20 



DISCOVERY 



the observer on the Monument, then it follows that the 

 origin of sound must be at some point which is about 

 370 yards nean-r the former place than the latter. 



The number of places in London which are this 

 distance nearer to Westminster Cathedral than to the 

 Monument is legion, but, if we mark all such points 

 on an accurate large-scale map, it is found thej' all lie 

 on one curve ; and, further, if the place where the 

 sound began is more than three miles away both from 

 Westminster and the Monument, the curve may be 

 taken as a straight line without any appreciable error. 

 This simplifies matters enormously. With two points 

 of observation, therefore, knowing exactly how soon 

 the sound arrives at one before the other, we have 

 information enabling us to draw on a map a straight 

 line, some point of which is the place we want to find 

 out. 



We shall now suppose we have a third observer, and 

 we shall place him in the window of a certain house in 

 Chelsea. He hears the noise of the explosion, we shall 

 suppose, exactly half a second after the man on West- 

 minster Cathedral. If follows that the source of sound 

 must be about 185 yards farther from him than from 

 the man at Westminster. With this fact before us we 

 are enabled to draw a second line on our large-scale 

 map of London, every point of which is 185 yards 

 nearer Westminster Cathedral than the house in 

 Chelsea. This line intersects the first one somewhere. 

 That somewhere is the only point which is common to 

 both lines, it is the only point in South London which 

 is both 370 yards nearer Westminster Cathedral than 

 the Monument, and 185 yards nearer the latter than 

 the house we selected in Chelsea. That is the point 

 where the sound originated. 



To locate a sound, therefore, we require to get posses- 

 sion of such facts as will enable us to plot on a map at 

 least two curves. For this we require three places of 

 observation. If we had six places of observation 

 we should obtain data for five curves, and if these all 

 intersected at one point, as in general they would, we 

 should be inclined to feel surer of the point we were 

 trying to locate, than if we had merely an intersection 

 of two lines. Similarlj', if we had a hundred places of 

 observation, and measured the differences in time be- 

 tween a sound arriving at these, we should be enabled 

 to draw ninety-nine curves. Accuracy in locating 

 sounds, however, does not depend upon having a 

 host of observation-stations, but, other things being 

 equal, upon having these stations as widely separated 

 as possible. 



When the French Army (thej' were the pioneers in 

 this work) discovered that this simple scientific principle 

 might be used with advantage in locating hostile 

 batteries, they commenced experiments. This was 

 done in October of 1914, when it looked as though the 



fighting in France was going to settle down to trench 

 warfare. Their work was so successful and promising 

 that tlie British took it up, and a year later an ex- 

 perimental sound-ranging section set to work at 

 Kemmel Hill, just south of the Ypres SaUent. This 

 unit developed the work in its own way, both scienti- 

 fically and tactically, and in 191O each army of the 

 Expeditionary Force on the Western Front had two of 

 these units assisting the Heavy Artillery. In 1917 

 and 1918 there were more than thirty sections working 

 on the Western Front, and a few in Salonika and in 

 Palestine also. 



The original officers were men trained in physics 

 who happened to be serving with their infantry or 

 artillery units in France or at home in 1914-15. Sound- 

 ranging, therefore, never ran the risk of becoming an 

 ultra-scientific series of pretty experiments, of interest 

 to research workers alone. It developed into a sj'stem 

 of real tactical value. 



A sound-ranging section consisted of four officers 



HoSTlUt B/VTTERlcS 



and about forty men. The disposition of their head- 

 quarters, and their sound detectors, with regard to the 

 front line, is shown in the figure. The personnel lived 



