71 
1910-11.] Relative Positions of certain Biological Types. 
however, been approximately calculated at a later stage, and found so small 
as almost to justify their exclusion on that ground alone. Of the angle of 
parietal curvature, the value for the Tasmanians only is available, so it 
has also to be excluded. The remaining twenty-four measurements and 
indices form the basis of the investigation. 
Taking each of these in turn, the various crania are arranged in order, 
according to the respective values of that measurement or index. The 
lowest award is zero, since the determination of an absolute zero is not 
possible, whilst the highest is one less than the number of “ entries ” under 
that head. The other awards are intermediate to these, and so allotted that 
the intervals between them are proportional to those between the actual 
values of the particular measurement or index for the various types of 
crania, whilst from the above choice of range the average value of the 
interval between successive awards throughout the series is unity. 
In Table II. are shown the proportionate values thus determined for 
each measurement or index. 
The total sum of the awards for each type on all the counts wherein it is 
represented is then determined, together with the total of the maximum 
awards obtainable by that type. The ratio of the former to the latter is 
taken to represent the relative position in the evolutionary scale. Thus this 
“ index of position ” has a range between zero and unity. The minimum 
value would indicate that the type whose index had that value occupied on 
each count the lowest position in the scale, whereas the maximum value 
would be obtainable only by a type that attained to the highest position 
on each count. 
The relative positions in the evolutionary scale of the various types of 
crania, as represented by the above “indices of position,” must, however, 
be regarded only as a first approximation, for, apart from the varying 
number of types represented, the values of the various measurements and 
indices as criteria have been taken as equal. Inspection of the differences 
in the orders as each of these is in turn adopted as the criterion shows this 
not to be the case. Thus it is necessary to attempt an estimation of the 
relative weight to be attributed to each of the measurements and indices 
with which this work is concerned. 
This is done by comparing each individual order with the composite 
order as given by the resultant relative positions in the last line of 
Table II. 
Table III. gives the resultant relative positions adapted for direct com- 
parison with the various individual series of Table II. This adaptation is 
made necessary by the absence of various measurements in Table I., and the 
