II

Mr Hammond has failed, I think, to bring anything forward of material value in support of his views on the circulation of hot water. He seems to have been more anxious to point out that there is no analogy between the emptying of a cistern by means of a syphon and the circulation of water in a heating apparatus, than he has been to explain why it is that the quickest circulation takes place in the highest parts of the apparatus, in preference to those coils of pipes that are less elevated above the boiler. If Mr Hammond can prove that it is not the additional weight of water in the return-pipes, but something else, that causes this to take place, he will be doing good service by at once correcting the error which has so long been entertained by hot-water engineers. I am quite willing to yield the point as to the ascent. I think I said it was immaterial. I would as soon have a vertical ascent as a slowr gradient. Mr Hammond's mode of fixing pipes will not be very likely to meet with much favour by either gardeners or hot-water engineers, if it was for nothing else but for their appearance. Fancy a house, say 60 feet long, with four rows of pipes along the front! On entering the house the flow-pipes commence a descent of, say 1 in 120 - that is, 6 inches in a length of 60 feet.

The return-pipe must have the same slow descent at least, so that when the latter reaches the point where the pipes enter the house, there will be a clear space of 12 inches between the flow and return pipes. This I think would not look so well, nor be so convenient in fitting, and moreover would occupy more room, than they would if fitted up on what I suppose we will soon have to call the "old system." I think it would be better to adopt a medium course, that of having the flow and returns both level inside the house, and so parallel to each other. This is done by many, and with as much success as those fitted with a slow ascent.

I am quite prepared to believe what "J. H" says in support of Mr Hammond's system. Hot water is wonderfully accommodating where it has but one way to circulate. I am acquainted with an old fitter who boasts of his firm having successfully heated a house for a gentleman with only one pipe for both flow and return. (He did not say it was a four-inch.) "J. H." will no doubt be done for ever with the "old system." Suppose, now, for the sake of experiment, he was to fix pipes to his boiler, and give them the usual ascent from, and descent to, the boiler, does he believe that he would get as good a circulation in the pipes that descend from the boiler as in those that-ascend? I venture to predict that the water would flow into the latter in preference to the former, if no check is put upon it in the shape of a valve. I shall not trespass further upon your valuable space. R. Inglis.

III

I have been rather interested the last few months in the question which has been raised by Mr Hammond on the circulation of hot water, and I think he is entitled to many thanks for bringing forward such a subject for discussion. I understand the main point raised in his first paper to be this: Is a continuous rise in the flow-pipe a hindrance to the circulation? Hot-water engineers say it is not, but I think a rational view of the case may prove the contrary. While quite agreeing with the correctness of Mr Makenzie's quotation from Mr Clarke's tables, I would only ask, "Why continue the rise throughout the whole length of the flow-pipe? Is it done to give an increased fall to the body of water in the return-pipe, or is it to assist the circulation in the flow? If the latter, I think the proper designation ought to be the force-pipe. Mr Hammond points out, in his reply to Mr Inglis (p. 210), "The water owes its expansion and relative lightness to heat, in the first instance, and as heat fails, the water contracts and becomes 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 goes on to say, that when this point is reached by a slow gradient, the water must have become colder than at the time it left the boiler; and of course the greater distance it travels, the greater is the decrease of temperature - that every inch it has to be raised thereto adds an extra tax on the "pushing and pulling" powers of the colder and heavier water in the return-pipes. With which I entirely agree; and not only must this be the case, but every inch it has to be raised in the length of the flow is a gradually increasing tax on its own powers of flow.

Mr Hammond says, at p. 57, "As soon as the fire acts on the boiler, the particles of water in contact with its inner surface bound upwards, until they come in contact with the inner surface of the upper side of the flow-pipes. Here they part with a portion of their heat, and become of greater specific gravity than they were at the time of starting on their upward course, and would now commence to descend towards the point whence they started, but that they are still lighter than the particles composing the body of cold water in the flows," etc. So they would, but I would also take into account the cold water rushing in from the return-pipes, which I think is one of the principal causes of circulation; and as an instance of the descent of the water after reaching the top of the boiler, and the absence of this return current, I need only mention the homely illustration of a common tea-kettle. Therefore I should consider that to have the highest point of circulation as near to the boiler as possible, and to have the return current with the greatest possible fall, would ' be the best means to insure a speedy circulation. This view is corroborated by your correspondent "J. H." in last month's number.

And this is also the conclusion, 1 think, Mr Hammond arrives at in disposing of "C. M.'s" theory of the circulation. The idea of hot-water engineers, that a continuous rise in the flow-pipe, no matter of what length, accelerates circulation of the water, may turn out to be, like many another popular idea, a fallacy.

Robt. Stevens.