3G2 Le Maistre's Method of making a fine Purple Colour [Nov. 



finest shade, and that the more concentrated the solution of the 

 alumina was the nearer did the shade approach to purple. 



The oxide of gold thus precipitated with the earths combines 

 with them in different manners, according as the solutions are 

 more or less diluted with water. When the sulphate of alumina 

 is dissolved in a great deal of water, the precipitate of gold and 

 alumina becomes bluish, and sometimes rose red, when dried; 

 and when exposed to the fire, produces only a violet colour. If, 

 on the other hand, we dissolve alum in as little water as possible, 

 the auroaluminous precipitate is yellowish, and becomes purple 

 when exposed to the action of heat. I found in the course of 

 these experiments, that sulphate of barytes, when mixed with 

 the alumina, gave it a body, and increased the lustre of the 

 colour. In consequence, I finally adopted the following process : 

 One part of dry muriate of alumina, one part of sulphate of 

 magnesia, four parts of muriate of barytes, and five partb of car- 

 bonate of soda, are each pulverized separately. The pounded 

 salts are mixed in a glass mortar, and a little water is added, 

 merely enough to moisten the mixture. Then a diluted solution 

 of gold is added by little and little, pounding the matter all the 

 time in the mortar, till the whole has acquired a light sulphur 

 yellow tint, and the consistence of cream. The poundh:g is 

 continued a long time to produce the decomposition of the salts 

 with as little water as possible. When no more effervescence is 

 perceptible, and when the salts cease to creak under the pestle, 

 a sufficient quantity of water is to be added for the complete 

 solution of the salts. This tedious process is essential to unite 

 the oxide of gold with the earths, and the whole success of the 

 operation (which is pretty capricious) depends upon it. The 

 precipitate is to be left for 24 hours in the mortar, stirring it 

 from time to time with a glass rod. It is then to be poured into 

 a saucer, oi other similar vessel, and left till the powder subsides. 

 The supernatant liquid is then drawn off with a syphon. The 

 deposit is then dried in the shade without washing it. 



The specimen (o), which accompanies this letter, was made in 

 this way. I have employed it in painting the draperies of small 

 pictures. 



The precipitate, when dried, is yellowish-white. The muffle 

 in which it is to be baked ought to be red-hot. The powder is 

 put upon a silver or porcelain plate, of the thickness of one or 

 two lines ; and it must be withdrawn from the fire the instant 

 that it acquires its purple colour. If it be- left too long exposed 

 •to heat, it acquires a tinge of violet. This is occasioned by the 

 salts which it still contains ; for after it has been washed, it may 

 be kept red-hot without losing any of its colour, which indeed 

 acquires greater lustre. These trials were made on a small 

 scale, and are certainly susceptible of being improved, bv exa- 

 mining with more care the proportions of sails which should be 

 employed. Though this lake appears to want intensity, the 



