1919] Pendleton: A Study of Soil Types 477 



ciation of this is shown in the many systems of soil classification that 

 have been proposed. 



Despite the foregoing facts that have been obtained showing the 

 divergent properties of different samples of one type presumably 

 alike, yet it must be admitted that soil surveys, even such as are no 

 more refined than those of the Bureau of Soils, have considerable 

 value for field use. 



It is felt that the additional effort required to modify the practices 

 of the Bureau of Soils in the mapping and classifying of soils would 

 be more than justified by the increased accuracy and usefulness of 

 the maps. To point out some of the causes of the present practices 

 and to give suggestions for possible methods of improvement, the 

 following discussion of the Bureau of Soils methods has been 

 ]) re pa red. 



Discussion of tin Bureau of Soils' methods. — The methods of map- 

 ping and classifying soils, as devised and used by the Bureau, have 

 resulted from some definite and important considerations. 



1. The necessity for keeping down the cost of surveying and map- 

 ping prevents the use of laboratory and culture methods in the study 

 of the soils classified, even if it were not for the fact that one of the 

 outstanding policies of the Bureau apparently denies the validity of 

 such studies in the classification of soils. This does not include the 

 mechanical analysis of soils, "which is not a separate laboratory deter- 

 mination, but a method of checking the field man's decision as to the 

 texture. It should also be added that some of the reports as published 

 in the Field Operations of the Bureau of Soils, for 1913, show the 

 subdivision of the soils into two groups based upon the CaCO s content. 

 Keeping clown the cost has also prevented the use of sufficient time to 

 map the soils correctly, even according to the criteria admittedly of 

 value in the system adopted. Many of the other methods of classify- 

 ing and mapping soils, even if applicable to most of the agricultural 

 regions of the United States, would be absolutely out of the question 

 on account of cost. 



2. The large and widely diversified area of the United States, and 

 the attempt to map representative areas in various parts of the coun- 

 try, early led to difficulties. There seemed to be a lack of understand- 

 ing as to what criteria to use in the classification of the soils. Re- 

 cently, some of the areas first mapped in the state of California have 

 been resurveyed. The texture, series, and province differences of the 

 early mapping seem not to have been clear. For example, we may con- 



