Ward—The Variations in Proportions in Bats. 651 
In the above table, in the column of averages, we find that the 
least variation is shown by the proportion of the first phalanx 
of the fourth digit compared to its metacarpal and next by the 
thumb to the forearm. In both these cases a short member is 
compared to a long one and the percentage would naturally be 
low. It is also noticeable that there is comparatively little dis¬ 
proportion shown in the comparisons of the metacarpals with 
one another because although while the lengths are considerable 
they are approximately equal. Wherever a long member is 
compared to a short one the percentage of variation becomes at 
once more elevated, a fact that must be borne in mind in order 
not to misinterpret this table. Although this method of show¬ 
ing the comparative variations of parts is faulty and liable to be 
misleading, yet after all it seems to me to be the fairest way 
to treat the matter. A comparison of these by differences in 
actual measurement expressed in millimeters would probably be 
more misleading as a few millimeters more or less on a long bone 
counts for little, whereas it may greatly alter the proportions of 
a short one. It should also be borne in mind that this table 
does not show the proportionate variation for the species but only 
for individuals. There are several other forms of comparison 
that it would be interesting to make; but my original object in 
making these measurements was to determine, for my own use, 
the value of such common descriptive terms as “- three- 
fifths the length of-,” “- slightly longer than- 
etc. Most of them are found to be unwarranted by the facts 
in the case. 
As the second metacarpal normally bears no phalanges it 
would be expected that its proportionate length would be less 
constant than the others and this we find to be the case. 
That the length of the tail compared to that of the head and 
body does not show as great variation as do the proportions of 
the first two phalanges of the digits is rather surprising. These 
latter show the greatest variations of any of the parts compared 
and the maximum of variation in five of the eight species com¬ 
pared, as well as in the average for all, is found in the proportion 
of the second phalanx to the first in the third digit, a measure¬ 
ment not unfrequently given by authors as among those diagnos¬ 
tic of species. 
From the various tables of species and the remarks under 
them, it will be seen that it is quite unwarranted to make com¬ 
parisons of the relative measurements of bones of bats for specific 
