Ward—The Variations in Proportions in Bats. 637 
that their position seemed to bear certain relations to the con¬ 
figuration of the caverns, and, consequently, to the distance to 
which it was possible for some faint trace of light to be re¬ 
flected. 
I would not dare hazard a guess at the number of bats then 
in the cave; but it was apparent that many millions must have 
lived there during the time that the guano had been accumulat¬ 
ing. 
On returning our experiences were much the same as when 
coming in. Approaching the entrance we found that a large 
number of bats had been driven before us and had clustered to¬ 
gether in a high, dark portion of the roof just before the tunnel 
debouched into the rotunda. Swinging my gun rapidly, so as to 
cover as large an area as possible, I fired both barrels, upon which 
a perfect shower of dead and wounded bats fell to the ground. 
The result of this double shot was 171 bats picked up, while 
probably many wounded ones escaped us, and a number too 
badly shot were not taken. We were soon out of the cave, and 
when I procured a collecting bottle to put insects in my surprise 
was great to find that there was not one of the myriads covering 
the floor of the cave to be found anywhere about our persons or 
paraphernalia. 
I have seldom found more than two species of bats inhabiting 
the same cave, and they generally belonged to different families. 
In this case the attractiveness of the situation and the abund¬ 
ance of insect life in the tropical vegetation surrounding the 
cave seem to have lessened the strife for existence to such a low 
ebb that thousands of each of these four species could live in 
harmony together in “theTittle hell.” 
The opposite graph exhibits, the correspondences and discrep¬ 
ancies in proportions as found in the thirty-three specimens 
measured. The more striking ones between length and tail as 
shown by the specimens from 814 to 822 have been alluded to. 
Although these two curves have, in reality, a pood many points 
of similarity, their differences are so striking as to convey at 
the first glance the impression that they have little in common. 
In some instances, such as from number 832 to 839, in com¬ 
paring the first and second phalanges of the third digit, the lack 
of agreement would suggest the possibility of the first measure¬ 
ments having by accident been advanced one place or of the 
second having been carried back an equal distance. There ap- 
