is constantly diminished by the resistance of the wire. This electricity moving more slowly than when the coil and battery were connected, and in the same direction as the battery current, is not able to maintain in the core, or in the primary or secondary coil, the magnetic power produced in them by the battery; but it maintains a part of it, and prevents the core, the primary and secondary coil, from losing their magnetism in an instant, and consequently diminishes the intensity of the secondary current. The condenser prevents the gradual diminution of the velocity of the electricity flowing in the primary coil at the moment its connexion with the battery is broken, and probably accelerates it; for in an instant after the battery connexion is broken, the end of the coil towards which the electricity is moving, and the plate of the condenser connected with it, become positive. This positive plate instantly renders the other plate negative: the latter then attracts electricity to the former with an enormous force, on account of their very close proximity; and if the plates of the condenser be of sufficient size, the electricity moving in the primary coil will be drawn with such force to the positive plate, that its velocity, instead of being diminished, will probably be increased. Thus the condenser removes the obstacle arising from the electrical state of the primary coil, to the sudden suspension of the magnetism of the core, and of the primary and secondary coil, and probably increases their magnetic power by accelerating the current in the primary coil after its connexion with the battery is broken. The condenser not only removes an obstacle to the instantaneous suspension of the magnetism of the core and of the primary and secondary coil, but it also supplies a force tending to destroy that magnetism; for as soon as all the electricity moving in the primary coil is drawn to the positive plate of the condenser, it instantly rushes back to the negative one through the primary coil, and is drawn to itself by that plate with an immense force; and in its passage through the primary coil, tends to reverse the magnetic poles of the core of the primary and secondary coil, and consequently to destroy their magnetism. Hence the effect of the condenser is to make the core, the primary and secondary coil, lose their magnetism instantaneously, and thus to increase the intensity of the secondary current, or the length of the sparks produced by that current. This explanation of the action of the condenser is confirmed by the effect which I have found it to produce on the electrical currents induced by the primary current in each section of the core, viz. an increase of their intensity. Now the only causes that can produce an increase of intensity, in these currents are, an increase of the magnetism of the core, and of the rapidity with which it loses its magnetic power, or either of these two causes. I have shown that the effect of the condenser is at least to increase the rapidity with which the core loses its magnetism, and probably to increase its magnetic power. Hence my explanation of the action of the condenser is confirmed by the effect of the condenser on the currents produced in each section of the core. The principle of this explanation of the action of the condenser suggested to me a new form of condenser, which I expected would act more powerfully than the condenser now in use. The new condenser was to consist, not of sheets of tinfoil, but of a large number of very thin sheets of iron, arranged in such a way that one-half of them would form one plate, and the other half the other plate of the condenser; and that the electrical current by which each plate would be charged, one positively, the other negatively, should not enter simultaneously each of the iron sheets forming the positive plate, nor leave simultaneously all the iron sheets in the negative one, but should flow through the whole length of each sheet, before entering into the next. In order to make a condenser of this kind, I got 112 sheets of iron, each 28 inches long, 10 broad, and about th of an inch thick. I intended to arrange them so that the current by which they would be charged, at the moment the connexion between the primary coil and battery would be broken, should flow successively through the whole length of the 112 iron sheets, or through one plate equal in length to the sum of their lengths, which exceeds 250 feet. Had I had time to make, as I intended, our iron condenser in this way, the iron plates would be strongly magnetized by the electrical current flowing through their entire length; and in losing their magnetism, would produce a powerful secondary current, tending to destroy or to reverse the magnetism of the core, and thus increase the intensity of the secondary current. In the ordinary condenser there is one electrical current tending to destroy the magnetism of the core: in an iron condenser made as I have described, there are two currents tending to produce the same effect; viz. the current arising from the rush of electricity from the positive to, the negative plate of the condenser, and the current caused by the demagnetization of the iron plates. In order to save time and trouble, I made our iron condenser in the ordinary way. When I have leisure I may make it in the manner I have just described. The seventh result consists in the discovery of some new facts regarding the condenser, which have not been hitherto noticed in any publication. First, I have found that the action of the condenser is feeble when the core is a solid bar of iron; secondly, when it is a coil of fine insulated iron wire not having a bundle of iron wire, or a coil of thick covered iron wire in the hollow part of it; thirdly, when the quantity of iron in the core is very great compared with the thickness of the primary wire. Secondly, I have found that the size of the condenser must be increased with the conducting power of the primary wire. Thus a thick primary wire requires a larger condenser than a thin one; a primary wire of copper requires a far larger condenser than one of iron of the same length and thickness; and a very short primary wire of any metal requires a condenser very much larger than that which is necessary for one of the same metal 100 feet long. I have found that when the primary wire is not more than 30 feet, a condenser of moderate size will not produce the slightest effect on it. Thirdly, I have found that when a condenser is very much larger than that which is required to produce the full effect of a condenser on a given coil, it not only does not increase the power of the coil, but it makes it less than it would be without a condenser, and sometimes destroys it; and that in general there is a limit to the size of the condenser, beyond which its effect on the coil will be diminished. Fourthly, I have found that a condenser so large as to diminish the power of a coil excited by one cell, will increase its power when the coil is excited by ten or twelve cells. Hence the same condenser will not answer for the same coil when batteries of very unequal powers are used. Then, every condenser should be made in such a way that the entire of it will produce the full effect of a condenser on the coil for which it is intended when the largest battery we wish to use is employed, and that a small or a large part of the condenser may be used when we wish to excite the coil by a weak or strong battery. I learned from Mr. Gassiot and M. Foucault during the late meeting of the Association, that they were aware of the necessity of making the condenser in this way. Maynooth College, Sept. 29, 1857. P.S. I have abstained from saying anything about the primary coil, because my experiments on it have not led me to a satisfactory conclusion, and not because I think the primary coils in common use incapable of improvement. I believe that they are very badly calculated to attain their object, and that they have been made on a false principle. XXXIX. On the Gyroscope. By JOHN BRIDGE, M.A.* THE gyroscope, as usually constructed, consists of a solid of revolution, whose axis forms a diameter of a circular ring into which its extremities a, a' are inserted: this ring is move* Communicated by the Author. able about a diameter at right angles to the axis of the solid by means of pivots b, b' inserted into another ring; and this second ring is moveable about a fixed vertical diameter c c' at right angles to bb'. The facts to be explained are the following 1. When the solid is in rapid rotation about a a', let a weight be hung to the first supporting ring at a. Then, instead of an accelerated motion about bb', we have a uniform motion about cc. The motion of a is in a direction a right angle in advance of that in which the weight alone would have caused it to move. The rate of this precessional motion about the axis cd does not depend on the elevation of the axis a a', but is increased when the force applied is increased, or when the rate of rotation of the solid of revolution is diminished. 2. When the outer ring, or the axis b b', is fixed, a a' will only be moveable in a vertical plane. If in this case, while the solid is in rapid rotation, a force be applied at a, it will produce very nearly the same effect as if the solid had been previously at rest. 3. If a a' be allowed to move with great freedom in one plane only, fixed with respect to the earth, the axis a a' will, during the rotation of the solid, oscillate about the meridian line of the plane in a manner similar to a common pendulum, the time of oscillation being a minimum when the plane in which a a' lies is parallel to the earth's axis. The following investigation contains an approximate solution of these problems. From Earnshaw's 'Dynamics,' art. 257, it follows, that if C be the moment of inertia about the axis of figure of a solid, and A that about any other principal axis, when the body is set in rotation with velocity w about an axis making an angle with the axis of figure, the effect of the centrifugal forces is the same as that of a couple, from which the angular acceleration of the C-A body originally at rest would be w2 sin 0 cos 0, tending Α to bring the principal axis towards the position of the instanta C-A A neous axis. If be small, this is nearly equal to -w20, which I will call co20. The First Problem. Let A be any point of the horizontal great circle on a sphere concentric with the solid, i the extremity of the instantaneous axis, a the extremity of the principal axis. , the spherical coordinates of i. Also let ia=0, and let ia make an angle x with the vertical circle through i. Then a being the angular acceleration which would result from the given weight if it were applied at right angles to the axis at a, we have (Airy's Tract on Precession, art. 12), since the change in @ arises entirely from the change in position of i, and not from the rotation about i. And the change in x resulting in part from the rotation about i, and in part from the change in position of i. By substituting (A) and (B) in (C) and (D), we obtain If 0 cos x=u and @ sin xv, the preceding may be replaced by the system of equations |