CAPE TOWN, South Africa -- High in the Silvermine nature reserve, proteas here and there unfurl skyward like floral fireworks in soft pink and yellow. Guy Midgley's eye is drawn elsewhere, though, to ugly brown lesions on the otherwise green landscape - dead protea plants.
"Nobody's quite sure what's going on," says Midgley, a plant scientist, scanning the bushy vegetation. But he suspects global warming is behind the recent protea "die-back," in which one-third of the plants have shriveled up in some parts of the Western Cape region.
Midgley bases his hunch partly on fresh evidence that rising temperatures threaten another species nearby: the quiver tree, so named because the semi-nomadic San, or "bushmen," carried arrows in its hollowed-out stems.
Half the quiver tree's range is in danger as warming turns semi-arid areas of Namibia and northern South Africa less hospitable, he says. The trees, which can live up to 350 years, are dying in the northern half of their range, according to a study in the September issue of Diversity and Distributions.
"It's the canary in the coal mine," said Midgley, a co-author of the study, calling it the first proof that a desert plant's whole range is shifting. "It's one species, but it's an index."
The research echoes findings on butterflies in Europe and frogs in Central America, bolstering the view that a hotter planet is affecting biodiversity far beyond the Arctic, said University of Texas biologist Camille Parmesan, who was not involved in the study.
As study co-author Lee Hannah of Washington-based Conservation International noted, "It's not just about polar bears anymore. Here is evidence that species in many other parts of the world are at risk."
The quiver tree, a species of aloe distantly related to aloe vera, shows the potential for a gradual migration toward cooler and wetter conditions: trees at the southern range are doing well. The fear is that the species will be squeezed so fast from the north that it could face eventual extinction. Plants, dependent on seed dispersal, naturally shift at a glacial pace.
A possible short-term measure would be to transplant some populations of trees, said Midgley, chief specialist scientist here at the South African National Biodiversity Institute's Global Change Research Group. The practice is controversial in conservation circles because it might mix genes and in extreme cases even create new invasive species that can harm ecosystems. It is also extremely expensive if done on a large scale.
Ultimately, he said, the solution is much bigger. "What we've really got to do is stop greenhouse gas emissions and slow down the rate of global warming. There is some temperature above which everything will go extinct."
The discovery of the quiver tree's peril came about after scientists heard rumors several years ago that the trees were inexplicably dying. A researcher named Wendy Foden went into the field to learn more about the distinctive tree, a succulent that can resemble an upright broccoli shoot and is also called the aloe tree.
A statistically clear north-to-south pattern of quiver tree mortality emerged from a detailed census and visual comparisons made using old photos. Then the research turned to why this was happening.
"We tested every explanation we could think of - grazing, fungal disease," Midgley said. Mining or other localized human impact was ruled out. "The only thing that made sense was global warming." He added that "desert plants are often seen as being incredibly resilient. They're not. They're very finely tuned to cope with the environment."
Parmesan said the tree's slow rate of southward migration is worrying. Even though plants migrate slowly, she said she would expect some progress. In her studies of butterflies in the 1990s, she found that extinction closer to the equator - in North Africa and Spain - was matched by expansion into the northern climes of Sweden and Finland.
Even though quiver trees cannot migrate as quickly as butterflies, she said, "it is a puzzle as to why they're gradually going extinct rather than showing any kind of shift."
That finding, she said, resembles what happened with the garishly yellow, black and red harlequin frogs of Central America, 75 percent of which have been killed by a fungus that benefited from rising temperatures. Frogs with the highest death rate lived just below the mountaintops, yet for some reason they did not move up to cooler elevations.
"Maybe," she said, "there is a category of organisms that for whatever reason is so habitat-specialized - we think they can move, but for them it's not habitat. Or they have such poor dispersal that it is being squeezed."
Back in Cape Town, Midgley is eyeing a type of fungus as a possible culprit in the alarming protea deaths, a phenomenon observed only within the past five years. He thinks temperature increases in the past decade might be making the much-loved plants more susceptible to naturally occurring fungus.
Proteas are grown in California for the American cut-flower trade, but South Africa is the global hot spot of their diversity. Numerous varieties have evolved over time in the region's Cape floral kingdom, the smallest but most diverse of the planet's six floral kingdoms. The king protea is South Africa's national flower, and the national cricket team is called the Proteas.
Midgley hopes to collaborate with scientists in Australia, home to a different group of proteas that is also under stress, to try to solve the mystery.
The known facts, including patchy die-offs in areas that are several hundred miles apart, point to some regional change in the environment, he said, as opposed to an invasive species or some other isolated cause.
"What is changing regionally? One thing we know is temperature," Midgley said. "It's very hard to come up with an explanation that fits all the facts - apart from a regional change in temperature and possibly rainfall."