( 114) 



The value of the merrmy lesistance used was 172,7 £1 in the 

 li(|iüd condition at 0°C. ; extrapolation from the melting point to 

 0°(J. by means of the temperature coefficient of solid mercury gives a 

 resistance corresponding to this of 39.7 ii in the solid state. At 

 4°.3 K. this liad sunk to 0.084 Ü that is, to 0.0021 times the resis- 

 tance which the solid mercury would have at 0° C. At 3° K. the 

 resistance was found to lia\ e fallen below 3 X 10~^ <2, that is to 

 one ten-millionth of tlie value which it woidd have at 0° C. As the 

 temperature sank further to 1°.5 K. this value remained the upper 

 limit of the resistance. 



The next step was obviously to look for the point at which the 

 resistance tirst becomes measurable as the temperature is raised. The 

 temperature of this point was found to be slightly more than 4°. 2 K. 

 at which the resistance was found to be 230 micro-ohms or one 

 hundred thousandth of the resistance (solid) at 0° C. As the tempera- 

 ture was raised to that of the boiling point (4°. 3 K.), the resistance 

 rose once more to 0.084 fi. This change took place more quickly 

 than the rate of change to which the formula given in the December 

 (February) Communication leads — exactly how much more quickly 

 is not yet known but it certainly seems to be increased very much 

 more rapidly. A point of inflection which does not appear in the 

 formula given — a formula which I regarded as incomplete also 

 on account of the method by which it was deduced, seems to occur 

 between the melting point of hydrogen and the Itoiling point of helium 

 in the curve which represents the resistance as a function of T. 



The more the up|)er limit which can be ascribed to the i-esistance 

 remaining at helium temperatures decreases, the more important 

 becomes the ol)served phenomenon that the resistance becomes prac- 

 tically zero. When the specific resistance of a circuit l)ecomes a 

 million times smaller than that of the best conductors at ordinary 

 temperatures it will, in the majority of cases, be just as if electrical 

 resistance no longer existed ujider those conditions. If conductors 

 could be obtained which could be regarded as being devoid of resis- 

 tance as long as their cross section was not excessively small, or 

 conductors of the smallest possible sections, either cylindrical with 

 diameters of the order of the wave length of light, or films of mole- 

 cular dimensions, whose resistance would be but small, if there had 

 no more to be reckoned with the Joule development of heat in 

 increasing the current in a bobbin to exceedingly high values, because 

 the development of heat in a circuit of constant current strength could 

 be made extremely small compared with the latent heat of vaporization 

 of the liquid which can be used for coohng, — then further experiments 



