Scientists turned literary detectives have unearthed a lost experiment in biodiversity that underpinned the thinking of Charles Darwin - and continues to have profound implications for the planet.

Reports of what was the first experiment in ecology, 40 years before the word was coined, have languished in the British Library for 175 years, according to the US magazine Science published today.

In Darwin's seminal work of 1859, On the Origin of Species, he wrote: "It has been experimentally proved that if a plot of ground be sown with one species of grass, and a similar plot be sown with several distinct genera of grasses, [in the latter] a greater number of plants and a greater weight of dry herbage can thus be raised." Unfortunately, he did not say when, or where, this study had been made.

Andy Hector of Imperial College London, and Rowan Hooper of the National Institute for Environmental Studies, decided to find out. In the last 10 years, experiments have confirmed Darwin's thesis - the more the diversity, the greater a landcape's productivity.

The two scientists looked at the draft for a much larger book started by Darwin, and then went to the British Library rare books collection, where they found the text, and the pressed and dried specimens of plants in the study.

It was conducted by George Sinclair, head gardener to the Duke of Bedford, who described it in an 1816 book and published results in the third edition of 1826. Sinclair grew a range of plant mixtures in 242 plots, each four feet square and enclosed in cast iron frames. Some of the soil analyses were provided by Sir Humphry Davy, the greatest scientist of the day.

"Darwin was interested in explaining where the diversity he observed in the world came from, and how it was maintained. That is still a central question in ecology," said Dr Hector yesterday.