EXPERIMENTS WITH ALFALFA AT YUMA 3 
siderable falling off in yields of alfalfa on parts of the project as 
compared with former years. Where such a condition occurred, it 
was observed that fields 3 or 4 years old gave the smallest yields of 
hay. A study of the soil indicated that the decrease in yields was 
not associated with alkali or a shallow water table, but was associated 
with a soil that was underlain by pure sand at depths ranging from 
14 to 34 inches. Blair suggests that this behavior of the alfalfa may 
indicate that the crop depends upon the soil to this depth for the 
first two years, by which time the taproots have reached the sandy: 
stratum where moisture conditions are not sufficiently favorable to 
support luxuriant growth. He further suggests that on such soils it 
may be necessary to plow under a crop of alfalfa every two or three 
pears and put the land into a cultivated crop before reseeding to 
alfalfa. 
Several years later complaints were general to the effect that land 
which had produced satisfactorily previously showed very marked 
decreases ae these fields were plowed, put into some other crop 
for a year or two, and then sown again to alfalfa. About the same 
time there were urgent requests for investigations to determine the 
causes of the apparent decline in yields of alfalfa seed. 
That there has been a decline in the yield of alfalfa is revealed by 
an examination of the statistics of the Reclamation Service which 
show a yield of 2 to 24% tons in recent years, as compared with about 
3 to 4 tons formerly (Table 1). The seed yields have not declined 
to the extent generally believed. As a matter of faet, the average 
yield for 1924 was greater than for 1911. It is true, however, that 
the yields from 1916 to 1919, inclusive, were nearly 100 pounds greater 
than since that time, with the exception of the year 1924. For the 
most part the yields during recent years compare favorably with those 
for the years 1911 to 1915. Such variations as do occur are doubtless 
largely attributable to seasonal conditions. 
EXPERIMENTAL RESULTS 
In an effort to ascertain a possible explanation for the decline in 
yields, particularly of hay, and for the shorter periods that satis- 
factory stands survive, a simple fertilizer experiment was begun at 
the United States Yuma Field Station, Bard, Calif., in March, 1919, 
on alfalfa that was sown in 1917. The test included duplicate plots 
recelving acid phosphate and sulphur, and also check plots. The 
acid phosphate was applied at the rate of 500 pounds and the sulphur 
at the rate of 200 pounds per acre. In a comparatively short time 
the effects of the acid phosphate were very marked, whereas the plots 
that were treated with sulphur showed no improvement over the 
check plots. During the season the plots receiving acid phosphate 
produced at the rate of approximately 1 ton per acre more than the 
check plots, and those receiving sulphur yielded slightly less than the 
checks. In another test the same year, plots sown that spring and 
recelying an application of acid phosphate yielded nearly a ton per 
acre more than the check plots in three cuttings. 
In January, 1920, an experiment was started including duplicate 
plots, one set receiving sulphur at the rate of 150 pose per acre, 
another set receiving acid phosphate at the rate of 625 pounds per 
acre, and a third set receiving barnyard manure at the rate of 11.2 
