340 DIRECT PROPERTIES OF MOLECULES 122 



if the number were 1,000 we should have 0-0005 / 1000 mm. 

 for the distance between adjacent particles ; and this agrees 

 fully with the value determined by the theory of gases. 

 A far better estimation, which rests on similar assump- 

 tions, is the calculation from the refractive index of a gas, 

 which has already been mentioned in 118. 



In addition to the method, described also in 118, of 

 calculating the size of molecules from the values of the 

 dielectric capacity, there are also other ways of deter- 

 mining it by electrical measurements. For this we can 

 make use of considerations which compare the energy of 

 the electrical forces brought into play with other kinds of 

 energy, just, indeed, as those we have mentioned in respect 

 of the capillary forces. By comparing his measurements of 

 the force of attraction between a pair of plates of zinc and 

 copper, which are electrified by contact with each other, 

 with the equivalent amount of heat, Lord Kelvin 1 finds 

 that it cannot be possible to make plates of these metals, 

 whose thickness is 30 times less than a millionth of a 

 millimetre. The atoms of zinc and copper, therefore, like 

 the molecules of air, have a size that is measurable in 

 millionths of millimetres. L. Lorenz 2 also makes use of 

 the electrical potential energy of a zinc-copper element and 

 the value of the energy necessary for the electrolysis of 

 water, and from these data he draws the conclusion that 

 the mean distance apart of two molecules of water in the 

 liquid state must be at least 10 times less than a millionth 

 of a millimetre. Oberbeck, 3 who investigated experiment- 

 ally the electromotive forces of thin layers, found that the 

 molecular forces of platinum are perceptible through layers 

 of other metals of from 1 to 2 millionths of a millimetre 

 in thickness. 



If these different methods that have been used to obtain 

 a limit of the divisibility of matter do not all lead to the 

 same value for the size of the particles, they yet agree, 

 without exception, in giving the thickness of an atom or 



1 Proc. Roy. Soc. 1860 ; Nature, 1870, i. p 551, ii. p. 56. 



2 Poog. Ann. 1870, cxl. p. 644. 



3 Wie'd. Ann. 1887, xxxi. p. 337. 



