussion: Before beginning this experiment, it was hypothesized that low density and therefore maximum amount of space would yield little evidence of intraspecific competition and conversely, that high density and therefore minimal amount of space would yield substantial evidence of intraspecific competition. This assumption was based off of a previous experiment that found that as density of a plant population increases, above ground biomass decreases (Wang 2005). It was also hypothesized that little to no added nutrients would hinder growth but fail to indicate intraspecific competition in lower density experimental units and conversely, higher amounts of added nutrients would more strongly influence growth but also more strongly indicate intraspecific competition in high density experimental units. After completion of the experiment, it has been concluded that units with the lowest density yielded the highest per capita biomass (Figure 1) and the highest proportion of flowers to plants (Figure 2) and is
  • 9. 9 therefore indicative of little intraspecific competition. It has also been concluded that units with the highest density yielded the lowest per capita biomass (Figure 1) and is therefore indicative of prevalent intraspecific competition. It cannot be stated with certainty that density strongly affected the proportion of flowers to plants (Figure 2) with units of highest density because the intermediate factor level (5 seeds) yielded a lower proportion of flowers to plants than the highest density factor level (10 seeds), however, the data still suggest, overall, that with increased density, intraspecific competition also increased. These results support the first part of the aforementioned hypothesis. Upon further investigation of experimental results and examining effects of added nutrients, it has been concluded that the second part of the aforementioned hypothesis is not supported because there is high variation within replicates and very little evidence of apparent trends in the data to state with certainty that amount of nutrients added had any effect on intraspecific competition. When considering Figure 1, added nutrients decreased per capita biomass in units with 3 seeds and 10 seeds, however, units with 5 seeds experienced, overall, more growth as a result of an increase in nutrients and therefore experienced lower intraspecific competition than units with 3 seeds and 10 seeds. When considering Figure 2, increased amounts of nutrients greatly lowered the proportion of flowers to plants in units with 3 seeds and 10 seeds, however, units with 5 seeds experienced, overall, more production of flowers per plant and therefore experienced lower intraspecific competition than units with 3 and 10 seeds. To design an experiment that better aims at supporting the second part of the previously stated hypothesis, many things could be done. To decrease amount of variance within replicate groups, more replicates could be added, possibly increasing sample size from 36 to 56, for example. Experimental units with 5 seeds appeared to be the outliers in the data sets and
  • 10. 10 experienced more overall effects of manipulation of factors. When studying proportion of flowers to plants, decreasing the amount of seeds from 3 to 2 in the low density units as well as increasing the amount of seeds from 10 to 12, perhaps, and leaving 5 seeds in the intermediate- level density units could allow for results that support the thought that highest density would yield lowest proportion of flowers to plants, intermediate density would yield the second lowest proportion of flowers to plants and lowest density would yield the highest proportion of flowers to plants. Along with increasing replicates to decrease variation, adding more factor levels for amount of nutrients added might yield better results with apparent trends that more strongly indicate what type of effect amount of nutrients has on marigolds as well as suggest an interaction between density and nutrient levels. It is possible that there was so much variation among replicates due to variation in amounts of water each plant received. Rather than watering plants every other day with different amounts of water based on soil conditions and temperature of the environment, plants could be kept in an environment with a more stable temperature and more access to sunlight as well as being watered everyday with lesser amounts of water that would remain consistent within all replicates throughout the duration of the experiment. Variation could also have been introduced when administering nutrients once per week. Cross contamination of pipettes used for administering nutrients could have occurred if another group borrowed pipettes for their solutions. Lack of precision and accuracy when delivering nutrients to plants could be another cause of variation within replicates. To correct for these causes of variation, more care can be taken when delivering nutrients to plants to ensure better accuracy and precision. Also, pipettes can be labeled and/or placed away so as not to allow use by other groups.
  • 11. 11 Literature Cited Berger, U. et. al. 2008. Competition among plants: concepts, individual-based modeling approaches, and a proposal for a future research strategy. Perspectives in Plant Ecology, Evolution, and Systematics 9: 121-135. SMART. December 3, 2014. http://www.smart-fertilizer.com/articles/potassium-in-plants. Wang, L. et. al. 2005. Effects of intraspecific competition on growth and photosynthesis of Atriplex prostrate. Aquatic Botany 83: 187-192.