Oecologia (2003) 134:278–283
DOI 10.1007/s00442-002-1101-7
BEHAVIOURAL ECOLOGY
Gy
rgy Abrusn
Morphological variation of the predatory cladoceran Leptodora kindtii
in relation to prey characteristics
Received: 27 May 2002 / Accepted: 6 October 2002 / Published online: 20 November 2002

Springer-Verlag 2002
Abstract The responses of invertebrate predators to
changes in the morphology of their prey, and especially
the responses for induced defences, are largely unex-
plored, compared with the vast amount of data on
predator-induced defences. This study demonstrates that
the size of the feeding basket, the anatomical structure
used to capture prey by the predaceous freshwater
cladoceran Leptodora kindtii, shows significant allometric
changes with the average body size of the prey (herbiv-
orous cladocerans) in six lakes of northeastern Poland.
Prey density influences the basket size only in adult
Leptodora individuals, whereas induced defences such as
helmets of Daphnia sp. seem to have no effect on
Leptodora’s morphology. In a feeding experiment it is
shown that the increase in the feeding basket enables
Leptodora to capture larger prey. The seasonal pattern of
morphological change, and its measurable benefit suggest
that the observed morphological variability of Leptodora
is phenotypic plasticity.
Keywords Multiple regression · Ceriodaphnia reticulata ·
Daphnia cucullata
Introduction
Research on adaptive phenotypic plasticity focuses
mainly on inducible defences, that have been documented
for several taxa from algae to vertebrates (Tollrian and
Harvell 1999a; Agrawal 2001). In cladoceran zooplank-
ton, phenotypic plasticity has been known for decades,
and cladocerans are among the most extensively studied
examples of predator-induced defences [see Tollrian and
Dodson (1999) for recent review].
G. Abrusn ())
Department of Hydrobiology, University of Warsaw, ul. Banacha 2,
Warsaw 02–097, Poland
e-mail: abrusan@hydro.biol.uw.edu.pl
Tel.: +48-22-5546437
Fax: +48-22-5546426
Since predator-induced morphological defences occur
in prey, also prey-induced “offences” may develop in
predators (Havel 1987). Diet-induced phenotypic changes
in the feeding apparatus are known for a number of
animal groups, particularly fish (Day and McPhail 1996;
Mittelbach et al. 1999; Hjelm et al. 2000, 2001; Magn-
hagen and Heibo 2001) and arthropods (Smith and Palmer
1994; Thompson 1992; Bernays 1986). Among zooplank-
ton, a plastic reaction is known for the predaceous rotifer
Asplanchna sieboldi – the transition between different
morphotypes (saccate-cruciform-campanulate) is deter-
mined by the presence of vitamin E in the available prey
(Gilbert 1980). However, the phenotypic and evolutionary
responses of predators to induced defences of their prey
has received little attention (Adler and Grnbaum 1999).
Leptodora kindtii is a common planktonic predator,
widely distributed in the Palearctic. Its main prey items
are other cladocerans (Rivier 1999; Browman et al. 1989;
Lunte and Luecke 1990). Induced defences in cladoceran
prey strongly influence the effectiveness of Leptodora
attacks: “defended” forms of Bosmina spp. (Hellsten et al.
1999) and Daphnia cucullata (Agrawal et al. 1999) were
shown to be much less vulnerable than the ones without
defences, thus adaptive responses of Leptodora for
changing prey morphology were expected.
The thoracic appendages of Leptodora form a “feeding
basket”, a structure used for prey capture and handling.
Feeding experiments with differently sized Leptodora (3–
11 mm) led Herzig and Auer (1990) to conclude, that the
size of the feeding basket is the driving variable in prey
selection.
There are two aims of this study: first, to demonstrate
that the size of the feeding basket of L. kindtii is variable:
allometric variation was expected in the size of the
feeding basket between lakes with different zooplankton
size, density and morphology (inducible defences). Sec-
ond, in feeding experiments the benefit of Leptodora’s
morphological change should be demonstrated.

Materials and methods
Field sampling and analysis of field data
Six lakes in the Mazurian Lakeland, northeastern Poland were
studied, ranging from shallow eutrophic, to stratified meso-
eutrophic lakes (Table 1). The sampling sites were located at the
deepest points of each lake, in the case of the two shallow ones the
geometric centres were chosen. Samples were taken with plankton
nets (mesh size 50 m) by vertical hauls from the whole water
column, or from the top of the anoxic zone. To collect L. kintdii in
sufficiently large numbers, oblique trawls were also taken with nets
of 400-m mesh size. All samples were preserved in 4% sucrose-
formalin. The lakes were sampled in 1998, in May (27–31), July
(12–15), August (12–15), September (12–15) and October (8–12).
All morphological measurements were made with an image
analysing system (MultiScan) connected to a stereoscopic micro-
scope. In herbivorous cladocerans two morphological parameters
were measured: carapace length (from the base of the tailspine, or
its equivalent to the top of the eye), and the total length of the body
(including helmet, if present). Relative helmet length was indicated
by the proportion of total length/carapace length, and its value
ranged from 1 (no helmet) to 1.6 (D. cucullata with strongly
developed helmet).Values of average carapace length were based
on a minimum of 100 randomly selected individuals, and represent
the average for all cladoceran species present in a particular lake, at
one sampling time. Average helmet length was calculated in a
similar way. Species with induced defences other than helmets (e.g.
Bosmina therisites) were present at very low densities, so statistical
analysis of these data was not performed. For each lake, the density
of filter-feeding cladocerans was calculated for every sample.
In L. kindtii three parameters were measured: body length (from
the end of the head to the tailspine bifurcation) along the axis of the
body, the base of the feeding basket, and body thickness, measured
at the connection of the thorax with the abdomen (Fig. 1). Since the
first males appeared in September, only females were measured. In
each sample usually 25–30 females were measured, though in some
lakes in May and October the density of females was so low that
only smaller numbers were collected (12–20 individuals). Alto-
gether, 765 Leptodora individuals and >4,000 other cladocerans
were measured in the preserved samples.
Statistical analyses
The field data were analysed by multiple linear regressions to
separate the effects of several explanatory variables on the feeding
basket and body thickness.
The influence of different explanatory variables on the feeding
basket was shown by standard partial regression coefficients (beta
weights, bn). The compared meso- and eutrophic lakes differ from
each other in so many parameters, that the analyses were restricted
to parameters of prey, for which a biologically reasonable
explanation for any significant effects may be easier to find. Two
analyses were performed with different explanatory variables:
Table 1 Some characteristics of the lakes studied
Lake Area Maximum Chlorophyll-aa
(ha) depth (g l–1)
(m)
Gołdopiwo 891 26 32
Iławskie 116 3 70
Kuc 95 28 4 5–6.75
Miłkowskie 20 15 51
Piłakno 237 56 6 3–6
Tynwałdskie 30 3 170
a
Data from August
279
Fig. 1 Outline of Leptodora kindtii. Body length (A), feeding
basket (B), body thickness (C)
1. The explanatory variables used: prey density, average carapace
length, and helmet length. In these regressions all the 30 data
sets from the six lakes and five sampling times were treated as
independent. When the data were pooled, the feeding basket had
larger variance in large individuals than in small ones, so
discrete size ranges of Leptodora were used in each regression
(see Table 2).
2. The explanatory variables used: the date of sampling and
Leptodora body length. Partial residual plots between the date
of sampling (shown as the day of the year) and basket size were
used to show the seasonal variability of the basket.
The influence of different explanatory variables on body
thickness was investigated using one multiple regression with all
data, with four explanatory variables: prey density, average prey
size, helmet length and Leptodora length.
Feeding experiments
Living Leptodora specimens were collected (August 2000) in Lake
Roś, a eutrophic lake in the Mazurian Lakeland, and in the
mesotrophic Lake Kuc (Table 1), with a plankton net of 320-m
mesh size. Leptodora individuals were starved for 24 h before
feeding. As prey, I used laboratory-reared Ceriodaphnia reticulata
clones. Only adult C. reticulata were used in the experiments, with
average carapace length 0.833€0.077 mm (n=153). The observa-
tions were carried out as described by Herzig and Auer (1990), at
18€1C. One hundred prey individuals were put in each of two
Petri dishes, containing 80 ml GF/C filtered lake water, with side
walls covered by black foil. Five Leptodora were added to each
Petri dish. Whenever a Leptodora captured prey, it was gently
transferred to a glass plate in a drop of water (with the prey in the
feeding basket), and a new Leptodora specimen was added to the
dish. When Leptodora had finished feeding on C. reticulata, the
Range of Range of cladoceran Range of mean
Secci depth density cladoceran size
(m) (individuals l–1) (mm)
2.1–4 18–135 0.393–0.711
0.45–0.9 20–596 0.275 – 0.438
11–46 0.384 – 0.614
1–1.75 82–1654 0.283 – 0.513
2–140 0.384 – 0.704
0.4–0.8 21–78 0.293 – 0.500
280
morphological parameters of both prey and predator were mea-
sured. The parameters measured in L. kindtii included: total length,
basket base size, and egg diameter. The number of eggs was
counted in the experimental populations only, because eggs fall out
of the brood sac when the samples are preserved in sugar formalin.
On Ceriodaphnia, only carapace length was measured (the carapace
of Ceriodaphnia remains nearly intact during Leptodora feeding).
During the measurements, Leptodora were narcotised with carbon-
ated water.

Results
The only variable that significantly influenced the body
thickness of Leptodora was its body length, explaining
85.5% of the variance (the overall regression explains
85.7% of the variance), thus body thickness was not
influenced by prey characteristics.
In the analysis of the feeding basket of animals
collected in the field, the type A multiple regressions
revealed that the size of the feeding basket correlated
significantly with average prey carapace length for nearly
all size ranges of Leptodora (Fig. 2, Table 2), the
exception being the largest (8–8.99 mm) size range.
Significant partial correlations were found between the
size of the feeding basket and prey density in three
regressions with adults, and in two cases partial regres-
sions of helmet length are also significant (Table 2). The
size of the feeding basket of Leptodora changed signif-
icantly (P<0.05) during the season in five of the six
studied lakes (Fig. 3).
The basket size of Leptodora from Lakes Kuc and Roś
used in the feeding experiment differed significantly
(ANCOVA, df=1,140, F=260.15, P<0.001, Fig. 4).
Feeding on the same prey, Leptodora from lake Kuc
(with larger baskets) captured larger C. reticulata
(0.852€0.068 mm) than individuals from Lake Roś
(0.787€0.076 mm, Fig. 4, t-test, t=–5.253, df=141,
P<0.001). The number of eggs in the two populations
differed markedly (Fig. 4, ANCOVA, df=1,106,
F=264.15, P<0.001), Leptodora from Lake Roś had
more (Fig. 4), eggs than mature individuals in Lake Kuc.

Discussion
The morphology of L. kindtii is variable, but this
variability is much less spectacular than the well visible
(and qualitative) defences of herbivorous cladoceran
species. From the three parameters of the prey examined
in the field populations, the average carapace length of Fig. 2 A A multiple regression with Leptodora of 5–5.99 mm
prey explains best the variability of the size of the feeding (Table 2) using carapace length and helmet size as explanatory
basket, while the effects of prey density or helmet size are variables. Prey density (n.s., see Table 2) was removed using
backward elimination procedure. B, C Partial regressions for the
visible only in large individuals (Table 2). The three same explanatory variables. Note that R2 values are partial
significant beta values for prey density are of comparable correlation coefficients. Relative basket size = bnXn+Residuals,
magnitude to those for prey size, and increase with where bn is the slope of the partial regression, Xn is the explanatory
Leptodora body length (Table 2). This suggests that the variable
morphology of large Leptodora is influenced also by the
density of prey, and not just prey size. Recently
Branstrator (1998) showed experimentally on differently
 281

Fig. 4 A Feeding basket sizes (circles) of the experimental

Leptodora specimens from lake Roś and lake Kuc, and the size
of captured prey (squares, note the difference in scales). B Clutch
sizes of the same specimens

Fig. 3 Seasonal changes in the size of the feeding basket (triangle)

and the size of herbivorous cladocerans (circles), in the six studied
lakes. The changes in the feeding basket size are shown as partial
residual plots, and the R2 values are partial correlation coefficients.
Filled circles show communities with Bosmina sp. dominance
(>80% of cladocerans), open circles show communities with
Daphnia sp. dominance (>80%), grey circles show communities
where both Bosmina and Daphnia were more abundant than 20%.
Note that in lakes Miłkowskie, Gołdopiwo and Kuc the basket size
show moderate nonlinearity, the relative basket size changes only
slightly after August
sized Leptodora that the profitability and ingestion
efficiency of captured prey increase with the ratio of
feeding basket size to prey size. Thus even when fed on
similarly sized prey, Leptodora with a smaller basket
ingests less than one with a larger basket. This is probably
caused by handling difficulties. The different ingestion
rates of individuals with differently sized feeding baskets
may explain the observed correlation of the basket size
with prey density in large Leptodora (Table 2) – at low
prey densities the increase in basket size improves
ingestion efficiency. Although the average helmet size
has statistically significant partial correlations with the
feeding basket in two regressions (Table 2), its beta

Table 2 Multiple regression parameters for different size classes of Leptodora kindtii. Significant (P<0.05) partial regressions are shown
with boldface. The feeding basket size was the dependent variable in all regressions
Leptodora length Independent variables (1–3) R2 F n
(mm) –1
Helmet size Log prey density (individuals l ) Average carapace length (mm)
b1 b2 b3
2–2.99 –0.18 0.02 0.58 0.24 5.64 45
3–3.99 –0.08 0.02 0.57 0.30 11.2 79
4–4.99 –0.14 –0.02 0.63 0.36 24.6 131
5–5.99 –0.18 –0.08 0.73 0.52 62.8 173
6–6.99 0.06 –0.30 0.50 0.53 59.1 159
7–7.99 –0.18 –0.49 0.37 0.61 46.4 93
8–8.99 0.02 –0.54 0.09 0.39 9.60 48
282
coefficients are so low (Fig. 2, Table 2), that these partial
regressions have no predictive power in practice. In
general, no convincing signs of adaptation to induced
defences of prey were found in Leptodora. A possible
reason for the lack of morphological response to changing
prey morphology may be, that species composition,
density, and size of the zooplankton community change
more rapidly than induced defences, and any possible
effect of the latter is overshadowed.
The size of the feeding basket changes during the
season, and it seems to follow the prey size rather than
any regular cycle (Fig. 3). The change in the morphology
of Leptodora was largest in May and July, when the
Bosmina-dominated prey communities changed to Daph-
nia-dominated ones in all studied lakes. From August
onwards, the morphological change is small. In contrast,
Manca and Comoli (1995) reported no significant
seasonal change in the outer diameter of the basket in
Leptodora from the large, oligotrophic Lake Maggiore
(Italy). A possible reason for this contradicting result is
that cladocerans in Lago Maggiore are much larger than
in the lakes studied in this survey. The maximal size of
Daphnia in Lago Maggiore is above 2.0 mm (Manca and
Ruggiu 1998) and selection may favour the largest
possible feeding baskets. In the meso- and eutrophic
lakes of this study, Daphnia individuals of carapace
length longer than 1.0 mm were only occasionally found.
The experiments demonstrated a benefit of the
increased basket size: a larger feeding basket enables
Leptodora to capture larger prey, when individuals of the
same body length but with different basket morphologies
(Fig. 4) are compared.
The presence of forms with a small feeding basket
implies that a large one is costly. A possible cost is the
energetic investment needed to build the additional tissue.
Such a cost may be substantial in species whose
morphology varies greatly (as crests in some daphnids,
Barry 1994); however, in the case of relatively small
morphological changes costs are not always measurable
(Scheiner and Berrigan 1998 ; Tollrian 1995). A cost may
also result from the maintenance of a structure (Tollrian
and Harvell 1999b). Altered morphology may result in
increased energetic cost of swimming, or smaller swim-
ming speed, since morphological changes influence the
hydrodynamic properties of the animal. The effect of
morphology on swimming performance of Bosmina sp.
has recently been demonstrated (Lagergren et al. 1997).
Preliminary analyses of the swimming pattern of Lep-
todora (Gy. Abrusn, unpublished data) suggest that the
increase in the feeding basket size increases measurably
the drag of the animal.
The question of whether the mechanism underlying the
morphological change in Leptodora is phenotypic plas-
ticity or the replacement of different genotypes (or both)
is unsolved. Any experimental separation of plastic and
genetic components of this change would require a
method of culturing Leptodora, that at present does not
exist. Genetic components of plastic reactions have been
shown for clonal organisms by many authors (Weider
1984; Parejko and Dodson1991; Bruno and Edmunds
1997), but only few studies have addressed the question
of the importance of phenotypic plasticity and genetic
variability in seasonal morphological changes (Lampert
and Wolf 1986 ; Stibor and Lampert 2000). In these
studies phenotypic plasticity seems to be more important.
Assuming that the morphological variability of Leptodora
is partly caused by phenotypic plasticity, at least two
factors can serve as proximal cues for the morphological
changes: (1) increased muscle power during prey han-
dling can result in larger muscles and exoskeleton – this
has been shown responsible for the changes of the feeding
structures in phytophagous insects (Bernays 1986) and
crabs (Smith and Palmer 1994); (2) the filter-screen areas
of several Daphnia species have been found to increase
when grown under food-limiting conditions (Koza and
Kořnek 1986; Lampert 1994) and the limbs of Leptodora
that form the feeding basket are homologous anatomical
structures. The observation that Leptodora from lake Kuc
has fewer eggs than specimens in Lake Roś (Fig. 5) also
suggest a reaction to scarce food conditions (but in this
case, one would expect strong dependence of basket size
on the density of prey, and less on average carapace
length).
Conclusions
This is the first study to demonstrate that the cladoceran
predator, L. kindtii, is morphologically variable, and its
morphology is correlated with the size and density of its
prey. Although a laboratory method for culturing Lep-
todora is still lacking, the seasonal pattern of the
morphological change and the presence of measurable
benefits suggest that the observed morphological vari-
ability is phenotypic plasticity.
Acknowledgements I thank Anna Jachner, Joanna Pijanowska,
Piotr Dawidowicz and the anonymous reviewers for comments on
the manuscript. The field sampling and the feeding experiments
were done at the Hydrobiological Station of the University of
Warsaw, Pilchy. The research was supported by a grant from the
University of Warsaw to Piotr Dawidowicz (BW 1420/25), and a
KBN grant (6PO4F 04717) to Gyrgy Abrusn.
References
Adler FR, Grnbaum D (1999) Evolution of forager responses to
inducible defenses. In: Tollrian R, Harvell CD (eds) The
ecology and evolution of inducible defenses. Princeton Uni-
versity Press, Princeton, N.J. pp 259–285
Agrawal AA (2001) Phenotypic plasticity in the interactions and
evolution of species. Science 294:321–326
Agrawal AA, Laforsch C, Tollrian R (1999) Transgenerational
induction of defenses in animals and plants. Nature 401:60–63
Barry MJ (1994) The cost of crest induction of Daphnia carinata.
Oecologia 97:278–288
Bernays EA (1986) Diet-induced head allometry among foliage-
chewing insects and its importance for graminivores. Science
231:495–497

Branstrator DK (1998) Predicting diet composition from body
length in the zooplankton predator Leptodora kindtii. Limnol
Oceanogr 43:530–535
Browman HI, Kruse S, O’Brien WJ (1989) Foraging behavior of
the predaceous cladoceran Leptodora kindti. J Plankton Res
11:1075–1088
Bruno JF, Edmunds PJ (1997) Clonal variation for phenotypic
plasticity in the coral Madracis mirabilis. Ecology 78:2177–
2190
Day T, McPhail JD (1996) The effect of behavioural and
morphological plasticity on foraging efficiency in the three-
spine stickleback (Gasterosteus sp). Oecologia 108:380–388
Gilbert JJ (1980) Developmental polymorphism in the rotifer
Asplanchna sieboldi. Am Sci 68:636–646
Havel JE (1987) Predator-induced defenses: a review. In: Ker-
foot WC, Sih A (eds) Predation. Direct and indirect impacts on
aquatic communities. University Press of New England,
pp 262–278
Hellsten M, Lagergren R, Stenson J (1999) Can extreme morphol-
ogy in Bosmina reduce predation risk from Leptodora? An
experimental test. Oecologia 118:23–28
Herzig A, Auer B (1990) The feeding behaviour of Leptodora
kindtii and its impact on the zooplankton community of
Neusiedler See (Austria). Hydrobiologia 198:107–117
Hjelm J, Persson L, Christensen B (2000) Growth, morphological
variation and ontogenetic niche shifts in perch (Perca fluvi-
atilis) in relation to resource availability. Oecologia 122:190–
199
Hjelm J, Svanb ck R, Bystrm P, Persson L, Wahlstrm E (2001)
Diet-dependent body morphology and ontogenetic reaction
norms in Eurasian perch. Oikos 95:311–323
Koza V, Kořnek V (1986) Adaptability of the filtration screen in
Daphnia: another answer to the selective pressure of the
environment. Arch Hydrobiol Adv Limnol 21:193–198
Lagergren R, Hellsten M, Stenson JAE (1997) Increased drag, and
thus lower speed: a cost for morphological difference in
Bosmina (Eubosmina)(Crustacea: Cladocera). Funct Ecol
11:448–488
Lampert W (1994) Phenotypic plasticity of the filter screens in
Daphnia: adaptation to a low-food environment. Limnol
Oceanogr 39:997–1006
Lampert W, Wolf HG (1986) Cyclomorphosis in Daphnia cucul-
lata: morphometric and population genetic analyses. J Plankton
Res 8:289–303
Lunte CC, Luecke C (1990) Trophic interactions of Leptodora in
Lake Mendota. Limnol Oceanogr 35:1091–1100
Magnhagen C, Heibo E (2001) Gape size allometry in pike reflects
variation between lakes in prey availability and relative body
depth. Funct Ecol 15:754–762
283
Manca M, Comoli P (1995) Seasonal changes in size of the feeding
basket of Leptodora kindtii (Focke) in Lago Maggiore as
related to variations in prey size selection. Limnol Oceanogr
40:834–838
Manca M, Ruggiu D (1998) Consequences of pelagic food-web
changes during a long term lake oligotrophication process.
Limnol Oceanogr 43:1368–1373
Mittelbach GC, Osenberg CW, Wainwright PC (1999) Variation in
feeding morphology between pumpkinseed populations: phe-
notypic plasticity or evolution? Evol Ecol Res 1:111–128
Parejko K, Dodson SI (1991) The evolutionary ecology of an
antipredator reaction norm: Daphnia pulex and Chaoborus
americanus. Evolution 45:1665–1674
Rivier I (1999) The predatory Cladocera (Onychopoda: Podonidae,
Polyphemidae, Cercopagidae) and Leptodorida of the world.
Backhuys, Leiden
Scheiner SM, Berrigan D (1998) The genetics of phenotypic
plasticity. VIII. The cost of plasticity in Daphnia pulex.
Evolution 52:368–378
Smith LD, Palmer AR (1994) Effects of manipulated diet on size
and performance of brachyuran crab claws. Science 264:710–
712
Stibor H, Lampert W (2000) Components of additive variance in
life-history traits of Daphnia hyalina: seasonal differences in
the response to predator signals. Oikos 88:129–138
Thompson DB (1992) Consumption rates and the evolution of diet-
induced plasticity in the head morphology of Melanopus
femurrubrum (Orthoptera: Acrididae). Oecologia 89:204–213
Tollrian R (1995) Predator induced morphological defences: costs,
life history shifts, and maternal effects in Daphnia pulex.
Ecology 76:1691–1705
Tollrian R, Dodson SI (1999) Inducible defences in Cladocera:
constraints, costs, and multipredator envirinments. In: Tollri-
an R, Harvell CD (eds) The ecology and evolution of inducible
defences. Princeton University Press, Princeton, N.J. pp 177–
202
Tollrian R, Harvell CD (eds) (1999a) The ecology and evolution of
inducible defences. Princeton University Press, Princeton, N.J
Tollrian R, Harvell CD (1999b) The evolution of inducible
defenses: current ideas. In: Tollrian R, Harvell CD (eds) The
ecology and evolution of inducible defenses. Princeton Uni-
versity Press, Princeton, N.J. pp 306-321
Weider LJ (1984) Spatial heterogenity of Daphnia genotypes:
vertical migration and habitat partitioning. Limnol Oceanogr
29:225–235