CHAP. II.] GEOLOGY. 



&quot; drift&quot; on the Saco river, thirty miles to the north of Ports 

 mouth, contain the entire skeletons of a fossil fish of the same 

 species as one now living in the Northern Seas, called the cape- 

 Ian (Mallotus villosus), about the size of a sprat, and sold abun 

 dantly in the London markets, salted and dried like herrings. I 

 obtained some of these fossils, which, like the associated shells, 

 show that a colder climate than that now prevailing in this re 

 gion was established in what is termed &quot;the glacial period.&quot; 

 Mr. Hayes took me to Kittery, and other localities, where these 

 marine organic remains abound in the superficial deposits. Some 

 of the shells are met with in the town of Portsmouth itself, in 

 digging the foundation of houses on the south bank of the river 

 Piscataqua. This was the most southern spot (lat. 43 6 N.) 

 to which I yet had traced the fossil fauna of the boulder period, 

 retaining here, as in Canada, its peculiar northern characters, 

 consisting of a profusion of individuals, but a small number of 

 species ; and a great many of those now abounding in the neigh 

 boring sea being entirely absent. It is only farther to the south, 

 and near the extreme southern limit of the drift, or boulder clay, 

 as at Brooklyn, in Long Island, for example, that a mixture of 

 more southern species of shells begin to appear, just as Professor 

 E. Forbes has detected, in the drift of the south of Ireland, the 

 meeting of a Mediterranean and Arctic fauna. 



Every where around Portsmouth I observed that superficial 

 polish in the rocks, and those long, straight, grooves or furrows, 

 which I before alluded to (p. 18), as having been imprinted by 

 icebergs on the ancient floor of the ocean. By the inland posi 

 tion of these fossil shells of recent species, the geologist can prove 

 that, at times comparatively modern in the earth s history, the 

 larger part of New England and Canada lay for ages beneath 

 the waters of the sea, Lake Champlain and the valley of the 

 St. Lawrence being then gulfs, and the White Mountains an 

 island.* But it is a curious fact that we also discover along this 

 same eastern coast signs no less unequivocal of partial subsidence 

 of land at a period still more recent. The evidence consists of 

 swamps, now submerged at low water, containing the roots and 

 * See my &quot;Travels in N. America, 1841-2,&quot; vol. ii. p. 142. 



