e assigning of a
regnal year to a definite date B.C. is clear enough (except in
occasional detail) from the conquest by Alexander onwards. Before that
time, in spite of successive efforts to establish a chronology, the
problem is very obscure. The materials for reconstructing the absolute
chronology are of several kinds: (1) Regnal dates as given on
contemporary monuments may indicate the _lengths of individual reigns_,
but not with accuracy, as they seldom reach to the end of a reign and do
not allow for co-regencies. Records of the time that has elapsed between
two regnal dates in the reigns of different kings are very helpful; thus
stelae from the Serapeum recording the ages of the Apis bulls with the
dates of their birth and death have fixed the chronology of the XXVIth
Dynasty. Traditional evidence for the lengths of reigns exists in the
Turin Papyrus of kings and in Manetho's history; unfortunately the
papyrus is very fragmentary and preserves few reign-lengths entire, and
Manetho's evidence seems very untrustworthy, being known only from late
excerpts. (2) The duration of a period may be calculated by
_generations_ or the probable average lengths of reigns, but such
calculations are of little value, and the succession of generations even
when the evidence seems to be full is particularly difficult to
ascertain in Egyptian, owing to adoptions and the repetition of the same
name even in one family of brothers and sisters. (3) _Synchronisms_ in
the histories of other countries furnish reliable dates--Greek, Persian,
Babylonian and Biblical dates for the XXVIth Dynasty, Assyrian for the
XXVth; less precise are the Biblical date of Rehoboam, contemporary with
the invasion of Shishak (Sheshonk) in the XXIInd Dynasty, and the date
of the Babylonian and Assyrian kings contemporary with Amenhotp IV. in
the XVIIIth Dynasty. The last, about 1400 B.C., is the earliest point to
which such coincidences reach. (4) _Astronomical data_, especially the
heliacal risings of Sothis recorded by dates of their celebration in the
vague year. These are easily calculated on the assumption first that the
observations were correctly made, secondly that the calendrical dates
are in the year of 365 days beginning on 1st Thoth, and thirdly that
this year subsequently underwent no readjustment or other alteration
before the reign of Euergetes. The assumption may be a reasonable one,
and if the results agree with probabilities as deduced from the