284 EXPLORATION OF THE CANONS OF THE COLORADO. 
could tell exactly what the salamander does at the instant he appeared, his 
motions being so quick that one cannot be quite sure; the general impression, 
however, is, that they are unloading their cheek-pouches. This is not at all 
improbable, for we know that they carry their food in these receptacles, and 
it seems a very natural way for them to bring their refuse sand to the surface, 
since they often have to transport it a distance of several feet. Still it is 
quite desirable to have other and more careful observations; for observers are 
apt to be deceived by their own eyes, especially in the light of preconceived 
opinions. 
The subterranean labyrinth constructed by this clever army of sappers 
and miners penetrates the pine-barrens and cultivated fields in every direction. 
An energetic salamander, with a slight knowledge of engineering, would find 
little difficulty, I suspect, in making an underground journey through Florida 
from the Atlantic to the Gulf of Mexico. The direction of the burrows may 
easily be traced by the loose hillocks of white sand which are thrown up 
along the line at intervals of three or four feet. These are the "dumps' 7 
made by the burrower in throwing out his refuse accumulations. Each con 
sists of about a peck of loose sand, and, by the casual observer, might easily 
be mistaken for an ant-hill. No opening is visible, but by digging under the 
hill a hole is found, the mouth of the adit to the main tunnel, which may be 
three feet below the surface if made in cold weather, but perhaps not more 
than six inches if in summer. One of these mounds is thrown up in a very 
few moments ; I have seen thirty raised in a single night on the line of one 
tunnel ; this would represent nearly one hundred feet of tunneling. I have 
seen one hundred and fifty in one continuous row raised in about two days ; 
this would make between four and five hundred feet of burrow completed in 
that short time apparently by one little animal, an amount of work which may 
seem incredible to one who has not watched the restless movements of these 
animated plows, which are seemingly as well adapted for piercing the sand as 
birds are for cleaving the air. The burrows are about two and one -half inches 
in diameter, barely large enough to admit a man's hand, and, as has been 
stated, are at various depths below the surface. They meander in all direc 
tions, except in straight lines; their builders being guided apparently only by 
their whims or their olfactories. They, no doubt, intersect each other at many 
