There is nothing like ocular demonstration to convince the public, and when one sees the cotton wool ascending in the tube of the model, it does look as if the ventilator must be effective. To test the value of such a model, a strong current of air measuring a foot square, produced by a fan or blower, should be directed against it, when it will be found that the current, like the wind, is neither discriminating nor obliging enough to blow just at the right point or upon the right side of the exit plates; but full against the openings between the plates immediately fronting the wind, and so the suction or aspiration is neutralised. After testing the value of the model in the current of air suggested, the head of the ventilator should be removed, and the air current directed over

the end of the open tube. On comparing the aspirating effects of the air current upon the open tube with those produced by the air pump ventilators there will be a most striking difference in favour of the open tube, amounting from two to five times, according to the form of the pneumatic ventilator employed, and to the shape of the end of the open tube. If the open tube is cut off like Fig. 4, and the wind blows against the left side, the aspirating effects are nearly twice as great as when it impinges against a pipe, Fig. 5, with a mouth level all round. An ordinary revolving lobster cowl is four times as effective as the best ventilator of the air pump variety for inducing a current of air by the wind, but either is capable of upsetting the ventilation of a closed building; and, as before stated, wind suction is too irregular

and uncertain, and often too violent, to be of any practicable use in ventilating churches and public buildings.

It has been stated that the extracting efficiency of an outlet ventilator depends upon the surfaces exposed to the aspiration of the wind, and not so much upon the diameter of the tube itself. Such a statement is most misleading. The volume of air aspirated by the wind must always be inversely as the resistance or friction encountered, and as the friction increases greatly with every decrease in the diameter of the tube the volume aspirated does depend primarily upon the diameter of the outlet shaft.

By cutting off a tube slantwise at the top as above, Fig. 4, it is seen that the great friction which the air encounters against the lip of the tube, Fig. 5, has been removed, and nearly double the air will be aspirated in consequence. The result is due to the decreased friction, and the extra volume aspirated is not proportional to the length of the slant, but to the lesser friction which the air encounters.

In the turret form of extractors the friction is not reduced in the head itself by increasing the height and exposing more aspirating surface, so the suction power of such a ventilator must be proportional to the diameter of the inside tube and the friction encountered in the head. The fact that the wind aspirates so much more powerfully upon an open tube than by means of any extracting ventilator, no matter how much surface is exposed, shows how great is the friction which the wind encounters in aspirating the air through the head of the ventilator.

What the condition of the atmosphere in large towns would be if there were no wind it is difficult to tell, and, during the summer and hot weather, the ventilation of houses and public buildings is very largely dependent upon the movement of the air due to the breezes and currents formed through the unequal heating of the earth's surface by the sun's rays.

The ventilation of churches and public buildings, in

summer, requires the attention and observation of the caretaker to be exercised to the fullest extent if he is going to accomplish his duties in a successful manner. All that is usually attempted, however, is to open certain windows or movable panes all over the building, and, having done this, it is often thought that the ventilation is as complete as can be attained under the circumstances. This is not the case, however, and, just as the careful noting of the temperature of the air outside a building is the most effectual precaution in attending to winter heating and ventilation in churches, etc., so is the appreciation and knowledge of small wind currents of the greatest importance to the caretaker in keeping the atmosphere of his building in the best condition during the summer time. Furthermore, he should note just what the wind effects will be upon the windows, doors, and other outlets or inlets of the structure according to the direction from which the wind blows.

The church or public building which has a delicate wind vane will furnish him with the direction of the wind, and he will be able to know if the force is appreciable from the movements of the vane. All new buildings ought to be so provided, but an intelligent man who once realises the value of air currents, properly directed, will soon discover the quarter from which the wind blows.

What the effects of the wind will be upon the doors, windows, ventilators, etc., is a different matter, and one which has not been previously determined. The author made a series of experiments extending over some years, and the results obtained were partly communicated to the British Association at the Bradford and Glasgow meetings in 1900 and 1901.

Having noticed the violent effects of the wind upon housetops, and the ventilators upon churches, etc., against which the wind impinged as it moved over a square or public park in London, and not being able to explain the reason of these

effects satisfactorily, a series of experiments were carried out upon a perpendicular cliff and promontory by the seaside. The results of such experiments showed that air as it impinges upon a perpendicular surface sticks, so to speak, and forms a cushion against the walls of houses, etc. As the air strikes a perpendicular surface with considerable velocity it flattens out, and tends to rise upward above the top of the building. A further upward movement is imparted to the air near the ground level by the forward motion of the upper stratum of air flowing over the tops of the buildings, and, during a moderate breeze, the flattening out against a perpendicular surface together with the upward twists caused by the wind currents above, give sufficient vertical movement to the air to neutralise the forward pressure of the wind, and the author frequently found he could hold a thread of silk within about two feet from the upper edge of a cliff uninfluenced by the wind, showing that no movement of air occurred at that point. When the wind exceeded 10 miles an hour, the velocity of the forward motion destroyed the protected area close to the edge of the cliff, but further inwards the onward movement was much restricted.

The results obtained from these experiments which throw the greatest light upon the reason why the wind aspirates so powerfully upon the housetops and ventilators upon buildings situated in front of open areas, and in the teeth of the prevailing winds, are the following. When the wind is what is called gusty, it has short wind waves with high crests; when the wind blows steadily, the wind waves are longer, and the crests inappreciable. When the high crest of the wind wave is approaching the top of a building it gives rise to a very powerful upward movement, hence it is that chimneys or ventilators upon buildings some distance from the front walls have great upward suction exerted upon them during the time when the crest of the wind wave is being reached, whilst the next moment as the crest passes and the hollow is ap

proaching there is no suction at all. When the wind blows horizontally across the ventilators on a building, it meets with much friction, and its aspirating power is considerably weakened. A partly vertical movement, such as is caused by the high wind wave when the crest is approaching, escapes much of this friction, and the volume of air aspirated is greatly increased, so that when a gusty wind travels with high velocity, the air drawn out of a ventilator or chimney is vastly greater. The partial vacuum formed in a church or other building during the upward aspiration of the wind is very appreciable, hence the down draughts which result when the crest of the wave is passing are also very marked. These results explain why it is that gusty winds are so detrimental to ventilation. In addition to the vertical effects of high wind waves it is noted, from the cliff experiments, that the movement of the upper wind strata and that of the air flattened out by the impact of the wind against a perpendicular surface, hinder the forward motion of the wind. It will be seen that some twenty feet away from the front of the houses facing a park there is a protected area, and here the upward suction of the high wind wave will exert the greatest aspiration. Hence it is that the chimneys upon houses so situated are so smoky.

Further experiments showed that the small whirlwinds towns were local, and due to the twisting motion given to the air by bends in a street or crescent, and by suction upon other streets leading out of those along which the wind was blowing. These small whirlwinds have a powerful upward movement, generally due to increased velocity of the wind just above the ventilating area. When wind blows along a wide street (see M. S, Fig. 6) and there are openings into other streets, R R R, at right angles to the direction of the wind, air is aspirated out of the cross streets. But the wind travelling over the tops of the houses, H H H, and over the streets situated at right angles, R R R, also aspirates air out