Their territory includes parts of the New Hebrides, the Loyalty Isles, New Caledonia, Viti, and a variety of smaller groups. These islanders are usually of mixed type, and are known as “Melanesians.” The natives of the Feejee Islands are an excellent specimen of these, and their archipelago forms the dividing line between the Papuan and Polynesian groups.[155]

3. The Melanesian Group.

The Melanesians, of all the islanders, present in individual cases the strongest likeness to the equatorial African Negro; yet among these there is that prevailing variability of type so frequent in insular peoples. Their color passes from the black of the typical Negro to the yellow of the Malayan; their hair, generally frizzly, may be quite straight and of any hue from black to blonde. These variations are in individuals or families, and are not owing to mixed blood.[156]

Unlike the Polynesians, the Melanesians are agricultural in habits, and sedentary. They build artistically decorated houses, are acquainted with the bow and arrow, occasionally make pottery, and construct shapely canoes, though not given to long voyages. The women are modest and chaste, and their religion is principally a form of ancestral worship.

The languages of these islanders betray their compound origin. In form and in the pronominal elements they stand related to the Malayan and Polynesian idioms, and in structure approach sometimes the richness of the former. In the Viti, for example, both prefixes and suffixes are employed, and the possessive is added to the noun. The root words are monosyllables or dissyllables, and drawn from the Papuan idioms, and the phonetics are much richer than the Polynesian.

These facts go to show that the Melanesians are physically and linguistically a mixed people, a compound of the woolly-haired black Papuas, whom we may suppose to have been the aborigines of Melanesia, with the smooth-haired, light-colored Malays, who reached the archipelago as adventurers and immigrants. As their tongues form, as it were, the second stratum of structure when compared with the Polynesian dialects, we can go a step further and say that the ethnic formation of the Melanesian islanders occurred subsequently to the construction of the Polynesian physical type and languages.[157]

The ethnic relationship of the various adjoining islanders to the Papuas has been studied by many observers, but its solution has not yet been reached. The Papuas themselves impressed Hale as partly Malayan—“a hybrid race,”[158] and Virchow calls attention to the fact that a broad zone of wavy-haired peoples intervene between the Papuas and the pure Malays, shading off into the Australians on the one hand and the Veddahs of Ceylon on the other.[159] This is very significant of the ethnic origin of the inhabitants of Australasia.

It is borne out by an examination of the Papuan languages. These are quite dissimilar among themselves, and appear to have been derived from a number of independent linguistic stocks. While these were originally distinct from the Malayan, it is a recognized fact that all the Papuan, and still more all the Melanesian dialects, have absorbed extensively from Malayan and Polynesian sources, and we are certain, therefore, that a similar absorption of Malayan blood has taken place.[160]

II. The Malayic Stock

Is by far the most important group of peoples with whom we have to do in the area we are now studying. Many ethnologists, indeed, set it up as a distinct race, the “Malayan” or “Brown” race, and claim for it an importance not less than any of the darker varieties of the species. It bears, however, the marks of an origin too recent, and presents Asian analogies too clearly, for it to be regarded otherwise than as a branch of the Asian race, descended like it from some ancestral tribe in that great continent. Its dispersion has been extraordinary. Its members are found almost continuously on the land areas from Madagascar to Easter Island, a distance nearly two-thirds of the circumference of the globe; everywhere they speak dialects with such affinities that we must assume for all one parent stem, and their separation must have taken place not so very long ago to have permitted such a monoglottic trait as this.