centuryCEinscription, recently discovered in Metropolis (Lydia) and palaeo-
graphically dated to the reign of Tiberius, comprises a table of Julian calendar
days and their equivalent in the calendar of the province of Asia (Engelmann
1999: 142–3), which confirms at least the existence of two-columnhemerolo-
giain the early Roman period. The marble stone it is inscribed on suggests that
it was set up in a public place (ibid.), not only for public use but also perhaps as
a monumental, symbolic statement of the harmonious relationship between
the institutions of the city (or province) and of the Roman Empire. The
hemerologia manuscripts, however, which contain many more columns,
were probably intended more specifically for scribes and imperial adminis-
trators, to help them with the conversion of multiple calendar dates into one
another—a task which in the Roman East was often likely to be needed. These
handbooks are remarkable not only in the quantity of information they
provide (especially as different calendars appear in the different manuscripts),
but also in the mere fact that they were produced. They are, in themselves, a
powerful testimony of calendar diversity in the Roman East, but at the same
time of the integration of these calendars into afixed and stable common
denominator, the 365-day year of the Julian calendar.
The calendars that are listed in thehemerologiavary considerably, in spite
of sharing the basic structure of the Julian calendar, i.e. a 365-day year (with,
implicitly, an additional day every four years).^78 Some are structurally identi-
cal with the Julian calendar, in particular the calendars of Antioch (and hence
of the province of Syria: see above, Table 5.3), Sidon, and (in Asia Minor)
Lycia, which differ from the Julian calendar only in their use of Macedonian
month-names and (for Antioch) an autumn New Year. Other calendars use a
sequence of months with similar lengths as in the Julian calendar, but begin-
ning the year on 23 September, i.e. Augustus’birthday; the most important of
these is the calendar of the province of Asia. Other calendars have a sequence
of 30- and 31-day months, but no 28-day month (like the Julian month of
February): these are the calendars of Tyre and Heliopolis (Baalbek). Others,
finally, are conterminous with the Alexandrian calendar (i.e. the Egyptian civil
calendar after its adaptation to the Julian calendar, on which see below),
may have been conventional among astronomers in this period. Earlier scholars have associated
the use of multiple calendars in ancient ephemerides with thehemerologia(Gerstinger and
Neugebauer loc. cit., Samuel 1972: 171–3); however, the calendars used in ephemerides are not
quite the same as inhemerologia(in particular the astronomical lunar calendar, and possibly also
the serial Julian calendar). Ephemerides andhemerologiaare different literary genres serving
very different purposes, and I would not necessarily assume any dependency between the two.
(^78) See, with tabulations, Samuel (1972) 174–8 and Bickerman (1968) 50; but the most
accurate and comprehensive tabulations remain Kubitschek (1915) 42–53. The additional day
(in leap years) and its position in the year are not indicated in the tables, but are implicitly there
because otherwise these calendars would not have remained in a stable relationship with the
Julian calendar.
Fragmentation: Babylonian and Julian Calendars 261