Suggestions for Cladogram Procedures

1. Arrange the taxa in groups according to the phylogeny you are learning in lecture.

Place those in the same phylum, phylogenetic group or subkingdom on the same branches.
Look up the relationships of each taxon in the list and construct a tentative branching pattern that reflects the major phylogenetic groupings of Animalia.
If there seems to be a group of three that you can't immediately divide into subgroups, you may decide later which two are more closely related, after more chracters and other relationships have been resolved.

2. Lay out the groups (represented by letters in the diagram below) on a draft page in a format like the one illustrated below.

All the taxa names should be in a straight line across one side of the page, with more similar ones closer together.

Sketch in a tentative branching pattern, guided by your preliminary organization in bullet 1 above, below the names, joining them in series of two-branched, "V"-junctions (example provided above).

Label each branching point with a number, and then the right and left branches from each point with the letters A or B (as in the illustration above).

3. Compile a list of characteristics which vary among the taxa and list their conditions for each taxon.

Start with the characters that define each of the major phylogenetic divisions you reviewed in bullet 1 above.
The condition of each of these traits, or characteristics, can listed under each taxon on a card or work sheet. If you keep characteristics in the same order and label them (e.g., "fate of blastopore:"), comparison will be easier.
Draw upon the hierarchical checklist (click to look at it) that I have prepared of useful characteristics and conditions.

There are many other characteristics that vary among these animals besides those listed, but the checklist covers most of those that have been most useful for building phylogenies of animal taxa down to about order level.

As you accumulate lists of character conditions in the checklist from lecture notes, Lab Guide, your textbook, and other sources, also keep an eye out for potentially useful characteristics that are not in the checklist.
Notice that each useful trait should have two conditions at each branching point, one of which will be derived and the other must be ancestral. For example, in comparing the parasitic with the free-living flatworms, the trait "epidermis" would be scored "ciliated with mucus glands" in the free-living Turbellaria, but "absorptive for gaining nutrition from host tissues" in the parasitic Trematoda and Cestoda. The former condition would be considered ancestral, partly because the possible outgroups Cnidaria larvae, rotifers, bryozoans, bivalves and many other spiralian protostomes have ciliated epidermis and/or mucus glands. Also, the absorptive epidermis in parasitic flatworms is more elaborate than ordinary, ciliated epidermis, in that (among other features) it is syncytial and has raised microvilli to increase its surface area, so it would be the derived condition. Zoologists have termed this a "neodermis" (= "new skin") and used it as the primary, shared, derived characteristic to define the clade of parasitic flatworms. At each branching point, you must decide which condition of a trait is derived, and use that one, but not the ancestral alternative, to justify a branch of the cladogram; ancestral conditions of traits have no value in analyzing phylogeny.

4. Starting from your initial, draft branching pattern, mark as many shared conditions as you can find in the lists for taxa in the more closely related pairs or groups.

5. Then bring in the next most closely related taxon and see how many of those shared features also occur in it.

If none are shared with the sister group but show up in other taxa in your list, try revising your branching hypotheses and re-arrange the taxa in your chart.

This is a trial-and-error process, a kind of puzzle or game. It is not unusual to find that two taxa (A and B) in a set of three (called A, B, and C) share several important characters that do not occur in the third (C), but also that a different two (say, A and C) also share several important characters that are not found in the other one (B). In order to resolve this dilemma and decide which two are more closely related, you must consider two possibilities:

A. Some of the shared characters are homologous but ancestral, that is, they were likely present in the most recent common ancestor of all three taxa, and lost during evolution of one of the taxa.

The best way to recognize this situation is to ask whether some fourth, less similar taxon also displays the character in question. This fourth taxon is called an "outgroup," because it is reasonably assumed to be more distantly related than any member of the main set under comparison, but closely enough related to provide some insight into the likely features of a more remote ancestor of the compared group.
For example, if you were comparing lizards, snakes, and mice, a reasonable outgroup would be a salamander. The salamander has two pairs of legs, so we can infer that two pairs of legs are an ancestral condition for the three amniotes, and therefore we will not be confused by the similarity between lizards and mice of having 4 legs, while snakes (which we believe are more closely related to lizards than mice are) have none.

B. Some of the important, shared characters are only analogous and convergent, that is, they are in a very similar condition in two different animals, but evolved independently to that condition from different ancestors. This situation can be confusing for even the most experienced cladistic systematists, and leads to many arguments among them. Some ground rules for recognizing convergences are offered below:

A general (but not absolute) rule is that embryology reveals convergences (a special case of "ontogeny recapitulates phylogeny"). If the character or structure in question arises from different parts of the embryos in each taxon, or in different ways, it is likely to have evolved independently and to be a convergence. On the other hand, if it develops in the same way from the same primordial area or structure in the embryo, it is probably either derived or ancestral.

However, two animals with a remote, common ancestor may independently evolve the same embryonic process, or a similar modification of the same basic structures for similar adaptations, long after their ancestors diverged. This is not common, but is believed to occur occasionally.

A somewhat shakier guideline is "probable adaptive convergence." If the two taxa are similar to each other, but each differs from many of its closer relatives, in habitat, feeding habits, or defensive strategies, etc., and of course if the shared condition in question is an obvious, functional adaptation to that habitat, feeding or defensive mode, etc., then the character or condition is probably convergent.

For example, the body size and contours are remarkably similar between ichthyosaur fossils, porpoises, certain sharks and swordfish. Each animal represents a different class of vertebrate (based on other characteristics), so the very similar body form must be convergent. All are swift, fish-eating inhabitants of the open ocean. The near relatives of each that occupy different habitats and/or feed differently (crocodiles, cows, rays, and grouper, for example) have very different body forms.
Unfortunately, many characters and conditions represent unknown functions or adaptations, or are not likely to be influenced by ecology (for example, sexually selected features), or may be a holdover from an unknown adaptation to some different ecological function in the taxon's evolutionary past. It is this all-to-common circumstance that is behind some of the bitterest phylogenetic controversies in zoology.

C. Often, the only practical resolution of such dilemmas is "the weight of the evidence." After we eliminate all reasonably justifiable, ancestral and convergent characters, then the two taxa which share the greatest number of remaining conditions are hypothesized to be the most closely related - that is, to have descended from a more recent common ancestor than either has shared with any other taxon under comparison. The underlying assumption is that those few, contradictory similarities of conditions with other taxa are actually ancestral or convergent, but we are unable to detect their nature with certainty. This means, of course, that you have to come up with a lot more than the minimum six distinguishing chatracteristics for a group of three taxa.

Last modified on June 18, 2004