The formation of new species is a gradual and not a sudden process, according to a team of biologists from the UK, France, Australia and the USA.
Their findings, from a study of birds on Pacific islands, are reported in today's Proceedings of the National Academy of Sciences (PNAS).
The "founder effect" theory, a controversial idea among biologists, says that speciation occurs suddenly due to a small influx of colonists founding new populations, in the process creating many new gene combinations and losing many others, in what is known as a "genetic revolution."
But according to the team's new evidence from fieldwork and computer modeling, the theory doesn't apply to island birds, and the way in which populations change their genetic diversity is a result of successive colonization events and long-term genetic drift.
"Our results indicate that speciation in island birds occurs gradually, not suddenly as a result of island colonization through founder effects," said Dr. Sonya Clegg of Imperial College London.
In order to carry out the work, Dr. Clegg and her collaborators visited a series of islands through the southwest Pacific.
Dr. Clegg said: "The result is exciting because this is the first time the theory has been tested using natural populations. Previous tests have used artificially introduced ones, which don't tell you much about how real biodiversity evolves.
"It's obvious that genetic changes can occur if a single pair of individuals founds a population, but the question is whether that really happens. Our results suggest that it doesn't."
The scientists tested the founder effect model in an unusual island bird species, the Silvereye (Zosterops lateralis), which has colonized a series of islands in the southwest Pacific from the Australian mainland during the last 200 years.
The Silvereye's pattern of colonization there was originally used to support the founder effect model, first proposed by Ernst Mayr in 1954. Though subsequently tested in lab studies and artificial populations, up until now the theory had been untested in a natural population because of the need to know the colonization date for a series of island populations.
The southwest Pacific islands benefit from a well-documented history of such events, and DNA samples were taken from birds captured on the islands to be compared with samples from Silvereye colonies on the Australian mainland and with samples from island populations of Silvereyes known to be founded over 3,000 years ago.
This allowed the scientists to make the crucial contrast between genetic changes in recent and old island populations.
Their computer models of evolution based on these data showed that single founder effects do not lead to strong genetic changes in the Silvereye population, contrary to the founder effect model.
Instead, genetic changes build up gradually -- either as a result of long-term genetic drift or through multiple bottlenecks. The authors estimate that the average successful founding flock would number more than 100 birds.
Laboratory-based studies have shown that founder effect models can work, but only under very restrictive circumstances, for example, with a very small number of colonists. A natural population with known colonization dates is very rare and most studies have been done on artificially introduced populations, which limit the value of subsequent results because humans have set the size of the colonizing group.
The research team included scientists from the Department of Biological Sciences and Natural Environment Research Council Centre for Population Biology, Imperial College, London; the Department of Zoology and Entomology, University of Queensland, Australia; Centre de Biologie et de Gestion des Populations, Campus International de Baillarguet, Montferrier/Lez cedex, France, and the Museum of Vertebrate Zoology, University of California, Berkeley.
The Australian Research Council and the Stuart Leslie Fund (Birds Australia) funded the work.
(Reference: Genetic consequences of sequential founder events by an island-colonizing bird. Authors: Sonya M. Clegg, Sandie M. Degnan, Jiro Kikkawa, Craig Moritz, Arnaud Estoup and Ian P. F. Owens. Proceedings of the National Academy of Sciences. May 28, 2002.)
Related websites:
Proceedings of the National Academy of Sciences (PNAS)
Imperial College of Science, Technology and Medicine
[Contact: Dr. Sonya Clegg, Tom Miller]
28-May-2002