THE ROTHAMSTED EXPERIMENTS 355 



regular amounts; that is, 600 pounds of potassium sulfate, 200 

 pounds of magnesium sulfate, and 400 pounds of sodium sulfate. 



In 1896 and in 1900, 600 pounds of basic slag phosphate were 

 applied instead of 500 pounds of acid phosphate. 



The sixty years' data from Agdell field are exceedingly valuable 

 in the study of many important soil fertility problems. No ex- 

 haustive discussion can be given here, but these results will be 

 referred to for many years at least as the greatest source of informa- 

 tion concerning the effect of long-continued crop rotation. A few 

 of the plainly indicated conclusions may be noted. 



(1) On the unfertilized land the rotation of crops does not main- 

 tain the fertility of the soil, the yields of every crop having de- 

 creased with the possible exception of beans. The yield of Swede 

 turnips dropped from about 10 tons per acre in 1848 to less than 

 2 tons in 1852, and never equaled 3 tons per acre afterward. 

 That is to say, the turnips have always been grown at a loss since 

 the first year, the best yields being scarcely worth harvesting. The 

 barley yields have decreased from more than 40 bushels, 1849, to 

 15 bushels as an average of the last 20 years, but the decrease 

 has been very gradual. The yield of legumes has been very irregu- 

 lar, but, with the exception of the beans in 1898, has markedly 

 decreased, the clover from 2.8 tons in 1850 to less than one half 

 ton per acre as an average of the crops grown during the third 

 2o-year period. 



The yield of wheat has been greatly influenced by several condi- 

 tions, but during the sixty years has decreased as an average by 

 8 bushels in the legume system and by 16.5 bushels in the fallow 

 system, if we assume that the difference between the averages for 

 the first 20 years and the last 20 years represents the decrease of 

 40 years. The lowest average yield is for the second 20 years, but 

 this period includes the abnormally low yields of 1879 (when the 

 best fertilized plots averaged only 13.5 bushels) and two other 

 rather poor years. 



It should be kept in mind, too, that the wheat crop comes in the 

 next year after legumes or fallow, and thus has the most favored 

 place in the rotation. 



(2) The application of mineral plant food has as an average main- 

 tained the yields of legumes and of the following wheat crops. 



