8
T. B. REYNOLDSON
had any preference for either type; a single leaf, stem or stone was
often found carrying the four species. These initial observations have
been supported by all later sampling and the results of surveys made by
other workers, e.g. Berg (1938). Mirolli’s data (1961) for the distribution
of Polycelis nigra, Dugesia lugubris and Dendrocoelurn lacteum in Lake
Maggiore have been analysed from this aspect. His main variations in
conditions were between exposure and shelter, between muddy, sandy
and stony substrata. He found triclads in general most abundant on
relatively sheltered, stony shores. When the distribution of the various
species was analysed by a x2 test incorporating Yates’ correction it was
found that this differed very significantly from random (P < 0.001);
all three species tended either to occur together or to be absent. A
major contribution to the value of x was derived from one cell where the
expected number of negative values was low, but those actually observed were fairly high. However, if the contribution from this cell is
omitted, x2 still had a significant value (P < 0.01 >0.001). Such a
result supports the observation that when these species of triclad are
present in a lake they occupy the same habitat.
In this sort of approach the larger the number and the greater the
variety of lakes it is possible to include, the better is the chance of
detecting pattern. This prohibited using a boat and meant restricting
the sampling to shallow, littoral zones which could be reached from the
shore. Actually, our experience suggests that shallow water up to a
depth of about 1 ft (30-5 cm) is the optimal zone for all these species,
especially on sheltered, stony shores. In exposed places they may be
concentrated in deeper water. These triclads follow changes in water
level quickly (Beauchamp, 1932; Taylor and Reynoldson, 1962), and I
have observed both species of Polycelis on stones overlying grass several
metres from the original level within 12 h of flooding. They do extend
into deeper water as shown by Berg (1938) but not into the sub-littoral
(Humphries, 1936; W. J. P. Smyly, private communication). Although
the data for stony shores reveal that the four species occur at approximately the same depth and zonation is not pronounced, Chodorowski
(1959, 1960) studying vertical distribution in a lake with a mud or sand
substratum and a rich flora found some evidence of zonation related to
plant type.
In selecting lakes for quantitative sampling, size is important for
several reasons as Gorham (1958) has shown. Large water bodies such
as Windermere and Loch Lomond not only provide more varied habitats than smaller lakes because of changes in soil type along the shore
with accompanying changes in the flora, but there is also greater variation in severity of wave action and other factors. It is also more
difficult to select the most appropriate shore without a lengthy search.
T. B. REYNOLDSON
had any preference for either type; a single leaf, stem or stone was
often found carrying the four species. These initial observations have
been supported by all later sampling and the results of surveys made by
other workers, e.g. Berg (1938). Mirolli’s data (1961) for the distribution
of Polycelis nigra, Dugesia lugubris and Dendrocoelurn lacteum in Lake
Maggiore have been analysed from this aspect. His main variations in
conditions were between exposure and shelter, between muddy, sandy
and stony substrata. He found triclads in general most abundant on
relatively sheltered, stony shores. When the distribution of the various
species was analysed by a x2 test incorporating Yates’ correction it was
found that this differed very significantly from random (P < 0.001);
all three species tended either to occur together or to be absent. A
major contribution to the value of x was derived from one cell where the
expected number of negative values was low, but those actually observed were fairly high. However, if the contribution from this cell is
omitted, x2 still had a significant value (P < 0.01 >0.001). Such a
result supports the observation that when these species of triclad are
present in a lake they occupy the same habitat.
In this sort of approach the larger the number and the greater the
variety of lakes it is possible to include, the better is the chance of
detecting pattern. This prohibited using a boat and meant restricting
the sampling to shallow, littoral zones which could be reached from the
shore. Actually, our experience suggests that shallow water up to a
depth of about 1 ft (30-5 cm) is the optimal zone for all these species,
especially on sheltered, stony shores. In exposed places they may be
concentrated in deeper water. These triclads follow changes in water
level quickly (Beauchamp, 1932; Taylor and Reynoldson, 1962), and I
have observed both species of Polycelis on stones overlying grass several
metres from the original level within 12 h of flooding. They do extend
into deeper water as shown by Berg (1938) but not into the sub-littoral
(Humphries, 1936; W. J. P. Smyly, private communication). Although
the data for stony shores reveal that the four species occur at approximately the same depth and zonation is not pronounced, Chodorowski
(1959, 1960) studying vertical distribution in a lake with a mud or sand
substratum and a rich flora found some evidence of zonation related to
plant type.
In selecting lakes for quantitative sampling, size is important for
several reasons as Gorham (1958) has shown. Large water bodies such
as Windermere and Loch Lomond not only provide more varied habitats than smaller lakes because of changes in soil type along the shore
with accompanying changes in the flora, but there is also greater variation in severity of wave action and other factors. It is also more
difficult to select the most appropriate shore without a lengthy search.