74 THE ROYAL SOCIETY OF CANADA 



In our communication of last year we gave the differences from 

 November, 1912, to April, 1913. 



The values so given for the two months of 1912 not included in 

 the above table are : 



1912 Mountain Seasonal 



November -2-125 +2-17 



December -1-367 +5-35 



It is our aim to show that relations exist between the mountain 

 differences and the seasonal variations. Our previous series of 

 records showed us that some such relation appeared to exist, but 

 unfortunately the break in the cable destroyed the continuity of the 

 observations. At the present time our records have not been suffi- 

 ciently continued to give us definite results. It is interesting to plot 

 the observation of the table when it may be seen that in general a 

 small mountain difference is related to a higher seasonal average. 

 It is quite remarkable to observe how characteristic are the various 

 records for each day of the kind of weather condition. As we have 

 pointed out before, an approaching change to warmer conditions is 

 shown usually by an inversion of temperature, several hours previous 

 to the temperature rise at the lower station. The same holds in general 

 for the approach of colder conditions. 



Dull, rainy or snowy weather gives us uniform difference curves, 

 which show very small and rapid fluctuations. 



We wish to thank Mr. A. A. Scott, M.Sc, for help in computing 

 the monthly averages 



