The first requirement asks for a vertical extension of all soil and waste pipes through the roof. This extension affords a ready outlet for all gases that would otherwise tend to accumulate inside the pipe system. In the case of soil pipes nothing short of an extension the full bore of the pipe will answer this purpose. It has been proposed, of late, to enlarge the soil pipe from the highest floor to the roof to six inches diameter, in order completely to prevent any stagnation of air in the pipe. Waste pipes should be enlarged from the point where they pass through the roof, to four inches diameter, as smaller outlets are liable, in cold climates to become obstructed by the freezing of condensed vapor. Plumbers sometimes use galvanized wrought iron or tin pipes for this extension, but this is decidedly bad practice. It should be of the same material as the main soil pipe, and its joints should be worked with equal care. The extension of soil and waste pipes should terminate at a distance from any windows, louvred skylights, or ventilating flues, and at least two feet below the top of the nearest chimney. It is desirable to have this extension as high as possible above the roof, so as well to expose the mouth of pipe to the influence of air currents. In order to prevent any obstruction of the soil pipe, plumbers often cover the mouth with a return bend. This, however, is objectionable, as it interferes with proper ventilation. Less bad is the plan of capping the soil pipe with a suitable fixed cowl, such as, for instance, Emerson's or Wolpert's ventilator. The best plan seems to be to do away entirely with any cover to the soil pipe mouth. Capt. Douglas Galton, in his book "Construction of Healthy Dwellings," says in regard to this question: "A tube or shaft with an open top acts best. It is, however, necessary to protect the top to prevent rain from entering the tube; but a cover tends more or less, according to its shape, to delay the current in the tube or shaft." This necessity of covering ventilating tubes or chimney tops to protect them from rain, does not exist in the case of soil pipes; these may only want Protection against malicious introduction of stones or similar articles. A galvanized iron, copper or brass wire basket set into the mouth of the soil pipe will answer this purpose.

There is no doubt that open-mouthed pipes have a better upward ventilation than pipes covered with cowls, if the wind blows horizontally or nearly so. Wolpert in his "Treatise on Ventilation and Heating" states the average useful effect in per cents, of the velocity of the wind, as derived from a number of experiments, to be:

68.6 per cent. for open-mouthed tubes, 51.9 per cent. for pipes capped with

Wolpert's new cowl, 35.8 per cent. for pipes capped with

Wolpert's old cowl, for a horizontal direction of the wind. In other words, the upward suction in a tube without any cowd is in the average equivalent to over 2/3 of the force of the wind, blowing over it in a horizontal direction. For pipes capped with Wolpert's new cowl it is only a little more than 1/2 of the wind force, and for the old cowl it is 1/3 of it. As an average for other directions of the wind Wolpert finds the upward draft in pipes covered with his new and old cowls to be 51.5 per cent. and 34.5 per cent., respectively, of the wind force.*

*The current of air in these experiments created by a powerful fan, the velocity of the current varying from 8 to 31 meters per second (from 17.9 to 69.3 miles per hour), equivalent to high winds and hur-rloanei respectlvely. The diameters of the cowls tested varied from 0.787 to 3.937 inches. it is to be regretted that the author did not extend his experiments so as to include much smaller velocities of current.

It is very likely that for the latter the percentage of useeful effet of cowls would be much smaller.

The result of an elaborate series. of about 100 experiments upon ventilating cowls, made on seven different days, at different times of the day, and under different conditions of wind and temperature, by Messrs. W. Eassie, Rogers Field and Douglas Galton, was as follows: "After comparing the cowls very carefully with each other, and all of them with a plain open pipe as the simplest, and in fact only available standard, the sab-committee find that none of the exhaust cowls cause a more rapid current of air than prevails in an open pipe under similar conditions, but without any cowl fitted on it. The only use of the cowls, therefore, appears to be to exclude rain from the ventilating pipes; and as this can be done equally, if not more efficiently, in other and similar ways, without diminishing the rapidity of the current in the open pipe, the sub-committee are unable to recommend the grant of the medal of the Sanitary Institute of Great Britain to any of the exhaust cowls submitted to them for trial."