Chapter 4: Principles of Classification and Systematics

Use the tree below to answer the following questions.

1. Circle those which are monophyletic groups:

ABC CDE EF EFG ABCD BCEF DEFG

2. Which pairs are more closely related? (circle the right answer)

AB or BC CD or EF EF or DG AB or EF

3. AB is the sister group to? _________

4. D is the sister group to? _________

Polytomies and Rotatable Nodes

In this lab, we will be dealing primarily with trees that have dichotomous branching patterns. That is, a pattern where only two descendants arise from a single node (branch point). But, in reality one often has to deal with a situation where there are polytomies on their tree. Polytomies are branching patterns where more than two descendants arise from a single node. Polytomies are often interpreted as representing uncertainties in the relationships of those groups.

What else might a polytomous pattern mean (evolutionarily speaking)?

Its important to note that branches on a tree can be rotated without changing the relationships on the diagram!

Draw an equivalent tree to the one in Figure 4.2. Use the straw kit to help you. (Hint: There are several right answers)

Characters

The study of phylogenetics groups organisms into groups (clades, as you learned earlier). A clade represents an ancestor and all of its descendants. Clades are united by shared derived characters (synapomorphies). But what is a character?

A character is any definite aspect of a particular organism. This means that any organism can have an almost infinite number of characters. So which ones do we use? Put another way, what makes a character appropriate for use in phylogenetic systematics?

At this time, the best characters are discrete characters. Discrete characters have a limited number of possible values. Discrete characters can either be binary or multistate. Binary characters are those that only have two states, like Present/Absent. For example, the ability to roll ones tongue is either Present or Absent. Or, a beetle may be colored either yellow or blue. Multistate characters on the other hand, have more than two states. For example, eye color can either be blue, brown or green, that's three character states.

Continuous characters such as size, are measurements on a continuous scale, and they are very hard to use. For example, the length of a femur can be 10.1cm, 12 cm, 10.4cm, 11.2cm, etc. The problem is, "Are these sizes "different enough" to group into separate groups? Or "similar enough" to group into the same group?" What is "different" and "similar" when it comes to size? Because these terms are so subjective, and with the current computer software, these characters just won't help us to define any groups.

The easiest way to do this is to create a code for the characters and their various character states and to use that information to make a character matrix.

Try this out with the example on the next page. If you get stuck, be sure to ask for help. Wait for the rest of the class. We'll go over these characters together. Draw the tree for the above critters below or on the other side of this page. Map on the characters. (Hint: There are several possible trees)

Character Terminology

An apomorphy is a derived character that is shared by all of the members of a clade, but is not possessed by the immediate ancestors of the clade. If a character is shared by more than one of the descendants, it is called a synapomorphy, or a shared derived character. If a character is possessed by only one descendant, it is termed a autapomorphy.

A plesiomorphy is an ancestral character. It is possessed by an ancestor and, possibly, any of its descendants. Basically, shared ancestral characters (synplesiomorphies) are not useful in phylogenetic reconstruction. For example, it would be folly to use "lungs" as a character for grouping all primates, because lungs arose much farther back in the tree of life as an adaptation to the colonization of land. Lungs are shared by all primates, but they are also shared by lungfishes and all other vertebrates that arose from the common ancestor of lungfishes and terrestrial vertebrates; thus, lungs are not derived at the level of primates. They are, however, a shaxed derived character for lungfish plus all the other things that have them: derived and primitive axe relative terms.

In addition, shared derived characters that developed convergently are not useful in phylogenetic reconstruction. Many animals have wings: birds, bats, pterosaurs, insects to name a few. Wings however, are made in very different ways in birds, bats etc. They axe analogous (similar) but not homologous (having the same genetic and developmental origins), and are therefore described as convergent. These different animals have "converged" in their evolution of simialr features. The wings of bats are probably a shared derived feature for bats, but wings in the general sense are developed convergently. Bats are mammals and insects are arthropods, and there is abundant evidence that birds axe derived dinosaurs.

The only characters that help us in phylogenetics are synapomorphies. Synapomorphies define clades. The others provide information about the taxa in question, but do little to inform us as to their evolutionary relationships.

Cladistics is a relative science. We are always conducting phylogenetic analyses on different scales (like the difference between your family tree, the primate tree and the tree of all life). It is, therefore, important to note that a synapomorphy at one level on a phylogenetic tree, could be viewed as a plesiomorphy on another scale.

Using the tree that you created, and the characters that you mapped on, answer the following questions...

1. What one synapomorphy defines clade ED (Include the character and the change in character state)?

2. Of all the characters analyzed, which one is an autapomorphy? What taxon has that character?

3. When looking at clade ED, what is an example of a plesiomorphic, or ancestral character for this group? (Remember that means that the character arose earlier than that clade, so its a character that ED shares with other taxa.)

Computer Exercises

Phylogenetics Using MacClade (by Jonathan Marcot)

We are now going to take what we have learned about phylogenetics in the lab, and go a step further using a computer program called MacClade. MacClade was developed by David and Wayne Maddison (from right here at the University of Arizona). It allows us to construct phylogenetic trees and analyze the evolution of organisms and their characters.

Constructing Phylogenetic Trees

One of the simplest, yet most useful attributes of MacClade is that it allows us to construct phylogenetic trees. To do this, all we need to do is supply MacClade with data (our character matrix) and it build trees using parsimony. For this exercise, we're going to use the matrix that we created in the lab, and reconstruct the phylogeny of our hypothetical organisms.

Open MacClade by double-clicking on the "MacClade 3.07 beta 6" icon in the "Items for GEOS203" folder.
You'll be confronted by a box saying something like "Don't publish anything using this program without the permission of the authors." Hold up your right hand, and say, "I promise not to publish the results of this lab." Then click on the OK bar...
When confronted to open a file, choose the "New" button. This opens a new file in which we'll put our character matrix.
What the screen should look like now is a spreadsheet. We have to tell MacClade how many taxa and characters we have in our matrix before we can enter the data. At the first cell (row 1, column 1) there's a gray shaded frame. Using the mouse, click on the bottom of that frame and drag it down until there are 5 rows (these will be your taxa). Now click on the right side of the frame and drag it out to the fifth column (these are for characters).
We are now ready to enter our data matrix. First, in the first column (the one without question marks) enter the names of your taxa from your handout (i.e. 'Species A,' etc.). If you want, you can enter the names of your characters (e.g. antennae, legs, etc.) in the first row (again, the one without question marks) on the top of the matrix, but this is not required (though it might help you keep track of your characters).
Now all we have left to do is enter the character state data that we collected in lab into the cells of the matrix that now have question marks. Each row of the matrix is one taxon (Species) and the characters are read across. Use the mouse to select the first cell in the first row and first column of the matrix (this time the cell with the question mark). Enter the state for the first character for Species A. Then use the right arrow to move to the second character. When you have entered all of the characters for Species A, MacClade will beep. This is frightening, but harmless. Simply move down to the next row (Species B) and enter its character data. Do this until your matrix is complete.

Finding the Correct Tree In any analysis, a nearly infinite number of trees can be made. So how we know we have the correct tree? The answer is that we never do. All trees generated in a phylogenetic analysis are merely hypotheses of how the group evolved. They are all equally valid, but usually a few, or if we're lucky just one, are plausible.

Just how plausible a tree is, is evaluated using a variety of various technical and mathematical methods. One of the simplest and most widely used method is parsimony. Parsimony postulates that evolution will take the shortest number of steps (or the shortest treelength) in an evolutionary pathway. For example, if a group of five animals possessed long floppy ears, it makes more sense to believe that there was one long floppy eared ancestor and they all evolved from it, rather than floppy ears evolving independently five different times. Using this principle, parsimony suggests that the shortest possible tree is the most probable.

So now that we have a character matrix, let's now build the tree. To do this we need to be in a different window (now we're in the Data Editor, and we need to get to the Tree Window). Go to the "Display" pull-down menu at the top of the screen. Pull it down to "Go to Tree Window." MacClade will tell you that no trees exist yet. To get around this, select the "Default Ladder" button.
There's a tree, but maybe not the best one. MacClade has a very primitive algorithm for finding more parsimonious trees, and now were going to use it to find a better tree than we have now (if there is one!). Use the mouse to click the "Tools" pull-down menu. Don't let go of the mouse button. Drag the flashing box down until the "Full Search Above" tool is high lighted (in the middle on the far left). The mouse icon has now changed from an arrow to a symbol (I am not really sure what it is). Move this icon so that its bottom (the little dot) is on the lowest branch on your tree (below the root). The branch will turn from black to shaded. When you see this, click the mouse button.
MacClade is now searching for better trees. When it is done, it'll beep at you twice, and tell you if it has found a better tree. It should say that "no trees were found" meaning that no trees were found that were better (of a shorter treelength = more parsimonious) than the current tree.

Analyzing Trees and Character Evolution

Reconstructing phylogeny is not MacClade's strength (that's not really what it's for). What it does do really well is allow us to analyze how different characters evolve on our phylogenetic tree.

Go back to the "Tools" pull-down menu. Again, don't let go of the mouse. Select the Arrow tool (in the upper right hand corner). This will allow you to manipulate your trees!
Go the "Trace" pull-down menu. Select "Trace Characters". You'll notice that your tree has suddenly become colorful. These colors represent the change in character state for a given character. On the lower right hand side of you screen should be a small box titled "Colors" It has a scroll bar that allows you to look at each character and the changes in character state mapped onto your tree.

Answer the following:

1. What is the treelength of the current tree? How does this tree compare to the one that you drew yourself?

2. Using the "arrow" make taxon A and taxon B sister groups. (Ask your TA to demonstrate how to do this). Write down the treelength. Look at some of the character changes.

3. Using the "arrow" make taxon C the sister to AB. Write down the treelength.

4. Using the "arrow" make taxon D the sister to ABC. Write down the treelength..

5. Using the "arrow" make taxon C the sister to AB. Write down the treelength.

6. Using the "arrow" make taxon B the sister to taxon D. Write down the treelength.

7. What is the least parsimonious of the trees you created in questions 1 - 6?

The Tree of Life!

The tree of life is a valuable web resource that was created by David and Wayne Maddison, and is maintained by Katja Sabine Schulz here at the University of Arizona. As you'll notice, some "branches" on the tree are more detailed than others. This is because some groups are better studied than others. The tree is always in a state of change, as new research helps us to better understand the relationships of organisms. As you can imagine, there is a lot of life out there and building and maintaining this page will be a life long project. Get to know the tree of life better by jumping onto Netscape and go to http://ag.arizona.edu/ENTO/tree/phylogeny.html.

To do the following exercises, select the green button "TO TREE".

1. Follow this pathway by clicking on the following, in the order listed: Eukaryotes: Mitochondrial Eukaryotes: Crown Eukaryotes: Metazoa: Arthropoda: Hexapoda: Insecta: Pterygota: Neoptera: Endopterygota. What group is the sister-group to the Lepidoptera (moths and butterflies)?

2. Start at the root and find your way to the Mammalia. Draw the phylogeny for the Mammalia.

3. Start at the root and find your way to the Fungi. Using the information available on this page, answer the following:

(a) By what epoch are all the modern Classes of fungi recorded in the fossil record?

(b) In what journal can I find the 1994 paper by Hass, Taylor and Remy?

Back to Chapter Four