EXPERIMENT STATION BULLETINS 



189 



Field 6. 



July 5th.. 

 July 26tli. 



Davenport Plot i. 



1st foot. 2nd foot. 3rd foot. 



Per cent. 



10.19 



4.03 



6.16 



Per cent, 



8.85 

 10.90 



-2.05 



Per cent. 

 11.41 

 10.47 



.94 



The Davenport plots 66 and 68 were te.sted separately and the figures given are 

 the averages of the two samples. Where samples were taken from the north end of 

 these plots the soil was mucky for the first foot in depth. The samples marked 

 third foot were in reality but six inches in depth, making a total depth of but two 

 and one-half feet instead of three feet. Except as indicated above the soil was a 

 loam with a clay .«.ubsoil. 



While these results are somewhat variable the average loss is quite marked, being 

 15.47 per cent, or more than one-seventh of the total moisture remaining in the 

 ground after the wheat was harvested. The loss must have occurred through 

 evaporation, as the clover upon the areas where the samples were taken had made 

 very little growth before July 26th. To prevent this evaporation a mulch, even if 

 no more than the stubble cut by a mower as before suggested, might be of a marked 

 benefit to the seeding. 



DRYING OUT OF SANDY SOIL BY CLOVER. 



On July 15th, 1897. an unmown clover plot was plowed for green manure. Five 

 days later the following amounts of moisture were found on tliLs plot and on an ad- 

 joining plot that had to that late date lain fallow: 



