September 17, 1915] 



SCIENCE 



389 



ing 0.01 per cent., and the writer has, with sug- 

 gestions from Professor S. F. Acree, attempted 

 to improve upon the technique employed by 

 studying the conductivity cells, the baths, 

 temperature regulation, the bridge and resist- 

 ances, the methods of making and handling 

 the solutions, and the so-called electrode and 

 polarization phenomena. This improvement is 

 now necessary in order to allow the writer to 

 use the conductivity method for the determi- 

 nation of ionizations and reaction velocities 

 in dilute solutions. He is indebted to Dr. 

 Curtis- and Dr. Wenner, of the National Bu- 

 reau of Standards, for much valuable advice, 

 and the fine work of Washburn^ and Bell 

 shows what great improvements can be made 

 in this line. 



The writer has used the excellent equip- 

 ment of the Bureau of Standards and some 

 fine apparatus loaned us by Leeds and 

 Northrup in making a fundamental study of 

 a large number of factors, some of which have 

 already been investigated in conductivity 

 work by physical chemists. As a result of 

 this work he has already greatly improved 

 the methods and has studied: (1) the cur- 

 rent from (a) induction coils, (b) a Holzer- 

 Cabot wireless generator, (c) a General Elec- 

 tric Company large generator, (d) a Siemans- 

 Halske generator for conductivity work, and 

 (e) a Vreeland oscillator furnished by Leeds, 

 Northrup & Co., which we have found to be 

 the best source of current yet tried, as it gives 

 a pure sine wave of uniform frequency which 

 can be varied very widely; (2) the voltage, 

 which when varied from 0.25 to 8 volts, has 

 shown no influence on the resistance of the 

 solutions measured so far in very clean cells, 

 but is very important in cells not entirely 

 clean; (3) the size and shape of the electrodes, 

 which have a very large influence on the 

 change of resistance and capacity of the cell 

 with change in frequency; (4) the material 

 used in making the electrodes (Pt, Au, Ag, 

 Cu, Zn, etc.), which is very important; (5) 

 the state of aggregation of the surface of the 



2 Ciirtis & Grover, Bureau of Standards Bulletin, 

 Vol. 8, No. 3. 



3 Jour. Am. Chem. Soc, 35, 177, 1913. 



electrodes, as in plain, gray and platinized 

 electrodes, which has a very great influence on 

 the capacity of the cell and change of resist- 

 ance with change in frequency; (6) the fre- 

 quency of the alternating current, which when 

 varied may change the resistance of some solu- 

 tions in some cells as much as 3 per cent.; 

 (Y) the high capacity of the cell as a conden- 

 ser, which is very important in decreasing the 

 change of resistance with change in frequency 

 and in obtaining a perfect minimum in the 

 telephone; (8) the valence and velocities of 

 the different ions; (9) the influence of the 

 concentration and the character of the electro- 

 lyte and the solvent on the change of resistance 

 with change in frequency; (10) the proper 

 use of a condenser or inductance in balancing 

 the capacity of the cell, and its influence on 

 the resistance and minimum in the telephone ; 

 (11) the construction of the cell in such a way 

 that no errors from evaporation and concen- 

 tration can be produced; (12) the use of a 

 tuned telephone attached to a stethoscope or 

 of double wireless telephones; (13) the con- 

 struction of a Wheatstone bridge with Curtis 

 resistances free from inductance and capacity, 

 kept automatically at constant temperature, 

 and arranged so that every resistance can be 

 cheeked against the others and against stand- 

 ard enclosed resistance; (14) the development 

 of especially good constant temperature baths 

 for such work; (15) the use of weight meth- 

 ods and special apparatus for making, keeping 

 and transferring solutions; (16) a number of 

 points connected with the proper use of all of 

 the apparatus, especially the electrical equip- 

 ment, to prevent errors arising from induc- 

 tion, capacity, skin effects, electrical leaks and 

 other factors; (17) our criterion of excellent 

 cells, namely that each one must be indepen- 

 dent of the above sources of error and give 

 readings constant to within 0.01 per cent., and 

 especially that whatever the solution used, the 

 ratios of the resistances in any two such cells 

 must he constant to within 0.01 per cent. 

 Only in this way can we be certain that the 

 electrode effects have been practically elimi- 

 nated and that we are measuring the true elec- 

 trical resistance of the solution with great 



