96 THE SCIENTIFIC PAPERS OF 



sent the resistances of insulation. The broken line shows the cor- 

 responding temperature in degrees Fahrenheit, calculated from the 

 observed copper resistances ; the abscissae represent time. As might 

 be expected from the limits between which the temperature varied, 

 no material change was perceptible in the inductive capacity of 

 the material.* 



In order more effectually to guard against and to discover at 

 an early stage, an undue accumulation of heat in any portion of the 

 coil, the Government electricians placed between the coils, at 

 regular intervals, certain instruments which may be termed resist- 

 ance thermometers, based upon the principle, that the resistance of 

 copper wire varies in a constant ratio with the temperature. The 

 instrument consists of several layers of insulated wire of a known 

 resistance, at the standard temperature, protected by an iron case, 

 and brought into connection "with the testing apparatus. By simply 

 measuring the resistances of these coils, deposited at different 

 places between the layers of the cable, any rise or fall of tempera- 

 ture throughout its mass is easily ascertained, by comparison of the 

 resistance observed with the original standard. This system was 

 found extremely useful, when applied, in observing the generation 

 of heat in a portion of the Malta and Alexandria cable, then on 

 board the Queen Victoria, which had been kept dry during twelve 

 days, and had attained a temperature of above 80 Fahr. ; although 

 the ordinary mercury thermometers, suspended in different parts 

 of the hold of the ship, only indicated, during the same period, a 

 maximum temperature of 60. 



The diagram of copper resistances in Plate 7 exhibits the tests 

 applied to the copper conductor in the Gutta Percha Works ; the 

 ordinates representing the resistances, in Siemens units, per 1,000 

 yards, of the various coils contained in Cable 6. It will be seen, 

 from this example, that the differences in the specific conductivity 

 of the copper vary up to 16 per cent. 



Of the two sections under consideration, the Alexandria and 

 Benghazi cable invites some further comment. It was laid at two 



* For further information on this subject, see the detailed reports by Dr. Allen 

 Miller and Messrs. Siemens, Halske and Co., upon the phenomena of the spon- 

 taneous generation of heat in this portion of the cable, which are given in 

 Appendix No. 11 to the "Report of the Joint Committee on the Construction of 

 Submarine Telegraphs." 



