270 ANNUAL REPORT SMITHSONIAN INSTITUTION, 1949 
which frequently sit on their haunches with the forefeet supporting 
the front part of the body (pl. 14, fig. 2). 
A position that would appear to us to be uncomfortable is regularly 
adopted by flying squirrels, which normally live in hollow trees. The 
body hangs in a vertical position, either head up or head down, sus- 
pended by the fingers or the toes, most of the load actually being on 
the fingernails or the toenails (pl. 3, fig. 1). They sleep in such poses 
at least a portion of the time. The hanging position is probably 
assumed as the result of having accommodated themselves to situations 
where there is no satisfactory ledge or bottom in the den, so that a 
tree cavity that would otherwise be unsuitable can be utilized by 
clinging to its rough surfaces. However, they now take this pose 
freely even when comfortable nests or rests are available. Of course, 
they also take other positions such as curled up in a ball or lying on 
the side or the back. 
The inverted position so often used by sloths (Choloepus, Bradypus, 
and Scaeophus), wherein they hang beneath a limb, is well known and 
is entirely suitable for their purposes. Bats generally hang head 
downward. 
HOMING INSTINCT 
When an animal has become established at a given location, it 
almost invariably develops a strong attachment for its home even 
though such home does not provide entirely satisfactory quarters. 
Almost every animal has a rather definite area in which it lives its 
entire life, and it rarely leaves this range except when forced to do so 
by very unusual circumstances. This is commonly shown when 
captive animals escape, for they usually remain nearby for some time 
and frequently go back into the cages if they are able todo so. People 
who are familiar with animals regularly take advantage of this trait 
by leaving the cage door open, refraining from frightening the animal, 
and placing food in the cage or nearby until the animal is recaptured. 
On one occasion I was keeping a specimen of the least short-tailed 
shrew (Cryptotis parva) in a jardiniere. I had observed it trying to 
leap out, but it is not adapted to leaping and was apparently unable 
to reach the top. After a few days I discovered that it had escaped 
during the night probably by superior jumping gained by persistent 
exercise. It was so tiny that it was useless to search for it in the 
house, so I waited until evening and then placed the jardiniere on the 
floor almost directly below the place at which it had been standing on 
my desk. JI crumpled around it an old blanket so that it formed a 
ramp or runway from the floor up to the top of the jardiniere. I then 
mashed a mealworm and dragged particles of it in a trail on the 
blanket from where it touched the floor to the top of the jardiniere. 
Within a couple of hours the little fellow was back in his home. 
