“In 1886 the _Alliance_ was supplied with new boilers, fitted with a system of forced draft designed by the Bureau of Steam Engineering. It was originally the intention of the Department to put six boilers in this vessel, as in the _Enterprise_ and _Nipsic_, but with the introduction of the forced-draft system, which was purely experimental, this number was reduced to four, having a total grate surface of 128 square feet. The boilers were designed to burn anthracite coal with natural draft, and were of course unsuited to the requirements of forced draft, the ratio of heating to grate surface being only 25.6 to 1, and the water surface and steam space being small. The maximum indicated horse-power developed on trial was 1022, but any attempt to run at this or at increased power for any length of time was attended with so much priming of the boilers that the trial had to be discontinued. Alterations were made in the boilers to prevent the priming, but no continuous trial was had previously to the sailing of the _Alliance_. The results obtained on a measured base were, however, sufficient to demonstrate the practicability of the system, and to show that a higher power could be maintained with the four boilers at forced draft than with the original eight boilers at natural draft. “The practical working of the system at sea presents no difficulty, as a recent run of the _Alliance_ has demonstrated. On a continuous run of ten hours, using only two boilers with sixty square feet of grate (the grate surface of each boiler having been reduced to thirty), the mean indicated horse-power was 668 and the maximum 744, being respectively 11.1 and 12.4 indicated horse-power per square foot of grate. There was an entire absence of priming, and no difficulty was experienced in operating the forced-draft apparatus, the length of the trial having been determined by the arrival of the vessel in port. The coal burned was Welsh, of fair quality, the consumption being 29.9 pounds per square foot of grate. “The efficiency of the system may be judged by the results obtained from an experimental boiler at the Washington Navy-yard. The boiler was of the marine locomotive type, and had a ratio of heating to grate surface of 32.73 to 1, with a water space of 245 and a steam space of 163 cubic feet. The coal burned was ordinary Cumberland Valley bituminous, and the evaporation, when burning as much as forty pounds per square foot of grate, was 6.61, while with 37.5 it was 7.24, and this with a moderate air pressure—1.5 inches in ash-pit and one inch on furnace door.”