Willis formed the Age and Area hypothesis during botanical field work in Ceylon where he studied the distributional patterns of the Ceylonese vascular plants in great detail. According to his hypothesis the extent of range of a species may be used as an indication of the age of that species. He also maintained that the "dying out" of species occurs rarely, and that new forms arise by mutation rather than by local adaptation through natural selection. Willis defined his hypothesis as: The Dutch botanist and geneticist Hugo de Vries supported the hypothesis, however it was criticised by the American palaeontologist Edward W. Berry who wrote it was contradicted by palaeontological evidence. Edmund W. Sinnott rejected the hypothesis and wrote "other factors than age share in the area occupied by a species". According to Sinnott factors inherent in the plant such as hardiness, adaptability and growth play an important part in determining distribution. Willis published the book Age and Area. A Study in Geographical Distribution and Origin of Species in 1922. The American entomologist Philip P. Calvert documented examples of the geographical distribution of insects that contradicted the hypothesis in a paper in 1923. On the subject in 1924, Berry wrote: In 1924, the American botanist Merritt Lyndon Fernald wrote that studies on floras of the northern hemisphere do not support the Age and Area hypothesis. Willis responded to the early criticisms and stated that his critics such as Berry and Sinnott had misrepresented his hypothesis. Willis claimed that his hypothesis should not be applied to single species but to groups of allied species. He wrote there was no rival hypothesis to his own to explain the botanical data and that his hypothesis had made successful predictions about flora distribution in New Zealand. The American ecologist H. A. Gleason praised the hypothesis for being testable in the field of phytogeography but came to the conclusion that it could not account for migration data. In 1926 Willis wrote a paper defending his hypothesis and responded to the criticism. Most scientists however had rejected the hypothesis for various reasons and according to the historian of science Charles H. Smith "The "age and area" theory attracted some interest for about twenty years, but support for it was clearly on the wane by the time of Willis's late books The Course of Evolution and The Birth and Spread of Plants."
The Course of Evolution
Willis published a controversial book on evolutionThe Course of Evolution by Differentiation Or Divergent Mutation Rather Than by Selection which was a sequel to his Age and Area. Willis questioned the adequacy of natural selection of chance variations as a major factor in evolution. He supported mutations as the main mechanism of evolution, and chromosome alterations to be largely responsible for mutations. He opposed Darwinian gradualism and favoured saltational evolution. The American ichthyologist Carl Leavitt Hubbs reviewed the book claiming Willis was advocating a form of orthogenesis: The American geneticist Sewall Wright similarly noted that Willis believed evolution was not the result of chance but an orthogenetic drive, and that he was a proponent of saltationism.
Publications
Studies in the Morphology and Ecology of the Podostemaceæ of Ceylon and India
A Manual and Dictionary of the Flowering Plants and Ferns
