became the model of the Syrian calendar (see above and Table 5.3). Similarly,
the calendar of Seleucia in Pieria (near Antioch) was identical with the Julian
calendar but with mostly Greek, non-Macedonian month-names. Closer still
to the Julian calendar were the calendars of Sidon and of Lycia, in which the
month of Dios—formerly thefirst month of the Seleucid Macedonian year—
corresponded to January,first month of the Roman year.^141
Some Near Eastern cities, such as Tyre and Heliopolis (Baalbek), adopted
more original calendars that only had in common with the Julian calendar a
fixed, 365-day year. The calendar of Tyre consisted of a run offive 31-day
months, starting with Dystros on 18 March, followed by a run of seven 30-day
months (Samuel 1972: 176).We do not know, however, whether the year was
considered to begin in Dystros, in Dios (on 18 November), or in any other
month. The dates of the beginning of months and their relationship to the
Julian calendar seem arbitrary and without rationale. The most likely expla-
nation is that at the time when the calendar was instituted, the old lunar
month of (say) Dios happened to begin on 18 November, which was adopted
as the starting point of the new calendar.
The calendar of Caesarea Maritima was probably also adapted to the Julian
year, but it is not represented in thehemerologiaand the sporadic evidence we
have—a handful of dates in Eusebius’history of the martyrs of Palestine—is
insufficient to provide a reliable reconstruction. It is clear, however, that it was
structured differently from the other calendars of the Near East.^142
The calendar of Heliopolis (Baalbek)The calendar of Heliopolis was in some ways unique, and has attracted much
attention—perhaps undeservedly—in modern scholarship. Its most striking
feature, at least as it appears in thehemerologia, is its unusual use of Babylo-
nian or Syriac month-names.^143 These names are inconsistently transliterated
into Greek in the varioushemerologiamanuscripts, sometimes even within the
(^141) Ibid. 174. For further analysis of Sidon and Lycia, see above, nn. 57–8.
(^142) A study of this calendar was made by Rey-Coquais (1978), who collated the evidence in
Eusebius’text (four Macedonian dates with days of the week, in short recension 1. 2, 4. 15, 6. 1, 7.
1 —but leaving out the Julian dates, which he convincingly identified as later interpolations—and
then reconstructed the calendar on the assumption that it was structurally similar to, albeit not
identical with, the calendar of Tyre (this assumption is of course unjustified). The dates in
Eusebius are not compatible with a lunar calendar; it is clear that the calendar consisted of
30- and 31-day months, but their distribution in the year remains unknown. See also the remarks
of Burgess (1999) 37–8n.7.
(^143) See Table 5.7. The nameOziris similar to the SyriacH:aziran(see above, Table 5.3); other
names are consistent with the Syrian calendar, except forAgandGelon(instead of Tishrei 1 and
Kanun 1 in the Syrian calendar respectively) which are local names.
Fragmentation: Babylonian and Julian Calendars 285