346 



The average crop divided by the average rainfall of the preceding 

 year shows that each inch of rain corresponds to about 800 hogs- 

 heads in the resulting crop; the extreme limits of variations are 713 

 and 877 hogsheads, so that in general Governor Rawson proposes 

 to predict the crop that will be gathered during the dry season, 

 February to May, each year by simply multiplying the ramfall 

 of the preceding calendar year by 800. The average uncertainties of 

 the crop thus predicted is very small, the extreme error being 28 

 per cent positive following the wet year 1861 and 4 per cent negative 

 for a certain dry year; therefore as an improvement on this method 

 he adopts the rule of adding 7 per cent for wet years and subtracting 

 7 per cent for dry years, the average year being that which corre- 

 sponds to 55 inches of rainfall. 



In supplementary calculations Rawson and Walcott show the 

 chances of a good crop as calculated from a large, small, or average 

 rainfall, respectively, for each month of the year, but I do not find 

 that they have at any time compared the crop with the total rainfall 

 for the whole eighteen months or growing period that immediately 

 preceded the crop, which comparison I have therefore made and 

 give in Table III. 



From all which it appears that large rains gives large crops, but 

 occasionally much smaller rains do also, so that it may reasonably 

 be suspected that here, as elsewhere, the sunshine must be considered ; 

 probably large rains are only of advantage when they occur at such 

 a time that they do not diminish the sunshine and in such a manner 

 that they do not wash the soil too severely. 



It would have been desirable to have stated these crops as yields 

 per acre rather than as total crops, but I find no stateihent of the 

 actual acreage in cane. Rawson gives only the total areas of the 

 six divisions of the island, Avhich sum up 107,000 acres; probably 

 two-thirds of this is planted in sugar cane, so that an inch of annual 

 rainfall corresponds to touott? oi' one-ninetieth of a hogshead of 

 sugar per acre. 



It is, however, more proper to reason upon this matter as follows : 

 Eleven j^oor crops- gave, according to Table I, an average deficit of 

 15 per cent; 12 good crops gave an average excess of 14 per cent; 

 the average rainfalls were 55.15 and 58.18, respectively. Therefore 

 an increase of 1 inch in rainfall corresponds to a gain of ^, or 10 

 per cent of an average crop. 



