you saw, gyration is absolutely essential to the effect which Mr. Boys thinks accounts for its stability.
We may, I think, dismiss the top from further consideration; but there is another instrument apparently much closer in its relation to the bicycle. I mean the gyroscope, or rather that form of it which Sir William Thomson calls a gyrostat. Its wheel is upright like the bicycle's (see Figs. 3 and 4). The lower part of
the ring which supports the wheel rests in a kind of trough, to the bottom of which is attached crosswise a piece of metal (best seen in Fig. 3) curved on the lower edge, and with two projecting wires by which it may be drawn back and forth in the plane of the wheel.
I now set the wheel in rapid motion—much more rapid than any bicycle-wheel can go; I place it on a smooth, hard surface—I have here a pane of glass—and leave it to itself. It begins at once, as you see, to revolve around a vertical axis. If it leans little, it revolves slowly; if it leans much, it revolves faster. It will not fall to the table, though I push it, or strike a hard blow. It resists with remarkable force. I now take it by the projecting wires and attempt to make it move in a straight course, as a bicycle does when it spins along the road. Instantly it falls. The rotation of the wheel on its axis was not in the slightest degree interfered with, but the stability vanishes the moment the rotation around the vertical axis ceases. Invariably it falls. Yet you observe the conditions are far more favorable for the effect of gyrostatic action than in the bicycle, for the mass of the rim of our gyrostat is many times heavier in proportion to its size, and its speed incomparably greater. I try the experiment over and over, the result is always the same. No amount of skillful management will make the instrument stay up for an instant if it has to move in a straight line. I submit that these experiments are proof positive that the sustaining power of the bicycle does not come from any gyroscopic action.