NOTES ON FLYING-MACHINES. 23 



various parts are tin ; the braces and ties are piano wire. It is 

 thought that a lighter structure could hardly be made, though 

 the distribution of the material might be improved. 



The wings were first made 29 inches wide, but it was found 

 they could only be flapped 16 times a minute, their length being 

 limited by the height of the room where the experiments were 

 carried on ; it is thought better results could be obtained with 

 longer and narrower wings. The length of each wing is 7' 3" 

 from the socket to the tip, and the length of the membrane is 4' 8", 

 the area of each is a little over 8^ square-feet ; this area, or the 

 number of revolutions must be largely increased, if the machine is 

 allowed to run along the ground. Thin tissue paper is undoubtedly 

 the best material for the wings, it is impossible to burst any of 

 the compartments by flapping. 



To measure the thrust of the wings a post was provided with a 

 vertical lever, the machine pushed against the short arm of the 

 lever and the long arm of the lever was hooked to a spring balance 

 and also carried a pencil that recorded the variations of pressure 

 on a sheet of paper pinned on a small table; tolerably uniform 

 motion was given to the small table by a clock and fan. A 

 weight was put on a cross lever attached to the long arm of 

 the vertical one, that stretched the balance spring an amount 

 corresponding to a thrust of 21 pounds, so that negative pressures 

 could be recorded. The pressures corresponding to four points in 

 each revolution were determined by a link that was dropped on 

 the top of the pencil by a system of levers actuated by four cams 

 on the crank shaft. 



This apparatus showed that there was a positive and negative 

 pressure due to the swaying backwards and forwards of the body 

 of the experimenter, the amount of which had to be determined 

 by taking some diagrams, whilst turning the handle with the 

 wings unshipped. These diagrams showed such unaccountable 

 distortions, that it was thought advisable to construct an apparat as 

 similar to Richard's Steam Engine Indicator, a weight being used 

 instead of a watch-spring to pull the drum back wards, as the 

 crank pin slacked the cord ; the wire backstay of the left wing 

 was carried over a block and fastened to a spring balance, and 

 moved the parallel motion of the indicator : the forestay of the 

 wing was removed to a point near the upper end of the connecting 

 rod, as the angular thrust and pull backwards of the connecting 

 rod masked the true thrust in the diagrams : counterpoises (not 

 shewn in the drawings) were also put to the wings. 



The scale of the diagrams was determined by weights of one, 

 two, and three pounds hung direct on to the centre of the length 

 of the wing membrane, the wing being horizontal and the weight 

 cord passing over a pulley. Fig. 4 is the mean of six cards 

 and shows that the total mean thrust of both wings is 3-07 pounds. 



