426 
PRINCIPLES OP GUNNERY. 
Mean 
lateral de¬ 
viation. 
Mean 
vertical 
deviation. 
Mean error 
in range, 
or mean 
longitu¬ 
dinal error. 
Mean 
lateral 
error. 
Mean 
vertical 
error. 
Mean 
absolute 
error. 
Method of 
estimating 
the accu¬ 
racy of 
small-arms 
and ord¬ 
nance. 
Example 
of testing 
guns for 
range and 
accuracy. 
each shot by the number of rounds fired. Thus, RR' represents the 
line of mean range. 
The mean lateral deviation is obtained by dividing the algebraical 
sum of the lateral deviations of each shot by the number of rounds 
fired. Thus, LI! represents the line of mean lateral deviation. 
The mean vertical deviation is obtained by dividing the sum of the 
vertical deviations of each shot from the horizontal line OX by the 
number of rounds fired. Thus, VF' represents the line of mean 
vertical deviation. 
The intersection of the lines RR' and LL' gives A, the point of 
mean impact on the horizontal target; and the intersection of the lines 
1!L" and VV gives A !, the point of mean impact on the vertical target. 
The mean error in range, or the mean longitudinal error, is obtained 
by taking the numerical sum of the differences between the range of 
each shot and the mean range, and dividing by the number of rounds 
fired. 
The mean lateral error is obtained by taking the numerical sum of 
the differences between the lateral deviation of each shot and the mean 
lateral deviation, and dividing by the number of rounds fired. 
The mean vertical error is obtained by taking the numerical sum of 
the differences between the vertical deviation of each shot and the 
mean vertical deviation, and dividing by the number of rounds fired. 
The mean absolute error is obtained by taking the sum of the radial 
distances of each shot from the point of mean impact, and dividing by 
the number of rounds fired. 
The mean absolute error may be estimated both on the horizontal 
and on the vertical target. 
In order to estimate the relative accuracy of small-arms, the mean 
absolute error is taken on a vertical target. To estimate the relative 
accuracy of ordnance, the mean longitudinal, lateral , and vertical errors 
are observed. 
As an example of testing guns for range and accuracy, a direct 
comparison of the shooting of a long 40-pr. and a short 40-pr. M.L. 
gun is given in detail. The length of bore of the long 40-pr. was 
22 calibres; that of the short 40-pr. was 18 calibres. 
