the correct orientation to a greater or less extent, which is obviously the fact, although the confusion may be very slight in some cases. In Fig. 1, B, the direction in which the birds will actually move when in flight as a flock is indicated by the parallel arrows. Thus in Fig. 1, B, the mutually related influence of the individuals prevents the dispersion that would occur as indicated by the direction of the arrows in Fig. 1, A. The averaging of the errors that take place, as indicated in B, is approximately governed by mathematical certainty, and as a matter of fact, in the drawing Fig. 1, B, the flight direction was determined by taking the mean of all the directions indicated by the arrows in "A."
Each bird is affected by the averaging of flight directions due to the mutual reaction of the individuals, and the reaction prevents false starts. As an extreme case of this correcting influence, consider a flock of birds proceeding northward after resting on an open sea in a fog; and suppose that one bird was so confused that it would have flown in the opposite direction (to the south) if alone. Then, in all probability, in spite of its individual inclination to fly south, this bird would be carried northward with the flock by the powerful principle of imitation.
It must be distinctly understood that this theory does not give an explanation of the sense of direction, but it does provide a mechanism which will prevent individuals of the flock from getting lost. The only assumption is that there are all degrees of right and wrong "bearings" among the individual birds of the flock.
Fig. 2. The curved lines in the left-hand diagram A represent the paths which the birds would follow due to unequal unbalanced wing power, if each was alone. In the right-hand diagram, B, the arrows represent the birds flying together and the direction is the mean of all deviations and represents the flight direction towards D as corrected by mutual reaction.
