324 ANNUAL REPORT SMITHSONIAN INSTITUTION, 1919. 
under all circumstances. Only a few regions of exceptional natural 
fertility afford illustrations of permanent tillage agriculture, like the 
slopes of ‘Vesuvius, the Valley of the Nile, and the plain of Hauran, 
east of the Jordan, where a loose volcanic soil has borne rich harvests 
of wheat through many centuries. 
Little of permanent, constructive improvement of natural condi- 
tions can be claimed except for the irrigating, terrace-building 
nations, and only one nation, the Japanese, has had the wisdom to 
maintain its forests, and yet has made great extensions of the area of 
cultivated lands which originally must have been very small. Tillage 
in most places has to be supplemented by some means of preserving 
or increasing the fertility of the soil, as the rice-growing nations of 
eastern Asia so well understand. In northern countries the keeping 
of domestic animals has contributed enormously to sustain crop pro- 
duction, but in many tropical countries the methods of temperate 
regions are not applicable and new adaptations are required. 
Conditions are found in many tropical countries where our methods 
of tillage agriculture prove more destructive and less permanent than 
the milpa system. On steep slopes or on lands that do not resist 
erosion, breaking and stirring the soil may result in serious injury 
with the first heavy rain, or gradual surface erosion may leave noth- 
ing but a sterile subsoil after a few seasons of tillage. Thus it is 
possible for lands to be destroyed or seriously injured by wrong ap- 
plications of tillage systems even more rapidly than by milpa agri- 
culture. It may even be claimed that the milpa system is more per- 
manent since it can be continued indefinitely if the cuttings and burn- 
ings are not too frequent, so that grasses do not become established 
in place of the ready vegetation. 
The newcomer’s theory of tropical development is that all of the 
difficulties arise from the failure to adopt the northern crops and 
methods. The tropical crops are unfamiliar and the lands appear 
uncultivated. No plows, harrows, drills, or cultivators are to be 
seen, perhaps only a few people, working with clumsy hoes or hack- 
ing the bushes with long knives, appearing so strange and casual that 
any backward condition is likely to be ascribed at once to the out- 
landish methods. The idea that tillage methods must be established 
in order to improve tropical agriculture is so nearly an obsession 
with people from northern countries that little consideration has 
been given to other possibilities, such as the intensive utilization of 
tree crops by methods that would maintain or increase the fertility 
of the soil and avoid unnecessary labor.’ 
That tree crops were not much used by primitive peoples is easy 
to understand when the temporary, seminomadic character of the 
7 Possibilities of intensive Agriculture in Tropical America, Proc. Second Pan-American 
Scientific Congress, Vol. III, pp. 573-579 (1915-16). 
