192 THE CLIMATIC FACTOR AS ILLUSTRATED IN ARID AMERICA. 



during the middle of the winter, although this was compensated for toward spring. Farther 

 south, however, storms which had gone far equatorward brought to Texas and northern 

 Mexico more than the usual amount of rain, while Yucatan also had a comparative abun- 

 dance of showers and of "northers." The conditions were by no means remarkable, but 

 they serve to illustrate the fact that variations in the tracks of the storm bring with them 

 important results in the way of variations in rainfall and temperature. The winter of 

 1911-12 was characterized by a relatively pronounced and long-continuing area of high 

 pressure over the central part of North America, and therefore the storms for a season went 

 far to the south. In 1912-13 exactly the reverse took place. The continental high-pressure 

 area was poorly developed, the storms moved over northerly tracks, and the northern 

 United States was uncommonly warm. 



Having seen that the general course of the storm tracks varies from year to year, our 

 next question is whether, if longer periods than a year are considered, the average track 

 shows any variation. Dunwoody some years ago began the investigation of this matter, 

 and, so far as data were available, made a map, figure 51, showing the average number 

 of storm centers passing over each 5 degrees square of the northern hemisphere. His map 

 shows that the number of storms is greatest in the region of the Great Lakes of North 

 America, and is large throughout all of the northern United States, southern Canada, 

 northwestern Europe, and Japan. Professor C. J. Kullmer, of the University of Syracuse, 

 has made a study of this map and has pointed out that it affords some most interesting 

 suggestions as to the possible relation between high civihzation and chmatic instabihty. 

 In a later pubhcation I shall consider this matter at length, but for our present purpose the 

 important matter is that his work enables us to compare the storm tracks at two periods 

 separated by an interval of 12 years. Inasmuch as his original data have never been 

 published, and as they are important not only for our present purpose, but for other 

 climatological studies, I have persuaded him to prepare the contribution which follows. 



