380 Mr H. Darwin, On a Self-recording Barometer. [Mar. 15, 



C than had entered by A, the gas bag would empty and the pen 

 would fall. Thus by opening and closing the tube G, the board 

 can be made to rise and fall, and to move the pen with it. The 

 barometer performs this function in such a manner, that the 

 vertical movement of the pen is three inches upwards for every 

 fall of atmospheric pressure corresponding to a head of one inch 

 of mercury. The barometer is in the form of an inverted 

 siphon, 0, H. The tube G is connected with a small glass tube 

 passing down the open limb of the barometer at G; it is sup- 

 ported by a bracket K fixed to the pressure board D at one-sixth 

 the distance of the pen from the hinges E. Now suppose the 

 end of the glass tube almost touches the surface of the mercury 

 in the open limb of the barometer, and that the flow of gas 

 through this small opening equals the flow entering by the tube 

 A from the main ; then no gas will enter or leave the bag and the 

 pressure board D will remain at rest. When the mercury in the 

 open limb of the barometer rises the leak will be diminished : 

 gas will enter the bag and will raise the pressure board, together 

 with the pen and the glass tube passing down the open end of 

 the barometer ; this will at once increase the leak and will 

 shortly again establish equilibrium. The converse will take place 

 when the mercury falls, and the end of the glass tube will 

 accurately follow the movement of the surface of the mercury. 

 If the atmospheric pressure increases by a head of one inch of 

 mercury the surface of the mercury falls half an inch in the open 

 limb of the barometer : the pen falls six times that amount, that 

 is, three inches. The instrument has only been constructed in an 

 experimental form, and it was not thought worth while to com- 

 plicate the instrument by putting an arrangement to prevent the 

 gas escaping into the room. But the leak is so small that with a 

 similar instrument working day and night for some weeks the 

 smell was not perceptible. 



I am indebted to Mr Deacon for the design of the pen. It is 

 used by him in his Waste Water Meter and in other self-recording 

 apparatus. These instruments are left for long periods without 

 being touched, the pen drawing the curves in a most satisfactory 

 manner. 



In the apparatus exhibited the tracing is made on paper cover- 

 ing a drum rotating about a vertical axis. An American clock 

 movement is fixed to the top of the drum, and the point of 

 rotation of the hands is on its axis. The hour hand projects 

 beyond the side of the drum, and rests against a fixed pin Q, 

 and is thus prevented from rotating. Consequently the body of 

 the clock and the drum with it are compelled to turn ; the usual 

 condition where the hands rotate and the body remains at rest is 

 reversed. 



