Some errors in the calculation of coefficient of inbreeding (F) and mean coefficient of inbreeding (α) values were detected in this article and have been re-calculated.

Page 3, Methods, paragraph 3, should read:

A coefficient of inbreeding (F) was calculated for each couple and mean coefficients of inbreeding (α) for the population then analysed for each period (Freire-Maia, 1974). Consanguineous marriages were classified according to the degree of relationship between spouses, i.e. double first cousins (F=1/8), first cousins (F=1/16), first cousin once removed (F=1/32), second cousins (F=1/64), second cousin once removed (F=1/128) and third cousins (F=1/256) (Lancaster, 2007). For all individuals and couples whose degree of inbreeding was not specified the value of F was taken to be 1/16, because since 1983 the dispensation requirement for marriage has been a 4th grade civil relationship, i.e. first cousins. All of the recorded degrees of consanguinity were converted from canonical to civil nomenclature.

Page 4, Results, paragraph 3, should read:

The values of F ranged from 0.000703 to 0.085508 per year, and the mean coefficient of inbreeding (α) was 0.014530.

Page 5, Results, paragraph 2, should read:

The values of F ranged from 0.000125 to 0.087281. The highest F value was in 1951 when 99.4% of marriages were endogamic, and 24.2% of these marriages were consanguineous. The mean coefficient of inbreeding (α) for the 2nd study period was 0.005670.

Page 6, Results, paragraph 2, should read:

In this period, the frequency of consanguineous marriage was 4.2%. The average F value was 0.001674 and ranged from 0.000250 to 0.098562 per year.

Page 10, new reference:

Lancaster, F. M. (2007) Genetic and Quantitative Aspects of Genealogy. URL: http://www.genetic-genealogy.co.uk/index.html