229 



The instrument used at the central met. Institution is 

 a Robinson anem ometer with long arms and a great mo- 

 ment of inertness. The mean velocity resulting from this 

 instrument is in the eolumn marked B, the mean velocity 

 from our anemometers is in the eolumn marked A. The 

 height of the Robinson anemometer is 22 meters above the 

 ground and also 1.5 lower than the instruments used in 

 this research. The last two columns give the values of the 

 velocity in the beginning and in the end and determine also 

 the zero-point for every curve. 



The first result is that the variations are very nume- 

 rous and that the wind also is, as Mr Langley has found it 

 to be, a series of gusts. 



It will however be seen that the number of variations 

 is augmented ivith the mecm velocity of the tvind. From 

 this general law only one curve makes an exception, viz the 

 curve for the 18 of January, if we take the mean velocity = 8 

 received with the Robinson anemometer; but if v^e take the 

 mean velocity from our anemometer 1.4 2, even this curve 

 will confirm the law. It is very difficult to settle what has 

 been the cause of this great inequality, for the direction of 

 the wind was E and the churchtower could not be a direct 

 hindrance, but it is possible that it could work indirectly by 

 reflecting the wind, being within a 50 meters of distance. 



Taking in consideration the curves for the 14, 15, 22, 

 23, 25 of January and both of April 28 vve fmd the varia- 

 tions great and numerous compared with the other curves 

 and it is easy to see that we can put them nearly in the 

 same order as the mean velocities. Putting the last 5 curves in 

 the first order with the most numerous variations, we can take 

 curves for the 14, 15 and 21 in the second, the curves for the 

 1(3 and 20 in the thlrd and fmally for the 18 an 19 in the fourth. 



Å remarkable fact is that the mean velocity, calcu- 

 lated as an average of the observations with the mill-type 

 anemometers is always less than this quantity observed with 

 the Robinson-anemometer, which must depend on its greater 

 moment of inertness. 



