The minerals that contain most arsenic are the white pyrites, orpiment, and cobalt. We have already considered the white pyrites: as to orpiment, it consists of sulphur and arsenic. Both these substances being very volatile, it is difficult to separate them by sublimation: yet, with proper management, and a due regulation of the fire, this separation may be effected; because sulphur sublimes a little more easily than arsenic. But it is more convenient, as well as more expeditious, to make use of some additament that hath a greater affinity with one of those substances than with the other. Fixed alkalis and mercury, both of which have more affinity with sulphur than with arsenic, may be very properly employed on this occasion. Cobalt is a mineral composed of arsenic, an unmetallic earth, and frequently bismuth: and as none of these are very volatile, except the arsenic, this may be easily separated from the rest by sublimation. The unmetallic earth which remains has, like that of the ore of bismuth, the property of giving a blue colour to any vitrifiable matters melted with it; whence it is conjectured, that cobalt and the ore of bismuth have a great resemblance, or are often blended with each other. Nevertheless, Mr. Brant, an ingenious Swedish Chymist, insists that they are very different: he pretends that the metallic substance contained in the true cobalt is a semi-metal of a peculiar nature, which hath been erroneously confounded with bismuth: and indeed he proves by a great number of curious experiments, related in the Memoirs of the Academy of Upsal, that these two metallic substances have properties that are essentially different: to that which is obtained from cobalt, he gives the name of _Regulus of Cobalt_. Besides the minerals already recited, there is found in the bowels of the earth another species of compound body, of which we have already taken notice; but which is supposed, with some degree of probability, to belong as much to the vegetable as to the mineral kingdom: I mean the _Bitumens_; which the best observations oblige us to consider as vegetable oils, that by lying long in the earth have contracted an union with the mineral acids, and by that means acquired the thickness, consistence, and other properties observable in them. By distillation they yield an oil, and an acid not unlike a mineral acid. Mr. Bourdelin has even demonstrated, by a very artful and ingenious process, that amber contains a manifest acid of sea-salt. See the Memoirs of the Royal Academy of Sciences. CHAP. XVII. _Explanation of the Table of Affinities_.