Challenges to Anticipate and Solve

  1. Challenge #1. Determining appropriate dependent variables. Students often have difficulty determining what dependent variables to measure. Time to emergence, body size at emergence, and emergence success can be measured in a reasonable time span. Students may suggest other offspring characteristics, such as lifespan, reproductive success, hatching rate, and sex ratio. Characters such as lifespan could be measured, but would add another two or more weeks to the experiment. Other dependent variables are appropriate, but difficult to measure (i.e., reproductive success and hatching rate). Finally, for other offspring traits like sex ratio, the predictions are not clear.
  2. Challenge #2. Identifying eggs on beans. Students sometimes have difficulty distinguishing between eggs on beans and frass. Eggs are larger and have a distinct dome shape when viewed with magnification. If students are shown examples of eggs on beans, they are better able to find them on their own. Typically, we do this by rotating through the room and showing individual students beans with eggs. Alternatively, the instructor could place small petri dishes with beans with eggs attached at each lab station.
  3. Challenge #3. Statistical comparisons. Students also have difficulty determining the appropriate statistical comparisons and then interpreting the results. After allowing the students to discuss the comparisons in groups, the instructor may want to review the possible comparisons and their interpretation. We have this discussion after the final data are collected. It could take place after the consensus experimental design is determined during the first lab period. However, given the time span between the initial lab period and the data analysis, students may have difficulty recalling the discussion if it takes place during the initial lab period.

Introducing the Experiment to Your Students

Prior to this exercise, we present the concept of adaptation to our students in lecture and distinguish adaptation from acclimation and phenotypic plasticity. We also discuss approaches for determining genetic differences among populations, including common garden and reciprocal transplant experiments.

Since this exercise is designed as a guided inquiry, the role of the instructor is to facilitate the discussion among members of each group and then the discussion of each group’s experiment in the class as a whole. We guide students to a consensus experiment that is carried out by the entire class. It is possible that individual groups could carry out their own experimental designs. In this case, each student will need to increase the number of replicates that they set up in order to have sufficient replication for statistical analysis. In our experience, most groups propose very similar experiments and therefore the process of coming to a consensus experiment for the class is more a confirmation of the ideas of each group rather than rejecting the ideas of most groups in preference to that of a single group.

When discussing the experimental design with the students, it may be helpful to review a little of the natural history of bean beetles and the materials that will be available to the students. Most of our bean beetle cultures have been reared on mung beans for many generations and then switched to either adzuki beans or blackeye peas (Vigna unguiculata) for several generations.

To determine if adaptation has occurred, students will want to conduct a reciprocal transplant experiment in which females from mung bean cultures and females from either adzuki bean or blackeye pea cultures are allowed to lay eggs on both types of beans. In effect, the experiment is a 2 x 2 factorial design with maternal host (mung or adzuki/BEP) and offspring host (mung or adzuki/BEP) as the two factors. Then, students will examine the life history traits of the resulting offspring.


Experiment Description

Consult the Laboratory Methods section of the Bean Beetle website (www.beanbeetles.org) for detailed information on growing cultures and handling techniques, as well as tips on identifying the sexes.

The experiment requires having dense cultures of bean beetles from which females can be isolated. Beetles should be from cultures reared on a natal host and from cultures switched to a new host several generations prior to the experiment. If new cultures are initiated approximately 2 months before the lab period, there will be sufficient time for two generations of beetles, which will result in dense cultures. When possible, we supply one culture of each type (natal and new host) to each group of students working in pairs. However, each culture should have sufficient beetles for use by multiple student groups. Sufficient cultures for one class section can be established in less than an hour. Once cultures are established, they do not need to be monitored or recultured until 2 months later. Cultures of bean beetles reared on different host types can be obtained from the authors.

Each student should set up a single replicate of each treatment combination of the reciprocal transplant experiment. To set up a replicate, a single female from a stock culture (either mung or adzuki/BEP) is added to a 35mm Petri dish with a monolayer of beans (either mung or adzuki/BEP). Oviposition will readily occur during a 48-hour period. Although most adult females in an active culture will have been inseminated, some females may have only recently emerged (and be infertile) and others are near the end of their adult life (and laid most of their eggs). Replication in the class will allow for failures in egg laying.

For offspring life history traits, one of the biggest confounding factors is the number of eggs laid on beans. If more than one egg is laid on a bean, then the larvae may compete for resources. As a result, only beans with single eggs should be used in tabulating data. Students may want to record the identity of the female that laid the egg to be able to consider differences among females in their analysis. However, data on female identity is not essential. Students can isolate beans of each species with single eggs into the wells of tissue culture plates or small Petri dishes. As the beetles emerge, students can record the offspring characteristics that they chose to measure. Accurate data on time to emergence and mass at emergence require that students check for emergence on a daily basis. As a result, measuring these life history traits may be feasible only in smaller, more advanced classes. The daily checks take between 15 – 30 minutes depending on the number of beetles that have emerged on that day. Students carry out these checks outside of class time. Emergence success could be determined on a single day (potentially as a 1 hour part of a longer lab period) after sufficient time for emergence (approximately 40 days). Therefore, emergence success is more tractable for larger classes.

Because the experiment is a 2 x 2 factorial design, the data are most appropriately analyzed with a multifactor ANOVA, assuming a normal distribution, which is commonly the case. If adaptation has occurred, we would expect a significant interaction effect between maternal host and offspring host, with offspring having higher fitness on the same host as their mother. In bean beetles, some life history traits differ between the sexes. Therefore, the analysis should be done for males and female separately. Students also could look at correlations between life history traits.


Questions for Further Thought

  1. Based on your results, have bean beetles adapted after a change in host species?

    Although students generally have little trouble in designing the appropriate experiment, they have more difficulty in interpreting the results in relation to adaptation. The instructor may need to help students understand what the main effects and the interaction effect are testing in the multifactor ANOVA.

  2. How might the results of your experiment been different if you had used a species of phytophagous insect that was specialized on its host plant to a greater degree? What if it were more of a generalist?

    The intent of this question is to get students to think about the fact that selection for adaptation should be stronger on specialists. However, at the extreme, you could not conduct this experiment with a specialist, because no offspring would be viable on an alternative host.

  3. Because bean beetles are an agricultural pest species, how could the results of your study be used to design an effective protocol for reducing the impact of bean beetles on stored beans?

    Beans could be stored as single species or mixed. Students could consider which method might be more effective in decreasing damage by bean beetles, if beetles can adapt to new hosts.

  4. If phytophagous insects are able to adapt rapidly to a new host, what might this suggest about the impact of these insects in monoculture versus polyculture agricultural systems?

    This question is similar to question 3, but asks students to think beyond bean beetles to other herbivorous insects.

  5. Many studies have examined whether female bean beetles exhibit a preference when given a choice of several host species on which to lay their eggs. Based on the literature on host preference, how might host preference influence adaptation of bean beetles to specific host species? Contrast this with a species in which females do not exhibit a preference for host species.

    This question allows students to examine the literature on host preference in bean beetles. Students can be sent to literature databases or to the bibliography found on the Bean Beetle website (www.beanbeetles.org/biblio/). Females do tend to show a preference for bean species from which their offspring can successfully emerge. They also seem to prefer their natal bean species. As a result, female preference might decrease the strength of selection due to a host switch as females would tend to avoid novel hosts. Female behavior may mediate selection due to novel hosts if both natal and novel hosts are available.

Assessment of Student Learning Outcomes

Student assessment through scientific writing is standard practice, and therefore does not warrant additional comments here. However, in addition to the scientific writing, in one year in which this experiment was tested at Emory University, students were given a pre-test/post-test to determine whether conducting the experiment changed their understanding of the scientific method. The pre-test was administered immediately before we discussed the experiment. The post-test was given the following week before any additional experiments were done in lab.

Sixteen of the 17 students in the course completed both the pre-test and the post-test.  We found a marginally significant increase in understanding of the scientific method from the pre-test to the post-test (Wilcoxon signed rank test, Z = -1.9, P = 0.06). The small sample size in the class may have contributed to the lack of clear significance. Of the 16 students who took both the pre-test and the post-test, 50% of them exhibited an increase in understanding of the scientific method and 31% (5 of 16) showed no change. In general, our results suggest that the experiment was effective in increasing students’ understanding of the scientific method.

Assessment on scientific method with scoring rubric

Below is the pre-test/post-test that was used for assessing student understanding of the scientific method. Also, we have included a rubric for scoring the test.

Ecology Laboratory

Name: ___________________________



Diagnostic Test (Answers are shown in italics.  Total possible score is 10 points.)

This is a non-graded test that will assess your understanding of experimental methods.Completion of this test is part of a research study so your responses are voluntary. Please print your name on this test so we can determine how your understanding changes as a result of your work in this course. Your instructor will be the only person who sees your test results and individual results will not be released.

  1. When conducting an experiment, what is the purpose of a control treatment?

    Total Points Possible = 3

    “Treatment with no manipulation” (1 point)
    “Treatment for comparison to a manipulation treatment” (2 points)

    Partial Credit

    “Normal conditions treatment” (0.5 points)

  2. What is the “null hypothesis” in an experimental study?

    Total Points Possible = 2

    “The hypothesis of no effect or no response to an experimental manipulation” (2 points)

    Partial Credit

    “The hypothesis that excludes or negates something from being true” (1 point)
    “The opposite of the hypothesis you wish to prove” (0.5 points)

  3. What is the purpose of replicating a given treatment in an experiment?

    Total Points Possible = 2

    “Replication ensures an adequate sample size for a given treatment” (1 point)
    “Replication improves the accuracy and reliability of the results” (1 point)

    Partial Credit

    “Replication is performed to reduce errors” (0.5 points)

  4. What is the function of conducting a statistical test on the results of an experiment?

    Total Points Possible = 3

    “Statistical tests permit us to establish the probability of the observed results occurring by chance alone” (1 point)
    “Statistical tests permit us to reject the null hypothesis if differences occur by chance alone less than 5% of the time” (2 points)

    Partial Credit

    “Statistical tests show whether there is a significant difference between the control and experimental treatment results” (1 point)

Evaluation of the Lab Activity

Students in the ecology course at Emory University (Fall 2006) were asked to rank each experiment they completed on a ten point scale with respect to how useful each experiment was in reinforcing their knowledge and understanding of the subjects covered in the lecture portion of the course.  A score of 10 meant that the study was the most useful.  In addition, the students were asked which studies were the most and least enjoyable and which study best increased their understanding of the scientific method. The students completed seven experiments (including the current experiment) plus an independent project during the semester.

Thirteen of the 17 students in the course completed the course evaluation.  The average ranking of the current experiment was 8, while the average ranking of all the experiments combined was 7.2. Twelve of the 13 students ranked this experiment higher than the average of all the experiments.  Furthermore, five of the students gave the experiment the highest ranking. Only one student considered it the most enjoyable experiment; however, no students considered it the least enjoyable experiment.  More than half of the students (7 out of 13) thought that this experiment was the one that best increased their understanding of the scientific method.  Of the remainder, four students thought that the independent projects were most important to their understanding of the scientific method.  Therefore, of the planned experiments conducted, this experiment was the most important at increasing student understanding of the scientific method, as self-reported by the students.

As noted above in “Tools for Assessment of Student Learning Outcomes,” students could be evaluated before completion of the experiment. For example, the proposed experimental designs that students bring to class could be collected and evaluated. In addition, after the class has discussed experimental approaches, students could be asked to write a minute paper explaining how reciprocal transplant experiments can be used to test for adaptation. Both of these types of formative evaluation would allow an instructor to determine how well students understand the concept of adaptation and how to test it, which may influence how instructors discuss the results of the experiment with their class at a later date.


Translating the Activity to Other Institutional Scales or Locations

(1) Translating this experiment to larger scales if you teach at a smaller school and vice versa,

Because bean beetles are easy to rear in large numbers and the materials for the experiment are inexpensive, this experiment could easily be scaled up to larger institutions. The main change would be to consider only emergence success rather than other life history traits, as evaluating emergence success does not require students to examine beans for beetle emergence on a daily basis.

(2) Using this lab in different regions of the country or world, in different seasons, or using different study species or systems,

Because the experiment is a laboratory experiment, it could be used in most regions. The only restrictions would be those associated with the shipment of bean beetles. However, at present, we (the authors) legally can ship bean beetles to all 48 states in the continental United States, the District of Columbia, Alaska, and Puerto Rico.

This experiment could be conducted with other phytophagous insects. Other phytophagous insects, such as tobacco hornworms, Manduca sexta, and the Brassica butterfly, Pieris rapae, that are easily reared in the laboratory and can be induced to use a variety of host plants could be used in this experiment. Both are commercially available. However, keep in mind that populations would have to be reared on separate host types for multiple generations before beginning the experiment.

(3) Using this activity to teach ecology to students with physical or other disabilities, and

Students with physical disabilities may have difficulties with this experiment, as it requires the manipulation of small insects. In addition, distinguishing between the sexes and identifying eggs on beans would be difficult for students with visual disabilities.

(4) Using this activity to teach ecology in pre-college settings (K-12).

Although this experiment has not been tested in pre-college settings, bean beetles have been used for other experiments in high school biology courses.