IMPROVEMENT OF THE SUGAR CANE. 123 
deductions from the consideration of the averages of all 
theresults. Now, with all deference, I submit that it is not 
possible to average discordant results. I have therefore, 
for my present purpose, worked out the results, omitting 
the returns which were obtained on two estates over two 
crops only, and those on plots which on the average of 
the five crops gave results below those obtained on the 
corresponding not manured plots, and which, therefore, 
we may fairly assume to have been of markedly lower 
natural fertility than the control plots. In each case the 
increases have. been calculated against the yields from 
the corresponding not manured plots and not against 
the average yields of all of them. The following are the 
results 1 have thus arrived at :— 
Average increase in tons of 
Bree No. of plot. canes per acre per annum on 
limed and not limed plots. 
Nitrate of soda 13 cwts, 
249 lbs. nitrogen. 3 5°83 
13 cwts. superphosphate 
25 0/o soluble. 
Nitrate of soda 3 cwt. 
49'8 lbs. nitrogen. ; 
13 cwts. superphosphate, 
Sulphate of ammonia | 
? 
602 
cwt. 23°06 lbs. nitro- 
ree 
3 cwt. superphosphate. 
fait: of ammonia 2 
cwts, 46°12 lbs. nitro- 
gen. 
1% cwt. superphosphate. 
Nitrate of soda 14 cwt, 
5 : 5°62 
Superphosphate 3 cwts. 5 f 5°49 
Sulphate of ammonia 1 
cwt. 5 6°09 
Superphosphate 3 cwt. 
These give us as the average annual increase, for in 
Qa 
