522 



NATURE 



[June 24, 1920 



the Western Electric Co. conducted a great ex- 

 periment from the American naval station at 

 Arlington, near Washington, and succeeded in 

 speaking to the Eiffel Tower (3700 km.), and 

 exceptionally to Honolulu (8000 km.). Triodes 

 were used as oscillators, modulators, and 

 magnifiers. The chain of apparatus was as fol- 

 lows : First came a triode oscillator of small 

 power, which was coupled by means of a high- 

 frequency transformer to the grid circuit of a 

 medium-sized tube. This grid circuit contained 

 also the secondary winding of a transformer, the 

 primary of which carried the currents from the 

 microphone. The anode circuit of this medium 

 triode therefore contained magnified modulated 

 current of the high frequency dictated by 



Fig. 7. 



the amall triode oscillator. The current was 

 next transferred by means of a transformer from 

 this anode circuit to the grid circuit of a bank 

 of medium-sized amplifying tubes connected in 

 parallel, and was again magnified, and finally it 

 was transformed once more into a circuit con- 

 taining the grids of more than 500 parallel con- 

 nected tubes. It ought to be remarked that great 

 progress has been made since 1914 in the manu- 

 facture of power bulbs, and that the experiment 

 can now be carried out on a larger power scale 

 with a far smaller array of tubes. 



The remaining example to be cited is the re- 

 cently erected 20-kw. plant of the Marconi Co. 

 During the past few weeks it has transmitted 

 good quality speech to a distance of 1500 miles. 

 In principle it is of the subtraction or absorption 



type. A diagram of the chief parts appears 

 in Fig. 7, from which are omitted all details 

 regarding the heating currents for the filaments 

 and concerning the rectifying of the high-voltage 

 current for the anode circuits. The oscillations 

 are generated in the circuit LC by a bank of six 

 large three-electrode valves in parallel marked O^, 

 and transferred to the antenna by the coupling fe. 

 The absorption tubes ire three in number, and 

 are shown at A3; their ihree anodes are large 

 enough to dissipate all the energy normally given 

 to the antenna. These absorption triodes are 

 controlled by applying to their grids the speech 

 electromotive forces after these have been magni- 

 fied by the successive triodes Mj and Mg. The 

 total consumption of power is 20 kw., including 

 all that necessary for 

 heating the filaments, the 

 height of the aerial is 

 400 ft., and the wave- 

 length 2750 metres. 



It will be noticed that 

 the above are all one- 

 way methods — that is to 

 say, the two persons 

 using two stations for 

 conversation must speak 

 in turn, and the listener 

 must wait for the 

 speaker to finish before 

 he switches over from 

 his listening circuit to 

 his speaking circuit. 

 This falls far short of 

 perfection. For perfect 

 telephony it is essential 

 that both persons shall 

 be able to speak and hear simultaneously if they 

 so desire, as in ordinary line telephony. For in- 

 formation about the most modern attempts at 

 duplex telephony a paper read recently before the 

 Institution of Electrical Engineers by Mr. P. P. 

 Eckersley should be consulted. Until a com- 

 plete duplex method is devised and proved, 

 wireless telephony must remain a somewhat irk- 

 some mode of communication for public use. The 

 ideal method will be such that a wireless station 

 on this side of the Atlantic could be connected 

 over land lines to, say, London, and a station on 

 the other side connected by land lines to New 

 York, and the persons using the apparatus would 

 be unable to tell that wireless telephony across 

 the ocean had been an agency in the transmission 

 of their voices. 



The Meteorology of the Temperate Zone and the General Atmospheric Circulation. 



By Prof. V. Bjerknes. 



TN Norway, since the year 19 18, an attempt 

 *~ has been made to base forecasts of weather 

 on the application of a very close network of 

 meteorological stations. The study of the corre- 

 sponding very detailed synoptic charts has led to 

 interesting results even for large-scale meteor- 

 NO. 2643, VOL. 105] 



ology. These are due especially to three young 

 meteorologists, J. Bjerknes, H. Solberg, and 

 T. Bergeron, who have been attached to this 

 service, and will return to the subject in detailed 

 papers. 



A very short summarv of some of the main 



