The first large-scale studies of mortality cast doubt on the idea that there is a biologically fixed life span for fruit flies and possibly for humans, suggesting that after a certain advanced age, a person's chance of dying may begin to level off. Results of the work, performed on fruit flies by researchers at the University of California, Davis; the University of Minnesota; and Duke University will be reported in the Oct. 16 issue of the journal Science. Previous studies -- on populations of people, most other mammals and birds -- found that death rates gradually accelerated with age throughout the life span of an individual. But the new studies suggest that beyond a certain advanced age, death rates slow down; in other words, there is no biological clock with a preset midnight hour. "Our research shows that the oldest individuals live progressively longer," said UC Davis entomology professor James Carey. "Rather than a decrease in expected life spans, there is an increase as they get older. This is the first study with sufficient numbers to produce confidence in the mortality patterns at old age." Carey and colleagues followed the lives and deaths of more than 1 million genetically diverse Mediterranean fruit flies (Medflies) to find out if the death rates at advanced ages would imply an upper life-span limit. They did not. "The main point is that the widely held idea about an exponentially increasing death rate doesn't apply when you do experiments on a large scale," said James Curtsinger, professor of ecology, evolution and behavior at the University of Minnesota. "Most gerontology experiments start with a couple hundred individuals, so by the time you get down to the longest-living 10 percent, you're talking about only a few individuals." In a companion study also published in Science, Curtsinger and his colleagues found results similar to Carey's in smaller, but genetically homogeneous, samples of fruit flies. "The notion that there's some fixed limit to a person's life, which you inherit from your parents, is overly simple," said James Vaupel, co-author of both reports and a population analyst at Duke University's Center for Demographic Studies. Vaupel, who also holds an appointment at the Odense University Medical School in Denmark, said basic genetic function in humans and flies is similar, so comparing mortality and survival rates for humans and flies is less of an "apples and oranges" comparison than it may seem. Indeed, he said, animal models are the foundations for much biological research on human aging and disease. Mortality and survival rates are the percentages of individuals dying, or surviving, at a given age. "In the field of aging and demography, these are extraordinarily novel studies," said Richard Suzman, chief of the National Institute of Aging's Office of the Demography of Aging, which sponsored the research. "Though there are obvious problems generalizing the results to the human population, these studies are changing how we think about the survival of people at very advanced ages. They were funded as part of NIA's initiatives to understand more about the growing numbers of the 'oldest old.'" Carey's group studied Medflies raised at a facility in Metapa, Mexico, where flies are reared and sterilized to help fight Medfly infestations in California and elsewhere. In each of several experiments, mortality rates increased until about 90 percent of the population had died, then leveled off until about 1 percent of the population remained, and then decreased. Thus the flies' likelihood of death followed the expected pattern up to a certain "old" age, but after a fly had passed that age -- about 40 days in one species -- the likelihood of death at a given point leveled off and, in some cases, decreased with increasing age. "Because there are no genetically defined stocks of Medflies, one possible explanation for Carey's results might have been that the genetically inferior ones -- the 'wimps' -- died early," Curtsinger said. "But we studied 11,000 flies consisting of 10 genetically homogeneous strains, and also saw a leveling off of the death rate for the oldest flies." Researchers have found similar mortality patterns in humans 85 years old and older, Vaupel said. For example, studies by other scientists at Duke's Center for Demographic Studies show potential average human life expectancies into the 90s. This work by Kenneth G. Manton and Eric Stallard, focused on selected populations that practice healthy lifestyles. Vaupel said the analyses of fruit fly survival and mortality patterns, together with other research, suggest that by the middle of the next century, the number of people in the United States older than 85 may rise from about 1 percent, or 2 million, to 15 percent to 20 percent, or 50 to 60 million. The prevailing idea of human life expectancy, however, holds that 85 years is the average natural upper limit of human life, according to Stanford University rheumatologist James Fries, a leading proponent of this view. According to that theory, there is a point beyond which healthy lifestyles lose out to the body's biological clock. "I call that argument for fixed life spans the 'time-bomb' model," Carey said. "In contrast, these results support the 'spaceship' model: if it's sturdy enough to make it to a faraway planet, it's probably sturdy enough to go a lot further." Vaupel said individuals who survive to old age may have genetic endowments giving them survival advantages. Another possibility, he said, is if a person is lucky enough to make it to old age, there may be a slowing in the biological changes that cause mortality rates to increase with age. The "luck" in reaching advanced age may now be more widespread because of recent advances in medicine and medical technology, as well as healthier lifestyles, he said. "Now, that doesn't mean that human beings are going to live to be 999 years of age starting tomorrow. It does mean that life expectancy, which in the United States today is about 75 years, could be significantly increased. Given current knowledge, if people began following good health practices today, they might live 90 years or even 100 years, on average," he said. Carey's co-authors are Pablo Liedo, director of Centro de Investigaciones Ecologicas del Sureste in Tapachula, Mexico; Dina Orozco, associate director of Programma Moscamed in Metapa, Mexico; and Vaupel. Curtsinger's co-authors are Hidenori Fukui and David Townsend, postdoctoral fellows in ecology, evolution and behavior at the University of Minnesota; and Vaupel.