250 General Strachey on the Correspondence of [May 31, 



Irrespective o£ its scientific interest, the conclusion thus adopted 

 would, if sound, be of no little practical importance, as it would supply a 

 means of indicating the probable recurrence of those seasons of excessive 

 drought which produce such terrible results in India, and from one of 

 which the Madras provinces are now suffering in an extreme degree. 



It is probably generally known that the conclusion which it has thus 

 been stated is to be drawn from the Madras observations had been con- 

 sidered to be established some years ago, in a more general manner, by 

 Mr. Meldrum, the Director of the Meteorological Observatory at Mauri- 

 tius, a paper by whom on this subject was read before the Royal Society 

 in 1873, and may be found in vol. xxi. of the ' Proceedings,' p. 297. 



As the numerical results of the method of treating the rainfall obser- 

 vation which Mr. Meldrum and Dr. Hunter have followed at first sight 

 may appear to support the conclusions they have adopted, it has seemed to 

 me desirable to examine the facts, with a view to arriving at an independent 

 opinion as to the trustworthiness or otherwise of those conclusions. I 

 shall first refer to the Madras observations and Dr. Hunter's results. 



The Madras register extends over sixty-four years, beginning with 1813. 

 The mean rainfall for the whole period is 48*5 inches. The deviations 

 from the mean vary from 30*1 ins. in defect to 39*9 ins. in excess. The 

 arithmetical mean of these deviations (disregarding the signs) is 12*4 ins. 

 The greatest difference between two consecutive years is 50*5 ins., and 

 the average difference 15*8 ins. 



Dr. Hunter, in order to test the point which he proposes to investigate, 

 divides the sixty-four years' observations into six cycles of eleven years, 

 and calculates the arithmetical mean of the successive years of the whole 

 series of cycles. The results are shown in the following Table, the 

 figures in which represent the differences of the several years' observations 

 from the mean of the whole : — 



Table I. 















Years of cycle. 











Number of cycle. 



















































1st. 



2nd. 



3rd. 



4th. 



5th. 



6th. 



7th. 



8th. 



9th. 



10th. 



11th. 





/1st cycle 



in. 

 - 3-4 



in. 

 -16-1 



in. 

 + 7-5 



in. 

 - 7-4 



in. 

 + 15-1 



in. 

 4-277 



in. 

 -12-2 



in. 

 4-21-5 



in. 

 - 1-4 



in. 

 +1M 



in. 

 -21-9 



m 



2nd „ 



-14-8 



+ 7-5 



4-12-2 



+39-9 



-10-6 



-11-6 



-16-1 



-4-2 



-30-1 



-11-4 



- 9-5 



o 



3rd „ 



- 70 



- 3-8 



4- 0-8 



+ 3-8 



-f 4-6 



4-101 



4- 9-8 



-120 



+ 1-8 



4-16-9 



-10-5 



Orig 

 observi 



4th „ 



+31-3 



+32-5 



4- 6-3 



- 8-7 



-11-6 



4-15-8 



4-24-2 



-12-7 



- 5-3 



—16-2 



- 1-5 



5th „ 



4- 44 







+ 6-6 



-20-9 



-11-3 



-10-3 



4- 6-1 



- 1-3 



- 6-9 



4- 2-9 



-241 





v 6th „ 



- 7-1 



-16-2 



+25-6 



4- 7-8 



+25-2 



4- 3-3 



4-14-4 



-11-4 



-270 







Mean difference 1 

 from the mean I 

 of 64 years ... J 



+ 0-6 



4- 0-7 



4- 9-8 



+ 2-4 



+ 1-9 



4- 5-8 



4- 4-4 



- 3-4 



-1.1-5 



4- 0-7 



-13-5 



