Porter — The Growth of St. Lowis Children. 2838 
TABLE No. 5. 
THE CALCULATION OF THE PROBABLE DEVIATION — FROM THE AVERAGE 
HEIGHT (118.36 oe ) of 2193 GrRLs, a 
Height at Intervals of : 3 ro 
2 Centimetres. 15 
141 and 142 Cm. 1 23.64 23.64 
139 oe 140 ee 
167 -** 138. ** 7 19.64 19.64 
185 =** 186 5 17.64 88.20 
ing. © [pa-- 10 15.64 156.40 
Lot isa 21 13.64 286.44 
e208! ASH se 28 11.64 325.92 
wage ** P28: 4° 19 9 761.56 
125 86 “1 2G <** 138 7.6 
i 4, eee oe 183 5.64 1032.12 
121 ics = 243 3.64 84.5 
119. 66-739 842 1.64 560.88 
bi? * iis & $21 0.36 115,56 
LG ** . Lige 297 2.36 700.92 
tig 114 222 4.36 967.9 
TEE 8 Fie 137 6.36 872.69 
1095 **. Gs 84 8.36 702,24 
FOOT: *& 108: = 42 10.36 435.12 
106 "306% 27 12.36 333.72 
Tos “* 104 — 8 14.36 114.88 
i102 "162 2 16.36 $2.72 
+A OG 1 18.36 18.36 
TGtal 42-55 2192 9487.77 
9487.77 
d = + 0. 845395 - = + 3.698 Cm. 
The distribution of the above series of the heights of girls, 
aged 8, is therefore characterized by a probable deviation of 
+ 3.7.cm.; that is, one against one may be wagered that no 
girl aged 8 will be taller than 122.06 cm. or shorter than 114.66 
cm. If the number of observations falling between A + d, 
A + 2d, A+ 3d......A+"d be noted, a complete picture 
of the individual observations in a series will be obtained. 
This observed distribution may then be compared directly 
with the distribution of the observations in an hypothetical 
series constructed according to the calculus of probabilities. 
The observed and the theoretical series should correspond, if 
the causes of deviation are purely accidental. It has already 
been said that such a comparison must be made before it can 
be known whether the observations in any series can be 
treated by the methods of the theory of probabilities. It is 
however not necessary to compare more than one of a num- 
