444 ANNUAL REPORT SMITHSONIAN INSTITUTION, 1955 



took to develop this region. There was great need for development 

 of the selva to the east because a few hundred miles to the southeast, 

 in northeastern Brazil, is an area where droughts are chronic. Even 

 during recent years thousands of villagers from this drought-stricken 

 region have been going some himdreds of miles by road to Sao Paulo 

 for work because of the too rigorous conditions in the northeast. 

 When in Sao Paulo in 1949 I remember clearly seeing the truckloads 

 of peasants coming in after a week or more of dusty travel by road 

 from the northeast, where they could no longer find any way to make 

 a living because of the drought in their arid and too often rainless 

 country. 



As a start in developing this Braganga region, the government built 

 a railroad east from Belem and assisted villagers to come in from the 

 dry regions of northeastern Brazil. Settlers came in, the land was 

 cleared, and crops were grown. The first crop following the clearing 

 of the forest was often very encouraging, but thereafter the crops were 

 extremely poor. It soon became evident that it was futile to cultivate 

 these very poor sandy upland soils continuously in an effort to develop 

 a permanent agriculture on them. 



Before his most untimely death in 1942, Geoffry Milne, soil chemist 

 to Tanganyika Territory, British East Africa, contributed greatly 

 to a better understanding of tropical soil-plant relationships. As a 

 result of his field studies in Trinidad and British Guiana ^ he pointed 

 out that "It is true that in better known floras there are always a few 

 'indicator' plants, but all plants must indicate something, and what 

 most of them could tell about the soil is quite insufficiently known." 

 Milne continues that — 



the difficulties due to the nature of the soil-plant relationship may be stated 

 briefly thus: 



(1) However faithfully a natural vegetation may reflect the soil conditions 

 that have promoted it, we usually propose to change or remove that vegetation 

 for the purpose of our uses of the land, and we shall thereby change the condi- 

 tions, perhaps fundamentally. A soil and its plant cover have interlocking 

 identities; and what we have thought of as forest soils or prairie soils cease 

 to be such within a short period after the trees have been cut or the sod turned. 



(2) The soil properties that are significant to a natural vegetation and receive 

 expression in it are not necessarily those that will be of most significance to 

 the crop proposed for replacing it. Crop plants will usually have a more urgent 

 demand for nutrients and may have a different range of tolerances. 



Milne visited a project on the island of Trinidad where natural 

 forest of mediocre value had been cleared and the land used for annual 

 crops. The settlers of East Indian stock had not been able to continue 

 cropping the land for more than a few years because the productivity 



' G. Milne, A report on a journey to parts of the West Indies and the United 

 States for the study of soils, February to August 1938, East African Agr. Res. 

 Stat, Amani, Tanganyika Territory, 78 pp., 1940, Government Press. 



