Evolution, extinction and biodiversity · Process
Natural selection
Not one animal in the population changes. The population changes, because of which animals had offspring — and that is a completely different mechanism from the one most people imagine.
Start here
Giraffes have long necks. How did that happen?
The obvious story is that giraffes stretched for high leaves, their necks lengthened a little, and they passed the extra length on. It is a clear, sensible explanation, it was the mainstream scientific view for fifty years, and it is wrong in a way worth understanding precisely.
What actually happened over those generations?
Neck length already varied. In hard times the longer-necked animals reached food the others could not, survived better, and had more offspring — who inherited the versions of the genes that made necks long. No individual giraffe ever grew a longer neck. The average moved because of who had calves and who did not.
At the bench · a hundred moths on a tree trunk
Run the generations
generation 0
Pale and dark moths resting on tree bark, hunted by birds that find whatever they can see. The population changes while no individual moth changes at all.
The tree bark
pale on top, dark below · one column per generation · oldest on the left
Pale 90%
Dark 10%
Nine moths in ten are pale, which is where the British population started.
Half the moths are pale and half are dark — not where the British population started, but the clearest place to watch a change from.
Neither colour has an advantage on patchy bark, so the proportions barely move however long you run it. This is the control: without a difference in survival there is no selection, only a population sitting where it already was.
The population is now overwhelmingly dark, and not one moth ever changed colour. Every moth that has ever lived in this simulation was born its colour and died its colour. What moved was the proportion, generation by generation, decided entirely by who was eaten.
On clean lichen the pale form is nearly invisible and the dark form stands out, so the population stays pale — or returns to pale, if you have switched back after running it sooty. Selection has no memory and no direction.
The proportions are shifting. Notice how little changes in any single generation and how much has changed after ten — that is the whole reason evolution is hard to see happening.
The process, in five steps
Every step is something you have already met.
Variation already exists
Individuals in the population differ, and the differences are there before anything changes. In the moths, pale and dark forms both existed long before the factories were built — dark ones were simply rare.
More are born than survive
Every population produces far more offspring than the food, space and shelter can support, so most of them will not live to breed. This is where the competition you met in Variation and competitive success comes in.
Some survive better than others
Whichever variations happen to suit the current conditions make survival more likely. Nothing is choosing; the environment simply kills a higher proportion of one kind.
Survivors reproduce and pass it on
The survivors are the ones that breed, and they pass on the versions of the genes they carry — which are, on average, the ones that helped them survive. This is heredity doing the work, and you met it in Passing it on: heredity.
Repeat, and the population shifts
A small change each generation, accumulated over many generations. Nothing dramatic happens in any single generation, and after fifty the population looks different.
Key fact
Natural selection: individuals vary; those whose variations suit the conditions survive and reproduce more; they pass those variations to their offspring; over many generations the population changes. Individuals do not change — populations do.
Think again
“Animals change themselves to suit their environment, and pass the change on.”
This is Lamarck's idea, and it deserves respect rather than mockery: it was a serious scientific theory, it explained the observations available in 1809, and Lamarck was right about the big thing — that species change over time — when most people still thought they did not. It fails on the mechanism. A blacksmith's children are not born with thick arms; a mouse that loses its tail has ordinary tailed offspring; and generations of people cutting their hair has produced nobody born bald. What an organism does during its life does not rewrite the DNA in its gametes, so it cannot be inherited. What natural selection needs instead is variation that is already there, arriving before the environment changes rather than in response to it. The moths on the bench above were not turned dark by soot; some were already dark and the soot changed which ones got eaten.
“The population needed to change, so it did.”
Nothing in this process is aiming at anything. There is no need, no goal, no trying — natural selection is the accumulated arithmetic of who happened to leave more offspring, and it cannot look ahead. The language makes this hard: it is almost impossible to describe evolution without saying a species developed a feature in order to do something, and every one of those phrases smuggles in a purpose that is not there. Watch the wording in the mastery ladder and in your own answers. The consequences of getting it right are real: a population facing a change it has no existing variation for does not conjure one up, it dies, and that is why extinction is common rather than rare. And it is why antibiotic resistance appears within years of a new drug — the resistant bacteria were already in the population before the drug arrived.
Mastery ladder
Not started yet.
Rungs 3 and 4 you mark yourself.
Rung 1 · What changes
After fifty generations on sooty bark, most of the moths are dark. What changed?
Rung 2 · The one that catches people
Why is "giraffes stretched their necks and passed the extra length to their calves" wrong?
Rung 3 · Explain it properly
Explain how a population of mainly pale moths became a population of mainly dark moths after the bark was blackened by soot. Use all five steps, and be careful not to say the moths changed.
Rung 4 · Take it somewhere new
A farmer sprays a pesticide that kills 99% of the insect pests. Within three years it barely works at all. Explain what has happened, why spraying more of it makes the problem worse, and what this shares with the moth bench.
Key note
Natural selection needs four things: variation that already exists in the population, competition because more are born than can survive, a difference in who survives and reproduces, and inheritance so the surviving variations are passed on. Repeat over many generations and the population changes. No individual adapts during its own life, and the process has no goal.
Going further
Ask how do we know? about the moths and you get the best answer in this book. Bernard Kettlewell tested the idea directly in the 1950s: he released marked pale and dark moths into a polluted wood near Birmingham and an unpolluted one in Dorset, recaptured what he could, and in each wood the better-hidden form came back in greater numbers. It was in every textbook within a decade — and then the method was taken apart. He had released moths by day, at densities no real wood would hold, and often onto exposed trunks, when peppered moths mostly rest higher up beneath branches. Those criticisms were fair, and for a while they were reported as though the conclusion had fallen with the method. It had not. Michael Majerus spent the 2000s running the experiment the way the critics said it should be run — moths in natural resting positions, at natural densities, with birds watched taking them — and got the same result, more strongly; the findings were published in 2012, after his death. That sequence is not an embarrassment for science, it is the whole of science in one example. A conclusion is only ever as good as the method behind it, so someone attacks the method, and a conclusion that survives a better test than the one that first produced it is standing on firmer ground than before.
Antibiotic resistance is natural selection happening fast enough to be a public health emergency. In any large population of bacteria a few individuals happen to carry a version of a gene that lets them survive a particular antibiotic — they were there before the drug was ever used, produced by ordinary random mutation. Give the antibiotic and it kills the rest, which leaves the resistant few with no competition and an empty patient to reproduce into. Bacteria divide every twenty minutes, so a population that is almost entirely susceptible in the morning can be almost entirely resistant within days. This is exactly the moth bench with the generation time reduced from a year to twenty minutes, and it is why finishing a course as prescribed, and not demanding antibiotics for a virus, are not fussiness. Every use of an antibiotic is a selection pressure applied to every bacterium in the person taking it.
Before this lesson
Connects to
At GCSE this becomes
- Darwin and Wallace, the evidence for evolution, speciation, mutation as the source of new variation, and selective breeding as the contrast case.
Where to next
Ask Mr Badmus AI
Want to check whether your explanation smuggles in a purpose?
The moth bench is a teaching model: survival differences are fixed percentages, the population size is held constant, and there is no mutation, migration or chance beyond the survival rates themselves.
Lesson content © MrBadmusAI.