The maps of currents (Figure 3) and winds of the Pacific Ocean show the trade winds that blow from east to west over the area of warm water encompassing most of the Pacific Islands, and steady currents of 8 to 35 km per day that flow in the same direction. Driftwood, rafts and square-rigged ships have been moved on these currents, witness the Kon-Tiki balsa raft (Heyerdahl 1978:185). In 1913, the Dagonar, a derelict sailing ship, took only 170 days to go 8000 km from Peru to the Tuamotus in this current (Hornell 1945). Heyerdahl’s basic premise was that early boats followed winds and currents, but he thought only of rafts and reed boats. Rafts and reed boats were known world-wide in Neolithic times so people could have drifted eventually to many unlikely places, and perhaps from Peru to Polynesia. Easter Island was known to the Incas and archaeological remains there strongly suggest that some aspects of South American culture and plants spread there, and perhaps into other parts of eastern Polynesia as well (Heyerdahl 1978). Probably the Asiatic chicken was taken to Ecuador via Polynesia, and the South American sweet potato, some cotton, gourds, and other plants travelled westwards by balsa raft into Polynesia. The arguments have raged for years but the evidence for transport in both directions before AD 1500 gets firmer every decade. Drifting downwind on an inefficient boat, however, is an inefficient way to colonize new islands because there is no return (Irwin 1992).

Just drifting in the Kuro Siwo current from Japan to the northwest coast of America takes 3-4 months, and many Japanese fishermen have survived this journey over the past few centuries (Hornell 1945). There are Japanese words in the coastal dialects of the northwest American Indians and ancient planked boats of Austronesian type on the islands off the coast of California. The current continues southwards along the Californian coast almost to the equator. One of the most controversial archaeological records is the discovery of the Valdivia complex on the coast of Ecuador, with pottery dated about 3200 BC resembling that in northeast Asia at that time (Estrada and Meggers 1961). In my view, this Pacific crossing was possible as a way of no return, and therefore unattractive, but some plants, the chicken (Langdon 1989) and craftsmen skills may have travelled this way. As a branch on this line, the current turns westward well out to sea off the coast of Oregon and regularly brought pine logs from the northwest American coast to Hawaii where they were stored and used for building large canoes. Possibly people also went that way occasionally, long before the Austronesians moved into Polynesia.

Rafts were known in the Marianas, Yap, Fiji and Melanesia when Western explorers arrived. There was a persistent tradition of sea-going sailing rafts in Tonga, there were transport rafts in Mangareva, and in New Zealand there were reed boats 18 m long made of bulrushes and flax. Bamboo rafts were commonly used in Japan, Taiwan and Indonesia, some with steering by fore and aft centre boards as in South America. Rafts were Neolithic if not older and perhaps humans are today all members of one species because gene-pools were continually mixed by raft crews.