510 Mr. D. M c Intosh on the Conductivity of 



form better suited to itself if it had been allowed to do so, as 

 it is in commerce. This consideration must affect, not only 

 the efficiency, but also the curves which Fleming and Petavel 

 obtained for the variations in the luminous intensity. 



Further, this action of the arc in modifying the wave form 

 may throw some light on the discrepancy between the 

 efficiency of alternate current arcs as determined in the labora- 

 tory and that stated to be obtained in practice. 



When an arc is run in the laboratory a large resistance is 

 almost certain to be put in series with it to ensure that degree of 

 steadiness which is essential to exact measurement, and hence 

 the arc cannot alter the wave form. In the commercial use of 

 arcs, on the contrary, the circumstances are widely different. 

 In this case, for economic reasons, the arc must form a large 

 percentage of the total " reactance " of the circuit, and there- 

 fore can easily alter the wave to the form required for the 

 greatest efficiency. 



It is interesting to note that the wave form giving the best 

 result for the arc is almost exactly the opposite to that giving 

 the best efficiency for transformers. In the former case a 

 flat-topped wave is best, while for the maximum efficiency of 

 transformers an exceedingly peaked wave is best, as lately 

 found by Dr. Rossler. 



This points to the building of alternators for use with 

 transformers in such a way as to give peaked wave-forms. 



In the case of the arc the building of machines to give the 

 most efficient wave form is not so necessary, since, generally, 

 the arc itself has the power of automatically converting any 

 wave form into the one best suited to its requirements. 

 Nevertheless, when the arc has to run in series with a large 

 resistance it is of the utmost importance for obtaining the 

 best efficiency that the machine should give a flat-topped 

 wave. 



City and Guilds of London Central Technical College, 

 April 2, 1896. 



LVIII. On the Calculation of the Conductivity of Mixtures of 

 Electrolytes having a common Ion, By Douglas M c Intosh, 

 Physical Laboratory, Dalhousie College, Halifax, U.S.* 



IN a paper published in the April number of this Maga- 

 zine (supra, p. 276) Prof. MacGregor showed how to 

 obtain, by a graphical process, from observations of the elec- 

 trical conductivity of a sufficient number of simple solutions 



* Abstract of a paper read before the Nova Scotia Institute of Science 

 on the 13th of April, 1896. Communicated by the Secretary of the 

 Institute. 



