It may be admitted that this explanation is one that at first sight seems to commend itself, for it appears to succeed in reconciling the later tradition with the early monuments. But difficulties in the way of its acceptance were at once pointed out.[23] The occurrence of the proper name Sharru-Gi-ili, "Sharru-Gi is my god," on the Obelisk of Manishtusu clearly proves that a king bearing the name of Sharru-Gi, and presumably identical with the Sharru-Gi of the new stele, preceded Manishtusu, king of Kish, for the deification of a king could obviously only take place during his lifetime or after his death.[24] Similar evidence has already been cited to prove that Urumush of Kish was anterior to Shar-Gani-sharri and Narâm-Sin, though his reign may not have been separated from theirs by any long interval.[25] Granting these conclusions, if Narâm-Sin had been the son of Sharru-Gi, as suggested by Père Scheil, Urumush would have been separated from Manishtusu by the Dynasty of Akkad, a combination that is scarcely probable. Moreover, the context of the passage on the tablet from Tello, on which the names of Sharru-Gi and Narâm-Sin are mentioned, though of doubtful interpretation, does not necessarily imply that they were living at the same time; they may have been separated by several generations. These reasons in themselves make it probable that Sharru-Gi was not the founder of Narâm-Sin's dynasty, but was a predecessor of Manishtusu and Urumush upon the throne of Kish. It has been further pointed out that in an inscription preserved in the Imperial Ottoman Museum at Constantinople the name of a king of Kish is mentioned, which, to judge from the traces still visible, may probably be restored as that of Sharru-GI.[26] The fragmentary nature of the text, which was found at Abû Habba during the excavations conducted by the Turkish Government upon that site,[27] rendered any deductions that might be drawn from it uncertain; but it sufficed to corroborate the suggestion that Sharru-Gi was not a king of Akkad, but a still earlier king of Kish. Since then I have recognized a duplicate text of the Constantinople inscription, also from Abû Habba, which enables us to supplement and to some extent correct the conclusions based upon it. The duplicate consists of a cruciform stone object, inscribed on its twelve sides with a votive text recording a series of gifts to the Sun-god Shamash and his consort Aa in the city of Sippar, and the early part of its text corresponds to the fragmentary inscription at Constantinople. Unfortunately the beginning of the text is wanting, as is the case with the Constantinople text, so that we cannot decide with certainty the name of the king who had the monument engraved. But the duplicate furnishes fresh data on which to base a conclusion.