286 MR, J. C. MAXWELL GARNETT 



he prepared silver of less specific gravity than that of normal silver, by depositing 

 that metal on the platinum electrode of a platinum-zinc battery. He also prepared 

 silver in suspension in water by chemical means, observing the characteristic amber 

 colour and noticing that precipitation could be accelerated by the addition of salt to 

 the water. 



VOGEL concludes (loc. cit., p. 337) that there are three forms of silver, (1) regular 

 dendritic silver [crystalline] ; (2) granular powdery silver [small spheres] ; (3) mirror 

 silver [amorphous]. He found that the second type " tended to the formation of a 

 coloured powder," but could be changed into the third type by pressure. He adds 

 (loc. cit., p. 441) that the silver precipitated by photography is of the second type, 

 and this is the view suggested in the preceding memoir (p. 417), because of the red- 

 brown transmitted colour and the green colour of the reflection from fogged photo- 

 graphic films, which, according to the analysis given above, 10, are the colours 

 exhibited by films of amorphous or granular silver,* of less than standard density. 



15. Allotropic Forms of Metal. 



In the course of the preceding investigations we have been led to recognise that 

 variation of the relative position of the molecules of a metal will cause the metal to 

 change colour, whether it be examined by reflected or by transmitted light. It has 

 been shown, for example, that mere variation in density causes gold in one state to 

 transmit green light, in another blue, in another purple, and, in another again, ruby. 

 Further, this discovery has led us to the conclusionf that, in order to account for the 

 properties of CAREY LEA'S anomalous silvers, it is not necessary to assume the 

 existence of an " allotropic " molecule of silver. The question thus arises : Are there 

 any other cases in which an allotropic molecule has been unnecessarily postulated ? 



EGBERTS- AUSTEN^ has collected particulars of a large number of supposed cases of 

 allotropic states of metals. We proceed to the examination of these particulars in 

 order to determine whether the effects, for the explanation of which the allotropic 

 molecule was postulated, are not merely those which, according to the analysis of 

 10 above, would be due to a decrease in the density of the metal in a granular or 

 amorphous state. 



In the first place, then, the discovery that metals in different states, corresponding 

 to different methods of preparation, possessed different densities and had widely 

 different properties, although chemical analysis could detect no change in the 



* Cf. figs. 9, 11 and 13, and also p. 282 above, where the same colours, exhibited by one of E. W. WOOD'S 

 silver films, are discussed. 



t Above, p. 285. 



J ' Metallurgy,' pp. 87 el seq. 



ROBERTS- AUSTEN defines "allotropy" as follows (loc. cit., p. 89): "The occurrence of elements in 

 .... allotropic states means that .... the atoms are differently arranged in the molecules." 



