D. G. Adelsberger, Baltimore, Md., writes:

"I have a contract for plumbing, etc , etc., of the university now building at Washington City. Mr. E. F. Baldwin (who is the architect and one of your subscribers) and myself, have had a consultation about the construction of the soil and vent pipes for water-closets, etc., etc. He has an opinion one way, and I another, so we agree to write to you for your opinion. Inclosed is a sketch showing the soil pipe running the highest at ridge of roof, and traps of closets all emptying into it and vent pipe the lowest which would cause a current of air to come down the vent pipe and follow the water, etc. from the closets. and vent the closet pipes and syphon up the soil pipe at highest point and make current of air circulate down the lowest or shortest pipe and up the longest.

"Now suppose the soil pipe to be placed where the vent pipe now is and would thus be the shortest, and place the vent pipe where the soil pipe now is and reverse the closet pipes and run them into the shortest pipe, the water from the closets and traps would taen run into the shortest pipe, thereby running against the current of air coming down the short pipe. I say coming down the short pipe because it would not do otherwise according to natural philosophy, the other pipe being the highest would cause the current of air to come down the shortest and lowest and go up the highest pipe which would be the water from closets running against the current of air.

"Will you please give me your opinion on the subject, which plan will ventilate the pipes best ?"

[We have reproduced the sketch as sent, which shows no traps, although they are mentioned in the letter.

Our correspondent is mistaken in supposing that the air will necessarily flow down the shorter and up the taller pipe.

The strength of a draft of a chimney or any other flue is proportional as much to its temperature as to its height, and in this case in whichever pipe the product is the greater of its height multiplied by the number of degrees that it is warmer than the outside air, the draft will be the stronger and the air will flow up. So that in this case if the shorter pipe is enough warmer the air in it will flow up and draw down in the taller. If two pipes or flues are equally high and warm no current will start, but if started it will continue, because the descending current will cool one flue, and being somewhat warmed before it escapes from the other the difference of temperature necessary for the flow will be maintained.

If the pipes shown in the sketch are all equally warm the air will circulate as our correspondent supposes.

The discharge from the water-closets will not, we think, materially affect the flow of the air in the pipes above the highest range. When one of the upper closets discharges, it will for the moment reverse the currents of air in all the pipes below it, but they will resume their normal direction as soon as the flow has passed. A discharge from one of the lowest closets would accelerate the air currents in the lower part of the vertical pipes, but would probably reverse the currents in the upper cross pipes and produce little if any effect above them. A discharge from the middle range would produce a combination of these two results.

If it were possible to get a fresh-air inlet at any point below, a much more efficient ventilation could be obtained.

If the lower inlet cannot be had, the arrangement shown is perhaps the next best thing, and we do not think it makes much difference which way the closets discharge as far as the ventilation is concerned, as the effect of their discharge will be only local and temporary.

It should be remembered that in taking in fresh air from above, the two pipes act as flues pulling against each other, and only the difference of their respective drafts is available for ventilation.]