FROM SELECTED DAYS DURING THE SEVEN YEARS 1898 TO 1904. 
333 
Table XVIII.—Barometric Pressure, Diurnal Inequality, Amplitudes, and Phase 
Angles (Unit for Amplitudes, O’OOOl inch). 
Cl. 
ai. 
C 2 . 
#2. 
Os- 
<* 3 . 
c 4 . 
a 4 . 
January. 
5 
1 
o 
199 
/ 
28 
99 
3 
o 
153 
/ 
59 
49 
3 
o 
345 
r 
38 
22 
5 
o 
210 
55 
February 
8 
6 
44 
37 
104 
8 
150 
12 
31 
4 
342 
53 
9 
8 
91 
42 
March . . 
40 
0 
28 
18 
115 
8 
150 
13 
18 
1 
332 
33 
13 
4 
24 
25 
April. . . 
82 
2 
30 
7 
117 
9 
151 
26 
9 
4 
170 
42 
11 
7 
- 8 
46 
May . . . 
108 
9 
29 
41 
103 
3 
148 
39 
23 
5 
160 
38 
5 
4 
-44 
46 
June . . . 
93 
0 
16 
13 
91 
7 
144 
6 
29 
5 
160 
44 
1 
1 
- 62 
1 
July . . . 
67 
0 
18 
12 
94 
0 
140 
20 
31 
6 
151 
4 
2 
6 
-51 
31 
August . 
94 
8 
6 
45 
92 
3 
140 
43 
13 
7 
156 
16 
15 
6 
-50 
18 
September . 
72 
5 
5 
19 
122 
3 
151 
32 
2 
3 
352 
45 
10 
7 
-20 
42 
October . . 
13 
9 
115 
23 
113 
6 
158 
28 
28 
6 
357 
18 
5 
9 
17 
16 
November . 
13 
6 
317 
32 
99 
7 
158 
10 
38 
8 
354 
51 
10 
1 
189 
1 
December . 
18 
1 
83 
35 
98 
2 
151 
57 
46 
3 
351 
34 
22 
7 
216 
1 
Year . . . 
47 
1 
21 
12 
103 
9 
150 
15 
9 
1 
359 
0 
2 
9 
282 
27 
Winter . 
4 
7 
44 
29 
100 
3 
153 
32 
41 
3 
348 
55 
12 
7 
199 
14 
Equinox. 
47 
5 
24 
44 
117 
9 
-j 
152 
52 
9 
6 
347 
17 
9 
9 
2 
30 
Summer. 
89 
9 
18 
8 
95 
2 
143 
35 
24 
5 
156 
58 
6 
1 
310 
15 
Comparing Tables XVIII. and V., we see that in both the amplitude of the 
24-hour term varies much with the season; but whereas in the case of P it is 
conspicuously largest in winter, that is precisely the season when it is least in B. 
Table XVIII. shows apparently a secondary minimum for c x in summer, but the 
smallest value in summer is 13 times the value in January. The fundamental nature 
of the difference between the annual variations of c x in B and P can in fact be seen at 
a glance on comparing the heavier and lighter full line curves in fig. 3, which repre¬ 
sent the monthly values expressed as fractions of the arithmetic mean. 
The phase angles a 4 in Table XVIII. show erratic variations in the winter months, 
which are very probably fictitious. Judging by the seasonal values, a x is larger in 
winter than in summer, but the difference is small compared to the corresponding- 
difference in Table V. In all the seasons a x is much less for B than for P, in other 
words, the hour of maximum for P is much the earlier. 
§ 31. In the case of the 12-hour term the differences between the annual variations 
in the amplitudes for B and P are comparatively small. These variations are 
illustrated by the dotted curves in fig. 3. 
The phase angles a 2 in Table XVIII. show fairly similar characteristics to those 
exhibited in Table V. In both cases the seasonal variation is comparatively small. 
The P phase angle is invariable the larger, i.e ., the maxima in the 12-hour curve are 
again earlier for P than for B; the difference in time is however on the average only 
slightly over an hour, though greater in summer than in winter. 
The 8-hour term in B is unusually large. In winter c 3 is in fact much in excess of c x . 
