Directory UMM :Data Elmu:jurnal:A:Applied Soil Ecology:Vol13.Issue3.Dec1999:
Applied Soil Ecology 13 (1999) 177±185
Toxicity and possible food-chain effects of copper, dimethoate
and a detergent (LAS) on a centipede (Lithobius mutabilis)
and its prey (Musca domestica)
Paulina Kramarz*, Ryszard Laskowski
Institute of Environmental Biology, Jagiellonian University, Ingardena 6, 30-060 KrakoÂw, Poland
Received 12 October 1998; received in revised form 23 March 1999; accepted 13 April 1999
Abstract
Toxicity of copper (Cu) and two biodegradable chemicalsÐan insecticide (dimethoate) and a detergent (LAS), was tested on a
common forest litter invertebrate carnivore Lithobius mutabilis and its prey, Musca domestica. The chemicals were mixed into
arti®cial food medium for house¯y larvae, and the centipedes were kept on contaminated OECD soil and fed contaminated
larvae. The concentrations of chemicals were selected to cover the whole range of survival of a prey animal: from no increased
mortality (control) up to 100% mortality during development from egg to adult. The house¯y larvae were signi®cantly more
affected than their predators by all three chemicals. The LC50 values for house¯y from egg to emergence were (in mg kgÿ1 dry
mass food): Cu, 708; dimethoate, 0.330; LAS, 110. In contrast, LC50 values for L. mutabilis for all three chemicals were above
the range of concentrations tested: Cu > 1500; dimethoate > 0.8; LAS > 10 000 (nominal concentrations in soil and house¯y
medium). No signi®cant sublethal effects on body weight or respiration rates were found in centipedes. This may suggest that
in contaminated areas, except for direct toxicity, a lack of food may be an important factor in reducing centipede populations.
The centipede prey, being mostly detritivores feeding directly on and living in a contaminated medium, are more exposed to a
toxicant than epigeic species such as centipedes. # 1999 Elsevier Science B.V. All rights reserved.
Keywords: Ecotoxicology; Pesticides; Copper; Soil invertebrates; Predators
1. Introduction
One of the least explored ecotoxicological phenomena is the in¯uence of toxicants on higher trophic
levels through the food chain. Although a number of
models have been proposed for this phenomenon,
none has been successfully implemented for terrestrial
ecosystems to-date. In earlier studies, it was postu*Corresponding author. Tel.: +48-12-633 63 77, ext. 447; fax:
+48-12-634 19 78.
lated that the most important process is biomagni®cation of pollutants along food chains. This would mean
that species occupying higher trophic levels are more
endangered by intoxication than those at lower trophic
levels. Such a point of view was widespread, especially in the 1960s and 1970s, and was presented in a
number of ecological handbooks (e.g. Odum, 1971;
Colinvaux, 1973; Collier et al., 1973). More recently,
this concept has been criticised by many authors (e.g.
Moriarty, 1975; Ernst and Joosse van Damme, 1983;
Beyer, 1986; Van Straalen and Van Wensem, 1986;
0929-1393/99/$ ± see front matter # 1999 Elsevier Science B.V. All rights reserved.
PII: S 0 9 2 9 - 1 3 9 3 ( 9 9 ) 0 0 0 2 3 - 2
178
P. Kramarz, R. Laskowski / Applied Soil Ecology 13 (1999) 177±185
Willamo and Nuorteva, 1987; Laskowski, 1991). It is
now mostly agreed that, depending on an animal's physiology, food preferences, trophic position and probably other as yet unrecognised factors, the intoxication
of carnivores with pollutants can either increase or
decrease with increasing position in a food chain (cf.
Dallinger et al., 1993). This leads to the conclusion
that, at least in terrestrial ecosystems, effects other
than direct intoxication may be more important.
Among them, one of the most interesting appears to
be changes in community structure (e.g. Poulton et al.,
1995; Korthals et al., 1996; Matthews et al., 1996; Abd
Allah et al., 1997). In the case of predatory species,
pollutants may affect food availability by reducing
population size of prey animals. This situation seems
especially likely in the soil/litter system, where many
prey animals are more intensely exposed to toxic
substances via direct contact with soil and the soil
solution than surface-living predators. Indeed, twospecies, prey±predator experiments on effects of toxicants on soil invertebrates have recently been advocated in a number of articles as re¯ecting the real-®eld
situation better and, thus, being more appropriate for
soil toxicity risk assessment (Kula et al., 1995).
In this experiment, toxicities of copper (Cu),
dimethoate and linear alkylbenzenesulphonate
(LAS) were tested on the common forest-litter invertebrate carnivore Lithobius mutabilis Koch. (Myriapoda) and its prey, Musca domestica L. (Diptera). The
chemicals chosen represent a broad spectrum of
anthropogenic pollutants, from biodegradable pesticide (dimethoate) and surfactant (LAS) to non-degradable metal (Cu). Centipedes are among the most
important group of terrestrial invertebrate predators.
According to Albert (1983), just two species, L.
mutabilis and L. curtipes, accounted for ca. 17±27%
of the total assimilation by predatory macroarthropods
in the forest ecosystems studied. Thus, effects of
toxicants on this group can be of particular importance
for the functioning of some ecosystems.
2. Methods
The centipedes (Lithobius mutabilis) were obtained
from a beech±pine forest of the Ratanica catchment,
located 40 km south of KrakoÂw, Poland. Samples of
the whole organic layer were sieved through a 1-cm
mesh sieve and the centipedes were collected from the
sieved material manually after transportation to the
laboratory. Before being used in the experiment, the
centipedes were kept for two weeks in 30 50
20 cm3 transparent plastic containers, 100±200 individuals in each, under constant laboratory conditions
at a temperature of 15 0.58C, relative humidity
80%, and a light : dark regime of 16 : 8 h. The
bottom of each container was covered with a layer
of a moist natural soil, 2±3 cm thick, with a 2±3 cm
layer of leaf litter on the surface. The animals were fed
on frozen house¯y pupae (Musca domestica), taken
from large cultures which were maintained in the
laboratory.
For toxicity tests on centipedes, adults of 15±45 mg
fresh mass (10±15 mm length) were selected from the
culture. Acute toxicity tests (mortality) were run on
groups consisting of three males and three females
kept in 16 11 6 cm3 plastic boxes, and sublethal
effects (respiration rate and body weight change) were
studied on animals kept separately in 11 7.5
4.5 cm3 boxes. All boxes were ®lled with 1 cm deep
layers of contaminated or uncontaminated (control)
standard arti®cial OECD soil (Lùkke and Van Gestel,
1998) at 50% water-holding capacity, and covered
with loose transparent plastic lids. The centipedes
were fed one house¯y pupa per individual, twice a
week. Six replicates per treatment were used in all
experiments except for `dimethoate-contaminated
food and soil', where only four replicates were used
due to an additional test which was run with only
contaminated soil (see below for details).
The medium for house¯y larvae was made from
515.6 g rabbit chaw, 20 g powder milk, 10 g sugar,
0.02 g baking yeast and 1 l of distilled water or
experimental solution. After mixing the ingredients,
the medium was left for 2 h at room temperature
before use in the experiment. The larvae were kept
in 120-ml plastic containers with 77 g of medium (wet
weight), covered with a ®ltering tissue. The boxes with
the larvae were kept under the same environmental
conditions as the centipedes.
In the tests on house¯ies, six replicates per treatment were used, and each replicate consisted of 40
freshly emerged house¯y larvae. Five days after transferring larvae to the experimental medium, a layer of
paper tissue was put on the surface of the medium to
provide larvae with a place suitable for pupation. This
179
P. Kramarz, R. Laskowski / Applied Soil Ecology 13 (1999) 177±185
®ve-day period was chosen based on earlier observations on the time required to start pupation in an
uncontaminated medium. When the pupae started to
turn dark-brown and no more new pupae were
observed, all the pupae were removed from the paper
tissue, and the remaining medium was searched for
specimens that had not migrated to the surface and
pupated inside the medium. All pupae were then
transferred to clean plastic boxes and covered with
a gauze that enabled observation of emerging adults.
After all imagines emerged, they were killed by
freezing and counted.
Copper was used as CuCl2 (PPH POCh, Poland),
dimethoate as a formulation of 400 g a.i. lÿ1 (Cheminova Agro A/S, Denmark) and LAS was the Marlon
A350 sodium alkylbenzenesulphonate with 70±90%
mass as C11 and C12 alkyl chains (HuÈls Danmark A/S).
The chemicals were mixed into the arti®cial food
medium for house¯y larvae at the same nominal
concentrations as into the OECD soil used in the
centipede tests. Soil was left to equilibrate for 24 h
at ca. 58C prior to use in the experiments. The concentrations of chemicals were adjusted so that they
covered the whole range of survival of a prey animal:
from no increased mortality (control), to almost 100%
mortality during development from egg to adult. In the
`soil food' experiments, centipedes were kept on
contaminated soil and fed house¯y pupae originating
from a medium which had been treated with similar
nominal concentrations of the test chemicals. Because
preliminary studies showed no increased mortality of
centipedes in dimethoate treatments at the range of
concentrations which house¯y larvae could survive,
an additional test was performed with only contaminated soil (`dimethoate-soil'), but at higher rates than
could be used in a house¯y test. Insuf®cient numbers
of centipedes forced us to reduce the number of
replicates in one of the tests, so that only four replicates were used in the `dimethoate-food soil' treatment.
The following nominal treatments were used in
`soil food' experiments (calculated per dry weight
soil or food medium):
Cu
0; 56; 167; 500; 1000; 1500 mg kgÿ1
Dimethoate 0; 0.012; 0.037; 0.111; 0.333; 0.800 mg
a.i. kgÿ1
LAS
0; 16; 80; 400; 2000; 10 000 mg kgÿ1.
The nominal concentrations of dimethoate in soil used
in the `dimethoate-soil' experiment were: 0; 1.65;
3.10; 6.25; 12.5 and 25.0 mg kgÿ1 (active ingredient
per dry weight basis).
All experiments, with the exception of Cu, were
terminated after 30-day exposure. In the case of Cu,
we also measured effects after 85-days exposure to
account for possible long-term effects caused by
accumulation of non-degradable Cu in the bodies of
centipedes fed the Cu-enriched diet.
The experimental cultures of centipedes were
checked daily for mortality, and the respiration rates
and body weights of individually kept specimens were
recorded weekly. The respiration rate of each individual was measured in a water bath at 158C for 1 h
using constant-pressure volumetric microrespirometers (Klekowski, 1975). The animals were weighed
to the nearest 0.0005 g (Precisa, Switzerland).
The LC50 and EC50 values for the response variables, y, were estimated using a `half-maximal
response' non-linear regression model (CSS-Statistica) as follows:
y b0 ÿ
b0
1 x=EC50 b1
;
where b0 denotes the expected response at the level of
dose saturation, b1 determines the slope of the function, and x the concentration of the chemical. After
setting b0 at 100 (100% response), there were two free
parameters estimated in the regression equation: LC50
and b1. The goodness of ®t was determined with the
least-squares method. If a hormetic effect was seen,
the following model was used:
y
b0 hx
;
1 b0 =2 hEC50 = b0 =2ECb501 xb1
where the parameter h allows for a stimulatory effect
at low concentrations (hormesis), b0 is the response in
control culture (here set to 100%), and b1 the slope
(Lùkke, 1995). The calculated regressions were then
plotted against nominal concentrations of chemicals in
house¯y medium or soil.
3. Results
The LC50 values for house¯y larvae for the period
from egg to pupation were (in mg kgÿ1 dry weight
180
P. Kramarz, R. Laskowski / Applied Soil Ecology 13 (1999) 177±185
Fig. 1. Effect of dimethoate (nominal treatments) on survival of housefly from eggs to imagines (a) and centipedes after 30 days of exposure
to low-level dimethoate in soil and food (b) or high-level dimethoate in soil only (c); body weight change (d) and respiration rates (e) of
centipedes exposed to low-level dimethoate contamination of soil and food (averages SEM and regression lines); 95% C.I. represents 95%
confidence interval for LC50.
food): Cu, 689; dimethoate, 0.326; LAS, 155. With
respect to the number of adults emerged, the LC50s
were as follows: Cu, 708; dimethoate, 0.28; LAS, 110
(Figs. 1±3). In contrast, LC50s of L. mutabilis were
higher than the highest concentrations tested for
all three chemicals: Cu, LC50 > 1500; dimethoate,
LC50 > 0.8; LAS, LC50 > 10 000 (Figs. 1±3). No
relationship between concentration and mortality
was observed in the `food soil' experiment for
dimethoate, while the estimated LC50 for centipedes
was as high as 13.8 mg kgÿ1 in soil contaminated
with high dimethoate concentrations (Fig. 1). Thus,
based on nominal concentrations, the house¯y larvae
were affected more than the predator by all three
chemicals.
The chemicals tested differed markedly in the way
they affected the survival of ¯y larvae and centipedes.
In house¯y larvae, the steepest response curve was
found for dimethoate, where no increased mortality
was observed up to 0.111 mg kgÿ1, yet at 0.8 mg kgÿ1
P. Kramarz, R. Laskowski / Applied Soil Ecology 13 (1999) 177±185
181
Fig. 2. Effect of copper (Cu; nominal treatments) on survival of housefly from eggs to imagines (a) and centipedes after 30 (b) and 85 (c) days
of exposure to Cu-contaminated soil and food; body weight change (d) and respiration rates (e) of centipedes exposed to Cu-contaminated soil
and food (averages SEM and regression lines); 95% C.I. represents 95% confidence interval for LC50.
no single larva survived to pupation (Fig. 1(a)). A
relatively sharp decline in survivorship was also
observed in the Cu-contaminated medium with no
increase in mortality up to 167 mg kgÿ1, and only
25% surviving at 1500 mg kgÿ1 (Fig. 2(a)). The most
shallow response curve was in the relationship of
larval survival with LAS concentration: survival
decreased gradually from 75% of the control at
16 mg kgÿ1, to >16% of the larvae entering the pupal
stage and 5% surviving to imagines at a concentration as high as 10 000 mg kgÿ1 (Fig. 3(a)). Thus, only
about a sevenfold increase in dimethoate concentra-
tion was suf®cient for house¯y survival to drop from
100% to 0, while for Cu a ninefold concentration
increase caused only a 75% increase in mortality,
and for LAS the house¯ies were able to survive over
at least a 10 000-fold range of concentrations.
The relationships between centipede survival and
chemical concentration in the soil and house¯y medium differed from those found for the house¯y. For
dimethoate, over the range of concentrations at which
house¯y larvae survived, there was no signi®cant
relationship between centipede mortality and concentration (Fig. 1(b)). This was true up to 5 mg a.i. kgÿ1
182
P. Kramarz, R. Laskowski / Applied Soil Ecology 13 (1999) 177±185
Fig. 3. Effect of linear alkylbenzenesulphonate (LAS; nominal treatments) on survival of housefly from eggs to imagines (a) and centipedes
after 30 days of exposure to LAS in soil and food (b); body weight change (c) and respiration rates (d) of centipedes exposed to LAScontaminated soil and food (averages SEM and regression lines); 95% C.I. represents 95% confidence interval for LC50.
soil, when mortality rates of the centipedes increased.
However, above this concentration, the curve was very
steep, so that at 25 mg a.i. kgÿ1 no single centipede
survived the duration of the experiment (Fig. 1(c)).
After 30 days, Cu did not affect centipedes signi®cantly at concentrations up to 1000 mg kgÿ1, but
above this concentration a very abrupt increase in
mortality was noted (Fig. 2(b)). After 85 days of
exposure to Cu-contaminated food and soil, the centipedes exposed to 56±1000 mg kgÿ1 survived considerably better than control animals, while the abrupt
increase in mortality above 1000 mg kgÿ1 was still
visible (Fig. 2(c)). For LAS, the relationship between
concentration and centipede survival was unclear and,
on average, only a slight decrease in survival was
observed. However, at a nominal concentration of
10 000 mg kgÿ1, survival of centipedes was still as
high as 85% of the control (Fig. 3(b)).
At the termination of the experiment, no signi®cant
sublethal effects (body weight change or respiration
rates) were found in centipedes (Figs. 1±3).
4. Discussion
Our studies indicate that L. mutabilis is relatively
resistant to all chemicals tested. All three chemicals
caused much higher mortality in cultures of its prey at
the respective concentrations. This suggests that in
contaminated areas lack of food rather than direct
poisoning may lead to disturbances in centipede populations. Also, predation pressure on resistant soil
invertebrates can drastically increase when other
invertebrates are affected by toxicants. This effect
would be expected especially for those chemicals with
the largest difference in LC50 between a predator and
its prey.
Among the chemicals tested in our studies, the
direct toxicity of LAS to centipedes appears negligible; hence, in this case exclusively secondary foodchain effects are to be expected. This result seems
reasonable, as centipede prey such as ¯y larvae, being
mostly detritivores feeding directly on and living in a
contaminated medium, are more exposed to a toxicant
P. Kramarz, R. Laskowski / Applied Soil Ecology 13 (1999) 177±185
than epigeic species such as centipedes. Biodegradable surfactants such as LAS exert toxic action mainly
through direct contact in water solution. Thus, soildwelling invertebrates are exposed to direct toxicity of
surfactants via the soil solution, while for surface
living forms there may be practically no effect.
There was also a substantial difference between
centipedes and house¯y larvae in their LC50s for
dimethoate. As dimethoate is highly toxic, both in
aqueous solution and as aerosol or vapour, the large
difference in toxicity was probably not caused solely
by the different life styles of house¯y larvae and
centipedes but also through dimethoate selective toxicity against the insects. It seems, thus, that although
dimethoate is considered a broad-spectrum insecticide
(Duf®eld and Aebischer, 1994; Duf®eld et al., 1996;
Huusela-Veistola, 1996), it is relatively safe for centipedes. As a consequence, secondary food-chain
effects may be more important with dimethoate as
in the case of LAS. On the other hand, dimethoate has
been found to be highly toxic to both, Coccinella and
Chrysopa (Dimetry and Marei, 1992). Thus, in the
case of dimethoate, the relative importance of direct
toxicity and secondary community-level effects to
carnivores probably depends not only on differences
in life styles (endo- or epigeic) but also on the relative
taxonomic positions of a predator and its prey: if a
pesticide is speci®c enough to affect the prey population almost exclusively, then mostly secondary effects
on predators can be expected. Consequently, it may be
anticipated that the larger the taxonomic distance
between the predator and its prey, the more important
the secondary effects will be.
For Cu, the LC50 for centipedes was also above the
range of concentrations tested and was estimated at ca.
1700 mg kgÿ1, while for house¯ies it was approximately one-third that value. This difference was,
however, substantially smaller than in the case of
dimethoate and LAS and suggests that, in this case,
direct toxicity may be important.
Ecotoxicological studies on the effects of toxicants
on terrestrial carnivorous invertebrates have followed
two directions. On the one hand, a number of studies
have been performed on concentrations of toxicants in
animals belonging to different trophic levels in the
hope of ®nding some `general rule', such as biomagni®cation, which would indicate which organisms are
the most endangered in contaminated ecosystems. On
183
the other hand, probably even more research has been
carried out on the side-effects of pesticides, mostly on
carnivorous beetles and spiders, but also on other
`bene®cial' arthropods. While the ®rst approach
apparently failed (cf. Beyer, 1986; Laskowski,
1991), the latter studies concentrated mostly on direct
toxic effects (e.g. Buchs et al., 1989; C,ilgi et al., 1996).
Although the biomagni®cation hypothesis did not
appear to be universal, other `food-chain' effects seem
possible. Among them, the most important from a
purely ecological point of view, are changes in interactions between populations of different species
exposed to toxicants (community-level effects) (cf.
Cohn and MacPhail, 1996). In many cases, the most
signi®cant among these interactions are predator±prey
relationships, because the abundance of prey ultimately determines the survival and fecundity of the
predators. For example, Purvis and Bannon (1992)
found that application of methiocarb-based slug pellets caused different effects in carabid populations,
depending on the season, and attributed these differences to purely indirect effects, such as prey availability and foraging behaviour. Possible interactions
between metal contamination and community-level
biotic factors were studied recently by Kiffney
(1996). This author found that metals can signi®cantly
in¯uence the relative importance of biotic factors in
altering freshwater macroinvertebrate community
structure: in streams dosed with metals, predation
intensity was greater than in controls. On the other
hand, Parmelee et al. (1993) found that soil-dwelling
predatory nematodes were more susceptible to Cu
contamination than their prey (also nematodes), which
resulted in reduced predation on herbivorous nematodes and, in consequence, greater numbers of nematodes compared to controls. Such results indicate how
important these interactions are and how inaccurate a
risk assessment may be if based solely on singlespecies, direct-toxicity tests.
The results obtained in this study, as well as those
reported by Parmelee et al. (1993) and Kiffney (1996),
add another dimension to laboratory-based risk assessment procedures. They suggest that, in some situations, there may be secondary community-level
effects, understood here as interactions between toxic
substances and biotic factors, rather than direct
poisoning, which can determine the ®nal impact of
a toxicant on a number of species in a polluted
184
P. Kramarz, R. Laskowski / Applied Soil Ecology 13 (1999) 177±185
ecosystem. The species particularly susceptible to
secondary trophic-level effects would probably be
the epigeic predators feeding on endogeic fauna. This
leads to the conclusion that selectivity of a particular
toxicant against particular species or a group of species, should only be the ®rst step of risk assessment
studies. A similar conclusion was also reached by
Duf®eld et al. (1996), who studied recolonisation
patterns of predatory invertebrates in ®elds sprayed
with insecticides.
Acknowledgements
We are grateful to all members of the SECOFASE
project for their co-operation and fruitful discussions
during project workshops. Special thanks are due to
Dr. Hans Lùkke, the Project Co-ordinator, for all his
help. We also wish to thank to Dr. Philip Heneghan for
a critical reading of the manuscript. The funding for
this work was provided by the EU Programme Environment 1990±1994 (the PECO Programme Contract
No. ERB-CIPD-CT93-0059).
References
Abd Allah, A.T., Wanas, M.Q.S., Thompson, S.N., 1997. Effects of
heavy metals on survival and growth of Biomphalaria glabrata
Say (Gastropoda: Pulmonata) and interaction with schistosome
infection. J. Mollus. Stud. 63, 79±86.
Albert, A.M., 1983. Energy budgets for populations of long-lived
arthropod predators (Chilopoda: Lithobiidae) in an old beech
forest. Oecologia 56, 292±305.
Beyer, W.N., 1986. A reexamination of biomagnification of metals
in terrestrial food chains. Environ. Toxicol. Chem. 5, 863±864.
Buchs, W., Heimbach, U., Czarnecki, E.A.D., 1989. Effects of snail
baits on non-target carabid beetles. Monographs of British Crop
Protection Council, vol. 41. pp. 245±252.
C,ilgi, T., Wratten, S.D., Robertson, J.L., Turner, D.E., Holland,
J.M., Frampton, G.K., 1996. Residual toxicities of three
insecticides to four species (Coleoptera: Carabidae) of
arthropod predators. Can. Entomol. 128, 1115±1124.
Cohn, J., MacPhail, R.C., 1996. Ethological and experimental
approaches to behaviour analysis: implications for ecotoxicology. Environ. Health Perspect. 104, 299±305.
Colinvaux, P., 1973. Introduction to Ecology. Wiley, New York.
Collier, B.D., Cox, G.W., Johnson, A.W., Miller, P.C., 1973.
Dynamic Ecology. Prentice±Hall, Engelwood Cliffs, NJ.
Dallinger, R., Berger, B., Gruber, A., 1993. Quantitative aspects of
zinc and cadmium binding in Helix pomatia: differences
between an essential and a nonessential trace element. In:
Dallinger, R., Rainbow, P. (Ed.), Ecotoxicology of Metals in
Invertebrates. Lewis Publishers, Boca Raton, Ann Arbor,
London, Tokyo. pp. 315±332.
Dimetry, N.Z., Marei, S.S., 1992. Laboratory evaluation of some
pesticides on the cabbage aphid, Brevicoryne brassicae, and
their side-effects on some important natural enemies. Anzeiger
fuÈr Schadlingskunde, Pflanzenschutz und Umweltschutz 65,
16±19.
Duffield, S.J., Aebischer, N.J., 1994. The effect of spatial scale of
treatment with dimethoate on invertebrate population recovery
in winter-wheat. J. Appl. Ecol. 31, 263±281.
Duffield, S.J., Jepson, P.C., Wratten, S.D., Sotherton, N.W., 1996.
Spatial changes in invertebrate predation rate in winter wheat
following treatment with dimethoate. Entomologia Experimentalis et Applicata 78, 9±17.
Ernst, W.H.O., Joosse van Damme, E.N.G., 1983. Umweltbelastung durch Mineralstoffe. Biologische Effekte. VEB Gustav
Fischer Verlag, Jena.
Huusela-Veistola, E., 1996. Effects of pesticide use and cultivation
techniques on ground beetles (Coleoptera, Carabidae) in cereal
fields. Ann. Zool. Fenn. 33, 197±205.
Kiffney, P.M., 1996. Main and interactive effects of invertebrate
density, predation, and metals on a Rocky Mountain stream
macroinvertebrate community. Can. J. Fisheries. Aquat. Sci. 53,
1595±1601.
Klekowski, R., 1975. Constant-pressure volumetric microrespirometer for terrestrial invertebrates. In: GrodzinÄski, W.,
Klekowski, R., Duncan, A. (Eds.), Methods for ecological
bioenergetics. IBP Handbook No. 24., Blackwell Scientific
Publications, Oxford, London, Edinburgh, Melbourne. pp. 212±
225.
Korthals, G.W., de Goede, R.G.M., Kammenga, J.E., Bongers, T.,
1996. The maturity index as an instrument for risk assessment
of soil pollution. In: Van Straalen, N.M., Krivolutsky, D.A.
(Ed.), Bioindicator Systems for Soil Pollution. Kluwer Academic Publishers, Dordrecht. pp. 85±93.
Kula, H., Heimbach, U., Lùkke, H. (Ed.), 1995. Progress Report
1994 of SECOFASE, Third Technical Report. Development,
improvement and standardization of test systems for assessing
sublethal effects of chemicals on fauna in the soil ecosystem.
National Environmental Research Institute, Denmark. pp. 166.
Laskowski, R., 1991. Are the top carnivores endangered by heavy
metal biomagnification? Oikos 60 387±390.
Lùkke, H. (Ed.), 1995. Effects of pesticides on meso- and
microfauna in soil. Ministry of Environment and Energy,
Denmark.
Lùkke, H., Van Gestel, C.A.M. (Eds.), 1998. Handbook of soil
invertebrate toxicity tests. John Wiley & Sons, Chichester, UK.
Matthews, R.A., Landis, W.G., Matthews, G.B., 1996. The
community conditioning hypothesis and its application to
environmental toxicology. Environ. Toxicol. Chem. 15, 597±
603.
Moriarty, F., 1975. Pollutants and animals: a factual perspective.
George Allen and Unwin Ltd., London.
Odum, E.P., 1971. Fundamentals of Ecology. Saunders, Philadelphia.
P. Kramarz, R. Laskowski / Applied Soil Ecology 13 (1999) 177±185
Parmelee, R.W., Wentsel, R.S., Phillips, C.T., Simini, M., Checkai,
R.T., 1993. Soil microcosm for testing the effects of chemicalpollutants on soil fauna communities and trophic structure.
Environ. Toxicol. Chem. 12, 1477±1486.
Poulton, B.C., Monda, D.P., Woodward, D.F., Wildhaber, M.L.,
Brumbaugh, W.G., 1995. Relations between benthic community
structure and metals concentrations in aquatic macroinvertebrates: Clark Fork, Montana. J. Freshwater Ecol. 10, 277±293.
Purvis, G., Bannon, J.W., 1992. Nontarget effects of repeated
methiocarb slug pellet application on carabid beetle (Coleop-
185
tera, Carabidae) activity in winter-sown cereals. Ann. Appl.
Biol. 121, 401±422.
Van Straalen, N.M., Van Wensem, J., 1986. Heavy metal content of
forest litter arthropods as related to body-size and trophic level.
Environ. Pollut. (A) 42, 209±221.
Willamo, R., Nuorteva, P., 1987. The role of heavy metals in forest
die-off. In: Anttila, P., Kauppi, P. (Ed.), Symposium of the
Finnish Research Project on Acidification (HAPRO). Ministry
of the Environment, Ministry of Agriculture and Forestry.
pp. 64±67.
Toxicity and possible food-chain effects of copper, dimethoate
and a detergent (LAS) on a centipede (Lithobius mutabilis)
and its prey (Musca domestica)
Paulina Kramarz*, Ryszard Laskowski
Institute of Environmental Biology, Jagiellonian University, Ingardena 6, 30-060 KrakoÂw, Poland
Received 12 October 1998; received in revised form 23 March 1999; accepted 13 April 1999
Abstract
Toxicity of copper (Cu) and two biodegradable chemicalsÐan insecticide (dimethoate) and a detergent (LAS), was tested on a
common forest litter invertebrate carnivore Lithobius mutabilis and its prey, Musca domestica. The chemicals were mixed into
arti®cial food medium for house¯y larvae, and the centipedes were kept on contaminated OECD soil and fed contaminated
larvae. The concentrations of chemicals were selected to cover the whole range of survival of a prey animal: from no increased
mortality (control) up to 100% mortality during development from egg to adult. The house¯y larvae were signi®cantly more
affected than their predators by all three chemicals. The LC50 values for house¯y from egg to emergence were (in mg kgÿ1 dry
mass food): Cu, 708; dimethoate, 0.330; LAS, 110. In contrast, LC50 values for L. mutabilis for all three chemicals were above
the range of concentrations tested: Cu > 1500; dimethoate > 0.8; LAS > 10 000 (nominal concentrations in soil and house¯y
medium). No signi®cant sublethal effects on body weight or respiration rates were found in centipedes. This may suggest that
in contaminated areas, except for direct toxicity, a lack of food may be an important factor in reducing centipede populations.
The centipede prey, being mostly detritivores feeding directly on and living in a contaminated medium, are more exposed to a
toxicant than epigeic species such as centipedes. # 1999 Elsevier Science B.V. All rights reserved.
Keywords: Ecotoxicology; Pesticides; Copper; Soil invertebrates; Predators
1. Introduction
One of the least explored ecotoxicological phenomena is the in¯uence of toxicants on higher trophic
levels through the food chain. Although a number of
models have been proposed for this phenomenon,
none has been successfully implemented for terrestrial
ecosystems to-date. In earlier studies, it was postu*Corresponding author. Tel.: +48-12-633 63 77, ext. 447; fax:
+48-12-634 19 78.
lated that the most important process is biomagni®cation of pollutants along food chains. This would mean
that species occupying higher trophic levels are more
endangered by intoxication than those at lower trophic
levels. Such a point of view was widespread, especially in the 1960s and 1970s, and was presented in a
number of ecological handbooks (e.g. Odum, 1971;
Colinvaux, 1973; Collier et al., 1973). More recently,
this concept has been criticised by many authors (e.g.
Moriarty, 1975; Ernst and Joosse van Damme, 1983;
Beyer, 1986; Van Straalen and Van Wensem, 1986;
0929-1393/99/$ ± see front matter # 1999 Elsevier Science B.V. All rights reserved.
PII: S 0 9 2 9 - 1 3 9 3 ( 9 9 ) 0 0 0 2 3 - 2
178
P. Kramarz, R. Laskowski / Applied Soil Ecology 13 (1999) 177±185
Willamo and Nuorteva, 1987; Laskowski, 1991). It is
now mostly agreed that, depending on an animal's physiology, food preferences, trophic position and probably other as yet unrecognised factors, the intoxication
of carnivores with pollutants can either increase or
decrease with increasing position in a food chain (cf.
Dallinger et al., 1993). This leads to the conclusion
that, at least in terrestrial ecosystems, effects other
than direct intoxication may be more important.
Among them, one of the most interesting appears to
be changes in community structure (e.g. Poulton et al.,
1995; Korthals et al., 1996; Matthews et al., 1996; Abd
Allah et al., 1997). In the case of predatory species,
pollutants may affect food availability by reducing
population size of prey animals. This situation seems
especially likely in the soil/litter system, where many
prey animals are more intensely exposed to toxic
substances via direct contact with soil and the soil
solution than surface-living predators. Indeed, twospecies, prey±predator experiments on effects of toxicants on soil invertebrates have recently been advocated in a number of articles as re¯ecting the real-®eld
situation better and, thus, being more appropriate for
soil toxicity risk assessment (Kula et al., 1995).
In this experiment, toxicities of copper (Cu),
dimethoate and linear alkylbenzenesulphonate
(LAS) were tested on the common forest-litter invertebrate carnivore Lithobius mutabilis Koch. (Myriapoda) and its prey, Musca domestica L. (Diptera). The
chemicals chosen represent a broad spectrum of
anthropogenic pollutants, from biodegradable pesticide (dimethoate) and surfactant (LAS) to non-degradable metal (Cu). Centipedes are among the most
important group of terrestrial invertebrate predators.
According to Albert (1983), just two species, L.
mutabilis and L. curtipes, accounted for ca. 17±27%
of the total assimilation by predatory macroarthropods
in the forest ecosystems studied. Thus, effects of
toxicants on this group can be of particular importance
for the functioning of some ecosystems.
2. Methods
The centipedes (Lithobius mutabilis) were obtained
from a beech±pine forest of the Ratanica catchment,
located 40 km south of KrakoÂw, Poland. Samples of
the whole organic layer were sieved through a 1-cm
mesh sieve and the centipedes were collected from the
sieved material manually after transportation to the
laboratory. Before being used in the experiment, the
centipedes were kept for two weeks in 30 50
20 cm3 transparent plastic containers, 100±200 individuals in each, under constant laboratory conditions
at a temperature of 15 0.58C, relative humidity
80%, and a light : dark regime of 16 : 8 h. The
bottom of each container was covered with a layer
of a moist natural soil, 2±3 cm thick, with a 2±3 cm
layer of leaf litter on the surface. The animals were fed
on frozen house¯y pupae (Musca domestica), taken
from large cultures which were maintained in the
laboratory.
For toxicity tests on centipedes, adults of 15±45 mg
fresh mass (10±15 mm length) were selected from the
culture. Acute toxicity tests (mortality) were run on
groups consisting of three males and three females
kept in 16 11 6 cm3 plastic boxes, and sublethal
effects (respiration rate and body weight change) were
studied on animals kept separately in 11 7.5
4.5 cm3 boxes. All boxes were ®lled with 1 cm deep
layers of contaminated or uncontaminated (control)
standard arti®cial OECD soil (Lùkke and Van Gestel,
1998) at 50% water-holding capacity, and covered
with loose transparent plastic lids. The centipedes
were fed one house¯y pupa per individual, twice a
week. Six replicates per treatment were used in all
experiments except for `dimethoate-contaminated
food and soil', where only four replicates were used
due to an additional test which was run with only
contaminated soil (see below for details).
The medium for house¯y larvae was made from
515.6 g rabbit chaw, 20 g powder milk, 10 g sugar,
0.02 g baking yeast and 1 l of distilled water or
experimental solution. After mixing the ingredients,
the medium was left for 2 h at room temperature
before use in the experiment. The larvae were kept
in 120-ml plastic containers with 77 g of medium (wet
weight), covered with a ®ltering tissue. The boxes with
the larvae were kept under the same environmental
conditions as the centipedes.
In the tests on house¯ies, six replicates per treatment were used, and each replicate consisted of 40
freshly emerged house¯y larvae. Five days after transferring larvae to the experimental medium, a layer of
paper tissue was put on the surface of the medium to
provide larvae with a place suitable for pupation. This
179
P. Kramarz, R. Laskowski / Applied Soil Ecology 13 (1999) 177±185
®ve-day period was chosen based on earlier observations on the time required to start pupation in an
uncontaminated medium. When the pupae started to
turn dark-brown and no more new pupae were
observed, all the pupae were removed from the paper
tissue, and the remaining medium was searched for
specimens that had not migrated to the surface and
pupated inside the medium. All pupae were then
transferred to clean plastic boxes and covered with
a gauze that enabled observation of emerging adults.
After all imagines emerged, they were killed by
freezing and counted.
Copper was used as CuCl2 (PPH POCh, Poland),
dimethoate as a formulation of 400 g a.i. lÿ1 (Cheminova Agro A/S, Denmark) and LAS was the Marlon
A350 sodium alkylbenzenesulphonate with 70±90%
mass as C11 and C12 alkyl chains (HuÈls Danmark A/S).
The chemicals were mixed into the arti®cial food
medium for house¯y larvae at the same nominal
concentrations as into the OECD soil used in the
centipede tests. Soil was left to equilibrate for 24 h
at ca. 58C prior to use in the experiments. The concentrations of chemicals were adjusted so that they
covered the whole range of survival of a prey animal:
from no increased mortality (control), to almost 100%
mortality during development from egg to adult. In the
`soil food' experiments, centipedes were kept on
contaminated soil and fed house¯y pupae originating
from a medium which had been treated with similar
nominal concentrations of the test chemicals. Because
preliminary studies showed no increased mortality of
centipedes in dimethoate treatments at the range of
concentrations which house¯y larvae could survive,
an additional test was performed with only contaminated soil (`dimethoate-soil'), but at higher rates than
could be used in a house¯y test. Insuf®cient numbers
of centipedes forced us to reduce the number of
replicates in one of the tests, so that only four replicates were used in the `dimethoate-food soil' treatment.
The following nominal treatments were used in
`soil food' experiments (calculated per dry weight
soil or food medium):
Cu
0; 56; 167; 500; 1000; 1500 mg kgÿ1
Dimethoate 0; 0.012; 0.037; 0.111; 0.333; 0.800 mg
a.i. kgÿ1
LAS
0; 16; 80; 400; 2000; 10 000 mg kgÿ1.
The nominal concentrations of dimethoate in soil used
in the `dimethoate-soil' experiment were: 0; 1.65;
3.10; 6.25; 12.5 and 25.0 mg kgÿ1 (active ingredient
per dry weight basis).
All experiments, with the exception of Cu, were
terminated after 30-day exposure. In the case of Cu,
we also measured effects after 85-days exposure to
account for possible long-term effects caused by
accumulation of non-degradable Cu in the bodies of
centipedes fed the Cu-enriched diet.
The experimental cultures of centipedes were
checked daily for mortality, and the respiration rates
and body weights of individually kept specimens were
recorded weekly. The respiration rate of each individual was measured in a water bath at 158C for 1 h
using constant-pressure volumetric microrespirometers (Klekowski, 1975). The animals were weighed
to the nearest 0.0005 g (Precisa, Switzerland).
The LC50 and EC50 values for the response variables, y, were estimated using a `half-maximal
response' non-linear regression model (CSS-Statistica) as follows:
y b0 ÿ
b0
1 x=EC50 b1
;
where b0 denotes the expected response at the level of
dose saturation, b1 determines the slope of the function, and x the concentration of the chemical. After
setting b0 at 100 (100% response), there were two free
parameters estimated in the regression equation: LC50
and b1. The goodness of ®t was determined with the
least-squares method. If a hormetic effect was seen,
the following model was used:
y
b0 hx
;
1 b0 =2 hEC50 = b0 =2ECb501 xb1
where the parameter h allows for a stimulatory effect
at low concentrations (hormesis), b0 is the response in
control culture (here set to 100%), and b1 the slope
(Lùkke, 1995). The calculated regressions were then
plotted against nominal concentrations of chemicals in
house¯y medium or soil.
3. Results
The LC50 values for house¯y larvae for the period
from egg to pupation were (in mg kgÿ1 dry weight
180
P. Kramarz, R. Laskowski / Applied Soil Ecology 13 (1999) 177±185
Fig. 1. Effect of dimethoate (nominal treatments) on survival of housefly from eggs to imagines (a) and centipedes after 30 days of exposure
to low-level dimethoate in soil and food (b) or high-level dimethoate in soil only (c); body weight change (d) and respiration rates (e) of
centipedes exposed to low-level dimethoate contamination of soil and food (averages SEM and regression lines); 95% C.I. represents 95%
confidence interval for LC50.
food): Cu, 689; dimethoate, 0.326; LAS, 155. With
respect to the number of adults emerged, the LC50s
were as follows: Cu, 708; dimethoate, 0.28; LAS, 110
(Figs. 1±3). In contrast, LC50s of L. mutabilis were
higher than the highest concentrations tested for
all three chemicals: Cu, LC50 > 1500; dimethoate,
LC50 > 0.8; LAS, LC50 > 10 000 (Figs. 1±3). No
relationship between concentration and mortality
was observed in the `food soil' experiment for
dimethoate, while the estimated LC50 for centipedes
was as high as 13.8 mg kgÿ1 in soil contaminated
with high dimethoate concentrations (Fig. 1). Thus,
based on nominal concentrations, the house¯y larvae
were affected more than the predator by all three
chemicals.
The chemicals tested differed markedly in the way
they affected the survival of ¯y larvae and centipedes.
In house¯y larvae, the steepest response curve was
found for dimethoate, where no increased mortality
was observed up to 0.111 mg kgÿ1, yet at 0.8 mg kgÿ1
P. Kramarz, R. Laskowski / Applied Soil Ecology 13 (1999) 177±185
181
Fig. 2. Effect of copper (Cu; nominal treatments) on survival of housefly from eggs to imagines (a) and centipedes after 30 (b) and 85 (c) days
of exposure to Cu-contaminated soil and food; body weight change (d) and respiration rates (e) of centipedes exposed to Cu-contaminated soil
and food (averages SEM and regression lines); 95% C.I. represents 95% confidence interval for LC50.
no single larva survived to pupation (Fig. 1(a)). A
relatively sharp decline in survivorship was also
observed in the Cu-contaminated medium with no
increase in mortality up to 167 mg kgÿ1, and only
25% surviving at 1500 mg kgÿ1 (Fig. 2(a)). The most
shallow response curve was in the relationship of
larval survival with LAS concentration: survival
decreased gradually from 75% of the control at
16 mg kgÿ1, to >16% of the larvae entering the pupal
stage and 5% surviving to imagines at a concentration as high as 10 000 mg kgÿ1 (Fig. 3(a)). Thus, only
about a sevenfold increase in dimethoate concentra-
tion was suf®cient for house¯y survival to drop from
100% to 0, while for Cu a ninefold concentration
increase caused only a 75% increase in mortality,
and for LAS the house¯ies were able to survive over
at least a 10 000-fold range of concentrations.
The relationships between centipede survival and
chemical concentration in the soil and house¯y medium differed from those found for the house¯y. For
dimethoate, over the range of concentrations at which
house¯y larvae survived, there was no signi®cant
relationship between centipede mortality and concentration (Fig. 1(b)). This was true up to 5 mg a.i. kgÿ1
182
P. Kramarz, R. Laskowski / Applied Soil Ecology 13 (1999) 177±185
Fig. 3. Effect of linear alkylbenzenesulphonate (LAS; nominal treatments) on survival of housefly from eggs to imagines (a) and centipedes
after 30 days of exposure to LAS in soil and food (b); body weight change (c) and respiration rates (d) of centipedes exposed to LAScontaminated soil and food (averages SEM and regression lines); 95% C.I. represents 95% confidence interval for LC50.
soil, when mortality rates of the centipedes increased.
However, above this concentration, the curve was very
steep, so that at 25 mg a.i. kgÿ1 no single centipede
survived the duration of the experiment (Fig. 1(c)).
After 30 days, Cu did not affect centipedes signi®cantly at concentrations up to 1000 mg kgÿ1, but
above this concentration a very abrupt increase in
mortality was noted (Fig. 2(b)). After 85 days of
exposure to Cu-contaminated food and soil, the centipedes exposed to 56±1000 mg kgÿ1 survived considerably better than control animals, while the abrupt
increase in mortality above 1000 mg kgÿ1 was still
visible (Fig. 2(c)). For LAS, the relationship between
concentration and centipede survival was unclear and,
on average, only a slight decrease in survival was
observed. However, at a nominal concentration of
10 000 mg kgÿ1, survival of centipedes was still as
high as 85% of the control (Fig. 3(b)).
At the termination of the experiment, no signi®cant
sublethal effects (body weight change or respiration
rates) were found in centipedes (Figs. 1±3).
4. Discussion
Our studies indicate that L. mutabilis is relatively
resistant to all chemicals tested. All three chemicals
caused much higher mortality in cultures of its prey at
the respective concentrations. This suggests that in
contaminated areas lack of food rather than direct
poisoning may lead to disturbances in centipede populations. Also, predation pressure on resistant soil
invertebrates can drastically increase when other
invertebrates are affected by toxicants. This effect
would be expected especially for those chemicals with
the largest difference in LC50 between a predator and
its prey.
Among the chemicals tested in our studies, the
direct toxicity of LAS to centipedes appears negligible; hence, in this case exclusively secondary foodchain effects are to be expected. This result seems
reasonable, as centipede prey such as ¯y larvae, being
mostly detritivores feeding directly on and living in a
contaminated medium, are more exposed to a toxicant
P. Kramarz, R. Laskowski / Applied Soil Ecology 13 (1999) 177±185
than epigeic species such as centipedes. Biodegradable surfactants such as LAS exert toxic action mainly
through direct contact in water solution. Thus, soildwelling invertebrates are exposed to direct toxicity of
surfactants via the soil solution, while for surface
living forms there may be practically no effect.
There was also a substantial difference between
centipedes and house¯y larvae in their LC50s for
dimethoate. As dimethoate is highly toxic, both in
aqueous solution and as aerosol or vapour, the large
difference in toxicity was probably not caused solely
by the different life styles of house¯y larvae and
centipedes but also through dimethoate selective toxicity against the insects. It seems, thus, that although
dimethoate is considered a broad-spectrum insecticide
(Duf®eld and Aebischer, 1994; Duf®eld et al., 1996;
Huusela-Veistola, 1996), it is relatively safe for centipedes. As a consequence, secondary food-chain
effects may be more important with dimethoate as
in the case of LAS. On the other hand, dimethoate has
been found to be highly toxic to both, Coccinella and
Chrysopa (Dimetry and Marei, 1992). Thus, in the
case of dimethoate, the relative importance of direct
toxicity and secondary community-level effects to
carnivores probably depends not only on differences
in life styles (endo- or epigeic) but also on the relative
taxonomic positions of a predator and its prey: if a
pesticide is speci®c enough to affect the prey population almost exclusively, then mostly secondary effects
on predators can be expected. Consequently, it may be
anticipated that the larger the taxonomic distance
between the predator and its prey, the more important
the secondary effects will be.
For Cu, the LC50 for centipedes was also above the
range of concentrations tested and was estimated at ca.
1700 mg kgÿ1, while for house¯ies it was approximately one-third that value. This difference was,
however, substantially smaller than in the case of
dimethoate and LAS and suggests that, in this case,
direct toxicity may be important.
Ecotoxicological studies on the effects of toxicants
on terrestrial carnivorous invertebrates have followed
two directions. On the one hand, a number of studies
have been performed on concentrations of toxicants in
animals belonging to different trophic levels in the
hope of ®nding some `general rule', such as biomagni®cation, which would indicate which organisms are
the most endangered in contaminated ecosystems. On
183
the other hand, probably even more research has been
carried out on the side-effects of pesticides, mostly on
carnivorous beetles and spiders, but also on other
`bene®cial' arthropods. While the ®rst approach
apparently failed (cf. Beyer, 1986; Laskowski,
1991), the latter studies concentrated mostly on direct
toxic effects (e.g. Buchs et al., 1989; C,ilgi et al., 1996).
Although the biomagni®cation hypothesis did not
appear to be universal, other `food-chain' effects seem
possible. Among them, the most important from a
purely ecological point of view, are changes in interactions between populations of different species
exposed to toxicants (community-level effects) (cf.
Cohn and MacPhail, 1996). In many cases, the most
signi®cant among these interactions are predator±prey
relationships, because the abundance of prey ultimately determines the survival and fecundity of the
predators. For example, Purvis and Bannon (1992)
found that application of methiocarb-based slug pellets caused different effects in carabid populations,
depending on the season, and attributed these differences to purely indirect effects, such as prey availability and foraging behaviour. Possible interactions
between metal contamination and community-level
biotic factors were studied recently by Kiffney
(1996). This author found that metals can signi®cantly
in¯uence the relative importance of biotic factors in
altering freshwater macroinvertebrate community
structure: in streams dosed with metals, predation
intensity was greater than in controls. On the other
hand, Parmelee et al. (1993) found that soil-dwelling
predatory nematodes were more susceptible to Cu
contamination than their prey (also nematodes), which
resulted in reduced predation on herbivorous nematodes and, in consequence, greater numbers of nematodes compared to controls. Such results indicate how
important these interactions are and how inaccurate a
risk assessment may be if based solely on singlespecies, direct-toxicity tests.
The results obtained in this study, as well as those
reported by Parmelee et al. (1993) and Kiffney (1996),
add another dimension to laboratory-based risk assessment procedures. They suggest that, in some situations, there may be secondary community-level
effects, understood here as interactions between toxic
substances and biotic factors, rather than direct
poisoning, which can determine the ®nal impact of
a toxicant on a number of species in a polluted
184
P. Kramarz, R. Laskowski / Applied Soil Ecology 13 (1999) 177±185
ecosystem. The species particularly susceptible to
secondary trophic-level effects would probably be
the epigeic predators feeding on endogeic fauna. This
leads to the conclusion that selectivity of a particular
toxicant against particular species or a group of species, should only be the ®rst step of risk assessment
studies. A similar conclusion was also reached by
Duf®eld et al. (1996), who studied recolonisation
patterns of predatory invertebrates in ®elds sprayed
with insecticides.
Acknowledgements
We are grateful to all members of the SECOFASE
project for their co-operation and fruitful discussions
during project workshops. Special thanks are due to
Dr. Hans Lùkke, the Project Co-ordinator, for all his
help. We also wish to thank to Dr. Philip Heneghan for
a critical reading of the manuscript. The funding for
this work was provided by the EU Programme Environment 1990±1994 (the PECO Programme Contract
No. ERB-CIPD-CT93-0059).
References
Abd Allah, A.T., Wanas, M.Q.S., Thompson, S.N., 1997. Effects of
heavy metals on survival and growth of Biomphalaria glabrata
Say (Gastropoda: Pulmonata) and interaction with schistosome
infection. J. Mollus. Stud. 63, 79±86.
Albert, A.M., 1983. Energy budgets for populations of long-lived
arthropod predators (Chilopoda: Lithobiidae) in an old beech
forest. Oecologia 56, 292±305.
Beyer, W.N., 1986. A reexamination of biomagnification of metals
in terrestrial food chains. Environ. Toxicol. Chem. 5, 863±864.
Buchs, W., Heimbach, U., Czarnecki, E.A.D., 1989. Effects of snail
baits on non-target carabid beetles. Monographs of British Crop
Protection Council, vol. 41. pp. 245±252.
C,ilgi, T., Wratten, S.D., Robertson, J.L., Turner, D.E., Holland,
J.M., Frampton, G.K., 1996. Residual toxicities of three
insecticides to four species (Coleoptera: Carabidae) of
arthropod predators. Can. Entomol. 128, 1115±1124.
Cohn, J., MacPhail, R.C., 1996. Ethological and experimental
approaches to behaviour analysis: implications for ecotoxicology. Environ. Health Perspect. 104, 299±305.
Colinvaux, P., 1973. Introduction to Ecology. Wiley, New York.
Collier, B.D., Cox, G.W., Johnson, A.W., Miller, P.C., 1973.
Dynamic Ecology. Prentice±Hall, Engelwood Cliffs, NJ.
Dallinger, R., Berger, B., Gruber, A., 1993. Quantitative aspects of
zinc and cadmium binding in Helix pomatia: differences
between an essential and a nonessential trace element. In:
Dallinger, R., Rainbow, P. (Ed.), Ecotoxicology of Metals in
Invertebrates. Lewis Publishers, Boca Raton, Ann Arbor,
London, Tokyo. pp. 315±332.
Dimetry, N.Z., Marei, S.S., 1992. Laboratory evaluation of some
pesticides on the cabbage aphid, Brevicoryne brassicae, and
their side-effects on some important natural enemies. Anzeiger
fuÈr Schadlingskunde, Pflanzenschutz und Umweltschutz 65,
16±19.
Duffield, S.J., Aebischer, N.J., 1994. The effect of spatial scale of
treatment with dimethoate on invertebrate population recovery
in winter-wheat. J. Appl. Ecol. 31, 263±281.
Duffield, S.J., Jepson, P.C., Wratten, S.D., Sotherton, N.W., 1996.
Spatial changes in invertebrate predation rate in winter wheat
following treatment with dimethoate. Entomologia Experimentalis et Applicata 78, 9±17.
Ernst, W.H.O., Joosse van Damme, E.N.G., 1983. Umweltbelastung durch Mineralstoffe. Biologische Effekte. VEB Gustav
Fischer Verlag, Jena.
Huusela-Veistola, E., 1996. Effects of pesticide use and cultivation
techniques on ground beetles (Coleoptera, Carabidae) in cereal
fields. Ann. Zool. Fenn. 33, 197±205.
Kiffney, P.M., 1996. Main and interactive effects of invertebrate
density, predation, and metals on a Rocky Mountain stream
macroinvertebrate community. Can. J. Fisheries. Aquat. Sci. 53,
1595±1601.
Klekowski, R., 1975. Constant-pressure volumetric microrespirometer for terrestrial invertebrates. In: GrodzinÄski, W.,
Klekowski, R., Duncan, A. (Eds.), Methods for ecological
bioenergetics. IBP Handbook No. 24., Blackwell Scientific
Publications, Oxford, London, Edinburgh, Melbourne. pp. 212±
225.
Korthals, G.W., de Goede, R.G.M., Kammenga, J.E., Bongers, T.,
1996. The maturity index as an instrument for risk assessment
of soil pollution. In: Van Straalen, N.M., Krivolutsky, D.A.
(Ed.), Bioindicator Systems for Soil Pollution. Kluwer Academic Publishers, Dordrecht. pp. 85±93.
Kula, H., Heimbach, U., Lùkke, H. (Ed.), 1995. Progress Report
1994 of SECOFASE, Third Technical Report. Development,
improvement and standardization of test systems for assessing
sublethal effects of chemicals on fauna in the soil ecosystem.
National Environmental Research Institute, Denmark. pp. 166.
Laskowski, R., 1991. Are the top carnivores endangered by heavy
metal biomagnification? Oikos 60 387±390.
Lùkke, H. (Ed.), 1995. Effects of pesticides on meso- and
microfauna in soil. Ministry of Environment and Energy,
Denmark.
Lùkke, H., Van Gestel, C.A.M. (Eds.), 1998. Handbook of soil
invertebrate toxicity tests. John Wiley & Sons, Chichester, UK.
Matthews, R.A., Landis, W.G., Matthews, G.B., 1996. The
community conditioning hypothesis and its application to
environmental toxicology. Environ. Toxicol. Chem. 15, 597±
603.
Moriarty, F., 1975. Pollutants and animals: a factual perspective.
George Allen and Unwin Ltd., London.
Odum, E.P., 1971. Fundamentals of Ecology. Saunders, Philadelphia.
P. Kramarz, R. Laskowski / Applied Soil Ecology 13 (1999) 177±185
Parmelee, R.W., Wentsel, R.S., Phillips, C.T., Simini, M., Checkai,
R.T., 1993. Soil microcosm for testing the effects of chemicalpollutants on soil fauna communities and trophic structure.
Environ. Toxicol. Chem. 12, 1477±1486.
Poulton, B.C., Monda, D.P., Woodward, D.F., Wildhaber, M.L.,
Brumbaugh, W.G., 1995. Relations between benthic community
structure and metals concentrations in aquatic macroinvertebrates: Clark Fork, Montana. J. Freshwater Ecol. 10, 277±293.
Purvis, G., Bannon, J.W., 1992. Nontarget effects of repeated
methiocarb slug pellet application on carabid beetle (Coleop-
185
tera, Carabidae) activity in winter-sown cereals. Ann. Appl.
Biol. 121, 401±422.
Van Straalen, N.M., Van Wensem, J., 1986. Heavy metal content of
forest litter arthropods as related to body-size and trophic level.
Environ. Pollut. (A) 42, 209±221.
Willamo, R., Nuorteva, P., 1987. The role of heavy metals in forest
die-off. In: Anttila, P., Kauppi, P. (Ed.), Symposium of the
Finnish Research Project on Acidification (HAPRO). Ministry
of the Environment, Ministry of Agriculture and Forestry.
pp. 64±67.