246 



REPORT 1844. 



tions per minute ; and these quantities increased in the following propor- 

 tions : — 



Feet per second. Revolutions of sails. 



12'5 gave 7 "5 



20 13 



28 22 



Here the ratios are 2*3, 1'7, 1-6, 1'2. Hence the ratio of the velocity of 

 the wind to the revolutions of the sails is continually decreasing. But there 

 are other forces interfering with the ratios of the velocity of the wind to the 

 velocity of the fly vvhich render deductions from this machine somewhat dif- 

 licult, such as the resistance of the air to the motion of the fly and such like, 

 all of which require considerable attention in the construction of anemome- 

 ters on this principle. 



Taking however the observations we have recorded with this machine as 

 fair examples of the nature and operation of its working, and of the kind of 

 information it would put us in possession of, were it so constructed as to 

 meet all these questions, they are upon the whole very valuable, and we 

 shall see how they may be further applied and received in the way of ap- 

 proximation to a correct result. 



If we suppose in Table XI. that the ratios had continued to increase in 

 column /as the spaces described by the pencil diminished, and in the pro- 

 portion of the preceding numbers, we should have for a descent of the pen- 

 cil at the rate of one division of the scale of measure per hour, a corre- 

 sponding velocity of wind equal to about ten feet and a half per second ; and 

 for a rate of "5 of a division of the scale of measure per hour, a corre- 

 sponding velocity of wind equal to about seven feet per second. Taking 

 then 10-6 feet per second as the velocity of the wind corresponding to the 

 descent of the pencil at the rate of one division of the scale per hour, which 

 was about the proportion found by a mean of observations, we may take 

 the velocity of the wind corresponding to half a division of the scale per 

 hour as 7 feet per second. If then we apply this result to the resultants in 

 Plate XXXV. we have in dividing these resultants by 8760, the total number 

 of hours in the year, the mean rate of the pencil per hour expressed in terras 

 of the scale of measure, and hence the general results will be as follows: — 



Mean direction of wind 



Integral result 



Rate in divisions of scale per hour. 



Feet per second 



Miles per hour 



The great annual movement of the air therefore would by observations with 

 this instrument be N. 1° east, at the mean rate of 3-9 miles per hour. 



In Table XII. will be found the effective amounts of wind for each month, 

 as deduced from the mean of the three years by Tables I. II. III., and from 

 these we may determine the mean monthly direction and velocity of the cur- 

 rent, at least so far as these observations are to be relied on. The winds m 

 quantity and direction in this Table, when laid off as already .shown in Plate 

 XXXV., will be found to give the following results :— 



1 



