CLIMATE OF HISTORIC TIMES 547 



stations in the United States where rainfall records go back as far as 

 1877. In some cases the records are imperfect and have been supple- 

 mented by those from neighboring stations. Elsewhere two or three sta- 

 tions which lie close together have been combined. The entire number 

 of stations is 183. The 33 years from 1877 to 1909 contain three sun- 

 spot cycles. The three years of minimum and of maximum spots in each 

 of the three cycles are the same that have been used in Kullmer's work 

 on storm tracks. They are as follows : 



Years of 

 inimum spots. 



1877 



1878 



1879 



Sun-spot 

 numbers, 



12.3 



3.4 



6.0 



Years of 

 maximum spots. 



1882 ■.., 



1883 



1884 



1893 



1894 



1895 



1905 



1906 



1907 



Sun-spot 

 numbers. 



i59.7 



........ 63.7 



63 . 5 



1888 



1889 



6.8 



, 6.3 



73.0 



84 9 



1890 



1900 



1901 



7.1 



9.5 



2.7 



78.0 



58.6 



52.8 



1902 



4.7 



64.5 



Average for nine years of minimum sun-spots, 6.5. 

 Average for nine years of maximum sun-spots, 65.5. 



For each station the average rainfall for the nine years of minimum and 

 for the nine years of maximum has been computed, and the departure of 

 this from the mean for the entire period covered by records has been ob- 

 tained. Then the difference between the departure when sun-spots were 

 few and when sun-spots were many has been ascertained. This has been 

 reduced to percentages of the mean rainfall and has been inserted on the 

 map. This is better than the use of the actual differences, since a change 

 of 2 inches in a place where the average rainfall is 10 inches is much 

 more important than a change of 2 inches where the rainfall is 50 inches. 

 By the use of percentages these figures would appear as 20 per cent in 

 one case and 4 per cent in the other. The shading of figure 17 has the 

 same significance as in figure 16. The two maps show certain differ- 

 ences, partly because the map of storms is based on a slightly larger 

 number of years than the other ; but much more because the rainfall map 

 includes not only the precipitation which accompanies ordinary cyclonic 

 storms, but that which comes in thunder-storms and other local sliowers. 

 In spite of minor differences the two maps show the same main fea- 

 tures. They indicate the general conditions which would be expected if 

 the conditions which now prevail at times of many sun-spots were to be- 

 come permanent. The most prominent feature is a large black area in 



