Porter — The Growth of St. Louis Children. — 299 
Table No. 11 and the curves constructed from it (Plates II, . 
III, IV) furnish considerable material for the study of the 
relation between median and average values. The curves of 
median minus average height agree with those of Dr. Bow- 
ditch in showing (1) a difference in the rate of growth of the 
same sex at different percentile grades, (2) an agreement in the 
rate of growth of the sexes from age 6 to 12, inclusive,* 
and (3) a marked difference in the rate of growth from age 12 
to age 16, beyond which the number of observations is per- 
haps too small for sure work. In the curves from both cities, 
moreover, the greatest asymmetry in girls is near age 13, and 
in boys near age 16, the culmination being a little later in St. 
Louis children, and the least asymmetry is near age 12. 
The asymmetry of Boston boys’ curves at age 15 and girls’ 
curves at ages 16 and 17 is somewhat greater than that of the 
St. Louis curves. It is further worthy of notice that the 
values in the case of St. Louis boys do not become negative. 
There is a general similarity also between the median minus 
average weights of the children of the two cities, the asym- 
metry of girls being again greatest about age 14 and of boys 
greatest at about age 16. Dr. Bowditch’s conclusions, quoted 
above, are therefore fully confirmed. 
e St. Louis curves on Plates II, III and IV should now 
be compared with each other. The maximum asymmetry of 
girls about 14 years and boys about 16 years old, appears in 
weight, height, height sitting, and span of arms, while the 
remaining curves exhibit no characteristic sexual differences 
during the period of prepubertal acceleration. The asym- 
metry is for the most part positive, weight, strength of 
squeeze and height of face from root of nose to point of chin 
being the only considerable exceptions. Sexual differences, 
aside from those already mentioned, are unimportant, save 
perhaps in length and width of head, in both of which the 
asymmetry of girls is greater than that of boys from age 6 to 
age 13. 
The errors in the median minus average values of height, 
weight and other single dimensions may be partially compen- 
* It should be remembered that Dr. Bowditch’s ages are recorded at last 
birthday, while mine are recorded at nearest birthday. 
