544 Intelligence and Miscellaneous Articles. 



The experiments lead therefore in all cases to conductivities 

 which are not greatly different. The smaller numbers are the 

 more important, for no circumstances can be conceived which make 

 the conducting power appear too small. 



From the time which elapses during and since the distillation 

 the numbers are in any case too large. An impartial consi- 

 deration would deduce from them for the conductivity of water 

 0-000000000025, or 1/40 -milliardth of that of mercury ; so that a 

 thread of water 1 mm. in length has the same resistance as a thread 

 of mercury of the same thickness which encircles the earth. 



According to this the resistance of an ohm is represented by a 

 layer of water with a cross section of 1 square millim. and a length 

 of about the 26 billionths of a millimetre. The " water- unit of 

 resistance" — a column of water a metre in length and a square 

 millimetre in cross section — has almost a resistance of 4-10 10 

 ohms. To produce the same resistance a copper wire a millimetre 

 square must have a length of 24*10 8 kilometres, a distance which 

 light would traverse in about 2'2 hours. If a semicircular elec- 

 trode a metre in diameter were sunk in the surface of a large mass 

 of water, the resistance would amount to about 12,000 ohms. 



A body with so small a conductivity may in many cases be 

 considered a non-conductor for voltaic electricity. 



Distillation in vacuo, as described above, has had the very 

 satisfactory result of leading by a far simpler method to a con- 

 ductivity almost one third of that previous^ found, or, as we may 

 say with some justification, to a water three times as pure. 



All that can be maintained with certainty is that the conduc- 

 tivity given above is again an upper limit. — Berichte der Akad. der 

 Wissenschaften zu Berlin, Oct. 23, 1884. 



THEIE DEPENDENCE ON TEMPERATURE. BY O. SCHUMANN. 

 The conclusions drawn by the author from a long series of 

 experiments are as follows : — 



1. Maxwell's formula gives, for a different arrangement of the 

 experiments, values for the coefficient at friction which show greater 

 deviations from each other than would correspond to errors of 

 observation, and this is particularly the case with high temperatures. 



2. By introducing a correction into Maxwell's formula, numbers 

 are obtained which, at ordinary temperature, exhibit a close agree- 

 ment with the method of transpiration. 



3. Owing to absorption, the method of transpiration gives too 

 high values of the coefficient of friction. With vapours the values 

 are too small for the same reason. 



4. The dependence of the coefficient of friction on the tempera- 

 ture increases with the temperature. 



5. The coefficients of all the vapours examined by me have almost 

 the same function of temperature. 



6. The relation found by Puluj to exist between length of path 

 and refractive index holds for the vapours of homologous ethers at 

 corresponding temperatures. — Wiedemann's Annalen, Nov. 1884. 



