September 23, 1897] 



NATURE 



505 



difference between the phase-changes in these two cases, but gave 

 no direct information as to the absolute magnitude of either. 



By depositing a wedge-shaped silver film on the outside of 

 one of the thick glass mirrors of a Jamin refractometer, and 

 observing the interference bands where they crossed from the 

 glass-air to the glass-silver surface, Quincke concluded that the 

 phase-change, when light is reflected in glass from a silver 

 surface, depends on the thickness of the silver, but was unable 

 to decide whether it was a quarter-wave acceleration or three- 

 quarters of a wave retardation for a thick silver film. The 

 reason of his uncertainty will be explained later. Wernicke 

 and Wiener analysed spectroscopically the light reflected from 

 thin transparent plates, the back surfaces of which were partly 

 silvered. The spectra obtained were crossed with black bands, 

 depending in number and breadth on the thickness of the plate ; 

 and these bands were displaced where the light was reflected 

 from the' silver surface. Wernicke, however, concluded that 

 the phase-change amounted to a quarter-wave acceleration, 

 whilst Wiener concluded that it was of the nature of a retard- 

 ation of three-quarters of a wave-length. Wernicke has since 

 stated that silver films could be obtained which would produce 

 either of these phase-changes, according to the nature of the film. 



A modification of Michelson's refractometer may be advan- 

 tageously used to study this vexed question. Light from a 

 lamp L, placed at the principal focus of a lens M, falls on the 

 thinly-silvered mirror A, part being reflected along the path 



through a funnel reaching to the bottom of the beaker. The 

 water was simply displaced upwards, and a few minutes after 

 the silvering solution had reached the level of the top of 

 the plate the stop-cock of the syphon was cautiously 

 opened, so as to slowly withdraw the silvering solution. The 

 flow should at first be moderately quick, but should decrease 

 later. Silver films will be found deposited on both sides of the 

 glass, that on the under side being the better ; the film deposited 

 downwards is generally very milky in appearance, and is fre- 

 quently spotted. If the glass is not washed with nitric acid 

 after the elastic band is placed round it, the silver will be 

 found to gradually shade off towards the clear glass. 



The following silvering solution may prove useful to those who 

 wish to make half-silvered mirrors ; it was used to form the 

 silver wedges on account of the slowness of its action. 



Take silver nitrate i gr. , dissolved in 20 c.c. of distilled water. 

 Add strong ammonia ("88), drop by drop, till the precipitate 

 formed is just re-dissolved. Add a solution of I "5 gr. caustio 

 potash (ordinary stick potash works well), dissolved in 40 c.c. of 

 water ; then ammonia, drop by drop, till precipitate is just re- 

 dissolved. Add 80 c.c. distilled water, and then a solution of 

 silver nitrate (strength unimportant), till a permanent precipitate 

 is just formed. Make the solution up to 300 c.c. I '8 grs. 

 milk sugar are dissolved with heating m 20 c.c. distilled water. 

 This solution is added to the above just before silvering is to 

 commence ; after a few minutes the whole will commence to 



A c D, and after reflection from the back surface of D, pursuing 

 the path D c A E, to the eye of the observer. The light originally 

 transmitted through the mirror A pursues the path A F A E. If 

 the two paths are equalised, brilliant interference bands are 

 ]>roduced.- It will be noticed that these bands are virtually 

 formed by the reflection of light at an air film, comprised be- 

 tween D and the image of F in A. Consequently both they and 

 the surface of d can generally be focussed simultaneously. 



Both I) and f are silvered on their back surfaces ; the film on 

 V is uniform, whilst that on D increases in thickness from one 

 side to the other, as indicated (with exaggeration) by the black 

 line a h (Fig. i). A horizontal strip of glass is also left un- 

 silvered across the middle of D. 



The method of obtaining the silver wedge was as follows. 

 The glass was first well cleaned with strong nitric acid, using a 

 small mop made of cotton wool plugged into the end of a glass 

 tube. An ordinary elastic band was then stretched round 

 the plate, and the whole was placed, inclined at an angle of 

 about 30", in a beaker, and distilled water poured in till it 

 reached the height of the top of the glass plate. A glass 

 syphon tube, reaching to the bottom of the beaker, was intro- 

 duced, the flow being capable of regulation by a stop-cock on 

 the outer limb. Silvering solution was then quickly poured in 



- For a simple method of constructing this apparatus, see N; 

 August 17, 1893, " Apparatus illustratine Michelson's method of ob 



lATURE, 

 abtaining 

 interference bands." Followinz Michelson, it was there st.ited that the 

 leiitral band is alwavs black. We have since determined that the colour of 

 the central band will vary according to the phase-changes produced at the 

 v.arious silver surfaces. We have been able to obtain a white central band. 



NO. 1456, VOL. 56] 



Fig. 2. —The thin edjj'^ of the wedge is to the left. A displacement of 

 the bands in the direction of the arrow is produced by lengthening 

 the path a c D(Fig. i). 



blacken. The glass to be silvered may be placed horizontally in 

 an ordinary evaporating basin, and the solution poured in till it 

 reaches the upper surface of the glass. The silver is deposited 

 upwards. For half-silvering about ten minutes is required, when 

 the temperature is about 15° C. For thick silvering the plate 

 should be left at least an hour ; we have frequently left it all 

 night with good results. The thick silver film produced may 

 be scrubbed with cotton wool under running water, and finally 

 left for some time in distilled water, when it may be removed 

 and left to dry. 



The above recipe for silvering was given to us some years 

 ago by Prof. Boys. It was originally due, we believe, to Prof. 

 Liveing. 



Fig. 2 is taken from a photograph of the bands produced, 

 using yellow light (the light from an electric arc passed through 

 a cell, made of signal green glass, filled with a saturated solution 

 of bichromate of potash) and a Cadett plate ; exposure, one 

 minute. 



The dark band across the figure represents the strip of clear 

 glass on which silver was not deposited. The bands, in passing 

 from the glass-air to the glass-silver surface are seen to be dis- 

 placed towards the thin edge of the wedge-shaped film, and it is 

 at once seen that the displacement depends on the thickness of 

 the film. In order to produce a displacement of the bands in 

 the same direction, the length of the path A D must be increased. 

 This, according to Wiener, would indicate that the light was 

 retarded on reflection from thick silver by three-quarters of a 

 wave-length. It will be noticed that the bands are continuous, 

 passing from the glass-air to the glass-silver surface. This is 



