obliterate the tradition of their true order, which even at the period
of the Vth Dynasty may have been completely forgotten. Manetho would
thus have introduced no strange or novel confusion; and this explanation
would of course apply to other sections of his system where the
dynasties he enumerates appear to be too many for their period. But his
reproduction of two lines of predynastic rulers, supported as it now is
by the early evidence of the Palermo text, only serves to increase our
confidence in the general accuracy of his sources, while at the
same time it illustrates very effectively the way in which possible
inaccuracies, deduced from independent data, may have arisen in quite
early times.

(1) Foucart, loc. cit.

In contrast to the dynasties of Manetho, those of Berossus are
so imperfectly preserved that they have never formed the basis of
Babylonian chronology.(1) But here too, in the chronological scheme,
a similar process of reduction has taken place. Certain dynasties,
recovered from native sources and at one time regarded as consecutive,
were proved to have been contemporaneous; and archaeological evidence
suggested that some of the great gaps, so freely assumed in the royal
sequence, had no right to be there. As a result, the succession of known
rulers was thrown into truer perspective, and such gaps as remained were
being partially filled by later discoveries. Among the latter the most
important find was that of an early list of kings, recently published by
Père Scheil(2) and subsequently purchased by the British Museum shortly
before the war. This had helped us to fill in the gap between the famous
Sargon of Akkad and the later dynasties, but it did not carry us
far beyond Sargon's own time. Our archaeological evidence also comes
suddenly to an end. Thus the earliest picture we have hitherto obtained