Observations of the United States Signal Service. 19 
e see that in each of these cases there was an area of high 
inwacter on the north, and this P ressure was generally suffi- 
cient to urge the wind with considerable force towards Indian- 
ola. We also see that in each case the temperature of the air 
at Indianola was below that of the water on the south side, and 
generally very much below it; and moreover the air over the 
Gulf contained an abundance of va por. This cause alone 
would be sufficient to produce a fresh breeze from the north at 
Indianola. 
The reason that violent winds are more common at Indianola 
than at stations further east on the Gulf of Mexico may be the 
greater dryness of the air near the Rocky Mountains. The 
annual rain-fall decreases from the Mississippi River to the 
ocky Mountains. and the winds must be dryer near Hi moun- 
tains than they are near the cee eed River. We find a cor- 
responding increase in the mean velocity of the winds as we 
advance westward from Mobile. At Mobile the mean veloc- 
ity of the wind is 6-0 miles per hour; at New Orleans it is 73 
ose at Galveston it is 9:3 miles ; and at Indianola it is 13°7 
mies, 
At several of the stations, one of the most noticeable pecu- 
liarities of these violent winds is the great oe of 
w 
sorthernst, ete. This seems to indicate that the diate of the 
wind is somewhat a by the configuration of the surface 
of the surrounding country. 
A high velocity « of the wind is not invariably associated with 
alow barometer. In fifty-seven cases (out of 250) the yey so 
was above thirty inches; in sixteen cases it was a 0°25 
inches ; and in three cases it was as high as 30°5 bashed; The 
following are the extreme cases: 
Wind. 
No. Date. Stations. Barom. Direc.| Fore. Temp. 
1 (1873, —~ 7 Breckenridge | 30°75 | N. | 40 | —17° 
2 2) Quebec 30°56 |N.E.| 41 1 
3 1872, Dec. ae Breckenridge | 30°50 N.W.! 40 | —23 
Tn each of these cases there was an area of low barometer 
within the limits of the United States, and also an area of hig 
barometer ; but in cases No. 1 and No. 3 the pressure at. Breck: 
ing 
vation, which seems to indicate that at the former date the 
center of high pressure was northwest of Breckenridge. | 
