E. Loomis— Observations of the U. S. Signal Service. 3 
while the fluctuations of pressure attending the progress of the 
storms of the middle latitudes are but little felt. Moreover it 
will be noticed that these great fluctuations of temperature 
occur more frequently in summer than in winter. Hence it 
may be inferred that the principal cause of these fluctuations 
is independent of the progress of storms. This will appear if 
we examine the cases at Wickenburg, which stands at the head 
of the list. The following are the cases in which the diurnal 
change of temperature at Wickenburg exceeded 60°: 
ax. Min. Diff, ax. Min. Diff. 
1877, » duly. 37,106" 49° 64° |) 189%. y Ang. 2, 109°. 39°. 92 
28) AIO y | 40. RO 2, 107 43 64 
29, 106 44 62 $. g108 44 ltt 
30, 106 43 63 13, 132. 46 - 86 
31, 106 30 6 14, 113 50 63 
Wickenburg is situated in a desert region where the annual 
rain-fall is only five inches; where, during the day, the sand 
becomes intensely heated by the sun’s rays; and where, on 
account of the dryness of the atmosphere the loss of heat b 
radiation at night must be about as great as at any other point 
of the earth’s surface. A similar condition of things exists at 
all the stations named in the table south of latitude 35°. 
At the stations further north, the rain-fall is generally small, 
and the air is ordinarily very dry. This remark will apply to 
all those stations at which the annual number of cases of 
these large fluctuations exceeds twelve. At some of the re- 
maining stations, the fluctuations of temperature resulting from 
the progress of great storms becomes appreciable, viz: at 
“<r Spee York Factory, Rockliffe, Yankton, Pembina, 
Bismark, Fort Garry, Geneva, Chatham and Parry Sound. 
Six of these stations are situated in that region where, after the 
passage of a center of low pressure with its high temperature, 
the cold winds from the north rush down with a violence un- 
known in any other part of the continent. It appears then 
G 
wich. The annual reports of the Signal Service show the 
aximum and minimum temperature for each day of the year 
for each of the stations of observation. I determined the 
average of the maxima for each month, and also the average of 
the minima, and called the difference the mean diurnal fluctua- 
tion for that month. When there were observations for several 
