Guppy[i] was the first to observe the need to look at the relative distribution of surnames and did a preliminary exploration of the subject in the late 19th-century. To make the scope manageable he limited his research to farmers, whom he considered the most geographically stable population group. He identified the farmers from the Kelly’s Post Office Directory for each county and worked out the relative frequency of each surname, recording those that exceeded 7 per 10000. This study did not compare the results to other time periods, however, it did suggest that individual surnames followed specific geographical patterns and that further investigation was warranted.
His work was followed by that of Leeson[ii] who recognised that one of the obvious uses for data from a one-name genealogical study was geographical mapping to confirm the geographical and etymological origins of a name and its subsequent spread. He used birth records from 1841 – 1880 and telephone directories from 1962 to demonstrate that different variants of the name Lee and Leeson had different geographical patterns which were relatively stable over time. The birth data however would have contained children who did not live to become adults and the telephone directories in 1962 did not cover a large portion of the population and included people who were not necessarily born with the name.
Telephone directories remained the major source of data and were used by Redmonds[iii] in the 1960s and 1970s to produce several volumes in a series, each relating to a different county. While this gave an overview of surnames within a county it did not give the overall picture of a surname. Hand drawn distribution maps were produced for Yorkshire, West Riding surnames but not for other counties – presumably proving too difficult without the use of current computer technology.
Lasker[iv] used marriage records from 1975 to study surname distribution, arguing that it is the ‘breeding population’ that is of primary interest when studying population genetics. He found that 93% of surnames occurred at a rate of 1 to 6 per 100,000 and that therefore rare surnames are common. The rarer a surname was the more likely it was to be localised. Mascie-Taylor and Lasker[v] analysed the distribution of 84 common surnames in England and Wales and concluded that most surnames would have a geographic distribution pattern different from a random sample of persons of all surnames. Lasker [vi] also produced the first publication with computer generated surname distribution maps, using the marriage records for the 100 most common surnames of England and Wales. It is of note that many of the patterns are similar to those described by Guppy[vii] nearly a century earlier. These works were all produced from the viewpoint of geneticists rather than genealogists.
Rogers[viii] produced a detailed book exploring the methodology of studying surname distribution from 1086 to 1995. He examined 100 English surnames in three stages: present day; Middle Ages to present day and medieval England. He found that with few exceptions, common surnames appear to be concentrated where they were six hundred years ago. The methodology still forms the basis of current surname studies although the sources used are out of date, especially with the increasing availability of online data such as censuses and civil registration indexes.
All of these studies have focussed on England and Wales and may not translate directly to Scottish surnames which were formed differently.[ix]
In conclusion, to gain the fullest picture of a surname and its origin one must study its earliest formation, etymology and geographical distribution over several time periods in conjunction with genealogical reconstruction.
[i] Guppy, H.B. (1890) Homes of family Names in Great Britain. London: Harrison and Sons. https://archive.org/details/homesoffamilynam00gupprich : accessed November 2015.
[ii] Leeson, F. (1964) The study of single surnames and their distribution. Genealogists Magazine. 14, December. pp. 405-412. Reprinted (1989) Journal of One Name Studies. 3(6) pp.174-181. www.one-name.org/journal/pdfs/vol3-6.pdf ; accessed September 2015.
[iii] Redmonds, George. (1973) Yorkshire, West Riding, English Surnames Series. Volume 1. Chichester: Phillimore.
[iv] Lasker, G. (1983) The frequencies of surnames in England and Wales. Human Biology. 55(2), May. p. 243.
[v] Mascie-Taylor, C.G.N. and Lasker, G.W. (1985) Geographical distribution of common surnames in England and Wales. Annals of Human Biology. 16. pp. 301-308.
[vi] Lasker, G.W. (1985) Surnames and genetic structure. Cambridge: Cambridge University Press.
[vii] Guppy, H.B. (1890) Homes of family Names in Great Britain. London: Harrison and Sons. https://archive.org/details/homesoffamilynam00gupprich : accessed November 2015.
[viii] Rogers, Colin D. (1995) The Surname Detective. Manchester: Manchester University Press.
[ix] Kennett, Debbie. (2012) The Surnames Handbook: A Guide to Family Name Research in the 21st Century. Stroud, Gloucestershire: The History Press. pp. 26-28.
Brimer-Brymer database last updated at 2016-07-02 00:29:43 with 4028 census records and 2856 individuals