688 STATE BOARD OF AGRICULTURE. 



tcutiou is drawn to the necessity of using the extended coils of the bridge 

 if a greater accuracy than .05 of 1% is desired. It is preferable to use a 

 bridge wire of rather low resistance as it can stand wear and heavy 

 treatment better. It will also hold its calibration more certainly. In 

 regard to the electrolytic cells, it may be said that their construction is of 

 utmost importance. These cells act as capacities and to balance this, a 

 condenser must be put in across the resistance in the adjacent arm. Cells 

 with platinized electrodes are necessary when working with a frequency 

 of GO cycles per second. Too much attention cannot be given to the detail 

 of maintaining a water bath at a constant temperature. A Freas water 

 thermostat which is accurate to .002 of a degree is efficient for this work. 

 A method for operating the apparatus described in this paper is sug- 

 gested and a few experiments added, to shoAV the degree of precision ob- 

 tained. This apparatus has proven efficient in our hands, and a measure- 

 ment can be made quickly and easily. Besides this, the results show a 

 high degree of precision. 



7. ACKNOWLEDGMENTS. 



We are especially indebted to Professors Smith and Williams of the 

 Physics Department of the University of Michigan for the use of appa- 

 ratus, laboratory space, and personal aid and criticism. It was in their 

 laboratory and with their apparatus that we were able to compare the 

 results obtained with the commutator and D. C. galvanometer, with those 

 obtained with the alternating current galvanometer. 



We also wish to acknowledge our indebtedness to Professor E. W. 

 Washburn of the University of Illinois who kindly let us make some 

 measurements wdth his apparatus, and who further kindly gave us blue 

 prints illustrating the setup and the various connections used by him. 

 We also wish to thank Professor A. R. Sawyer of the Michigan Agri- 

 cultural College for the loan of various smaller pieces of apparatus w^hen 

 these were necessaiy in the two years of our investigations. He has also 

 given us per.sonal attention and advice and kindly read the manuscript. 



The authors are indebted to Leeds & Northrup Co. for permission to use 

 Figures 1 and 2, and for the loan of an electrotype of the alternating 

 current galvanometer. 



The mechanical drawings, Figures 1, 2, 3, 4, 5, 6, 9 and 14 were made 

 by P. G. Andress of the class of 1916, Michigan Agricultural College. 



