Klondike miners call it muck, an unromantic name for the thick layer of black silt that fills most of the valleys, and must be stripped away before the gold-bearing gravels lying underneath are exposed.

When this muck is moved, it sometimes uncovers a different sort of pay dirt -- ancient soils called paleosols that can hold a wealth of information about past environments. Usually buried deep underground, these paleosols can contain traces of pollen and fragments of small bones from the plants and animals that once lived in the area.

Last summer a paleosol uncovered in the Klondike offered up a soil scientist's version of a bonanza. Scott Smith is one of a small number of scientists who study paleosols. He was working on another project in the Klondike when he heard about a discovery on another claim in the area.

Just before he visited the site, the miner had ripped off a section of muck and uncovered large stumps from a buried forest that could be tens of thousands of years old. The stumps, still rooted in the ancient forest floor, were possibly the remains of white spruce trees.

The wood has not been dated yet, but Smith is guessing that the trees were alive between the last two major glacial periods in the Yukon. He says that he has never seen anything like these stumps from this interglacial time period, which ended about 30,000 years ago.

During the last ice age, massive glaciers covered most of the Yukon except for the northern and central regions. The Klondike was part of Beringia, a vast unglaciated area that stretched east from Siberia to the Mackenzie River. If glacial ice had scoured the Klondike, it would have dispersed the rich concentrations of placer gold there and churned up well-preserved paleosols.

It was thought that the scraggly forests surviving in the Klondike between glacial advances degenerated into tundra as the climatic conditions became more severe. This buried forest could help paint a different picture of the vegetation in this area during the ice ages.

The origins of the muck that helped to preserve this ancient forest are still something of a mystery as well, but are thought to be the remains of loess, or wind-blown silt, that once blanketed much of Beringia. Scientists think that loess eroded off the uplands and collected in the valley bottoms where it froze solid, preserving any organic matter mixed in with it.

Usually soil scientists must study fossil pollen, not tree stumps, to determine what plants once grew in an area. Pollen can keep its characteristic shape for millions of years, and can be used to identify the genus of different plants.

The animals that lived in the area can be determined by looking for bones mixed in with the soil. The remains of small animals such as lemmings and ground squirrels are found most often. Fragments of beetle shells are also good indicators as beetles live in very specific environments.

Soil scientists also compare paleosols with modern soils to determine what sort of climatic conditions prevailed when the soils were still active. For example if they find paleosols in which the humus has been swirled around and mixed up, they know that they are looking at cryosols, which are formed by the churning action of permafrost.

At the other end of the spectrum are podzols, bright red soils that are characteristic of moist forests. "You can see the same soils all over the world in temperate humid conditions," says Smith. In each case the soil type can be linked to a specific climatic condition.

The most amazing fossil evidence that Smith has ever seen was on Axel Helberg Island, located west of Ellesmere Island at latitude of 80 degrees north. Ten years ago he studied a fossil forest of deciduous conifers that had been uncovered there. The trees, known as dawn redwoods, are relatives of the sequoia trees that grow in California today.

"What is so incredible is that there was once a huge forest in the high arctic with trees one metre in diameter," he says. The trees grew in a swamp forest with ferns and other lush vegetation.

Even though the climate would have been warmer then, the length of the days year-round would have been the same, with three months of continuous daylight in the summer and three months of darkness in the winter.

The paleosols and fossil forest were well preserved because they were located on a delta below a range of large mountains. Periodically floods would bury the forest floor with sediments from the mountains, preserving the organic matter underneath.

"You can find fossil wood that looks like it was deposited ten years ago. You can even burn it," he says.

Until last year Scott Smith worked for Agriculture Canada in the Yukon. He now works for the Pacific Agri-Food Research Centre in Summerland, B.C., but will continue to come north regularly to study paleosols. For more information on this work you can contact him at smithcas@em.agr.ca.