148 ON THE STORM EXPERIENCED THROUGHOUT THE UNITED STATES 



Having given all the information I have been able to collect respecting the 

 storm, and assigned, as far as practicable, its limits, I proceed to analyze more 

 minutely its phenomena. Let us take, then, first, the observations of the wind. 

 In making this comparison, certain general principles should be borne in mind. 

 One is, that the direction of the wind is liable to be sensibly influenced by the 

 inequalities of the surface of the earth. In a shallow stream of water, flowing 

 over a rocky bottom, the course of the particles of water is often very devious. 

 An effect similar in kind, though much greater in degree, must be expected 

 from an elastic fluid like air. This is strikingly exhibited in the narrow and 

 straight streets of cities, with high buildings on each side. The wind must 

 here blow in the direction of the streets, or not at all. So, also, a mountain 

 gorge; the straight bed of a river with high banks; the shore of a lake, or the 

 ocean; or a mountain ridge might be expected sensibly to influence the direc- 

 tion of the atmospheric current. To this cause is doubtless to be ascribed the 

 fact, that at stations very moderately removed from each other, the prevalent 

 winds often differ sensibly in direction. 



Again, the direction of the wind is exceedingly variable. I mean that its 

 direction varies not merely from day to day, and from hour to hour, but from 

 minute to minute, and from second to second. When the wind is at all fresh, 

 it is rare that it blows sensibly from the same direction for five successive se- 

 conds. I have been the more struck with this fact from having a vane attached 

 to a revolving shaft, to which is secured a graduated circle. The precise 

 amount of the oscillations is thus easily measured. I am accustomed, at each 

 observation, to note the extreme excursions during an interval of about five 

 minutes. This range of the vane may be termed a measure of the variableness 

 of the wind, and from an average of the year, does not differ much from fifty 

 degrees. In repeated instances it has amounted to ninety, and even more, de- 

 grees. Now, one who judges of the direction of the wind from a single glance 

 at a vane, is liable to mistake the mean direction by twenty-five or thirty de- 

 grees, because the vane, at the instant observed, may point thus much aside of 

 its mean position. To this add the probable error of judgment when angles are 

 estimated entirely by the eye, and it will not appear strange if two individuals^ 

 neither of them particularly careless in observing, should, at the same place, 

 and at the same hour, sometimes differ by forty-five degrees in their estimate 

 of the wind's direction. Indeed, at five-sixths of the stations from which ob- 



