In various parts of Melanesia, from New Guinea to Fiji, are found relics of a prehistoric culture or cultures, which cannot as yet be dated or correlated satisfactorily. An important region covers the east end of New Guinea, New Britain and the adjacent archipelago, where the remains found include ornamented pottery (somewhat similar to that of prehistoric Japan), engraved skulls, obsidian implements and stone pestles and mortars, some of which occur at a considerable depth in river gravels and all of which are quite distinct from the products of recent inhabitants. In the same region, but within a more limited area, are low stone circles, lines and heaps of stones, now used as squatting places and, in part at least, of recent origin. Rock paintings and engravings are numerous in the district behind Port Moresby, some of them coated with a film of calcium carbonate, indicating a certain an tiquity. Megaliths are widely distributed and irrigation terraces are found throughout Melanesia. In the Solomon islands (Bou gainville and San Cristoval) upright stones occur singly and in groups ; in San Cristoval some villages are bordered with stone walls or platforms, and numerous stone-faced and earth mounds, containing shafts leading to burial chambers, and sometimes sur mounted by stone statues, small dolmens or upright stones, are still in use. They may be related to Polynesian burial platforms. Their form and the mortuary customs associated with them have led to a comparison with early Egyptian mastabas ; attention has also been called to a supposed resemblance of certain pig-tail figures to Egyptian royal statues, but this seems too remote to justify the emphasis laid upon it by some authorities. In the New Hebrides the antiquities include menhirs, stone sacrificial "tables" or dolmens, walls and high platforms, and obsolete types of pot tery ornamented in a great variety of styles. There is an excep tional development of stone walls and buildings in the Banks islands, some, however, of recent construction. In general, chrono logical evidence is almost entirely lacking for these monuments, and the difficulties of distinguishing between ancient and modern are increased by the fact that parts of Melanesia are still living in a "megalithic period." In New Caledonia stone statues (in caves in Lifu) and a few "dolmens or trilithons" have been reported. Petroglyphs are common to all these groups, and particularly nu merous in New Caledonia, where they take the form of human and animal figures, spirals and other patterns. Some of the better
structures may be attributed either to late waves of Indonesian immigrants, or to reflex Polynesian influence, which has affected many of these islands in comparatively recent times; the nanga of Fiji so closely resemble Polynesian marae as to leave no doubt of their common origin.
Migrations and Hypotheses.—There is general agreement, on the evidence of oral traditions supported by physical, linguistic and botanical evidence, that the ancestors of the Polynesians—a composite stock including Caucasian, Mongolian and Negroid elements—must have come originally in a series of migratory waves from the Asiatic continent. Their starting-point cannot be definitely fixed; hypotheses vary as between India (Ganges or Chota Nagpur), Assam, and the Cambodia-Siam region of Indo China. In favour of the latter view is the fact that a people speak ing an allied language (Mon-Khmer) and having physical affin ities with the Polynesians is still living there. They had probably settled in the Indonesian archipelago by the latter part of the first millennium B.C., whence they proceeded in at least two main streams of migration by way of Melanesia or Micronesia to Fiji, which they reached in about the 5th century A.D. Thence eastern Polynesia and New Zealand were settled in successive voyages during the 7th to 14th centuries. Hawaii seems to have been reached first by a direct migration from Indonesia and subse quently by a branch of the main wave from south-eastern Poly nesia ; this conclusion is supported by archaeological study which recognizes two distinct types of heiau (marae), the later of which corresponds more closely with the Tahiti-Marquesas type.
The hypothesis that Oceania served as a highway for the dif fusion of an "archaic civilization," originating in Egypt in the 3rd millennium B.C., and including the practices of sun worship, mum mification and the building of megalithic structures ; and that the stone monuments of Oceania are attributable to the bearers of this culture-complex, has been a subject of considerable controversy. If tradition is right in asserting that Polynesian migrations did not reach the eastern Pacific before the 6th or 7th centuries A.D., and if, as appears probable, these people were the builders of the monuments in that area, it is evident that the older civilizations of America cannot have been inspired by them. On the other hand, since they were in contact with Asia till a relatively late period, they can hardly have escaped the influence of old-world culture or have failed to carry some of its elements with them into the Pacific. The view that Oceania was peopled by migration from America finds little support at the present time. But the theory has recently been advanced that the monuments and culture of eastern Oceania may owe something to the influence of Central and South America, the coasts of which were quite probably visited by the intrepid Polynesian navigators. The absence of pot tery has been used by others as an argument for the very early isolation of the Polynesians; the loss of this art was, however, an inevitable result of prolonged wanderings among coral islands, where the material with which to practise it was not available.