observed during the five weeks' summer holiday; the "control children" gained on an average .37 kilos (_i.e._, at the rate of .074 kilos per week), while the children fed at school lost .46 kilos.[493] The accompanying chart illustrates the rate of increase of the two groups of children. Apart from the increase in weight, the improvement in the general appearance and carriage of the children who received the meals "was more or less apparent in all, and very obvious in some of the children, who visibly filled out and brightened up."[494] The reverse process was equally apparent after the summer holidays. Footnote 492: "The average gain per year of children of this class and size," Dr. Crowley points out, "is not more than two kilos (4 lbs. 6 oz.) for the whole year." (_Ibid._, p. 9.) Footnote 493: _Ibid._, pp. 9-11. As Dr. Crowley points out, several points have to be considered in interpreting the effect on weight. "The increase in the weight of children normally varies greatly at different seasons of the year," and "at any given season fluctuates much, sometimes, comparatively, even from week to week. The proportional increase in weight varies with the age of the child, or rather with the weight to which the child has already attained." (_Ibid._, p. 8.) Footnote 494: _Ibid._, p. 8. [Illustration: Chart illustrating the average gain or loss in weight--during the intervals shown--of the children who were fed at Bradford. The broken line shows the average increase in weight--during the same time--of the "Control Children."] At Northampton, in 1909, a similar experiment was conducted under the supervision of the Medical Officer of Health. Forty-four children were given breakfast and dinner for fourteen weeks, and weighed weekly, together with forty children of the same social class who were not receiving meals. At the beginning of the experiment the average weight of the fed children was 1.71 kilos less than that of the "controls"; in the second week their average gain was much greater, and by the end of the fourteenth week the difference in weight was reduced to 1.02 kilos. During the Easter holidays of ten days in which no meals were given, the children who had previously been fed lost in weight while the "controls" gained.[495] Footnote 495: Report on the Working of the Education (Provision of Meals) Act up to March 31, 1909, pp. 14-15.