428 Bayliss and Starling, 



placed in connection with the capillary by turning the stopcock, the jj 

 air is compressed, but photographs taken of this seem to show that 

 the heat formed escapes as rapidly as it is produced, for the volume 

 of the air-space does not further diminish under continued exposure to 

 the same pressure, as it would do if it had been heated, and moreover 

 the rate of compression is probably not sufficiently rapid for it to be i 

 adiabatic. 



Let us now see how far the instrument fulfils other conditions of 

 a good manometer. 



I. As to mass moved. It is obvious that a great advantage is 

 gained by the abolition of levers etc., for recording the movement; 

 and, besides the diminution of mass, the photographic method has the 

 further advantage of giving curves whose ordinates are straight lines 

 instead of arcs of circles, as in all cases where levers moving around 

 a centre are employed. The only mass moved is the volume of fluid 

 forced into and out of the capillary; this we have measured, and find 

 that, for 100 mm mercury increase of pressure, there is a volume of 

 fluid moved equal to 0-0335 cub. mm. In this respect it compares very 

 favourably even with Hürthle's small "Gummimanometer", the corre- 

 sponding volume of fluid in which is equal to 90 cub. mm. i) 



II. As to Rapidity of Movement, or Inertia. The most rapid 

 rate of change of pressure we were able to produce was one of 

 4750 mm Hg. per second, and to this our instrument responded accu- 

 rately, although with a few vibrations before coming to rest. Hürthle's 

 instrument, undamped, could move at a rate of 10 000 mm Hg. per 

 second. In this respect therefore, we can only say that our instrument 

 responds to the most rapid rate at which we have tested it. 



in. As to Mohilitij. In respect of latent period, we have made 

 no measurements, but that the instrument can i-espond to rapid 

 changes of pressure is shown by the fact that the number of vibrations, 

 produced in ihe instrument itself by very suddenly turning on the 

 pressure-bottle, amounted to 63 per second. 



IV. As to Aperio dicity. When the tap connecting the manometer 



') Pflüger's Archiv. XLIII, p. 409. 



