THE ROYAL ARTILLERY INSTITUTION. 
53 
all calibres of our modern artillery. The grains should be as compact 
in shape as possible, approaching as nearly to the cube or sphere as 
granulated powder can be expected to come. 
The size and proportion of grain can be readily ascertained with the 
sieve. The limits of size of each powder being known, all that is 
required to decide if the powder be of the proper size, is to sift it on 
the two sieves which define its size; it must all pass the one, and all 
be retained on the other. For example, if the powder be ft.L.G.—the 
size of which is between a 4 and an 8 mesh—the powder, when 
sifted on a 4 mesh sieve,* must all pass through it, and when placed 
on an 8 mesh sieve, must all remain on it. We say all retained on it, 
but practically a small quantity will always pass the lower size of sieve. 
This arises from the fact that, in the granulating machines, the powders 
are separated on sloping sieves, which do not allow the smaller particles 
to pass through so easily as horizontal ones would. The quantity, how¬ 
ever, which passes should be very small; in the B.L.G., for instance, 
it must not exceed one-sixteenth of the whole. 
The sifting we have described, though it shows that the powder 
is within the limits fixed for its size, does not however convey any 
information as to the proportion of different sized grains contained in the 
powder. To explain what is meant:—A sample of B.L.G. must 
pass a 4 mesh and be retained on an 8; but the sample may either 
consist entirely of grains just small enough to pass the 4; or, again, 
of grains just large enough to he retained on the 8. In the first 
case, the sample would consist entirely of large grains; in the second, 
entirely of small , the effect of which would be very different when fired. 
Powder made in the granulating machine, generally presents a tolerably 
regular gradation of sizes in the grain ; and it is the object of the proof 
officer to discover if this gradation of grain be properly preserved in the 
sample before him; for too great a proportion of small grains may have 
been either accidentally or purposely introduced. The sample must 
therefore be sifted into the different sizes of grains of which it is com¬ 
posed. Thus, R.L.G. is sifted on a 4 mesh, a 6 mesh, and an 8 mesh; 
and the proportion of grain retained on them should be as follows :— 
t should he retained oil the 6 mesh. 
4 a a 8 n 
But, as said above, a small quantity will generally pass the 8 mesh, 
which must not however exceed ^ of the whole quantity sifted. 
If 1 lb. be sifted, it should therefore be distributed nearly as follows :— 
ozs. 
4 mesh to 6 mesh. 12 
6 „ 8 „ . 3 
Pass 8 „ . 1 
Total. 16 
* It is perhaps hardly necessary to say, that by a 4, an 8, a 16 mesh, &c. sieve, is intended a 
sieve with 4, 8, or 16 divisions to the inch linear; not 4j 8 } or 16 openings in the square inch* 
Thus, a 4 mesh sieve has 16 openings in the square inch; an 8, 64; a 16, 256; and so on* 
