522 HARPER: CLIMAX VEGETATION 
itself. Not only is shade conducive to the accumulation of humus 
and the retention of moisture, but the shade, humus, and moisture 
together favor the development of bacteria, fungi, worms, insects, 
and numerous other inconspicuous organisms, both animal and 
vegetable, which work on the soil in such a way as to increase the 
available nitrogen and otherwise improve it.* This, in short, is 
the regular way in which biotic succession (as defined by Cowles, 
and distinguished from regional and topographic succession) pro- 
ceeds, the world over. 
For the Florida hammock peninsulas, if not for all the other 
cases referred to, the key to the situation can be expressed in a 
single word: FIRE. All the long-leaf pine forests of which I have 
any knowledge (and I have seen them in seven of the nine states 
and about 200 out of some 300 counties in which Pinus palustris 
is found), even the more or less isolated ones in the mountains of 
Georgia and Alabama, bear the marks of frequent fires. In pre- 
historic times these fires were doubtless set by natural agencies, 
probably mostly lightning, and each spot perhaps did not get 
burned over oftener than once in several years, on the average. 
Although fires may not be started by lightning on any one square 
mile oftener than once in several decades, a fire once started in 
the grassy carpet of an unbroken pine forest might easily spread 
over several square miles, so that every acre of such forest if not 
protected in some way would be likely to be burned over every 
few years. Forest fires are now usually set by man,{ sometimes 
purposely and sometimes accidentally, but the increase in number 
of fires due to this cause has been partly counterbalanced by the 
numerous highways, clearings, etc., which serve as barriers to fire, 
so that the frequency of fire at any one point in the pine barrens 
may not be over two or three times as great as it was in prehistoric 
times, and the geographical distribution of forest fires in the south- 
eastern states has probably not been changed materially. 
* See in this connection a paper by J. W. Harshberger in Science II. 33: 741-744. 
12 My IogII. 
T Bot. Gaz.51:171-181. 1011. 
t The Indians are supposed to have burned the pine woods for centuries before 
the white man arrived, and done it more regularly, but the differentiation between 
pine barrens and insular and peninsular hammocks probably antedates the whole 
human race. 
