Original Research Article Geography of Surnames in the Azores: Specificity and Spatial Distribution Analysis RITA CABRAL, 1 CLAUDIA C. BRANCO, 1 SONIA COSTA, 1 GIANUMBERTO CARAVELLO, 2 MIRO TASSO, 2 BERNARDO R. PEIXOTO, 1 AND LUISA MOTA-VIEIRA 1 * 1 Molecular Genetics and Pathology Unit, Hospital of Divino Espı ´rito Santo, Ponta Delgada, Sa ˜o Miguel Island, Azores, and Instituto Gulbenkian de Cie ˆncia, Oeiras, Portugal 2 Dipartimento di Medicina Ambientale e Sanita ` Pubblica, Sede di Igiene, Universita ` di Padova, Padua, Italy ABSTRACT In order to obtain a better understanding of the genetic structure of the Azorean population, a specificity and spatial distribution analysis was performed, based on 2,454 different surnames present in the Azorean telephone directory (2002). We considered as specific surnames those with an absolute frequency ratio equal to or higher than 50%. The results revealed 51 specific surnames in the whole archipelago. The smallest island presents the only surname with 100% specificity (Pedras). In addition, Sa ˜o Miguel island, which contains 54% of the Azorean population, has the highest number of specific surnames (25 specific surnames). The spatial distribution analysis was used to detect genetic similarity between municipalities through the calculation of spatial autocorrelation (Moran’s I coefficient). Of 240 surnames included in the analysis, 113 showed statistically significant patterns. Five different patterns were obtained, of which the most relevant was isolation by distance and depression (41.6%). However, 43.4% had no defined pattern. The overall correlogram shows a majority of positive values for distances lower than 49 km and between 269–309 km, indicating high similarity between closer municipalities and between distant municipalities whose populations show historic and sociocultural affinities. In conclusion, our data are in agreement with the historical background of the Azorean popula- tion. Am. J. Hum. Biol. 17:634–645, 2005. # 2005 Wiley-Liss, Inc. Ever since Crow and Mange (1965) put forward their isonymy-based model, infor- mation about surnames has been increas- ingly used in research on human population genetics (Colantonio et al., 2003; Legay and Vernay, 2000). For the majority of European societies, surnames can be considered a highly polymorphic locus linked to the Y-chromosome, resembling neutral alleles transmitted by males (Sokal et al., 1992). In addition, surnames reflect the traditions and culture in which they were formed, i.e., populations that share the same surnames usually share same cultural background (Lucchetti and Soliani, 1989). Surnames represent a precious instru- ment for the assessment of population struc- ture, because large databases are easily acquired and are believed to be selectively neutral. Indeed, they were used to estimate inbreeding changes in a population, to mea- sure the degree of population subdivision, and to analyze changes in genetic relation- ships between populations (Pettener et al., 1998). Due to their high polymorphism, sur- names permit an effective means of charac- terization, even when populations belong to a geographically limited area (Caravello and Tasso, 1999). Moreover, they represent a pre- cious instrument for evaluating the affinity among different human groups which also live in a restricted geographical area, whereas biological markers, in that case, usually afford poor differentiation. The use of surname fre- quency and distribution to assess the popula- ß 2005 Wiley-Liss, Inc. *Correspondence to: Luisa Mota-Vieira, Molecular Genetics and Pathology Unit, Hospital of Divino Espı ´rito Santo, 9500-370 Ponta Delgada, Sa ˜o Miguel Island, Azores, Portugal. E-mail: lmotavieira@hdes.pt Received 12 November 2004; Revision received 2 June 2005; Accepted 20 June 2005 Published online in Wiley InterScience (www.interscience. wiley.com). DOI: 10.1002/ajhb.20424 Grant sponsor: Direcc ¸a ˜o Regional da Cieˆncia e Tecnologia, Azores; Grant sponsor: Fundac ¸a ˜o para a Cieˆncia e a Tecnologia, Operational Programs, Science, Technology, and Innovation, Portugal; Grant number: SFRH/BD/12254/ 2003. AMERICAN JOURNAL OF HUMAN BIOLOGY 17:634–645 (2005)