MID-MORNING LUNCH IN RURAL SCHOOLS 39 



These findings are given in terms of the mean gains per week, actual and per- 

 centage, for boys and for girls at each age. The normal rate of growth of children 

 is known to vary with sex and with age. Therefore, the experimental and control 

 groups could not be compared as a whole. Because of the length of time re- 

 quired to obtain stripped weights of so large a number of children, the time 

 elapsing between the first and last weighings varied more or less. The total 

 gains have therefore been reduced to gains per week to make them comparable. 

 In the first year in school D the fall weighings were not made until December or 

 January. Both the total gains and weekh- gains were lower in this group than 

 in the same school the next year, probably because of the seasonal variations 

 usually found in the growth of children. For this reason the records for the 

 growth of the children in school D during 1930-31 have not been included in the 

 summary. The rate of growth of children at any age varies more or less with 

 their stature. To eliminate this variable, the weekly gains were calculated in 

 terms of percentage of the fall weight or height. 



As has already been stated, no influence of the feeding on gains in height is 

 indicated either by the actual or percentage gains per week of the boys or of 

 the girls. Greater gains in weight were found on the whole, however, in the fed 

 children from six to twelve years of age. Numbers of older children were too 

 few for a comparison to be significant. With the exception of the one receiving 

 evaporated milk, the fed groups were not large, and the numbers at each age, 

 therefore, were relatively small, which probably explains some of the fluctuations 

 found. The mean gains of the evaporated milk group were more consistent and 

 for the most part greater than gains observed in the controls. That these gains 

 were less than those observed in the other fed groups may have been due in part 

 at least to differences in stature. Children who are tall not only ga'n more but 

 gain at a more rapid rate than short ones, and the majority of the children given 

 evaporated milk were obviously shorter than children of the same ages in other 

 groups. 



To eliminate the possible racial factor, gains made in school D during the final 

 year when a lunch was given were compared with those of the preceding year 

 when the children were not fed (Table 13). With the exception of six-year-olds, 

 the percentage gains per week were greater lor the last year, and this improve- 

 ment with the giving of a lunch was on the whole slightly more than that for all 

 experimental children. The amount of this increase in gain in weight seems really 

 somewhat surprising, as it varied from 10 to 50 per cent above that made by the 

 controls. The difference in the six-year-olds is due not to less growth in the fed 

 children at this age but to a relatively greater growth by the controls than w^as 

 found at any. other age. The explanation of this is not apparent. 



It is quite generally thought that older children need extra feeding less than 

 little ones, and therefore that supplementary feeding rs more important for the 

 lower grades. The results of this investigation do not confirm this popular 

 belief. Age did not seem to have any influence on the proportion of children 

 who improved, as shown by the medical examination, nor did the growth of the 

 younger children appear to improve any more than that of the older ones. On 

 the contrary, the greatest percentage increase in gains in weight was made by 

 the twelve-year-old girls and the eleven- and twelve-year-old boys, while no im- 

 provement was evident in the six-year-olds. 



