12 #. Loomis— Observations of the U. S. Signal Service. 
74 85™, Oct. 20, 1873, and corresponds to No. 20, table II of 
the present Article. At that date the center of low pressure 
was moving almost directly towards the center of the rain-area. 
Plate I accompanying the present Paper, shows the isobars for 
Oct. 21.1, 1873, and the dotted line shows the area over which 
the rain-fall for the preceding eight hours amounted to at least 
one-fifth of an inch. The rain-fall at Cleveland was °75 inch; 
Alpena 58 inch; Rochester ‘49 inch, and Saugeen 49 inch. It 
will be perceived that during these eight hours the storm center 
had been moving almost exactly towards the center of gravity 
of this rain-area. There was also a rain-area extending along 
the New England coast from Boston to Eastport which did not 
appear to exert any appreciable influence upon the progress of 
the storm. I have made a similar comparison for each date 
in table IT and find that in each case the storm center was 
moving nearly towards the center of the rain-area. In more 
than half of the cases the storm appeared to be moving exactly 
towards the center of the rain-area. In four of the cases the 
rain center appeared to be a little westward of the storm path, 
and in twelve cases it appeared to be a little eastward, but in 
only two or three cases did it deviate as much as 45° from the 
direction in which the storm center was moving. This coinci- 
dence seems to favor the conclusion that in a great storm the 
condensation of the aqueous vapor is an efficient cause which 
controls the movement of the winds. 
Table III shows results very different from table II. In six 
of these cases no rain was reported at any station within the area 
of low pressure during the preceding eight hours; in 23 of the 
cases the total rain-fall during the preceding eight hoursatall the | | 
stations within the low area was less than half an inch; and in 
only five of the cases did the total rain-fall in eight hours exceed 
one inch, and in each of these five cases there appears to have 
been 
a special reason for the greater rain-fall. In No. 11 the | 
rain center was about 600 miles northeast of the center of low 
pressure, and the succeeding observation shows that there was 
another low center in Canada which mainly controlled the 
movement of the winds throughout this rain-area. In No. 29 
the character of storm No. IV had already changed, and ‘the 2 
subsequent course of the storm was nearly east. In Nos. 34, 
35 and 36 the greater rain-fall is partly explained by the prox- _ 
imity of the low center to the Gulf of Mexico. In No. 39 the — 
rain-fall is given for a period of 24 hours. Thus wesee thatan — 
area of low pressure may be formed with very little rain, and ap- 
parently with none at all. Moreover in these cases the storm cen- 
ter did not generally follow the rain-area but moved away from it. 
Plate II accompanying this paper shows the isobars for Jan. 4, 
1877, at 4" 35™ P. M., indicating a low center near Ft. Sully. 
