Introduction
Introduction:
Many seeds will partake in a period of dormancy that delays the seed from germinating until environmental conditions are more favorable. The dormancy is displayed in one of two types. The first of the two categories of seed dormancy affecting the germination of seeds is called embryo imposed dormancy from immature seed development or specific chemicals within the embryo affecting its germination, and the second category is seed coat-imposed dormancy from a hard seed coat, impermeability, or chemicals in the seed coat itself (Kimura et al, 2015). With many seeds, it can be assumed that if plants are picked when ripe, seed immaturity is not likely to be a factor and instead the seed coat is the most likely responsible factor for seed dormancy. This dormancy may be long-lasting, especially in harsher climates such as Alaska, where the winters are extremely cold and dry exposing seeds to unfavorable environmental conditions for germination, thus prolonging dormancy. This causes the seeds to be slow to propagate and often times fail when growing a garden in Alaska.
It is common practice to soak the seeds for long periods of time as a priming method to partially hydrate the seeds for breaking dormancy (Wagner and Oplinger, 2016), however, this method does not always successfully break dormancy in seeds subjected to harsh climates like Alaska with its long cold winters and seeds that are left out for long periods of time with excessive drying making the priming methods have lower success rates. To counteract the dormancy of seeds and initiate germination some have resorted to scarification methods. Scarification can be through mechanical, water, or chemical means to break down the seed coat, allowing for increased permeability with water absorption to initiate germination. Scarification may be accomplished through physical abrasions with sandpaper, boiling, or placement in a concentration of sulfuric acid (Murphy, 2011). Many homeowners do not have access to sulfuric acid, and in parts of rural Alaska that are accessible only by plane, regulations prohibit the shipment of harsh substances, like acids, by air. As a result, this study focuses on physical scarification methods with easily accessible materials in rural Alaska.
Kleyheeg and associates (2018), performed a similar experiment testing the effects of mechanical scarification on 250 different species, where they scarred the seed coats using sandpaper until the seed coats were visibly damaged. It was found that seeds germinated more often when scarred, suggesting that the mechanical abrasion of the seed coat, demonstrated with sand paper, stimulates germination but it did not distinguish by how much the germination lengths change by (Kleyheeg et al, 2018). Additionally, it has also been shown that in C. Colocynthis seeds, a gourd species with a hard seed coat, scarification methods attributed to significant increases in the germination rates of the seeds (Saberi et al, 2018). Furthermore, multiple researchers, including Jones and colleagues (2016) have claimed the benefits of scarification for increasing germination rates in dormant seeds, despite this, all of these researchers have yet to investigate the effects of scarification on the germination lengths for improved plant growth. As a result of this observation, the following experiment was developed to go beyond simple germination rates and instead determine the effects of mechanical scarification on germination growth lengths in a common hard seed plant grown in Aniak Alaska, an area subjected to the harsh cold winters of Alaska. For the purposes of reduced time requirements for the experiment in the class, as well as ease of obtaining seeds in the Bush of Alaska where shipping can take up to a month to deliver, Zea mays corn kernels were used and acquired from the small market store in Aniak Alaska. Corn was chosen for its relative commonality as a plant grown by gardeners.
For the purposes of the experiment, 30 control group seeds and 30 experimental seeds (scarification group) were compared for significant differences in the germination abilities of the two groups with the following hypothesis.
Null Hypothesis: There is no difference in average germination lengths, measured by tail length, when Zea mays corn seeds are subjected to scarification and when they are not.
Ho: GC = GEx
Alternative Hypothesis: There is a difference in average germination lengths, measured by tail length, when Zea mays corn seeds are subjected to scarification and when they are not.
HA: GC ? GEx
Methodology:
This experiment was conducted in Aniak Alaska with the site consisting of a large wall in my apartment. To perform this experiment the following items were acquired: one bag of Zea mays corn kernels, 60 plastic Ziplock resealable bags, a paper towel roll, 2 pipettes, a 10 mL graduated cylinder, two 150 mL beakers, water, 1 roll of tape, 1 permanent marker, 1 ruler with mm measurements, 1 large book, and 2 pieces of sandpaper. To ensure a constant environment the seeds were placed in a room kept at 68o F with equal access to light for all seeds.
After all materials were collected, 30 of the plastic bags were labeled 1-30 with a C for the control group and the other 30 bags were labeled 1-30 with an Ex for the experimental group. I used Brawny pick a size paper towels, choosing the smallest size for all 60 bags so that all the paper towels were the same size. These paper towels were folded in half and place in each of the Ziploc bags. To reduce bias in seed selection, the container with the seeds was shaken and 60 seeds were chosen at random. Of the selected seeds, I placed 30 of them in a 150 mL beaker of water containing 100 mL of water overnight, before each one was placed between the folded paper towel in the bags labeled with a C for the control group. The other 30 had a layer of the seed coat removed by mechanical scarification using sandpaper.
To create the same amount of scarification on all 30 seeds, the seeds were placed on top of a piece of sandpaper, with all seeds laying down the same way and another piece of sandpaper placed on top. Then a book was placed over the top of the seeds and second piece of sandpaper so that all seeds are under the book. This book created the needed weight to keep all the seeds in place and the same amount of force on each seed for the same amount of mechanical scarification being introduced to each seed. I moved the book and sandpaper forward and backward for 2 minutes until I saw a layer of the seed coat removed for the mechanical scarification. The seeds were then placed in an “EX” labeled 150 mL beaker filled with 100 mL of water overnight before being placed in the bags labeled Ex between the folded paper towel.
After placement of the seeds in each bag between the folded paper towel, 10 mL of water was measured with a 10 mL graduated cylinder and pipette onto the paper towels in each bag until every part of the paper towel was equally moist. The bags were then taped on the largest wall in the room and the germination lengths were measured and recorded with a ruler after 10 days of time. To measure the germination lengths the root was stretched straight across the edge of the ruler and measured to the mm mark.
Results and Discussion:
At the conclusion of 10 days, all the data was collected and analyzed for the gemination lengths of the Zea mays corn seeds. The following data table shows the results of the experiment with 4 seeds in the control group not germinating and 8 seeds in the experimental group treated with mechanical scarification showing zero growth, although seed 16 in the control group did look as if it was trying to germinate but could not break free of the hard seed coat.
Table 1 – Raw data of germination lengths (mm) of the corn seeds within each treatment group after 10 days, along with the mean, median, and standard deviation for the values in each treatment group of 30 seeds. Table and statistical values were created using Excel (Microsoft, 2013).
Seed #
Control Group Lengths (mm)
Experimental Group Lengths (mm)
1
142
5
2
0
16
3
106
106
4
130
103
5
122
81
6
90
92
7
115
150
8
133
111
9
142
0
10
70
71
11
90
0
12
45
54
13
109
58
14
0
102
15
105
124
16
0
111
17
17
111
18
135
105
19
59
115
20
153
0
21
109
105
22
118
0
23
84
131
24
105
76
25
118
57
26
122
0
27
106
120
28
0
18
29
142
125
30
116
105
Mean (mm)
92.8
75.1
Median
106
92
STDV
48.0
48.8
Figure 1 – Histogram showing the frequency of each growth amount for both the control and the scarification treatment group created using Excel (Microsoft, 2013).
Furthermore in Table 1, it is clear from the means that the average germination lengths, measured by tail length, are different when Zea mays corn seeds are subjected to scarification and when they are not, however, the standard deviations are similar. Additionally, Figure 1 above shows the differences in the frequencies of the two groups at each germination length interval with a normal distribution not seen. In order to see if there is a significant difference between the two treatment groups, a Shapiro-Wilk test was performed to investigate the degree of the non-normal distribution seen in the histogram. Using the Shapiro-Wilk function in the Real Statistics Resource Pack (Zaiontz, 2013), a W value of 0.847 and a p-value of 0.005 was given which is greater than the alpha value of 0.05, and so it can be concluded that with a 95% confidence interval the data for the control group is not normally distributed. When conducted on the treatments group, the Real Stats function (Zaiontz, 2013) for the Shapiro-Wilk test gives a W value of 0.869 and a p-value of 0.005. Because the p-value is greater than the alpha value of 0.05 it can be concluded that with a 95% confidence interval that the data for the experimental group also is not normally distributed. Nonetheless because the data for the two groups are both skewed to the right, according to Whitlock and Schluter (2015, pg.376), a sample size of 30 will give “reasonably accurate” answers when using the two-sample t-test to compare the means of the two groups, especially since the standard deviations are equal for the two.
With these assumptions in mind, a two-sample t-test with equal variances was performed giving the following information in Table 2.
Table 2 – Two-sample T-test assuming equal variance results for the difference between the control and experimental group
Control Group
Experimental Group
Mean
92.77
75.07
Variance
2244.05
2327.58
Observations
30
30
Pooled Variance
2285.81
Hypothesized Mean Difference
0
df
58
t Stat
1.43
P(T
