( 575 ) 

 TABLE XXII. 



If ^^ c cak'uhilo for Ihe different values the coeflieiejit / for v.\n 

 DKM Waals's forinuhi ^) : 



/'/.- J 



WO liiid with iiuTeashig- leiiij)erature regiilarlv (lecreasijiii- values of 



1 ■ 

 /. It ill the (le\eloi)ineiil in pouers ot — , \ve take (^Jie term more, 



so that we arri>'(' at the formula: 



hui 



P 





and we calculate tiie values of /' and ƒ/ which ,^•i^"e the i)est a;i:ree- 

 ment, we iind verv nearlv ƒ/ = ƒ. Then 1 have [)ut : 



/<uf z= r -„- 



/'/.• ■ V- 



and found ƒ' =2.1114'. \U means of this we obtain the values for 

 ft given iji column ( . 'J'he agreement is (|iiite satisfactorv, onlv for 

 the temperature /{(P.S'i (\ there is a considerable de\iation : the same 

 is also foujid in the coinparison of the saturation volumes so that 

 here \ve have probablv to do with an error of ol)ser\ation. 

 From this we find al the critical point: 



From the dclcrminalions of the isothermals we cannot \\ ith cer 



1.(310. 



taintv deri\(' a dcliiiitc \ariation of ^ , 1 at the same volume with 



the temperature, as mav l)e best derived from table XX. Therefore 1 



/'Of' \ 

 have derived for the different d(Misiti(;s from the isothermals 



yd I J, 



of 3Ü°.98 C. and 4SM0 C. and fiMiiid the fol](»wiiii: values: 



^^ Contiuuitiit 1 p. 15b. 



