Having for the last few years spent considerable time in experimenting on the circulation of hot water (as an amateur), I was pleased to read Mr Hammond's remarks in the February number of ' The Gardener,' and I unhesitatingly endorse the same, as they agree with the results of my experience. His remarks are so lucid, that it would be impossible for me to render them more intelligible had I felt inclined.

This season I have made an alteration from the old system to that described by Mr Hammond - viz., the water, immediately it leaves the boiler, runs down an incline to the houses to be heated, and the pipes round the houses are on a level, and drop about 13 inches on entering the boiler. The valves (Messenger's) are fixed to the returns about a foot distant from the boiler. Any person interested would be pleased at the rapid circulation denoted by the thermometrical readings during its circuit, and the slight variation of temperature between the water in the flow and return as it enters the boiler, compared with others where the flow rises above the level of the boiler and then drops as a return.

Any persons who doubt this statement are welcome to give me a call and judge for themselves.

J. H.

The Mills, Swallowscliffe, Salisbury, Wilts.

In last month's ' Gardener' two of your correspondents question the correctness of my statements in ' The Gardener' for February in reference to the above subject; but after reading carefully what they say, I fail to see that they prove any of my statements wrong. Mr Inglis is of opinion that I "make too much of the inclination of the water to form a returnrcurrent in the flow-pipe, and too little of the fact that the water in the return-pipes is very much heavier, - that this heavy column is pulling, as it were, a lighter one, and at the same time is 'pushing it from below." Now, as regards this sentence, I may say that I consider both columns are equally concerned in the process of circulation; but I object to the water in the lighter becoming colder than it was at the time of leaving the boiler until the highest point of the apparatus is reached. As by it doing so before reaching, this point, the relative difference in the weight of the water in the two columns is lessened, and, consequently, circulation is retarded.

Mr Inglis next says, "that whether the rise in the pipes is made by a slow gradient, or by a verti-tical rise, will not make material difference to the circulation; but every foot that they are thus elevated above the boiler gives greater force to the circulation of the water." This sentence is a very important one bearing on the subject under discussion, and if its conclusions are right, then all I have said on the matter is wrong, and therefore worthless. Let us examine the first clause. It says - whether the rise in the pipes is made by a slow gradient, or by a vertical rise, will not make material difference to the circulation. Now, I maintain that, when the highest point of the apparatus is reached by a slow gradient, the result is a return-current in the flows. In an apparatus of a few hundred feet of piping, this return-current is not so marked as in an apparatus where the piping extends to thousands of feet; but it occurs in both. The water owes its expansion and relative lightness to heat in the first instance, and as heat fails, the water contracts and becomes relatively heavier; it therefore follows that the heated volumes of water should reach the highest points of the apparatus before any diminution of their temperature takes place; and when this point is reached by a slow gradient of, say 200 feet, I think Mr Inglis will admit that they cannot do so without becoming colder than they were at the time of leaving the point on which the fire acts.

And I think he will also admit that every inch the water has to be raised after it commences to cool is an extra tax on the pushing and pulling powers of the colder and relatively heavier water in the return-pipes. The slow gradient is therefore to be avoided as much as possible, as it leads to a waste of force, and, consequently, retards circulation.

We will now look into the last clause, which says that every foot the flow-pipes are elevated above the boiler, gives greater force to the circulation of the water. Mr Inglis is not singular in holding this opinion. Hot-water engineers are very decided about elevation of the flows above the boiler being essential to rapid circulation of the water in the apparatus. They will sink a hole in the earth 20 or 30 feet deep - unless prevented by some insurmountable obstruction - to obtain elevation for the flows above the boiler, and after the flows reach the surface from so low a level, they are again elevated to their farthest points from the boiler in the various compartments to be heated, so that circulation may be more rapid still. Thus it might be supposed that the rapidity of circulation was proportionate to the elevation of the flows above the point on which the fire acts. Let us see if this is really so, by supposing 5 feet to be the distance to which the flows are elevated above the boiler in an apparatus of, say 3000 feet of piping, and that it takes an hour and a-half from the time the fire is lighted until all the water in the apparatus has passed through the boiler.

Now, if the elevation had been 10 feet instead of 5 feet, would the water have made the circuit in less time - say an hour and a quarter; and if the elevation had been 20 feet, would it have performed the journey in three quarters of an hour 1 If the elevation of the flows above the boiler determines the rate of speed at which the water moves, then it would be possible to attain elevation sufficient to drive the water through the apparatus at a speed equalled only, by that of the electric spark.

The rapidity of circulation of the water in the heating apparatus is not measured or determined by the elevation of the flows above the point on which the fire acts, but by the difference of the specific gravity of the volumes of water in different points of the apparatus. The volume of least specific gravity will at all times occupy the highest, and that of greatest specific gravity the lowest, point in the apparatus, unless prevented from doing so through the defective adjustment of the latter, which is not of unusual occurrence. Mr Inglis seems to think there is some analogy between circulation as it takes place in the heating apparatus and emptying a cistern by means of a syphon. A little consideration will convince him that the two processes are not analogous. Circulation does not empty the apparatus. The latter is an endless tube full of water. At one point of this tube heat is applied, causing the water to expand and become relatively lighter; while at all other points the water is contracting and becoming relatively heavier - hence circulation.