Mr Forbes's description of a New Anemometer. 37 



not represented in the figure, to shelter one side from the wind 

 and expose the other to its force. On the axis of the sail is 

 the pinion «, working into the wheel 6, which turns the endless 

 screw c. By these two transfers of power any required di- 

 minution of velocity may be produced. The box B contains 

 a number of spherules to be dropt at short intervals by means 

 of a mechanical contrivance, and in the course of their fall are 

 exposed to the deflecting force of the wind, which deflection is 

 measured by means of the circular rings, d, e^ff, e\ d, in 

 the board C D, of which a horizontal plan is given in Fig. 8, 

 showing by the distribution of the cavities that both the 

 direction and velocity of the wind will be measured at once, 

 being divided into eight, or as many points of the compass as 

 may be desired, and into three or more measures of deflection. 

 It is thus apparent, that, in the estimation of force, the results 

 are wholly independent of any mechanical friction, since all 

 4he wheel- work connected with it is merely employed for drop- 

 ping the balls ; which is thus contrived : The bottom of the 

 box B is composed of several parts, a section of which is 

 given in Fig. 4 : the centre part Q is the real bottom, and is at- 

 tached to the sides. It has the thickness of one of the balls, 

 and has a cylindrical perforation q, of the same diameter, as 

 represented also in Fig. 5, which is a horizontal section of the 

 same. United by a common axis g (Fig. 4,) which passes freely 

 through the centre of the bottom Q, are the two plates P and R, 

 the horizontal surfaces of which are represented at Figs. 6 

 and 7, and into the edge of the former works the endless screw 

 c (Fig. ^,) by which the plates P and R are slowly turned 

 round, being on a common axis, so fixed that while the aper- 

 ture r covers the cylindrical hole in the plate Q, it is closed be- 

 low by the stop p in the plate P, and the reverse, so that it is 

 obvious, that, as soon as the cavity q is insulated above, the 

 ball which was received through the opening r will drop, 

 which will be repeated once every turn, the recurrence of 

 which may be varied at pleasure by the number of teeth in the 

 wheel-work. 



We must, however, observe that the number of pellets drop- 

 ped into the circular rings in Fig. S, even should these rings be 

 calculated for spaces corresponding to equal increments of 

 velocity, will not be proportional to the times during which 



