AND DIRECTION OF THE WIND AT SEA. 459 



which any particulai- wind is made up ; and although, in a mere mechanical poin^ 

 of view, a wheel of great weight would tend to equalize and mean all the cur- 

 rents of different intensity, still it can only do so with a certain amount of loss, 

 and with the total omission of all very light impulses ; and the only way accu- 

 rately to sum up all those separate little quantities, is to employ an instrument 

 which shall be as sensible as a feather, and take full and immediate account of 

 the slightest motion of the atmosphere. 



After trial of floats 2, 3, 3^, 4, and 6, inches in diameter, the 4-inch ones were 

 considered as being the best ; and the hemispherical shape was also preferred, 

 as giving the greatest per centage of velocity with the least weight of material 

 and smallest side resistance, as well as offering the shape, of all others, of the 

 easiest and truest execution, and best understood everywhere. 



Various experiments were tried, of making the floats more or less conical, in 

 order to diminish the pressure of the wind on their backs ; but though that point 

 was most eminently obtained, still the advantage was outweighed by the neces- 

 sary increase of weight accompanying the increase of surface, the greater side 

 resistance to the wind, and the diminished pressure on the concave side. 



In the month of January this year, I had the opportunity of trying the value 

 of the revolutions of this anemometer, in company with Captain Cockbubn. The 

 instrument was mounted on the top of a cab, clear of the driver's head, and 

 driven at a pretty uniform speed of above seven miles an hour, first forwards 

 and then back, on two miles of the London Road ; the object being to measure 

 the artificial wind produced by the motion of the vehicle, which would of course 

 be equal to a natural wind blowing with the same velocity in the contrary 

 direction. The first day there was a rather strong breeze, which would have 

 completely vitiated the experiments, but that, as it was blowing almost exactly 

 in the direction of that part of the road which was traversed, we expected to be 

 able to eliminate its effects by taking a mean of the numbers given in going and 

 returning. 



When going, having the wind with us, the instrument, which measured then 

 only the difference between the velocities of the wind and the cab, made 209 

 revolutions in one mile ; but in returning, measuring the sum, it gave 921 revolu- 

 tions in the same distance. The mean of these, or 565, when multiplied by 3'] 415 

 feet, or the space described by the centre of the float in one revolution, gives 

 a velocity not exactly J, but ~^ of that of the wnd. 



The second day was all but perfectly calm ; it was at the commencement of 

 the long-continued frosty weather ; and a proof of the general stillness of the air 

 was offered in the dense, unnatural manner in which the smoke was accumulated 

 and remained suspended over the city. In going out, 558 revolutions were made in 

 one mile ; and in coming back, 551 . The mean of these, or 555, gives a velocity 



VOL. XVI. PART IV. 6 b 



V 



