212 BELI. SYSTEM TECHNICAL JOURNAL 



Visual observations made with the cathode ray oscillograph, which 

 unfortunately furnishes no permanent record of transient effects, con- 

 firmed the strip records in regard to the reality of there being side-band 

 phase variations. From equation (17), it is seen that if these varia- 

 tions occur the fundamental of the detected tone signal at the receiver 

 will not bear a fixed phase relation to that detected from the trans- 

 mitting antenna current while if there are no such changes the phase 

 between these two tones would remain constant. The locally detected 

 tone and the tone detected from the transmitting antenna current and 

 brought to the receiving station over telephone wires, were applied 

 to the two pairs of deflecting plates in the cathode ray oscillograph 

 Since the deflections caused by these two pairs of plates are at right 

 angles to each other the resulting Lissajous figure from two sine waves 

 of the same frequency will be a slanting line, an ellipse or a circle 

 depending on their phase and amplitude relation. The actual figures 

 were observed to change progressively through this range of shapes, 

 the changes following roughly the magnitude and rapidity of the 

 fading. The effect of amplitude changes on such figures is quite 

 distinct from the effect of phase changes and there was no difificulty 

 in separating out the evidence of large phase changes. 



Considering only the above theories and facts there appears to be a 

 reasonable basis for a conclusion that the best form of radio transmis- 

 sion for use in broadcasting is single side-band with carrier suppression. 

 But on practical grounds we do not believe such a conclusion is justified. 

 The fading and distortions which we have made much of in the pre- 

 ceding pages are not experienced by the majority of broadcast listeners 

 when they listen to local stations. To require these listeners to provide 

 themselves with more complicated and expensive receivers, simply to 

 allow more distant or less favorably situated listeners to obtain better 

 reception, seems neither reasonable nor desirable. The art offers 

 several other possible avenues toward improvement much less difficult 

 of application and it must be remembered that radio broadcasting is 

 already reaching a degree of standardization and a volume of existing 

 receiving equipment which rules that changes must come slowly and 

 without serious prejudice to the existing order. 



Conclusions 



Subject to the limitations imposed by the scope of our investigations 

 the following conclusions may be drawn: 



Fading can be quite sharply selective as to frequency and the evi- 

 dence points tow^ard wave interference as the cause. 



