ARVICOLA AMPHITBIA. 203 
dition of limestone caves as the cause of their introduction, 
viz., that of being traversed by small rivers, which in some 
limestone countries lose themselves in cavities, and after 
running through a series of subterranean caverns, reappear 
on the surface at a distance from the spot where they first 
disappeared. A change in the relative levels and other 
physical features of a country, and a variety of other modi- 
fying circumstances, might afterwards alter the relation of 
a cavern to such subterraneous stream. 
Dr. Buckland, however, is disposed to refer the intro- 
duction of the smaller Mammalia, as well as the larger 
ones, to the agency of the hyenas. He says,* ‘‘ The ex- 
treme abundance of the teeth of Water-rats has also been 
alluded to; and though the idea of hyenas eating rats 
may appear ridiculous, it is consistent with the omnivorous 
appetite of modern hyenas, and with the fact, quoted by 
Johnson, that they feed on small animals, as well as car- 
rion and bones; nor is the disproportion in size of the 
animal to that of its prey greater than that of wolves 
and foxes, which are supposed by Captain Parry to feed 
chiefly on mice, during the long winters of Melville Island. 
Hearne, in his ‘ Journey to the Northern Ocean,’ mentions 
the fact ‘of a hill, called Grizzly Bear Hill, being deeply 
furrowed and turned over like ploughed land, by bears in 
search of ground squirrels, and perhaps mice, which con- 
stitute a favourite part of their food.” If bears eat mice, 
why should not hyznas eat rats? Our largest dogs eat 
rats and mice; jackalls occasionally prey on mice, and 
dogs and foxes will eat frogs. It is probable, therefore, 
that neither the size nor aquatic habit of the Water-rat 
would secure it from the hyenas. They might occasion- 
ally, also, have eaten mice, weasels, rabbits, foxes, and 
* Reliquiz Diluviane, p. 33. 
> 
