"had grown up with black-walnut, another and different growth from that
prostrated by the force of the storm." In this instance, there were no
neighboring trees, except perhaps at distant intervals, from which the
nuts of the black-walnut could have been derived, unless they had been
promiscuously strewn by the tornado along its entire track. But it is,
unfortunately, not stated that the tornado occurred at that opportune
season of the year when the nuts were properly matured for planting.

In many parts of the United States, particularly in the South and West,
the paths of local tornadoes--those sweeping the native forests long
before the axe of civilization invaded them--may still be traced by the
alternations of timber growths, extending for long distances, and
through forests where there were no neighboring trees from which it was
possible that their seeds could have been derived. One of these
tornadoes the writer traced many years ago (as early as 1837) in South
Alabama, and he is satisfied, both from observation and reading, that
the instances are rare, if not altogether exceptional, where the clean
path of a tornado, through any of our primitive forests, has been
succeeded by the same growth of timber as that borne down by the winds.
Where the path of this ancient tornado of Alabama swept through a pine
forest, a clean growth of oak was buttressed on either side by pine;
and _vice versa_, where it swept an oak forest. And it is certain that
the tornado, whenever it may have occurred, could have exhibited no such
discriminating freak as alternately to distribute acorns in pine
growths, and pine cones in oak growths, either to make good a scientific
theory or balk an unscientific one.

Professor Agassiz, in passing through a dense young spruce forest some
years ago, on the south shore of Lake Superior, noticed that the ground
was thickly strewn with fallen birch trunks, showing that their place had