Faculty of Biological Sciences, University of Leeds

These pages have been left in this location as a service to the numerous websites around the world which link to this content. The original authors are no longer at the University of Leeds, and the former Centre for Human Biology became the School of Biomedical Sciences which is now part of the Faculty of Biological Sciences.

Evolution and Taxonomy

Dr Bill Sellers


Welcome to the "Human Evolution" course. We, the course organisers hope that you will find it both interesting and enjoyable - but unfortunately, before you can really get to grips with the fun stuff, you will need some background theory. that's really what today's lecture is all about.

The course is about human evolution. This means that at some stage, preferably fairly early on, you need to have some idea about evolution, and it's also helpful to have an idea what we mean by "human". The first half of this lecture is an introduction to the central ideas of evolution, and the second half is about how we group up animals, including humans, so that we all know what we are talking about when we talk about them. Discussion about how similar and how different humans are from other animals makes up a lot of the subsequent course.


Change through time

The word "evolution" merely means "change through time". It doesn't imply a direction, nor, does it necessarily imply improvement, merely change. In this context, evolution refers to the observation that the animals in the world have not always been the same as those that we see around us today. How do we know that this is the case?


We know that distributions of animals change. We have historical records of bears in England 1000 years ago. We find bones in caves containing lemming (# slide of lemming) and reindeer (#slide of reindeer) bones with no historical record of either being present here. We have film sequences of the Tasmanian wolf only fifty years old, and there are none around today (# slide of Tasmanian wolf).
When we look at piles of sediment that have accumulated in a cave for example, we find that the bones present in the top layers are the most similar to what we expect to find today: chicken and rabbit bones for example, which indicates that this sediment has come in since Roman times when both the chicken and rabbit were first introduced into Britain. These bones just look like old bones. As you dig deeper, you find that the bones begin to become "mineralized". The colour changes and the bone generally becomes heavier, as the calcium and various other chemicals in the bone are slowly dissolved away and replaced by other chemicals from the sediment. These bones are fairly obviously older, and it is in these layers that we start to find (in the UK) animals such as hippopotamuses and lions (# slides of hippo). In a very large column of sediment, the bones found in the lower layers becomes completely different from those found today: very large cats and elephant like creatures, or whatever (# slides of mammoth). Clearly, when these sediments were laid down, the fauna was quite different from today: animals were similar, and the range of animals was similar, but the actual animals themselves were noticeably different in form.

In other places, there are rocks that look rather like solidified sediment. We assume that these are columns of sediment that have become solidified over a very long period of time due to various geological processes. These rocks contain very bizarre animals very unlike anything seen today. Sometimes, giant lizard like creatures are common (# dinosaur slides). At other sites, only fish are found (# early fish picture), even though the current location is miles from the nearest water.

Dating techniques
You may have noticed that I have avoided talking about ages. I've used terms such as "old" and "very old". This is because techniques for dating rocks, fossils and bones accurately are relatively recent.

The earliest method of dating is "stratigraphic dating". This follows on directly from the column of sediment. Anything nearer the top is younger than anything further down. This gives you an idea of the relative ages of rocks, and by estimating the rate of deposition of sediment you can attempt to calculate an absolute age (# slide of early geological time scale). It's not terribly accurate, but it is very intuitive. There aren't any deposits of sediment that cover the whole age of the earth continuously, so that you need to look at the change in the fossils in one deposit, and match them with fossils in other deposits to attempt to build up a full picture.

With the discovery of radioactive decay, other more precise dating techniques have been possible. The best known is radiocarbon dating which works well on organic material and relies on the proportions of a radioactive and non-radioactive form of carbon. For older materials, other radioactive forms can be used: potassium, uranium etc. Each series covers a different time scale, and is useful in particular geological circumstances. None are without their problems, but they can give much better estimates of absolute ages than stratigraphy alone.

There are also other techniques based on thermo-luminescence, or magnetic field reversals that can be used. Used together, these have provided a widely accepted set of dates for various rock layers and fossil animals (# slide of geological age).

Controlling Change

Problems with random change

As I emphasized at the beginning if this lecture. Evolution doesn't imply improvement, or any sort of direction itself. Merely change. However, from our examination of the fossil record, it is clear that animals and plants have become a great deal more sophisticated over the years, and it certainly seems that more recent variations have a tendency of replacing earlier versions. Very many varieties no longer exist - indeed the average "life span" for a species seems to be of the order of a few million years. From various mathematical considerations, it seems highly unlikely that all this change can occur purely by chance. Fortunately, Charles Darwin (# slide of C.D.) came up with a mechanism that explains the apparent direction of evolution extraordinarily well.

Natural selection

The Darwinian argument, "Evolution by natural selection", is extremely clever, and nowadays seems almost self-evident. It is based on empirical observations of natural history and is supported by a wealth of evidence.

[text and diagrams from the NHM book]

* `Survival of the fittest'

* Fitness

* Adaptation

Genetic drift etc.

However, it is now thought that although a large proportion of genetic diversity can be ascribed to natural selection, it is now realised that random genetic drift plays an important non-directed role since certain features show variation without any fitness change. In addition with the discovery of entities like retro-viruses that incorporate their DNA into the host genome, it is clear that there are other mechanisms that act directly at a genetic level.


Classification is simply the ordering of organisms into groups, and giving them names. Before anyone was particularly bothered about evolution, this tended to be a very simple exercise: we'll put all animals that swim in one group (fish & whales); flying animals in another (bats & birds); and the ones that climb trees (monkeys & squirrels). Linnaeus expanded on this a little by using more than one characteristic in his groupings, but nevertheless, there was always dispute about how to produce a "natural" grouping.

When evolution became accepted, it became clear that the obvious way of grouping organisms was by their evolutionary relationship - a huge family tree, if you like, showing how the various animals have descended from common ancestors and grouped accordingly.

Phylogenetic reconstruction

This, then, is the goal. But how do we achieve it?

The best/easiest way is to look at the fossil record and find all the ancestral groups. (# diagram p.46 Evolution) This is precisely what has been done for horses. Unfortunately, this is normally not possible. For most animals, there are just not enough fossils available for this sort of analysis.

There are other problems too. Although a family tree is a natural way of grouping organisms, we still need to decide on what we are going to use as the smallest group. The answer to this (usually), is the species, but this begs the question: "What is a species?"

What is a species?

Generally, a species is defined as a sexually interbreeding (or potentially interbreeding) group of individuals normally separated from other species by the absence of genetic exchange. This is the "biological species concept". This is fine, in theory, but in practice, there are problems. (# diagram of closely related animals - Lemur fulvus subspecies) Group A can mate with group B and produce offspring. Group B can mate with group C and produce offspring, but group A and group C can't mate. In addition, it doesn't help define what a species is for fossil animals where mating can't be observed. And finally, it is no help for defining species in organisms that don't reproduce sexually (# slide of garlic plant). A number of other species concepts have been postulated to overcome these problems, but none of them are perfect.

My personal view is that the concept of a species is a completely arbitrary construct that humans have created. Organisms can be thought of as a continuum of genetic variation, and we use species as a way of naming regions in that continuum for our own convenience. The size of these regions is roughly consistent, but there is definitely overlap at the edges, or even gaps. As long as we are consistent in what we call these regions, then we can still use them for practical purposes like conservation management, and it means that we can stop bickering quite so much over whether a particular animal is or isn't in the same species as another... The concept of a species is OK, just remember that it tends to be fuzzy round the edges.

Practical Reconstruction

Firstly, evolution (change over time) doesn't have to lead to a branching pattern. A single group can change gradually without splitting into two distinct groups (# diagram horseshoe crabs, Evolution P.209). This process is called Anagenesis. However, the much more interesting problem is the reconstruction of the branching pattern, where species split into two or more groups. This is called Cladogenesis and is what gives us our family tree. (# diagram P.212)

Determination of pylogenetic trees is difficult in practice because the common ancestors are usually long extinct and the fossil record is inadequate. However, the relationship can be inferred by looking at common, inherited characteristics: the more morphological, embryological, behavioural, physiological, biochemical, genetic and chromosomal inherited characteristics that organisms have in common, the more likely they are to have descended from a common ancestor.

Merely sharing common features is not enough since they may derive from different evolutionary causes:


This is what we want. The feature is shared because it derives directly from a common ancestor. For example the bony features of the forelimbs in vertebrates. (# diagram P.39)


The similar feature occurs in different species, but it is not present in their immediate common ancestor. For example, anteater-like features in various different mammalian lineages (# diagram P.211). These shared features are very much functional adaptations.


Similar to parallelism, but the ancestral lineages differed for a considerable period of time. For example vertebrate and octopus eyes, or the hydrodynamic morphology of marine predators from the widely separated fish, reptile and mammalian classes (# diagram P.40).


Obviously, homologies are what we need to consider to reconstruct phylogenies. However, they are not always easily separated from the other 2. Consider the convergence example: the shape of the pectoral fins in these animals is very similar due to convergence. However, there is a great deal of homology there two. Especially between the reptile and the mammal due to a common land vertebrate ancestor.


Taxonomy isn't only for evolutionary reconstruction. We need fairly stable names and grouping for practical purposes such as conservation. Groupings make animals easier to remember and identify, and we don't want it all to change every time someone decides that actually humans are closer related to chimpanzees than chimps are related to gorillas. This means that official naming schemes tend to lag somewhat behind the current thrust of research. There is quite a bureaucracy preventing everyone from renaming animals at a whim, and there are international efforts to try and maintain some consistency. Even so, there are generally several alternative classification schemes around for groups of animals that seem to last about 5 years until the next big name in a particular field writes the latest review paper on that specific taxonomy. A good text books will tell you which scheme it is using, and a really good textbook will list several alternatives so you can make up your own mind.


Evolution happens. There is very good evidence for change in the life forms that inhabit the earth over long periods of time.

Charles Darwin's theory about the origin of species through natural selection explains the apparent direction of evolutionary change extremely well. It is probably one of the most widely accepted theories in modern biology.

For practical purposes, we need to group animals. Animals that can interbreed are grouped as species. Species are grouped in a tree structure that more or less attempts to mimic the evolutionary process. We attempt to use the interbreeding idea for extant animals, but for fossil forms we use similarity in shape.

Unfortunately, every scientist has their own personal preference when it comes to classification. That's life!

This page is maintained by Steve Paxton