4 
practice was carried on till shells were observed to pierce the parapet. 22 
light and 22 heavy plugged shell were fired. The average penetration was 18 feet. 
After this the practice was continued with filled shell from the light guns at 
the loose snow, and from the heavy guns at the rammed snow. 
The effect of the filled shell on the parapet was the same as observed in the 
previous year. After firing 88 light and 32 heavy shell, the parapet (70 ft. long) 
was destroyed half-way down, though it still afforded some cover. On examining 
the parapet at the end of the practice, the plugged shell was found at a depth of 
18 and 22 feet, the depth being greater in the loose snow. Splinters of the filled 
shell were found at a depth of 6 and 20 feet. 
The results of the 1891 practice shewed :— 
(1) That plugged shell would pierce 18 feet of snow. 
(2) That 22 feet of rammed snow and 25 feet of loose snow would give 
cover against field artillery projectiles. 
(3) That splinters of shell did not penetrate more than 20 feet. 
(4) That it is extremely hard to lay accurately at snow works. 
These results indicate the best form of snow parapets as one of loose snow 
25 feet thick. 
The range of course in actual warfare would be much greater than 700 yards. 
The difficulties of laying would also be immensely increased at longer ranges, 
especially against parapets artificially concealed (by screens, &c.). 
The penetration moreover, varying inversely as the range, enables us to fix 25 
feet as the maximum thickness, which might be largely diminished at long ranges 
especially if time was short. 
The striking force of a projectile is represented by its momentum on-impact, 
which varies as the square of the velocity. 
These experiments, therefore, do not give us much information as to the effect 
of fire at medium or long ranges against such works. 
The remaining velocity of the-17-pr. at 3500 yards is hardly one-fourth of its 
velocity at 700 yards. 
Next year it is proposed to carry out further experiments at medium, and long 
ranges. 
