Biology Lab Report
Plant damage by herbiovry needs to be studied especially with potential negative impacts from climate change. The impact of caterpillars on cucumber leaves was measured under three different situations. Firstly when the leaves were damaged with holes, secondly when saliva was added or thirdly when leaves were both damaged and saliva was added. The hypothesis tested was ‘There will be an increase herbiovry on damaged and saliva treated plants compared to saliva and no damaged or damage and no saliva treated plants. The leaves were photographed before treatment and after, caterpillars were added to the leaves under controlled conditions. The area of caterpillar damage was measured. The leaves with saliva showed increased herbivory both with or without damage from holes. (120 words)
The competition between herbivores and plants has been a consistent part of plant ecology throughout history.  Plants could either be eaten, become extinct or adapt by developing strategies to protect themselves from herbivores and pathogens. Many of the adaptations have been observed but with the potential impact of climate change observations and understanding the mechanisms that protect plants from stresses such as herbivores must be better understood. 
Understanding signs of what makes a plant more vulnerable to herbivory will become very useful and can take on a large range of variation. Volatile organic compounds are emitted after an herbivore or pathogen attack in order to attract the enemies of the attacker so the plant will be better protected.  “Chemical and biological processes dictate an individual organism's ability to recognize and respond to other organisms. A small but growing body of evidence suggests that plants may be capable of recognizing and responding to neighboring plants in a species specific fashion.”  The most common way of interplant communication is receiving a volatile cue rom a neighboring plant that has been damaged. 
Research has added herbivores to a system in order to observe the results. Stress has been induced on plants and other species by adding a caterpillar, applying methyl jasmonate or causing mechanical damage.  Research has proven that the adaptations include abilities to warn neighbours of herbiovry. In fact plants have an ability for enhanced resistance after initial (herbivory) attack nearly universal in plants.”  Usually the “plants sense a volatile cue from a damaged neighbour and induce defensive metabolites, sensitivity to future damage, or anatomical structures in order to defend themselves from their herbivores.”  For example research under field conditions “have demonstrated repeatable increases in the herbivore resistance of plants growing downwind of damaged plants.”  Other adaptations are epigenetic. Epi is from the Greek and means above, over or outer; therefore epigenetic refers to cellular phenotypic (as well as gene expressions) that are initiated due to a cause outside of a plant’s DNA. 
Photography has been successfully used to better understand the interactions between plants and herbivory in research. [9, 10] Therefore this experiment included photography in the methodology in order to determine the effect of damage or saliva on cucumber leaves after the introduction of caterpillars to their environment.
Purpose and Summary
The purpose of this laboratory was to evaluate cucumber leaves before and after they were treated in order to determine how much more damage a caterpillar would cause due to the treatment. The cucumber leaves were photographed pre-treatment and post treatment. In that way the change in area after exposure to the caterpillar could be measured and compared.
There will be an increase herbiovry on damaged and saliva treated plants compared to saliva and no damaged or damage and no saliva treated plants.
- 48 cucumber plants, that is 12 per treatment
- 48 Zip lock bags, to enclose one leaf per plant
- 96 caterpillars, 2 per leaf (we have a maximum of 100)
- Pins, blue tack, cotton buds and elastic bands
Briefly the methodology used was to photograph the cucumber leaves before and after treatment in order to measure the area of the damage. The cucumber leaves were photographed pre-treatment and post treatment. Leaves were damaged by poking holes in them with a hole-punch. Saliva from three people was used to place on leaves during the appropriate treatment. Pearl, Katelyn and Claudius donated saliva after their names were chosen from a container at random by the lecturer, Clare. The leaves were photographed prior to treatment and photographed again post treatment in order to assess the damage.
(Week 7, Monday, 2 pm) The leaves were photographed prior to treatment. The treatment consisted of damaging some of the leaves with a hole-punch and putting saliva on leaves on them.
(Week 7, Wednesday, 12-2 pm) For each treated leaf per plant one zip lock bag was used. The treated leaves were handled very carefully. A cotton bud was placed inside each enclosed treated leaf to make sure the plastic would not collapse because that would make the treated part of the leaf inaccessible to the caterpillar. Two caterpillars were added to the treated leaves. The blue tack (zip lock device) was closed and the caterpillars were not allowed to escape.
(Week 7, Friday, 2-5 pm) The leaves were photographed so the amount of area eaten and/or damaged by the caterpillars could be measure. Download and learn how to use Image J.
(Week 8) Set up for photographing the leaves was done. Image J was used to analyse the photos. Each person was allocated an equal number of photos to analyse.
- +Damaged –Saliva which is referred to in the data as 1.00 Damage/ .00 Saliva
- -Damaged +Saliva which is referred to in the data as .00 Damage / 1.00 Saliva
- +Damaged +Saliva which is referred to in the data as 1.00 Damage/ 1.00 Saliva
- -Damaged –Saliva which is referred to in the data as .00 Damage/ .00 Saliva because it is the control group
- (-) means without; in other words no holes have been punched and/or no saliva has been applied.
- (+) means with; in other words holes have been punched and/or saliva has been applied.
Data variable / recorded
The data variable (in other words the data recorded is the amount of area that was damaged by the caterpillar. The area was measured by evaluating photographs of the leaves post treatments and after the caterpillars had been exposed to the leaves.
The following statistical analysis was done because these are appropriate for biological studies. The results of the statistical analysis are discussed in the next session.
- The Univariate Analysis of Variance In general the ANalysis Of VAriance (ANOVA) refers to a group of statistical models that help a researcher understand the amount of difference among groups of data and between groups of data. This was used because the differences in the means of the four different treatments needed to be calculated. A univariate (one factor) model was used because only one variance (treatment) was evaluated.
- Levene’s Test of Equality of Error Variances is sometimes referred to as a test of the homogeneity of the variances. Levene’s test assumes that the variances between groups are equal. Degrees of freedom were evaluated to understand whether the null hypothesis (assuming equality of variances) could be accepted or rejected. If the null hypothesis is rejected the source of variances should be evaluated.
- Tests of Between-Subjects Effects that were calculated using ANOVA included Type III Sum of Squares, degrees of freedom (df), the mean square, the F-statistic (F) and Significance (Sig).
- The Type III Sum of Squares provides R squared values which help define if the statistical model compares to the measured data. The R Squared values which show a very good fit are equal to 0.99 or 0.98.
- Degrees of freedom help assess variance by comparing 1 group to the number of other groups that could have variance. So usually, for example, 10 samples have a degree of freedom of 10 - 1 = 9.
- The Mean Square is used to prove or to reject the null hypothesis by comparing the means of each treatment to the means of the other treatments.
- The F-statistic should help find the best-fit datasets of a population.
- The significance is very helpful in evaluating the success of an experiment because the statistic helps identify the degree to which external variances (or errors) affected the results as opposed to random errors which cannot be controlled.
- Custom Hypothesis Tests will indicate non-random errors.
- Estimated Marginal Means takes into account the sample size affect on the error when it is calculated by ANOVA. When sample sizes are different then the mean is evaluated in proportion to the sample size for which it is the mean. ANOVA is a statistical model that helps evaluate data in one experiment when samples consist of different sizes.
Replicates were performed for each sample in order to measure reliability and therefore the variance and errors were reported in this section and discussed in the next section.
The dependent variable used throughout the statistical analysis was the Area of leaf damaged by the caterpillar. The univariate analysis of variance was conducted between the subject factors presented in Table 1; in other words (a) between the damaged leaf and the non-damaged leaf and (b) between the saliva and without saliva samples.
The key to the notation for treatments in Table 1 is
- (Group 1) 1.00 Damage/ .00 Saliva,
- (Group 2) 00 Damage / 1.00 Saliva,
- (Group 3) 1.00 Damage/ 1.00 Saliva, and
- (Control Group) .00 Damage / .00 Saliva.
The mean area of the experiment of the control group (no damage, no saliva) is 3.71. The mean area for no damaged leaves but salvia was applied equals 5.26. Whereas the mean area for the damaged leaves without saliva mean was 5.13. The mean area for leaves having both damage and saliva was 5.63. The standard deviations were large for each of the four groups. For the control group the standard deviation is almost 4 times the value of the mean. For Group 2 the standard deviation was approximately 3 times the mean. The value of the standard deviation for the mean was the least for Group 2 but it was close to twice the amount of the mean.
The Levene’s test of error variances was calculated to test the null hypotheses. The null hypothesis tested was that the error variance of the area damaged by the caterpillars was equal in all the groups. The F-statistic (F) is 0.641 showing three degrees of freedom (df) and it is significant at the 0.593 level. The null hypothesis can usually be rejected when the F-statistic is than 0.05.
Five statistical tests were calculated to determine the between subjects effects and the results are listed in Table 3. Note that the corrected model reported F equal to 0.045. The percentages for significance were approximately 2% for the intercept, 82% for Group 1, 80% for Group 2 and 90% for Group 3. R-Squared before adjustment was reported to be equal to 0.003 and after adjustment to be equal to -0.071.
The Damage Helmert Contrast results (K matrix) reported the lower bound of the 95% Confidence Interval for difference to be -8.93 and for the upper bound equal to 7.14 based on the dependent variable. The contrast test was calculated by the same five tests as the between-subjects. The F-statistic was equal to 0.051 and Significance was reported as equal to 0.823. The errors calculated for the Sum of the Squares, degrees of freedom and mean square tests were so large that the values are not useful.
Figure 1 Profile plots the Area* of the leaves treated with saliva and with no saliva
*Area of the leaf damaged by the caterpillar
As much as possible was done to reduce the amount of confounds in the experiment but like any qualitative methodology a great deal of variance is introduced into research. For example volunteers to donate the saliva used were chosen by putting everyone’s name in a hat and then drawing out three names. In that way the bias or unfairness in the selection for saliva samples was reduced. But the experiment is difficult to control because there were differences between the two caterpillars. Measuring the amount of saliva or the chemical components in the saliva from the three volunteers was not done.
The standard deviations reported in Table 1 are very large compared to the means. For Group 1 the mean ± Std. Deviation is 5.123.7±9.82; for Group 2 it is 5.26±16.28; for Group 3 it is 5.64±8.91; and for the control group it is 3.7±15.95. These standard deviations are too large so further analysis included adjustments by ANOVA were necessary to understand the data better.
The Levene’s test of error variances (testing the null hypotheses) reported the F-statistic to be equal to 0.641 with three degrees of freedom (df1) with Significance at 0.593. In order to reject the null hypothesis the rule of thumb is that F < 0.05. The Levine test suggests that the null hypothesis is true which would mean that the error variance for the area of the leaves damaged by the caterpillars was equal among the groups.
The value for Variance was calculated with five tests for between-subject effects considering the dependent variable again to be the area of the leaves damaged by the caterpillars. The corrected model reported the F-statistic equal to 0.045 with a Significance of 0.987(or approximately 99%). Therefore F indicates that the dependent variable is equal only 60% of the time meaning that 40% of the time they were not equal. The R-squared value was not used because the adjusted square only equalled -0.071 which was not a reasonable value but instead reflected external or internal error in the experimental procedure. The other tests were not useful due to the large value reported for error in each of the three remaining tests. (See table 3)
Interplant cues have been observed between plants to communicate herbivory disturbances. This experiment evaluated and compared three different treatments between each other and a control group. Leaves from cucumber plants were used and divided into four groups: the control group, the group damaged by holes being punched into them only a group treated with saliva only, and a group of leaves damaged with holes and treated with saliva. Replicates were performed for each sample in order to measure reliability and therefore the variance and errors were reported in this section and discussed in the next section. ANOVA was used to measure variance and errors between the samples and within the samples. The results were found to be only significant at the 60 %level. This is understandable because of the errors introduced when doing an experiment for the first time. Also many confounds were evident in the experiment although as much as possible was done to decrease the impact of the compounds. The experiment should be repeated so the procedure can be designed so each time it is preformed the process is consistent.
Finally evidence was observed supporting an assumption that the leaves with saliva were more attractive to the caterpillars so they ate a larger area of the leaves with saliva in both Group 2 (with zero hole punch damage but treated with saliva) and Group 3 (with both hole punch damage and treated with saliva). Therefore the hypothesis that increased herbiovry on damaged and saliva treated plants compared to saliva and no damaged or damage and no saliva treated plants- was not altogether correct. The saliva on the leaf led to increased herbivory regardless of the damage done to the leaves by the students.
 Jander G, Howe G. (2008) Plant Interactions with Arthropod Herbivores: State of the Field, Editorial Plant Physiology 146: 801-803
 He W-M, Li J-J, Peng P-H (2012) Simulated Warming Differentially Affects the Growth and Competitive Ability of Centaurea maculosa Populations from Home and Introduced Ranges. PLoS ONE 7(1): e31170. doi:10.1371/journal.pone.0031170
 Campos WG, Faria AP, Oliveira MGA, Santos HL. (2008). Induced response against herbivory by chemical information transfer between plants. Review, Brazilian Society of Plant Physiology, 20(4) : 257-266.
 Broz AK, Broeckling CD, De-la-Pena C, Lewis MR, Green E. Callaway RM, Sumner LW, Vivanco JM. (2010). Plant neighbour identity influences plant biochemistry and physiology related to defence. BMC Plant Biology, 10: 115. http://www.biomedcentral.com/1471-2229/10/115
 Pearse IS, Porensky LM, Yang LH, Stanton ML, Karban R, et al. (2012) Complex Consequences of Herbivory and Interplant Cues in Three Annual Plants. PLoS ONE 7(5): e38105. doi:10.1371/journal.pone.0038105
 Rasmann (et al., 2012: 854) Rasmann S, De Vox M., Casteel CL, Tian D., Halitschke R. et al. (2012) Herbivory in the Previous Generation Primes Plants for Enhanced Insect Resistance. Plant Physiology, 158: 854-863. www.plantphysiol.org/cgi/doi/10.1104/pp.111.1867831
 Baldwin IT, Kessler A, Halitschke. (2002) Volatile signalling in plant-plant-herbivore interactions: what is real? Current Opinion in Plant Biology, 2: 1-4
 Riddihough G. and Zahn L. (2010) What is epigenetics? Science 29 Special Issue, 330(6004: 611. doi:10.1126/science.330.6004.611
) Peiffer M Felton G.W. (2009) Do Caterpillars Secrete “Oral Secretions”? Journal of Chemical Ecology, 35: 326-335.
] MithoferA and Boland W. (2008) Recognition of Herbivory-Associated Molecular Patterns. Plant Physiology, 146: 825-831.