Prof. J. Perry on Telephone Circuits. 673 



in sea-water is small, and it has been difficult to explain these 

 high values. They may be due to the comparative absence 

 of physical impurities in the air over the sea. In the present 

 series of experiments with artificially ionized air, the highest 

 values were found when the air had been well cleaned by a 

 heavy fall of rain. 



3. When smoke is mixed with the air current it may be 

 shown that charged centres are present in numbers about 

 equal, even when radium is not present. These smoke ions 

 have feeble mobility, and a strong electric force is necessary 

 to detect them in the testing vessel. 



4. The diminution of the ionization current by smoke 

 (Owens' experiment) is not due to the more rapid recombina- 

 tion of small ions, but to the fact that the small ions tend to 

 disappear, and that they are replaced by large ions which 

 move and recombine more slowly. 



In the case of a cloud of smoke or fumes containing 

 charged centres, the number of these centres does not appear 

 to be greatly modified by the action of intense y rays. Whilst 

 some become neutralized, others acquire a charge, so that the 

 number is to a large extent independent of the intensity of 

 the radiation. 



5. The presence of dust, smoke, mist, or other centres, 

 charged or neutral, in the air, causes a transformation from 

 small to large ions. In this way the total number of ions 

 present may be increased, whilst the conductivity is 

 diminished. In any region of air where a charge of one 

 kind is predominant, the effect of the presence of centres, 

 neutral or charged, is to increase and accentuate the excess. 

 This tendency must have an important influence in the 

 variation of the potential gradient and on the production of 

 thunderstorms. 



Montreal, 



January, 1910. 



LXXII. Telephone Circuits. 



By Professor John Perry, F.R.S* 



ABOUT a year ago Mr. Sidney Brown told me that he 

 wished to place contrivances at equal distances in 

 telegraphic or telephonic circuits and that it was necessary 

 to make calculations, beforehand, because in a telephone line, 

 say, of three hundred miles or more with similar contrivances 

 inserted every one or two miles, it was practically impossible 



* Communicated by the Physical Society : read February 25, 1910. 

 Phil. Mag. S. 6. Vol. 19. No. 113. May 1910. 2 X 



