The mode by which this is generally accomplished in England is by fixing a copper terminal rod (four or five feet long), on to the side of the top of the chimney shaft. This method is open to one serious objection: if the wind should happen to blow the stream of smoke or heated vapour in a direction opposite to the terminal rod, the electric discharge might go down the chimney shaft and effect considerable damage. By far the best plan is that shown in fig. 35. It consists simply of an iron or copper cap, to the centre of which is attached the terminal rod. This latter, however, is by no means essential, and may be said to be merely placed on the top for ornament. A structure of such small circumference really wants no terminal rod, the most important thing being to provide a copper rope or band conductor of sufficient size to carry any electric discharge in safety to the ground. It will conduce greatly to the strength and stability of such a conductor if it be built up together with the chimney shaft, and fastened into the brickwork by clamps on the plan shown in fig. 36. A conductor of this kind should be made of copper rope or band of much greater calibre and weight than that used for ordinary buildings. That made of seven solid wires twisted together (see fig. 37) being the best. [Illustration: Fig. 36.] A theory propounded some years ago by the late Prof. Clerk Maxwell, F.R.S., one of the most eminent physicists in Europe, deserves some notice here, perhaps more from its ingenuity than its practical accuracy. On investigation, it proves to be a revival of an old presumption that it is possible to protect a powder magazine or other building from the effects of lightning by having its roof, walls, and ground floor surrounded with a covering of sheet metal, or a network of lightning conductors, and disconnecting the said covering or network from the earth, or even insulating it by means of a layer of asphalt or some similar substance. Prof. Clerk Maxwell argues that the presence of a lightning conductor induces a larger number of electric discharges in its immediate neighbourhood than would occur provided no conductor was present, although at the same time these discharges are rendered less intense and smaller by reason of the existence of the conductor. Therefore, it is possible that fewer discharges take place in the area just outside the radius of the conductor. Reasoning from this, Prof. Clerk Maxwell considers that an ordinary lightning conductor tends rather to mitigate the accumulation of electricity in the clouds than to protect the building on which it is placed. [Illustration: Fig. 37.] He says: ‘What we really wish to prevent is the possibility of an electric discharge taking place within a certain region--say, in the inside of a gunpowder manufactory. If this is clearly laid down as our object, the method of securing it is equally clear.