62 HARPER 



York and Virginia. About 20 species are confined to the coastal 

 plain of the southeastern states, but not to the pine-barrens. 

 Half a dozen or so are not known outside of Georgia. About 25 

 species extend inland to sandy bogs in Middle Georgia and the 

 neighboring mountains, and fifteen or twenty have a wide dis- 

 tribution in similar habitats in the glaciated region of the north. 1 

 Ten others grow almost anywhere in the Eastern United States. 

 About 15 are supposed to range southward to tropical America. 

 This is rather anomalous, for there can be no Lafayette and 

 Columbia formations in the tropics; and doubtless some of our 

 plants will hereafter be found to be specifically distinct from their 

 tropical allies. 



5. Branch-Swamps. 



If we start at the head of any little valley in the Altamaha Grit 

 region and go down-hill, we come in a very short distance, after 

 passing through dry, intermediate and moist pine-barrens, to a 

 branch-swamp occupying the trough of the valley. A branch- 

 swamp, like any other swamp, 2 is characterized by the pre- 

 dominance of trees and shrubs, presenting a marked contrast 

 to the pine-barrens immediately adjoining. The difference in 

 vegetation is not to be explained by differences in the amount of 

 water in the soil, for branch-swamps are not much if any wetter 

 than the moist pine-barrens, and trees and shrubs are by 

 no means confined to wet places. The explanation is doubtless 

 to be found in the humus which the swamps contain. What 

 little humus is formed in the pine-barrens of course tends to 

 accumulate at the bottom of the valleys, and gradually prepares 

 the soil for the growth of broad-leaved woody plants. These 

 plants as they arrive naturally produce more humus, and there is 

 every reason to believe that the branch-swamps are tending to 

 increase their area in this way, independently of topographic 

 or climatic changes, though the process is probably so slow 

 that it would not be perceptible in a single life-time. The trees 

 and shrubs are of course accompanied by herbs which thrive in 



1 See Rhodora 7: 69-80. April, 1905. 

 See footnote on page 21. 



