STEVENSON— INTERRELATIONS OF THE FOSSIL FUELS. 85 



Rashleigh"^ found 4 feet of peat in the upper part of the section, 

 while the merely vegetable soil of Colenso's section is " solid 

 black fen." 



The presence of trees unassociated with peat has been regarded 

 by some as evidence of allochthonous origin, as reforestation of an 

 area after entombment of the peat seemed improbable. But refores- 

 tation is comparatively rapid on a new surface, provided only that 

 there be moisture. During the great Missouri River flood of 1903,"'' 

 the water, diverted from the channel by obstructions piled against a 

 railroad bridge, swept over a wide area. Crops were ruined and a 

 nursery field, near Topeka, Kansas, was covered with sand, which 

 buried the young trees. But within three months, the naked fields 

 were green with young cottonwoods, growing from seed blown in 

 after subsidence of the flood waters. Even dunes, consisting of loose 

 sand, become covered with vegetation and eventually with forest. 



Reforestation is rapid even amid untoward conditions. Seventy- 

 five years ago, the White Mountains of New Hampshire were cov- 

 ered with a dense forest, mostly spruce. Lumbermen denuded a very 

 great part of the surface and their labors were supplemented by 

 forest fires, which destroyed trees elsewhere. Where the soil was 

 burned off, so as to be washed away and to expose the glaciated sur- 

 face, nothing grew ; but elsewhere the restoration was rapid. Plants 

 of various types took prompt possession and prevented erosion. They 

 were succeeded by birch and cherry, in whose shade the conifers 

 grew. On the neglected farms of that region, one finds all stages 

 of restoration, from pasture lots invaded by sturdy weeds to the 

 forest of fir^ and spruces, which have overcome the birches. The 

 conditions are similar in Ontario, as Miller and Knight"' have 

 shown. Their statement respecting one area is : 



" Years ago, the area was visited by heavy fires which destroyed all but 

 a few of the pine trees that were numerous and made the area important for 

 its timber. On the part of the lake referred to, a few red pines and one or 



115 p. Rashleigh, " An Account of the Alluvial Depositions at Sandry- 

 cock," Trans. R. Geol. Soc. Cornzvall, Vol. II., 1822, p. 282. 



11^ H. C. Frankenfield, " The Floods of the Spring of 1903 in the Missis- 

 sippi Watershed," Bull. M, U. S. Weather Bureau, 1904, p. 62. 



11" W. G. Miller and C. W. Knight, " Pre-Cambrian Geology of South- 

 eastern Ontario," Toronto, 1914, p. 18. 



