74 Dr. Meyer Wilderman on the Velocity of 



Crompton and Fisher to be 0*000005. If the temperature of 

 the room varies only 1 to 2 degrees during the whole series 

 of experiments, which is commonly the case with all labor- 

 atories, no correction for the bridge-wire is required. When 

 we investigate dilute solutions the whole series is carried out 

 within 3 or 4 cm. of the bridge; for the velocity of reactions 

 15-20 cm. are required. If we arrange the resistances so that 

 the freezing-point of the solvent comes very near to the middle 

 of the bridge, even greater variations in the temperature of 

 the room will not affect the results, for the expansion of the 

 wire and the exceedingly small change in the resistance of 

 the two parts of the bridge compensate each other. Another 

 great advantage gained by it is that the calibration of the 

 wires on the scale can be made more accurately and more 

 easily. To calibrate a wire to 0*01 mm. entails much tedious 

 difficult work, and in the end it cannot be relied upon as if it 

 had been made only to 0*1 mm. The results of the calibration 

 can also be relied upon for a longer period than in the case 

 of a thinner wire. To prevent oxidation of the manganin 

 wires, they were gilt on the proposal of Crompton and 

 Fisher ; however, oxidation does not occur at ordinary 

 temperatures, and becomes apparent only at 100° C. Besides 

 the two thick manganin wires used for the measurements of 

 temperature, a thin manganin wire, 100 cm. long, the length 

 between 10 and 90 of the scale being equal to 0*451 ohm, 

 was stretched in the air and used as a rheostat. The tem- 

 perature-coefficient of it is 0*0000035 (Mr. Fisher found 

 0*0000025), and only a relatively small part of it is inserted 

 in the circuit. Its purpose is to bring the indication of the 

 point of equilibrium on any desired place of the thick wires, 

 so that it plays the part of the upper chamber in the Beckmann 

 mercury-thermometer for freezing- and boiling-points. It was 

 found necessary to clamp the thin wire between two grooves 

 3*81 cm. long, cut on the copper carriage /3, and to keep it 

 so during the whole series of experiments to ensure accuracy 

 of the results. The thin wire must not be used for delicate 

 measurements of temperature by moving its sliding-carriage, 

 for the reason that on moving the copper carriage and 

 bringing it back to the same place, quite the same result 

 cannot be obtained, On the two thick wires only we are 

 able to investigate all problems concerning freezing-point 

 depressions or elevations of boiling-points, &c, to the limits 

 of 10°, and during the whole series of observations we need 

 not change any plug, or move the sliding-carriage on the 

 thin wire, or make new contacts. Thus we ensure that no 

 errors arising from these sources can possibly affect the results. 



