GoalUse a model and evidence to test natural selection, judge explanations, and say what the evidence can and cannot show.
New words
selection pressure · a factor in the environment, such as a predator, disease or drought, that affects which individuals surviveprediction · a statement of what you expect to happen if an idea is rightevidence · observations and measurements that support or contradict an ideamodel · a simplified version of something, used to test an idea; every model has limits
Explanation
A scientist tests an idea by making a prediction that could turn out to be wrong. For natural selection the prediction is that when a selection pressure acts on a population with inherited variation, the share of the individuals with the advantageous feature goes up from one generation to the next. Scientists measure real populations, for example the finches before and after a drought. They use organisms that reproduce fast, such as bacteria and fruit flies, so that many generations can be seen. They use fossils to see changes over very long times. And they use models, such as the bag game below. In a model chance plays a part: a lucky few may survive, so the numbers go up and down. If the population is small, a type can be lost by chance. Natural selection can only work with the variation that exists, and new versions of genes come only from rare changes in DNA. So where there is no variation, there is no change. Evidence supports an idea more strongly when it comes from many different sources.
Kind of evidence
What it shows
A limit
measuring a real population over time
change happening now
only a short time, and a few species
bacteria in the laboratory
many generations in days
may not match large animals
fossils in rock layers
changes over millions of years
the record has gaps
a model such as the bag game
how selection works in a simple case
simpler than the real world
Worked examples
One run of the bag game below. L and S counters start at 10 each, with L better at surviving. Counts after each generation. The L counters usually rise, but not every time: in generation 3 there was no change, and in generation 6 there was a fall, owing to chance.
Group of finches measured
Mean beak depth (mm)
adults before the drought
9.1
adults that survived the drought
9.7
the chicks of the survivors, one year later
9.6
Imaginary data on the beak depth of finches on an island, from before a drought and after it. What pattern is there?
Prediction
Could it be wrong?
Is it a good test?
‘Birds with deeper beaks survive in a drought, and their chicks have deeper beaks too.’
yes: the chicks might not have deeper beaks
yes
‘Birds do what they need.’
no result could show that it is wrong
no
A good test of the idea: it makes a prediction about a trait that is inherited. Here two predictions are compared.
Class activity
Generations in a bag. With an adult. You need 10 beans (type L: long roots) and 10 buttons (type S: short roots) as ‘plants’, a paper bag and a die. Each generation: (1) For each counter roll the die once. S survives if it shows 1, 2 or 3. L survives if it shows 1 to 5. Take away the ones that do not survive. (2) Each survivor leaves 2 seeds of its own type: put 2 counters of that type in the bag in place of each survivor. (3) The field only has room for 20 plants: if there are more than 20 in the bag, take out 20 without looking, and the rest do not find space. (4) Write down the number of L and S counters. Repeat for 6 generations and draw a line graph. Predict first. Then repeat the whole game, and compare. Next try again with both types surviving on 1 to 5. What do you expect and why? What do the die, the bag and the room for 20 stand for? Name two things in which the model is unlike real populations. Keep small objects away from small children.
Practice
1
In the bag game, what would you expect if both counters survive on 1 to 5, and why does this help you to judge the game?
The shares of L and S would stay about the same on average, with random ups and downs, because neither has an advantage. This shows that the rise of L in the first game is caused by its better survival, not just by the rules.
2
Three pupils explain why rats have become resistant to a poison used to control them. Nodira: ‘The rats tasted the poison and learned to survive it.’ Leo: ‘A few rats already had a resistance gene. They survived, bred and passed it on.’ Aisha: ‘The poison changed the genes of every rat.’ Who is right, and what is wrong with the other two?
Pupil
Explanation
Nodira
The rats tasted the poison and learned to survive it.
Leo
A few rats already had a resistance gene. They survived, bred and passed it on.
Aisha
The poison changed the genes of every rat.
Leo is right. Nodira: learning a behaviour does not change the genes in the gametes. Aisha: the poison does not change every rat’s genes to protect them; the resistant rats existed before.
3
A farmer sprays a weedkiller on a field every year, and after ten years it no longer kills some weeds. Explain this by natural selection.
The weeds vary, and a few have genes that make them resistant. The weedkiller kills the others. The resistant ones produce seeds and pass on the genes, so each generation has more resistant weeds.
4
Imaginary data on sea snails on a shore where crabs have recently arrived are in the table. Does the data support natural selection? Give a reason, and say what extra evidence would make the case stronger.
Group of sea snails measured
Mean shell thickness (mm)
adults before crabs arrived
2.1
adults that survived two years of crab attacks
2.8
the young of the survivors, one year later
2.7
Yes: the survivors had thicker shells (2.8 mm against 2.1 mm) and their young did too (2.7 mm), which suggests that shell thickness is inherited and the crabs selected for it. Stronger evidence: more years and more shores, and a check that the thicker shells are caused by genes and not just by the food or water.