492 SIR DAVID BREWSTER ON THE COLOURS OF THE SOAP-BUBBLE. 



As the colours of the soap-bubble cannot be well studied on the bubble itself 

 even when defended from the action of the air by " a clear glass," as Newton 

 did, I have employed plane, convex, and concave films of all sizes, up to 3£ 

 inches in diameter, as obtained upon glasses, conical or cylindrical, or upon 

 tubes of glass or metal closed at one end, or open at both. I have also employed 

 wires of different metals bent into various shapes, rectangular, triangular, and 

 curvilinear, but in the following experiments I have principally made use of 

 a cylindrical wine-glass, which gives a film 1^ inch in diameter, and unless 

 otherwise stated, I have obtained the films from the common solution of soap and 

 water, sufficiently strong to give an ordinary bubble. 



I. Phenomena of Colour in a Vertical Plane Film. 



1. When the film has been newly formed and held vertically for a second or 

 two, it exhibits at its apex six, seven, or eight horizontal bands of colour, the 

 uppermost white and orange-red, with a little black above them, being the first 

 order of Newton's scale. In a few seconds the different orders increase in width, 

 descending, and becoming in succession, as shown in Figs. 1-6, eight, seven, six, 

 four, three, &c, in number, the film when unprotected generally bursting when 

 four or five bands cover the whole of it, as shown in Figs. 4 and 5. Previous to 

 bursting, the black portion of the first order of colours has appeared at the 

 apex, as shown in Fig. 3. 



In order to observe the changes in the state of the bands, after their number is 

 reduced to five or four, we must protect the film by a watch-glass, or a piece of 

 plate glass. When this is done, the eighth, seventh, and sixth orders of colours 

 gradually disappear, being succeeded by the fifth, fourth, third, and second orders, 

 which, along with the first, cover the film. After a while the third and second 

 orders disappear, and the film is covered, as in Fig. 7, with the black and white of 

 the first order. In a few minutes the white of the first order disappears, and 

 the black band covers the whole film. Before the black film is complete, it often 

 advances with an irregular margin, and throws out filaments into the white 

 band, as shown in Figs. 5 and 6. It very frequently includes also minute systems 

 of rings of different colours, and portions of the white of the first order of all 

 shapes, having within them, in constant motion, small portions of the black 

 matter. These white portions gradually disappear, leaving behind them the most 

 beautiful silvery dendritic forms, which move about till the film bursts. 



When the black band is formed from a disturbed condition of the coloured 

 bands, where a great number of separate black portions are slowly united, there 

 may be seen a portion of the black space much darker than the rest, with a 

 beautiful margin of white spots, and accompanied with one or two circular spaces 

 of equal blackness, and surrounded also with white spots, so small as to require a 

 lens to see them. This phenomenon is shown in Fig. 7. The deep black colour- 



