2GJ: Prof. 1\. ^X. Wood on tJie Electrical Resonance • 



waves, or a resonance within the molecule, as in the case of 

 the aniline dyes. Though I have not yet devised any crucial 

 exi)eriment which enables me to distinguish between the two, 

 I have observed ditt'erences in the behaviour of these metal 

 films from that of films formed of dye-stuffs, which makes it 

 seem extremely probable that the action is different in the 

 two cases. 



To these arguments may be added the one that it seems 

 unlikely that so many different metals should exhibit such 

 similar molecular absorption, and that the granular condition 

 should be essential to the production of colour, for I have 

 succeeded in detecting the granules in every case by means 

 of the microscope. 



Let us now consider some of the differences between the 

 action on light of these coloured films and films formed of 

 aniline dyes. A film of granular silver can be formed which 

 resembles very closely a film of cyanine, both having absorp- 

 tion-bands in the middle of the visible spectrum. If a very 

 thin film of silver is prepared which exhibits a light purple 

 colour by transmitted light, the colour is much less marked, 

 /. €. the absorption is less, when the plate is held at an angle 

 of say 30 degrees with the incident light than it is at normal 

 incidence. Moreover, the colour changes from purple to red, 

 the absorption-band appearing to move down the spectrum a 

 trifle. Possibly a simple broadening of the band would 

 produce this same colour change owing to the greater sensi- 

 tiveness of the eye for red. A cyanine film does not exhibit 

 this peculiarity, the colour of the transmitted light remaining 

 nearly constant for varying angles of incidence. 



The change in depth and hue of the colour when the inci- 

 dence is made oblique is probably due to the fact that the 

 absorption is different when the incident light has a com- 

 ponent of the electric vector perpendicular to the surface, 

 which^ of course, is not the case when the incidence is normal. 

 If the light is polarized to start with, this is found to be the 

 case, the colour of the transmitted light being bluish-purple 

 when the electric vector is perpendicular to the plane of 

 incidence, and red when it is parallel to this plane in which 

 case there is a component normal to the surface. These 

 changes are not obtained with the film of cyanine. In the 

 course of a recent discussion of the problem which I have 

 had with Professor J. J. Thomson, he expressed the opinion 

 that the capacity of a sheet of resonators would be greater, 

 and the period of vibration consequently longer, when the 

 electric vector was parallel to the sheet. This is in agreement 



