188 BELL SYSTEM TECHNICAL JOURNAL 



that which may exist over a large part of the broadcast area at a 

 distance from the transmitter, while in the Westchester region we 

 have an extreme and rather special circumstance. Field strength 

 surveys have shown that there are indications of a daytime inter- 

 ference pattern over the Riverhead area but this pattern, such as 

 it is, appears to be irregular and to lack the definition which makes 

 the Westchester pattern so remarkable. 



On the basis of the Westchester data alone we might build up a 

 theory to the effect that night-time shifts of the stable daylight 

 pattern were in some way responsible for quality distortion fol- 

 lowing the departure of daylight. Such a thought applied to the 

 Riverhead case does not seem so reasonable since here the pattern is 

 about one-quarter as distinct in terms of the ratio of maxima to 

 minima values as the Westchester pattern. If, however, we presume 

 that quality distortion may be expected in areas where daytime 

 signals arrive considerably attenuated or so interfering as to simulate 

 such an attenuated condition both situations are satisfied. After a 

 consideration of the evidence at present available, such a conclusion 

 seems attractive; that is, a daytime wave interference pattern alone 

 is only an agency in night-time quality distortion in so far as its 

 minima in combination with the general shadow effect are responsible 

 for a low signal directly transmitted. Perhaps, in other words, the 

 daytime field strength is a measure of direct night-time transmission, 

 there existing in combination with this direct path at night a second, 

 variable route of greater eflfective length. Probably close to the 

 transmitter the "direct wave" is large compared to the "indirect" 

 but shadows or interference may materially modify the ratio. 



Night Distribution of Field Strength 



By receiving simultaneously at several points the signal coming 

 from a distant transmitter, it ought to be possible to detect the 

 movement in space of these interference bands we have been dis- 

 cussing. The question immediately arises as to how far apart these 

 distributed receivers can be placed without giving us an entirely dis- 

 continuous and misleading picture. For the first step toward record- 

 ing space variations, in the vicinity of the Riverhead testing station, 

 the receivers were spaced 1/16 wave length (30.5 meters), as illus- 

 trated in Fig. 33. It is necessary in making such determinations to 

 transmit a single radio frequency, since we have already found that 

 the interference bands for one component of a modulated wave are 

 likely to be in a different position than those for another. 



