The Fertilizers Used. 



The following prices are used, as representing approx- 

 imately the average cash price in local markets during the 

 last few years : 



Per Ton. 



Acid phosphate (14 per cent, available) $14.00 



Cotton seed meal $30.00 



Kainit $14.00 



Prices naturally vary in different localties. Any one can 

 substitute the cost of fertilizers in his locality for the prices 

 given above. 



In each experiment three plots were left unfertilized, these 

 being plots 3, 7, and 11. When these yields differed widely 

 the experiment was classed as inconclusive The increase on 

 plots 4 to 6 is calculated on the assumption that the grada- 

 tion in fertility is uniform from plots 3 to 7; likewise the 

 increase is calculated for Plots 8 to 10 inclusive.* The fol- 

 lowing- table shows what kind and amounts of fertilizers 

 were used on certain plots; the number of pounds of nitro- 

 gen, phosphoric acid, and potash supplied per acre by each 

 fertilizer mixture; and the percentage composition and cost 

 per ton of each mixture, the latter being given in order that 

 these mixtures may be readily compared with various brands 

 of prepared guanos. 



*In other words instead of calculating the increase merely 

 by subtracting the yields of any plot from the average yield of 

 the three unfertilized plots (which would be incorrect and mis- 

 leading unless all three unfertilized plots afforded practically 

 the same yield), the following method is used as a means of 

 making allowance for variations in the natural fertility of the. 

 different plots: — 



(-1). The difference between the yields of unfertilized 

 plots 3 and 7, or between unfertilized plots 7 and 11 is divided 

 by 4, because this difference must be distributed over the four, 

 intervening plots. 



(2). This quotient is then added to the yield of the poor- 

 est Of this unfertilized pair, thus giving the corrected or calcu- 

 lated yield (if unfertilized), for the fertilized plot adjacent to 

 the poorest unfertilized one. Similarly the yield of the poorest 

 unfertilized plot is increased by twice and three times the abore 

 quotient as a means of calculating the corrected unfertilized yield 

 on the plots occupying respectively second and third, positions from 

 the poorest unfertilized plot of the pair. 



(3) Now these calculated yields (if the plots were unfer- 

 tilized) are subtracted in regular order from the corresponding 

 actual yield, thus giving the most accurate measure known for the 

 increase due to tie fertilizer. 



