(150.) Between the evaporating power of the boiler, and the magnitude of surface it exposes to the action of the furnace, there is a relation which, like that of the grate surface, has never been ascertained by any certain or satisfactory experimental investigation; much less have the different degrees of efficiency attending different parts of the boiler-surface been determined. That part of the surface of the boiler immediately over and around the grate, is exposed to the immediate radiation of the burning fuel, and is therefore probably the most efficient in the production of steam. The tendency of flame and heated air to rise, would naturally bring them in the flues into closer contact with those parts of the boiler-surface which are horizontal in their position, and which form the tops of the flues, than with those which are lateral or vertical in their position, and which form the sides of the flues. In a boiler constructed like that already described, the flue-surface therefore, which would be most efficient, would be the concave bottom of the boiler extending from the fire-bridge to its remote end. In some boilers, especially those in which steam of high pressure is produced, the form is cylindrical, the middle flue being formed into an elliptical tube the greater axis of which is horizontal from end to end of the boiler. It seems doubtful, however, [Pg263] whether in such a boiler the heat produces any useful effect on the water below the flue, the water above being always at a higher temperature, and therefore lighter than that below, and consequently no currents being established between the upper and lower strata of the water. It was considered by Mr. Watt, but we are not aware on what experimental grounds, that from eight to ten square feet of heating surface were sufficient to produce the evaporation of one cubic foot of water per hour. The practice of engine-makers since that time has been to increase the allowance of heating surface for the same rate of evaporation. Engine-builders have varied very much in this respect, some allowing twelve, fifteen, and even eighteen square feet of surface for the same rate of evaporation. It must, however, still be borne in mind, that whether this increased allowance did or did not produce the actual evaporation imputed to it, has not been, as far as we are informed, ever accurately ascertained. The production of a given rate of evaporation by a moderate heat diffused over a larger surface, rather than by a fiercer temperature confined to a smaller surface, is attended with many practical advantages. The plates of the boiler acted upon by the fire are less exposed to oxydisation, and the boiler will be proportionally more durable.