* For an ideal picture of what may have been the beginnings of that civilization, see Delitzsch, Die Entstehung des altesten Schriflssystems, p. 214, et seq.  I will not enter into the question as to whether it did or did not come by sea to the mouths of the Euphrates and Tigris.  The legend of the fish-god Oannes (Berossus, frag. 1), which seems to conceal some indication on the subject, is merely a mythological tradition, from which it would be wrong to deduce historical conclusions.

** Agade, or Agane, has been identified with one of the two towns of which Sippara is made up, more especially with that which was called Anunit Sippara; the reading Agadi, Agacle, was especially assumed to lead to its identification with the Accad of Genesis x. 10, and with the Akkad of native tradition.  This opinion has been generally abandoned by Assyriologists, and Agane has not yet found a site.  Was it only a name for Babylon?

[Illustration:  040.jpg MAP OF CHALDAEA]

There was nothing serious to fear from the Guti, on the branch of the Tigris to the north-east, or from the Shuti to the north of these; they were merely marauding tribes, and, however troublesome they might be to their neighbours in their devastating incursions, they could not compromise the existence of the country, or bring it into subjection.  It would appear that the Chaldseans had already begun to encroach upon these tribes and to establish colonies among them—­El-Ashshur on the banks of the Tigris, Harran on the furthest point of the Mesopotamian plain, towards the sources of the Balikh.  Beyond these were vague and unknown regions—­Tidanum, Martu, the sea of the setting sun, the vast territories of Milukhkha and Magan.* Egypt, from the time they were acquainted with its existence, was a semi-fabulous country at the ends of the earth.

* The question concerning Milukhkha and Magan has exercised Assyriologists for twenty years.  The prevailing opinion appears to be that which identifies Magan with the Sinaitic Peninsula, and Milukhkha with the country to the north of Magan as far as the Wady Arish and the Mediterranean; others maintain, not the theory of Delitzsch, according to whom Magan and Milukhkha are synonyms for Shumir and Akkad, and consequently two of the great divisions of Babylonia, but an analogous hypothesis, in which they are regarded as districts to the west of the Euphrates, either in Chaldaean regions or on the margin of the desert, or even in the desert itself towards the Sinaitic Peninsula.  What we know of the texts induces me, in common with H. Rawlinson, to place these countries on the shores of the Persian Gulf, between the mouth of the Euphrates and the Bahrein islands; possibly the Makse and the Melangitso of classical historians and geographers were the descendants of the people of Magan (Makan) and Milukhkha (Melugga), who had been driven towards the entrance to the Persian