MADE WITH A NEW ANEMOMETER. 809 



this line, a third; and so on. In this manner I obtain a continuous 

 line, which represents the course of the wind as long as it is con- 

 tinued. Such lines were drawn for February and March 1837; those 

 for March are exhibited in Plate VII, which represents the curves for 

 March, drawn for the stations, at the Society, at the Observatory and at 

 Edinburgh. In all the cases the observations experienced interruptions, 

 which make it difficult to draw any general conclusions from them. But 

 we may remark that in February the wind blew almost constantly from 

 a more westerly point at the Observatory than at the Society. It is 

 not difficult to conceive this result to be occasioned by the peculiar 

 circumstances of the Anemometer at the Observatory: but it is also 

 possible that it may be a general fact that such differences obtain at 

 neighbouring places, in consequence of the direction of valleys, &c. 

 Further observation alone can clear up this and similar points. 



It has been deemed an important point by Meteorologists to obtain 

 the mean direction of the wind at a given place for a given time, for 

 instance, a year. Kiimtz in his Meteorologie, Vol. ii. p. 218, has 

 collected several results of this kind. But in these researches the force 

 of the wind has entirely been left out of the account, and each wind 

 was reckoned according to the number of days which it blew. It is 

 clear that such a procedure is entirely fallacious; for the high wind 

 of one day may be greater, with regard to every possible effect, than 

 the gentle breezes of a week. The mean annual direction is probably 

 constant at each place within certain limits: and the mean directions 

 at different places are perhaps connected by certain general j-elations, 

 depending upon the quantity of fluid transferred, and upon other atmo- 

 spherical conditions, which may hereafter be found to be important 

 elements of meteorological speculation. But it is not at all likely that 

 this will, hold if the mean direction be taken without reference to the 

 strength of the wind; and no mode of measurement can be good for 

 this purpose which does not give the whole quantity of the aerial cur- 

 rent, depending both upon velocity and upon time. 



The Anemometer here referred to is, as I have said, the only one, 

 so far as I know, which has been constructed with the view of thus 



rrS 



