132 BRIDGMAN. 



was a second possibility of short circuit. The initial resistance of the 

 arrangement was of the order of one megohm. Pressure was not 

 pushed higher than 8000 kg., in order not to freeze the water. At this 

 pressure the resistance had dropped to 35000 ohms. There were large 

 polarization effects, and on releasing pressure the resistance did not 

 recover its initial high value. 



These experiments can only justify the conclusion, therefore, that 

 under high pressure iodine does not at any rate become metallic in 

 its conductivity, but the specific resistance remains high. It is quite 

 possible that the relative resistance may suffer large changes, but the 

 probability is small that the change of relative resistance is as high 

 as it is for black phosphorus. 



The iodine used for this experiment was Kahlbaum's, previously 

 dried in vacuum. The platinum electrodes were approximately 

 1 cm- each in area, and 3 mm. distance from each other. These 

 dimensions, together with the value of the minimum resistance 

 recorded above, allow a minimum value to be set for the specific 

 resistance at 8000 kg. of about 100,000 ohms per cm. cube. The 

 correct value is doubtless many fold greater. The specific resistance 

 of iodine under ordinary conditions seems too much affected by 

 impurities to allow of its accurate determination, and I have not been 

 able to find a value anywhere recorded. 



" Chromel a." This is an alloy for high temperature resistance 

 units essentially similar to the alloys known more familiarly under the 

 name of "Nichrome." "Chromel A" is made by the Hoskins Co. of 

 Detroit, and has the composition 80% nickel and 20% chromium. It 

 was furnished by the manufacturer in the form of a wire 0.005 inches 

 diameter, and was double silk covered by the New England Electrical 

 Works. I wound it for these measurements into a coreless toroid of 

 1 18 ohms resistance at 0°. It was seasoned for the measurements by 

 keeping it at 135° for four hours, and by a preliminary application of 

 2000 kg. 



The effect of pressure is in the normal direction, that is, the resist- 

 ance decreases with increasing pressure, but the effect is very small, 

 smaller than any which I have previously found. The maximum dis- 

 placement of the slider of the Carey Foster bridge was 4 cm., so that 

 the sensitiveness of the measurements was not greater than one part 

 in 40,0. Within the limits of error the relation between pressure and 

 resistance is linear to a maximum pressure of 12000 kg. At 0° the 

 two points at the highest pressures were irregular, probably because 

 of viscosity in the transmitting medium, and at 90° there were hystere- 

 sis effects amounting to 4% of the total pressure effect. 



