the Filar Anemometer. 89 



hardly detect a smaller interval than a half tone, so that 

 speeds are not definable closer than about 6 per cent. This, 

 however, in view of the excessively variable character of a 

 gust of wind, is no serious disadvantage. 



(B) Copper wire, 0*126 cm. diameter, 190 cm. long. — ■ 

 I heard all the chromatic intervals from c' to c" to d" to ef'\ 

 which corresponded to the limiting speed used. The frequency 

 is thus from n = 261 to above w = 1305, indefinitely ; and the 

 speeds are from 1*7 met./sec. to above 8*5 met./sec, or from 

 3*8 miles per hour to above 19 miles per hour. Comparisons 

 with Strouhal's formula were made for g' 9 V ', d ,r , g ff y c f// , 

 checking the pitch in the two cases. All the notes were clear 

 and strong. StrouhaFs lower limit was here about n= 600, 

 so that the microphone again detects inaudibly low notes. 



(C) Copper wire, 0*240 cm. diameter, 190 cm. long. — Faint 

 cello-like notes were heard, ranging from c to c f to c" . The 

 pitch was checked for g, c', and e f . The limit of speeds used 

 corresponded to c" , above which the notes would, doubtless, 

 have come out more ringing. The lower limit of frequency 

 was thus actually n = 130, whereas Strouhal failed to hear 

 tones below n = 500 about. I did not enter into this case at 

 much length, because of the confusion produced by overtones. 

 Very regular rotation was needed to bring out the low 

 notes. 



(D) Brass wire, 0*020 cm. diameter, 190 cm. long. — The 

 behaviour of this wire was very peculiar. No sounds were 

 obtained except for tensely drawn wire, and I then heard the 

 exceedingly shrill notes b 1Y , c v , c$ y , d Y , dP, e r , f v , /#v, ranging 

 therefore in frequency from ??. = 3915 to n = 6000 nearly. A 

 comparison of pitch with Strouhal's data was made for b w and 

 d Y . Curiously enough, Strouhal in the case of these thin wires 

 actually heard a whole range of low sounds from less than 

 n = 2000 to about n — 3000 } and none above. The fact that 

 I had to draw the wire tensely to affect the microphonic plates 

 may be adduced in explanation. Apart from the range there 

 is no anomaly. It is probable that these wires are not 

 massive enough to actuate the microphone. 



I have stated that my reasons for not going to higher 

 speeds than about 21 miles per hour were merely the imper- 

 fections of my whirling arm. The last example shows forcibly 

 enough that excessively high pitches are by no means beyond 

 the scope of the microphone. If the note n = 6000 had been 

 heard from the first ware (^ = 0*072 cm.), it would actually 

 mean a speed of about 22 met./sec. or 49 miles per hour. 

 This is fairly a gale. If heard from the second wire 

 (d = 0\126 cm.), it would mean a speed of 87 miles per hour 



