10 BULLETIN 3 4, PORTO RICO EXPERIMENT STATION 
lowest and the highest yields, respectively, throughout the course 
of the experiment. 
Of the five plats, No. 3 showed the greatest increase in yield. The 
four fertilized plats differed only in regard to potash, its omission, 
form, or replacement by sodium. If the notable increase in yield of 
plat No. 3 were due to the application of potassium chloride, an in- 
crease in yield would be expected to follow either the application of 
potash in the form of potassium sulphate in plat No. 2 or the applica- 
tion of chlorine in the form of sodium chloride in plat No. 4. A com- 
parison of the total yield during the second and third periods with 
that of the first period shows that plat No. 1, receiving no potash, 
made practically the same increase in yield as plat No. 2, which re- 
ceived potassium sulphate. Certainly no benefit appeared to have 
resulted in plat No. 2 from the application of potash. The ratios 
of total production of first period to that of combined second and 
third periods were as 1 to 2.926 for plat No. 1 and 1 to 2.922 for plat 
No. 2, which is a difference of less than one-half of 1 per cent. The 
ratio for plat No. 4, which received sodium chloride, was 1 to 2.62. 
This plat showed very little increase in yield throughout the dura- 
tion of the experiment and gave the lowest yield of any plat in the 
second and third periods. Since in the experiments at the San Jose 
plantation the second highest production was from the plat which 
received sodium chloride alone, it would be unreasonable to attribute 
the poor production of plat No. 4 to the application of sodium 
chloride. In view of these contradictory results it would be illogical 
to ascribe the increase in yield in plat No. 3 to the application of 
potassium chloride. 
The unfertilized plat gave the highest yield of any in both first 
and third periods, but was surpassed by one plat in the second period. 
The production of the second and third periods as a unit showed a 
slight decline for the check and an increase for all fertilized plats. 
That a wide variation existed between the individuals comprised in 
a plat is shown by dividing the plats into north and south halves 
and considering each half plat as a unit. A comparison of the yield 
of the last with that of the first period shows that the north half 
of plat No. 1 decreased 10 per cent, whereas the south half 
increased 75 per cent. In like manner, the halves of plat No. 2 
showed increases of 4 and 22 per cent, respectively; plat No. 3, 
increases of 60 and 67 per cent ; plat No. 4, a decrease of 11 per cent 
in the north half and an increase of 30 per cent in the south half; 
and plat No. 5, an increase of 5 per cent in the north half and a 
decrease of 4 per cent in the south half. Thus, a group of 9 or 10 
palms receiving the same fertilizer treatment as a second group of 
9 or 10 palms responded very differently from the latter in most of 
the comparisons. These inconsistencies are such as to indicate the 
presence of a potent factor or factors veiling the effects of the ferti- 
lizers. If the difference in yield between the first and last periods 
be calculated on an annual basis, it is seen that the increase per palm 
was 6.2 nuts for plat No. 1, 4.4 nuts for plat No. 2, 13.8 nuts for plat 
No. 3, 1.8 nuts for plat No. 4, and 0.4 nut for plat No. 5. An average 
annual increase of less than 7 nuts per palm followed annual applica- 
tions of 8 to 12 pounds of high-grade fertilizer costing from 15 to 25 
cents at the local fertilizer agencies. An additional cost of 5 cents 
