the higher up you rise in the air, and thus by means of the barometer you can tell to what height you have gone. The barometer is also useful in telling us when bad weather is at hand. When the barometer falls, that is to say, when the top of the column of mercury gets lower in the tube and especially when it falls quickly, we may expect bad weather. On the other hand, if the mercury remains steady and high we may expect a continuance of fine weather.

33. Air-pump.-We have already spoken about taking the air out of a jar, how this is done by the air-pump. You will see how this instrument acts by means of the figure. But first of all I must tell

Fig. 17.

you what is meant by a valve. A valve is just a tightly fitting trap-door that closes a hole, and that can only open in one way-upwards, for instance. You have, most of you, seen trap-doors in floors that open upwards. Now in the figure you see to the left a belljar full of air, which fits tightly upon a plate. You see too coming out from the middle of the plate a tube which opens into the bell-jar on the left side, and into the cylinder or barrel on the right, and thus connects the two together. You see also a piston or plug that can move up and down in the cylinder or barrel.

Finally, you see two valves or small and tightly fitting trap-doors, one of which is placed where the tube enters the bottom of the cylinder, while the other is in the piston itself. Both of these valves open upwards and not downwards.

Now suppose we start with the piston at the bottom of the cylinder, and the valves shut, and begin to pull the piston up. In doing so we make an empty space which the air on all sides will try to fill up if it possibly can (Art. 29). The air from above will try to press into this space, but it will not be able to get in, and all it can do will be to press against the outside of the upper valve and keep it tightly shut, since the valve does not open downwards. The air from the bell-jar will succeed better, for it will rush through the tube and press open the lower valve which opens upwards, and then get into the empty space. Let us now suppose that we have got the piston to the top of the cylinder, and that we are beginning to press it down. The push that we give to the piston, the piston gives to the air; and the air in its turn communicates this push to the lower valve, which is kept shut. But the air within is more successful with the upper valve, for it pushes this open; and so, as we continue to push down the piston, all the air that was in the cylinder below it is pushed out through the upper valve or trapdoor. But this air which we have pushed out was part of the air that was originally in the bell-jar, so you see that in the first double or up-and-down stroke of the piston we have succeeded in squeezing out part of the air of the jar. Let us now repeat the same process, that is to say, raise the piston again, and the air from above will shut the upper valve, while the air from the

bell-jar will rush along the tube, push open the lower valve and fill the empty space which we make when raising the piston. And when the piston descends once more, the lower valve is kept shut, while the air within pushes open the upper valve and gets out, and thus in every double stroke we get rid of part of the air in the bell-jar. Of course it is quite necessary in working the pump that the piston shall fit quite tightly into the cylinder; for, if not, the air will get in from without, and therefore we shall not succeed in getting the air out from within. I have now told you the way in which the air-pump works, but you must not expect every air-pump to be precisely like the figure I have given you; the principle, however, of all air-pumps is the same, although the appearance may be very different in each. 34. Water-pump.-Having now told you about the air-pump, let us return for a moment to the barometer. You have seen how the pressure of air is just strong enough to hold up a column of mercury about thirty inches high. But water is much lighter, bulk for bulk, than mercury, and we might therefore expect the pressure of the air to hold up a much longer column of water than one of thirty inches. In truth, the pressure of the air will hold up a column of water very nearly thirty feet in height.

This will enable you to understand the mode of action of the common pump. In the figure on the next page you have a sketch revealing the interior of such a pump. Below we have the reservoir from which we wish to pump the water up, and we have a tube leading from this reservoir up into the barrel of the pump. In this barrel you see a piston which fits tightly into the

barrel, and in this piston there is a valve opening upwards, while at the bottom of the barrel there is another valve also opening upwards. In fact, the barrel of the ifting pump is quite similar to that of the air-pump, and we may begin by supposing that the piston is at the bottom of the cylinder. Let us now raise up the piston, and just as in the air-pump, the air above will press down the upper valve and keep it shut. The air in the tube will on the other hand rush up through

the lower valve in order to fill up the empty space made by raising up the piston. When we lower the piston again, just as in the airpump, the lower valve will be shut, and the valve in the piston will open and let out some air. In fact, we are now pumping out the air from the barrel and the tube. But meanwhile, what is the water in the reservoir doing? The air from without continues pressing on the surface of the water in the reservoir; but as we have been taking away the air in the tube, this pressure of outer air is no longer counterFig. 18. balanced by that of the air in the tube; the outer air will therefore find itself unopposed, and will drive up the water into the tube, until at last, when all the air is taken away, the whole tube will be filled with water. This water will then enter the pump barrel through the lower valve.

But all this will not take place if the distance between the surface of water in the reservoir and the

lower valve be more than thirty feet. For you have just been told that the pressure of the air will support a column of water thirty feet high, but if the column be higher than this it will not support it. So that if there be a greater distance than thirty feet between the surface of the reservoir and the pump barrel, the water will refuse to enter into the barrel, and do what you can you will not be able to entice the water quite up into the barrel. If, however, the distance be not more than about twenty-six or twenty-seven feet, the pump will work well, and you will get the water to enter the barrel. Suppose now that you have got the barrel filled with water, and that you are pressing down the piston. As you do this the pressure you give the piston will be communicated by the water to the lower valve, which will be kept closed. On the other hand, the pressure of the water will force open the upper valve which opens upwards, and the water will get above the piston. Next time when you pull up the piston, you will pull up this water with it, and it will empty itself through the spout of the pump, and the water will now come out of the spout at every stroke.

* EXPERIMENT 32.-To enable you to see with your own eyes what goes on in a common pump, take a model in which the pump barrel is made of glass, so that you can see into it. You will thus see that when we raise the piston, the upper valve shuts and the under one opens, while, as the piston descends, the under valve shuts and the upper valve opens. You quite understand that the piston of the pump must fit tightly on to the barrel, because otherwise the air will get in from above and prevent the action. Sometimes,