70 SCIENCE PROGRESS 



always the same — the existence of metallic sodium in the form 

 of small particles scattered through the mass. The yellowish 

 specimens are not resolved by the ultramicroscope, but the blue 

 varieties show well-defined particles, according to Cotton and 

 Mouton, of a brownish red colour. They are not arranged so 

 regularly as they are in the glasses already considered ; some- 

 times they form clusters or beads, at other times they are 

 arranged in straight lines, which lie along the cleavage planes. 

 The artificially coloured specimens act just like the natural ones, 

 and can be modified in the same ways. 



There is some evidence in favour of the hypothesis men- 

 tioned above, that the coloration is due to the presence of 

 particles of sodium. Thus, they are prepared by heating the 

 colourless crystals in sodium vapour; and further, the crystals 

 lose their colour when heated to a temperature near the melting- 

 point of sodium. It may, at first sight, appear strange that 

 sodium should be present, as such, in moist air. Siedentopf 

 supposes that each particle is surrounded by a protecting 

 pellicle, but Cotton and Mouton think that this is unnecessary, 

 and that the phenomenon is explained " by the extreme slowness 

 with which diffusion of the gas occurs in these ultramicro- 

 scopic openings." This is supported by some experiments of 

 R. W. Wood. When sodium vapour is condensed in a vacuum 

 tube, films are obtained which present very striking colours. 

 When the pressure rises the film changes, and with it the 

 colour. In Wood's experiments the vacuum tubes communicate 

 with the atmosphere by means of very long, fine capillary tubes. 

 It was found that the rate at which the pressure increased, as 

 shown by the change in colour of the film, was very slow, 

 indicating the slow rate of diffusion in such fine tubes. 



Liquids. — Having shown how the ultramicroscope has been 

 successfully applied to the study of solids, such as glasses 

 coloured by metals and coloured salts, we will pass on to the 

 consideration of liquids. Here a new experimental difficulty is 

 encountered. In the case of solids there are practically no 

 particles present in the substance except those to be examined. 

 It is very difficult, however, to free liquids from extraneous 

 particles. Ordinarily distilled water usually contains an ap- 

 preciable quantity of impurity, and if exposed to the air for 

 any length of time it becomes contaminated, and the number 

 of such particles may equal in number those to be observed. 



