statements which can be checked are found to be in error. Whole sections of the Armenian narratives seem to bear no relation to known facts, and the tradition which they were following is apparently not a direct one. We have seldom cited these sources, both because it would be unsafe to follow their unsupported statements and because there was no point in adding them only where other and better authorities could be quoted. Further detailed study by some scholar well grounded in historical method and equipped with the necessary linguistic training to handle the original manuscripts may well yield important new facts.
Little information of a political character regarding the Parthians is given by Arabic writers; they seem to have possessed neither sources nor reliable tradition going back this far. Some details of the downfall of the empire are preserved by the Arab historians, and among stories in the literature there are mentioned titles and remnants of the political organization of the early Sasanian state which almost certainly descend from its predecessors. On the other hand the Syriac writers had excellent sources. Very probably these included histories of the early Christian bishoprics along with marginal notes on such manuscripts. In addition the Syriac documents possess the advantage of having originated under Parthian rule in the heart of the western empire. The Talmud supplements the fragmentary classical refer-
