Richard Melville Cormack, 1935–2026
本文是理查德·科马克的讣告，回顾了他作为标记重捕理论先驱的生平与学术贡献，包括提出Cormack-Jolly-Seber模型、处理捕获概率异质性等，对统计生态学研究者有参考价值。
Richard Cormack passed away on 20 February 2026, aged 90. Richard was one of the giants who developed the theory of mark-recapture. He was the first person to develop a non-deterministic model for open populations, preceding work by Jolly and Seber; most survival models for mark-recapture are developments of the ‘Cormack–Jolly–Seber model’. Richard was born on 12 March 1935, the only child of William Sloan Cormack and Jean Wilson Cormack. His father was Principal of Stow College of Engineering in Glasgow. From the age of 7, Richard attended Glasgow Academy, obtaining six Scottish Highers at the young age of 14. He left 3 years later, and entered directly into the second year at King’s College, Cambridge, intending at the time to be a theoretical astronomer. Despite the tutelage of Fred Hoyle amongst others, Richard pivoted into the realm of mathematics at the end of his second year, and after securing first class honours in Special Mathematics from the University of London as an external student in 1954, and second class honours in Mathematics from Cambridge in 1955, Richard left Cambridge in 1956 with a Distinction in the Diploma in Mathematical Statistics. Richard’s PhD, undertaken while a lecturer at Aberdeen, and supervised by David Finney, was titled ‘Some Problems of Genetic Equilibria’, and was completed in 1961. Richard’s period at Aberdeen (1956–66) coincided with a golden era for statistics there, and in addition to David Finney, his colleagues included Bill Brass, Peter Fisk, David M.G. Wishart, Michael Sampford, Robert Curnow, George Jolly, and Andrew Rutherford (the last four being members of the ARC Unit of Statistics). In common with a number of these colleagues, he moved to Edinburgh in 1966, holding a Senior Lectureship there until 1972, when he became Professor of Statistics at St Andrews. Richard’s introduction to mark-recapture was a study of fulmar survival rates, carried out by George Dunnet. Richard’s analyses were published in British Birds in 1963, but it was his paper in Biometrika in 1964, laying out a general framework for estimating survival rates from marked animals, that established his reputation. It was the first paper to solve a non-deterministic model for open populations. In the following year, George Jolly and George Seber independently developed the Jolly–Seber model, which also allowed estimation of numbers of births and population size. Subsequently, survival models for mark-recapture data that use the framework of Richard’s 1964 paper have become known as Cormack–Jolly–Seber models, after that term was adopted in a 1992 monograph by Jean-Dominique Lebreton, Ken Burnham, Jean Clobert, and David Anderson. It is noteworthy that Richard and George Jolly met daily over coffee and played squash together during the 1960s, yet their trail-blazing papers were written entirely independently. Each was a statistician teaming up with separate groups of field biologists, and their research interactions were very much directed at what was needed by those groups, rather than reaching out to statistical colleagues. Richard’s ground-breaking contributions continued when he addressed the issue of heterogeneity in capture probabilities, publishing a test for heterogeneity in Biometrics in 1966. Then in 1972, in another Biometrics paper, he showed the logic behind capture–recapture estimates, making the methods more accessible and understandable to the user community. In 1981, jointly with Philip North, Richard published important insights into mark-recovery models. At about the same time, he commenced his work on loglinear models for mark-recapture, which led to papers in Biometrika in 1984 (with Ron Sandland) and 1991 (with Peter Jupp), and in Biometrics in 1989, and additionally, to four book chapters. There was also a sequence of Biometrics capture–recapture papers in the 1990s: on modelling covariates (1990), on interval estimation (1992), and on variance estimation (1993). After retirement in 1994, his publications in mark-recapture were as co-author in epidemiology studies. In 2016, a special issue of Statistical Science (31, 2) was devoted to the 50th anniversary of the capture–recapture papers by Cormack, Jolly, and Seber and the issue also included an interview with Richard that gives further details of his life (Buckland, 2016). Richard also published on other diverse topics, often with scientists from other disciplines. His 1971 book, The Statistical Argument, was used as a general introductory text by many universities. His 1971 review of classification, read to the Research Section of the RSS and later appearing in JRSSA, is a classic, and while his 1988 exposition on statistical challenges in the environmental sciences (also in JRSSA) has had substantially less impact, it too showed the incisiveness that those who knew him associated with Richard. Many colleagues witnessed his incisiveness and forthright manner over the years, yet socially, he was courteous and kind, and excellent company. His contributions to a wide range of committees, working groups, visiting groups, and scientific organizations (including council member for NERC and the Freshwater Biological Association) were substantial. For the Royal Statistical Society, he was elected as a fellow in November 1956, was a Council member from 1980 to 1984, a member of the Research Section Committee from 1972 to 1975 and 1986 to 1987, a member of the Honours Committee from 1994 to 1995, and an assessor on the Examinations Board from 2000 to 2004. He was elected a member of the ISI in 1962 and a Fellow of the Royal Society of Edinburgh in 1974. He held various offices within the Biometric Society, as Secretary of the British Region from 1970 to 1977, Regional President from 1991 to 1992 and President of the International Society from 1980 to 1981. Richard’s achievements were not restricted to the academic field. He was school captain for cricket, and played twice for Glasgow Schools against Edinburgh Schools. He also played twice for Cambridge University 2nd XI, was an Aberdeen University blue, winning the Scottish University Championship, and played for Aberdeenshire. He was a Cambridge half-blue as goalkeeper at lacrosse, winning the South of England League and the all-England Iroquois Cup. He also played for Cambridge University’s second team at chess. He was also active musically at Cambridge, organizing the Consolidated Operas Incorporated at King’s in his final year, and he sang in two concerts at King’s College. Richard married Edith Whittaker on 1 September 1960, at King’s College Chapel, Aberdeen. Edith (who passed away in 2024) was a plant ecologist, and a past chairperson of the Fife & Kinross Branch of the Scottish Wildlife Trust and of the Friends of St Andrews Botanic Garden Education Trust. Their son Andrew (who passed away in March 2023) was Chief Regulatory Advisor to the Joint Information Systems Committee (Jisc), while their daughter Anne is a trainer in Carbon Literacy. Photography was a passion of Richard’s for many decades. He was a lecturer and judge for 40 years for the Scottish Photographic Federation and was placed on their roll of honour. He had exhibitions in Dundee (Land of the Berbers) and St Andrews (Growth and Form) in 2010, and gave many talks. Hill-walking was another passion, and he climbed more than 200 Munros. After retirement, he and Edith went on many extended treks in mountainous regions abroad, mostly to Morocco, on trips led by Hamish Brown. These included the first known British ascent of Jbel Fazzaz. Richard’s other interests included natural history and gardening. Richard was instrumental in building up strength in statistical ecology in the UK, by proposing the formation of a Mathematical Ecology Group as a joint group of the British Ecological Society and the British Region (as it was then) of the Biometric Society. He and Michael Usher were joint secretaries, and the first meeting was at Imperial College London in April 1974, when Peter Greig-Smith, Pierre Dagnelie, Gavin Ross, and M.V. Brian gave talks under the topic of ‘Classification in Ecology’. A discussion session was led by Mark Williamson. This group met regularly for several years, encouraging new researchers into the field, and encouraging collaboration. Richard firmly established the University of St Andrews as a centre for statistical ecology. It is a tribute to his contributions that St Andrews now boasts the largest group of statistical ecologists in the UK National Centre for Statistical Ecology, and one of the largest in the world. Richard’s presence at seminars and other scientific meetings will be greatly missed.