A good deal of interest has been shown recently in the statement that grounding a wireless telegraph station impairs considerably the efficiency of the station. The London " Electrician " in its issue of December 29, gives an abstract of a recent German paper by Herr J. S. Sachs, detailing the author's experiments to arrive at some definite conclusions in this matter. The work was suggested by Prof. Drude, and the system employed was due to the latter. It consisted of a primary circuit containing a spark-gap and a secondary circuit with its ends connected, respectively, to an air wire and a balancing capacity. The primary was a single turn of thick wire interrupter with a spark-gap. Its condenser was made of lead-foil fastened to a glass plate. The secondary was inside the primary and possessed ten well insulated wires wound in a single layer on an ebonite ring 8.7 centimeters in diameter.
The air wire consisted of a brass tube three metres long and 1.4 centimetres in diameter. The balancing capacity was a metal plate. The coils were designed to resonance, and the wave-length was computed to be thirty-one metres. The receiver differed from the sender only in having a constantan - iron thermo-couple in the position of the sender's epark-gap. The experiments were generally carried out in the open court in front of the Physical Institute of Giessen. The distances betweeu sender and receiver varied from twenty-five to fifty metres. Headings were taken by galvanometers placed inside the institute building, to which wires were run from thermo-couple. In order to make allowance for the variableness of the energy radiated from the sender a coil called a standard coil, not in resonance with the sender, was kept in a fixed position relative to the sender's winding and was provided with a thermo - junction similar to that in the receiver. This junction was also connected to a galvanometer.
The mode of experimenting was to pass a current through the induction coil for the same period in each observation of a set. The deflections of the galvanometers connected with the two thermo-couples, divided the one by the other, gave the figures which were used in drawing conclusions.
In this manner were investigated the effect of varying the relative positions of the various parts of the sending apparatus, and the effect of symmetry of the two ends. Another series of experiments was performed to contrast the behavior of the receiver when sender and receiver were wholly insulated, with its behavior when the balan ing plates at both ends of the thirty-metre stretch were connected to a plate buried one metre in the gronnd. When insulated the balancing-plates were one metre above the ground. It was found that the readings of the galvanometer were about twice as great when the whole apparatus was insulated as when the apparatus was earthed.
In order to examine the influence of the earth on the propagation of waves the author used the principle that this influence must alter with alteration of the apparatus above the ground. In dry, frosty weather, with air wires vertical and balancing-plates horizontal the maximum efficiency occurred when the apparatus was three metres above the ground. The effect at this height was more than four times the effect when the plates were only ten centimetres above the ground. On the other hand, in wet, cloudy weather, the maximum occurred at about one metre elevation, the effect at this height being about two and one-half times the effect at the height of ten centimetres. In both cases the maximum value was maintained with slight change, on further raising the apparatus to four or five times above the ground.
Experiments conducted to show the effect of distance on transmission gave values obeying the inverse square law. The distances varied between twelve and twenty-four metres.
The author's conclusions are stated thus: The earth's surface is, for waves of thirty-one metres, a strong absorbing and a weak reflecting medium. The connection to the earth of sender or receiver is greatly prejudicial to transmission. Insulating it is decidedly favorable. It is desirable to install the apparatus as high above the earth as possible. The integral effect at the receiver varies inversely as the square of the distance of transmission.-"Electrical Review."