A simple experiment, capable of yielding results which shall be at least comparative, has long been sought after by large consumers of coal and artificial fuel abroad in order to ascertain the relative calorific power possessed by each description, as it is well known that the proportion of mineral matter and the chemical composition of coal differ widely. The determination of the ash in coal is not a highly scientific operation; hence it is not surprising that foreign merchants should have become alive to the importance of estimating its quantity. While, however, the nature and quantity of the ash can be determined without much difficulty, the determination of the chemical composition of coal entails considerable labor and skill; hence a method giving the calorific power of any fuel in an exact and reliable manner by a simple experiment is a great desideratum. This will become more obvious when one takes into consideration the many qualities and variable characters of the coals yielded by the South Wales and North of England coal fields. Bituminous coals--giving some 65 per cent, of coke--are preferred for some manufacturing purposes and in some markets. Bituminous steam coals, yielding 75 per cent, of coke, are highly prized in others. Semi-bituminous steam coals, yielding 80 to 83 per cent, of coke, are most highly valued, and find the readiest sale abroad; and anthracite steam coal (dry coals), giving from 85 to 88 per cent, of coke (using the term "coke" as equivalent to the non-volatile portion of the coal) is also exported in considerable quantity. Now the estimation of the ash of any of these varieties of coal would afford no evidence as to the class to which that coal belongs, and there is no simple test that will give the calorific power of a coal, and at the same time indicate the degree of bituminous or anthracitic character which it possesses. In order to obtain such information it is necessary that the percentage of coke be determined together with the sulphur, ash, and water, and these form data which at once show the nature of a fuel and give some indication of its value. To ascertain the quantity of the sulphur, ash, and water with accuracy involves more skill and aptitude than can be bestowed by the non-professional public; the consequence is that experiments entailing less time and precision, like those devised by Berthier and Thompson, have been tried more or less extensively. In France and Italy, Berthier's method--slightly modified in some instances--has been long used. It is as follows: