a a es 
HARPER: COASTAL PLAIN OF THE CAROLINAS 361 
palustris, which is preéminently the tree of the pine-barrens, very 
gradually gives way to P. Zaeda northeastward, disappearing en- 
tirely before reaching the James River. In fact there are in the 
Carolinas some limited areas which might reasonably be called 
pine-barrens where the pines are all P. Zaeda.* This is probably 
never the case within the range of P. Elliottii. 
A notable exception to the general thinning out of the pine- 
barren vegetation northeastward is found in the southern corner of 
North Carolina, where Pinus palustris becomes again the prevail- 
ing pine, and the flora is perceptibly richer than it is a little to the 
southwest as well as to the north, with little if any corresponding 
increase in diversity of habitats. Some of the species characteriz- 
_ing this minor pine-barren center are enumerated below. 
Within the pine-barrens the local diversity of the vegetation is 
of course governed mainly by the slight inequalities of the surface. 
The greater part of the area consists of dry and moist pine-bar- 
rens, the latter mostly occupying broad shallow depressions or 
nearly flat areas, rather than evident slopes as in the Altamaha 
Grit region of Georgia. The ponds of course contain their own 
characteristic flora, quite different from that of the moist pine- 
barrens. All the streams are bordered by swamps, the width of 
each of which is approximately proportional to the volume of the 
Stream. As in all pine-barren regions, most of the angiospermous 
trees are confined to the vicinity of streams, and the swamps ot 
rivers originating above the fall-line have quite a different flora 
from those of the pine-barren streams, as will be illustrated below 
in discussing the distribution and habitat of certain species. 
Scattered through the flat parts of North Carolina, and to a 
lesser extent in adjacent ey are many “ — ot Amese 
known to extend beyond Miasiasipes (See Torreya 6: 200, n0G 
1906.) Ser 
ferrulata, vate geminata, Cliftonia monophylla and ear ges a irae have a 
Similar distributio 
. ivilization | ‘ia of course changed the relative abundance of these trees to some 
extent, but in making the statements in this paragraph I have tried to reconstruct the 
primeval conditions as far as possible. 
Often spelled ‘* pocosons,’’ The use of this term seems to be almost ore to 
€astern North Carolina, but it appears to some extent in other states, though with vari- 
ations in meaning in different localities, as in the case of several other native a8 
names, like swamp, hammock, prairie, bay, savanna, brake, etc. The northernmost 
Tecord of such a word which I have come across is in York County, Virginia, where 
