266 MISS MARIETTA PALLIS ON THE 
healthiness of appearance of the reed from the open reed-swamp stage to the 
time when, as Plav, it gradually declines in size. Some tests which I applied 
seem also to confirm this. I had all the reed from a unit area, 4 square 
metres (6 feet 63 inches across), cut and weighed, green, on the spring- 
balance (see р. 235). The results were :— 
These few tests indicate that the output of reed in weight per unit-area 
varies relatively little whether the reed be tall-stout, tall-slender, medium- 
slender, etc. Apparently, therefore, according to these tests, the major 
individual is more or less equally vigorous at each of these periods, which 
would not be the case were it poisoned. 
(b) Because, when the reed becomes detached, there is already an accumu- 
lation of fluvio-lacustrine ooze covering the bottom of the lake—ooze that 
contains a considerable amount of organic matter proceeding from the reed 
itself—yet there is no change in the size or healthy appearance of the reed 
that has become detached ; so strictly localized an effect appears improbable 
on the assumption of the presence of poison. 
(c) Because though, in general, the periphery of a Plav or reed-swamp is 
liable to better conditions of aeration than its centre, yet the reed-swamp 
abutting on open-water often becomes detached first (many Plavs abut 
on open-water), and also the reed of the interior of a Plav may become 
slender later than the reed at the edge of a Plav*. Were poisoning the cause 
of death of the basal rhizomes, and therefore of detachment, or of the change 
in size, one would expect the stage, whether of detachment or of decrease 
in the size of the shoots, to take place, as a rule, in the interior of a swamp 
or of a Plav. 
Secondly, as regards the factor overcrowding. This does not seem to 
operate unfavourably, as one of the most crowded reed-beds—that around 
the grind of Rosulet (see No. V. of the table on p. 265)—-is full of vigour, 
and produces 164 lbs. (about 7:4 kilog.) of aerial shoots per unit-area. 
Nor do I think that crowding in itself—that is, without injury to health— 
has the effect of causing the production of small shoots. Among the short 
variants are some exceptionally short ones, yet the reed as a whole never 
becomes reduced to their length; it dies before, though buds still exist from 
which, one would suppose, yet smaller shoots could be produced. Apparently 
when a definite stage in the compoundness of the reed-system is reached, 
the reed major unit dies owing to vital exhaustion. 
Thirdly, as regards change in the relation of the surface of the Plav to 
that of the water-level. There is practically no such change so, long as the 
Plav remains floating. In this respect, Plav, “fixed Plav," and fen differ. 
Fourthly, that death is not caused by inter-competition of the reed with 
the accompanying plants, is quite evident. The accompanying plants, 
* Of course, wind or waves affect the edge of a swamp more than the interior; hence, 
possibly, slightly less ripe reed may become detached there. 
