124 



SCIENCE 



[N. S. Vol. XL. No. 1021 



yards aad nnirseries, and land occupied by farm 

 buildmgs. 



TABLE II 



States Bast of the Mississippi Biver 



States West of the Mississippi Biver 



West North 

 Central . . 



West South 

 Central. . 



Mountain . . . 



1879 

 1889 

 1899 

 1909 



1879 

 1889 

 1899 

 1909 



1879 

 1889 



1879 

 1889 



26.8 

 29.1 

 24.8 

 23.1 



13.4 

 20.0 

 20.6 

 15.9 



21.1 

 21.3 



22, 

 26.5 



18.6 

 17.3 

 18.7 

 21.7 



24.0|17.7 



28.9 

 30.9 

 32.0 

 27.5 



17.0 

 20.2 

 25.8 

 21.4 



1.32 

 1.26 

 1.34 

 1.33 



0.82 

 1.35 

 1.48 

 1.03 



1.13 

 1.36 

 1.59 

 1.73 



0.47 

 0.41 

 0.39 

 0.27 



30.5 1.45 

 28.4 1.49 

 31.4ll.44 

 35.3|1.73 



If tlie area of cereals bore a constant ratio 

 to all improved land, the final conclusion 

 miglit have been correct in spite of tbe error 

 in method used, but this is far from the case. 

 Other crops have increased much more rapidly 

 than cereals. The area of cereals increased 3.5 

 per cent., and other crops increased 22 per 

 cent., in ten years. 



The truth is that the area of cereals har- 

 vested increased 3.5 per cent, (not 15.4 per 

 cent.) vchile the bushels of cereals increased 

 1.7 per cent. 



Another serious error involved is in the use 

 of figures for the entire United States. A 

 large amount of arid land in the Dakotas, 

 Nebraska, Kansas, Oklahoma and Texas that 

 was not farmed in 1899 is now planted to 

 crops and lowers the average yields for the 

 entire country. 



Nor is it safe to use total bushels of cereals 

 as a measure of production. The normal yields 

 of oats and wheat in bushels are not the same 

 and the proportion of land planted to each is 

 very far from constant. 



In order to study the question, we must deal 

 with the individual crops grown in some par- 

 ticular region. The accompanying table gives 

 such a comparison with the states grouped by 

 the method used in the last census. The pro- 

 duction of cereals in bushels and averages for 

 the United States are included for compari- 

 son with the article by Hopkins, although the 

 writer does not consider either of these figures 

 safe ones to use, for reasons given above. The 

 yield of hay and forage shows a decided in- 

 crease, but again this is made up of a number 

 of crops whose normal yields are different, so 

 that a shift in kind of crop changes the yield. 



In the states east of the Mississippi Eiver, 

 comparatively little new land has been added 

 to farms in the last twenty years. For this 

 reason these states are the ones that give the 

 best information as to changes in crop yields. 

 The highest yield of cereals ever reported 

 by the census for New England, the East North 

 Central, and South Atlantic, states is the crop 

 of 1909. In the Middle Atlantic states, the 

 highest yield ever reported is for 1899 with 



