152 Mr. R. Appleyard on Dielectrics. 



Results. 



The results of the tests are given in Table I., which requires 

 little explanation. The u Capacity/' " Resistance," and " Elec- 

 trification " of paraffin-paper and mica are there tabulated at 

 the various temperatures. 



Capacity. — As standards of capacity two condensers were 

 used. These were mica of 1 and 0*5 microfarad capacity 

 respectively. They were submerged in water and kept always 

 at the same temperature, 27° C, which is just above the 

 maximum air temperature. The object of using two standards 

 was to avoid shunts and to simplify the wofk. 



In the paraffin-paper condensers there appears to be an 

 irregular variation of capacity with temperature. It is too 

 small to affect them in practical telegraphy; but it is suffi- 

 ciently large to prohibit their use as standards. We must 

 suppose that the softening of the wax, and the consequent 

 approach or recession of the tinfoil sheets, is responsible for 

 part of these changes, to which must be added the variation of 

 the specific inductive capacity of the dielectric. It is curious 

 that with paraffin-paper condensers the capacity diminishes 

 as the resistance diminishes. With gutta-percha or india- 

 rubber core the result is nearly always the reverse of this. 



Owing to the vibration of the building in which the tests 

 were made, the capacities could not be measured with a greater 

 accuracy than 1 per cent. ; but within this limit it is safe to 

 say that throughout the whole range of temperature mica 

 retains a constant capacity. 



Temperature Coefficients. 



It is usual to assume * that the resistance R T of a dielectric 

 at any temperature r is represented by an equation of the form 



R T =Ra T , 



and that similarly, at some other temperature 0, 



Rfl = Ra*; 

 so that 



R T =R ea T -*, 

 or 



logR T =logR<,+ (T-0)log« (1) 



From this equation, taking the fifteen observations of Table I. 

 two at a time, in alternate order, and putting them succes- 



* See Appendix II. 



