SHRAPNEL FIRE. 
179 
The following will therefore be the number per square foot at the 
different areas, as above stated:— 
TABLE IV. 
Length of Burst in yards. 
Guns. 
20 yds. 
40 yds. 
60 yds. 
80 yds. 
100 yds. 
120 yds. 
140 yds. 
160 yds. 
180 yds. 
200 yds. 
per 
per 
per 
per 
per 
per 
per 
per 
per 
per 
sq. ft. 
sq. ft. 
sq. ft. 
sq. ft. 
sq. ft. 
sq. ft. 
sq. ft. 
sq. ft. 
sq. ft. 
sq. ft. 
9-pr. 
1-6 
•4 
•17 
•1 
•06 
•043 
•03 
•024 
•02 
•015 
16-pr. 
2-5 
•62 
•26 
•15 
•1 
•07 
•05 
•04 
•03 
•024 
If now the area of the upright figure of a man is taken, it will be 
found that, if these tables are correct, when a shell is burst 20 yds. 
short of a line, each man in the area covered by its fire is hit, in the 
case of a 9-pr. by nearly 16, and of a 16-pr. by nearly 25 bullets. 
These are direct hits, and do not take into account the bullets of the 
lower portion of the cone which ricochets. 
Here is a manifest waste of power: our object must naturally be 
not to disable one man by 16 or 25 hits, but 16 or 25 men by one 
hit each. 
In fact, the practical question to be solved at this point is—What is 
the distance at which a shrapnel shell ought to be burst in front of a 
body of troops with an extended front, so as to have the best chance 
of disabling the largest number possible ? 
Combining the calculations we have already made with the practical 
experience of the experiments at Shoeburyness (Table I.), the fol¬ 
lowing may be said to be the results at different lengths. Throughout, 
the effect on the front row only is taken into account, and the number 
of hits shown in the Report has been reduced by J, to allow for the 
difference between a 9-ft. and 6-ft. target. 
At 800 yds. range 
Length of burst, 50 yds. short. 
Spread of cone (see Table II.) = 7 yds. 
Area of target covered, 7 X 3 x 6 = 126 sq. ft. 
■«t p . 7x3x12 1n 
No. of men m area-—-= 12. 
21 
Taking the number of direct hits for this area and length of burst 
from Table IV., they will be 126 x *44 = 54. At such a range they 
would doubtless be all throughs. One-third the average number of 
throughs in the Report = 63 ; the balance? after deducting the direct 
hits, is therefore 63 — 54 = 9, and these are due to ricochet. 
Under the conditions, the whole of the lodges may be taken as 
effective bullets. They are the result of ricochet, and to it therefore 
9 + 15 == 24 effective hits may be credited. 
Now, the whole number of pieces in the shell is 136, and deducting 
