90 BULLETIN 699, U. S. DEPARTMENT OF AGRICULTURE. 
In Table LIV the average annual increase per acre for each treat- 
ment is figured both by the Ohio system and also by comparing the 
treated plots with the average of all the check plots. These two 
systems of comparison lead to very different conclusions. By the 
Ohio method acid phosphate appears to much greater advantage than 
raw rock, even though the latter is considerably less costly. By the 
second method of comparison the applications of raw rock phos- 
phate show increases in yield practically as good and in some cases 
better than those obtained from equal applications of acid phosphate. 
Another point worthy of serious consideration in this experiment 
is that the applications of raw rock were not very liberal for such a 
relatively insoluble material, the effectiveness of which depends 
largely on its thorough distribution in the soil. Such distribution 
must be brought about principally by heavy applications. Accord- 
ing to analysis, the soil on which this experiment was conducted was 
relatively low in phosphoric acid (0.08 per cent P,O;) yet even had 
none of the phosphate rock added during the 17 years of the 
experiment been removed by crops or lost in other ways, the total 
quantity of phosphoric acid present in the upper 9 inches of the 
soil would only have been increased to 0.1 per cent. The crops grown 
were such as would remove an amount of phosphoric acid equal ap- 
proximately to one-half of that added in the form of raw rock so at 
the expiration of 17 years the soil should have contained about 
0.01 per cent more phosphate than at the beginning of the experi- 
ment, an amount which in spite of thorough tillage and frequent cul- 
tivation was probably not uniformly distributed in the soil. An ex- 
periment similar to the one just described, except using heavier ap- 
plications of phosphate rock would be interesting, and yield valuable 
data on this subject. 
Another experiment with raw rock phosphate in which a ie 
year rotation of corn, oats, and clover was followed, was begun at 
Wooster in 1905.1. The following description of the field employed, 
and the history of the experiment which is known as the Lime and 
Floats Test, is taken directly from Circular No. 144 of the Ohio 
Station: 
The land had been under the regular rotative cropping of the farm since its 
occupation by the Station, and for a considerable period before, and was in good 
condition—12 tons of manure per acre had been plowed under for corn in 1904. 
Three sections of 26 plots each are included in the test, the plotS containing 
one-twentieth acre each. 
For the crops of 1905, section A (north end) was manured at the rate of 6 
tons per acre only (because of the recent application above mentioned), limed, 
fertilized, and planted to corn. Section B was sown to soy beans instead of 
clover, the beans to be followed by rye in the fall and corn in 1906. Section C 
1Qhio Agr. Expt. Sta., Circular No. 144, pp. 92-93 (1914). 
