The recent application of microfocus X-ray computed tomography to the upper back spiral of the Antikythera Mechanism has revealed that there was inscribed on it a Metonic cycle of 19 solar years = 235 lunar months named Phoinikaios, Kraneios, Lanotropios, Machaneus, Dodekateus, Eukleios, Artemisios, Psydreus, Gameilios, Agrianios, Panamos and Apellaios (Freeth, Jones, Steele, and Bitsakis, Nature 454 (2008), 614-617, with extensive Supplementary Notes). After a brief introduction to Greek lunisolar calendars, we present confirmatory evidence that this calendar was that of Corinth, or one of Corinth’s colonies in NW Greece, or a member of the Epirote League, which may have adopted this calendar from one of Corinth’s colonies in Epirus, possibly from Ambrakia in the time of Pyrrhos. In particular, we will demonstrate the invalidity of the reasons recently given by P. Cabanes, Tekmeria 10, 2010, pp. 249-260, for supposing that the calendar of the Mechanism and that of Epirus were different. After eliminating previously misread epigraphic evidence that there was a month called Haliotropios (“solstitial”) in the calendars of Epidamnos and Epirus, we present direct and indirect evidence from Syracuse, Issa, the Peloponnesos and Rhodes that the pan-Dorian month Karneios = Kraneios generally fell around the time of the autumnal equinox, and hence that Phoinikaios fell in the early autumn and Lanotropios in the late autumn, rather than around the winter solstice. We also demonstrate that the calendar of the Mechanism cannot have been the calendar of Syracuse (an ancient Corinthian colony which was the home of Archimedes, who is known to have built orreries), since the Syracusan calendar had in place of Machaneus and Agrianios months called Apollonios and Damatrios, respectively.

This presentation is based on work co-authored by Paul A. Iversen.