Certain Minnesota Soils. 397 



for the red sandstone soil in New Haven, while the next in 

 order is about the same as that of Satterly's Cambridge soil, 

 the amounts increasing until the highest is nine times the New 

 Haven value. The thorium variation is not so great, but it is 

 very marked, and apparently quite independent of the radium 

 content. 



This variation in radio-activity is probably not due to a very 

 wide dissimilarity in the actual radio-active content of the soils, 

 but rather to varying ratios of " free " emanation to total ema- 

 nation produced. This, in turn, depends on the superficial 

 area of the particles in the soil. In his book, " The Soil," A. 

 D. Hall states that a coarse, sandy soil (like Nickerson) has a 

 superficial area of about 11,000 square feet per cubic foot, while 

 a very fine soil (like Cyrus) may have as much as 70,000 square 

 feet per cubic foot ; these limits correspond sufficiently closely 

 to the observed radium limits, 13 and 80. 



It so happened that the soils examined are sharply differen- 

 tiated into two groups from the standpoint of fertility. Those 

 marked "A" are extremely fertile, while the "B" soils range 

 from distinctly mediocre to low fertility. A sharp break in 

 both the radium and thorium -contents will be noticed at the 

 boundary between the two groups. This accords very well 

 with the results of the experiments in artificial enrichment, 

 mentioned in the introduction. The conclusion seems justifi- 

 able that the hygroscopic properties of a soil are a valuable 

 index of its fertility, not merely because the presence of mois- 

 ture has of itself a beneficial effect on vegetable growth, but 

 also because the presence of moisture increases the radio- 

 active emanating power of the soil, and because, furthermore, 

 the ability to retain moisture implies fineness of texture, which 

 also means great emanating power, so that at least part of the 

 effect is due to a greater proportion of "free" radium and 

 thorium emanations. 



Summary. 



A method has been devised by which the radium content of 

 a soil may be determined from a sample of about fifty pounds. 



This method, together with the thorium method, previously 

 described, was used to determine the radio>active content of 

 some thirteen typical Minnesota soils. 



Without exception, the very fertile soils were found to be 

 richer in radium and thorium emanations than soils of inferior 

 fertility, which is in accord with the results of experiments on 

 artificial enrichment of soils. 



In conclusion I wish to express my hearty thanks to Dr. F. 

 J. Alway of the Department of Agriculture for his invaluable 

 assistance. 



Physical Laboratory, University of Minnesota, January 9th, 1915. 



