continued to spin for thirty-five minutes. On being spun (after the manner of spinning a humming-top), on the table of an air-pump, it was covered with a glass receiver, from which the air was then removed, and the top continued to spin during the space of two hours and sixteen minutes. Mr. Roberts, of Manchester, a few years ago made a top which would spin in the air forty-two minutes. He made another, and, in order to give it a neat appearance, covered it with lacquer; when he found it would not spin more than seventeen minutes. He removed the lacquer, and the top continued to spin as at first. He found that the lacquer, although it improves the appearance of surfaces, yet imparts to them a vast number of minute roughnesses, scarcely, if at all, appreciable by the touch, yet sufficient to offer so much additional resistance to motion in the air. We will now return to the common form of the pegtop, and endeavour to explain the means by which the top is enabled to rise from the oblique position (which it always assumes more or less when first set spinning), into the truly vertical position, which produces the effect called sleeping, when the motion is so steady that it scarcely seems to move. When the top is sleeping, its centre of gravity is situated perpendicularly over its point of support; but in rising from an oblique to a vertical position, the top must have its centre of gravity raised. The force which effects this change has been a subject of contest in the philosophy of the peg-top, and we believe that Dr. Paris was the first to offer a satisfactory explanation of it. He considers it to depend upon the form of the extremity of the peg, and not upon any simple effect connected with the rotating or centrifugal force of the top. If the peg were to terminate in a fine (that is to say, in a mathematical) point, the top could never raise itself. Let ABC (fig. 1.) be a top spinning in an oblique position, having the end of the peg, C, on which it A X B spins, brought to a point. It will continue to spin in the direction in which it reaches the ground, without the least tendency to rise into a more vertical position and it is by its rotating force that it is kept in this original position for, if we conceive the top divided into two equal parts (A and B), by a plane passing through the line XC, and suppose that at any moment during its spinning the connection between those two parts were suddenly dissolved, then would any point in the part, A, fly off with the given force in the direction of the tangent, and any corresponding point in the part B, with an equal force in an opposite direction. While, therefore, these two parts remain connected together during the spinning of the top, these two equal and opposite forces, A and B, will balance each other, and the top will continue to spin on its original axis. Hence the rotating or centrifugal force can never make the top rise from an oblique to a vertical position. IM a Fig. 1. Fig. 2. E top will roll round But in order to be satisfied that the change in position depends on the bluntness of the point, let A BC (fig. 2.), be a top spinning in an oblique position terminating in a very short point with a hemispherical shoulder Pa M. It is evident that in this case, the top will not spin upon a, the end of the true axis, Xa, but upon O, a point in the circle P M, to which the floor IF is a tangent. Instead, therefore, of revolving upon a fixed and stationary point, the upon the small circle P M, on its blunt point, with very considerable friction. the force of which may be represented by a line, O P, at right angles to the floor I F, and to the spherical end of the peg of the top. Now, it is the action of this force, by its pressure on one side of the blunt point of the top, which causes it to rise in a vertical direction. Produce the line OP, till it meets the axis C; from the point C draw the line CT perpendicular to the axis a X, and T O parallel to it; and then, by a resolution of forces the line T C will represent the part of the friction which presses at right angles to the axis, so as gradually to raise it in a vertical position, in which operation the circle P M gradually diminishes, by the approach of the point P to a, as the axis becomes more perpendicular, and vanishes when the point P coincides with the point a, that is to say, when the top has arrived at its vertical position, where it will continue to sleep without much friction or any other disturbing force, until its voluntary motion fails, and its side is brought to the earth by the force of gravity. -S. M. THE PUMP. The PISTON ROC CYLINDER EVERY boy knows what a squirt is, and how it is used. You pull up a rod by a ring at the top, hold the nose in water, and then raising the squirt, push the rod down, and the water is forced out in a stream. Now, let us look into the squirt and see how this is done. I have cut a squirt straight down from the top to the bottom, that you can see how it works. rod that moves up and down is called a "piston-rod," because it works a round sort of button, called a piston, fixed on the bottom of it, and covered with thread so as to make it fit tightly to the sides of the barrel, and keep the air from passing between it and the barrel (or, as it is properly called, the cylinder). When you put the nose of the pipe into a bucket of water, no air can get in through Fig. 1. the opening, because the water closes the entrance; and as soon as you pull up the handle you leave the barrel empty of air, or cause what is called a vacuum. But as the air is pressing with great weight upon the surface of the water in the bucket, it pushes it up into the vacuum, till it is filled. You can easily try for yourselves how water will be forced up to fill a vacuum. Take a common tea saucer and fill it with water, and then get an empty tumbler; put a little bit of lighted paper into it, and turn it gently upside down with its mouth into the water, and you will see the water run up into the tumbler till it is nearly full; because you have burnt up most of the air in the tumbler and made a vacuuni. A pump is on the same principle. There is a piston and a piston-rod; the air is drawn out of it, a vacuum is made, and the water rushes up into the barrel or cylinder. But a pump is too heavy to be lifted out of the water each time to squirt it out, and it would be a great waste of time, if we could do so. Another contrivance is made, then, to send the water out of the top part of the barrel without letting in theair, and this is by means of valves. Let us put one of these valves into our squirt, just where the pipe of the nose goes out of the barrel. You see it fits like a of a bottle, and keeps any air or water that gets into the barrel from running down into the pipe. At the point A there is a hinge, so that this valve can open upwards like the lid of a box, but it cannot open Now, if we pull up the piston-rod, raises the valve, and fills the barrel. get it out? We might put a pipe into the barrel at B, and then, when the piston got above B, the water would run out, but only in a very small quantity, because air would come in at the spout and the water would stop rising in the barrel; so we make a spout higher up, above the piston, at C. We must now see the way in which the water gets above the piston. There is a valve again in the piston, and this valve only opens upwards. As soon as the piston has been drawn to the top, and the barrel below it is filled with water, we push the piston down again, and the pressure of water shuts the lower valve A, and as the piston goes down it forces up the valve in the bucket, D, (fig. 2), and the water rises into the upper part of the barrel. We now pull Fig. 4. the bucket up again, and a fresh lot of water rushes in below; but the weight of water above shuts down the valve in the bucket, and so it is raised as the bucket ascends, until it flows out of the spout. This will be easily seen by the two figures at the side. In fig. 5 the water has been raised by the first lifting of the piston to the top of the barrel. The bucket is now descending for some more, and the water below it is rising through the valve, A, D B into the upper part of the barrel, the pressure of the water closing the valve B. We now raise the bucket again by another stroke of the handle, water |