September 22, 1905.] 



SCIENCE. 



357 



appreciating the analogy of diffusion and 

 heat conduction, placed the phenomenon 

 on a satisfactory theoretical basis, and 

 Fick's law has since been rigorously tested, 

 in particular by H. F. Weber (1879). 



The development of diffusion from a 

 physical point of view followed Pfeffer's 

 discovery (1877) of osmotic pressure, soon 

 after to be interpreted by vant' Hoff 

 (1887) in terms of Boyle's and Avogadro's 

 laws. A molecular theory of diffusion was 

 thereupon given by Nernst (1887). 



DYNAMICS. 



In pure dynamics the nineteenth century 

 inherited from the eighteenth that un- 

 rivaled feat of reasoning called by La- 

 grange the 'Mecanique Analytique' (1788), 

 and the great master was present as far as 

 1813 to point out its resources and to watch 

 over the legitimacy of its applications. 

 Throughout the whole century each new 

 advance has but vindicated the preeminent 

 power and safety of its methods. It tri- 

 umphed with Maxwell (1864), when he 

 deduced the concealed kinetics of the elec- 

 tromagnetic field, and with Gibbs (1876- 

 78), when he adapted it to the equilibrium 

 of chemical systems. It will triumph again 

 in the electromagnetic dynamics of the 

 future. 



Naturally there were reactions against 

 the tyranny of the method of ' liaisons. ' 

 The most outspoken of these, propounded 

 under the protection of Laplace himself, 

 was the celebrated 'mecanicjue physique' 

 of Poisson (1828), an accentuation of Bos- 

 covich's (1758) dynamics, which permeates 

 the work of Navier, Cauchy, de St. Venant, 

 Boussinesq, even Fresnel, Ampere and a 

 host of others. Cauchy in particular spent 

 much time" to reconcile the molecular 

 method with the Lagrangian abstractions. 

 But Poisson 's method, though sustained by 

 such splendid genius, has, nevertheless, en 

 more than one occasion — in capillarity, in 



elastics— shown itself to be untrustworthy. 

 It was rudely shaken when, with the rise of 

 modern electricity, the influence of the 

 medium was more and more pushed to the 

 front. 



Another complete reconstruction of dy- 

 namics is due to Thomson and Tait (1867), 

 in their endeavor to gain clearness and 

 uniformity of design, by referring the 

 whole subject logically back to Newton. 

 This great work is the first to make sys- 

 tematic use of the doctrine of the conserva- 

 tion of energy. 



Finally, Hertz (1894), imbued with the 

 general trend of cotemporaneous thought, 

 made a powerful effort to exclude force 

 and potential energy from dynamics alto- 

 gether—postulating a universe of concealed 

 motions such as Helmholtz (1884) had 

 treated in his theory of cyclic systems, and 

 Kelvin had conceived in his adynamic gyro- 

 static ether (1890). In fact the introduc- 

 tion of concealed systems and of ordered 

 molecular motions by Helmholtz and Boltz- 

 mann has proved most potent in justifying 

 the Lagrangian dynamics in its application 

 to the actual motions of nature. 



The specific contributions of the first 

 rank which dynamics owes to the last cen- 

 tury, engrossed as it was with the applica- 

 tions of the subject, or with its mathemat- 

 ical difficulties, are not numerous. In 

 chronological order we recall naturally the 

 statics (1804) and the rotational dynamics 

 (1834) of Poinsot, all in their geometrical 

 character so surprisingly distinct from the 

 cotemporary dynamics of Lagrange and 

 Laplace. We further recall Gauss's prin- 

 ciple of least constraint (1829), but little 

 used, though often in its applications su- 

 perior to the method of displacement; 

 Llamilton's principle of varying action 

 (1834) and his characteristic function 

 (l'"?4, 1835), the former obtainable by an 

 easy transition from D'Alembert's prin- 



