among other things, husks of barley and millet, and fragments of mammalian bones, including those, no doubt, of the domesticated sheep and goats and cattle painted on the pottery.13 It is therefore apparent that at an extremely remote period a knowledge of agriculture extended throughout Egypt, and we have no reason for supposing that it was not shared by the contemporary inhabitants of Sumer.
The various theories which have been propounded regarding the outside source of Sumerian culture are based on the assumption that it commenced abruptly and full grown. Its rude beginnings cannot be traced on the banks of the Tigris and Euphrates, but although no specimens of the earliest form of picture writing have been recovered from the ruins of Sumerian and Akkadian cities, neither have any been found elsewhere. The possibility remains, therefore, that early Babylonian culture was indigenous. "A great deal of ingenuity has been displayed by many scholars", says Professor Elliot Smith, "with the object of bringing these Sumerians from somewhere else as immigrants into Sumer; but no reasons have been advanced to show that they had not been settled at the head of the Persian Gulf for long generations before they first appeared on the stage of history. The argument that no early remains have been found is futile, not only because such a country as Sumer is no more favourable to the preservation of such evidence than is the Delta of the Nile, but also upon the more general grounds that negative statements of this sort cannot be assigned a positive evidence for an immigration."14 This distinguished ethnologist is frankly of opinion that the Sumerians were the congeners of the pre-Dynastic Egyptians of the Mediterranean or Brown race, the eastern branch of which reaches to India and the western to the British Isles and Ireland. In the same ancient family are included the Arabs, whose physical characteristics distinguish them from the Semites of Jewish type.
Some light may be thrown on the Sumerian problem by giving consideration to the present-day racial complexion of Western Asia. The importance of evidence of this character has been emphasized elsewhere. In Egypt, for instance, Dr. C.S. Myers has ascertained that the modern peasants have skull forms which are identical with those of their pre-Dynastic ancestors. Mr. Hawes has also demonstrated that the ancient inhabitants of Crete are still represented on that famous island. But even more remarkable is the fact that the distinctive racial type which occupied the Palaeolithic caves of the Dordogne valley in France continues to survive in their vicinity after an interval of over twenty thousand years.15 It is noteworthy, therefore, to find that in south-western Asia at the present day one particular racial type predominates over all others. Professor Ripley, who summarizes a considerable mass of data in this connection, refers to it as the "Iranian", and says: "It includes the Persians and Kurds, possibly the Ossetes in the Caucasus, and farther to the east a large number of Asiatic tribes, from the Afghans to the Hindus. These peoples are all primarily long-headed and dark brunets. They incline to slenderness of habit, although varying in stature according to circumstances. In them we recognize at once undoubted congeners of our Mediterranean race in Europe. The area of their extension runs off into Africa, through the Egyptians, who are clearly of the same race. Not only the modern peoples, but the Ancient Egyptians and the Phoenicians also have been traced to the same source. By far the largest portion of this part of Western Asia is inhabited by this eastern branch of the Mediterranean race." The broad-headed type "occurs sporadically among a few ethnic remnants in Syria and Mesopotamia".16 The exhaustive study of thousands of ancient crania in London and Cambridge collections has shown that Mediterranean peoples, having alien traits, the result of early admixture, were distributed between Egypt and the Punjab.17 Where blending took place, the early type, apparently, continued to predominate; and it appears to be reasserting itself in our own time in Western Asia, as elsewhere. It seems doubtful, therefore, that the ancient Sumerians differed racially from the pre-Dynastic inhabitants of Egypt and the Pelasgians and Iberians of Europe. Indeed, the statuettes from Tello, the site of the Sumerian city of Lagash, display distinctively Mediterranean skull forms and faces. Some of the plump figures of the later period suggest, however, "the particular alien strain" which in Egypt and elsewhere "is always associated with a tendency to the development of fat", in contrast to "the lean and sinewy appearance of most representatives of the Brown race".18 This change may be accounted for by the presence of the Semites in northern Babylonia.
Whence, then, came these invading Semitic Akkadians of Jewish type? It is generally agreed that they were closely associated with one of the early outpourings of nomadic peoples from Arabia, a country which is favourable for the production of a larger population than it is able to maintain permanently, especially when its natural resources are restricted by a succession of abnormally dry years. In tracing the Akkadians from Arabia, however, we are confronted at the outset with the difficulty that its prehistoric, and many of its present-day, inhabitants are not of the characteristic Semitic type. On the Ancient Egyptian pottery and monuments the Arabs are depicted as men who closely resembled the representatives of the Mediterranean race in the Nile valley and elsewhere. They shaved neither scalps nor faces as did the historic Sumerians and Egyptians, but grew the slight moustache and chin-tuft beard like the Libyans on the north and the majority of the men whose bodies have been preserved in pre-Dynastic graves in the Nile valley. "If", writes Professor Elliot Smith, "the generally accepted view is true, that Arabia was the original home of the Semites, the Arab must have undergone a profound change in his physical characters after he left his homeland and before he reached Babylonia." This authority is of opinion that the Arabians first migrated into Palestine and northern Syria, where they mingled with the southward-migrating Armenoid peoples from Asia Minor. "This blend of Arabs, kinsmen of the proto-Egyptians and Armenoids, would then form the big-nosed, long-bearded Semites, so familiar not only on the ancient Babylonian and Egyptian monuments, but also in the modern Jews."19 Such a view is in accord with Dr. Hugo Winckler's contention that the flow of Arabian migrations was northwards towards Syria ere it swept through Mesopotamia. It can scarcely be supposed that these invasions of settled districts did not result in the fusion and crossment of racial types and the production of a sub-variety with medium skull form and marked facial characteristics.
Of special interest in this connection is the evidence afforded by Palestine and Egypt. The former country has ever been subject to periodic ethnic disturbances and changes. Its racial history has a remote beginning in the Pleistocene Age. Palaeolithic flints of Chellean and other primitive types have been found in large numbers, and a valuable collection of these is being preserved in a French museum at Jerusalem. In a northern cave fragments of rude pottery, belonging to an early period in the Late Stone Age, have been discovered in association with the bones of the woolly rhinoceros. To a later period belong the series of Gezer cave dwellings, which, according to Professor Macalister, the well-known Palestinian authority, "were occupied by a non-Semitic people of low stature, with thick skulls and showing evidence of the great muscular strength that is essential to savage life".20 These people are generally supposed to be representatives of the Mediterranean race, which Sergi has found to have been widely distributed throughout Syria and a part of Asia Minor.21 An interesting problem, however, is raised by the fact that, in one of the caves, there are evidences that the dead were cremated. This was not a Mediterranean custom, nor does it appear to have prevailed outside the Gezer area. If, however, it does not indicate that the kinsmen of the Ancient Egyptians came into contact with the remnants of an earlier people, it may be that the dead of a later people were burned there. The possibility that unidentified types may have contributed to the Semitic blend, however, remains. The Mediterraneans mingled in Northern Syria and Asia Minor with the broad-headed Armenoid peoples who are represented in Europe by the Alpine race. With them they ultimately formed the great Hittite confederacy. These Armenoids were moving southwards at the very dawn of Egyptian history, and nothing is known of their conquests and settlements. Their pioneers, who were probably traders, appear