account of the influence of winds, the danger from dust, and the need of drawing the air from the purest source possible. It has been found that if this air can be drawn from the south or warm side of a building there will be a decided saving in fuel. Carpenter says:

It may be demonstrated by a properly protected thermometer that the average day temperature of air is higher on the south than on the north side of a building. The difference often reaches 10° F. An average of 5° F. would make it highly advantageous to take the air from the south rather than from the north side of a building. If an average rise of 35° F. is needed in the air temperature in ventilating work, then one-seventh of the heat required for that rise could be gained by choosing a south as against a north location for the inlet. (Heating and Ventilating Buildings, by Dr. Rollo C. Carpenter, pp. 451–452.)

The fresh air must be brought into the bottom of the heating chamber so that even on windy days there will be no possibility for reverse currents, and that at all times there will be as little hindrance to the easy movement of the air as possible.

All things considered, especially when an ample radiating surface is installed, when the ducts leading to the rooms are mathematically proportioned and not too long, and when proper means are afforded for humidifying the air, hot-air furnaces are fairly satisfactory in the milder climates of our country, and if carefully kept in repair there is no valid argument totally condemning them. The main troubles come through overheating, lack of humidification, and unbalanced system of ducts leading to the various rooms and neglect in upkeep.

It is unnecessary to state or discuss in this case the various rules used for determining the exact amount of radiating surface needed to meet easily and safely the demands to be made upon any system of heating.

These are matters for the expert engineer to determine, and a school board will act wisely when it seeks and pays for such advice. This caution, however, ought to be given: Some expert engineers— indeed, many of them-know little about the demands of schools, and it is always better to select for consultation one who has made a careful study of the peculiar needs of schoolroom heating, and especially one who has no connection with manufacturers of heating systems. One duty, and a very important duty indeed, which such an expert adviser ought to be called on to perform, is to test the plant when completed and be sure it fulfills the contract signed.

5. HOT-WATER HEATING.

In the past few years hot-water heating systems have been installed in greater numbers in school buildings than theoretically one could have anticipated, and yet there are some major difficulties with this system when applied to the heating of school buildings.

The one outstanding disadvantage and danger is that during severely cold weather, during vacations, the radiators, which, of course, are necessarily filled with water, will freeze and burst unless emptied or the fires kept burning. Were it not for this particular difficulty there would be no hesitancy in recommending hot-water systems for medium-sized school buildings, for in many respects it is the best system of central heating thus far devised. In the first place, it is far more regular and steady than either steam or hot air. It is more easily regulated, it is free from noise and "air-scorchings" and more economical of fuel. It is a little slower, however, than either of the other systems, but once the rooms are heated variation is very slight. It has one distinct advantage over steam. In mild weather, when only a slight increase over outside temperature is demanded, small fires will produce the result and keep heat going that in a steam boiler would produce no effect because obviously the fires must be heavy enough with steam heating plants to develop steam before any heat can be supplied.

Speaking in general terms, it is not safe to install a hot-water system in very cold climates, but in mild climates it can be used safely and satisfactorily if all necessary precautions are taken and the system is installed under the guidance of a competent engineer. There is far less expense attached to keeping up a carefully planned hot-water system of heating than any other known system. Naturally, it requires a greater radiating surface for a given space than steam, but offsets this difficulty with a much steadier and more regular heat. There will probably be in the next few years a rather striking growth in the number of hot-water heating systems installed in school buildings in the milder parts of the country, and this will doubtless be in the line of advancement.

6. STEAM HEATING.

The most generally used method for heating the larger school building is that of direct or indirect radiation or both from the radiators of a steam heating plant. Within the past 10 years lowpressure systems of steam heating have been installed in a great majority of the new buildings constructed. The one great advantage of steam heating lies in the fact that it is much easier to regulate the heat in the various rooms, and to proportion it properly, especially in large buildings, than any other system. Ten years ago it was practically universal to use, in connection with a steam heating plant, a plenum fan system to blow the air over the radiators into the various rooms. Later the split system was developed, in which a part of the radiating surface is located in the various rooms to be heated, and the rest left in heating chambers and con

nected up with the various rooms by inlet ducts through which fans operate to force in warm fresh air. This development took place on the theory that proper ventilation could be secured by the plenum system in which doors and windows of classrooms were always kept closed and air forced in and out by means of the fans. Recently, thoroughgoing investigations have somewhat changed our attitude in this regard and at the present time there is a growing tendency to depend more completely on direct radiation without the aid of the fan system. This point will be discussed more completely under

the general heading "Ventilation."

Steam radiators placed in the schoolroom are often disconcertingly noisy. The hammering or slapping sounds which are not infrequently heard are very annoying and distracting alike to the teacher and the pupils. No way has been devised to get rid entirely of these disturbances, for they seem to be caused by the rising steam catching and driving back the returning stream of water produced by the condensation of the steam already used. In addition to this noise in any system of steam coils, the valves get leaky and the escaping steam will set up confusion, or the dripping water will injure the building.

All of the so-called vacuum systems of steam heating are designed to prevent this noise, and in addition to manage the steam supply in the radiators so that the temperature will be automatically regulated. But systems of this type, which I have seen in use and for which large claims have been made, are not free from beating noises, nor will they at all times properly regulate the temperature. Theoretically, they may be all that they claim, and if perfectly set and furnished may suffice in practice; but I am told by a distinguished engineer that thus far there is surely a discrepancy between the claims made for these systems and their practical workings.

Steam heating is better adapted to cold climates than it is to those parts of our country where mild to medium weather prevails through the greater part of the winter season. This system of heating, however, frequently requires more care than any other system now in general use. This is largely due to the fact that the radiators, of which a great number are required, are often quickly heated and as quickly cooled. This rapid expansion and contraction necessarily throws much strain upon all connections and when these are opened they are frequently hard to close. The boilers of course require expert care, especially where high pressures are used both to insure. safety and to guard against deteriorations. Leaking flues and steam fittings require the services of expert mechanics and this is expensive service. But taking it all in all the best advice to give, especially in the colder climates of the country, is to install a low-pressure system of steam heating and to depend largely, if not entirely, upon direct

radiation and a thoroughgoing system of thermostats to regulate the heat in each room. Of course the utmost care must be taken to proportion and apportion the radiators in the several rooms and to make certain that the disturbing noises or "water-hammerings" are eliminated as completely as possible.

It is essential that every board of education upon whom the duty of building schoolhouses rests should see to it that the best of expert advice be followed in installing heating plants. Otherwise one can not expect satisfactory results.

7. THERMOSTATS.

When the heating of schoolrooms is wholly or is part dependent upon indirect radiation and warmed air introduced into the classroom, it is a vital necessity to have an adequate system of thermostats to regulate the temperature of the air or rather to maintain a proper temperature inside the classroom.

Such devices depend in one way or another upon the principle of expansion by heat and contraction by cold. The practical difficulty to overcome has been that of hitting upon a medium sufficiently sensitive to respond and yet steady enough to prevent rapid fluctuations. Some use mercury as the medium, others sensitive metals of a more rigid form, and still others liquids of such a composition that they will vaporize at the temperature required. Generally, those using mercury depend on making and breaking an electric circuit in which an electromagnet acts upon the dampers to close or open the entrance for hot air, and vice versa to open or close the duct permitting the entrance of cold or tempered air. Those using liquids depend directly upon the force generated by the vapor to regulate the dampers, while those using metals of a rigid sort depend upon the management of compressed air to do the work required. These appliances have brought great relief to both the teachers and the children, for without them the teacher was compelled to keep watch on the thermometer to prevent the room from either becoming too cold or too hot. When thermostats are properly installed and the heating is of the indirect sort mentioned, the temperature of a schoolroom can be regulated, and, if the heat is always to be had, will keep the room at a temperature not varying more than 2° F. at any time. One of the most successful kind may be described briefly as follows: Somewhere in a convenient place in the basement is a tank containing compressed air, from which small pipes radiate to the various schoolrooms, where they connect with a thermostat properly located and carefully adjusted. One part of the apparatus visible in the schoolroom shows by means of a pointer whether the room is being cooled

or warmed. A neat covering containing a thermometer shields the rest of the apparatus and at the same time furnishes a ready means for testing its accuracy. The thermometer has no organic connection with the thermostat. The air is supplied to the tank containing the compressed air by an air pump automatically regulated. The force applied to this pump is usually that of the water in the pipes of an ordinary city water supply. If this is not available, some other source of regular power must be used. If the temperature of the room is below that required, the opening in the end of this tube is kept closed by a tight-fitting valve or plug held in place by a spring. When the air reaches the temperature desired, a tongue of specially prepared metal expands, and, by means of the mechanism connected with it, opens the valve and permits the air to escape through another tube, at the end of which is a small air-tight chamber connected on one side with a diaphragm, or metal bellows. These are in the basement, where the changes in the diaphragm can be observed by the fireman. The pressure of the air within this cavity forces up this diaphragm, which acts on a lever connected with the dampers in the air ducts leading to the schoolroom, and thus cuts off a part or all of the warm air and at the same time opens the damper to the cold-air chamber, and thus through the pressure of the fan allows the cold or tempered air to be driven into the schoolroom until that temperature is reached which causes the metal tongue to contract. The spring will then cause the valve to close the tube leading to the compressed-air tank and at the same time permit the air in the diaphragm to escape back through the tube into the outer air. The diaphragm will then be compressed by a spring attached to it and the dampers will be reversed, allowing warm air again to enter the schoolroom. This plan will prove successful if carefully guarded, and the apparatus is so made that it will not be constantly changing and through too great a range of temperature.

There is this to be said about all thermostats with which I have had experience in school buildings: They are complicated bits of apparatus and must be guarded with care and will not prove successful unless the caretaker thoroughly understands their construction and knows how to adjust them. It is a waste of money and dangerous to the health and comfort of school children to install a system of thermostats and then to put them under the control and supervision of a janitor or a teacher who does not understand the principle used in their construction or the mechanism devised to utilize the principle involved. A janitor selected for political purposes or simply because he can sweep and build fires is never likely to handle them well. Here, then, let me repeat what is said elsewhere, the janitor of a modern and thoroughly equipped school building must