Circular No. ii. November 1907. 



MASSACHUSETTS AGRICULTURAL EXPERIMENT STATION. 



CHEMICAL ANALYSIS OF SOILS. 



WILLIAM P. BROOKS, Director. 



Every year brings to the Station numerous inquiries in relation to 

 the chemical analysis of soils, as well as a considerable number of 

 requests for such analyses. It seems best, therefore, to present a 

 brief statement as to the probable value of such analyses, and to state 

 the policy which will be followed in relation to work of this character. 

 There exists much misapprehension as to the value to the indi- 

 vidual of a chemical analysis of the soil which he is to use for crops. 

 Persons writing in relation to this matter as a rule make a statement 

 to the effect that they desire a chemical analysis in order to learn 

 what fertilizer is needed. The results of such analyses do not, as a rule, 

 afford a satisfactory basis for determining manurial requirements. The 

 chemist, it is true, can determine what the soil contains, but no ordi- 

 nary analysis determines with exactness what proportion of the sev- 

 eral elements present is in available form for the crop. Indeed, there 

 is no such thing as a constant ratio of availability. While one crop 

 may find in a given soil all the plant food it requires, another may 

 find a shortage of one or more elements. Further, on the very same 

 field, one crop will find an insufficient amount of potash, another 

 may find enough potash for normal growth, but insufficient phos- 

 phoric acid, and a third may suffer from an insufficient supply of 

 nitrogen. 



Most of our soils are of mixed rock origin, and as a rule possess 

 similar general chemical characteristics, provided they have been 

 farmed under usual conditions. The manurial and fertilizer require- 

 ments are determined more largely in most instances by the crop 

 than by peculiarities in the chemical condition of the soil. The 

 chemical analysis of soils then does not, as a rule, afford results 

 which have a value commensurate with the cost, and as a rule this 



