C 213 ] 



XIX. On the Thermal Properties of Normal Pentane. — 

 Part II. By J. Rose-Innes, M.A., B.Sc, and Sydney 

 Young, B.Sc, F.R.S* 



[Plate II.] 



IN the first part of our paper on this subject, read before the 

 Physical Society last December f, we published a large 

 array of figures giving the relations between the volume, 

 temperature, and pressure of normal pentane ; it was sug- 

 gested that a reasonably good agreement between calculation 

 and experiment might be secured by using the formula 



RT f , el I 



T i e ~\ 



\ v + k~gv~ 2 f 



^ v [_""" v + k~ gv~ 2 J v(y + k) 



(Phil. Mag. xlvii. p. 367). This formula was first found by 

 one of us for isopentane, but it seemed likely that it would 

 do equally well for normal pentane ; and arguments were 

 adduce:! showing that in such case the values of R and of 

 l-r-e might be taken to be the same for the two isomers. 

 The question then arose whether I and e for normal pentane 

 could be separately taken as equal to the values already found 

 for isopentane, and the question was provisionally answered 

 in the negative (loc. cit. p. 367). Two methods were employed 

 in dealing with this matter, but neither of them could be 

 regarded as conclusive. 



The first plan was to plot the differences of — 2 against 



v~& ; these differences grew smaller as the volume was made 

 larger, but it did not seem likely that they vanished altogether 

 when the volume was made infinite. The " wobbling " at 

 large volumes, however, made it impossible to speak with any 

 certainty one way or the other. 



The second plan adopted was to take the three constants R, 

 e, and I the same as for isopentane, and then to calculate suitable 

 values for the two remaining constants k and g from the 

 experimental data near the critical point. The isothermals 

 were then calculated from the formula, and it was found that 

 deviations occurred between calculation and experiment 

 amounting to nearly 2 per cent. Accordingly the hypothesis 

 on which the calculations were founded was considered to be 

 most probably incorrect. 



This second method of investigating the question has 

 several advantages. We have to deal with the experimental 



* Communicated by the Physical Society : read May 26, 1899. 

 t Phil. Mag. [5] vol. xlvii. p. 353. 



Phil. Mag. S. 5. Vol. 48. No. 29] . Aug. 1899. Q 



