them, it will be necessary at all the principal stations, and, perhaps, at all the stations where cars are left, that there should be either a stationary boiler, or else one or more locomotives specially devoted to heating cars, and that pipes or hose should lead from the boiler to those points where cars will stand when they are to be heated. For this purpose an ordinary low-pressure boiler would be sufficient, and it could also be used to heat the station. This method is adopted by the Boston and Albany at Boston and Springfield, and by the Connecticut River at Springfield. It would certainly be the method to be adopted by most roads at all stations where they have occasion to leave cars in the cold for any length of time. Some roads, especially the Old Colony, are, however, obliged to leave cars at so many stations that it would be at least a hardship to be obliged to have a stationary boiler at each one of them. It is an open question whether there is any other feasible means of heating up cars at the smaller stations where only two or three cars are left. It having been suggested that oil stoves might be used for this purpose, some experiments were made to determine their heating powers, with results which were not encouraging. The odor also would be objectionable if they were not properly trimmed. AMOUNT OF STEAM USED IN HEATING CARS. There are all sorts of opinions and statements in regard to the amount of steam taken from the locomotive to do the heating. Some claim that the steam cannot be spared, and that the engine cannot make her time if called on for this extra duty, and others that it makes no perceptible difference in the running of the locomotive. Some even go so far as to say that all the steam required for heating can be furnished through a hole in the boiler no larger than a pin. hole. That neither extreme is correct is shown by the following experiments. It did not seem to be worth while to undertake any very precise work to compare the amount of steam used, when this, that, or the other appliance is used, or this, that, or the other size or amount of pipe, for it does not seem that, in the present stage of development of steam-heating, such information is required. The winter was so far advanced when the investigation began that there was not sufficient time to perform the work in such manner L as to obtain indications as to the effect of each appliance upon the amount of steam used, and, moreover, at the beginning it was not at all certain whether the amount of steam used would form an element of any importance in the question. Such an investigation would have involved some very careful work in getting the same conditions of temperature outside and inside of the cars, and also in studying the effect of the wind. Other conditions also would have been required, which would have been at best difficult to obtain, and which could only have been obtained, if at all, with an empty train, and, if this were used, the results would not be those of practice. The attempt, therefore, has been made to obtain, as nearly as may be, the amount of steam used in an ordinary run in cold weather, leaving the train hands to manage the heating, as they usually do, which means that the cars are sometimes too hot and sometimes too cool, but generally somewhere near right. It also means that if they become too hot they will be cooled off, either by shutting off steam or by opening ventilators or windows. - In view of all the above, the problem has been to determine, approximately, the amount of steam used with sufficient exactness to show whether it is likely to be a serious tax upon the locomotive. The apparatus used consisted of two lengths of six inch flange pipe, as shown in the accompanying cuts, bolted together with a brass disk between them, this disk containing a two inch hole, into which is screwed a nicely-made circular orifice. The steam enters the apparatus at one end and passes through the orifice, and then from the other end of the apparatus it passes into the main train pipe. The pressures in the different parts are regulated by two globe valves, first controlling the admission of the steam to the apparatus, and the second placed just beyond the apparatus, to control its pressure on the train, so as to make it correspond to that ordinarily used. The pressures on each side of the orifice were shown by test gauges, and these and also the boiler gauge were read every five minutes during the run. Afterwards, when the engine was in the round-house, the same conditions were repeated as nearly as possible, with the following exceptions: the Instead of delivering the steam into the train-pipe it was delivered into a short pipe with another globe valve, which regulated its |