By Rebecca Morelle
BBC News science reporter
In a field in Rothamsted , Hertfordshire, lies a grid of grassy plots.
The field is divided into plots
Some are scattered with colourful flowers, others plush and green, and a few with a felt-like carpet of vegetation.
But this pretty spot has a remarkable history, for it is the site of the world's oldest ongoing ecological experiment.
The Park Grass Experiment has had a huge impact on the study of biodiversity and ecology - and this year marks its 150th anniversary.
The experiment was set up in 1856 by John Lawes - founder of the research institution now known as Rothamsted Research - and his scientific collaborator, Joseph Gilbert.
It was originally established to answer a crucial agricultural question of the day: how do different fertilisers affect hay yield.
A hay field was divided up into 20 grids, each measuring about 20m by 60m (70x195ft), and each plot was treated with various combinations of fertilisers - including different types of nitrogen fertilisers, ammonium sulphate, sodium nitrate, and later, manure; and they also added phosphates, potassium and even silica.
A bonus for botanists
"You could say the original reason they had started it had been satisfied by the 1880s," said Professor Jonathan Silvertown, an ecologist from the Open University, Milton Keynes, who has worked intermittently on the Park Grass Experiment for the last 30 years.
"It was completely clear by then what the effects of fertilisers on hay were, but something they hadn't expected was that you had all of these changes of species composition on different plots.
Harvesting Park Grass in 1941
"At that point, they noted that the experiment was now of more interest to botanists than it was to agriculturists."
As the decades passed, some changes to the original grid were made.
In 1903, most of the plots were split and one half was treated with lime to test the effects of acidification.
In 1965, most plots were divided into four sub-plots, and the pH of the soil was adjusted to pHs of five, six and seven, or left unchanged.
Over the years, the plots began to look dramatically different - and scientists were able to make some astonishing discoveries from studying these numerous combinations of nutrients, acidity, grasses and plants.
One hundred and seventy publications have resulted from the Park Grass Experiment to date.
Professor Silvertown highlighted two important findings.
He said the experiment had shown an interesting relationship between nutrients and species' populations.
"You might think the more nutrients you add, the more species you get - when in fact completely the opposite happens. The more nutrients you add, the fewer species you get. And although perhaps quite a lot of people know that these days, it was discovered at Park Grass," he said.
This is because the plants that can use the nutrients best, he said, push out the competition.
Plants like dandelions grow more readily in some plots
"The Park Grass Experiment has also revealed that evolution has taken place in the plots," Professor Silvertown told the BBC News website.
"For example, you have a plot that has become very acid because it has had ammonium sulphate added to it for 150 years and next to it is a plot that has no ammonium sulphate added.
"And some of the plants have evolved adaptations to the plot they are growing on - so if you move plants to another plot they grow much better or worse depending on the soil conditions. This reveals the power of natural selection and how it can shape adaptation on a very local scale."
While field-study has been ongoing, so has the diligent collection of plants and soils from the plots, resulting in a collection of 150 years worth of samples.
Dr Keith Goulding is an ecologist at Rothamsted Research, and, like Professor Silvertown, has been working with the Park Grass Experiment for 30 years. He says this wealth of archived material has allowed research into some unexpected areas.
Samples of soil and hay are taken and stored
"Some scientists from Southampton analysed plutonium, uranium and caesium on the grass samples," he said.
"They were able to identify plutonium coming from nuclear tests, and because of the sensitive equipment they used, they could tell exactly which nuclear tests it came from.
"We've done the same tests for other atmospheric pollutants, like dioxins and PCBs, to see how these have changed over the last 150 years."
More recent research has included looking into the impact of climate change on ecology, by investigating the impact of increasing CO2 levels on the grasses.
Dr Goulding says the long-term nature of the project makes it impossible to assess where the next discoveries will be - but the long timescales are essential for understanding future ecological problems, especially surrounding issues such as climate change
"So many of the environmental problems that we face are long-term," he said.
The differing plants show the boundaries between plots
"Short-term experiments are very good for answering specific questions, but if we want to develop truly sustainable environmental systems we need to look at them in the long-term.
"Our experiments and others like them around the world are the best way to do this."
Professor Silvertown agrees: "We need to be able to understand the past in order to predict the future.
"If we can understand how plant communities are influenced, for example, by climate, then you can predict what might happen in the future."
It seems unlikely that Lawes and Gilbert could have predicted that their original agricultural endeavour would remain active for a century and a half, let alone yield the advances in ecology that it has.
Yet the 2.8-hectare (6.9 acres) site is still answering the questions that arise from our ever-changing environment, and Professor Goulding believes it will do so well into the future.
"It won't be me stood here asking the questions in 150 years' time; but I'm certain somebody else will be."