We could use any means of classification to organize the world's fossil and living species, and answer the above question. Organisms could be grouped on the basis of size, whether they lived on land or in the sea, or even by color. However, one of the main tenets of comparative biology is that there is order in nature: an order which manifests itself in patterns of similarity of appearance among all the organisms on the Earth.

There are two fundamentally different ways of explaining this striking similarity, which are often dialectically opposed in western thought. One relies on the belief that all the different types of organisms are the result of order imposed by a divine omnipotence. The other approach is to look at the similarities between groups of organisms, and to see them as a manifestation of the degree of relatedness existing between these organisms due to descent with modification (e.g. evolution). In other words, all life is descended from a common ancestor (or at least a limited number of common ancestors) and there is a process called evolution which is responsible for the splitting of lineages and the divergence of form that results in the diversity of life. [There is also the occasional combining of lineages as in the symbiotic organelle theory]. Since life has been evolving for 4 billion years without human observers we cannot possibly know the exact evolutionary history of life. However, we can make inferences about the evolutionary relationships of organisms on the basis of their shared similarities, because the traits present in an ancestor tend to be passed on to its descendants.
 
 

As you remember from class a character is a feature or thing which we can examine or label. A character which is an innovation developed in an ancestor of a group is called a derived character relative to the characters seen in the ancestors of the founder of the new group. The ancestors of the founder of the new group are said to have at least one primitive character relative to the derived character. The character which is derived because it is an innovation in the ancestor of a group, is also a primitive character with respect to the members of that group. A character shared by all members of the group is, as you might expect, a shared character. If the character is both in the new state, and shared by the members of the group in question, it is a shared derived character.

If a group of organisms is believed to have shared a common ancestor, the group containing that ancestor and all of its descendants is called a monophyletic group. A monophyletic group must be recognized by the presence of at least one shared derived character. A character found only in one of the groups being studied is called a unique derived character, and does not help us relate this group to any other group. An example of a unique derived character for humans is frontal sex.

The fact that organisms from different species resemble each other does not necessarily mean that they are closely related. They might resemble each other because they share a large number of primitive characters. On the other hand, they might share a character that evolved independently in the groups as a result of convergent evolution. This resemblance may be due to different lineages of organisms adapting to very similar environments. Similarities which evolve through means other than descent are called analogous characters.

The wings of a bat, a bird, and a butterfly all perform the same function, and have similar form. However, on the basis of many other dramatically different characters we can conclude that this aerodynamic limb evolved independently in these three organisms from ancestors who did not have such a structure. Analogous characters such as those just described are not used to group organisms in an evolutionary classification.

When we have an array of organisms and begin our search for shared derived characters, we need to know which characters are primitive for all of the organisms we are examining. We can do this by looking for the characters every member or almost every member seems to have then look at a group of organisms outside the group in question (an outgroup), and see what characters are shared with the group in question. Outgroups allow the polarity of characters (e.g. primitive to derived) to be established.

Similarities between organisms which do share a common ancestor are called homologous characters. Relative to groups not possessing these characters they are also shared derived characters uniting the group which have them. The front limbs of a dog, a bird, a whale, and a human perform very different functions, yet they share a common anatomical structure: all have a single large bone, the humerus, which is attached at one end to the shoulder and at the other end to two smaller bones, the radius and the ulna. The same bones are present in the wings of bats and a birds. If the front limbs of each of these organisms had evolved independently from different ancestors without front limbs, it would be hard to imagine that such striking similarities would have arisen. It is far more reasonable to conclude that these similarities are present because the common ancestor of all these organisms had the same type of front limb with the same bones. On the basis of this assumption all of these organisms are classified together as tetrapods and their front limbs are called homologous structures. Thus, this type of front limb unites these different animals as a shared derived character.

Two other commonly recognized schools of systematics other than cladistics, which groups organisms only by shared derived characters, are: evolutionary systematics, which groups by shared derived and shared primitive characters; and phenetics, which groups by convergent characters, shared derived characters, and shared primitive characters.
 
 

The mechanics of the process consists of a set of hypotheses and tests. First we construct a hypothesis of relationship for the organisms in question. This hypothesis can come from anywhere: general resemblance, a whim, or an authoritative text. Second, we look for characters which allow us to define groups. Third, we look for the distribution of primitive characters which stand in contrast to the derived characters. Fourth, we look for unique derived characters which define each of the organisms. Fifth, we construct a cladogram and hang the distribution of the characters on it. OK - now we have groups defined by shared derived characters and we have our cladogram with our characters. It is now time to test the hypothesis by looking at the characters which could define groups other than the hypothesis in question. These characters are in conflict and must be explained by some ad hoc argument other than simple descent from a common ancestor. If you need more ad hoc arguments to justify your cladogram than you have shared derived characters supporting your cladogram, your cladogram must be discarded. If your cladogram survives this test, the next step is to look for more characters and hang them on your cladogram and see how they fit. If they do not and there are a lot of them, again your hypothesis fails and you must look for a new and better one. This criterion that allows the selecting of the hypothesis which requires the fewest number of ad hoc hypotheses is called the principle of parsimony, and it is the hallmark of science in general. Ultimately, we think something is true, whether it is in general life or in systematics, when it has survived a very large number of tests.