SOILS OF THE ARID AND HUMID REGIONS. 391 



which no other form of combination can be assigned. Since some of 

 the alumina undoubtedly takes part in the formation of such zeolites, 

 the silica must to that extent be withdrawn from the estimate made for 

 kaolinite. While it is impossible to make any definite numerical al- 

 lowance for this fact, it clearly will tend in many cases to increase 

 materially the amount of alumina that must be assigned to the hydrate 

 condition. It will be noted that in most cases given, the alumina per 

 cent is rather large. 



The relatively large number of such cases shown in the table 

 for California soils is not a matter of accident; for even a cur- 

 sory glance at the columns of analyses of California (and 

 Washington and Montana) soils, shows that the cases in 

 which the alumina exceeds the silica in amount are rather pre- 

 dominant, while the reverse is the case in the humid region. 1 

 But it must not be inferred that the reverse relation is not also 

 frequently observed even in the arid region ; it occurs in fact 

 in close proximity to the localities where some of the most 

 striking instances of excess of alumina over soluble silica have 

 been found. 



Thus Nos. 86 1 and 863 from the neighborhood of Grass Valley, 

 which show this excess most strikingly, occur within 15 miles of local- 

 ities which show almost the reversal of the numbers given for the two 

 former, and at a level of about a thousand feet lower. It would seem, 

 on the whole, that the excess of alumina occurs most frequently in con- 

 nection with soils formed from eruptive rocks ; in the case referred to, 

 from volcanic ash. It will require more detailed study to detect the 

 causes of these marked differences. 



Retention of Soluble Silica in Alkali Soils. It is somewhat surprising 

 that, contrary to the expectation one would naturally entertain, the 

 alkali lands, so frequently rich in the carbonates of the alkalies that 

 would dissolve free silica, on the contrary, show most frequently an 

 excess of soluble silica over alumina. This is probably to be explained 

 from the very liberal opportunities afforded in the alkali soils for the 

 formation of complex zeolitic masses by the retention in soil of the 

 soluble alkali salts, and the abundance of lime always present in them. 

 As already stated, we usually find in alkali soils a very large proportion 



1 See for comparison the data given in vols. 5 and 6 of the report of the Tenth 

 Census of the United States. 



