Everybody knows it was Sir Ronald Aylmer Fisher (1890 –1962) who explained the basis of the 1:1 sex ratio at birth. The account came from his book on natural selection written in 1930. If males are produced in excess then females become more desirable; if females are produced in excess then it pays for parents to produce males, all things being equal. Over time the sex ratio must settle at 1:1. What I will not deal with here are circumstances in which all other things are not equal and the sex ratio may be modified in some species between conception and birth.
Fisher’s explanation of the sex ratio, the ‘rarer-sex effect’ is a famous principle of evolutionary biology. Except it was not Fisher’s. Andy Gardner of the University of St Andrews has recently published reminders that the first person to get the explanation right was John Austin Cobb in 1914. Gardner has also demonstrated that Fisher knew of Cobb’s earlier work and quoted it in another publication but did not name Cobb in his 1930 book. In those days scientific books were far more personal commentaries on the state of play in a particular field rather than the annotated bibliographies that many have become. Fisher wrote his book in the manner of the time, as an advanced textbook, stating his views across the whole field. Cobb was not omitted by design or oversight for the simple reason that there are very few references to the work of others—or to Fisher himself—anywhere in the book.
That Cobb and not Fisher was responsible for the ‘rarer-sex effect’ was uncovered in the late 1990s by A.W.F. Edwards in Cambridge. However, major reviews and books published since then have continued to overlook Cobb, continuing to credit Fisher for the insight.
In his paper Cobb clearly lays out the argument:
If we take the sex-ratio at birth it appears at first sight that the numbers of the sexes born will become equal. For if there are more born of one sex, say, the male, a female will have a greater chance of finding a mate than a male. There will be more matings, therefore, among the descendants of mothers of females than amongst the descendant of mothers of males. The mothers of females will therefore be better тергеsented in the third generation, and as their characteristic is assumed to be inherited, there will be a tendency for the sex-ratio to diminish until it reaches equality in numbers between the sexes at birth.
But who was John Austin Cobb? Cobb was a man in that enviable category of ‘gentleman scientist’. He had sufficient money not to have to work for a living. Edwards and Gardner found something of Cobb and his life, and I have managed to add a little more but we only have an outline of his life and the papers he wrote but little else to explain how he became interested in statistics and the mathematical treatment of evolution and other matters.
John Austin Cobb was born in the village of Sheldwich, Kent on 27 November 1866, the son of a farmer. He was educated at Haileybury (1879-1884) taking the intermediate examination for admission as a solicitor in 1884. In January 1885 he matriculated in the University of London but there appears to be no indication that he completed his studies, nor of for how long he was a member of the university or the subjects he took. In 1889 he qualified as a solicitor. In 1891 he married Helen Isabel Marrs in Minneapolis. Why he had gone to the USA and what he did while there are not known but the family appeared to move between England and the USA for a time in the 1890s. For example, the first child was born in Surrey in 1892, the second in Minneapolis in 1896 and the third and fourth in Surrey in 1898 and 1899. From Minneapolis in 1896 he published a paper in Nature.
|
Only photographs of John Austin Cobb as a boy appear to be in circulation. The one shown here taken in 1880 at the age of 14 is also in Gardner's paper |
The family lived at 108 Church Road, Richmond, including, in the 1901 Census, Cobb’s American mother-in-law. The house (which can be seen on Google Earth) is a substantial one, now divided into flats and he is shown as employing four domestic servants. He is shown in the 1901 and 1911 censuses as living on ‘private means’ which indicates he was not working for a living in any capacity.
I do not know whether Cobb ever worked. In the marriage certificate of one daughter he is described as ‘manufacturer (deceased)’ but since that was the occupation of the groom’s father, it is possible there was some confusion. In a translation from the German of his daughter’s death certificate he is described as a ‘barrister’, a misinterpretation of the word for lawyer in German.
John Austin Cobb died on 23 October 1920, aged 53, at ‘Hotel Messena, 11 Rue Bachaumont, France’ (an address in Paris). His address in London was given as Portland Hotel, Great Portland Street, formerly of Richmond.
Gardner lists the following publications by Cobb:
Cobb JA. 1896. Measurement of crabs. Nature 55, 155. doi:10.1038/055155b0
Cobb JA. 1905. Halation. Nature 73, 54. doi:10.1038/073054c0 [an exploration of the phenomenon caused by light reflecting back through a film layer onto photographic emulsion, suggesting Cobb had an interest in photography.]
Cobb JA. 1908. The effect of errors of observation upon the correlation coefficient. Biometrika 6, 109 doi:10.2307/2331561
Cobb JA. 1913. Human fertility. Eugenics Review 4, 379–382.
Cobb JA.. 1914. The problem of the sex-ratio. Eugenics Review 6, 157–163.
Cobb JA. 1914. Sex ratio. Review of Reviews 50, 128.
Cobb JA. 1914. The alleged inferiority of the first-born. Eugenics Review 5, 357-359.
Cobb became well-known in the eugenics world for his paper on differences in fertility (actually fecundity) between classes and its implications for future generations. It all made perfect sense but only if fecundity and intelligence were determined entirely genetically—the downfall of eugenics in most of its many manifestations. Poor Cobb would now have been ‘cancelled’ by the misguided zealots who bathe in wilful ignorance.
I have been unable to find any publications by Cobb after 1914. He suffered two family disasters around this time. His daughter, Sybil Josephine, died while at school in Dresden, Germany, in 1913, aged 14. His son, John Eldridge, an observer in 21 Squadron, Royal Flying Corps, was killed near Ypres in 1917 when his aircraft suffered engine failure on take off.
Cobb had two other children. Mildred Isabel (1892-1981) married an American lawyer in England and then lived in the U.S.A. They had three daughters. Roland Marrs Cobb (1898-1948) was a Royal Navy officer who served at the Battle of Jutland; he retired as a Commander. He married in 1930 and had a son, Dudley Marrs Cobb in 1931.
There is a chance that more information on Cobb will emerge and shed light on how he acquired his interests and statistical expertise. In the latter respect, he impressed Karl Pearson, by his note showing how errors in x and y affect the correlation coefficient—a point I heard discussed at a symposium just before covid.
In conclusion, Andy Gardner writes highly of Cobb’s contributions to the ‘rarer-sex effect’. It is certainly time to recognise the seminal contributions of this unaffiliated gifted amateur of the early 20th century.
Edwards AWF. 1997. The Galton Lecture: The Eugenics Society and the development of biometry. In Essays in the history of eugenics (ed. RA Peel), pp. 156–172. London: Galton Institute.
Edwards AWF. 1998. Natural selection and the sex ratio: Fisher’s sources. American Naturalist 151, 564–569 doi:10.1086/286141
Fisher RA. 1930. The Genetical Theory of Natural Selection Oxford: Clarendon Press.
Gardner A. 2023. The rarer-sex effect. Philosophical Transactions of the Royal Society B 378, 20210500. https://doi.org/10.1098/rstb.2021.0500
Gardner A. 2023. R.A. Fisher on J.A. Cobb’s The Problem of the Sex Ratio. Notes and Records of the Royal Society doi:10.1098/rsnr.2023.0067