In both the engines every known means must be adopted to secure efficiency:--1, An excess of boiler power; 2, expansion permanently, say at ⅓; 3, steam of not less pressure than 20 lbs., and I should prefer 25 lbs.; 4, that cylinders, particularly top and bottom, slide chest, and steam pipes, be all jacketed, and the jacket supplied with steam from an auxiliary boiler of at least 10 lbs. more pressure than that of main boilers; and it would be very desirable to make some experiments to determine whether it is not worth having a heating apparatus to heat the steam immediately before it enters the cylinders.[147] _July 17._--After a long conference with Mr. Field, I continue of the opinion that it would be well to apply about three-fifths of the power to the screw and two-fifths to the paddles, and probably, as the vessel gets light, diminishing a little the expenditure of power on the paddles, and keeping up the full power on the screw. Mr. Field is not in favour of increasing the pressure of steam beyond 12 lbs. or 15 lbs., on the ground that all the mechanical difficulties increase rapidly without a corresponding advantage, particularly where size and weight are not so important. There seems much truth in this.... The possible advantages of a slight increase are not sufficient to justify the risk of the possible new difficulties in a work on so large a scale. Nothing uncertain must be risked. These arguments do not apply to the jacketing and heating, which Mr. Field also deprecated, or rather discouraged, simply on the ground of the trouble and difficulty of effecting it, but he admitted that all experience went to show the advantages of it; and as to the difficulties, which I could not see, they involve no other risk than that of being useless: they cannot do mischief. The heating of the top and bottom of the cylinders, I think, must be particularly important in a short-stroked engine working expansively. In a cylinder of 80 inches diameter and 40 inches stroke, having regard to the _time_ of contact, the area of the bottom will be nearly equal in effect to the surface of the cylinder. _July 19._--After much consideration, I think I feel satisfied that the best construction will be to have strong bulkheads every 30 feet or thereabouts, this distance being dependent on what is required for one set of boilers and its stock of coals; these bulkheads being carried right up wherever practicable--I think every alternate one may be--and then place the main ribs of the ship, and even at least two main deck beams, _longitudinal_ instead of _transverse_.