Genetic analysis of a Sicilian population using 15 short tandem repeats
Calo et al. (2003)
Link to Full Text
Other authors, studying classical markers (Piazza et al. 1988; Cavalli-Sforza and Edwards 1969) and surnames (Guglielmino et al. 1991), have stressed the existence of genetic differentiation between the eastern and western areas of Sicily and, in general, a remarkable internal variation. In addition, the boundaries found within the Palermo district have been stressed as a sort of economic boundary in a study focusing on surnames (Zei et al. 1993). The variability that has been found could be a consequence of the various dominations that Sicily has undergone: Sycanians, Siculi, and Elymians to begin with (Piazza et al. 1988), followed by Greeks, Romans, Normans, and Arabs (Sandier et al. 1978; Beretta et al. 1986). Among these, Arab domination seems to have had a very strong genetic impact.
Further analysis shows within the same cluster a certain degree of affinity between Egypt and the populations of Sicily. The relationship between Sicilian and North African populations is controversial in population genetics (Piazza et al. 1988; Rickards et al. 1992; Rickards et al. 1998). Our data seem to confirm the hypothesis of Sandler et al. (1978) that underlines the African contribution to the Sicilian gene pool, because of the high frequencies of Hbs, cDe, and Fy (a-b-). In a paper on mtDNA, Semino et al. (1989) found support for this hypothesis, dating back to the introduction of black slaves by Phoenicians and Romans and to the later influxes of Arab immigrants.
A genetic boundary, in fact, clearly divides Sicily from north-central Italy and from northern European populations, besides the other populations from the western Mediterranean basin.
To begin with, most of the research cited in parentheses is from the 60s, 70s, 80s and early 90s, and based on antiquated methods of analyzing blood groups that are now known to be under selective pressure, like HbS, Fy(a-b-) and cDe. A problematic Semino paper is also cited, and two references are even surname analyses pertaining to "economic boundaries," which have nothing to do with genetics at all. Needless to say, none of this is reliable evidence for anything.
Despite the claims of Arab and black admixture being detected in the present data, no Middle Eastern or sub-Saharan African populations are sampled in the study — only two North African populations and other Europeans were tested. The neighbor joining tree (unfortunately not reproduced) is said to group the Sicilian samples on one side with Egyptians among them, and the western/central European samples on the other side with Moroccans among them. The best explanation for this arrangement is the known divide between the eastern and western Mediterranean in terms of Neolithic ancestry, which affects both the European and African coasts and runs vertically through Italy, dividing north from south.
However, the authors use this inconclusive Sicilian-Egyptian connection anyway, and for double duty. First they imply that it serves as evidence of Arab admixture, which is problematic in itself because Egyptians are not particularly good representatives of Arabs. Then in the very next paragraph, they say it confirms the earlier findings of Semino and Sandler alleging African admixture. But Egyptians are North African Caucasoids, while both of those studies deal with markers prevalent in Negroids from sub-Saharan Africa. So in the span of two paragraphs, the authors jump from "Arabs" to "Egyptians" to "North Africans" to "Africans" to "black slaves" and then back to "Arab immigrants," as if these were all synonymous. Very sloppy.
Finally, all of these supposed foreign influences in Sicily are only referred to vaguely using qualifiers and indefinite language like "seems" and "a certain degree." The adjective "strong" is used once, but we know from past experience how a word like that can be an exaggeration that actually amounts to very little admixture. So even if we were to accept the authors' outdated references and ludicrous interpretations, they would have to be taken with a grain of salt for that reason alone.