1877.] the Rainfall with the Sun-spot Period. 257 



Table Till. 



Cycles of 



Intervals of cycle. — Two hours. 



oue day. 



0. 



2. 



4. 



6. 



8. 



10. 



12. 



14. 



16. 



18. 









-38 



- 5 



+53 



+83 



+45 



+ 10 



+32 



+52 



+78 



+54 



- 6 



-59 







2nd „ 



-87 



-58 



-21 



+13 



-11 



-12 



- 9 



+ 1 



+ 16 



+ 5 



-32 



~2 



3rd „ 



-72 



-43 



+ 2 



+19 



- 2 



-16 



-13 



+21 



+57 



+42 



o 



— 40 j 









4th , 



-63 



-28 



+18 



+47 



+22 



- 7 



+ 4 



+20 



+64 



+37 



-13 



-46| 







5th ., 



-52 



-29 



+ 9 



+24 



+ 7 



-21 



- 3 



+35 



+56 



+35 



-13 



-74 1 



Mean difference ] 

 from mean of L 

 60 obs | 



-62 



-33 



+12 



+37 



+12 



- 9 



+ 2 



+26 



+54 



+35 



-13 



-59 



In this case it will be found that the mean deviation (disregarding 

 sign) of the whole of the observations from the arithmetical mean of 

 the whole (which is here zero) will be 30-3. 



The differences between the mean values, given in the last line of 

 Table Till., for the two-hourly periods and the individual observations 

 are as follows :— 



Table IX. 



Cycles of 



Intervals of two hours. 



one day. 



0. 



2. 



4. 



6. 



8. 



10. 



12. 



14. 



16. 



18. 



20. 



22. 





+24 

 -25 



+28 

 -25 



+41 

 -33 



+46 

 -24 



+33 

 -23 



+19 

 - 3 



+30 

 -11 



+26 

 -25 



+24 

 -38 



+19 



-30 



+ 7 







2nd „ 



-19 



-13 



3rd „ 





-10 



-10 



-18 



-14 



- 7 



-15 



- 5 



+ 3 



+ 7 



+ 11 



+ 14 



4th „ 





+ 5 

 + 4 



+ 6 



+10 

 -13 



+ 10 

 - 5 



+ 2 



+ 2 



- 6 



+10 



+ 2 



+ 2 







+13 



-15 



5th „ 



+10 



- 3 



-12 



— 5 



+ 9 











_ 



Mean difference, ] 

 disregarding L 

 the signs J 



14 



14 



19 



22 



17 



9 



13 



14 



15 



12 



7 



11 



Mean of mean differences 14. 



Here a true periodicity existing, the mean of the differences, or, as I 

 before called it, the cyclical deviation, is reduced to 14. Moreover the 

 several differences for the separate observations for the successive hours 

 are well below the mean variation 30, and none of the separate results 

 is so high as that figure. 



If we had such a series of sixty observations as is contained in 

 VOL. xxvi. t 



