44 On a Species of Plumbago from Africa, 



With a view to delermine the quantity of iron the ore 

 contains, the following expcrinient was made. 



Third Experiinent. 8 grains of the ore were mixed with 

 about 70 erains of oxymuriate of potash (previously fused) 

 and heateu in a platina crucible. At a dull red heat the mix- 

 ture became an ignited mass : the heat was increased until 

 all the oxvgcne was expelled from the salt. The ore was 

 not much affected by this treatment. About 150 grains of 

 nitre were now introduced, and the heat raised to a strong 

 red : after half an hour the crucible was examined, a quan- 

 tity of the ore slill remained. A fresh portion of nitre was 

 added, and the heat raised and continued for about two 

 hours, during which time fresh portions of nitre were oc- 

 casionally introduced. 



The crucible was ^low examined, and exhibited to the 

 eye no remains of the ore. The solid matter was treated 

 with diluted test muriatic acid, and after being digested for 

 some time at a moderate heat, it wholly dissolved except 

 a quarter of a grain, which proved to be principally silex 

 with a little dust from the charcoal. Theacid solution was 

 treated with a slight excess of annnonia, and after a short 

 time a whiliLdi flocculent precipitate appeared. It was col- 

 lected on a filter, washed, and heated to a strono; red in a 

 platina crucible. It hardened in the fire, and had now the 

 characters of alumine coloured by iron. It was bruised to 

 powder and digested in diluted test muriatic acid ; the 

 greater part of it remained insoluble, and was principally 

 alumine with a little silex. The acid solution was now 

 treated with prussiate of potash, and after some time the 

 prussiate of iron was collected on a filter, washed and dried. 

 As no ascertainable quantity of it could be separated from 

 the paper, the filter was burnt in a platina crucible, and a 

 red heat continued until all the carbonaceous matter was 

 consumed. The oxide of iron obtained, weighed a quarter 

 of a grain, and was partly attracted by the magnet. 



This experiment affords onlv an imperfect approximation 

 as to the real quant'ty of iron in the ore. If the iron be 

 considered as existing in the ore in the metallic state, it can 

 scarcely be estimated at more than two or three per cent. 



The preceding experiments furnished distinct evidences 

 a? to the nature of this substance, that it is a species of 

 plumbago, and consists almost entirely of carbonaceous 

 matter. In its general characters, it seems to bear a strik- 

 ing resemblance to the artificial products of some iron fur- 

 naces in the East IndieSj ivell known by the name of Kish 



iron. 



