484 M. Carey Lea — Allotrojpic Silver. 



with a bright brick red color. The difference between this 

 and the true solution as originally formed is extremely well 

 marked. 



Dextrine is a very variable substance and different speci- 

 mens act very differently. Common brown dextrine seems to 

 do better than the purified forms. 



Convenient proportions are as follows : in two liters of water 

 forty grams of sodium hydroxide may be dissolved and an 

 equal quantity of dextrine, filtering if necessary. Twenty- 

 eight grams of silver nitrate are to be dissolved in a small 

 quantity of water and added by degrees at intervals. Com- 

 plete solution readily takes place. Although the liquid con- 

 tains less than one per cent of metallic silver it appears 

 absolutely black, when diluted, red, by great dilution yellowish. 

 With some specimens of dextrine the solution remains clear, 

 with others it soon becomes a little turbid. 



Perhaps the most interesting reaction which this solution 

 shows, is that with disodic phosphate. A little phosphate is 

 sufficient to throw down the whole of the silver although both 

 solutions are alkaline. When a gram of phosphate in solution 

 is added to 100 c.c. of silver solution the color becomes bright 

 red, sometimes scarlet, and the whole of the silver is presently 

 precipitated. This precipitate on the filter has a color like that 

 of ruby copper, which color it retains during the first washing, 

 but after a few hours' washing with distilled water the color 

 changes to a deep Nile green and at the same time it becomes 

 slightly soluble, giving a port wine colored solution. With 

 more washing this solubility may disappear. 



It is a general fact that all these forms of silver, however 

 various their color, have both a body and a surface color and 

 these two colors tend always to be complementary. The body 

 color is that shown by the precipitate while still moist ; it is 

 also visible when a thin coat is brushed over paper, a coat so 

 thin that light passes through it, is reflected by the paper and 

 returned again through the film. But when a thick and opaque 

 film is applied, the body color disappears and only the comple- 

 mentary surface color is visible. 



So in the case of the precipitate by phosphate, when the 

 substance resembling ruby copper is spread thickly on paper it 

 dries with a bright green metallic surface color. But when 

 the substance itself becomes green by continued washing it 

 assumes on drying a dark gold or copper color, the surface 

 color changing with the body color and maintaining its com- 

 plementary relation. In observing these effects one is con- 

 stantly reminded of certain coal tar colors, both by the great 

 intensity of coloration and by the metallic surface color. I am 



