Joshua Everett's Mini Wisconsin Fast Plant Lab Report

Everett Wisconsin Fast Plant Lab Report

The Effect of Artificial Selection on Wisconsin Fast Plants

The purpose for performing this lab is to model artificial selection within Wisconsin Fast Plants. As a class, we chose the characteristic of average leaf length of the plants to be selected for in which a change in the average leaf length would be predicted in the offspring from the parent generation. The research question we are trying to answer is: does the selection for the average leaf length influence the phenotypes of the offspring? The results of the experiment showed that the research question is not supported by the data that my classmates and I collected.

In the beginning of the experiment, a parent generation of fifty seven Fast Plants was grown within planting units each having four pods. In each pod a diamond wick, planting soil, fertilizer seeds, and fast plant seeds were placed respectively in this order. They were grown continuously for approximately 40 days before the offspring plants were grown. All plants, however, were place on top of a water reservoir to ensure the plants maintained ideal water levels. All plants were also under cool-white fluorescent lights to allow plants to perform metabolic process and grow. After approximately five days the average leaf lengths for each of the plants were recorded (Figure 1). The averages for the leaf lengths vary continuously from plant to plant. (Figure 1). There is no relationship shown between the average leaf length for each plant in which there is no bell curve. The average leaf lengths for the plants are heavily concentrated between four to six millimeters and between eight and ten millimeters. The outliers for this graph are the average leaf length between six and eight millimeters and from ten to twenty millimeters. (Figure 1). The most concentrated averages and the outliers are not integrated in the graph to where a pattern is shown. The data is not normally distributed because if it was, a bell curve would be shown in which the sloping sides would contain the outliers and the curve would contain the average leaf lengths that  are most concentrated by the plants. Another account for why the data is skewed is because the standard deviation is 3.39 which is higher than the standard deviation for the survivors (1.26) and F1 generation (1.81). There is more variation of the data here which explains why the data is not more evenly dispersed. Even though a pattern is not depicted, the curve of the bell curve represents the mean of all the average leaf lengths for the parent generation plants which is 8.839181287.  

Figure 1

From the information from figure one, it was decided by my classmates and I that the leaf lengths under 12 mm was to be selected against. To implement this selection pressure, the class trimmed the leaves of all the plants that had a leaf length of twelve millimeters and below. The outcome the my classmates and I were expecting to see in the survivor parent population were all plants with leaf lengths of  twelve millimeters and over. The average leaf lengths for each plant for the survivors after selection were recorded (Figure 2). There is a significant decrease in the population of the parent population to the survivor population in which the parent population has 57 plants and the survivor population only has 8 plants. The frequency of higher average leaf length is significantly lower than in the original parent population when compared to the survivor population. To make this conclusion, the means of the parent population and the survivor population was compared. The parent population mean is 8.84 and the survivor population mean is 13.88. The average leaf length increased after the selection pressure was implemented. Because all leaf length under twelve millimeters were selected again, only leaf lengths above twelve millimeters were left. This caused a shift in the graph from eight to ten millimeters in the parent population to 13.5 millimeters to 14.5 millimeters. In addition, the dispersion and variation of the data for the survivor population is smaller than that of the parent population. This is accounted from through the standard deviations of the two populations. The standard deviation for the parent population is 3.39 but the survivor population has a standard deviation of 1.26. The smaller the standard deviation, the less variation and dispersion of the data. The standard deviation also accounts for the lack of fluctuation in the graphs because of the lack of variation and dispersion which is why the graph lacks a bell curve. On another note, the SEM of the survivor population is 0.45 and the 95% confidence interval for the survivor population is from 12.98 to 14.78. For the parent population, however, the SEM is 0.45 and the 955 confidence interval is from 7.94 to 9.74. The confidence intervals show the range of the where the means will lie if the experiment is performed again. Since the confidence intervals from the survivors population and the parent population do not overlap, I can conclude that the means for both populations are significantly different from each other. This would imply that the selection pressure that was implemented was a success.

Figure 2

Using the survivors population, my classmates and I performed sexual reproduction with the plants by cross pollinating the plants using bee sticks. If the process is done correctly the plants will produce produced seed pods which  are then used to plant the F1 generation. After letting the offspring grow for approximately nine days, we measured their leaf lengths for each plant a recorded them (figure 3).

The graph now depicts a bell curve in which the data is evenly distributed and an average leaf length can be roughly seen. The mean of the F1 generation is 6.462121212 which is different from the parent population and the survivor population in which their means were 8.84 and 13.88 respectively.  The average leaf length shifted along the graph in which it is not the same as the survivor population but it is similar to the parent population in which the average leaf length now ranges from 6 millimeters to 7.5 millimeters. The means of the parent population and the F1 generation are more similar than that of the survivor population. The standard deviation for the F1 generation is 1.81 which is higher than the survivors standard deviation which is only 1.26. The standard deviation is high for the F1 generation because there were more plants in the F1 generation (33) than the survivors generation (8) which increase the variation in the plants. To ensure that the mean of the F1 generation is not similar to the survivor population, the 95% confidence intervals were compared. The definition of similar or not similar is determined through 95% confidence intervals. The null hypothesis that is used to see if the survivors generation is similar to the F1 generation for is that there is no significant difference between the mean of the original population and the mean of the F1 generation.The SEM of the original population is 0.4492766416 and the SEM for the F1 generation is 0.3146201419. These SEM’s provide the range in which the confidence interval will range based on 5% built in error if the experiment was performed again. The upper and lower confidence intervals for the original survivor population is 14.78 and 12.98 respectively. The upper and lower confidence intervals for the F1 generation is 7.091361496 and 5.832880928 respectively. These confidence intervals do not overlap at any point in time which means that is a significant different between the mean of the survivor generation and the F1 generation. The null hypothesis failed to be rejected in this situation.

Figure 3

Discussion

Conclusions:

The results of the experiment does not support the research question because even though the means are different from each other, the mean of the  F1 generation is lower than the mean for the original population and the survivor parent population. If directional selection was inflicted on the parent population and affected the phenotypes of the offspring, we would expect for the  mean of the offspring to be greater than the original parent population and similar to the mean of the survivor parent population. We directionally selected for plants with average leaf lengths of 12mm and higher in which we wanted to have offspring with average leaf lengths of 12mm or larger but that was not the case. The average leaf lengths are instead smaller than the survivor parent population. The 95% confidence interval for the parent population is from 7.94 to 9.74 and the confidence interval for the F1 generation is from 5.83 to 7.09. The confidence intervals do not overlap which means that the means of the two populations are significantly different from each other. They are significantly different from each other in which the F1 generation has a lower mean than the parent population.  Through these comparisons, I can conclude that the research question was not fully supported by the experiment. After continually analysing the results of the experiment, I came to the conclusion that genetics did not have a role in this experiment but the results were rather an act of randomness. I had an outcome of what I wanted the outcome of the experiment to be and the results were not matching up. I wanted the mean of the F1 generation to similar to the mean of the survivors or even higher. In this experiment we were focused on phenotypes and genotypes so genetics was not a factor in this experiment in which genes of larger average leaves would not have passed down to the next generation. There were a couple limitations of the experiment. Specifically, one limitation would include that not all of the parent generation seeds grew into plants for the selection process to take place and to produce offspring. The seeds that did not bloom could have possibly had leaves of 12 mm or higher that could have affected the results of the experiment in some way.

Experimental Evaluation:

I do not feel 100% confident in the results I obtained because we only did one trial of the experiment. Performing multiple trials of the same experiment can confirm or refute the results obtain in the first trial. Performing multiple trials solidifies your experiment and an average of some sort can be calculated to incorporated all the results of each trial. Looking back on the lab, I would have been more organized in my notes and data. Many times throughout the experiment there was confusion on what day of the experiment we were on and there was confusion on what data belonged to which plant. At times there was a lot of confusion but as the experiment went on, it became easier to anticipate what we were doing. I do not believe there were any intentional human errors in this lab. The class followed the procedure to the experiment very well. Any sources of error in this lab comes from random errors. When planting the send and growing the parent generation, an error in planting the seeds, like forgetting fertilizer, could affect the growth of the plant which could affect the results of the experiment. Another source of error would be not keeping the plants hydrated at all times. Sometimes the water reservoir was not seeping up water through the cloth to be pulled up to the plant through the time. Many times we found our plants dry and hard. This characteristic of the soil of the plant could also affect growth and development and reproduction. Lastly, another source of error could have derived from measuring, Many times, it was difficult and very tedious to measure each leaves length. Error in measurement could have affected the selection and the over results of the experiment. I do not have any changes or improvements to make for this lab. The only thing I would do different is to just perform this experiment without the distractions of any other activities. In class, a lot of time was spent on other activities focused on other parts of the curriculum. Many time the students of the class forgot about the plants are lost track of the step of the procedure we were on. This is another potential error source that I would want to eliminate if I were to perform this experiment again.