ON WIRELESS ECHOES. 433 



favourable ones for the occurrence of echoes, and this has actually been found to be 

 the case. Also the theory, as will readily be seen from the slides, suggests that con- 

 ditions should be more favourable for the production of echoes at the Equator than 

 in temperate latitudes — and, again, the prediction has been fulfilled. However, the 

 results obtained in experiments at the Equator have, unfortunate^, at the same time 

 brought evidence which is a little disturbing and difficult to reconcile with the theory 

 as first put forward. Because of this we must consider it a little more in detail. 



In May of last year, a French astronomical expedition went out to Paulo Condore 

 in Indo-CTiina to make observations on the total eclipse of the sun — and a party of 

 French wireless engineers accompanied the expedition to make observations on the 

 influence of the eclipse on wireless transmission generally, and on short-wave echoes 

 in particular, if such could be heard at the site chosen. 



A short-wave transmitter, operating on a wave length of 25 metres, and situated 

 on board ship, was employed for the special tests for echoes. The transmissions 

 consisted of musically-modulated continuous waves, the actual signals being two 

 dots, sent each half minute. Echoes were heard frequently during the half-minute 

 intervals, the time-lag ranging from five to twenty-five seconds. Throughout the 

 four-hour period beginning at noon, local time, echoes were observed to follow 

 practically every signal, and, at times, the strength of the echo was approximately 

 one-third that of the original received signal. Stormer's prediction of the suitability 

 of equatorial sites for observations on echoes was, therefore, amply borne out. 



On the day of the solar eclipse, however, it was found that echoes disappeared 

 between 13 hr. 51 min. and 13 hr. 64 min. 29 sec, while the eclipse total began at 

 13 hr. 53 min. 45 sec, and finished at 13 hr. 68 min. 25 sec. There is, therefore, no 

 doubt whatsoever that the two effects were related. The eclipse of the sun was 

 unfavourable for the production of echoes of long delay. 



A particularly interesting effect was observed during one of the tests after the 

 eclipse. When signals were being sent out every half minute, the operator forgot to 

 send one particular signal. This was noted by the receiving operator, but he actually 

 heard an echo due to the signal sent out 40 seconds earlier ! 



Thus, we have now the additional evidence that the effects of sunlight seem a 

 necessary accompaniment to the production of echoes. Now we know that the effect 

 of sunlight is to increase the ionisation in the Kennelly-Heaviside Layer so that the 

 difficulty of its penetration should be correspondingly greater. Such evidence, 

 therefore, seems to be against Stormer's theory. But if we try to suggest an alternative 

 explanation we get into almost equal difficulties. We know that in a densely ionised 

 medium the waves are slowed up so that they no longer travel with the velocity of 

 light. Is it possible for the waves to have travelled actually in the layer for 20 seconds 

 not having gone really far from the earth's surface ? So far as we can see, such slow 

 motion would mean very marked absorption of the waves, and yet we have to account 

 for the intensity of the echo being sometimes as much as one-third that of the original 

 signal. The matter is, indeed, as puzzling as it is fascinating. No decisive answer 

 to the problem can yet be given. We must, if we can, wait patiently for those 

 Octobers and Februarys when conditions appear to be most favourable for the 

 occurrence of echoes, and accumulate more data. We now have highly developed 

 methods of finding the direction in which wireless waves arrive at a station, and I hope 

 that when these can be applied in the reception of Hals-Stormer echoes a good deal 

 of suggestive information will be obtained. 



I think it is, perhaps, most fitting to conclude our survey of short-wave wireless 

 phenomena at this point, leaving the problem of these echoes of long delay with you 

 as an inviting subject for your own speculation. A tussle with such a problem will, 

 at any rate, illustrate for you the healthy state of the subject of wireless at the 

 present day. 



We are told that the mythical nymph. Echo, because of her incessant chattering, 

 was deprived of speech to the extent that she could only repeat exactly the question 

 put to her. She could neither answer nor give information in any way. 



I think, however, that we can claim for some of our wireless echoes a little more 

 than that. They repeat their original, it is true, but they sometimes do it with 

 sufficient lack of exactitude and at such significant intervals as to give away some 

 of Nature's secrets. Remembering, therefore, the objects of our Association, I hope that 

 you will agree that they have formed a fitting subject for our consideration this 

 evening. 



1930 FP 



