540 ANNUAL REPORT SMITHSONIAN INSTITUTION, 1961 



tinents, whereas these same groups in recent times are restricted to a 

 part of one continent only. We must conclude that the distribution of 

 Gocos nucifera does not establish anything in favor of Heyerdahl's 

 theories, since no certainty exists as to its original area. Heyerdahl 

 also agrees with tliis conclusion. Buck (1937) cited a Polynesian 

 myth in which the coconut is said to have originated from the head 

 of the demigod Tuna after he was beaten and killed; the nut still 

 shows the mouth and eyes of Tuna. This story of course does not 

 cast any light on the situation. Eiesenfeld (1951) has called atten- 

 tion to the fact that archeological investigations in Peru have never 

 brought to light the remains of coconuts. 



Up to the present, only the three useful plants discussed above can 

 cast doubt upon the final conclusions of Alphonse de Candolle (1883) 

 in his classical work "Origine des plantes cultivees." De Candolle, in 

 that work (English translation of 1885, pp. 461, 462) wrote: "In the 

 history of cultivated plants I did not find any indication as to contact 

 between the populations of the Old World and of the New World 

 before the discovery of America by Columbus." And he added, 

 "Between America and Asia perhaps two transports of useful plants 

 took place : one by man [sweetpotato] , the other either by man or by 

 currents [coconut]." 



Much dependence is placed upon cotton by Heyerdahl. In Poly- 

 nesia some wild species occur ( Gossypimn taitense Pari, and G. toTnen- 

 tosum Nutt.), and Heyerdahl here refers to the investigations of J. B. 

 Hutchinson, R. A. Silow, and S. G. Stephens (1947). These investi- 

 gators found that the Old World species of cotton possess a haploid 

 number of 13 large chromosomes, the wild American species of cotton 

 have 13 small chromosomes, and the cultivated American cotton 

 has 26 chromosomes: i.e., 13 large and 13 small ones. This means, 

 according to them, that the cultivated American cotton is allopolyploid 

 and originated by hybridization of Asiatic and American cotton. 

 Since the cultivated cotton was known in America in pre-Columbian 

 times, they assumed that the old, civilized, American populations 

 introduced cotton on their voyages from Old World countries, and 

 then developed the hybrids. Heyerdahl agrees with this conclusion 

 and thinks it probable that this cotton reached America by the south- 

 ern Atlantic. Hutchinson, Silow, and Stephens, moreover, stated that 

 the wild Polynesian Gossypimn species, considered to be endemics, had 

 26 chromosomes as in the American-cultivated cotton, and they also 

 argued that Gossypium taitense was not a distinct species but a mere 

 form of the American G. hirsutum var. punctatum (Schum.) J. B. 

 Hutchinson et al. According to Heyerdahl we can thus arrive at only 

 one conclusion: the migrating Peruvian population brought with 

 them, from Peru, the cultivated cotton. In Polynesia, later on, the 

 custom of cotton spinning was lost and the Polynesians took up the 



