o06 SIR DAVID BREWSTER ON THE FIGURES OF 



elliptical film is about four times greater than its minor axis ; but if we in- 

 crease the angle of the planes, the minor axis will gradually increase till it 

 becomes equal to the major axis, the oval plane becoming rectangular when the 

 planes are inclined 180°, a position which cannot be experimentally obtained. 

 While this change is going on, the four curvilineal films, to which the sides of the 

 elliptical plane are attached, are gradually diminishing, and disappear at 180°. 

 During this expansion of the elliptical film, it is not stretched and made thinner, 

 because it appropriates to itself the fluid of the four curvilinear films, which at 

 180° it extinguishes. If we now diminish the angle of the rectangles, the enlarged 

 oval plane will gradually become more elliptical, giving back its fluid to the four 

 curvilineal planes, till at 90° the plane film resumes its normal size. 



If, when in this position, we diminish the angle of the rectangular planes, the 

 minor axis of the elliptical plane will gradually diminish. At 45° it will become 

 a straight line and disappear. The elliptical plane will start into the angle of 

 135°, pushing towards their wires the four curvilineal films, placing itself be- 

 tween them, and in an enlarged state appropriating a portion of their fluid. 



Remarkable as these phenomena are, there is one still more remarkable, which 

 requires the testimony of the eye to make it credible. If in the normal or rectan- 

 gular position of the rectangles we blow upon the oval film, or between the 

 curvilineal ones, a bubble of the proper size, it will replace the system of films 

 with a hollow curvilineal cube, the sides of which will project beyond the faces of 

 the vertical cube, which, having plane faces, would not project beyond the wires. 

 Within the four triangular spaces at the upper side of this cube, will be four 

 summits where the black spot of the first order in Newton's scale will be produced, 

 and at which the bubble will burst. If we now hold the wires vertically, the 

 cubical bubble will burst, and the system of liquid films which it expelled will 

 reappear, as if it had left its ghost behind it, to recover the elements which the bubble 

 had appropriated I When the wires were held horizontally this resurrection of 

 the normal system of films did not take place, owing, I presume, to the bubble 

 not bursting symmetrically. The same results will be obtained when the incli- 

 nation of the rectangles is above 90°. 



If we introduce the bubble into the system of wires when empty, the system of 

 liquid films is equally reproduced ; but the experiment succeeds better when the 

 bubble is laid upon the oval film, as it thus appropriates the fluid of the different 

 films, and when it bursts there is a greater quantity of fluid for their re-forma- 

 tion. For the same reason, the reproduction of the films is produced if the bubble 

 is burst when it is strongest, by thrusting into it a piece of blotting-paper. 



The same restoration of the figures of equilibrium, produced in the tetrahedral 

 and quadrangular systems shown in Figs. 6, 7, 8, 9, and 10, may be effected by 

 introducing bubbles of the proper size, whether they are empty or occupied by 

 their respective films. 



