Evolution is a fundamental principle in biology, yet students, teachers, and the public at large all too often misunderstand the way it works. I introduce a hands-on exercise that emphasizes tree-thinking and phylogenies to organize biodiversity. During the activity, students observe and investigate the patterns and processes of macroevolution by first building unique specimens through gradual, stepwise changes in characters. They then switch specimens with another group and, by observing shared characters, hypothesize the evolutionary relationships of the specimens by drawing phylogenies. The exercise has been used for several years, and pretest–posttest results confirm that it significantly improves student understanding of macroevolution and phylogenetics.

Introduction

In On the Origin of Species, Darwin introduced the idea that all species are related through common descent by using a branching evolutionary tree as a model (Darwin, 1859). This concept was revolutionary, and yet it was almost instantaneously explained and expanded upon, linking the fields of taxonomy, embryology, and biogeography (Mayr, 2001). The modern field of phylogenetics seeks to understand and map this branching pattern of evolutionary history for all species, using diagrams called “evolutionary trees” or “phylogenies” (Edwards & Cavalli-Sforza, 1964; Camin & Sokal, 1965). Phylogenies provide an efficient way to examine how life evolved and to organize its diversity. They are increasingly used across fields of biology and are becoming much more common in textbooks (Cately & Novick, 2008). However, both macroevolution and phylogenetic trees prove difficult for students to grasp fully, and little time is dedicated to teaching students how to read phylogenies (Baum et al., 2005; Cately & Novick, 2008; Sandvik, 2008; Phillips, 2012). Thus, alternative concepts and poor acceptance of macroevolution continue to invade the classroom among both students and teachers (Meir et al., 2007; Robbins & Roy, 2007; Scott, 2009; Allmon, 2011). Perhaps as a consequence, acceptance of evolution in the public at large has fallen to ~40% in the United States (Miller et al., 2006), and there is pervasive controversy over whether or not common descent should be taught in schools (e.g., Kitzmiller et al. v. Dover [2005]; Selman v. Cobb County School District [2006]). Therefore, educators should strive to make understanding evolutionary biology, including phylogenetics, an accessible and fun learning experience for students, so that they may better grasp these concepts, better interpret the figures in their books, and – hopefully – increase their acceptance of evolution as fact.

Objectives

The goals of this exercise are for students to

  • recognize that large changes in phenotype can occur via accumulation of small changes over generations;

  • know the scientific jargon of phylogenetics and taxonomy, such as nodes, branches, monophyly, synapomorphy, sister species, convergence, and loss;

  • be able to interpret phylogenies; and

  • be able to create phylogenies based on shared characters.

Explanation of the Exercise

In this exercise, students use specimens with unique characters to recreate evolutionary histories, much like the “Caminalcules” – imaginary creatures with a complete evolutionary history, including a fossil record – designed by Joseph H. Camin to test phylogenetic principles (Sokal, 1983a, b) and since used extensively to teach phylogenetics (e.g., Gendron, 2000). In this exercise, students themselves design and build the specimens to be used by other groups. Therefore, students not only learn how to draw a phylogeny based on shared characters, but they also observe how phenotypic diversity evolves through small, gradual changes over time (i.e., macroevolutionary processes).

Building a History

After a short lecture on the basics of phylogenetics, students work in groups (usually four at one laboratory bench) to create an evolutionary progression of specimens using modeling clay and arts-and-crafts supplies such as beads, pipe cleaners, wood shapes, feathers, googly eyes, and so on. Each group starts with a small clay sphere that represents the common ancestor of all future taxa (Figure 1, stratum 1). This ancestor gives rise to two lineages, resulting in two new specimens built by the students (Figure 1, stratum 2). Each specimen is built separately and kept. I have deliberately omitted the amount of time between strata, but it should be stressed that this level of morphological change usually takes many, many generations, although speciation can happen in as short a time as two generations (Coyne & Orr, 2004). These new specimens inherit the basic body form and characters of their ancestor, but students add three or four newly evolved characters to each. The process repeats four times, with one “speciation” event per stratum (Figure 1, strata 2–5). Each new specimen inherits all, or at least most, of the traits from the ancestors in its lineage, while gaining three or four new traits. (Students are often reluctant to add so many characters, but this is necessary to make accurate hypotheses about their relationships later. For the same reason, convergence and loss should be kept to a minimum, without underemphasizing the importance of these phenomena in nature.) The result is a complete evolutionary sequence consisting of five strata with 10 ancestral specimens, leading to five operational taxonomic units (OTUs; labeled A–E in Figure 1). Each group of students must keep note of the relationships of their specimens and draw their phylogeny (Figure 2).

Once the evolutionary sequence is complete and its evolutionary history is mapped, the ancestors (strata 1–4) go extinct. In other words, students disassemble all the specimens except the five OTUs, which are given names that do not reveal their relationships. The names are written on an accompanying label, and the specimens are then rearranged on the bench to simulate dispersal and to avoid having their locations indicate their relationships (Figure 3).

Figure 1.

Depiction of the pattern of evolution students might create by beginning with the ancestral phenotype (a clay sphere) and building the lineages that descend from it. At every level, there is one “speciation event.” Each descendent specimen inherits all the characters from its ancestor, while evolving three or four unique characters (to simplify the figure, not all the characters are pictured here). In the end, there are 10 ancestral specimens leading to five extant specimens.

Figure 1.

Depiction of the pattern of evolution students might create by beginning with the ancestral phenotype (a clay sphere) and building the lineages that descend from it. At every level, there is one “speciation event.” Each descendent specimen inherits all the characters from its ancestor, while evolving three or four unique characters (to simplify the figure, not all the characters are pictured here). In the end, there are 10 ancestral specimens leading to five extant specimens.

Figure 2.

The evolutionary history of the operational taxonomic units (OTUs) created in Figure 1 could be represented by these phylogenies. Phylogenies A and B are synonymous.

Figure 2.

The evolutionary history of the operational taxonomic units (OTUs) created in Figure 1 could be represented by these phylogenies. Phylogenies A and B are synonymous.

Figure 3.

After students build their specimens, the ancestral taxa go extinct. The remaining specimens are rearranged and are given names that do not reveal their relationships.

Figure 3.

After students build their specimens, the ancestral taxa go extinct. The remaining specimens are rearranged and are given names that do not reveal their relationships.

Watching the Detectives

When two or more groups have built and labeled their specimens, they can switch places. (If there is an odd number of groups, three must rotate in a triad, rather than switch in pairs.) Each group's new task is to hypothesize the evolutionary history of the new specimens by drawing a phylogeny based on shared characteristics. To do this, students must categorize characters in a way that is informative (Table 1); if they are too specific, the characters they describe will be possessed by only one taxon (autapomorphies), whereas being too general will result in characters shared by all taxa (symplesiomorphies), neither of which is useful in determining the relationships among taxa. Ideally, students should focus on characters shared by at least two, but not all, taxa (synapomorphies), because these hint at a recently shared evolutionary history that is separate from the other taxa. Students can build a character matrix, but given the relatively small number of characters, matrices tend to be ambiguous. By comparing the number of shared characters among pairs of taxa, students then draw a phylogeny depicting the evolutionary relationships of all the OTUs. They also label their tree to indicate on which branch the characters mostly likely evolved (Figure 4).

Table 1.
Example character table used to determine which taxa have the most traits in common (OTU = operational taxonomic unit).
OTU 1 = present; 0 = absent
CharacterWhassupHermioneMarkThursdayRandolf
1. Clay body 
2. Oval wings 
3. Two pairs of oval wings 
4. Legs 
5. Star 
6. Small circles 
7. A row of small circles 
8. Cross 
9. Spikey tail 
10. Loop on head 
11. Flat disk 
12. Round body 
13. Loop legs 
14. Arms 
15. Striped body 
OTU 1 = present; 0 = absent
CharacterWhassupHermioneMarkThursdayRandolf
1. Clay body 
2. Oval wings 
3. Two pairs of oval wings 
4. Legs 
5. Star 
6. Small circles 
7. A row of small circles 
8. Cross 
9. Spikey tail 
10. Loop on head 
11. Flat disk 
12. Round body 
13. Loop legs 
14. Arms 
15. Striped body 
Figure 4.

An example phylogeny depicting relationships among the five specimens and a parsimonious (i.e., requiring the fewest changes) timing of character evolution that might be hypothesized by students.

Figure 4.

An example phylogeny depicting relationships among the five specimens and a parsimonious (i.e., requiring the fewest changes) timing of character evolution that might be hypothesized by students.

Unlike for natural systems, these student systematists can verify their hypothesized phylogeny by comparing their tree to the evolutionary history mapped by the creators of the specimens. Students complete a homework assignment in which they draw and compare the two trees and answer questions about the characters they used. This provides closure seldom experienced by scientists and reinforces the concepts of convergence and loss, and that nodes can be rotated without changing the meaning of the tree. It also stresses that phylogenies are hypotheses based on observation and may be rejected or modified given new evidence. Alternatively, some of the ancestral specimens may be kept and serve as a fossil record to which students can compare their hypothesized tree. A complementary fossil record provides strong support for evolutionary relationships inferred from shared, derived characters.

In addition, this exercise can be used to address several principles of macroevolution, such as (1) how small, gradual changes can lead to the diversity of life; (2) that speciation does not always lead to replacement; and (3) that extant taxa can be used to reconstruct the past. The specimens can also be used to teach taxonomic hierarchy and classification by instructing students to organize them into species, genera, families, and so on. Some students may be “lumpers” while others are “splitters.”

Effect on Student Learning

Methods

This exercise has been implemented by multiple professors as part of the freshman sequence of biology at Georgia Gwinnett College (GGC) in Lawrenceville, Georgia, for 3 years. GGC is an open-access public school in the University System of Georgia with small class sizes and an extremely diverse student body (e.g., 38.7% white, 31.4% black non-Hispanic, 15.6% Hispanic, 21% nontraditional age, and ~50% first-generation college student). In order to assess the effects of this exercise, student performance before and after the exercise was compared using scores on a pretest and posttest given as part of the normal course assessment. The pretest and posttest contained 33 multiple-choice questions, six of which pertained to phylogenetics. Questions on the pretest and posttest were identical. The pretest and posttest were taken online in the first week of the semester and the last week of the semester, respectively (pretests and posttests were separated by about 13 weeks, with the activity taking place in week 5). Students were given credit for completing each test, not for their scores on those tests.

Analysis was performed only on the six phylogeny-related questions, which referred to an example phylogeny and tested student understanding of monophyly, common ancestry, and shared characteristics (see  Appendix). Student performance across all sections (28 sections taught by 11 professors) from the fall 2013, spring 2014, and summer 2014 semesters were combined. Although the amount of class instruction on phylogenetics varied, much of the coverage of phylogenetics took place during this exercise. Many students who took the pretest did not take the posttest, because they withdrew from the course or chose not to take it, and therefore the sample sizes of the tests are different (pretest: n = 487; posttest: n = 330). Student withdrawals might artificially inflate improvement between the tests if higher-achieving students were more likely to take both tests (although those who took both tests scored no better on the pretest than those who took only the pretest; Student's t = 1.3, df = 412, P = 0.2). Unfortunately, the online system used to tally grades did not allow separation of individual student scores for each question for all classes, but I was able to do this with the classes I taught. Therefore, to correct for potential biases due to nonrandom student withdrawals, I separately compared the results of students who took my classes, because analysis could be limited to the scores of students who took both the pretest and posttest. The results from students who took my classes were derived from three sections in three semesters (fall 2013, spring 2014, and summer 2014) and from two sections in fall 2014. Chi-square tests of independence (χ2) were used to evaluate differences in the ratio of correct versus incorrect answers between the pretests and posttests.

Results

Students were significantly more likely to choose the correct answer on each question on the posttest than on the pretest, when assessed across all sections (Table 2) and in my classes alone (Table 3).

Table 2.
Percentage of students from all sections who answered pretest and posttest phylogeny questions correctly (n = 487 on the pretest, n = 330 on the posttest). Fewer students took the posttest than the pretest because they withdrew from the course during the semester or chose not to take it.
Question TopicPretestPosttestχ2P
1 Recognizing monophyly 23.8 35.6 13.1 0.0003 
2 Recognizing the most recent common ancestor of a subset of taxa 33.8 54.6 34.6 <0.001 
3 Recognizing the common ancestor of all the taxa 41.0 55.2 100 <0.001 
4 Recognizing shared characters 50.2 71.5 37.6 <0.001 
5 Recognizing shared characters 46.8 65.5 27.9 <0.001 
6 Recognizing synonymous trees 36.1 62.0 53.2 <0.001 
 All questions combined (total score) 38.6 60.9 236 <0.001 
Question TopicPretestPosttestχ2P
1 Recognizing monophyly 23.8 35.6 13.1 0.0003 
2 Recognizing the most recent common ancestor of a subset of taxa 33.8 54.6 34.6 <0.001 
3 Recognizing the common ancestor of all the taxa 41.0 55.2 100 <0.001 
4 Recognizing shared characters 50.2 71.5 37.6 <0.001 
5 Recognizing shared characters 46.8 65.5 27.9 <0.001 
6 Recognizing synonymous trees 36.1 62.0 53.2 <0.001 
 All questions combined (total score) 38.6 60.9 236 <0.001 
Table 3.
Percentage of students in my classes who answered pretest and posttest phylogeny questions correctly. These results are limited to students who took both the pretest and the posttest (n = 76).
Question TopicPretestPosttestχ2P
1 Recognizing monophyly 36.8 77.6 26.7 <0.001 
2 Recognizing the most recent common ancestor of a subset of taxa 67.1 85.5 7.29 0.007 
3 Recognizing the common ancestor of all the taxa 69.7 96.1 20.6 <0.001 
4 Recognizing shared characters 60.5 90.8 20.0 <0.001 
5 Recognizing shared characters 54.0 81.5 13.6 <0.001 
6 Recognizing synonymous trees 59.2 92.1 12.0 <0.001 
 All questions combined (total score) 58.0 88.2 72.6 <0.001 
Question TopicPretestPosttestχ2P
1 Recognizing monophyly 36.8 77.6 26.7 <0.001 
2 Recognizing the most recent common ancestor of a subset of taxa 67.1 85.5 7.29 0.007 
3 Recognizing the common ancestor of all the taxa 69.7 96.1 20.6 <0.001 
4 Recognizing shared characters 60.5 90.8 20.0 <0.001 
5 Recognizing shared characters 54.0 81.5 13.6 <0.001 
6 Recognizing synonymous trees 59.2 92.1 12.0 <0.001 
 All questions combined (total score) 58.0 88.2 72.6 <0.001 

Discussion

Macroevolution is not intuitive, and students often struggle to visualize evolutionary history. Phylogenies are useful tools to help us understand these patterns, but they require students to learn specific jargon, such as parsimony, and concepts, such as the equivalency of rotating nodes. Many textbooks are beginning to incorporate phylogenies as they discuss evolution, the relationships among genes, the origins of diseases such as HIV, and the organization of biodiversity; however, students receive little training in reading and analyzing evolutionary trees. Indeed, textbooks often refer to phylogenies with little or no description on how to properly interpret them (Cately & Novick, 2008). To be proficient, students should receive training and practice in drawing and interpreting phylogenies (as an example of a text that attempts to explain phylogenies to a beginner, see Zimmer & Emlen, 2013, chapter 4).

Here, I have described an exercise that gives students hands-on experience with the process of macroevolution, as well as a chance to apply phylogenetic concepts in a realistic way. A basic assessment showed that it successfully improved students’ knowledge of relevant terms and ability to analyze a phylogeny (Tables 2 and 3). The relatively low success rate on the posttest across sections (61%) may reflect low motivation of students on an out-of-class assignment that is scored for participation only. However, given that the exercise described herein takes place during the fifth laboratory of 15, posttest results are potentially indicative of 10 weeks of retention. This extended length of time between tests may also reduce bias caused by multiple exposures to the same questions; however, this was not controlled for explicitly. The depth of instruction provided by this exercise is flexible, and therefore it is appropriate for many levels. Indeed, it has been implemented in high school classes and at several undergraduate levels (freshman, sophomore, and junior–senior). It has also been used with both STEM and non-STEM majors. It is best performed in a 3-hour period but can be done in less time if students have had previous exposure to drawing phylogenies. Alternatively, the specimens can be preserved for later sessions. (I can provide a fully instructive manual upon request.)

This exercise shows students how macroevolution works and introduces them to phylogenetic patterning in a fun and interactive way. It can also be used to demonstrate taxonomy, the causes and conditions leading to speciation, common descent, homology and convergence, and even paleontological patterning. Students get to express their creativity and watch as every bench creates something new. Taken together, the class shows how small changes accumulate over time to form the diversity of life; after all, each bench starts with a single ball of clay, but they end up with a world of diversity (Figure 5). Students often comment that this is one of their favorite exercises, and their results on posttests indicate that they better understood the concepts behind it. More importantly, if we can make it easier for students to accept the science, then we might make strides to remove some of the stigma surrounding macroevolution in the public at large.

Figure 5.

Representative diversity created by students during this exercise.

Figure 5.

Representative diversity created by students during this exercise.

I thank Zach Vanhoose, Sallie Vanhoose, and Jennell Talley for testing the initial idea for the exercise; Kefyn Cately and Manuel Leal for encouraging me to practice it on their students; Ali Witt and Cierra Fowler for grading so many tests; Kimberly Hays for encouragement; and the 1108 faculty at GGC for adopting and assessing the lab. Thanks to Peter Sakaris for his statistical advice. Thanks also to Jennell Talley, Fengjie Sun, Jennifer Davis, and the anonymous reviewers for helpful comments on the manuscript.

References

References
Allmon, W.D. (
2011
).
Why don't people think evolution is true? Implications for teaching, in and out of the classroom
.
Evolution: Education and Outreach
,
4
,
648
665
.
Baum, D.A., DeWitt Smith, S. & Donovan, S.S.S. (
2005
).
The tree-thinking challenge
.
Science
,
310
,
979
980
.
Camin, J.H. & Sokal, R.R. (
1965
).
A method for deducing branching sequences in phylogeny
.
Evolution
,
19
,
311
326
.
Cately, K.M. & Novick, L.R. (
2008
).
Seeing the wood for the trees: an analysis of evolutionary diagrams in biology textbooks
.
BioScience
,
58
,
976
987
.
Coyne, J.A. & Orr, H.A. (
2004
).
Speciation
.
Sunderland, MA
:
Sinauer Associates
.
Darwin, C. (
1859
).
On the Origin of Species by Means of Natural Selection, or the Preservation of Favoured Races in the Struggle for Life
.
London
:
John Murray
.
Edwards, A.W.F. & Cavalli-Sforza, L.L. (
1964
).
Reconstruction of evolutionary trees
. In V.H. Heywood & J. McNeill (Eds.),.
Phenetic and Phylogenetic Classification
(pp.
67
76
).
London
:
Systematics Association
.
Gendron, R.P. (
2000
).
The classification and evolution of Caminalcules
.
American Biology Teacher
,
62
,
570
576
.
Kitzmiller et al. v. Dover Area School District et al. (400 F.Supp.2d 707, M.D.Pa., 2005).
Mayr, E. (
2001
).
What Evolution Is
.
New York, NY
:
Basic Books
.
Meir, E., Perry, J., Herron, J.C. & Kingsolver, J. (
2007
).
College students’ misconceptions about evolutionary trees
.
American Biology Teacher
,
69
,
71
76
.
Miller, J.D., Scott, E.C. & Okamoto, S. (
2006
).
Public acceptance of evolution
.
Science
,
313
,
765
766
.
Phillips, B.C., Novick, L.R., Cately, K.M. & Funk, D.J. (
2012
).
Teaching tree thinking to college students: it's not as easy as you think
.
Evolution: Education and Outreach
,
5
,
595
602
.
Robbins, J.R. & Roy, P. (
2007
).
Identifying & correcting non-science student preconceptions through an inquiry-based, critical approach to evolution
.
American Biology Teacher
,
69
,
460
466
.
Sandvik, H. (
2008
).
Tree thinking cannot be taken for granted: challenges for teaching phylogenetics
.
Theory in Biosciences
,
127
,
45
51
.
Scott, E.C. (
2009
).
Evolution vs. Creationism: An Introduction
.
Berkeley, CA
:
University of California Press
.
Selman v. Cobb County School District. (449 F.3d 1320, 11th Cir., 2006).
Sokal, R.R. (
1983a
).
A phylogenetic analysis of the Caminalcules. I. The data base
.
Systematic Zoology
,
32
,
159
184
.
Sokal, R.R. (
1983b
).
A phylogenetic analysis of the Caminalcules. II. Estimating the true cladogram
.
Systemic Zoology
,
32
,
185
201
.
Zimmer, C. & Emlen, D.J. (
2013
).
Evolution: Making Sense of Life
.
Greenwood Village, CO
:
Roberts
.

Questions used for the pretest and posttest assessment. Questions were identical for both tests.

Which of the following forms a monophyletic group? A and C 
C and D 
A, B, and C 
B, C, and D 
Which pair of individuals shares the most recent common ancestor? C and D 
A and B 
B and C 
A and C 
A and D 
Which specimen is the common ancestor to all the others? 
Which characteristic is most likely shared by A, B, and C but not D? 
Which specimen is most likely to have the characteristic z? 
All of the above are equally likely. 
Which phylogeny shows the same relationships as the original? 
graphic
 
Which of the following forms a monophyletic group? A and C 
C and D 
A, B, and C 
B, C, and D 
Which pair of individuals shares the most recent common ancestor? C and D 
A and B 
B and C 
A and C 
A and D 
Which specimen is the common ancestor to all the others? 
Which characteristic is most likely shared by A, B, and C but not D? 
Which specimen is most likely to have the characteristic z? 
All of the above are equally likely. 
Which phylogeny shows the same relationships as the original? 
graphic