50 



TEMPERATUEE. 



Removal of the Cold Surface Layers hy a Ligltl Local Wind—Septemher 22 to 25 {fujvre 

 18). —The thermograph trace shows many interesting examples of a sudden pufE of wind causing 



12 ^15 2ZnD.2Ci 



25 RD. 



i 24th. lb 



Z5ffl.SEPTEMBER 1911 



Fig. 18. Wind and Temperature. 



an instantaneous rise of temperatui-e of many degrees. These sudden rises prove conclusively 

 that there must be warm air very near to the cold air which is so suddenly removed. 

 There is no reason to believe that there was warm air anywhere near in a horizontal direc- 

 tion, hence the warm air could only have been part of a warm layer lying above the cold 

 surface layer. The period September 22 to 25 contains several good examples, and the 

 thermograph trace for this period is reproduced here. A detailed discussion is unnecessary, 

 the relationship between the wind and the temperature is obvious. 



Cold Blizzard during the Summer— February 14, 15, and 16, 1912 (figure 19).— In this 

 diagram also the relationship between the wind and the temperature is obvious. But we have 



Fig. 19. Wind and Temperature. 



a different effect from those previously considered. Here we see that the direction of the 

 wind plays the chief part in deciding the temperature. The two short periods of noitheriy 



