370 



SCIENCE PROGRESS 



These difficulties are to be got over by damping, which delays 

 the effect of the wind but excludes the effect of momentum. 

 We are not yet at the end of our experiments on this point 

 but I give here some of the results obtained. I may say at 

 once that irregularities are as pronounced as the most pessi- 

 mistic investigator could anticipate. Fig. 19 may be appealed 

 to with confidence in justification of this statement. 



We have divided all the diagrams obtained from the vector 

 apparatus into four types : 



Type 1. When the wind is steady and no change of im- 

 portance takes place either in velocity or direction. 



Type II. When the changes in velocity and in direction 



t>>» n j T,f,t m , 



Ht'd 3» ft 



obeve ord. 



T,f,t I 



r,(,i e 



s ■ -t.n y •'">'*'" F * }m'tth 



Fig. 20. — Four types of vector diagrams of wind changes during intervals 

 from 15 sec. to 2 min. 30 sec. 



are independent of each other. This is a common type when 

 the apparatus is undamped and gives rise to diagrams of 

 rectangular shape. 



Type III. When the changes of direction are large and 

 rapid but accompanied by small changes in velocity giving rise 

 to diagrams with sharp Vs. 



Type IV. When both velocity and direction are subject to 

 damping and a fairly smooth oval diagram is formed. 



These types are represented in fig. 20. 



We may pursue this investigation in co-operation with air- 

 men who can estimate the difficulties of manoeuvring in different 

 conditions recorded by an anemometer or other special apparatus, 

 until we can assign a limit of gustiness in which it is not 

 safe for an air pilot to take his craft. We have seen that the 



