654 
PROFESSORS A. W. REINOLD AND A. W. RUCKER 
constitution of the films therefore cannot possibly have differed much from that of the 
liquids from which they were formed, and even if this were not so the changes in the 
refractive indices would have been too small to produce appreciable errors (see Phil. 
Trans., 1881, p. 485). When the films were thinning the field of view was traversed 
both by bands of colour due to the interference produced by the thin films, and by 
others due to the thick plates. To prevent confusion we propose to restrict the term 
interference bands to the first of these, and to call the second fringes. The instrument 
was so adjusted that the fringes were vertical and widely separated. Cross wires were 
introduced into the middle of the field of view, the compensator was placed in the 
vertical position and the central black fringe, exhibited when white light was used, 
was brought up to the vertical wire by slightly altering the orientation of the mirrors. 
A sheet of ruby glass was then interposed between the lamp and the mirrors, and the 
angles through which it was necessary to turn the compensator, to bring up to the 
vertical wire the fringes which were right and left of the central one, were measured. 
The mean of these values was taken as the angle corresponding to a wave length of 
red light, or according to our previous measurements to 615xlO~ G millims. (Phil. 
Trans., 1881, p. 454). 
The red glass having been removed, the “ zero ” or reading of the compensator when 
the central black fringe touched the vertical at its intersection with the horizontal 
cross wire was determined. It was then necessary to break a number of films in one 
of the tubes without in the least shaking or disturbing the apparatus. For this pur¬ 
pose one or two stout sewing needles had been enclosed with the films. A strong- 
electromagnet was now used to move these, and a known number of films having thus 
been broken, the displacement of the black line was determined. In breaking the 
films care was always taken to leave if possible two or three unbroken at each end of 
the tube. Disturbance by the irruption of air from the outside, if the contact between 
the ends of the tubes and the glass plates was imperfect, was thus prevented. 
If then 
T be the average thickness of a film, 
n the number of films broken, 
ji the refractive index of the liquid, 
\ the wave-length of red light, 
8 the angular displacement of the compensator necessary to restore the central 
fringe, after the rupture of the films, to the position it previously occupied; 
and 
a the angle (as above defined) corresponding to a wave-length of red light, 
we have the equation 
nf— l)T=-|x. 
