Ecotoxicology and Environmental Safety 52, 75}81 (2002)

Environmental Research, Section B

doi:10.1006/eesa.2002.2149, available online at http://www.idealibrary.com on

Joint Toxicity of Linear Alkylbenzene Sulfonates and Pyreneon Folsomia fimetaria

John Jensen*�� and Line E. Sverdrup-*National Environmental Research Institute, Department of Terrestrial Ecology, P.O. Box 314, Vejls~vej 25, Silkeborg, DK-8600, Denmark;

and -Jordforsk, Norwegian Centre for Soil and Environmental Research, N-1432 As s, Norway

Received December 31, 2001

Surfactants may enhance the biodegradation of hydrophobicsubstances in soils. This has partly been attributed to an increasein the bioavailability, brought about by the presence of surfac-tants. The aim of this study was to examine the ecotoxicologicale4ects of the detergent linear alkylbenzene sulfonate (LAS) andthe polycyclic aromatic hydrocarbon pyrene, alone and incombination, using the survival and reproduction of the collem-bolan Folsomia fimetaria as endpoints. The EC50 and EC10 were803 and 161 mg kg�1 for LAS, and 23 and 15 mg kg�1 forpyrene. If LAS was able to increase the bioavailability of pyreneto springtails, it was expected that the combined e4ect of the twosubstances would exceed the e4ect found for each of the com-pounds tested separately. However, the results showed no e4ectof LAS on the toxicity of pyrene in the concentration rangetested (1+750 mg LAS kg�1 dry weight). Both the toxic unitconcept and the isobologram method indicated that an additiveapproach would be the most useful when assessing the risk ofthese two compounds. � 2002 Elsevier Science (USA)

Key Words: collembolan; springtails; ecotoxicity; sewagesludge; mixtures.

INTRODUCTION

There is an obvious contradiction between the generalpractice of evaluating ecological risk of contaminants on thebasis of toxicity data from organisms exposed to singlesubstances and the fact that organisms living in and oncontaminated soil typically are simultaneously exposed tovarious contaminants. The extrapolation from standardizedlaboratory conditions to real-world situations generatesvarious uncertainties of which joint e!ects of chemicals areone of the major qualms (Posthuma et al., 1997). Theseuncertainties are normally accounted for by introduc-ing safety or assessment factors (Chapman et al., 1998).

� To whom correspondence should be addressed. Fax: #45 89201413.E-mail: john.jensen@dmu.dk.

75

However, this approach is, in its current use, very oftenconservative and applied to all kinds of pollution regardlessof the nature and history of the contaminants. Joint e!ectsof toxicants may be similar to (additive) or stronger(synergy) or weaker (antagony) than expected from theobserved e!ects of separate exposure. The responses tosimultaneous exposure to multiple chemicals depend on theconstituents of the mixture and may vary signi"cantly, e.g.,additive e!ect of narcotic compounds (Hermens et al., 1984)or possible synergy between di!erent stressors (e.g., H+jer etal., 2001). Such variation calls for a "rm scienti"c basis forcoping with chemical mixtures in risk assessment. Whereasthe toxicity of mixtures are relatively well studied in theaquatic environment (e.g., Kraak et al., 1993; Enserink et al.,1991; McCarthy et al., 1992), the toxicity of chemical mix-tures in soils is not very well investigated. Within the soilenvironment di!erent types of interactions between chem-icals may occur. These include chemical and physico-chem-ical interactions with other constituents of the soil, thusa!ecting sorption and bioavailability, physiological interac-tions in the organisms a!ecting uptake from the soil or porewater, or mechanistic processes at the target receptors(Calamari and Alabaster, 1980; Van Gestel and Hensbergen,1997). Chemicals present in mixtures may interfere witheach other at each of these steps.

Polycyclic aromatic hydrodrocarbons (PAHs) constitutea variable group of compounds, all made up of 2}7 benzenerings. PAHs are not produced for any speci"c use, butdue to both natural and man-made processes they arewidespread in the environment. They may occur in largequantities in industrial products from the petro-chemicalindustry, the coke factories, and coal tar processing, and increosote wood preservatives. The emission of PAHs into theenvironment is a result of spills, or incomplete combustionof fossil fuels such as coal, oil, petrol, and wood. PAHsemitted to the air from coal and petrol combustion at, e.g.,gas works and cooking plants or by car and air tra$c aremainly bound to aerosols and hence able to travel long

0147-6513/02 $35.00� 2002 Elsevier Science (USA)

All rights reserved.

76 JENSEN AND SVERDRUP

distances (BjoK rseth et al., 1979). PAHs are typically presentas complex mixtures at contaminated sites, land"lls, oragricultural areas receiving sewage sludge. Some PAHs arepotentially carcinogenic and the fate of these compounds istherefore of high public interest. With n-octanol}water par-titioning coe$cients (log K

��values) between approxim-

ately 3.4 (naphthalene) and 7.7 (indeno(1,2,3,-cd)pyrene)(Sims and Overcash, 1983), PAHs are relatively insoluble inwater and are therefore primarily associated with the partic-ulate phase, in particular the organic matter (Means et al.,1980). The removal of PAHs from the soil takes place fromabiotic loss (e.g., leaching, hydrolysis, photodegradation,volatilization) and microbial degradation. Volatilization isa signi"cant removal pathway for the two- and three-ringedPAHs (Bossert and Bartha, 1986; Park et al., 1990), whereasPAHs with four or more rings are much less a!ected byabiotic processes. Biodegradation is also signi"cantly higherfor two- and three-ringed PAHs (Wilson and Jones, 1993;Sims and Overcash, 1983). The half-lives of PAHs in theenvironment strongly depend upon the contamination his-tory of the soil and parameters such as temperature, pH,redox conditions, soil structure, and nutrient supply(Maliszewska-Kordybach, 1993; Manilal and Alexander,1991; Mihelcic and Luthy, 1988; Coover and Sims, 1987).In general, many of the PAHs have half-lives longerthan 100 days. Whereas the bioavailability and leachingof the PAHs with "ve or more rings is very limited,due to their high absorption to soil material, some leachinghas been observed for the smaller compounds, such asnaphthalene and the three-ringed PAHs (Grenney et al.,1987).

The low bioavailability of many PAHs limits the deve-lopment of e$cient biodegradation technologies in remedi-ation processes. Chemically and biologically producedsurfactants are known to enhance the solubility of hydro-phobic organic compounds like PAHs (e.g., Aronstein et al.,1991; Abdul et al., 1991). They are therefore occasionallyused when remediating, for example, coal tar-contaminatedsoils. The mechanisms behind surfactant-enhanced biode-gradation and the interactions between the microorganism,the PAHs, and the surfactants are not fully understood. Forexample, more information on how the surfactants in#uencethe survival and activity of the microbial degraders isneeded before surfactants can be applied successfully asa bioremediation technique (Willumsen, 1998). Increasingthe bioavailability of organic contaminants like PAHs maynot only make them more available for microbial degrada-tion, but may also make them more available for uptake inother soil-dwelling organisms, like plants and soil fauna.Furthermore, surfactants may on their own pose a risk tosoil living species (Jensen et al., 2001). Simultaneous expo-sure to surfactants and PAHs may also occur in sewagesludge-amended soils. For example, linear alkyl benzenesulfonate (LAS) detergents can be found in a large number

of household products, and for this reason they are presentin high concentrations in sewage sludge. The soil concentra-tion in sludge-amended soils will typically be below 10 mgkg��, but may occasionally reach more than 50 mg kg��

(Jensen, 1999).In this article the reproductive output of the soil micro-

arthropod Folsomia ,metaria exposed to di!erent concen-trations of a commercial LAS and pyrene, alone and incombination, is investigated.

MATERIALS AND METHODS

Test Substances and Sample Preparation

Pyrene (desiccate) was obtained from Sigma}Aldrich.Acetone (J. T. Barker, HPLC quality) was used as solvent.The LAS sodium salt used was an aqueous solution with anactive matter concentration of 20.3% (w/w) (Isorchem113/S-Na, Condea Augusta S. p. A., Milan, Italy). Thepercentage distribution of the linear alkyl chain wasC

��(14%), C

��(34%), C

��(31%), and C

��(21%). Pyrene

was dissolved in acetone in a stock solution correspondingto the highest test dosage. For each of the test concentra-tions used, dilutions from the stock solution were madeusing acetone; 4 mL of acetone with test substance wasadded to each container. Two control samples with pureacetone or water was also prepared. Following the additionof test substance in acetone, the containers were weighedand the weights recorded. The solvent was evaporated un-der a fume head for 24 h, and the containers were thenweighed again to make sure the solvent was totally evapor-ated. LAS, dissolved in water at concentrations correspond-ing to the dosages tested, was then added so that thewater content equaled 50% of the water-holding capacityand mixed thoroughly into the soil. The soil wasthen transferred to plastic cylinders (5.5 cm high, diameter6 cm) with a 1-mm mesh in the bottom. During incubationthe test containers were closed at the top and bottom withlids.

Experimental Soil

A natural Danish soil (Askov soil), dried at 803C for 24 h,was used. Samples were sieved through a 2-mm mesh priorto use. The Askov soil is a sandy loam, and it has thefollowing particle size distribution: coarse sand (200}2000 �m) 38.4%, "ne sand (63}200 �m) 23.6%, coarse silt(20}63 �m) 10.0%, "ne silt (2}20 �m) 12.3%, clay ((2 �m)13.0%. The humus content of the soil was 2.8%, and thetotal content of organic carbon, 1.6%. The soil pH (pH-H

�O) was 6.2, the soil density was 1.135 (g/cm� dry soil), and

the total cation exchange capacity (CEC) was 8.14 meq100 g��.

TABLE 1Nominal and Measured Concentrations in One of the Tests

Perfomed with Pyrene

Nominal Measured pyrene concentrations (mg/kg dry weight)concentrations(mg/kg dry weight) Beginning of test End of test

6 4.9$0.2 5.2$0.360 58$3 56$1

77JOINT TOXICITY OF LAS AND PAHs TO SPRINGTAILS

Test Organisms

Collembolans (F. ,metaria) were synchronized, mass-bredin petri dishes containing a regularly dampened charcoal/plaster of Paris mixture, kept in darkness at 20$13C, andfed on bakers's yeast. The springtail F. ,metaria is widelydistributed and common in several soil types ranging fromsandy to loamy soils and from mull to mor soils. It hasan omnivorous feeding habit, including fungal hyphae,bacteria, protozoa, and detritus in its food. Adults of F.,metaria are 0.8}1.4 mm long and weigh 10}30 �g (dryweight). The test animals used in the tests are permanentlycultured at the Danish National Environmental ResearchInstitute. F. ,metaria is sexually mature after 18 days, reach-ing the sixth instar. They reproduce sexually. Synchronizedcultures were obtained by incubating adult animals in newpetri dishes for a few days, after which the adults wereremoved. The "rst eggs hatch after approximately 10 days.After 3 days, unhatched eggs were removed, leaving 0- to3-day-old juveniles. Sexual di!erence between male andfemale is di$cult to determine before 23}26 days afterhatching, and this age group is therefore used for the experi-ments.

Principle of Test

The tests were performed according to a procedure des-cribed by Wiles and Krogh (1998). Based on a preliminaryrange "nding test, a range of concentrations was selected forthe reproduction test. In the reproduction test, a geometricseries of "ve concentrations or mixtures were used, in addi-tion to the control samples. Four replicates were used perconcentration. Test concentrations in the pyrene and LAStests were 240, 120, 60, 30, and 15 mg pyrene kg�� soil dryweight and 800, 400, 200, 100, and 50 mg LAS kg�� soil dryweight. In the mixture studies four exposure series wereused. Two tests were performed using the same test concen-trations of pyrene as in the standard test, but with theaddition of 1 and 10 mg LAS kg��, respectively, to allreplicates. In addition, two tests were performed where thetest concentrations of both LAS and pyrene were varied. Inthe "rst of these tests, LAS#pyrene concentrations were5#0.8, 15#2.4, 50#8, 150#24, and 500#80 mg kg��,corresponding to EC

��proportional values of �

�: �

for theLAS:pyrene mixture. In the second test, LAS#pyrene con-centrations were 7.5#0.6, 2.5#1.8, 75#6, 225#18, and750#60 mg kg��, corresponding to EC

��proportional

values of �

: �

for the LAS:pyrene mixture.Following the sample preparation, ten male and 10 fe-

male collembolans were added to each replicate. The testswere conducted at 20$13C under lighting of about400}800 lux with a 12/12-h photoperiod. The exposureperiod was 21 days, and test containers were opened after7 and 14 days for feeding and, in case of water loss, addition

of water. At the end of the test period, the organisms wereextracted using a controlled-temperature gradient extractorbased on the principles described by MacFadyen (1961) andPetersen (1978). Extraction was started at 253C and thetemperature was increased automatically every 12 h by 53C.After 12 h at 403C the extraction was "nished. During theextraction, animals were collected in cool collecting vessels(53C) with a bottom layer of plaster of Paris/charcoal.

Counting and Statistical Analysis of Results

The collembolans were counted using digital imageprocessing (DIP) and image analysis procedure as describedby Krogh et al. (1998). The data produced by the imageanalysis software (Anakron, Roskilde) were analyzed usingthe computer program SAS/INSIGHT Release 6.12 (SAS,1999). Estimations of the 50 and 10% e!ect concentrationsfor reproductive output (EC

��and EC

��values, respective-

ly) were done by logistic regression in SAS/INSIGHT Re-lease 6.12 (SAS, 1999). Data were "tted according to thefollowing formulas, which showed a good "t to the dose}response curves: For the EC

��: juv"k (1#(conc/EC

��)�)��;

for the EC��

: juv"k (1#(1/9)* (conc/EC10)�)��. No-observed-e!ect concentrations (NOEC values) were as-sessed using ANOVA and Dunnett's procedure (at a 5%signi"cance level) on the data. The 50% e!ect concentra-tions for lethality (LC

��values) were estimated by probit

analysis in SAS/INSIGHT Release 6.12 (SAS, 1999).

RESULTS

Veri"cation of exposure concentrations was performedfor two levels of pyrene in one of the tests. As seen inTable 1, the measured concentrations of pyrene were closeto nominal. It was therefore decided to base all e!ectconcentrations on nominal concentrations. The chemicalanalyses also indicated that there was no or very littledegradation of pyrene during the 3-week test period. Toevaluate the repeatability of the F. ,metaria test results,three standard tests were performed using pyrene as thetoxicant. The results (Table 2) reveal that the estimatedtoxicity is quite stable, and that the EC

��-value seems to be

the most stable of the e!ect measures used.

TABLE 2Toxicity Tests with F. fimetaria Exposed to LAS or Pyrene

Estimated e!ect concentrations (mg/kg dry weight)

LC��

(95% CI) EC��

(95% CI) EC��

(95% CI) NOEC

Pyrene (1) 63 (58, 73) 22 (16, 28) 16 (8, 24) 15Pyrene (2) 79 (65, 315) 24 (23, 24) 17 (15, 17) 15Pyrene (3) 82 (75, 90) 24 (18,27) 12 (0.4, 20) 15LAS '800 803 (560, 1000) 161 (113, 209) 200

Note. LC��

is the concentration estimated to cause 50% lethality,EC

��/EC

��and NOEC values are based on reproductive output.

FIG. 1. Isobolodiagram of the EC��

values derived from exposing thespringtail Folsomia ,metaria to mixtures of LAS and pyrene.

78 JENSEN AND SVERDRUP

The e!ect of LAS on springtail mortality and reproduc-tion is found in Table 3. The highest test concentration wasnot high enough to cause 50% mortality, and the LC

��could therefore not be estimated. The results of the com-bined exposure studies are found in Table 3 and as iso-bolographs in Figures 1 and 2. An isobologram representsiso-e!ect lines (isobols) of all possible mixtures of the twocompounds. The shape of the isobols represents a speci"ctoxicological response or interaction; i.e., concave curveindicates synergy, a straight line represents concentrationaddition, and a convex line corresponds to antagonism. Thedata in Fig. 1 (combined e!ects at the EC

��-level) and Fig. 2

(combined e!ects at the EC��

-level) reveal a linear relation-ship, indicating that the e!ects of LAS and pyrene areadditive. This conclusion is also supported by the resultsfrom the toxic unit approach (Table 3). The average sum oftoxic units falls within a range 1.35 and 1.05, depending on

TABLE4ect Concentrations for LAS and Pyrene (PY

with Di4erent Mixture

E!ect concentrations (mg kg dr

LC��

LAS Pyrene TU LAS P

'800 * 803* 75 *

817 65 (1,92� 169490 78 (1,66� 1681 66 (0,89� 110 70 (0,95� 10

Average (1,35

Note. The e!ect concentrations for pyrene are the average of the three tes� E!ect measures determined from concentration series of LAS}pyrene mi� E!ect measures determined from concentration series of LAS}pyrene mi� E!ect measures determined from a mixture of 1 or 10 mg LAS kg�� and

the toxicity measure. One toxic unit equals the LC��

/EC��

/EC

��for a speci"c substance, depending on the endpoint

used for the examination of joint e!ects. As toxicunits (TU) are dimensionless, the toxicity of a mixture canbe described as the sum of the TU contributed by eachcomponent, TU being the ratio between the concentrationof a chemical in mixture and the threshold value, e.g.,EC

��or EC

��, of that chemical. In situations where the sum

of TU is approximately one, the toxicity of the mixture iswhat could be predicted from chemicals having additivee!ects.

E 3R) Alone and in the Four Tests Performed

s of LAS and Pyrene

y weight or toxic units (TU))

EC��

EC��

yrene TU LAS Pyrene TU

* 161 *

23 * 1514 0,81� 64 5 0,73�

27 1,37� 78 13 1,35�

24 1,03� 1 17 1,14�

23 1,00� 10 14 1,00�

1,05 1,05

ts presented in Table 2.xtures with a "xed ratio in toxic units (EC

��) of �

:�.

xtures with a "xed ratio in toxic units (EC��

) of ��:�.

concentration series of pyrene in the range 15}240 mg kg��.

FIG. 2. Isobolodiagram of the EC��

values derived from exposing thespringtail Folsomia ,metaria to mixtures of LAS and pyrene.

79JOINT TOXICITY OF LAS AND PAHs TO SPRINGTAILS

DISCUSSION

The chemical analyses revealed that very little or nopyrene disappeared during the 3-week test. This is in linewith several degradation studies, which generally "nd a slowdegradation of pyrene and other four-ringed PAHs (Simsand Overcash, 1983; Park et al., 1990). The disappearancerate of LAS was not measured in this study. However,Jensen (1999) reviewed the fate and e!ect of LAS in theterrestrial environment, and reported half-lives of LAS be-tween 1 and 3 weeks under normal environmental condi-tions. A signi"cant degradation of LAS is therefore expectedto occur during the 3-week test period.

Test results for the three pyrene tests indicate that theestimated toxicity values are quite stable, and that theEC

��-value seems to be the most stable of the e!ect

measures used. However, concerning mortality, there is nota complete overlap between the 95% con"dence intervals ofthe three tests. Less than 50% mortality was observed at thehighest test concentration of 800 mg kg�� in the LAS toxi-city study. A low acute lethal toxicity of LAS was also foundby Holmstrup et al. (2001). In a number of tests with vari-able soil types and LAS formulations they also found theLC

��for F. ,metaria to be higher than 800 mg kg��. How-

ever, they found e!ects of LAS on reproduction at concen-tration levels lower than those observed in this study, i.e.,average EC

��and EC

��values of approximately 95 and

440 mg kg��. Holmstrup and Krogh (1996) also failed to"nd any mortality of springtails below 1000 mg LAS kg��,whereas their EC

��and EC

��values of 147 and 737 mg

LAS kg�� is in accordance with the observations in thepresent study. The e!ects of pyrene and other PAHs on soilinvertebrates are not well studied. Sverdrup et al. (2001)observed e!ects of pyrene within the same concentration

range as the results from this study. They found LC��

,EC

��, and EC

��values of 53, 16, and 10 mg kg��, respec-

tively. All values were slightly lower than those in this study(Table 2).

The combined e!ect of LAS and pyrene mixtures wastested to examine whether these substances, which typicallyare present in sewage sludge simultaneously, would haveany synergistic e!ect on springtails or if additivity could beassumed. A synergetic toxicological action was not ex-pected, but it was hypothesized that the presence of LAScould increase the bioavailability of pyrene, and in thatrespect increase its toxicity. In the current study, no en-hanced toxicity was found, and both the toxic unit concept(Table 3) and the isobologram method (Figs. 1 and 2)indicated that a simple additive model would be the mostuseful when assessing the risk of these two compounds.Although some PAHs are suspected to be genotoxic, themain toxic action of both LAS and pyrene is likely to benonpolar narcosis, i.e., a general perturbation of biomem-branes (Sverdrup et al., 2001; Jensen, 1999). Although somestudies have revealed that this is not always the case (e.g.,Warne et al., 1989), it is generally believed that the toxicityof nonpolar narcotic toxicants are additive, at least at levelsbelow their thresholds.

The toxicity of surfactants in mixtures has received con-siderable attention since they are used as oil dispersants andas additives in pesticides to reduce the surface tension of theformulations. Lewis (1992) reviewed the e!ect of mixtureson the toxicity of surfactants to aquatic species. He foundthat oil}surfactant mixtures usually were more toxic thanexpected from exposure to the oil alone.

For metals it has been found that interaction among thechemicals outside the body are important and partiallydeter toxicity (Posthuma et al., 1997). To the authorsknowledge it has not been determined how detergents mayin#uence the toxicity or bioavailability of PAH to soilinvertebrates, for example, by increasing their uptake. How-ever, several studies have veri"ed that detergents, andespecially the nonionic detergents, may enhance the biode-gradation of PAHs, probably by increasing the bioavailablefraction (Tiehm, 1994; Aronstein and Alexander, 1992; Mad-sen and Kristensen, 1997). The lack of augmentation inPAH toxicity by the presence of LAS observed in this studymay be explained by a number of reasons. It is a known factthat surfactants may increase the apparent solubility ofPAHs (Edwards et al., 1991; Tiehm, 1994). However, thisability is closely related to the formation of surfactantmicelles. In general, concentrations of surfactants below thecritical micelle concentration (CMC) have little or no e!ecton the solubility of hydrophobic substances (Haigh, 1996).The CMC depends on a number of variables, but in generalnonionic surfactants have much lower CMC than anionicsurfactants such as LAS. Whereas the CMC for LAS isapproximately 410 mg L��, the CMC for a typical nonionic

80 JENSEN AND SVERDRUP

surfactant is in the range 20 mg L�� (Haigh, 1996). A soilpore water concentration of approximately 400 mg L�� isexceeded in the highest LAS concentrations used in thesemixture studies, as the test concentration range 75}750 mgLAS kg�� approximately corresponds to 500}5000 mgLAS L�� in the soil pore water, assuming that all LAS willbe present in the water phase. Nevertheless, the stimulatinge!ect of LAS on solubility of organic contaminants is likelyto be smaller than the ones reported for nonionic surfac-tants. Fox et al. (1997) found, for example, that at concentra-tions typically found in sewage sludge, LAS had no e!ect onthe biodegradability of 2,4-dichlorophenoxyacetic acid and�-naphthol.

In contrary to freshly spiked soils very little pyrene isexpected in the water phase of naturally aged soils. Madsenand Kristensen (1997), for example, did not "nd any degra-dation of pyrene in coal tar-contaminated soils that werenaturally aged, whereas signi"cant degradation was ob-served when surfactants were added to the soil. They alsofound that the enhancement of PAH degradation was lessconvincing when easily degradable surfactants were added.Although not veri"ed by chemical analyses a half-life ofLAS of 1 week is realistic under the relatively optimalconditions in the current study (203C and aerobic incuba-tion) (Jensen, 1999). This may also partly explain why theobserved toxicity in the mixed exposure did not di!er fromadditivity, although the initial pore water concentrationexceeded the CMC of LAS in two studies. Further investiga-tions using nonionic surfactants with CMC lower than thatof LAS have been initiated to reveal whether synergistice!ects of combined exposure to surfactants and organiccontaminants on soil fauna are likely.

CONCLUSIONS

Both pyrene and LAS were found to be toxic to collem-bolans. The e!ect levels were similar to those of previouslyreported studies and generally far above what would befound in soils outside contaminated areas. When exposedsimultaneously to di!erent LAS}pyrene mixtures, spring-tails responded to the contaminants in a way that could bepredicted by assuming an additive toxicity. The concentra-tions of LAS and pyrene is well beyond what would beexpected, for example, in soils recently amended with sew-age sludge. It may hence be concluded that the current useof assessment or safety factors in ecological risk assessmentis likely to account for potential derivation from additivityamong substances acting by narcosis and present in sewagesludge.

ACKNOWLEDGMENTS

Financial support from the Norwegian Research Council (GRUF), the&&Centre for Sustainable Land Use,'' and the &&Centre for Biological

Processes in Contaminated Soil and Sediment'' (BIOPRO) under theDanish Environmental Research Program is acknowledged. We also thankZdenek Gabor, Trine G. S+rensen, Elin J+rgensen, and Karen Kjvr Jacob-sen for assistance in the cultivation of test organisms; Paul Henning Kroghfor statistical assistance; and Torben Nielsen, Ris+, Denmark, for chemicalanalyses.

REFERENCES

Abdul, A. S., Gibson, T. L., and Rai, D. N. (1991). Selection of surfactantsfor the removal of petroleum products from shallow sandy aquifers.Ground =ater 28, 921}926.

Aronstein, B. N., and Alexander, M. (1992). Surfactants at low concentra-tions stimulate biodegradation of sorbed hydrocarbons in samples ofaquifer sands and soil slurries. Environ. ¹oxicol. Chem. 11, 1227}1233.

Aronstein, B. N., Calvillo, Y. M., and Alexander, M. (1991). E!ects ofsurfactants at low concentrations on the desorption and biodegradationof sorbed aromatic compounds in soil. Environ. Sci. ¹echnol. 25,1728}1731.

BjoK rseth, A., Lunde, G., and Lindskog, A. (1979). Long-range transportof polycyclic aromatic hydro carbons. Atmosph. Environ. 13, 45}53.

Bossert, I. D., and Bartha, R. (1986). Structure}biodegradability relation-ships of polycyclic aromatic-hydrocarbons in soil. Bull. Environ.Contam. ¹oxicol. 37, 490}495.

Calamari, D., and Alabaster, J. S. (1980). An approach to theoreticalmodels in evaluating the e!ects of mixtures of toxicants in the aquaticenvironment. Chemosphere 9, 533}538.

Chapman, P. M., Fairbrother, A., and Brown, D. (1998). A critical evalua-tion of safety (uncertainty) factors for ecological risk assessment.Environ. ¹oxicol. Chem. 17, 99}108.

Coover, M. P., and Sims, R. C. (1987). The e!ect of temperature onpolycyclic aromatic hydrocarbon persistence in an unacclimated agricul-tural soil. Hazard=aste Hazard Mater. 4, 69}82.

Edwards, D. A., Luthy, R. G., and Liu, Z. (1991). Solubilization of polycyc-lic aromatic hydrocarbons in micellar nonionic surfactant solutions.Environ. Sci. ¹echnol. 25, 127}133.

Enserink, E. L., Maas-Diepeveen, J. L., and van Leeuwen, C. J. (1991).Combined e!ects of metals: An ecotoxicological evaluation.=ater Res.25, 679}687.

Fox, K. K., Chapman, L., Solbe, J., and Brennand, V. (1997). E!ect ofenvironmentally relevant concentrations of surfactants on the desorp-tion or biodegradation of model contaminants in soil. ¹enside Surfact.Det. 34, 436}441.

Grenney, W. J., Caupp, C. L., and Sims, R. C. A. (1987). MathematicalModel for the fate of hazardous sunstances in soil: Model describtionand experimnetal results. Hazard=aste Hazard Mater. 4, 223}239.

Haigh, S. D. (1996). A review of the interaction of surfactants with organiccontaminants in soil. Sci. ¹otal. Environ. 185, 161}170.

Hermens, J., Canton, H., Janssen, P., and de Jong, R. (1984). Quantitativestructure}activity relationships and toxicity studies of mixtures of chem-icals with anaesthetic potensy: Acute lethal and sublethal toxicity toDaphnia magna. Aquat. ¹oxicol. 5, 143}154.

H+jer, R., Bayley, M., Damgaard, C. F., and Holmstrup, M. (2001). Stresssynergy between drought and a common environmental contaminant:Studies with the Collembolan Folsomia candida. Global Change Biol. 7,485}494.

Holmstrup, M., and Krogh, P. H. (1996). E!ects of an anionic surfactant,linear alkylbenzene sulfonate, on survival, reproduction and growth ofthe soil-living collembolan Folsomia ,metaria. Environ. ¹oxicol. Chem.15, 1745}1748.

81JOINT TOXICITY OF LAS AND PAHs TO SPRINGTAILS

Holmstrup, M., and Krogh, P. H. (2001). E!ects and risk assessment oflinear alkylbenzene sulphonates (LAS) in agricultural soil. III. Sub-lethale!ects on soil invertebrates. Environ. ¹oxicol. Chem. 20, 1673}1679.

Jensen, J. (1999). Fate and e!ects of linear alkylbenzene sulphonates(LAS) in the terrestrial environment*A review. Sci. ¹otal Environ. 226,93}111.

Jensen, J., L+kke, H., Holmstrup, M., Krogh, P. H., and Elsgaard, L.(2001). E!ect and risk assessment of linear alkylbenzene sulphonates(LAS) in agricultural soils. V. Risk assessment of LAS in sludgeamended soils. Environ. ¹oxicol. Chem. 20, 1690}1697.

Kraak, M., Schoon, H., Peeteres, W. H. M., and van Straalen, N. M.(1993). Chronic ecotoxicity of mixtures of Cu, Zn, and Cd to the zebramussel Dreissena polymorpha. Ecotoxicol. Environ. Saf. 25, 315}327.

Krogh, P. H., Johansen, K., and Holmstrup, M. (1998). Automatic countingof collembolans for laboratory experiments. App. Soil Ecol. 7, 201}205.

Lewis, M. A. (1992). The e!ects of mixtures and other environmentalmodifying factors on the toxicities of surfactants to fresh-water andmarine life. =ater Res. 26, 1013}1023.

MacFadyen, A. (1961). Improved funnel-type extractors for soil arthro-pods. J. Animal Ecol. 30, 171}184.

Madsen, T., and Kristensen, P. (1997). E!ects of bacterial inoculationand noionic surfactants on degradation of polyaromatic hydrocarbonsin soil. Environ. ¹oxicol. Chem. 16, 631}637.

Maliszewska-Kordybach, B. (1993). The e!ect of temperature on the rateof disappearance of polycyclic aromatic hydrocarbons from soils.Environ. Pollut. 79, 15}20.

Manilal, V. B., and Alexander, M. (1991). Factors a!ecting the microbialdegradation of phenanthrene in soil. Appl. Microbiol. Biot. 35, 401}405.

McCarthy, L., Ozburn, G. W., Smith, A. D., and Dixon, D. G. (1992).Toxicokinetic modeling of mixtures of organic chemicals. Environ.¹oxicol. Chem. 11, 1037}1047.

Means, J. C., Wood, S. G., Hassett, J. J., and Banwart, W. L. (1980).Sorption of polynuclear aromatic hydrocarbons by sediments and soils.Environ. Sci. ¹ech. 14, 1524}1528.

Mihelcic, J. R., and Luthy, R. G. (1988). Degradation of polycyclic aromatichydrocarbons under various redox conditions in soil-water systems.Appl. Environ. Microbiol. 54, 1182}1187.

Park, K. S., Sims, R. C., Dupont, R. R., Doucette, W. J., and Matthews,J. E. (1990). Fate of PAH compounds in two soil types: In#uence ofvolatilization, abiotic loss and biological activity. Environ. ¹oxicol.Chem. 9, 187}195.

Petersen, H. (1978). Some properties of two high gradient extractors forsoil microarthropods. Natura Jutlandica 20, 95}121.

Posthuma, L., Baerselman, R., Van Veen, R. P. M., and Dirven-VanBremen, E. M. (1997). Single and joint e!ects of copper and zinc onreproduction of Enchytraeus crypticus in relation to sorption of metalsin soils. Ecotoxicol Environ. Saf. 38, 108}121.

SAS Institute (1999). &&SAS/STAT User's Guide, Version 8.'' SASOnlineDoc, Cary, NC.

Sims, R. C., and Overcash, M. R. (1983). Fate of polynuclear compounds(PNAs) in soil}plant system. Residue Rev. 88, 1}68.

Sverdrup, L. E., Kelley, A. E., Krogh, P. H., Nielsen, T., Jensen, J., Scott-Fordsmand, J. J., and Stenersen, J. (2001). E!ects of eight polycyclicaromatic compounds on the survival and reproduction of the springtailFolsomia ,metaria L. (Collembola, Isotomidae). Environ. ¹oxicol. Chem.20, 1332}1338.

Tiehm, A. (1994). Degradation of polycyclic aromatic hydrocarbons inthe presence of synthetic surfactants. Appl. Environ. Microbiol. 60,258}263.

Van Gestel, C. A. M., and Hensbergen, P. J. (1997). Interaction of Cd andZn toxicity for Folsomia candida in relation to bioavailability in soil.Environ. ¹oxicol. Chem. 16, 1177}1186.

Warne, M. J., Connel, D. W., et al. (1989). Prediction of the toxicity ofmixtures of shale oil components. Ecotoxicol. Environ. Saf. 18, 121}128.

Wiles, J. A., and Krogh, P. H. (1998). Tests with the collembolans Isotomaviridis, Folsomia candida and Folsomia ,metaria. In Handbook of SoilInvertebrate ¹oxicity ¹ests (H. L+kke and C. A. M. van Gestel, Eds.),pp. 131}156. Wiley, Ba$ns, West Sussex, England.

Willumsen, P. A. (1998). Surfactant-Enhanced Biodegradation of PolycyclicAromatic Hydrocarbons. Ph.D. thesis. National Environmental ResearchInstitute, Roskilde, Denmark.

Wilson, S. C., and Jones, K. C. (1993). Bioremediation of soil contaminatedwith polynuclear aromatic hydrocarbons (PAHs)*A review. Environ.Pollut. 81, 229}249.