28 BULLETIN 699, U. S. DEPARTMENT OF AGRICULTURE. 
DURATION OF EXPERIMENTS. 
Because of the numerous conditions just discussed, some of which 
-are beyond control and some of which vary in spite of the exercise 
of the greatest care, it is obviously unwise to draw conclusions from 
field experiments which have been conducted for a short time only.¢ 
Climatic conditions vary so greatly from year to year that in order 
to gain an accurate knowledge of the value of a fertilizer material 
the same crop should be grown several times on the same land. 
Where proper systems of rotation are practiced it takes from 6 to 
20 years to accomplish this end. In studying the effect of the less 
soluble fertilizers, time plays a very important part. Not only does 
the material often become more soluble, but it becomes more thor- 
oughly distributed in the soil from year to year, and hence is more 
readily available to the root systems of crops. 
RESULTS OF EXPERIMENTS. 
ALABAMA. 
No field experiments with raw rock phosphate continuing beyond 
three years have been conducted by the Alabama station. Two series _ 
of cooperative one-year experiments, however, were carried on, one in 
1891, consisting of 25, and the other in 1892, consisting of 35 experi- 
ments. In these experiments equal amounts of acid phosphate and raw 
rock phosphate were compared but they were applied at a rate (240 
to 300 pounds per acre) considerably below that at which the latter 
should prove effective. In these experiments the average yields of 
the acid phosphate plots were appreciably greater than those of the 
raw rock plots, though the latter showed considerable gains over 
the average of the checks. 
A field experiment conducted for three years was reported by the 
Alabama station in 1913.3 The general scheme and results of this 
experiment were in accord with those just mentioned, but the yield 
of the raw rock plot the third year was almost identical with that 
of the acid phosphate plot. 
CONNECTICUT. 
The work of the Connecticut station with raw ground rock phos- 
phate has been very limited, none of the experiments having been 
conducted sufficiently long to warrant repetition in detail. An ex- 
periment began in 1887 * and continued for three years® on the same 
1Ala. Agr. Expt. Sta., Bul. No. 23 (1891). 
2Ala. Agr. Expt. Sta., Bul. No. 34 (1892). 
8 Ala. Agr. Expt. Sta., Bul. No. 173, p. 189 (1918). 
Conn. Agr. Expt. Sta., Ann. Rept. for 1888, pp. 110-117 (1889). 
® Conn. Agr. Expt. Sta., Ann. Rept. for 1889 (1890). ; 
