649 
Ward—The Variations in Proportions in Bats. 
visited many tunnels and saw many colonies of bats, some of 
them probably containing several thousand individuals, all be¬ 
longed to this one species with the exception of a few individuals 
of a then undescribed form, Pipistrellus veraecrusis (Ward), 
found living in close proximity to M. velifer, and a few indi¬ 
viduals of Corynorhinus macrotis (Le Cbnte) living separate 
from each other as well as from the other species. 
In the foregoing diagrams and notes are presented the com¬ 
plete data of the observed variability of these eight species of 
bats. From it the reader may judge for himself how much 
variation appears to be normal and how much may be consid¬ 
ered exceptional. 
In order to present the amount of proportional variation in 
these various .species of bats, to facilitate a comparison of one 
species with another and to exhibit the degrees of average of 
maximum variability of the several comparisons, the following 
table has been constructed. 
It will be noticed that the relative amount of maximum vari¬ 
ation of any comparison varies more or less in the different spe¬ 
cies so that a comparison that may be subject to little variation 
in one species may be less stable in another. 
The percentage of the first term of the comparison is calcu¬ 
lated upon the second; thus with a head of 22 mm. and a length 
of 50 mm., the former is computed as 44, while a 22 mm. length 
of head with a 57 mm. length of head and body is computed as 
38.5 and the range of variation is the difference, or 5.5 per cent. 
When a member is compared with a shorter one a variation in 
absolute length will make a much greater variation in percentage 
than were the comparison made with a member of equal or 
greater length. I have followed the order of comparing a mem¬ 
ber with its proximal supporting member, or in case of metacar- 
pals, considering the second as proximal to the third, etc.. 
