A Fertility Program for Celery Production


increase in yield over that of the 6%; fertilizer; and finally, that
an 18% potash fertilizer produced a slightly higher yield in
three out of five years than did the 12% formula, but not enough
higher, in the average year, for the yield increase to pay for
the additional potash applied per acre. This can serve to
illustrate the economic law of diminishing returns as applied
to fertilizer usage.
 From the standpoints of both yield and net profit per acre,
 therefore, a fertilizer analyzing 12% potash, properly balanced
 with nitrogen and phosphate, is suggested for celery. The ratio
 of nitrogen to potash which has given best results has been
 found to be 1 to 4. The ratios of phosphate to the other two
 nutrient materials will be discussed in the next section. Muriate
 of potash is recommended as the most economical potash source,
 although sulfate of potash and hardwood ashes are suitable in
 all respects except their higher cost per unit of potash.

 INFLUENCE OF PHOSPHATES (PHOSPHORIC ACID)
 UPON YIELDS AND COSTS
 These peat and muck soils are also very low in phosphates,
 perhaps even lower than in natural potash content. Further-
 more, the availability of phosphate to celery-or to any other
 plant-is dependent upon other factors, largely the reaction and
 calcium content of the soil. At the normal soil reactions usually
 found in this region (slightly acid to slightly alkaline) part of
 the phosphate forms a relatively insoluble compound with cal-
 cium or with some other soil constituent, and thus becomes less
 available to the plant roots.
 Many sources of phosphatic fertilizers are used and varied
results may be expected on different types of soils. In Area 1
of these experiments several phosphate sources were compared,
including 16% superphosphate, colloidal phosphate, floats, for-
eign basic slag, and domestic basic slag. Of these, floats and
colloidal phosphate were found to be definitely unsatisfactory
materials, regardless of cost. The other materials all were
more or less satisfactory, the domestic basic slag in particular
showing up well. However, it is practically impossible to com-
pare these basic slag materials with the superphosphates on
the unit PO, basis, because of a rather wide variability in the
phosphate content of slags from various sources, and because
the guaranteed analyses of the slags are usually much lower