THE CORRELATIONS OF AREAS OF MATURED CROP AND THE RAINFALL. 385 



PART III. 



CORRELATIONS OF THE SEASONAL RAINFALLS. 



§ 10. From the work done in Part I of the paper, it was easy to deduce the 

 correlations of the total rainfall in April to September with the total for October to 

 March. For the 20 years discussed the results were : — 



Sialkot 



Zafarwal 



Raya 



Pasrur 



•6061 -f -095 



•4943 + -114 



■4468 + -121 



•3919 + -128 



These correlations are from 3 to 6 times greater than their probable errors, and it 

 was clearly a matter of high interest and importance to discover whether these corre- 

 lations were really due to a definite relationship between the seasonal rainfalls or to 

 the fact that the years dealt with constituted a special group. As a matter of fact it 

 was found that, at any rate for the part of the Punjab dealt with, the 20 years 

 1887-8 — 1906-7 form a very special group, and can by no means be taken to represent 

 the result of a random sampling from among all years. 



For the purpose the data for 8 stations in the North- West of India were obtained 

 from the Meteorological Department. 1 



The results which are given below very soon showed how completely the years 

 1887-8 — 1906-7 failed to be representative of the secular rainfall. 





*« 



' r R 



! 



V R 



R P 



,r R 



V R 



N 



r 



Peshawar 



676 



4-l8 



61-79 



6-28 



3-50 



55-66 



45 



•0496 



Shahpur 



10-65 



4'45 



41-81 



3-40 



2-52 



73-9i 



53 



— • 15 14 ±-093 



Kohat 



11-49 



4-68 



40-75 



6-36 



4-00 



62-87 



45 



— '076 



Lahore 



15-63 



6-35 



40-64 



3-76 



2-23 



59-16 



45 



•1895 + -097 



Beawar 



18-26 



7-i4 



39-12 



1-09 



1-07 



97-87 



5i 



•0610 



A j mere 



I9-54 



6-75 



34-57 



i-66 



I'26 



76-09 



44 



—•1665 + -099 



Nagpur 



41-23 



io-ii 



24-52 



4-00 



2-78 



69-62 



53 



•1874 + -og 



Jubbulpore 



49-36 



I3-59 



27-54 



3-93 



3-37 



85-87 



■ 



62 



■1240 + -084 



1 I am indebted to the courtesy of Mr. J. H. Field, Imperial Meteorologist, for the selection of the specially long 



