Pestic. Sci. 1998, 52, 119È125

The Fate of Imidacloprid in Tobacco Smoke ofCigarettes Made from Imidacloprid-TreatedTobaccoTerrence Clark,1 Ernst Ewald Ro� mer2 & Georg Schepers2*Kau�mann,11 Bayer AG, Crop Protection Development, Institute for Metabolism Research and Residue Analysis,D-51368 Leverkusen-Bayerwerk, Germany2 INBIFO Institut fu� r biologische Forschung GmbH, 3, D-51149 Ko� ln, GermanyFuggerstra�e

(Received 4 July 1997 ; accepted 27 August 1997)

Abstract : The residues and metabolites of radiolabelled imidacloprid [1-(6-chloropyridin-3-ylmethyl)-N-nitroimidazolidin-2-ylideneamine], formulated as awettable powder containing 250 g kg~1 active ingredient diluted with water andadministered to tobacco plants, were studied in sidestream and mainstreamsmoke, in the ash and butts after smoking cigarettes. An almost completerecovery of radioactivity (93É5%) was achieved. The highest amounts of radioac-tivity were found in the butts and sidestream smoke. The two dominant com-pounds identiÐed after smoking were unchanged parent compound and carbondioxide. A total of 76% of the recovered radioactivity was identiÐed. 1998(SCI.

Pestic. Sci., 52, 119È125, 1998

Key words : imidacloprid ; metabolism; cigarettes ; smoke ; chloronicotinyl ;radiolabelling

1 INTRODUCTION

Imidacloprid [1-(6-chloropyridin-3-ylmethyl)-N-nitro-imidazolidin-2-ylideneamine] is a relatively new, highlye†ective systemic and contact insecticide exhibiting lowmammalian toxicity. It is currently the most importantrepresentative of a new type of insecticide, namely thechloronicotinyl class of compounds. These compoundsinterfere with the signal transmission in the insect nervesystem acting at a nicotinic acetylcholine receptor(nAChR).1h3 Imidacloprid controls a wide range ofinsect pests including aphids, whiteÑies, thrips, scales,leafhoppers and planthoppers.

Imidacloprid has been registered for use in more than70 countries worldwide and has a favourable safety andtoxicology proÐle. Its metabolism in a wide variety ofcrops, such as tomatoes, potatoes, apples, corn, cotton,rice and others, has already been investigated andreported for registration purposes.4

* To whom correspondence should be addressed.

In order to check that pesticides have no adversee†ect on the taste of cigarettes from the smokerÏs pointof view, Ñavour studies on cigarettes produced frompesticide-treated tobacco have to be conducted bycigarette manufacturers. Prior to initiating these studies,it is prudent to ensure that undesirable pesticide-derivedpyrolysis or combustion products are not transferred inrelevant amounts in smoke. Within that framework, theobjective of this study was to investigate the fate ofimidacloprid after soil drench and spray application totobacco plants with regard to the transfer of imid-acloprid, its metabolites and pyrolysis products into thesmoke of cigarettes made from leaves of theseimidacloprid-treated plants.

2 MATERIAL AND METHODS

2.1 Test chemicals and reference compounds

[pyridinyl-14C-methyl]imidacloprid with a speciÐcradioactivity of 1É15 MBq mg~1 and a radiochemical

1191998 SCI. Pestic. Sci. 0031-613X/98/$17.50. Printed in Great Britain(

120 T errence Clark et al.

purity of 99% was supplied by Bayer AG, Elberfeld,Germany and, prior to application, it was formulated asa wettable powder containing 250 g kg~1 active ingre-dient (WP 25). Additional nonradiolabelled referencecompounds supplied by Bayer AG, Monheim, Germanywere imidacloprid, the guanidine 2 (1-(6-chloropyridin-3-ylmethyl)imidazolidin-2-ylideneamine), the urea 3( 1 - ( 6 - chloropyridin -3 -ylmethyl ) imidazolidin -2 -one ) ,the oleÐn 4 (1-(6-chloropyridin-3-ylmethyl)-N-nitro-imidazolin-2-ylideneamine), the nitrosimine 5 (1-(6-chloropyridin - 3 - ylmethyl ) - N - nitrosoimidazolidin - 2 -ylideneamine) and the 4-hydroxy compound 6 (1-(6-chloropyridin-3-ylmethyl)-2-(nitroimino)imidazolidin-4-ol) (Fig. 1).

2.2 Application

One soil application and three spray applications weremade to a total of 11 tobacco plants. Owing to legalconstraints on the use of radioactivity within the glass-house facility, two separate experiments were conductedin parallel. At harvest, all sampled leaves from bothexperiments were combined and treated as one sample.The soil applications were made on 16 September, theÐrst spray applications on 26 October, the second on 2November and the third on 8 November, all in 1993.The stability of the parent compound in the applicationsolutions was conÐrmed by thin-layer chromatography(TLC) before and after each application, both for thesoil application and the three spray applications.

2.2.1 Soil applicationThe amount applied to the soil around each of the 11tobacco plants individually grown in 15-litre pots was80 mg of WP 25 prepared from radiolabelled imid-

Fig. 1. Structures of parent and reference compounds.

acloprid in 120 ml water, i.e. 20 mg active ingredient(AI), which is equivalent to 23É0 MBq.

2.2.2 Spray applicationIn order to ensure that, at the time of harvest, alltreated leaves were, as far as possible, at the same matu-rity state, the four leaves at the base of the plants werenot treated. The eight leaves immediately above thesewere all individually sprayed using a retouch micro-sprayer (EFBE airbrush, Boldt, Hannover, Germany).The leaves above these eight were not treated.

The same procedure was followed for each applica-tion : water (96 ml) was added to 134É4 mg of the WP25 prepared from radiolabelled imidacloprid. Analiquot (1 ml) of the resulting suspension was applied toeach individual leaf. This was equivalent to spraying0É35 mg AI per leaf. For the three applications, thisyielded a total amount applied per leaf of 1É05 mg AI,and per plant 8É4 mg AI, which was equivalent to9É66 MBq.

The total amount applied to the soil and the 11plants amounted to 1249É6 mg WP 25, 312É4 mg AIwhich was equivalent to 359É26 MBq.

2.3 Sampling and conditioning

A total of 88 leaves was harvested two weeks (22November 93) after the last spray application, of which24 leaves (1380 g fresh weight) at the same stage ofmaturity were selected for air-curing and preparation ofcigarettes for the smoking experiments.

2.4 Preparation of cut tobacco and cigarettes

Lamina of air-cured (four months, 22(^1)¡C, 50(^5)%RH, air Ñow 1 m s~1) and conditioned (approx. threedays, 22(^1)¡C, 85% RH) tobacco leaves were cut witha modiÐed tissue chopper (McIllwain tissue chopper,The Mickle Laboratory Engineering Company, Gom-shall, UK) at 35 cuts per inch.

Commercially available Ðlter sleeves were Ðlled withapproximately 0É8 g of the cut tobacco and the Ðltercigarettes conditioned (approximately two days,22(^1)¡C, 65% RH). The tobacco content of each singlecigarette was determined by weight. Before loading thesmoking machine, the Ðlters were cut o†, resulting inÐlterless cigarettes of approx. 68 mm length.

2.5 Smoking

Each of 10 smoking experiments was performed usingfour cigarettes which were consecutively smoked on asingle-port total-recovery smoking machine (PhilipMorris, Richmond, VA, USA) in as close conformity aspossible with ISO 3402,5 ISO 33086 and ISO 4387.7After lighting with a halogen spot lamp, the cigarettes

Fate of imidacloprid in smoke from treated tobacco 121

were smoked down to a butt length of 23 mm, the pu†volume being 35 ml. The cigarettes were pu†ed everyminute.

The smoking machine worked essentially as pre-viously described.8 The cigarette was placed in thecigarette holder of a glass smoking chamber which wassupplied with air or nitrogen from pressurized cylinders.In each experiment, the Ñow rates for mainstreamsmoke (MS) and sidestream smoke (SS) were adjustedwith the aid of Ñowmeters to 1050 and 1500 ml min~1,respectively, before lighting the Ðrst cigarette, and theÑows were controlled by computer-driven solenoidvalves. During the 2-s pu† duration, the MS path wasopened and the SS path closed. The MS passed througha glass-Ðbre Ðlter (Cambridge Filter Pads, 44 mm diam-eter, Filtrona, Reinbek, Germany) and the resulting gas/vapour phase of the MS through various traps asdescribed in Section 2.6.2. For the time period betweentwo pu†s (58 s) the MS path was closed and the airpassed around the cigarette collecting the SS from thesmoking chamber into the SS path of the smokingmachine. The SS passed through an identical series ofÐlter and traps to the MS. Extensive Ñushing of the gas/vapour duct of the MS path between the pu†s was per-formed with nitrogen and of the SS path during puffingwith air. After the last pu† of each cigarette, thecigarette was extinguished by Ñushing the smokingchamber with nitrogen, after which the whole systemwas extensively Ñushed with nitrogen for 4 min.

2.6 Determination of total radioactive residue (TRR)and recovered radioactivity (RR)

Radioactivity in liquid samples was quantiÐed by liquidscintillation counting (LSC) (Tri-Carb 2550 TR,Canberra-Packard, Frankfurt a.M., Germany). Radio-activity in solid samples was quantiÐed after com-bustion of the samples or aliquots thereof in a sampleoxidizer (OX 500, Harvey Instruments via ZinsserAnalytic, Frankfurt a.M., Germany), and the resultingcarbon dioxide was trapped in Oxysolve C-400 (ZinsserAnalytic, Frankfurt a.M., Germany). The total radioac-tive residues (TRR) were calculated on a mgimidacloprid-equivalent per kg tobacco basis.

The recovered radioactivity (RR) was calculated asthe sum of the total radioactivity found in the butt, inthe ash, in the trapped smoke fractions of both MS andSS, and that remaining in the apparatus. This value wasnormalized to the total radioactivity in all cigarettes,calculated from the amount of tobacco in thesecigarettes and the TRR determined in the cigarettetobacco (see Section 2.6.1).

2.6.1 Determination of T RR of the tobacco in thecigarettesAll the tobacco in each of Ðve cigarettes was removedfrom the paper and individually homogenized in a

blender. The radioactivity in three aliquots(approximately 50 mg each) of the homogenizedtobacco of each cigarette was determined after sampleoxidation.

2.6.2 Determination of T RR in butts, ash and smokefractionsThe four butts (including paper) of each individualexperiment were combined, weighed and homogenized.The ash from the four cigarettes in each individualexperiment was combined, weighed and thoroughlymixed. The TRR in butts and ash was determined aftersample oxidation.

The Ðlter pads containing the particle phases of theMS and SS of the four cigarettes in each experimentwere removed from the respective Ðlter holders, placedinto glass bottles containing acetone (20 ml), and soni-cated for 20 min. Each Ðlter was washed with acetone(3] 10 ml). The acetone suspensions of the particlephases were combined with the respective washing solu-tions in 50-ml volumetric Ñasks and Ðlled up to themark with acetone. The radioactivity in the acetonesolution was determined by LSC in Quicksafe N(Zinsser Analytic, Frankfurt a.M., Germany). The radio-activity in the air-dried extracted Ðlters was determinedafter sample oxidation.

The traps for polar compounds in the gas/vapourphase of MS and SS contained dilute hydrochloric acidsolution (170 ml, 4 mM). After smoking the fourcigarettes of each experiment, the radioactivity in thetraps was determined by LSC in Aquasafe 300 Plus(Zinsser Analytic, Frankfurt a.M., Germany).

The traps for carbon monoxide in the gas/vapourphase of MS and SS contained copper (I) chloride(0É8 M) dissolved in hydrochloric acid (170 ml ; 4 M).After smoking the four cigarettes of each experiment,the radioactivity in the traps was determined by LSC ina gel formed from sample aliquots, Quicksafe A (ZinsserAnalytic, Frankfurt a.M., Germany), and water.

The traps for apolar compounds in the gas/vapourphase of MS and SS contained isooctane (170 ml). Aftersmoking the four cigarettes of each experiment, theradioactivity was determined by LSC in Quicksafe N(Zinsser Analytic, Frankfurt a.M., Germany).

The traps for carbon dioxide in the gas/vapour phaseof MS and SS contained aqueous sodium hydroxide(170 ml, 1 M). After smoking the four cigarettes of eachexperiment, the radioactivity was determined by LSC inAquasafe 300 Plus (Zinsser Analytic, Frankfurt a.M.,Germany). In order to conÐrm that the radioactivity inthe sodium hydroxide traps originated only fromtrapped carbon dioxide, representative aliquots of 10 mlwere taken and any trapped carbon dioxide was pre-cipitated as barium carbonate by adding aqueousbarium chloride (20 ml ; 0É5 M) containing ammoniumchloride (0É4 M). The precipitate was Ðltered o† undersuction. The precipitate on the Ðlter was washed with

122 T errence Clark et al.

barium chloride/ammonium chloride solution (10 ml).This washing solution was combined with the Ðltrate,thus ensuring that the precipitation was complete.Radioactivity in the Ðltrate was determined by LSC.

At the end of experiments 5 and 10, the total-recoverysmoking machine was extensively washed. Washing wasperformed separately for the MS and SS paths. Allducts, tubings, Ðlter holders, the smoking chamber aswell as the trapping bottles were washed using meth-anol, except the trapping bottles for polar compounds,carbon monoxide and carbon dioxide, which werewashed with water. The respective washing solutionsobtained after experiment 5 were combined with thoseafter experiment 10. It was considered that the meth-anolic washing solution represented particle phasedeposits, whereas the aqueous washing solution rep-resented deposits from the gas/vapour phase. For eachof the four combined washing solutions, radioactivitywas determined by LSC. It was assumed that one-tenth of this radioactivity remained in the apparatusduring one experiment ; this activity was taken intoaccount when calculating the recovery of each singleexperiment.

2.7 Quantitation of imidacloprid, its metabolites and itspyrolysis products by two-dimensional TLC

The homogenized tobacco from the butts of the Ðrst Ðveexperiments was extensively extracted by maceration infour successive steps : with methanol ] water (20 ml ;1 ] 1, by volume), with methanol (15 ml), and withdichloromethane (2] 15 ml). After the last extractionstep, the suspension was Ðltered under suction and theextracted butt tobacco was washed on the Ðlter withdichloromethane (5 ml) and methanol (5 ml). Thedecanted supernatants and wash Ñuids were combined.The biphasic extraction Ñuid was Ðlled up with meth-anol until a single phase was obtained. The TRR wasdetermined in the extract as well as in the air-driedextracted Ðlter. The extract was used for TLC analysis.

The acetone extracts of the MS particle Ðlters of theÐrst Ðve experiments were combined and the pooledsolution analyzed by TLC. This was similarly repeatedfor the SS particle Ðlter. The methanol washing of theSS path of the apparatus was used as it was.

Two-dimensional TLC was performed using silica gelplates (Kieselgel 60 thickness 0É25 mm,F254 ,20 ] 20 cm; Merck, Darmstadt, Germany) after spotapplication. The solvent systems were as follows : A : 1stdimension : chloroform ] methanol ] acetic acid]water (65] 25 ] 3É5 ] 3É5, by volume), 2nd dimension :butan-1-ol ] acetic acid] water (80] 20 ] 20, byvolume) ; B : 1st dimension : ethyl acetate] toluene] methanol ] acetic acid (80] 20 ] 20 ] 1, byvolume), 2nd dimension : ethyl acetate] propan-2-ol ]water(65 ] 23 ] 12, by volume). The plates were care-

fully dried between developments.The evaluation of the TLC plates was performed after

exposure for three days to a phosphor screen by radio-metric scanning (BAS 2000 Bioimaging Analyser, Fujivia Raytest, Straubenhardt, Germany). The quantitativedistribution of metabolites was determined by using the“TinaÏ software, version 2.08a (Raytest). Co-chromatographed reference compounds were visualizedon the TLC plates by UV light at 254 nm.

3 RESULTS AND DISCUSSION

3.1 Tobacco and cigarettes

The TRR in the air-cured tobacco of the cigarettes was126É4(^7É5) mg imidacloprid equivalents kg~1(mean^ SD). The cigarettes used in the experimentshad a tobacco weight of 0É782(^0É065) g (mean ^ SD).A mean number of 8É5 pu†s (SD: 1É2) was obtained bysmoking the cigarettes following as closely as possiblethe pertinent ISO standards.

3.2 Radioactivity in butt, ash and smoke fractions

From the total radioactivity in the air-cured tobacco,93É5% was recovered after cigarette smoking. The nor-malized results for the individual smoke fractions aregiven in Table 1. Of the total RR (i.e. 100%), the buttcontained 39É3% and the ash 5É0%. Radioactivity in theash represents most probably SS components whichwere adsorbed on the large surface of the ash. A total of34É6% of RR was identiÐed as carbon dioxide and 0É5%as carbon monoxide. From the remaining RR, 5É3%was found in the MS (0É2% in gas/vapour phase and5É1% in particle phase) and 15É3% in the SS (7É0% ingas/vapour phase and 8É3% in particle phase).

The TRR in the MS was 10É2 mg kg~1 (6É5 mg kg~1in particle and 3É7 mg kg~1 in gas/vapour phase),60É3 mg kg~1 in the SS (10É5 mg kg~1 in particle and49É8 mg kg~1 in gas/vapour phase), 49É6 mg kg~1 inthe butt and 6É3 mg kg~1 in the ash.

3.3 Quantitation and identiÐcation of imidacloprid, itsmetabolites and pyrolysis products

3.3.1 ButtsThe butts were extracted with an efficiency in recoveryof radioactivity of 95É6% (TRR: 47É4 mg kg~1) ; 4É4%of the radioactivity (TRR: 2É2 mg kg~1) remained unex-tracted. The major compound identiÐed by two-dimensional TLC was unchanged imidacloprid (Fig. 2),accounting for 72% of the radioactivity in the butt(Table 2). Furthermore, compounds 2, 3, 4, 5 and 6 werealso identiÐed by two-dimensional TLC in the buttextract, but in minor amounts. These results reÑected

Fate of imidacloprid in smoke from treated tobacco 123

TABLE 1Distribution of Radioactivity in the Di†erent Smoke Frac-tions Obtained after Smoking Cigarettes Prepared from

Imidacloprid-Treated Tobacco

Recovered radioactivityaFraction (% of total)

Mainstream smokeParticle phaseFilter extract 4É6 (^1É5)Not-extractable 0É2 (^0É2)Wash 0É3b

Subtotal 5É1 (^1É6)Gas/vapour phase trapsDilute acid 0É2cIsooctane \ 0É05cCuCl/HCl (for CO) 0É1cNaOH (for CO2) 2É7 (^0É4)Wash \ 0É05b

Subtotal 2É9 (^0É4)Sum 8É0 (^2É0)

Sidestream smokeParticle phaseFilter extract 5É8 (^0É9)Not-extractable 0É3 (^0É1)Wash 2É2b

Subtotal 8É3 (^1É0)Gas/vapour phase trapsDilute acid 6É7 (^0É6)Isooctane 0É2cCuCl/HCl (for CO) 0É4 (^0É1)NaOH (for CO2) 31É9 (^4É1)Wash \0É05b

Subtotal 39É4 (^4É8)Sum 47É7 (^5É6)

Butts 39É3 (^6É5)Ash 5É0 (^1É3)

Total 100É0 (^6É7)

a Normalized results ; mean(^SD) of 10 experiments. Recov-ered radioactivity (93É5%) is based on radioactivity incigarettes. Total radioactive residue is equal to 126É4 mgimidacloprid-equivalents kg~1 tobacco. Deviations in the lastdigit of sums may arise from rounding.b n \ 1c SD \ 0É05

those obtained in other plant metabolism studies forregistration, a summary of which has been publishedpreviously.4

3.3.2 Mainstream smokeThe particle Ðlters in the MS path of the smokingmachine were extensively extracted with a mean extrac-tion efficiency of radioactivity of 95É3% (TRR:5É8 mg kg~1) ; 4É7% of the radioactivity (TRR:0É3 mg kg~1) remained unextracted. The radioactivityrecovered by washing the MS ducts and Ðlter holder ofthe apparatus accounted for 0É3% (TRR: 0É4 mg kg~1).

Fig. 2. Two-dimensional TLC, radiometric evaluation of thebutt extract. TLC was performed using solvent system B.

Numbers refer to compounds shown in Fig. 1.

This fraction was not further analyzed. The major com-pounds identiÐed by two-dimensional TLC were theurea compound 3 and unchanged imidacloprid (31 and23% of the radioactivity, respectively, see Fig. 3 andTable 2).

No further analysis of polar compounds (trappedin dilute acid solution, 0É2% of RR) and of apolarcompounds (trapped in isooctane, 0É04% of RR)was performed. The radioactivity trapped incopper(I)chloride/hydrochloric acid (0É06% of RR) wasassumed to be carbon monoxide, as this reagent is con-sidered to be speciÐc for trapping that compound.9 The

Fig. 3. Two-dimensional TLC, radiometric evaluation of theMS particle Ðlter extract. TLC was performed using solvent

system B. Numbers refer to compounds shown in Fig. 1.

124 T errence Clark et al.

TABLE 2Distribution of Imidacloprid, Its Metabolites and Pyrolysis Products in the Di†erent Smoke Fractions

Recovered radioactivity (% of total)

MS SS

Particle phase Particle phase

Compound Filter W ash Filter W ash Butt Ash SumbGas/V apour Gas/V apour

phase phase

Imidacloprid (1) 1É2 È È 0É8 0É2 È 28É3 È 30É4Guanidine compound (2) n.d.a È È n.d. n.d. È 1É3 È 1É3Urea compound (3) 1É6 È È 2É0 0É3 È 1É2 È 5É0OleÐn compound (4) n.d. È È n.d. n.d. È 2É2 È 2É2Nitrosimine compound (5) n.d. È È n.d. n.d. È 0É3 È 0É34-Hydroxy compound (6) n.d. È È n.d. n.d. È 1É6 È 1É6CO È È 0É1 È È 0É4 È È 0É5CO2 È È 2É7 È È 31É9 È È 34É6UnidentiÐed 2É0 0É3 0É2 3É3 1É7 7É0 4É3 5É0 23É8

Totalb 4É8 0É3 3É0 6É1 2É2 39É3 39É2 5É0 100É0

a n.d. : not detected.b Deviations in the last digit of sums may arise due to rounding.

radioactivity trapped in sodium hydroxide (2É7% of RR)was precipitated with barium chloride to an extent of98É8%, thus indicating that almost all of the trappedradioactivity was carbon dioxide.

3.3.3 Sidestream smokeThe particle Ðlters in the SS path of the smokingmachine were extensively extracted with a mean extrac-tion efficiency of radioactivity of 94É6% (TRR:7É3 mg kg~1) ; 5É4% of radioactivity (TRR:0É4 mg kg~1) remained unextracted. The major com-pounds identiÐed by two-dimensional TLC were theurea compound, 3, and unchanged imidacloprid (24 and9% of the total radioactivity in the SS particle phase,respectively, Table 2). The radioactivity recovered bywashing o† the condensate from the smoking chamber,the ducts to the Ðlter, and the Ðlter holder of the SS was2É2% of RR. In this wash fraction, only unchangedimidacloprid (8%) and compound 3 (14%), representing2 and 4%, respectively, of the total radioactivity in theSS particle phase (Table 2), was identiÐed by two-dimensional TLC.

The trapping solution for polar compounds (diluteacid solution) contained 6É7% of the RR. Analysis ofthis fraction by two-dimensional TLC yielded onlydi†use radioactivity (not in peaks ; data not shown) andwas therefore not investigated further. No furtheranalysis of apolar compounds (trapped in isooctane,0É2% of RR) was performed due to the low amount ofradioactivity. The radioactivity trapped incopper(I)chloride/hydrochloric acid (0É4% of RR) wasassumed to be carbon monoxide. A major portion ofthe radioactivity was trapped in sodium hydroxide(32% of RR). By precipitation with barium chloride it

was shown that all of the trapped compound (99.8%)was carbon dioxide.

4 CONCLUSIONS

Using the smoking apparatus described in this pub-lication, an almost complete recovery of radioactivity(93É5%) was achieved. The amount of carbon dioxidepresent in the SS represented almost 60% of the radio-activity found in total in MS and SS and was by far themain component of the smoke fractions. This indicatedthat imidacloprid was primarily degraded to carbondioxide during smoking. The only other componentsidentiÐed in the smoke fractions were unchanged parentcompound, the urea compound, 3, and small amountsof carbon monoxide. Further metabolites identiÐed intrace amounts in the butts have also been previouslyidentiÐed in a variety of plant metabolic studies. Intotal, 76% of the recovered radioactivity (TRR:96.1 mg kg~1) was identiÐed.

ACKNOWLEDGEMENTS

We gratefully acknowledge the skilful technical assist-ance of J. Pu� tz and P. Colianni (Bayer AG) for thein-life part, and G. Fleger, B. Mende, D. Demetriou andF. Radtke (INBIFO GmbH) for the smoking and ana-lytical part of the study.

REFERENCES

1. Liu, M.-Y., Latli, B. & Casida, J. E., Imidacloprid bindingsite in Musca nicotinic acetylcholine receptor : interactions

Fate of imidacloprid in smoke from treated tobacco 125

with physostigmine and a variety of nicotinic agonists withchloropyridyl and chlorothiazolyl substituents. Pestic.Biochem. Physiol., 52 (1995) 170È81.

2. Bai, D., Lummis, S. C. R., Leicht, W., Breer, H. & Sattelle,D. B., Actions of imidacloprid and a related nitromethyleneon cholinergic receptors of an identiÐed insect motorneurone. Pestic. Sci., 33 (1991) 197È204.

3. Liu, M.-Y. & Casida, J. E., High affinity binding of [3H]imidacloprid in the insect acetylcholine receptor. Pestic.Biochem. Physiol., 46 (1993) 40È6.

4. Klein, O., The metabolism of imidacloprid in plants andsoil, Abst. 8th IUPAC International Congress of PesticideChemistry, Washington DC, 4È9 July 1994, p. 157.

5. International Organization for Standardization, Interna-tional Standard ISO 3402, T obacco and tobacco productsÈAtmospheres for conditioning and testing, 3rd edn, 1991.

6. International Organization for Standardization, Interna-tional Standard ISO 3308, Routine analytical cigarette-smoking machineÈDeÐnitions and standard conditions, 3rdedn, 1991.

7. International Organization for Standardization, Interna-tional Standard ISO 4387, CigarettesÈDetermination oftotal and nicotine-free dry particulate matter using a routineanalytical smoking machine, 2nd edn, 1991.

8. Jenkins, R. W., Clavis, M. K., Newman, R. H. & Morrell,F. A., The quantitative recovery of smoke from radioac-tively labeled cigarettes. Int. J. Appl. Radiation Isotopes, 22(1971) 691È4.

9. Budavari, S., OÏNeil, M. J., Smith, A. & Heckelmann, P. E.,T he Merck Index. An encyclopedia of chemicals, drugs andbiologicals. 11th edn. Merck & Co., Rahway NJ, 1989, p.275.