Figures 19 to 23 show no consistent plan for distributing practice. With 5 × 5 (Fig. 19) the amount of practice increases from the first treatment in grade 3 to the end of grade 6, so that the distribution would be better if the pupil began at the end and went backward! With 7 × 7 (Fig. 20) the practice is distributed rather evenly and in small doses. With 6 × 7 and 7 × 6 (Fig. 21) much of it is in very large doses. With the divisions (Figs. 22 and 23) the practice is distributed more suitably, though in Fig. 23 there is too much of it given at one time in the middle of the period. [Illustration: FIG. 19.--Distribution of practice with 5 × 5 in the first two books of the three-book text E.] [Illustration: FIG. 20.--Distribution of practice with 7 × 7 in the first two books of text E.] [Illustration: FIG. 21.--Distribution of practice with 6 × 7 or 7 × 6 in the first two books of text E.] [Illustration: FIG. 22.--Distribution of practice with 72, 73 ... 79 ÷ 8 or 9 in the first two books of text E.] [Illustration: FIG. 23.--Distribution of practice with 81, 82 ... 89 ÷ 9 in the first two books of text E.] POSSIBLE IMPROVEMENTS Even if we knew what the best distribution of practice was for each ability of the many to be inculcated by arithmetical instruction, we could perhaps not provide it for all of them. For, in the first place, the allotments for some of them might interfere with those for others. In the second place, there are many other considerations of importance in the ordering of topics besides giving the optimal distribution of practice to each ability. Such are considerations of interest, of welding separate abilities into an integrated total ability, and of the limitations due to the school schedule with its Saturdays, Sundays, holidays, and vacations.