THE ELECTRICAL RESISTANCE OF THIN LIQUID FILMS. 
451 
The refractive index no doubt varied from film to film, as the composition of the 
liquid from which they were formed was often considerably altered by the absorption 
or evaporation of water. As a knowledge of the amount of water thus gained or lost 
was one of the objects in view, it was necessary to assume as a first approximation 
that the refractive index was constant. The method of testing and improving upon 
this assumption will be explained hereafter. 
IV. — Revision of Newton’s Table of Colours. 
The determination of the thickness of the films involved a knowledge of (in addition 
to the refractive index) (l) the angle of incidence, (2) the colour of the film, (3) 
the thickness of a plate of ah' exhibiting that colour when illuminated by light at 
normal incidence. The method of measuring the angle of incidence will be described 
when an account is given of the apparatus employed (see p. 461). It is sufficient to 
say here that the films experimented upon were in the form of cylinders with 
vertical axes. Light was reflected upon them from a fixed mirror ; a brightly 
illuminated vertical band was thus produced, and the thickness at any point was 
determined by the colour exhibited by this band. To increase and test the accuracy 
of the observations, two mirrors and therefore two beams of light, incident at 
different angles, were used. 
The apparent thickness of a film is defined to be the thickness of a plate of air 
which shows, when illuminated at normal incidence, the same colour as that displayed 
by the more obliquely illuminated of the two bands. A good deal of labour was 
saved in the calculations by the use of this quantity instead of the real thickness, 
which could of course be readily deduced from it. 
Newton, in describing the rings which bear his name, gives two lists of colours. 
In the first* seventeen colours are enumerated in the first four orders. In the second, 
which is reproduced in Watts’ ‘Dictionary of Chemistry’ (Art. Light), this number is 
increased to twenty-six. In this more extended list many tints are included which 
shade into those nearest to them by gradations too subtle to be readily appreciated; 
and since for purposes of measurement it is of little use to retain the names of colours 
unless the eye is able to distinguish with some approach to certainty where they begin 
and end, the following list, but little more extended than Newton’s first, was em¬ 
ployed. The colours in brackets are those which are not included in Newton’s first 
list. The blue of the first order is omitted. 
First order: Black, white, yellow, [orange], red. 
Second order : Voilet, blue, green, yellow, [orange], red. 
Third order: Purple, blue, green, yellow, red, [bluish red]. 
Fo urth order : Green, [yellowish green], red. 
* Newton’s ‘Opticks,’ 3rd Edition, London, 1721, pp. 174 and 206. 
MDCCCLXXXI. 3 N 
