ON THE ANEMOMETERS OF PLYMOUTH. 



249 





w 



tlie annexed figure, and its magni- 

 tude = \/(3134.)^+(14'23)2 = 3442 

 nearly ; this divided by 8760, the 

 number of hours in the year, gives 

 a rate of '4 per hour in divisions of 

 the scale of measure for the descent 

 of pencil, which by previous de- 

 ductions from Table XI. would in 

 taking \ a division of the'scale of 

 measure = to 7 feet per second, 

 give a velocity of 5*6 feet per se- 

 cond, or 3'8 miles per hour, a simi- 

 lar result to that already arrived at, 

 with the exception of a little devia- 

 tion in the direction. 



Such are the principal results of 

 the working of this instrument du- 

 ring the years 1841, 1842, 1843; 

 and could we feel assured that it 

 had worked with perfect accuracy, 

 we should doubtless have arrived 

 at very important results. This is 

 however unfortunately not the case, 

 and therefore these results can only 

 be taken in the way of rough approximations. 



The integral effects, in fact, have been deduced on the supposition 

 that the revolutions of the fly are proportional to the velocity of the wind, 

 whereas we see by Table XI. that they only approach this ratio in certain 

 cases, and differ from it considerably in others. The relative effects also of 

 the wind on the fly and vane are unfavourable to a very correct direction, in 

 consequence of a tendency in the top to be thrown occasionally a little ob- 

 lique to the course of the current, so that it is possible that the direction is 

 not always to be depended on. Professor Airy, in his experience of this ane- 

 mometer at Greenwich, thinks it cannot be depended on for direction ; he 

 says it turns very heavily in azimuth, an objection however to which the in- 

 strunient at Plymouth is not liable since it was last set up. 



With all these difficulties, however, I am encouraged to hope that the ex- 

 amples just given of its operation for three years, together with the deduc- 

 tions we have made by a careful analysis of the register, may not be alto- 

 gether without claims to attention ; beside that they hold out inducements 

 to ingenious mechanics and meteorologists to improve its construction, and 

 extend the principle to more perfect machines, from which certainly the 

 most important consequences would result. To render an anemometer of 

 this kind perfect, experience has taught us, that it should be a permanently 

 fixed machine, and set up in a commodious and convenient form ; it should 

 consist of several separate parts, all of them independent of each other, viz. 

 a vane for registering direction only, unembarrassed by any other mechani- 

 cal detail ; the most simple and effective form of vane is that of a flying-fish 

 with extended fins, balanced at about one-third from the head. Secondly, a 

 powerful fly or horizontal windmill, so contrived as to revolve in one direc- 

 tion only and register the integral of the wind. The power of this fly should 

 be in such a ratio to the friction of the machine as will admit of the latter 

 being thrown out of the calculation, so that it could always revolve with a ra- 

 pidity proportional to thewinds'velocity, and with which it should heabsolutely 



