Born at Nantwich in Cheshire, Mather graduated from the University of Manchester in 1931. He then joined the John Innes Horticultural Institution at Merton, Surrey, where the chromosome theory of heredity was then being developed. Here Mather investigated chromosome behavior, especially crossing over, his research being influenced by his association with Cyril Darlington.

Mather gained his PhD in 1933 and then spent a year at the plant breeding institute, Svalöf, Sweden. Experience at Svalöf convinced him that characters that vary continuously through a population are extremely important in breeding work. On his return to England he took up a lectureship at University College, London, under Ronald Fisher, who was developing statistical techniques that could be used to analyze such quantitative variation.

In 1938, after a year with T. H. Morgan in America, Mather returned to John Innes as head of the genetics department. It was already appreciated that quantitative variation is governed by many genes, each of small effect, and Mather termed such complexes ‘polygenic systems’. He demonstrated that by applying selection to continuously varying characters one could greatly increase the range of variation beyond that found in the normal population. Continuous variation cannot be analyzed satisfactorily by conventional segregation ratios and Mather thus applied statistics to his results, terming this combination ‘biometrical genetics’.

In 1948 Mather became professor of genetics at Birmingham University, where he remained until his appointment as vice-chancellor at Southampton University in 1965. As founder of biometrical genetics he wrote a number of books on the subject, which he greatly developed during his time at Birmingham in collaboration with J. L. Jinks. Mather returned to Birmingham in 1971 as honorary professor of genetics.