There are two very different systems of using steam in greenhouses. One may be called the gravity system, and the other the high pressure system. Both have their advocates, but circumstances will often decide which is the better system to use for your particular place.

Frequently the end of a range of glass may abut on a steep bank or ravine, or a sharp descent, where it is little expense to cut away the bank and place your boiler sufficiently below the level of the greenhouses to afford the best conditions for the gravity system. We know of such places in our locality, and one well known place in Philadelphia where such circumstances exist. It may be economy in installation to adopt the gravity system. But where there is not such a slope, and if the site of the boiler-house is about the same level as the greenhouse floor, to adopt the gravity system you would have to excavate ten or twelve feet, then perhaps drain the hole at great expense. Under these circumstances I must pronounce this excavation, with all its terrible inconveniences and expense, the height of folly. It is absolutely unnecessary. I have never known anyone using the high pressure system to convert their system into gravity, but I do know men of large experience who, after they had their boilers installed down in a deep hole, changed to the pressure system and expressed themselves as being foolish for ever going to the expense of excavation, hauling out ashes, etc.

Now suppose you do use the gravity system. The boiler should be low enough so that the return pipe in your houses, the pipe that carries the condensed steam, is four feet above the level of the water in the boiler, and if it is six or eight feet so much the better.

Scarcely any two ranges of glass are built alike, so it would be impossible to give any definite directions for arrangement of pipes. Under the gravity system the main steam pipe leading from the boiler needs to be larger than with the pressure system, because the steam is not so dry and hot. For 20,000 feet of glass surface we think that not less than a 6-inch flow should be used, while we have proved that a 3-inch under pressure is large enough. The main pipe can either run along the end of the range of connected houses, or up one of the houses in a trench free of water. In either case it should be well covered with asbestos, or you will waste heat as well as have too much of it in certain parts of the house. Your main steam pipe can always be reduced in size as its steam is used. If you start with a 4-inch pipe, after supplying steam for two houses it can be reduced to a 3-inch, and after two more are served to a 2-inch. I don't know that it makes any difference whether your main pipe rises gradually and evenly, or whether it drops slightly to the farthest point, but would prefer that it rose slightly. There is a difference of opinion as to where the heating pipes should be placed. We lately learned the opinion of a most successful rose grower who said all he wanted were pipes on the side walls. We know equally successful and larger growers who run a 1 1/4-inch around the edge or side of every bench. And I am inclined to believe there is a more even and better distribution of heat when the pipes are well distributed. Few growers of roses or carnations nowadays advocate heating pipes beneath their benches, but a pipe running along the side of a bench or bed is a different thing. Hook plates are now being made to build into the low brick or cement walls of semi-solid beds and there I think is a good place for a heating piper Some who use steam will run five or six 1 1/4-inch pipes up the length of the house and provide at the farther end one 1 1/4-inch or 1 1/2-inch return for the condensation. In anything less than 300 feet this is not the best plan. If there is a drop of a few inches to the farther end of the house and the same drop back, then run two or three flow pipes, or more as the case may be, and the same number returning. The latter pipes, which we will call returns, will radiate as much heat as the flows, for with three or four pounds pressure they will be full of live steam and there will be no condensing till the steam enters the main return, which is in most cases on or just below the surface of the house.

Although not an advocate of overhead pipes, I have found in the use of steam that two 1 1/4-inch pipes, one on each side of the house, two or three feet below the glass, are a great help in very severe weather. We have winds and frost and blizzards occasionally when you can almost feel the frost falling from the glass. Then it is that the overhead pipes are a blessing, moderating the cold before it strikes the plants. In more favored climates these overhead pipes may not be necessary, but they cost but little. We run a pipe along over the plants to a farther end of the house and, dropping it there, return along the side of a bench.

I think in some establishments there are not valves enough used. There should be a valve on every run of pipe, not only on the end where you let in the steam but on the end where it enters your main condensed. If you have occasion to turn off a certain run of pipe and have only a valve at the feed end, you will by closing that valve soon have a vacuum in the pipe, because the steam will soon condense. Then the pipe fills up with any water there is in the system. This does no harm at the time, but when occasion arises to run that pipe again you have a lot of cold water to dispose of and that must go back to the boiler, needing more fuel to raise the water to the boiling point. Put on valves where-ever there is any chance of their being useful. In case of a cracked pipe or a broken joint through expansion, the absence of a valve may cause you to shut down your whole system while repairs are made. Those using cast iron or metal gutters will find a steam pipe running along a foot beneath the gutter will make a considerable difference in the disappearance of the snow. I think now I have said enough on this system and will proceed to consider the other, which we called high pressure and with which I have had much more experience.