niMl'AIMSuN OF PLATINTM AM' c\s TH KKM< ).MKT1 l>. 



99 



:itti-m|itril to collect further evidence on the subject, before proceeding to the final 

 reduction of our results. 



( ' \I.I.K\DAK and GRIFFITHS in re-determining the Ixiiling-point of sulphur made 

 no attempt to deduce any formula for the variation of this point with prefigure, and, 

 in their subsequent work, apply the one deduced by REONAULT from his observations 



made ill 1862. 



As the results of this investigation of REONAULT have been differently interpreted 

 by several observers, it may be worth while here to state exactly what experiments 

 UKUNAULT made on the subject. The primary object of his work was to determine 

 the influence of large variations of pressure on the lx>iling-points of a number of 

 substances, rather than to deduce formulae representing accurately over a limited 

 range the variation for each substance. He made altogether eight experiments with 

 sulphur at pressures between 250 and 3000 millims., the four nearest to the standard 

 pressure of 7CO millims. l>eing as follows: 



Temperature 011 uir scale. 



418-70 

 440-30 

 447-71 

 186-61 



Pressure in millims. 



467-45 



679-97 



763-04 



1308-54 



In the carrying out of these experiments REONAULT says he had considerable 

 difficulty, due to violent boiling and also to superheating of the vapour, especially at 

 lii^li pressures. 



From the eight experiments REGNAULT calculated a formula for the change of 

 temperature with pressure over the whole range ; from this GRIFFITHS finds the 

 value of dt/dp at 760 millims. to be 0'082. 



It happens, however, that the experiment made at 763 millims. is one, the result of 

 which diverges more from the calculated value than almost any other, and therefore 

 the value to be taken as the boiling-point at 760 millims. is appreciably uncertain. 



The most probable value for this point, as deduced from these observations of 

 REGNAULT, is given by different authorities as 448'38, 448'34, and 447'48. 



In view of this uncertainty, and also of the fact that the Meyer tube apparatus is 

 so entirely different in its construction from that employed by REGNAULT, we deemed 

 it advisable to obtain some further evidence as to the validity of the application of 

 REGNAULT'S value of dt/dp to our experiments. As our own observations happened 

 to be ah 1 made within a small pressure range, we selected from the records of the 

 platinum thermometers in regular use at Kew Observatory, the results of the different 

 determinations of the sulphur-point made with thermometers K.I and K.3, and from 

 these, calculated by least squares for each thermometer a formula representing the 



o 2 



