at Lemington Priors. 259 



Quantity hardly fufficient to precipitate. After boiling there 

 is no decompofition ; nor is there any if the liquor be eva- 

 porated to half its original bulk. 5. Tincture of galls ftrikes 

 a purple colour before the water is boiled : after boiling 

 there is a precipitate like wife, but of a dull lemon colour } 

 after a partial evaporation the colour approaches to white- 

 nefs. In each of the two laft cafes, the liquor gradually ac- 

 quires a deep yellow tinge. In thefe experiments, we have 

 none of the ordinary figns of iron, except the carbonat (11.), 

 though its exiftence in the water has been proved beyond 

 queftion (3). . 



V. MANGANESE IS DISSOLVE*) IN THIS WATER. 



As zinc is known to have a ftroriger affinity to acids than 

 the other metals have, I hoped by its mean to obtain fome 

 information on the caufe of the facts I have defcribed. 1. I 

 boiled, therefore, fome of the water in a glazed vefTel, in 

 contact with fome pieces of zink : a fmall precipitate was 

 formed, but enough to be collected for examination with 

 the blow-pipe. This was fufed with borax ; and a globule 

 was formed of a rich red colour, precifely that communi- 

 cated by manganeft*. Continuing the fufion, the colour 

 vanifhed ; nor could I make it re-appear by the yellow 

 flame of the blaft : probably becaufe the manganefe was 

 friixed with another metal. The globule was removed to a 

 filver fpoon ; and, then, by fufion, it regained the red co- 

 lour. To avoid the poffibility of error, a little manganefe, 

 fufed with borax in the fame manner, was placed befide the 

 red globule ; and no difference could be obfervcd in the co- 

 lour of the two globules, except a flight variety in the in- 

 tenfity \. %. The Urnon-coloured precipitate, formed by the 



tincture 



* For the properties of this mineral fee Scheele's Eflay on Manganefe j 

 cr Bergman on the White Ores of Iron. 



t I am aware of the obfervation of Bergman, that zink does not preci- 

 pitate ibtutions of manganc(c. Sec his EfTays, to!, lii. p. 414* But, be- 

 S 2 fides 



