Univicrsitv of Arizona 347 



of the percolate. This involves much analytical work. Attempts 

 were made to parallel the laboratory studies by pot cultures which 

 at first proved unsatisfactory due to the difficulty of preventing 

 leaching when the j)Ots were irrigated and the consequent change 

 of concentration of alkali in the soils. Successful pot culture 

 studies in this climate require that the pots be sunk in soil to pre- 

 vent too high temperature and excessive drying. Benches have 

 now been constructed in the screened garden in which the pots 

 Tire sunk in sand at the level of the surrounding soil and any perco- 

 lating water due to heavy irrigating is caught in receptacles and 

 returned to the pots. The pots are paraffined to prevent losses by 

 transfusion. 



The percolation experiments with gypsum have been espe- 

 cially interesting. When a percolation test is made comparing 

 untreated University Farm soil with samples to which the theoreti- 

 cal amount and half that amount of gypsum have been added, it is 

 found that the second half of the gypsum applied has two or three 

 times the elTect of the first half in promoting percolation. Large 

 plot experiments are now being conducted which are planned to 

 test this result in a practical way. Several lands at the University 

 Farm have been divided into numerous small plots each of which 

 has been analyzed to a depth of three feet and the necessary 

 amount of gypsum calculated separately for each plot of 1500 to 

 2000 square feet. These lands had been treated previously by 

 applying gypsum uniformly over the surface, but without reclaim- 

 ing them successfully. After the proper amount of gypsum has 

 been applied, the lands will be leached by confining the water 

 on the more alkaline areas. Without gypsum, percolation is very 

 slow, altho the soil is a very fine sand. In the laboratory water 

 applied an inch deep to the wet soil in 10-inch flower pots and 

 covered to prevent evaporation has stood for two or three weeks 

 without entirely disappearing. It appears from the investigations 

 in the laboratory that light or insufficient applications of gypsum 

 would be unprofitable. On some areas it is necessary to apply 30 

 or even 40 tons of gypsum to the acre. Under some conditions 

 this would be prohibited, and never could be considered for large 

 areas. Gypsum beds, however, are available near the University 

 Farm and hauling is done by the farm teams when other work is 

 light. Small areas of black alkali in otherwise good lands, as is 

 the condition at the University Farm, would often warrant the 

 expenditure of several hundred dollars for reclamation. How per- 

 manent the effect will be remains to be shown. The groundw^aters 



