126 THE GEOGRAPHICAL DISTRIBUTION OF ANIMALS. 
breeding without these narrow limits, although frequently 
blown into the lower vallies. In a zone below these lofty. 
summits, but not extending to the base of the mounta 
Argynnis Montinus Seudd. is found, yet never at the sum 
mit with the Chionobas, nor about the base with the specie 
of the Canadian or Virginian faune.* Many other spe 
of insects, and many plants, are restricted in the samem 
ner. It cannot be that. these species are thus restricted 
their distribution merely by some primary, innate princi 
which prevents their diffusion ; for, like their marine re 
tives, they have not always been thus restricted. A 












zal 
7 
ba 
= 
land as it is in the ocean, for land species are not so 
left fossil in their ancient homes ; and, as there are no 
thentic records of land animals existing through the Gh 
epoch, we can go back no farther than its decline. Yetit 
worthy of remark, that the arctic land fauna of the Tertiary f 
period, like the marine, was probably cireumpolar ; and 
the gradually advancing coldof the glaciers would have drit 
many arctic plants and animals southward, and, living just 
beyond the border of the ice belt, they would have fol 
it back with the glacial decline. 
_ At the close of the Glacial epoch, the fauna and the 
of New England must have been very much like that of 
As the migra 
as a€rial is] 
evidence in the northern Species scattered along its P 
M * S. H. Scudder, Remarks on some Characteristics of the Insect Fauna of the W 
ountains, Journal of the Boston Society of Natural History, Vol. VIL 
t Packard, Glacial Phenomena of Labrador, p. 256, 
