Some attempts have been made to displace the induction coil in wireless telegraphy altogether by a specially made dynamo. These machines can produce either alternating or continuous currents, in fact the alternating current dynamo is really simpler than the more familiar continuous-current machine. The difficulty is, however, to run it sufficiently fast to produce sufficiently rapid alternations. Nicola Tesla made an alternator (to give the alternating current dynamo its short title) which could produce 1500 alternations per second, while Mr W. Duddell made one which produced 120,000, but neither was satisfactory for the work in question. Could such a machine be made, it would be invaluable, for it will be apparent that a continuous succession of waves would be formed by it and not a succession of short trains of waves such as is produced by the induction coil and spark-gap. The difficulties are not electrical, but mechanical. It seems doubtful if a machine will ever be made to run with sufficient rapidity which would not knock itself to pieces in a very short time. [Illustration: FIG. 11.--The simplest form of wireless antenna.] Small alternators are used sometimes, however, to supply alternating current to the primary of an induction coil, or transformer, as it is more often called in its larger sizes. The interrupter is only needed when the primary current is continuous--from batteries, for example. Alternating current needs no interrupter, and so that bother is removed. The alternations of a hundred or so per second, which are quite the common thing with alternators, are just what is needed to excite an induction coil. Consequently small machines of this kind are to be found in many stations. A Danish inventor, Valdemar Poulsen, has adopted an altogether different method of producing electrical oscillations, which method is the distinctive feature of his mode of telegraphy. He takes advantage of a curious effect of passing current between two rods, one of which is carbon, so as to form an arc such as we see in arc lamps. My readers are already familiar with the term "shunt" in connection with electrical matters, and so will perceive at once what is meant when a second circuit is said to be arranged as a shunt to the arc. The accompanying diagram will in any case make the matter clear. The current comes along from the battery or continuous-current dynamo to a hollow rod of copper which, to prevent it being melted, has cold water continually circulating inside it. Thence the current jumps across to a carbon rod, forming an arc between the two rods, and returns whence it came. In its journey it traverses the coils of an electro-magnet, the poles of which are one each side of the arc. This tends to blow the arc out, as a puff of wind blows out a candle, an effect which a magnet always has upon an electric arc.