ethnography and archaeology of these islands will obtain definite evidence towards the solution of this problem.
If the future should show the association elsewhere of the three features of burial-place, pyramid and image of stone, the case for a common origin will become still stronger should it reveal a motive for the meaning of the image similar to that of San Cristoval. In this island the stone image is regarded as the representative of the dead chief who is buried in the pyramidal mound it surmounts, and it is believed that the soul or ghost of the dead man has its abode in the image. This suggests a motive for the presence of the statues on the ahu of Easter Island. In other parts of Melanesia, as in Ambrim, we know that the ghost of a dead man is believed to go into an image in human form. In Ambrim these statues are not situated on burial-places, but are connected with an institution called the Mangge, an organisation which has a cult of the dead as its motive. Ambrim and San Cristoval suggest that the great statues of Easter Island were made in order that they should act as abiding-places of the ghosts of the dead over whose bodies they watch.
I can now return to the problem raised by the presence of two chief kinds of ahu. There are two main possibilities. One is that the two modes of burial belong to different cultures. The other, and the more probable, is that the pyramidal form of burial-place was that proper to the culture of the people who introduced the practice of stone-working, but that when the statues increased in size the original form of pyramid was unable to bear their weight, and a process was set in action which produced the platforms or terraces upon which the images stood. It is quite in keeping with all we know of Oceanic culture, however, that the people should have continued to use the pyramidal form when no image was made, and that the sentiment in favour of monuments of this kind was of considerable strength is suggested by the use of the
