162 THE DEVELOPMENT OF RAW COTTON 



the optimum conditions of cultivation for a certain site 

 and year can be determined. It still remains to answer 

 the question as to how the plants behaved under those 

 conditions ? 



The cotton crop is thoroughly misleading in its appear- 

 ance. A field may appear to have but little ripe cotton 



Uncertaint * n ^' anc ^ ^ e ^ ^ e ^ u ^ ^ P en b us which are 

 of Subjective obscured by the leaves. Another field may 

 Opinions. appear to be flowering profusely, and yet 

 be about to stop flowering almost entirely. The best 

 experts of those who spend their time in travelling 

 about a cotton district and reporting on the crop con- 

 dition are well aware, and will admit, that they cannot 

 estimate the yield of cotton ripe in any particular field 

 to within 10 per cent. Yet there is scarcely any crop 

 which has been so entirely discussed on purely general 

 ideas as to its appearance. 



The question arises as to whether any more accurate 

 data are worth obtaining in the early stages of develop- 

 ment, and the answer is most certainly 



Elimination a ffi rmat i ve . Take, for example, the case 

 of Accidents. 



of a rain-storm occurring at an unusual 



time in the first season of experimental work ; it is desir- 

 able to know whether the results obtained in that year 

 are still generally applicable to future years, or whether 

 they have been entirely falsified by the storm. To justify 

 expenditure on elaborate small- plot experiments, it must 

 be shown that their results can be made of general sig- 

 nificance, and applied forthwith to other possible sites 

 and seasons. 



