CAUSES OF DRIFT. 231 
and 10°, which might have been inferred from the inaccuracy of the 
practice, was noticed both at the battery and on the range. 
Now as the special shot were unsteady when fired even at 5°, in 
consequence of the gun not giving a sufficiently high velocity of rotation 
for flat-headed projectiles, I suggested that the remaining special shot 
should be fired at 2°, 3°, and 4°; this was done on the 9th and 18th 
April, 1864, and very decisive results were obtained. The following 
were the mean deflections obtained on the 9th April, Table V. ;— 
9° go 4° 
yds. yds. yds. 
Servicershot. SMe ts. meaec susie ek orci teen 2 left ‘2 left line, 
Special shot (centre of gravity towards point) 2°8 left 2°3 left 3 46 left. 
The mean deflections on 18th April, Table VI., were, 
ZO g2 4° 
yds. yds. yds. 
SISTAAICOTS AO th sab SotontoncnonadnoLascalsansagdubodonpoddon 16 left 1-4 right 1-9 right. 
Special shot (centre of gravity towards base) 3-4 left 78 left 8-46 left. 
On both days the wind blew from right to left and will account for 
some of the service shot having a slight deflection to the left ; every one 
who has seen much practice is aware that the right deflection of the 
service shot is hardly sensible at low angles of elevation, and that a 
little wind blowing from right to left is sufficient to counteract it. 
I have purposely omitted to notice No. 1 round with service shot, 
Table VI., for the deflection 8 yds. left, is evidently a very wild one, 
there being no such eccentric graze (at 2°) recorded in any of the practice 
returns given in Tables ILL, LV., V. and VI., or in numerous other 
practice returns I have purposely consulted. 
On the 9th April, Table V., it was noticed that at 2° the special shot 
were steady during flight ; at 3° they were unsteady near the end of the 
range ; and at 4° they were unsteady for some distance before grazing. 
As might have been supposed the left deflections of the special shot 
with preponderance behind were greater than those of the special shot 
with preponderance in front. 
The results of the experiments shown in the preceding Tables are, I 
conceive, sufficient to establish the following facts :— 
(1) That elongated projectiles with rounded or pointed heads, have a 
tendency to deflect to the right at ordinary ranges when fired with right- 
handed rotation. 
(2) That similar projectiles fired with Jeft-handed rotation, tend to 
deflect to the left. (Table II.) 
(3) That the drift of cylindrical or flat-headed projectiles ig in the 
opposite direction to that of projectiles with rounded or pointed heads 
fired with similar rotation ; for instance ; if a cylindro-conoidal and a 
flat-headed shot be both fired with right-handed rotation, the drift of 
the former will tend to the right and the latter to the left. 
(4) That the drift of ordinary service shot fired from the 40-pr. 
Armstrong gun is hardly sensible at 1000 yds. ; for it may be seen from 
Tables V. and VI. that a slight breeze from left to right was sufficient 
to counteract the tendency of the shot to bear off to the right. It also 
appears from the Tables that flat-headed shot require a very much 
higher velocity of rotation than cylindro-conoidal projectiles, and that 
the former lose their velocity very much sooner than the latter, 
