April 10, 1908] 



SCIENCE 



569 



The principal object in the design of this 

 cell is to present the concept of "molec- 

 ular conductivity" of an electrolyte in 

 such a clear and concise form that no one 

 who is capable of making electrical meas- 

 urements can misunderstand it. The ap- 

 paratus consists of a strong glass tube pro- 

 vided with a small side tube for filling. 

 Both ends are ground plane and closed by 

 platinum-faced electrodes, the whole being 

 firmly clamped together in a suitable 

 framework. Good insulation between the 

 electrodes is provided by making a portion 

 of this frame of ebonite. Crushing strains 

 due to the unequal contraction of glass and 

 metal are avoided by the use of heavy rub- 

 ber washers, one on each side, which take 

 up the extra length without much increase 

 in pressure. It has been found that a 

 moderate pressure is sufficient to prevent 

 leakage between the ends of the glass tube 

 and the metal electrodes. The apparatus 

 is supported by four short legs, one at each 

 comer. 



The resistance of the column of liquid 

 contained in this tube is measured by the 

 method of Wheatstone's bridge, using a 

 telephone and alternating current, and the 

 specific conductivity, c, or the conductivity 

 of a centimeter cube of the solution from 

 one face to the opposite one, is computed 

 in the usual way. 



The molecular conductivity, jtt, of the 

 electrolyte is then defined as the conduct- 

 ivity of a centimeter cube of the solution 

 per gram molecule of salt within this cube. 



In symbols, 



IX = o/m, 



where m denotes the concentration of the 

 electrolyte expressed in gram molecules 

 per c.c. of the solution. 



Heat of Evaporation of Water: A. W. 



Smith, University of Michigan. 



The heat of evaporation of water is one 

 of the important constants of nature. It 



has far-reaching practical applications, es- 

 pecially at high temperatures, while at low 

 temperatures it enters into many problems 

 of the physicist, the chemist and the 

 meteorologist. Unfortunately, few deter- 

 minations have been made at ordinary tem- 

 peratures, and to fill this want my recent 

 work was undertaken. The method em- 

 ployed was to draw a stream of dry air 

 through the water and determine the 

 amount of water evaporated by again dry- 

 ing the air and weighing the water thus 

 collected. Heat to maintain a constant 

 temperature was supplied by an electric 

 current and the results are expressed in 

 terms of international joules per gram of 

 water evaporated. 



The present pape: gives the final cor- 

 rected results expressed in terms of "mean 

 calories" as well as in joules. The true 

 value of the E.M.F. of a Clark cell at 15° 

 C. is probably much nearer 1.433 volts than 

 the legally accepted value, 1.434 volts. It 

 is further shown that for this purpose at 

 least, the most probable value of J is 

 4.1836 joules per mean calorie. Using 

 these factors brings my results and those 

 of other investigators at both higher and 

 lower temperatures into perfect agreement. 

 It is, therefore, possible to give the values 

 of the heat of evaporation of water for the 

 entire range of temperature from 0° C. to 

 100° C. with a very high degree of cer- 

 tainty at the lower temperatures and 

 probably not over 0.2 per cent, error at 

 100° C. 



The values of the heat of evaporation 

 determined by the various investigators are 

 plotted on a sheet of accurately engraved 

 cross-section paper, and the nearest smooth 

 curve is drawn through the points thus 

 located. From this curve are then ob- 

 tained values corresponding to tempera- 

 ture intervals of 5° C. These values 

 decrease from 596.3 calories at 0° to 538.0 

 at 100°. 



