516 PROCEEDINGS OF THE AMERICAN ACADEMY. 



a calibrated thermo-couple. The apparatus was ingenious, but the 

 conduction of heat into the calorimeter from the entering tube must 

 have been considerable, and heat was undoubtedly gained also by 

 direct radiation from the black interior surface of the ebonite chamber 

 at the opening in the side of the calorimeter. 



He concluded fi-om his experiments, which he wished to be con- 

 sidered as merely preliminary, that the true value for the heat of 

 vaporization of water (not including the heat given out in cooling the 

 water) is about 540 calories. Unfortunately, his data, corrections, 

 and calculations, are not given. 



Kahlenberg,^^ in an almost simultaneous investigation, sought to 

 retain all the good qualities of the method of Berthelot, and at the 

 same time endeavored to eliminate its sources of error by an ingenious 

 modification, using the heat generated from an electric current con- 

 veyed by a resistance wire immersed under the liquid to be vaporized, 

 in order to produce the vaporization. Thus superheating is avoided, 

 as well as the presence of any flame near the calorimeter. This 

 method, undoubtedly one of the best heretofore used, has yielded 

 fairly good results, which are nevertheless undoubtedly somewhat 

 vitiated by premature condensation. 



Other methods which should be considered are the electrical methods 

 of Marshall and Griffiths, ^^ of Marshall and Ramsay, ^i of Brown, ^2 

 and of Henning.23 



The method of Marshall and Griffiths is exceedingly complicated, 

 and the heat of vaporization of but one substance, benzene, has been 

 obtained. The values were determined for the temperature of 50°C., 

 40°C., 3()°C., and 20°C., and the value at the boiling point obtained by 

 extrapolation. Upon this value for benzene, thus obtained, was 

 based all the work done by Marshall and Ramsay. This latter 

 method was merely a comparative method. The electric current was 

 sent through two similar vaporizers in series, and the amount of liquid 

 distilled from each was weighed. The results obtained all depended, 

 therefore, upon the value of the heat of vaporization of one substance, 

 and for this one substance different experimenters had found values 

 differing by over sixteen per cent. The investigators state that they 

 were unable to obtain any satisfactory results for water, perhaps because 

 of electrolysis and its consequent consumption of electrical energy. 



" Kahlonberg, Journal Phys. Chem., 5, 215 (1S95). 

 20 Marshall and Griffiths, Phil. Mag. (5) 41, 1 (1896). 

 " Marshall and Ramsav, Phil. Mag. (.5) 41, 38 (1896). 



22 BrowTi, Jr. Chem. Soc, 83, 987 (1903). 



23 Ilemiing, Ann. d. Phys. (4) 21, 849 (1906). 



