If, however, by a suitable adjustment of capacity and inductance, we make the natural time-period of oscillation of the receiving aerial circuits agree with those of the transmitting aerial, within certain limits the former will only be receptive for waves of the frequency sent out by the transmitter. It is quite easy to illustrate this principle by numerous experiments. It can be done by means of an apparatus devised by Dr. Georg Seibt for showing in an interesting manner the syntonisation or tuning of two electric circuits. This consists of two bobbins, each consisting of one layer of insulated wire wound on a wooden rod (see Fig. 22). Each of these bobbins has a certain electrical capacity with respect to the earth, when considered as an insulated conductor, and it has also a certain inductance. If, therefore, electromotive impulses are applied to one end of the bobbin at regular intervals, electrical oscillations will be set up in it, and, as already explained, if these are timed at a certain rate, the bobbin will act like a closed organ-pipe to air impulses and oscillations of potential will be accumulated at the opposite end, which have much greater amplitude than the impressed oscillations at the end at which they are applied. We can make the existence of the amplitude oscillations of potential evident by attaching to one end of the bobbin a vacuum tube, which will be illuminated thereby, or by terminating it by a pointed piece of wire, so that an electrical brush can be formed at the point, if the potential variations have sufficient amplitude. We arrange also another closed oscillation circuit, consisting of two Leyden jars and a variable inductance coil and a pair of spark balls which are connected to an induction coil. In this manner we can set up oscillations in the discharge circuit of these Leyden jars, and we can vary the time period by altering the inductance and the capacity. If we denote the capacity of the jars in the microfarads by the letter C and the inductance in centimetres of the discharge circuit of the jars by the letter L, it can then be shown that the number of oscillations per second denoted by _n_ is given by the expression--[61] n = (5,000,000,000) / ([\sq]{CL}).