Science: Ever heard a sex-crazed spider buzz?

The British Library's sound archive has been recording animals, birds and insects for 30 years. What can we learn about life on earth from the snapping, vibrating and howling of beasts?

Going to the cinema can hold a special fascination for a band of sharp-eared naturalists who target wildlife sounds. In The Madness of King George, for instance, the sound of a blackbird singing in a garden scene especially delighted Richard Margoschis, a retired environmental health officer from Warwickshire, who noted the artful use of his very own recording of the avian songster.

A similar sense of deja vu struck the veteran recordist Patrick Sellars a few years ago, during a screening of the American film version of Lord of the Flies. As the shipwrecked boys explored their tropical island, the sound-track conveyed the distinctive lilt of the curlew along with a host of other British birds. Film producers can be indiscriminate users of one of the world's great audible treasures, housed in the British Library's National Sound Archive in London.

Tucked away in the archive are some of the most intriguing and mysterious sounds ever detected by the human ear. Cacophonous or melodious, haunting and oddly compelling, their full meanings remain elusive. Thirty years old this year, Britain's unique wildlife sounds archive is now the most comprehensive aural record of life on Earth.

Nowhere else can you choose between 120,000 recordings featuring snapping shrimp, head-banging termites and the ever-so quiet songs of fruit flies. Dip into the archive and hear tropical spiders tapping their feet to the Flat Eric beat of their mating-dances. Or tune in to the sex-crazed British spider that buzzes as it hangs from oak trees.

There are bandicoots, bats, bumblebees, whales, wolves, walruses, egrets, elephants, earthworms - yes, even earthworms make faint, staccato notes in regular clusters. The challenge is to capture each animal vocalist on shellac, vinyl, reel-to-reel, cassette tape, digital mini-disc - whatever is the latest in audio-recording technology.

Ever since sound-recording techniques were first invented, just over a century ago, pioneering enthusiasts have traversed the globe lugging up to 700kg of equipment, just to catch the fleeting utterances of the world's wildlife. Trophy-hunting is part of the motivation, but there are strong scientific and conservation angles, too. A vintage Seventies recording from the archive, the bubbly call of a randy Norfolk frog, could soon make this neglected amphibian a local hero, if only posthumously. The Norfolk colony has now apparently died out, but the old recording is helping scientists to determine whether these amphibians were in fact, rather than edible frogs introduced from the Continent, "pool" frogs that had somehow survived the ice ages in remote Norfolk glacial ponds known as pingos.

Britain may in fact have had seven, not just six, native amphibians, suspects Julia Wycherley, of the Surrey Reptile and Amphibian Group. She has confirmed that the Norfolk croakers really were pool frogs, and is now comparing their calls with those of living European pool frogs. Any tell-tale signs of a distinctive Norfolk dialect in the Seventies recording would lend support to the theory that these sadly deceased frogs were a long-isolated relic population.

Among songbirds, the study of local dialects across Europe is now so fashionable that the archivists have begun urging experienced sound-recorders to take their recording paraphernalia along when they go on holiday in Europe. "We tell them, please go for common species that we get at home," says Patrick Sellars, who founded the archive in 1969, "but it is darned hard to get them to do it."

Field ecologists also frequently raid the archive for soundbites useful in "play-back" experiments. In this handy technique, researchers can quickly census breeding birds by generating virtual rivals; the broadcast of a provocative song right under the beak of a territorial male is guaranteed to elicit an indignant vocal response and sometimes resident birds even attack the loudspeaker. In judicious hands, the technique can do more than count birds; it can also lure them to safety. Conservationists recently used play-back tapes from the archive to trap the last remaining 20 pairs of the Seychelles warbler. The plan was to save them from certain death on their rat-infested island, and the experiment has been a great success. Established on a rat-free island nearby, the warblers now number 200 pairs.

Clues to evolutionary trees also lurk in animal utterances, as taxonomists increasingly realise. One recent amateur recording from Christmas Island included many birds that have never been taped before, says Richard Ranft, curator of the wildlife sounds archive.

"The recordings are not top quality, but there's still a lot you can do with a Sony Walkman and microphone. Before these recordings, no one even knew what these creatures sounded like." Now specialists can compare the voice of the rare Christmas Island hawk owl with similar species to help to determine whether the island bird is indeed a distinct species that needs special protection. It's too late for the Hawaiian o'oa'a. On tape in the archive is the lovely song of the last male, which sang alone for two seasons - the last time in 1983.

Already, the distinctive chirrups of the hundreds of species of European grasshoppers and crickets are proving to be good indicators of habitat quality and climate change. Learn to recognise a grasshopper's stridulations - the sounds it makes by rubbing bits of its anatomy together - and you can instantly spot a pristine grassland. Count the timing and frequency of the songs, and you have a natural weather gauge.

But so far most insects have eluded the naturalist's microphone. The archive has only 650 species of insect on tape - a tiny proportion of the many tens of thousands, perhaps even millions, that must make some kind of noise.

"It's a potentially vast subject that brings new technical challenges," says Dr Ranft. A whole new science of "macro-recording", on a par with the close-up skills of macro-photography, is needed to appreciate the fruit fly's whispered love song, which carries only to the distance of one centimetre.

And many insects rely on the transmission of sound waves not through the air but through mechanical contact. When one specialist stumbled across the songs of plant-hoppers - sap-suckers that infest rice crops in south-east Asia - he discovered "a whole world of communication filled with messages that can't be heard in air".

These tiny insects create vibrations strong enough to shake whole plant stems and generate extraordinary rhythms, says Dr Ranft. "When recorded mechanically by the equivalent of a gramophone needle stuck in a stem, you hear these amazing sounds, like someone drumming on a tree." The science of bio-acoustics is flourishing as never before, following recent technological advances. In one promising spin-off, voice-recognition programmes designed to respond to human voices are being adapted to help census animal populations remotely, and automatically. Stick a microphone on a satellite and it may be possible to follow the spring migration of, say, swifts flying over London from their wintering grounds in Africa. Monitoring international movements is crucial if we are to conserve animals that move across such vast distances, Dr Ranft says.

Already, the global tracking of individual whales looks possible, using information recently released by the American military. Thousands of hours of deep-sea recordings from hydrophones, planted all over the Atlantic and Pacific sea beds to detect Soviet submarines, are being made available to biologists eager to follow the migrations of whales via their distinctive songs. And the recent discovery that African elephants communicate by infrasound, at frequencies pitched too low for the human ear, probably explains how social groups keep together even over hundreds of miles of savanna.

"It turns out they've been shouting their heads off using sounds we can't hear," says Dr Ranft. "If Asian forest elephants, and maybe even rhinos, do the same, we may be able to monitor the health of their populations by listening in on these low-frequency rumbles." The possibilities are "mind-boggling".

Could sophisticated sound-recordings, coupled with hi-tech computer systems, one day make Dr Dolittle fantasies come true? Could we ever really "talk to the animals"? To date, human interjections into wild conversations remain at a primitive level.

One of the archive's erstwhile clients, Westminster City Council, plays starling distress calls in Leicester Square to try to shield cinema-goers from the droppings of thousands of roosting birds. Airports also frequently broadcast distress calls to scare away gulls, while apple-growers do the same to keep bullfinches out of their orchards.

These tactics work, says Ranft, but only up to a point. After a while the scare factor wears off, and sometimes the broadcasts even backfire. "The birds may learn that when they hear alarm calls, it means there's a huge abundance of food about, and they make a beeline for the sound."

Clearly, even canned avian screaming is open to many interpretations. So listen up, humans, and keep the tapes rolling. Forget all that "look and learn nonsense" when, all around us, there's so much more to hear and understand.

Further information at: www.bl.uk/ collections/sound-archive/ wild.html. A double CD of animal sounds `Wild Britain' is on sale at the British Library, ISBN 0712303472. Order from Turpin Distribution Services: call 01462 672555 or e-mail: turpin@rsc.org