It is a common argument of Creationists and others who know little of biology to claim that random mutations in a genome cannot produce an animal which is fitter than its ancestors. According to this argument, all mutations are harmful. So, I thought I'd post a simple analogy to show how this argument fails. Suppose, in some imaginary world, we have an animal whose complete genome is dictated by six "letters", which can each be either "H" or "T". You can think of this genome as being based on six coin tosses, if you like, where each coin can come up either heads (H) or tails (T). Imagine that in this particular environment, the "ideal" animal would have the following genome: HHHHHH Further, assume that for this animal, the more Hs there are in the genome, the better adapted the animal will be to its environment. For simplicity, we assume this animal reproduces without sex, so most of the time its offspring have exactly the same genome as the parent. However, occasionally, one-letter mutations can occur in the genome during reproduction. when this happens, the particular letter in the sequence which changes does so entirely at random (think of flipping a coin). So, for example, an animal with the genome: THHTTH will produce offspring with an identical genome most of the time, but occasionally, it might produce something like: THHTHH instead (where here the second last letter in the code has undergone a random change). So far, we have a system where there is no progression from less fit animals to more fit animals, because it is as likely that an animal with the genome THHTTH will produce offspring with genome TTHTTH (less fit) as it is that it will produce offspring with genome THHTHH (more fit). Now, put natural selection into the picture. Suppose our animal with genome THHTTH has 7 offspring, with the following genomes: 1. THHTTH 2. TTHTTH 3. THHTTH 4. THHTTH 5. THHTHH 6. THHTTH 7. THHTTH These offspring all compete with each other for limited resources such as food. Let's assume that there is enough food for only 4 out of 7 of these animals. Which ones, on average, get the food and survive to reproduce? Looking at the list, animals 1,3,4,6 and 7 all have the same genome, and so are equally well adapted to their surroundings. But animal 2 has a less well adapted genome, so it is less likely to be able to compete successfully with any of the other animals. And animal 5 has a "fitter" genome than any of the others, so it is a bit more likely to get the food. A reasonable outcome to expect would be that the following animals survive: 1. THHTTH 3. THHTTH 5. THHTHH 7. THHTTH Natural selection has eliminated animals 2,4 and 6. What happens in the next generation, when the surviving animals reproduce? Animals 1,3 and 7 have offspring in a similar pattern to their parent. But most of animal 5's offspring have the same genome as animal 5 itself. So, in the next generation, we might have, for example: 15 animals with genome THHTTH 5 animals with genome THHTHH 8 animals with some other genome But now, most of the 5 animals with THHTHH genome will out-compete the animals with genome THHTTH, and probably some of the other 8 animals with the different genomes. In the generation after this one, or perhaps the one after that, we will most likely find that animals with genome THHTHH will outnumber animals with genome THHTTH (which the common ancestor of all these animals had). So, we are now at a situation where most of the animals in this population are better adapted to their environments than a few generations ago. How has the change come about? It is a 2 step process: 1. Variation is produced randomly in the offspring. 2. Natural selection culls those who are "less fit", leaving those who are better adapted to their environment. So, contrary to Creationist or "Intelligent Design" claims, progression in a species, or evolution of new species, does not require any intelligent planning. Random changes combined with natural selection are enough in themselves. Speaking of the evolution of new species from old, a new species can come about simply when the genome diverges far enough from the original so that creatures with the two different genomes can't interbreed any more. The example I have given is not a particularly good illustration of that, since I assumed asexual reproduction, but it is easy to see that if we introduced sex for this species, then after many generations it may well be the case that an animal with genome HTHHTT may not be able to breed successfully with an animal which has, for example: THHHTT Any questions or comments?