It has long been suspected that a marked change in natural conditions must have taken place during historic times throughout considerable areas in Central Asia. The present comparatively arid condition of Mongolia, for example, is in striking contrast to what it must have been in the era preceding the Mongolian invasion of Western Asia in the thirteenth century, and travellers who have followed the route of Alexander's army, on its return from India through Afghanistan and Persia, have noted the difference in the character of the country at the present day. Evidence of a similar change in natural conditions has now been collected in Russian Turkestan, and the process is also illustrated as a result of the explorations conducted by Dr. Stein, on behalf of the Indian Government, on the borders of the Taklamakan Desert and in the oases of Khotan. It is clear that all these districts, at different periods, were far better watered and more densely populated than they are to-day, and that changes in climatic conditions have reacted on the character of the country in such a way as to cause racial migrations. Moreover, there are indications that the general trend to aridity has not been uniform, and that cycles of greater aridity have been followed by periods when the country was capable of supporting a considerable population. These recent observations have an important bearing on the Sumerian problem, and they have therefore been treated in some detail in Appendix I. The physical effects of such climatic changes would naturally be more marked in mid-continental regions than in districts nearer the coast, and the immigration of Semitic nomads into Syria and Northern Babylonia may possibly have been caused by similar periods of aridity in Central Arabia. However this may be, it is certain that the early Semites reached the Euphrates by way of the Syrian coast, and founded their first Babylonian settlements in Akkad. It is still undecided whether they or the Sumerians were in earliest occupation of Babylonia. The racial character of the Sumerian gods can best be explained on the supposition that the earliest cult-centres in the country were Semitic; but the absence of Semitic idiom from the earliest Sumerian inscriptions is equally valid evidence against the theory. The point will probably not be settled until excavations have been undertaken at such North Babylonian sites as El-Ohêmir and Tell Ibrâhîm.