Graph E shows clearly the essential facts in this relation. 

 Two points stand out: 



1. Weight increases with height. The only exception is at 

 the extremes of height in our series, where the number of cases 

 is entirely too small to give a correct picture of the interre- 

 lation. Beginning with 55 inches among the boys and 56 

 inches among the girls, and continuing upward for the boys 

 without exception, and for the girls to height 67 inches, the 

 weights increase steadily with every successive inch of height. 

 The actual increase in weight for every inch of height varies 

 somewhat for the several heights, and also for the two sexes. 

 In Table 8 (page 16) we give figures which present the relation 

 of weight to height, smoothed so as to eliminate some of the 

 fortuitous variations that make the lines in the graph rather 

 irregular. 



2. The girls are heavier than boys for any given height up 

 to about 63 inches. Beyond this height, the boys outweigh 

 the girls. The sole exception is the extreme upper limit of 

 height, where the fewness of cases again mars the reliability 

 of our data. This phenomenon is consistent with, although 

 somewhat different from, the findings of other authors. Bow- 

 ditch, for example, points out that girls are heavier than boys 

 up to 58 inches. Above that point the reverse is true. He 

 finds that the height of 58 inches is attained on the average in 

 the 14th year. The fact that our figures cover the age- period 

 of 14 to 16 only may in part explain the difference in the values 

 of the crossing point in our investigation. 



The above relations, which cover the two years combined, 

 are, with but slight exceptions, true for each of the individual 

 years. In the first year, namely, 14 to 15, the point where 

 the boys overtake the girls is about 62.6 inches. In the next 

 year it is approximately 63.4 inches. 



As might be expected from the smaller number of cases in 

 each individual year, the fluctuations in the weights for the 

 successive inches of height are somewhat more marked than 

 in the two-year period. It has therefore seemed desirable, in 

 order to correct for the fluctuations resulting from the paucity 

 of data, to "smooth" the series of weights. These figures are 

 given in Table 8: 



15 



