Mountains cover about 25 percent of the Earth's land surface. They are found on every continent and at every latitude. And because of their sharp elevational gradients, they are centers for biodiversity. In the White Mountains, for example, plants typical of the Mojave Desert lie surprisingly close — less than two vertical miles — to plants found near the Arctic Circle. In between lie other distinctive biomes, with limber and bristlecone forests rubbing elbows with sagebrush steppe and pinon-juniper woodlands. Given all the opportunities for short-distance migration here, it is hard to imagine a place where, in coming years, the jostling for advantage will be more intense.

Over time, it is expected that more species will do what Millar's pines are doing — move in response to rising temperatures. As a result, the number of species at any given elevation is expected to increase. Project GLORIA was founded after botanists at the University of Vienna sampled the plant communities on more than two dozen summits in the Alps and compared them to historical records. They identified nine species that were moving upslope at a pace that ranged from three to 12 feet per decade. Between 1994 and 2004, the summit zone of 11,497-foot-high Mount Schrankogel experienced an 11 percent gain in species diversity. Last month, at a conference held in Perth, Scotland, GLORIA's organizers reported that other mountainous areas in Europe are recording similar gains.

An increase in biodiversity may sound like a positive development, but in this case, it raises concern. In general, the higher you go, the cooler it gets, but that's true only in a relative sense. As the world warms, the high places are warming, too. Eventually, many scientists fear, an unknown number of species will find themselves trapped on mountaintops, unable to move any higher. By the end of this century, some could even face local or regional extinction. It's a scenario with such apocalyptic overtones that Rob Klinger, a U.S. Geological Survey ecologist based in nearby Bishop, calls it the Rapture Hypothesis.

But high-living species haven't run out of options just yet. Craggy terrain, as all mountain-dwellers know, harbors an expansive envelope of climatic conditions — warmer, colder, wetter, drier — that do not conform to elevational lines. West-facing slopes are both windier and rainier than those that face east. And north-facing slopes receive less direct sunlight than south-facing ones, making them moister and cooler. Now, new measurements are giving quantitative heft to these differences. Whether a slope faces north or south is not a trivial thing, says ecologist Stuart Weiss of the Creekside Center for Earth Observation in Menlo Park, Calif. "In fact, it's equivalent to about 1,500 feet of elevation and more than 5 degrees Fahrenheit in mean annual temperature."

There is also the widespread phenomenon of cold-air downwelling. At night, after rocks shed the heat they've absorbed during the day, the air at high elevations quickly cools. Then, because this cool air is denser than the warmer air below, it starts rolling downhill, into valleys and ravines. These "cold holes," as they're sometimes called, can prove surprisingly persistent. They occur in summer as well as in winter and on every scale imaginable. A mountain meadow, though relatively flat, has enough topographical roughness that a cold summer night can cause frost to form in some spots but not in others.

Within an area no larger than a football field there can be dramatic differences. Recently, for example, Weiss and Chris Van de Ven, a geologist from Michigan's Albion College, deployed an array of 26 iButtons, spaced several hundred feet apart, to generate a temperature map of the area around Crooked Creek. At dawn, they discovered, from late July through early October, temperatures on the valley floor are a good 15 degrees Fahrenheit lower than on surrounding slopes. Even in August, nighttime temperatures frequently dip into the 30s, sometimes below freezing.

The chill may help explain why young pines, though moving downslope in many areas, continue to avoid the floor of the Crooked Creek Valley. By contrast, the much-higher ridge I visited with Millar and Westfall seems almost balmy. There, the iButton record for the 2008 and 2009 growing season shows that nighttime minimums rarely dipped below 40 degrees Fahrenheit, while daytime readings frequently climbed into the 50s. Physiologically, this is important. In order to grow, trees apparently require a substantial stretch of time when temperatures stay reliably above 40 to 45 degrees. Where summer temperatures dip too low, they cannot make new pith, new heartwood or new cambium.

Still, it's hard to attribute all the observed tree growth to just one factor. Some experts, for example, point to the fertilizing effect of carbon-based gases. (For photosynthesizing plants, carbon is an essential building block; as carbon builds up in the atmosphere, many trees, shrubs and flowering plants are expected to grow significantly faster.) Then, too, as one descends in elevation, decreasing soil moisture becomes an important constraint. In the White Mountains, the precipitation gradient is steep, dropping from 20 inches a year on the peaks to a scant 5 inches on the Owens Valley floor. As a result, limber pines mostly peter out below about 9,400 feet, partly because of the decline in precipitation and partly due to higher rates of evaporation driven by warmer temperatures.

But there is an intriguing exception, Millar notes. A small colony of young limber pines can be found more than 2,500 feet lower, on the steep north-facing side of the Owens River Gorge. It's certainly cooler in the gorge, and moister, too, but that being the case, why haven't limber pines found their way there before? The same question could be asked about all the other young pines Millar has found deep-diving into Great Basin ravines. The answer, she suspects, can be found in the past, when the region's climate was significantly cooler and wetter. As recently as a century ago, these same ravines were likely choked with willows and cottonwoods, making it all but impossible for limber pines to take root. Now that the climate is warmer and drier, the riparian vegetation has died off, opening up habitat for young pines.

Whoosh. Whoosh. Whoosh. Despite his large frame, John Smiley pirouettes with the grace of a ballet dancer as he swipes his gauze net through the air. Soon he's caught the fluttery thing he spotted out of the corner of one eye. It's a clouded sulfur butterfly, pale lemon in color, with paper-thin wings and delicately patterned eyespots. He's delighted with the find — the species is new to this study — but does its presence here mean anything? The veteran biologist shrugs. A common butterfly in many parts of the United States, the clouded sulfur has been spotted before in the White Mountains, albeit a few hundred feet lower.

Smiley is the associate director of the White Mountain Research Station. He is also the leader of the annual butterfly count. I've been trailing behind him all day. Our first stop, a rocky meadow beneath White Mountain Peak, yields a fair number of butterflies called Shasta blues, though not quite as many as Smiley found here two years ago. Then, he and his colleagues counted a total of 875 Shasta blues in the space of just half an hour. The previous record was 230 sighted over a 24-hour period. "The Shasta blue is not a common butterfly," Smiley reflects. "So what we may have found here is a core area for the species, one from which it spills out to colonize other areas. Maybe Shasta blues do really well only in high alpine locations."