ERRORS ABOUT ROOTS. 99 



farther laterally than do the branches. Perhaps 

 it would be safe to say that the roots of trees 

 extend on the average to a distance equal to the 

 total height of the tree. They sometimes fall 

 short of that limit, but often exceed it, especially 

 in the open ground where trees do not grow as tall 

 as in a forest. I have, in the forest, traced roots of 

 the American Elm to a distance of one hundred and 

 twenty feet, which was about the height of the tree. 

 Professor Burrill tells of an elm, the roots of which 

 filled a tile drain 450 feet away. 



The depth to which roots penetrate depends very 

 much on the character cf the soil. If it is dry 

 and porous they extend deeper than if it is wet. 

 Thus in a swamp the roots of trees grow mainly 

 near the surface, and the trees are easily over- 

 turned by the wind when the surrounding trees are 

 cut away. Such trees are liable to die whenever 

 the swamps in which they grow are drained, 

 while trees of the same kinds on upland thrive 

 equally well in the drier soil. It is not, there- 

 fore, always true because certain trees are found 



v 



mainly in swamps that they naturally prefer 

 such locations. The fact that a moist soil and 

 climate favors the formation of roots near the 

 surface has recently been well illustrated in a 

 comparison of the roots of forest trees in the 

 eastern states with those of the same species grow- 



