Introduction (written for students)

The interactions between herbivores and their host plants are often complex, involving plant chemical and physical defenses, herbivore foraging behaviors, and many other factors. Most plants are attacked by several to many different types of herbivores. Each herbivore may feed in a different manner or on different plant tissues, causing different types of feeding damage. One of the more unique plant-herbivore interactions is the formation of galls. Galls are modified plant tissue stimulated by the oviposition and feeding activities of certain insects and spider mites. They result when the cells around the damaged area grow larger or divide more often than normal cells. As the insect feeds on the plant, it becomes surrounded by this abnormal plant growth. The insect continues to feed from within the gall, which protects it from many (but not all!) of its natural enemies. Other organisms, including viruses, bacteria, fungi, nematodes, and mites, may induce plant galls, but insects are the most common gall formers.

Galls can be used to test a number of interesting ecological and evolutionary questions about plant-herbivore interactions. The hypothesis that host plant quality affects herbivore densities and community structure was tested by Fritz et al. (1987b). As predicted, both densities of individual sawfly species and the relative abundances of these species varied among clones of arroyo willow. Additional data showed that shoot size is an important plant trait affecting gall densities: larger shoots have higher sawfly densities (Fritz et al. 1987a.) Since galls act as nutrient sinks (Nakamura, et al. 2003, Price et al. 1987), larger galls should provide more nutrients and therefore increase the success rate of the galling insect. Investigating the relationship between gall size and gall success (e.g., percent emergence) would provide a test of the generally supported hypothesis that plant galls are adaptive for the galling insect. The mechanisms through which habitat affects the density of galling insects were investigated by Fernandes and Price (1992). Lower rates of parasitism and fungal attack of galls may be at least partly responsible for higher gall densities in xeric (dry) environments compared with mesic (moist) habitats.

Willow trees (genus Salix) are attacked by several gall-forming herbivores. Gall midges form galls on buds, and sawflies form galls on leaves and shoots. Studies for this lab will be conducted at Engelhorn Pond on the Central Washington University campus where many of the willows have leaves with elongate, reddish capsules emerging from the leaf surface (Image of Gall – upper surface; Image of Gall – lower surface). These are caused by sawflies of the genus Pontania. Sawflies are not actually flies but relatives of bees and wasps (Order Hymenoptera). Adult females oviposit (lay eggs) into the leaf tissue. The egg hatches into a larva, which feeds on the leaf tissue while enclosed in the gall. When the larva has completed its development, it chews a hole in the gall and departs. See weblink in References to “Forest and Timber Insects in New Zealand” for pictures of egg, larva, pupa, and adult Pontania.

Willows are also eaten by a variety of free-feeding invertebrate herbivores. Lace bugs suck sap from leaves, spider mites chew leaves, and flea weevils chew on leaves and new shoots. You may find other insects feeding on the willows at the study site.

During this lab, you will (collectively) test a number of hypotheses about the gall-forming sawflies on willows. Particular questions chosen by student groups, in consultation with the instructor, may include:

• Do some trees have more galls than others?
• Do female sawflies avoid ovipositing on leaves that already have galls?
• Are galls on leaves with several other galls less successful that single galls?
• Do chewing herbivores avoid leaves with galls?

In the process, perhaps you will become expert cecidologists (students of plant galls)!

______________________________________________________________

Materials and Methods (written for faculty)

Study Site(s):

We are fortunate in being able to walk to our study site, a very small reserve (<1 ha) across the street from the Biology Building at the western edge of Central Washington University’s campus. Engelhorn Pond formed as a borrow pit in the 1920’s when gravel was excavated for use in the construction of the Interstate 90. The pit filled with water from runoff and groundwater and vegetation (including willows) colonized the site. Although very small, and nearly surrounded by university buildings, this site offers an urban refuge for ducks and other wildlife. As a wetland it garnered the attention of The Nature Conservancy, which purchased the site and donated it to the Biology Department. The pond is the dominant feature of the site and is surrounded by willow trees (Pacific Willow, Salix lasiandra). Gall densities vary from year to year, but usually galls are fairly abundant on the leaves. Numerous other sites are possible inside the city limits where willows grow along streams and irrigation canals. At your location, any site where willows grow and you can easily find galls would be appropriate. You might also consider sites with other plant species that harbor galls (such as poplars, goldenrod, maple trees).

Overview of Data Collection and Analysis Methods:

Introduction to the Study System

When we first get to the field site, I show students galls on the willows. Have a few students carefully open the galls (with a pocket knife or thumbnails) to find a sawfly larva. You could collect a few larvae in advance of the lab, and let students view them under a microscope during the lab introduction, or take a portable dissecting scope or simple hand lens with you to the field.

Hypothesis 1: all teams

• Background: Plants vary in their chemical and physical traits as a result of differences in genetic makeup as well as varying environmental factors such as soil moisture and nutrient levels, amount of sunshine, ambient temperature, and exposure to plant pathogens and herbivores. Therefore, one might expect to find differences among plants in the number of herbivores that feed on them (either because of differences in susceptibility to herbivory or differences in nutritional quality that attracts herbivores).
• Hypothesis: Level of herbivory by sawflies varies among willow trees (i.e., some trees have more galls than others).
• Method: Randomly sample 20 leaves from each of 2 trees. Record the number of galls on each leaf. Do not include aborted galls (substantially smaller than the “typical” galls).
• Analysis: For each tree, calculate the mean, and standard deviation of the mean, of the number of galls per leaf. Keep data for each individual tree separate. Then add data for the individual trees you sampled to a class list of means and standard deviations for each tree sampled. Compare the mean and standard deviation of number of galls/leaf among the various trees. Do trees vary significantly in the extent of herbivory by sawflies?

Additional Hypotheses

In addition, each team will choose one of the other hypotheses described below, or can design a new hypothesis, to test. Read through these hypotheses to see which interests you most, or discuss ideas with your instructor. You may come up with interesting hypotheses based on your initial observations of the trees and galls, or even by reading the titles of some of the journal articles in the References section. Alternatively, your instructor may assign a hypothesis to each team to be sure each hypothesis gets tested by your class. See Appendix [PDF] (211 KB) for suggestions on random sampling, descriptive statistics, and statistical tests. When you “sample” leaves, please avoid removing them from the branch so that other teams may sample the same plants and we leave the willows as undisturbed as possible.

Hypothesis 2

• Background: If sawflies compete for plant nutrients (sugars, proteins, lipids), there might be selective pressure for the evolution of behaviors to avoid competition. Females might avoid ovipositing on leaves that already have galls, so that leaves with one gall are more common than leaves with multiple galls. On the other hand, if some leaves are better food sources than others, galls may be clumped on the good leaves. We can test these ideas against the alternative hypothesis that females oviposit independently of other oviposition events.
• Hypothesis: Females select leaves for oviposition independently of whether the leaf has other galls.
• Method: Count the number of galls per leaf on 20 randomly chosen leaves on each of 5 trees. Fill in the following table (“# leaves” column) with tick marks as you observe each leaf. Combine your data with the class data from question #1 to get the total number observed (f) for each row in the table.



# Galls/leaf	# Leaves	Class Data from Q #1	Total = Observed (f)	Expected (ˆf)	Deviation from expected ( |f-ˆf| )
0
1
2
3
4
5
6
7+
Total	100		n=



• Analysis: This test is sensitive to small expected frequencies in a cell (ˆf < 5), so you may need to group together cells (e.g., 5-6 galls/leaf) to make sure ˆf ≥ 5. Then see how closely f matches ˆf. Are there more or fewer leaves with just one gall than expected? Are the galls distributed independently of one another on leaves? See Appendix [PDF] (211 KB).

Hypothesis 3

• Background: When sawflies form galls on willow leaves, plant nutrients become more concentrated in the galls. If several galls are formed on a leaf, the sawflies might compete for these plant nutrients, and each sawfly might get fewer nutrients than if there were no other galls on the same leaf.
• Hypothesis: Galls on leaves with competing galls are less successful than galls on leaves with only one gall.
• Method: Determine for 50 single galls (only 1 gall/leaf) and 50 "multiple" galls (share their leaf with other galls) whether the sawfly was successful (an exit hole indicates success) or not (no exit hole). Enter the number of galls that fit into each category in the following table (you can use tick marks):



Number of galls/leaf	Successful Emergence?
	Yes	No	Total
Single			50
Multiple			50
Total			100



• Analysis: G-test of independence. See Appendix [PDF] (211 KB). Are single galls more successful?

Hypothesis 4

• Background: Willows have leaf-chewing herbivores as well as gall-formers. Do these different types of herbivores feed on the same leaves (because the selected leaves have low chemical or high nutrient concentrations), avoid each other (to avoid competition), or feed independently of each other?
• Hypothesis: Leaves with galls are more likely to have leaf damage from chewing herbivores than are leaves without galls (leaf chewers do not select leaves independently of galls).
• Method: Randomly sample 5 leaves with galls and 5 leaves without galls from each of 10 trees. Record whether each leaf has damage from chewing insects.



Galls present?	Chewing Damage Present?
	Yes	No	Total
Single			50
Multiple			50
Total			100



• Analysis: G-test of independence. See Appendix [PDF] (211 KB). Is gall presence independent of chewing damage?

Hypothesis 5

• Background: Construct your own hypothesis to test. Discuss your ideas with the instructor before you begin; your instructor can give you advice on methods and analysis.

______________________________________________________________

Questions for Further Thought and Discussion

Conclusions about hypothesis 1:

• What did we find as a class?
• Do trees vary much in the level of herbivory by sawflies? Justify your answer by referring to the class data.
• Give at least one explanation for our result.

For additional hypotheses tested by individual teams:

• Did you accept or reject your null hypothesis?
• How did the results of your statistical test lead to this conclusion?
• What did you conclude about sawfly herbivory on willows based on your data?
• Give at least one ecological explanation for the result you obtained, and the possible ecological and evolutionary consequences of that result to the plant and to the animal.
• What have other investigators concluded about this hypothesis? Use your library’s research databases or Google Scholar (http://scholar.google.com) to find and read at least 2 journal articles to support your statement.

______________________________________________________________

References and Links

• Boecklen; W. J., P. W. Price, and S. Mopper. 1990. Sex and drugs and herbivores: sex-biased herbivory in arroyo willow (Salix lasiolepis). Ecology 71:581-588.
• Boecklen, W. J., S. Mopper, and P. W. Price. 1991. Size and shape analysis of mineral resources in arroyo willow and their relation to sawfly densities. Ecological Research 6:317-331.
• Clancy, K. M., P. W. Price, and C. M Sacchi. 1993. Is leaf size important for a leaf-galling sawfly (Hymenoptera: Tenthredinidae)? Environmental Entomology 22:116-126.
• Clancy, K. M., and P. W. Price. 1989. Effect of plant resistance, competition, and enemies on a leaf-galling sawfly (Hymenoptera: Tenthredinidae). Environmental Entomology 18:284-290.
• Clancy, K. M., P. W. Price and T. P. Craig. 1986. Life history and natural enemies of an undescribed sawfly near Pontania pacifica (Hymenoptera: Tenthredinidae) that forms leaf galls on the arroyo willow, Salix lasiolepis. Annals of the Entomological Society of America 79:884-892.
• Craig, T. P., J. K. Itami, and P. W. Price. 1990. The window of vulnerability of a shoot-galling sawfly to attack by a parasitoid. Ecology 71:1471-1482.
• Craig, T. P., J. K. Itami, and P. W. Price. 1989. A strong relationship between oviposition preference and larval performance in a shoot-galling sawfly. Ecology 70:1691-1699.
• Craig, T. P., P. W. Price, J. K. Itami. 1986. Resource regulation by a stem-galling sawfly on the arroyo willow. Ecology 67:419-425.
• Felt, E. P. 1965. Plant galls and gall makers. Hafner Pub. Co., New York, New York. (SB767 F4 1965)
• Fernandes, G. W., and P. W. Price. 1992. The adaptive significance of insect gall distribution: survivorship of species in xeric and mesic habitats. Oecologia 90:14-20
• Fritz, R. S., W. S. Gaud, C. F. Sacchi, and P. W. Price. 1987a. Patterns of intra- and interspecific association of gall-forming sawflies in relation to shoot size on their willow host plant. Oecologia 73: 159-169.
• Fritz, R. S., W. S. Gaud, C. F. Sacchi, and P. W. Price. 1987b. Variation in herbivore density among host plants and its consequences for community structure. Oecologia 72: 577-588.
• Hartley, S. E. 1998. The chemical composition of plant galls: are levels of nutrients and secondary compounds controlled by the gall-former? Oecologia 113:492-501.
• Johnson, W. T., and Lyon, H. H. 1988. Insects that feed on trees and shrubs. Comstock Publ. Assoc., Ithaca, New York, USA.
• Kause, A. E., E. Haukioja, and S. Hanhimaki. 1999. Phenotypic plasticity in foraging behavior of sawfly larvae. Ecology 80:1230-1241.
• Kokkonen, K. 2000. Mixed significance of plant vigor: two species of galling Pontania in a hybridizing willow complex. Oikos 90:97-106.
• Kopelke, J. P. 2003. Natural enemies of gall-forming sawflies on willows (Salix spp.) (Hymenoptera : Tenthredinidae: Euura, Phyllocolpa, Pontania). Entomologia Generalis 26: 277-312.
• Larew, H., and J. Capizzi. 1983. Common insect and mite galls of the Pacific Northwest. Oregon State University Press, Corvallis, Oregon, USA.
• Mani, M. S. 1964. Ecology of plant galls. W. Junk, The Hague, Netherlands.
• Nakamura, M., Y. Miyamoto, and T. Ohgushi. 2003. Gall initiation enhances the availability of food resources for herbivorous insects. Functional Ecology 17:851-857.
• Nyman, T., H. Roininen, and A. Widmer. 1999. Evolution of gall morphology and host-plant relationships in willow-feeding sawflies (Hymenoptera: Tenthredinidae). Evolution 64:526-533.
• Price, P. W., and K. M. Clancy. 1986. Interactions among three trophic levels: gall size and parasitoid attack. Ecology 67:1593-1600.
• Price, P. W., G. W. Fernandes, and G. L. Waring. 1987. Adaptive nature of insect galls. Environmental Entomology 16:15-24.
• Price, P.W., T. Ohgushi, H. Roininen, M. Ishihara, T. P. Craig, J. Tahvanainen, and S. M. Ferrier. 2004. Release of phylogenetic constraints through low resource heterogeneity: the case of gall-inducing sawflies. Ecological Entomology 29: 467-481.
• Roininen, H. and J. Tahvanainen. 1989. Host selection and larval performance of two willow-feeding sawflies. Ecology 70:129-136.
• Wagner, M. R., and K. F. Raffa, eds. 1993. Sawfly life history adaptations to woody plants. Academic Press, San Diego, California. (SB945.S3 S29 1993)
• Whitham, T. G. 1983. Host manipulation of parasites: within-plant variation as a defense against rapidly evolving pests. Pages 15-41, in R. F. Denno and M. S. McClure, editors. Variable plants and herbivores in natural and managed systems. Academic Press, New York, New York, USA.
• Woodman, R. L., and P. W. Price 1992. Differential larval predation by ants can influence willow sawfly community structure. Ecology 73:1028-1037.

Web Links

• The Solidago Eurosta Gall Homepage: A Resource for Teaching and Research. Useful information, photographs, illustrations, clips of an educational video (Goldenrod and the Gallfly), maintained by Dr. Warren Abrahamson.
  http://www.facstaff.bucknell.edu/abrahmsn/solidago/main.html
• Forest and Timber Insects in New Zealand, No. 45: Willow Gall Sawfly. Nice photographs of Pontania galls on willow leaves, and egg, larva, pupa, and adult stages of sawfly.
  http://www.forestresearch.co.nz/PDF/Ent45Pontaniaproxima.pdf
• USDA, Agricultural Research Service: List of Plant Galls Web Sites
  http://www.wcrl.ars.usda.gov/cec/teaching/galls.htm
• North Carolina State University: Galls and Gall Makers.
  http://www.cals.ncsu.edu/course/ent591k/galls.html
• Colorado State University Cooperative Extension: Insect and Mite Galls
  http://www.ext.colostate.edu/pubs/insect/05557.html

______________________________________________________________

Tools for Assessment of Student Learning Outcomes

Informal Assessment

While students are deciding which hypothesis to test, I wander among groups and ask them some simple questions to assess whether they are grasping the main concepts and hypotheses.

• Does this hypothesis make sense?
• What other factors might influence the distribution of galls besides the one hypothesized?
• What factors should you try to control in your study?

While they are collecting and analyzing data, I informally walk among teams of students and ask them simple questions such as:

• Do you have enough data? If not, should you increase your sample size? How would you know if your sample size is sufficient to answer the question?
• What did your statistical tests tell you?
• How confident are you in your data and conclusions? Explain why.

We discuss the results of hypothesis 1 as a class, after putting the means and standard deviations on the board. I ask students whether trees vary in the extent of herbivory by galling sawflies.

Formal Assessment

I assess student learning primarily by requiring a formal oral report (similar in style to paper presentations at ESA meetings). I give students a guideline (below) for what material they should include in the talk, and give them the grade sheet I use to grade their oral report (see Oral Report Grade Sheet). Assessment of their data analysis is via written data summary and analysis they hand in at the time of the oral report. Students also evaluate oral presentations by other teams, providing me an opportunity to see if they understand what other teams found.

ORAL PRESENTATION

Your team will present a short, concise (5-10 min.), well organized oral presentation (in PowerPoint) based on the additional hypothesis you tested.

Use the following outline to prepare for your talk:

• Introduction/Background — sufficient information from the literature to lead up to your hypothesis
• Hypothesis — stated clearly; give both null and alternative hypothesis; what predictions follow logically from this hypothesis?
• Methods of data collection — overview of how you collected data to test your hypothesis, including sample size (number of leaves, trees, galls, as appropriate), how you selected units to sample; what you recorded
• Statistical test — which test did you use, and what does it test? (e.g., t-test tests for difference in mean between 2 groups)
• Results — spend some time thinking about the best way to present your data. Graphs are typically better than tables. Make sure the axes are labeled with large lettering that can be easily read. You want your audience to be able to quickly see any patterns in your data. Give results of statistical tests, and whether you accepted or rejected your null hypothesis.
• Interpretation — what do your data mean? Did they support the hypothesis or not? Very briefly, how confident are you in your data and conclusions? (large enough sample size? Errors minimized?) How do your results square with what is known about herbivory/galls from the literature? Cite at least 2 journal articles that you have read.

Your grade will be based on the criteria listed in the Oral Report Grade Sheet.