668 Professor J. A. Fleming [June 4, 



These dischargers, however, have not stood the test of prolonged 

 practical use, and we cannot say therefore that they are comparable 

 in value for telegraphic purposes with the well-proved inventions of 

 Mr. Marconi. 



In connection with spark telegraphy it has been clearly seen, 

 lately, that much can be done by attention to details of construction 

 to increase the number of oscillations in each wave train in the case 

 of spark apparatus, in other words, to lessen the damping by obviating 

 energy losses in all parts of the apparatus. It is not a matter of 

 indifference what kind of glass we use in Leyden jars or what form 

 of stranded wire we employ in oscillation transformers, or type of 

 spark discharger. By appropriate selection of apparatus, we can 

 considerably increase the number of oscillations in clamped trains of 

 small amplitude, and therefore increase the possibilities of utilising 

 the principle of resonance. 



Before leaving the subject of the antenna we may notice some 

 recent improvements in directive antenna, that is, in devices for 

 more or less confining the radiation to one direction, and for locating 

 the position of the sending station. 



In a previous discourse explanations were given of the property 

 of a closed or partly closed antenna of radiating more in some direc- 

 tions than others, and the action of Marconi's bent antenna was 

 described. Two other inventors, Messrs. Bellini and Tosi have taken 

 advantage of this fact to construct antennas of a very interesting 

 character. They erect an antenna consisting of tAvo wires, each bent 

 into a triangular form, the top ends nearly meeting, the planes of 

 these triangles being at right angles to one another, and both of them 

 vertical. The nearly closed antenna circuits are then inductively 

 coupled with a condenser circuit, which is capable of being swivelled 

 round in various directions. If the said condenser circuit is placed 

 in such a position as to be coupled with one of the triangular antenna, 

 it will cause the maximum radiation to take place in the plane of that 

 antenna, but none at all at right angles to it. If it is coupled with 

 the other antenna, it will cause radiation to take place to a maximum 

 degree in the plane of that second antenna. If, however, the 

 oscillatory circuit is placed in an intermediate position, so as to act 

 inductively upon both the nearly closed triangular antennae, then it can 

 be shown both mathematically and experimentally that the radiation of 

 the combined system is a maximum in the direction of the plane of 

 the oscillatory circuit which is coupled with the antenna. Hence, 

 with such a combined antenna, we have it in om* power to create 

 radiation most strongly in one direction, although not entirely sup- 

 pressed in all other directions. By combining together, however, a 

 single vertical antenna with two nearly closed circuit antenna at right 

 angles to one another, Messrs. Bellini and Tosi have constructed a 

 complex antenna which has the property of producing radiation 

 almost entirely limited to one-half of the circumjacent space (see 

 Fig. 12). It therefore corresponds to a certain extent in effect to 



