CAUSES OF THE GLACIAL PERIOD 571 



years of maxima since 1877 the average may have been 21. If we sup- 

 pose that the conditions which accompany sun-spots are magnified six- 

 fokl, the number of storms would be 26, and this is the figure used in 

 plotting the lower curves. 



If we examine the various groups of curves, we see that they all agree 

 in certain important features. Beginning with the curves for longitude 

 60° to 65°, we see that at times of minimum spots, as represented by the 

 upper curve, there is a regular increase of storms as one approaches lati- 

 tude 45° to 47°. The place where the storms are most numerous is indi- 

 cated by the letters B + S. In the next curve, the average for all years, 

 the high point, B + S becomes flattened. In the third curve, represent- 

 ing conditions as they now exist at times of many sun-spots, the region 

 of greatest storminess has moved two or three degrees north of its loca- 

 tion in the minimum curve, while on the other side, southward, there has 

 developed a somewhat prominent angle. Finally, in the lower curve we 

 see that if the conditions which accompany many sun-spots at the present 

 time should be exaggerated sixfold, tliere would be a sharply defined 

 maximum of storminess between latitudes 47° 30' and 50° and another 

 maximum of less proportions between latitude 42° 30' and 45°. Turning 

 to the next set of curves for longitude 70° to 75°, we see the same con- 

 ditions repeated. The only essential difference is that here we have a 

 liint of a double maximum even in the curves for the actual conditions as 

 they exist today. The same is true of the curves for longitude 80° to 85° 

 and 90° to 95°. The latter presents some slight irregularities which 

 seem to be related to the curious little tongue which projects southward 

 from the main area of northern increase of storminess in figure 7 and 

 elsewhere. In the groups of curves for longitudes 100° to 105° and 110° 

 to 115°, especially the former, the two belts of storms stand out dis- 

 tinctly even in the upper curves. Finally, in the last group the condi- 

 tions of the first are quite closely repeated. 



Summing up the whole matter, we see that except for the extreme 

 borders of the continent the curves of present storminess all show not 

 only the main crest of what we have termed the boreal storm belt, which 

 is indicated by the letter B, but also some faint indications of the smaller 

 subtropical belt, which is indicated by the letter S. If we turn to the 

 lower line in each group, we see that in all cases, without exception, there 

 are two distinct crests. Invariably also there is a tendency for these two 

 crests to separate from one another as the conditions which accompany 

 maximum sun-spots become intensified. The boreal crest generally 

 moves only 2 or 3 degrees from its present position, while the subtropical 



