GIBBS' PAPERS I AND II 57 



have had) or would he have gone into the matter of the electroly- 

 tic cell, or the theory of dilute solutions, or the mass law? Per- 

 haps he would have followed the lead of the address of the 

 Chairman and confined himself chiefly to contributions of others. 



It is not without interest that in the period from 1872 to 1891 he 

 is not recorded as offering any course on thermodynamics which 

 could be presumed to include any of the matters in his thermo- 

 dynamic papers, although from 1886 on he announced a course 

 on the a priori deduction of thermodynamic principles from the 

 theory of probabilities, which in view of his paper of 1884 (Gibbs, 

 II, Pt. II, p. 16) may safely be assumed to have dealt with 

 statistical mechanics. Was he concentrating his attention, as 

 Clausius and Maxwell had done and as Boltzmann and Kelvin 

 were doing, on the attempt to deduce thermodynamic behavior 

 from dynamical properties of matter and possibly to find some 

 equation expressing the thermodynamic functions of a body of 

 variable composition other than perfect gases? It is not often 

 that we find a great scientist neglecting in his lectures his own 

 most important contributions at a time when they are of as 

 great interest to others as Gibbs' contributions were to the ris- 

 ing physical chemists of the decade from the early eighties to 

 the early nineties of the past century. Certainly the subject 

 matter of his Papers I and II to which he gave half his time 

 during the year 1899-1900 in the course above summarized was 

 no more difficult, no less suitable for instruction than the courses 

 he did offer on mathematical physics to students who could not 

 have been expected to have much if any physics beyond the first 

 general course, or much if any mathematics beyond the differ- 

 ential and integral calculus.* 



It has been seen that Gibbs, as he taught thermodynamics, 

 late in his life, made much use of the pressure-volume diagram, 

 discussed briefly the entropy-temperature and pressure-temper- 

 ature diagrams, but ignored the volume-entropy diagram (except 

 as its properties may be considered to be implied in those of the 

 thermodynamic surface). He made no use of the concept of 



* The list of courses offered by Gibbs from 1872 to the time of his 

 death is given in my "Reminiscences of Gibbs by a Student and Col- 

 league" in the Scientific Monthly, 32, 210-227, (1931). 



