STRUCTURE AND HISTORY ОЕ PLAV. 263 
that is to say, thick unbranched, or almost unbranched, roots, often of great 
length. Many of these roots are constricted at frequent intervals. 
The Minor Individual. 
The reed-shoots forming Plav, as already stated, are of several sizes. A 
Plav may be formed of one thicket of reed, the shoots of which are more or 
less of one size, or of several thickets, each formed of a distinctive size 
of reed and sharply separated from the others. There is, naturally, some 
variation of size within a thicket, but the dominant difference in size of the 
reed of different thickets, even though contiguous, is so marked in the delta 
of the Danube that it suggests the presence of several varieties of reed 
(see Pls. 18-21). 
The reed groups itself quite naturally into two main divisions as regards 
size. In the one group are all the reeds which I call “ giant,” or, better, 
“stout,” since there are short ones amongst them: these I regard as the 
juvenile form. They reach a height of about 5:15 m. (about 17 feet), as 
measured from the surface of the Plav. The other group includes the aged 
form of the reed, the slenderer and shorter reed, the Shitka Kamish of the 
fishermen, which varies in height from 1:2 m. or less to about 3:3 m. (about 
4 feet to about 11 feet). А reed-shoot, whether tall or short, can generally 
be assigned to either group, stout or slender, at a glance. Each of these two 
main groups can be further subdivided into tall-stout, tall-slender, short- 
slender, etc. In this case too the reed of each thicket is mere or less 
uniform in size *. The height-classes of the slender reed are more numerous 
and eonspicuous than those of the giant reed. 
That difference in size of the reed of the delta of the Danube does not 
signify difference of variety is shown, apart from the identity of the 
flowering parts :— 
(1) By the fact that the reed forming open and closed reed-swamp is 
almost invariably f stout, whereas slender reed is a Plav plant. Were 
slender reed therefore a distinct Plav variety, the obvious position in the 
Plav for its rhizomes would be above those of the stout reed and never at 
the base of the Plav, since Plav is formed of detached swamp reed, which is 
stout reed. Slender reed, however, when present in a Plav, occurs at its 
base—it occupies the position of a swamp plant without having been one. 
* In a smaller way, these divisions are also to be seen in Norfolk, around Sutton and 
Barton Broads for instance. 
t I found one exception to this rule at Ghola Kossa, a low grind on the south-west 
side of Lake Rogu. A narrow row of slender reed, whose total length was about 
3:15 m. (about 10 feet), was growing there in water about 1'65 m. (about 5 feet) deep 
on September 6th, 1913. І can offer no explanation for this exception. 
Many fragments of marine shells occur at Ghola Kossa, viz.:— Cardium edule, Linn., 
Mytilus edulis, Linn., mixed with fresh-water shells such as Planorbis corneus, Linn. 
