[M’LEoD & BARNES] NOTES ON THE DIFFERENCE OF TEMPERATURE 7 
for the mountain difference curve is shown and our conclusions are 
thereby impaired. They can only be regarded as a suggestion until we 
ean obtain continuous records of a much greater period. It may be 
observed, however, that both periods of large temperature differences on 
the mountain are accompanied by a predominance of cold weather con- 
ditions. Thus, from July, 1903, to March, 1905, every month was below 
the average with the exception of three, and this occurs along with a 
period of very large temperature differences. 
Accompanying and following the small temperature differences from 
February to June, 1905, we find months very close to the average in 
temperature conditions. From December, 1905, to August, 1906, we 
find a series of small differences, including some less than had been 
previously observed. We find that over the same period of time only 
two months out of twelve are below the average and some very warm 
months were experienced. 
Since last September we have been observing larger temperature 
differences again, and the Observatory records show colder conditions 
prevailing. In order to show a possible relation between the mountain 
records and the Observatory temperatures we have plotted in Figure 2 
a succession of six-month averages. The values given for each month 
represent an average for three months on either side. In one case over 
the short period between the two breaks in the records we have pro- 
duced the difference curve sufficiently to make up the six months average. 
This can be done without much error. It will be seen that there is 
an indication of some intimate relation between the temperature of the 
high level and the character of the season which can only be elucidated 
by further records extending over a longer period of time. 
