2;6 



NA TURE 



[April 22, 1909 



I have often been asked why I adhered to the practice 

 of transmitting series of the letter S for these tests. The 

 reason is that the switching arrangements at the send- 

 ing station at Poldhu were not constructed at that time 

 in such a manner as to withstand long periods of opera- 

 tion — especially if letters containing dashes were sent — 

 without considerable wear and tear, and that if S's were 

 sent an automatic sender could be employed. Moreover, 

 the immediate object of these experiments was -not' to 

 transmit actual messages across the ocean, but to ascertain 

 the possibility of detecting the elTects of electric waves at 

 a distance of 2000 miles. 



The result obtained, although achieved with imperfect 

 apparatus, was sufficient to convince me and my co-workers 

 that by means of permanent stations (that is, stations not 

 dependent on kites or balloons for sustaining the elevated 

 conductor) and by the employment of more power in the 

 transmitters it would be possible to send messages across 

 the Atlantic Ocean with the same facility with which they 

 were being sent over much shorter distances. 



About two months later, in February, 1902, further tests 

 were carried out between Poldhu and a receiving station 

 on board the American liner Philadelphia, en route from 

 Southampton to New York. The sending apparatus at 

 Poldhu was the same as that used for the Newfoundland 

 experiments. The receiving aerial on the ship was fixed 

 to the mainmast, the top of which was 60 metres above 

 sea-level. As the elevated conductor was fixed, and not 

 floating about with a kite, as in the case of the Newfound- 

 land experiments, good results were obtained on a syntonic 



receiver, and the signals were all recorded on tape by the 

 ordinary Morse recorder. On the Philadelphia readable 

 messages were received from Poldhu up to a distance of 

 1551 miles, S's and other test letters as far as 2099 miles. 



Although I never had the slightest doubt in my mind 

 as to the genuineness of what was accomplished between 

 Poldhu and Newfoundland, the results obtained on • the 

 Philadelphia amply prove that the station at Poldhu was 

 capable at that time of transmitting signals to a distance 

 of at least 2000 miles, which is the distance separating 

 Cornwall from Newfoundland, and that if it was practic- 

 able to send a message over 2000 miles of sea from shore 

 to jhip, it should also be practicable to send it over the 

 same space of ocean from shore to shore. 



A result of some scientific interest which I first noticed 

 during the tests on the s.s. Philadelphia was the very 

 marked effect of sunlight on the propagation of electric 

 waves over great distances.' 



At the time of those tests I was of opinion that this 

 effect might have been due to the loss of energy at the 

 transmitter by daytime, caused by the dis-elect'rification 

 of the highly charged transmitting elevated conductor 

 operated by the influence of sunlight. I am now inclined 

 to believe that the absorption of electric waves during dav- 

 time is due to the ionisation of the gaseous molecules of 

 the air effected by ultra-violet light, and as the ultra- 

 violet rays which emanate from the sun are largely 



1 Proc. Rov. Soc, Ixx. 



"A Note . 

 :tic Impuls, 



absorbed in the upper atmosphere of the earth, -it -is prob- 

 able that the portion of the earth's atmosphere which-is 

 facing the sun will contain more ions or electrons than 

 that portion which is in darkness, and therefore, as Prof; 

 J. J. Thomson (Phil. Mag., August, 1902, Ser. 6, iv., 

 p. 253) has shown, this illuminated and ionised air will 

 absorb some of the energy of the electric waves. 

 • The fact, remains that clear sunlight and ■ blue skies, 

 though transparent to light, act as a kind of fog to powerful 

 Hertzian waves. Hence the weather conditions prevailing 

 in this country are usually suitable for long-distance wire- 

 less telegraphy. 



Apparently the amplitude of the electrical oscillations and 

 the lengths of waves radiated have much to do with the 

 interesting phenomena, small amplitudes and long waves 

 being subject to the effect of daylight to a less degree than 

 large amplitudes and short waves. I never considered that 

 this daylight effect .would be an insuperable obstacle to 

 Transatlantic telegraphy, as sufficient sending energy could 

 be used during daytime to riiake up for the loss of range 

 of the transmissions. 



Turning again to Newfoundland, I ought to add that 

 the experiments could not there be continued or extended 

 in consequence of the hostile attitude of the Anglo-.American 

 Telegraph Company, which claimed all rights for tele- 

 graphy, whether wireless or otherwise, in Newfoundland. ' 



FICA 



NO. 2060, VOL. 



However, us I had received an offer of assistance from 

 the Canadian Government, it was decided to resume the 

 tests between Great Britain and Canada, and these tests 

 were very greatly facilitated by the subsidy of 16,000/. 

 granted by the Canadian Government to support my ex- 

 periments. The construction of another long-distance 

 station was, therefore, commenced at Glace Bay, in Nova 

 Scotia, and .very extensive, tests and experiments were 

 carried on with Poldhu during the latter part of 1902. 



Contemporaneously with the construction of the station 

 at Glace Bay, alterations and modifications were executed 

 at Poldhu. Four wooden lattice towers, each 210 feet high, 

 were erected at the corners of a square of 200-feet side. 

 The towers carried insulated. triatic stays, from which was 

 suspended a conical arrangement of four hundred copper 

 wires forming the aerial, put. up in sections, so that more 

 or less could be employed (Fig. 8). The buildings for the 

 generating plant were placed in the middle of the space 

 between the . to\ycrs. Additional machinery was obtained; 

 and alterations carried out in accordance with the experience 

 obtained from' previous tests. 



Identical towers and aerial arrangements were at that 

 time adopted at- the stations at Glace Bay, and at the 

 similar installation in course of erection at Cape Cod, 

 Mass. 



In most of the experiments carried on from Poldhu the 

 capacity of the sending condenser was one-thirtieth" of. a 

 microfarad, the spark-length i? inches, and the wave-length 

 3600 feet. In these and subsequent tests the double con- 



