1919] Pendleton: A Study of Soil Types 377 



same or similar when examined from the laboratory standpoint. For 

 example, we may take the Hanford fine sandy loam, which is one of 

 the types that has been used in the present study. According to the 

 criteria of color, mode of formation, origin (as judged by the presence 

 of mica), nature of subsoil, texture, etc., this soil has been found and 

 mapped in a number of areas that have been mapped in this state. 

 But will these various bodies of soil, from widely separated portions 

 of the state, when judged by laboratory and greenhouse studies on 

 samples as nearly representative as possible, appear to be the same or 

 similar ? 



The types selected for such a study as this should fulfil the follow- 

 ing conditions : first, they should have at least a reasonably wide dis- 

 tribution in the state so as to have been mapped in a number of differ- 

 ent soil survey areas; and second, the several types should be repre- 

 sentative of different classes of soils (clays, loams, sandy loams, etc.), 

 so that contrasts could be obtained between the types. 



In the collection of samples it was aimed to obtain representative 

 samples from each of a number of bodies of soil of the types selected ; 

 not to obtain possible variations from the ideal in any one body. In 

 the laboratory the soils were compared with regard to their physical 

 composition in the surface horizon, to their chemical composition in 

 three horizons, and to their relative bacteriological activities. In the 

 greenhouse the soils (surface horizon only) were placed in large pots 

 and their comparative ability to produce various crops was studied. 



Xo claim is made that these criteria should be the ones used in 

 determining the systematic classification of soils or in determining 

 the relative fertility of the soils. They were merely used to determine 

 how nearly the soils classed under a given type name agree from the 

 standpoints named. 



DISCUSSION OF RESULTS 



The bacteriological and chemical determinations were run in dupli- 

 cate so that the figures presented are averages. It is considered that 

 this gives fairer figures for comparison, especially since the determina- 

 tions were run on separate samples, and not on aliquots of a single 

 solution from a single sample. 



There is a very important factor which should always be kept in 

 mind especially when considering the bacteriological and greenhouse 

 comparisons. This is the factor of the probable error. Though the 



