FUNDAMENTAL PROPERTIES OF THE ELEMENTS RICHARDS. 211 



with its interpretations leads me to feel that the idea is highly sug- 

 gestive and helpful in stimulating new search after truth and in cor- 

 relating and codifying diverse facts. By such fruit are hypotheses 

 justified. 



The relation between heat of reaction and change of volume 

 stimulates interest in chemical thermodynamics and curiosity as to 

 the mechanism of the output of energy during chemical change. A 

 search for accurate data wherewith to reason about this question 

 soon revealed the uncertain nature of many of the figures. Here, in 

 the domain of thermochemistry, as in those of atomic weights and 

 compressibilities, new methods were needed in order to attain pre- 

 cise results. Accordingly, a device was adopted which at one stroke 

 annihilates the pernicious "cooling correction" — the worst foe to 

 accuracy — by merely causing the temperature of the jacket around 

 the calorimeter to change in temperature at the same rate as the 

 calorimeter itself. There are several ways in which this may be 

 accomplished; among these ways the following was chosen as the 

 best method for a chemical laboratory. The calorimeter, inclosed in 

 a slightly larger water-tight vessel, with tubes above — a kind of sub- 

 marine — is immersed under the surface of dilute crude alkali in a 

 pail. Thermometers inside and out enable one to adjust the tem- 

 peratures at the same point. The reaction is then started in the calo- 

 rimeter, and at the same moment and at a corresponding rate acid 

 is dropped into the dilute alkali in the pail, so that the two tempera- 

 tures inside and out keep pace with one another. Thus there is no 

 loss of heat from the inside vessel; the thermochemical reaction is 

 strictly adiabatic. This method has already been used at Harvard 

 with very encouraging outcome in determining a wide variety of 

 thermochemical data, heats of combustion of hydrocarbons, of solu- 

 tions of metals in acids and of neutralization, specific heats of solu- 

 tions, and also of the elements at very low temperatures, and finally 

 latent heats of evaporation. 1 It has proved itself especially valuable 

 in the study of slow reactions, where the cooling correction may 

 become a large portion of the total result. The effort is being made 

 to apply to this experimentation concerning chemical energetics the 

 same degree of care which has recently been attempted in the revi- 

 sion of the atomic weights, and although on account of the greater 

 complexity of the problem the percentage accuracy thus far reached 

 has not equaled that in the case of atomic weights, one can not help 

 thinking that the proportional gain over previous investigations is 

 perhaps as great in this case as in the other. 



t Richards, in collaboration with Henderson, Forbes, Frevert, Mathews, Rowe, Jesse, Burgess, and 

 Jackson, Proceedings American Academy, 1905, vol. 41, p. 3; 1907, vol. 42, p. 573; 1908, vol. 43, p. 475; 

 1911, vol. 46, p. 303; Journal American Chemical Society, 1909, vol. 31, p. 1275; 1910, vol. 32, pp. 208, 432, 

 1176; Zeitsch. physikal. Chem., 1905, vol. 52, p 551; 1907, vol. 59, p. 531; 1909, vol. 70, p. 414. 



