scattered in all directions. In one block two furrows from shells were observed 
but the block remained whole. 
Such very great resistance points to the usefulness of ice as a material for forti¬ 
fication. But as we have seen above the shell when it strikes the ice throws off 
from it an immense number' of splinters, which largely add to the number of 
destructive fragments, and this property of ice renders it decidedly disadvantageous 
for the above-mentioned purpose. 
In anticipation of this result, the ice, as we have said above, was covered with 
snow 1 foot thick ;' but this snow was scattered by the first shot that struck the 
section: this happened partly no doubt owing to the snow that was used being 
very crisp, almost like fine sand. Possibly soft snow would have lasted longer, 
and would have prevented the ice throwing off so many splinters. 
In the 5th section the snow bags were much scattered and the parapet des¬ 
troyed half-way down, but it still offered a certain amount of cover. 
We now indicate the conclusions come to from these experiments :— 
(1) A parapet of ordinary snow of sufficient thickness, affords a very solid 
protection from field gun projectiles. Although the 1st section, 14 feet thick was 
pierced by the empty shell, the filled shell only penetrated 9 feet; this shews that 
a parapet 14 feet thick would be perfectly safe. 
(2) A similar result may be expected from a snow bag parapet; but taking 
into consideration the immense number of bags and the quantity of snow to be 
dug, this form of parapet seems very disadvantageous. 
(3) Trampling the snow and wetting it do not diminish the thickness of para¬ 
pet necessary. 8 feet is not sufficient. 
(4) ,Ice unless strongly faced with snow, is no use for a parapet; but if a 
durable facing could be arrived at, we might come to a different conclusion. 
Unfortunately the 1st section, which from the simplicity of its construction 
would be the most useful form, was hit by only a few projectiles, so that no 
opinion could be formed of the effect on it of a large number. 
Experiments in 1891. 
From the results of the previous year’s experiments, it was decided to confine 
this year’s experiments to parapets of snow thrown up in the ordinary way, and 
of snow trampled down as it was thrown up. 
The object of the practice of 1891 was to fix the minimum thiokness of parapet 
that would be impervious to field artillery projectiles. The maximum thickness 
of the experiments of 1890 viz., 14 feet, having proved insufficient. 
With this view it was decided to make two parapets of trapezoidal form, with 
a maximum thickness of 30 feet and a minimum of 15 feet. Two such parapets 
were constructed, one of snow thrown up in the ordinary manner, and the other 
of trampled snow; both parapets were joined at their 30 feet extremities and 
formed so to speak one parapet 30 feet thick in the middle and 15 feet at the 
ends—height throughout 5 feet. Looking from the battery the right half of the 
parapet was of ordinary snow and the left of trampled snow. The sections of the 
parapet were divided by a board to assist in aiming. The practice was carried 
out from two 17-prs. and two 30-prs. of the 1st Brigade, with 60 17-pr. common 
shell and 54 30-pr. shell; range 700 yards. At such a range very accurate 
practice was to be expected but as a matter of fact it was not so, the reason being 
that there was no aiming point. 
It seems that it is very difficult to lay at snow targets even at such short 
ranges. 
The practice commenced with plugged shell. Orders were given to commence 
firing at the centre of the target and to pass on to the flanks ; the heavy guns to 
fire at the loose snow portion, and the light guns at the rammed snow. The 
