Whether this estimate of an average horse's power be correct or not, in reference to the actual work which the animal is capable of executing, is a matter of no present importance in its application to steam-power. The steam-engine is no longer used to replace the power of horses, and therefore no contracts are based upon such a comparison. The term horse-power, therefore, as applied to steam-engines, must be understood to have no reference whatever to the actual animal power, but must be taken as a term having no other meaning than the expression of the ability of the [Pg289] machine to move the amount of resistance above mentioned through one foot per minute. (172.) It has been already explained (67.) that the conversion of a given volume of water into steam is productive of a certain definite amount of mechanical force, this amount depending on the pressure under which the water is evaporated, and the extent to which the expansive principle is used in working the steam. It is evident that this amount of mechanical effect is a major limit, which cannot be exceeded by the power of the engine. If the steam be not worked expansively, then the whole power of the water, transmitted in the form of steam from the boiler to the working machinery, will be a matter of easy calculation, when the pressure at which the steam is worked is known. A table, exhibiting the mechanical power of a cubic foot of water converted into steam at various pressures, expressed in an equivalent number of pounds' weight raised one foot high, is given in the Appendix to this volume. Where much accuracy is sought for, the pressure at which the steam is used must be taken into account; but by reference to the table it will be seen, that when steam is worked without expansion, its mechanical effect varies very little with the pressure. It may therefore be assumed, as has been already stated, that for every cubic inch of water transmitted in the form of steam to the cylinders, a force is produced, represented by a ton weight raised a foot high. Now, as 33,000 lbs. is very nearly 15 tons, it follows that 15 cubic inches of water converted into steam per minute, or 900 cubic inches per hour, will produce a mechanical force equal to one horse. If, therefore, to 900 cubic inches be added the quantity of water per hour necessary to move the engine itself, independently of its load, we shall obtain the quantity of water per hour which must be supplied by the boiler to the engine for each horse-power, and this will be the same whatever may be the magnitude or proportions of the cylinder.