604 Mr. J. Viriamu Jones [May 24, 



piece of platinum, the mean solar second on particular bodies of the 

 solar system. 



The time will come when we shall be able to take another step 

 forward, when we shall be in a position to make the unit of mass a 

 derived unit with reference to the units of length and time as 

 fundamental through that universal uniformity of nature — the law of 

 gravitation ; the units of length and time being based on some more 

 permanent phenomena of the universe (the properties of the sodium 

 molecule have, for instance, been suggested) than the length of a 

 particular bar, and the motion of the earth on which we live. 



But we cannot take this step forward yet. We do not know 

 either the gravitation constant or the velocity of light with sufficient 

 accuracy to enable us to define by means of them units precise enough 

 to meet the needs of practical life. Yet the fact that if we knew 

 these constants we could make an important advance in the realisation 

 of a system of absolute measurement is an excellent reason why tho 

 masters of physical measurement should apply themselves to thoir 

 precise determination. 



The use of a derived unit of mass would have this among other 

 advantages, it would rationalise the dimensions of the electric and 

 magnetic units in length and time, and relegate the irrationality to 

 the ether, to which I believe it properly belongs. 



We must, however, for the present be content with a less com- 

 plete reference to the most permanent phenomena of the universe, and 

 by absolute measurement to-night I mean measurement in terms of a 

 unit derived with reference to the unit of length, time, and mass as 

 fundamental — i. e. a unit conditioned only by these units, and the 

 properties of the ether. 



"There is such a unit of electrical resistance — a unit derived with 

 reference to the units of length and time as fundamental — the magni- 

 tude of which is independent of the properties of any particular 

 conductor, any particular coil of wire, and is conditioned only by the 

 units of length and time, and the properties of the ether. 



It may, indeed, at first sight seem astonishing that a quantity so 

 different can be expressed in terms of a length and a time. Yet so it 

 is. We can, in virtue of experiments made 63 years ago in this 

 Institution by Michael Faraday, measure the resistance of a con- 

 ductor with the help of no other standardising instruments than a 

 tape or measuring machine and a clock. An electrical resistance is 

 always proportional to a certain velocity, and if the magnetic permea- 

 bility of the ether be taken to be unity, the number expressing the 

 electrical resistance will be the same as the number expressing that 

 velocity ; the unit of electrical resistance may then be taken to be 

 the resistance corresponding to unit velocity — i. e. on the C.G.S. 

 system a velocity of 1 c. per second, and any electrical resistance 

 may be conveniently expressed in terms of this unit as so many 

 centimetres per second. 



[You will observe that I do not say electrical resistance is a 



