The BBC's Costing The Earth programme investigates the safety of chemicals in the environment and how they could be altering the behaviour of wild animals.
Are synthetic chemicals leaching into the environment and changing behaviour?
The humble stickleback lives under constant threat from predators such as herons.
Many times a day, a stickleback, eyeing a piece of food, will have to decide whether it is worth breaking cover to get the food, and risk getting eaten itself.
The over-bold stickleback will not live long, but the excessively timid will starve.
Just how bold the little fishes are depends, it seems, on what kind of water they were brought up in.
Alison Bell, from Glasgow University, UK, has reared sticklebacks in water that contains tiny concentrations of a synthetic oestrogen - the kind of level of pollution that you might easily find in the environment.
She found these fish were significantly more willing to take risks than those that were reared in pure water.
Alison Bell's work is one of a growing number of studies that show that animal behaviour can be influenced by environmental chemicals.
And it suggests that these effects might be seen at very low concentrations of chemicals - levels which, until now, everyone thought were simply too low to have an effect.
"There are now lots of studies that show exposure to low levels of chemicals can affect behaviour," she tells BBC Radio 4's Costing The Earth programme.
"We are barely scratching the surface."
Stuart Rhind and colleagues, at the Macaulay Land Research Institute near Aberdeen, did something similar.
They took two sets of sheep, and grazed one set on ordinary grass. The second group they grazed on land that had been fertilised with sewage sludge.
Sewage sludge contains a complex mixture of chemicals that go under the general description of "endocrine disrupters" - chemicals that somehow fool the body into thinking they are natural hormones (again, the levels in the sludge are very low).
Stuart Rhind could find no differences at all in the bodies of the two sets of sheep, or their lambs. What he did find, however, was that the lambs born to the ewes grazed on the sewage sludge behaved differently to their uncontaminated counterparts.
The male lambs behaved in a much more "feminine way" - which, in lambs, means being more inquisitive.
"It's not the fact that there has this been this particular change of behaviour that matters," explains Rhind. "It's the fact that behaviour has been altered in some way that matters, because it implies that the developing offspring have had their brains altered."
Boys and girls
If tiny amounts of these chemicals, applied to animals while in the womb, can alter their brains, what about humans?
It is much harder to do a controlled experiment with babies, but there is one intriguing piece of research from the Netherlands.
Paediatricians there measured the concentration of two endocrine-disrupting chemicals in the blood of pregnant women. The levels were in the range you would expect to get from eating a normal mixed diet.
Then they looked at the behaviour of the children once they reached school age.
They found that the boys whose mothers had had the most exposure were more likely to play in a "girly" way - with jewellery, and dolls, for instance; while the girls whose mothers had been most exposed were drawn to boyish toys such as guns and trains.
According to investigator Dr Nynke Weisglas Kuperis, "it was a significant effect, but it is very subtle".
These and many other results with laboratory animals have led some scientists to call for a radical overhaul of the way chemicals are tested for safety.
They claim to be able to detect adverse effects at levels far below those deemed safe by the regulators.
So why aren't the rules changing? Perhaps because this is an extraordinarily tricky area of science.
While some researchers find these effects, others who have tried to repeat the experiments in large, well-controlled trials, have found none at all.
The people who do not find the effects are, by and large, those working in industry - a fact that leads some academic scientists to suspect a conspiracy.
But the fact remains that, until these low-dose effects can be consistently replicated, we are left with worries and suspicions, but not proof.
For more, listen to Costing The Earth on BBC Radio 4 on Thursday, 28 April, at 2100 BST. The edition will be archived on the programme website.