246 
MESSRS. R. H. FOWLER AND C. N. H. LOCK ON 
They represent the size of the initial disturbance which upsets the nose-on motion, 
and vary irregularly in any one group, as might be expected. Their general size, 
however, is remarkably consistent from group to group. The corresponding values 
for the stable rounds and rough values for some of the rounds here analysed have been 
discussed by us in a previous paper.* The further results here given confirm the state¬ 
ments of that paper ; the mean value of b for any group is practically constant some¬ 
where between 0-01 and 0*02 for all groups fired. 
§ 10. Allowance for the Effect of Cards. 
In obtaining figs. 1 and 2 and Table I., but not in other cases, a small correction 
has been made to the value of the couple to allow for the impulsive action on the shell 
when it strikes a card. The amount of this correction was calculated from a few special 
rounds fired with cards on the far screens only. The following argument indicates 
that the effect should be roughly proportional to the couple due to the air and 
independent of other factors except the number of cards and the muzzle velocity. 
The effect of each card may be considered as an impulsive force whose moment about 
the centre of gravity is roughly proportional to sin d. If the cards are (as they were) 
uniformly distributed in space and therefore in time, there will be a total impulse per 
second which is proportional to sin S, and if the cards are not too far apart this is 
equivalent to a steady couple roughly proportional to the couple due to the air. The 
correction worked out in the case of the stable rounds at from 3 per cent, to 4 per cent, 
depending on the -muzzle velocity. In view of the preceding argument the same 
corrections are applied here. 
§ 11. Concluding Remarks. 
In view of possible future experiments it is of interest to compare the merits of the 
stable and unstable rounds for the purpose of this analysis. A large proportion of 
the rounds analysed in (A) had a stability coefficient s of about 1 • 8. The advantages 
of this are that the theory of the motion is practically complete, and in addition to 
the values of / M as a function of v/a rough values of the damping factors were obtained, 
which could be greatly improved if a longer range were available. The maximum 
yaw, however, is small, so that small errors in the determination of the yaw are 
important, while the periods are short so that, unless the cards are close together, it is 
difficult to draw curves through the observation points with sufficient certainty to 
determine the periods accurately. 
For the unstable rounds here analysed the theory is imperfect as regards the 
determination of damping. But a small change in the value of the couple M will 
* ‘Proc. Camb. Phil. Soc.,’ vol. XX, p. 311, 1921. 
