SOUND-RANGING 85 



conditions, is of a surprisingly high order of accuracy. The 

 error is often less than ten meters and rarely more than twenty- 

 five meters at a distance of from five to eight miles. The reason 

 for this high order of accuracy of the average is probably the 

 following: Systematic errors, such as those due to a careless 

 survey of the sound-ranging base or to errors in the timing 

 device, etc., are practically non-existent and all errors are hap- 

 hazard in character. The relative excellence of each location 

 of a series may be judged from the character of the intersec- 

 tion on the plotting board as described earlier in this article 

 so that a fairly correct weighted average may be formed by 

 counting locations, estimated as correct to within fifty meters, 

 three times, those estimated as correct to within one hundred 

 meters, twice, and those to within one hundred and fifty meters, 

 once. Whatever unsystematic error due to wind and tempera- 

 ture has been introduced will affect the weighted average value 

 far less than it affects the individual values, for such errors 

 will tend to cancel each other's effect. 



As a result of the study of the errors in the sound-ranging 

 locations of scores of enemy batteries it appears that the 

 section commander should report to the artillery the average 

 of all previous locations of a battery rather than the latest 

 location or even the best location as judged from the character 

 of the intersection of the strings on the plotting board. (There 

 are many instances showing that the average of five or more 

 locations, no one of which was correct to within 150 meters, 

 was accurate to within 50 meters.) Of course in mobile war- 

 fare averages should not be taken nor should they be taken 

 in position warfare if there is any reason to suspect that the 

 gun is not occupying a fixed emplacement. 



The great accuracy of the average of a series of sound- 

 ranging locations was not suspected during the war even by 

 those engaged in this service ; had it been recognized an incident 

 like the following would have been impossible. In the 

 St. Mihiel sector there was an enemy battery position which 

 was repeatedly reported by sound-ranging as active. The 



