COMPARISON MF I'l.ATlNTM AND <JAS THKUMOMKTKKS. 75 



(c) Di-tiTiiiin'it.'.tn, of the Coefficient of Pressure of the TJieivnometric Reservoir. 



The capillary tube fused to the thermometric reservoir had, at the commencement, 

 a length of about 250 millims. The volume of this tube was determined by weighing 

 a thread of mercury occupying in it a length of 200 millims. ; the weight of mercury 

 contained in 1 millim. of length was found to be 1 '69185 gramme. The calibre of the 

 tube was afterwards studied by GAY-Lu88AC*8 method between the two extreme 

 points and 250. The calibration corrections thus obtained had to be applied in the 

 reduction of the observations on the coefficient of pressure. To measure this coefficient 

 the same' method is followed as for the determination of the coefficient of external 

 pressure of mercury thermometers. 



The thermometric reservoir is placed in a glass tube filled with water, and closed 

 by a cork pierced with a hole, through which passes the capillary attached to the 

 reservoir. The space between the reservoir and the external tube can, by means 

 of stop-cocks, lie put into communication either with the atmosphere, or with a large 

 exhausted vessel. The reservoir itself being filled with water up to a certain scale 

 division, observations are made of the displacements of the meniscus produced by 

 varying the external pressure by nearly an atmosphere. 



The observations effected under these conditions gave, after all reductions, the 

 following value for the variation Aw of the volume of the reservoir, which corresponds 

 to a variation of external pressure equal to a millimetre of mercury, 



Av = 0-006 228 microlitre. 



This value of the coefficient of pressure was employed for the calculation of a table 

 giving the variations of volume of the reservoir corresponding to the changes of 

 internal pressure observed in the course of the experiments. 



(d) Determination of the "Dead Space" ("Espace Nuisible"). 



The determination of the volume of the space occupied by the gas not exposed to 

 the same temperature as the reservoir presents peculiar difficulties. It is of extreme 

 importance that the limits of this space should be well defined, which, however, 

 cannot be done quite rigorously. 



The "dead space" may be divided into two parts: (1) the space occupied by the 

 gas in the closed branch of the manometer between the mercury touching the point 

 and the lower surface of the stopper, and (2) the internal volume of the capillary tube 

 between the plane of the stopper and the part of the tube which penetrates into the 

 heating apparatus. The curvature of the mercury-meniscus in the closed branch is 

 necessarily somewhat variable, and as the diameter of the tube is 16 millims., small 



L 2 



