﻿252 Mr. J. H. T. Roberts on a Method of 



funnel constitutes the air-tube, the funnel and the tap T t are 

 hermetically sealed into the reservoir, and the tap T\ is closed 

 after the reservoir is filled with mercury. The carriage 

 containing the reservoir is supported in a strong wooden 

 tower, about 3 feet in height, and rests upon horizontal iron 

 bars, placed in holes at any desired height. 



The lower gla^s vessel consists of two sphe v ical parts B, 

 (diameters about 9 and 8 cm. respectively), which are filled 

 with the gas to he experimented upon, the oas being supplied, 

 through drying-tubes, to the side tube D. The other side 

 tube leads to a manometer, which should contain some liquid 

 giving off little vapour: Fleuss pump-oil may be used for 

 lower pressures, and mercury for higher pressures. 



The upper end of the vessel is closed by a well-fitting 

 rubber stopper, through which passes a short length of glass 

 tubing, about 5 mm. bore; at the upper end of this tube is a 

 small piece of rubber tubing which can be closed to exclude 

 air until an experiment is begun. This is important, as the 

 presence of a very small amount of foreign gas has a con- 

 siderable effect upon the transpiration time *. The capillary 

 is contained within this glass tube, and is secured at its lower 

 end with soft wax. 



The tap T 2 is to turn the mercury absolutely on or off, T 3 is 

 to regulate the rate of inflow, and T 4 is to empty the vessel 

 after an experiment. The mercury passes up the curved 

 tube E, and enters the vessel at a fixed height, and therefore 

 at a constant rate, once the pressure inside the vessel las 

 become constant ; the gas is forced through the capillary tube, 

 and its pressure is indicated on the manometer. The tap T 2 

 being full on, T 3 is adjusted until a suitable pressure is 

 obtained within the vessel, and everything is allowed to 

 attain a steady state ; with the dimensions given, this state 

 will be attained before the mercury reaches the lower part, of 

 bulb B ; the time is then observed for the mercury to pass 

 marks at the lower and upper ends of bulb B, the volume 

 between the marks being determined subsequentlv. 



If the adjustment of the tap T 3 be preserved, and the 

 mercury again passed through, the rate of inflow of the 

 mercury, and of the efflux of the gas, will be practically 

 the same as before, for pressures on the manometer which 

 are small compared with the statical pressure due to the 

 mercury. Thus the pressures required to force different 

 gases through the same capillary at the same rate can be at 

 once compared. 



The dependence of pressure upon length for a constant 

 * Graliam (Joe. cit.). 



