Section III., 1907. [3] Frans. R. 8. C. 
1.—Further Notes on the Difference of Temperature between Mount 
Royal and the McGill College Observatory.—Local Temperature 
Forecasting. 
ah Proressors C. H. McLeop, Ma.E., and H. T. Barnes, D.Sc. 
McGill University, Non call 
(Read May 15th, 1907.) 
Since we had the honour to present our second paper’ on this 
subject before Section III last year, we have obtained a series of 
remarkably perfect records which has extended almost unbroken through- 
out the year. This includes traces for last summer, which are interest- 
ing, owing to the rather unusual heat experienced in this locality. 
Results for the two previous summers have unfortunately not been 
obtained owing to reasons set forth in our second communication, but 
an analysis of the past summer’s records has shown that the differences 
between the mountain-top and Observatory were small. A study of the 
complete records for the year has confirmed our views of the value of 
these records for local temperature forecasting, as explained and illus- 
trated in our previous papers. 
We have observed that temperature changes affect the high level 
thermometer first, and that the time interval between the first indica- 
tion of a change on the mountain instrument, and the change at the 
lower level varies from four to twenty-four hours. So accustomed are 
we now to follow the changes that it has become a matter of common 
observation with us. 
This idea of forecasting from temperature readings at a high alti- 
tude is not new. We believe, however, that the feature which makes 
the method successful is new, and that is, as we have frequently pointed 
out, the differential reading must be obtained so that the relative changes 
may be seen at a glance without the necessity of subsequent plotting 
Professor Cleveland Abbe, of the Weather Bureau in Washington, 
maintained, for seventeen years a station at the top of Mount Washing- 
ton, from which, he states, he was frequently able to forecast weather 
changes by means of observations telegraphed daily from the summit. 
Eventually, however, the station was discontinued, as the cost was found 
to be not fully compensated by the value of the work. After carefully 
studying the records obtained in this work for the months of January, 
February and March for the years 1871 to 1886, and published in the 
Monthly Weather Review for 1891, Professor Hazen states in regard 


1 Trans. Roy. Soc. Can., Vol. XII, p. 141, 1906. 
