548 PROCEEDINGS OF THE AMERICAN ACADEMY. 



form devised by him, before he had published his description which 

 appeared in the Journal of Physical Chemistry, April, 1901. 



This apparatus is an ingenious modification of Berthelot's method, in 

 which the suhstance is heated and volatilized by means of an electric cur- 

 rent within the body of the liquid, thus avoiding errors due to external 

 heating. Through the kindness of Professor Kahleuberg and Profes- 

 sor Trevor we are able to show in this connection the form of this 

 apparatus. 



In the publication referred to, the apparatus is described as follows : 

 " The retort consists of a tube 17 centimeters long and 5.5 centimeters in 

 diameter, into the bottom of which is fused a tube which fits into a con- 

 denser with a ground glass joint. At the other end of the inner tube 

 are two large lateral openings. Glass tubes pass through the cork at 

 the top, and into these are fused the ends of the spiral of platinum wire. 

 This spiral consists of about 40 centimeters of fairly stout platinum wire, 

 to the ends of which are welded short heavy jjieces of platinum rod, and 

 these rods are in turn fused into the glass tubes. Long, rather heavy 

 copper wires pass down into the glass tubes, at the bottom of which 

 they are connected with the ends of the platinum rods by means of a few 

 drops of mercury. The calorimeter is covered with a heavy piece of 

 asbestos board and the retort is enclosed in asbestos and cotton batting." 



A current from eight to ten amperes, regulated by a rheostat, gave suf- 

 ficient heat to vaporize from 20 to 30 grams of the oil in from 5 to 6 

 minutes. The calorimeter, about 2500 cubic centimeters capacity, was 

 made of thin nickel-plated sheet copper. It was elliptical in form, to 

 conform to the shape of the condenser, and was provided with a cop- 

 per stirrer. The capacity of the calorimeter was reduced, as shown 

 in the figure, by bending closer together the glass tubes leading from 

 the body of the condenser. Temperatures were taken on a Beckman 

 thermometer. 



The water equivalent of the calorimeter, condenser, stirrer, and ther- 

 mometer were found to be 185 grams, practically the same value as the 

 equivalent calculated from the weights and specific heats of the parts of 

 the apparatus. 



As mentioned above, this form of the apparatus is limited in this work 

 by the fact that when oils of boiling points higher than 125° are volatil- 

 ized, the sudden change in temperature at the water line of the condenser 

 is so great that glass will not stand it. 



The following results were obtained with a few members of the 

 series C n H 2n+2 , as the mean of several observations : 



4 



