METHODS OF SOIL CULTIVATION . 1417 



farming to the maintenance of crop yield have been made on 240 farms in 

 Chester Countj^ Pa., and on 303 farms in Central Illinois. Use has also 

 been made of the data compiled for 377 other farms in Chester, Co. Pa. 

 and 300 farms in Ivcnawee Co. ]Mich. The studies consisted in comparing 

 the productive capacity of the various farms by means of their " crop 

 index " which ma}' be defined as the crop yields of a particular farm ex- 

 pressed in percentage of the average crop yields in the community. 



Results of the investigations indicated : 



i) That there seems to be in all districts so far studied an optimum 

 percentage of the crop area of the farm which' can be devoted to a single 

 class of crops and maintain maximum }delds. Even under the rather in- 

 tensive types of farming prevaiUng in Chester Co. Pa. and in lycnawee Co. 

 Mich., and imder the more extensive type of farming in Central Ilhnois, 

 the optimum area of intertilled crops in each case falls wdthin 5 per cent 

 of each other, the range being about 32 to 36. 



2) That when more than this percentage of area is devoted to a single 

 class of crops, yields decrease even where there is an increase in the number 

 of live stock per acre. 



3) That in Chester Co. Pa. the optimum percentage area for peren- 

 nial grass (timothy and clover) is about 36 per cent of the crop area of the 

 farm. 



4) That data of tliis character make it possible to construct a crop- 

 ping system which should under average conditions, produce ihaximum 

 yields with a given quantity of manure. In the case of Chester Co. such a 

 system constructed from' data brought out in the 1912 survey corresponds 

 very closely to the practice of those farmers who maintain high yields and of 

 those who have made highest profits. 



These conclusions were arrived at as follows : the relation between 

 the crop yield, the percentage of the crop area of the farm in intertilled 

 crops, and the average number of live stock per acre was determined for 

 Chester Co. Pa. and for Central Illinois (Tables I and II). 



The relation of non intertilled crops to crop yield was next investiga- 

 ted (Table III) and finally that of perennial grass (timothy and clover) 

 to crop yield (Table IV). 



The results show that the best distribution of area amongst the diffe- 

 rent classes of crops occurs when 10 per cent is allowed for crops not in the 

 rotation such as garden, orchard, .soiling crops etc., 36 per cent is under 

 intertilled crops, 36 percent in perennial grass and the remaining 18 percent 

 in annual crops not intertillied. To illustrate how nearl)' this corresponds 

 to the actual practice of the most successful farmers. Table V shows the crop 

 area percentages of the 27 Chester Co. farms having the highest yield and 

 of the 27 having the lowest yield. There is a close agreement between the 

 cropping systems .selected for high crop yields andt hat producing the high- 

 est profits and this agreement indicates that a well balanced type of agri- 

 culture jjrevails in Chester Co. Where this condition does not prevail 

 high yields are obtained to the sacrifice of profit or high profits are made at 

 the sacrifice of soil fertility. 



