PHILLIPS — THE NORTHEEN VIRGINIA DEER 
131 
A word of explanation is necessary in regard to the way the various 
measurements were taken. 
1. Palatal length. (From back of posterior molar to gnathion.) 
2. Audito-basal length. Owing to the fact that the posterior end 
of many of these skulls was damaged in the taxidermist shop, in process 
of getting out the brains, the condylion could not be utilized in taking 
the condylo-basal length. Accordingly I selected the anterior edge 
of the auditory canal as the posterior point and the gnathion as the 
anterior point, calling this the audito-basal length. 
3. The length of lower tooth row. 
4. The zygomatic width (greatest width outside zygomata). 
5. The length of upper tooth row. 
6. The greatest width of orbit. 
7. The mastoid width (greatest width between mastoid prominences). 
8. Length of nasal bones. 
9. The length of antler along outer curve. 
The probable error is also given in each case for the mean, the stand- 
ard deviation, and the coefficient of variability. 
A summary of all these measurements is given below in the form of 
a table, the measurements appearing in the order of their uniformity, 
those having the smallest coefficient of variability (C. V.) appearing 
first. 
Under each measurement appears: 
1. The number of individuals upon which the calculation is based. 
2. The actual mean size of the whole series. 
3. The absolute maximum size. 
4. The absolute minimum size. 
5. Measurement of the type of borealis in Museum of Comparative 
Zoology. 
6. The standard deviation, or index of variability (see Davenport, 
Statistical Methods, John Wiley & Son, 1899, page 15) which gives the 
average deviation of all individuals from the mean. 
7. The coefficient of variability (see Davenport) wherein the devia- 
tions from the mean of all the different individuals are reduced to a con- 
crete number, which renders the different skull measurement directly 
comparable, one with another. This is the important figure to note, 
and the only one of special interest in determining the relative value 
for systematic purposes of the various measurements here studied. 
