Time is running out for rare Himalayan butterflies, says David Spencer Smith

The world's greatest concentration of high mountains lies within the Karakoram range in western Himalaya. It is an arid land of scree slopes and snow-capped mountainsides defining glaciated valleys, with melt water feeding the tributary of the river Indus.

It might seem a strange place to study butterflies; scientific attention is more usually focused on the fast dwindling faunas of the tropics - but in 1994, the University Museum, Oxford, and the Pakistan Museum of Natural History, Islamabad, started to map the distribution of butterflies in the Karakoram's ancient kingdom of the Hunza valley.

During the last century, study of butterflies of the Indian subcontinent and central Asia was apportioned through "Great Game" politics. From 1865-75 Russia occupied the Central Asian khanates, (most of these princely dominions in present Uzbekistan), and a succession of Russian lepidopterists passed through. These men got the lion's share of Central Asian work.

ÁñÁ«ÊÓƵ

Fieldwork was hazardous: British collector Charles Swinhoe, posted to Qandahar during the second Afghan war, noted: "I took a trained collector with meI So long as he lived, my collection increased: but unfortunately he was murdered one morning by a GhaziI" Even after the partition of Pakistan and India, Hunza and its adjacent valleys remained inaccessible. Work on their natural history became feasible only after the opening, in 1986, of the Karakoram Highway, a collaboration between the Pakistani and Chinese military. Now the Karakoram offers a unique opportunity to study closely related species that have diverged from common ancestors - population islands of butterflies separated by great mountains.

In Britain, it is easy to predict when adult butterflies will appear, but the workings of the seasonal clock of the Karakoram are more obscure. The China-Pakistan border crosses the Khunjerab plain at 16,000 feet; in July 1994 this was carpeted with alpine flowers - purple legumes, white saxifrages - and the few butterflies of high altitude - clouded yellows, fritillaries and blues - flew among them.

ÁñÁ«ÊÓƵ

The next year, on the same date, a cold wind blew in from China, no plants were in bloom and snow was falling. How do the butterflies of high altitude, together with their larval food plants and nectar sources, fit together in this strange ecosystem?

We have found many kinds of butterfly, probably the result of different groups being isolated by the high mountains. One group, Parnassius, is found from France to Japan and in the North American Rockies. We have found the butterfly Parnassius charltonius in the main Hunza valley and two side valleys at 10,000 to 13,000 feet. This species encompasses more than 20 geographical races or "subspecies", each supposedly distinct in wing pattern and diverging from a common ancestor through genetic isolation.

Once we have mapped the distribution of butterflies in Hunza then we can carry out further research. DNA sequencing techniques can be used to trace the butterflies' genealogical history. Are populations of a Polyommatus - a distant relative of the common blue - found above 10,000 feet in four apparently isolated valleys genetically distinct? If so, over what time period have they diverged? Have the Shimshal mountains, and the tangle of peaks centred on 26,000-foot Disteghil Sar, and the Hispar glacier, isolated a Parnassius found at 16,000 feet at Khunjerab from a population at 13,000 feet 120 miles south on the Deosai plateau of Baltistan?

Hunza has been peopled for at least 2,500 years. Not only was the valley cut off from the rest of the world, but the isolation of peoples within it resulted in the use of three separate languages. However, the Karakoram Highway is changing Hunza; high pasture is being damaged by goats, and the alpine vegetation of Khunjerab, designated a National Park primarily to protect the snow leopard and Marco Polo's sheep, is under threat from cattle and yak grazing.

ÁñÁ«ÊÓƵ

We have started to assess the wealth of insect life in Hunza; knowing what is there is a prerequisite for conservation strategy. Understanding the distribution of butterflies - indicators of the preservation or degradation of the environment - is essential.

David Spencer Smith was formerly Hope professor of zoology/entomology, University of Oxford. He works at Oxford University Museum.