^{page 85 note *} It may be noted that the comparatively large deviation shown in Mr. Trachtenberg's Table VI, J.I.A., vol. lii, p. 48 (Endowment Assurances), seems to be very largely attributable to the complete break in the size of his data between the first half and the second half of the unexpired term—see his Table II, p. 46. This disturbs the progression in the 3rd group, which group accounts for a large proportion of the total error.

^{page 85 note †} I.e. for purposes of demonstration only: it is not necessary to find the straight lines in practice. It is important to note that if ø = α + βx + γx² , the fitted straight line will not be α + βx (which would give a curvilinear deviation of constant sign).