CHAPTER III. 
TRUSTWORTHINESS OF THE MATERIAL; ITS LIMITATIONS. COMPARISON OF 
MEDIAN AND AVERAGE VALUES, 
The question first to be decided in the discussion of an 
anthropometrical series is whether the individual observations 
are so related one to another that they constitute accidental 
deviations from a middle value. The method of answering 
this question by comparing the series of observations with a 
series constructed according to the theory of probabilities has 
already been described, and it has been stated incidentally that 
_ the heights of St. Louis girls aged 8 agreed with the theoretical 
series in their distribution. Where such an agreement exists, 
the individual observations are to be regarded as approxima- 
tions of a middle value which is the type of the series. It is 
not necessary to make this comparison at more than one age, 
or in more than one dimension, for it is known that if one 
series in a group like that with which we have to deal shows 
this agreement the other series will be found to do the same. 
In the present case, additional evidence of the correspondence 
between observation and theory could be furnished, were it 
required, by Table No. 10, containing the probable deviation 
) 
