132 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 146 
of what they were doing. The lands they had passed through in north- 
eastern Asia were in all probability occupied sparsely by small mi- 
gratory bands. Once across the Bering Strait, similar small bands of 
hunters would probably encounter each other with sufficient frequency 
to simulate the demographic situation they had left behind, and to ob- 
scure the actuality that they were invading mountains, plains, and 
river valleys where human beings had never passed before. 
By about 5000 B.C. the immigrants had learned to exploit the fish 
and shellfish of seacoasts, and the fauna and flora of forests and 
plains throughout the Americas. In the process of making maximum 
use of the foods naturally available, they learned to conserve and 
gradually to improve what they found. The earliest surviving evidence 
of this trend comes from Mesoamerica, where diminutive cobs of 
maize and seeds of squash and beans have been found in preceramic 
deposits dating between 5000 and 4000 B.C. In South America, the ear- 
liest known indications of incipient agriculture come from the shell 
middens of coastal Peru, where lima beans and bottle gourds appear 
between 3800 and 3000 B.C. The difference of some 1000 years be- 
tween these initial dates leads to the suspicion that the impulse toward 
cultivation may be traceable to a Mesoamerican influence, the first of 
several to be felt on the Pacific coast of South America. 
The origin of domesticated cotton, present on the Peruvian coast 
after about 2500 b.c, offers a different kind of problem. Although 
possessing Asiatic features, its derivation via transpacific contact has 
not met with general acceptance among American archeologists. New 
evidence of Jomon-like pottery on the coast of Ecuador, with an ini- 
tial date of between 3000 and 2500 b.c, makes it necessary to re- 
examine the question. It could be argued that this plant was introduced 
together with pottery onto the coast of Ecuador, where climatic con- 
ditions rule out its preservation, and that it then diffused southward 
to coastal Peru. The fact that pottery making did not also diffuse does 
not invalidate this hypothesis; it can easily be imagined that cotton 
for fishing line and nets contributed more to the security of life than 
pottery, which was more difficult to make and more easily broken than 
other types of containers already in use. What we need to know is 
whether cotton was utilized in Japan as early as 3000 B.C. 
Maize appears on the coast of Peru in a preceramic context around 
1400 B.C., but it had no significant cultural effect, the variety perhaps 
being poorly adapted to the arid coastal environment. This time we 
can be rather sure that the source was Mesoamerica by inference from 
archeological evidence on the coast of Ecuador. Here a complex of 
distinctive ceramic traits was introduced around 1500 B.C. possibly 
