Smith RN (1982) Forensic applications of high-performance liquid chromatography. In: Saferstein R.(ed.) Forensic Science Handbook. New Jersey: PrenticeHall.

Smith RN and Vaughan CG (1976) High-pressure liquidchromatogrphy of cannabis. Quantitative analysis ofacidic and neutral cannabinoids. Journal of Chromato-graphy 129: 347}354.

Tracqui A, Kintz P andMangin P (1995) Systematic toxico-logical analysis using HPLC/DAD. Journal of ForensicSciences 40: 254}262.

Weiss J (1995) Ion Chromatography, 2nd edn. New York:VCH.

White P (1998) Crime Scene to Court. The Essentialsof Forensic Science. Cambridge: Royal Society ofChemistry.

FORENSIC TOXICOLOGY: THIN-LAYER(PLANAR) CHROMATOGRAPHY

I. OjanperaK , University of Helsinki, Helsinki, FinlandCopyright^ 2000 Academic Press

The selection of appropriate analytical methodologyfor forensic toxicological investigations depends onthe scope of the laboratory. Postmortem forensictoxicology investigates the cause of death, and conse-quently a very broad-range screening is needed todetect all potential poisons. In trafRc toxicology,only such substances are relevant which may impairthe drivers ability to control the vehicle. Dopingcontrol focuses on those substances that have beenbanned by the International Olympic Committee.Prisoners and rehabilitation clinic patients are testedfor psychotropic drugs, whereas the US MandatoryGuidelines for Federal Workplace Drug TestingPrograms are limited to the major drugs of abuse,cannabis, cocaine, amphetamine, opiates and phen-cyclidine.

Thin-layer chromatography (TLC) has found ex-tensive use in forensic toxicology since the early1960s when the famous book of Stahl made thetechnique well known. The Rrst edition of theclassic laboratory manual by AS Curry, Poison De-tection in Human Organs from 1963 (Charles C.Thomas, SpringReld, IL), still relies on paperchromatography but the second edition in 1969 util-izes TLC as a major technique for drugs. Gaschromatography (GC) and gas chromatography}mass spectrometry (GC-MS) in the 1970s, and espe-cially high performance liquid chromatography(HPLC) in the 1980s gradually began to replaceTLC but today the planar technique is having a re-naissance due to the progress in instrumentation andsoftware. From 1990 to 1996, 26% of publishedTLC applications were in the Reld of medical, clinicaland biological analysis, which also includes forensictoxicology.

The commonly recognized advantages of classicalmanual TLC are high throughput, low cost, easysample preparation and versatile visual detection pos-sibilities. Instrumental TLC extends the scope to re-producible quantitative analysis and allows the utiliz-ation of in situ UV spectral information for identiRca-tion. The main disadvantage of TLC is low chromato-graphic resolution, which can be partly overcome byinstrumental techniques. Another disadvantage isthat quantitative calibration curves are not reproduc-ible enough to be stored, making it necessary toco-analyse several standards along with samples oneach TLC plate. Most of the substances frequentlyencountered in forensic toxicology can be readilyanalysed by TLC. These include therapeutic drugs,drugs of abuse, pesticides and naturally occurringalkaloids, which are all relatively small molecularweight organic compounds with functional groupsamenable to visualization by colour reactions.

It is practical to divide the discussion of TLC inforensic toxicology into two categories, the broad-scale screening analysis and target analysis. The for-mer approach is related to the concepts of systematictoxicological analysis or general unknown, i.e. thesearch for a rational qualitative analysis strategy forhundreds of potential poisons. TLC drug screening isoften performed in urine or liver, where the drugconcentrations are higher than in the blood. In targetanalysis, the aim is speciRcally to detect and oftenalso to quantify a substance or a limited number ofsubstances.

Broad-scale Screening Analysis

Chromatographic Systems

Evaluation of systems The rational selection of TLCsystems for screening analysis differs from the opt-imization of the separation of a few-component

III /FORENSIC TOXICOLOGY: THIN-LAYER (PLANAR) CHROMATOGRAPHY 2879

Table 1 TLC systems for broad-scale toxicological screening analysis

Mobile phase Stationary phase Correction hR cF Applicationstandards

1 Chloroform}acetone Silica gel Paracetamol 15 Acidic and neutral drugs80#20 Clonazepam 35

Secobarbital 55Methylphenobarbital 70

2 Ethyl acetate Silica gel Sulfathiazole 20 Acidic and neutral drugsPhenacetin 38Salicylamide 55Secobarbital 68

3 Ethyl acetate}methanol}conc. ammonia

Silica gel Hydrochlorothiazide 11 Acidic and neutral drugs

85#10#5Sulfafurazole 33Phenacetim 52Prazepam 72

4 Methanol}water Silica gelRP 18

Diazepam 16 Acidic and neutral drugs65#35 Secobarbital 35

Phenobarbital 54Paracetamol 74

5 Methanol}water}conc. hydrochloric acid

Silica gelRP 18

Hydroxyzine 20 Basic, amphoteric andquaternary drugs

50#50#1Lignocaine 46Codeine 66Morphine 81

6 Toluene}acetone}ethanol}conc. ammonia

Silica gel Codeine 16 Basic and neutral drugs

45#45#7#3Promazine 36Clomipramine 49Cocaine 66

7 Ethyl acetate}methanol}ammonia

Silica gel Morphine 20 Basic and neutral drugs

85#10#5Codeine 35Hydroxyzine 53Trimipramine 80

8 Methanol Silica gel Codeine 20 Basic and neutral drugsTrimipramine 36Hydroxyzine 56Diazepam 82

9 Methanol}ammonia Silica gela Atropine 18 Basic and neutral drugs100#1.5 Codeine 33

Chlorprothixene 56Diazepam 75

10 Cyclohexane}toluene}diethylamine

Silica gela Codeine 6 Basic and neutral drugs

75#15#10Desipramine 20Prazepam 36Trimipramine 62

aImpregnated with 0.1 mol L1 KOH and dried.

mixture. In screening systems, the most importantfeatures are the distribution of RF values across theplate, the reproducibility of the measurement of thosevalues, and the correlation of chromatographic prop-erties between systems. The computational methodscapable of taking into account these features includediscriminating power, mean list length (MLL), in-

formation content, quotient of distribution equalityand principal component analysis. The (MLL)method has found widespread use, and it can also beused in computerized substance identiRcation. TheMLL approach is not related to the separation num-ber (SN), i.e. the number of spots that can be separ-ated by a system with a certain resolution. Table 1

2880 III /FORENSIC TOXICOLOGY: THIN-LAYER (PLANAR) CHROMATOGRAPHY

Table 2 Visualization reagents for broad-scale screening analysis

Reagent Application

Bratton-Marshall reagent (diazotation and coupling) Benzophenones (from benzodiazepines), sulfonamides7-Chloro-4-nitrobenzo-2-oxa-1,3-diazole (NBD-Cl) Amphetamines, amino acids2,6-Dibromoquinone-4-chlorimide (Gibbs reagent) Pesticides2,6-Dichlorophenol-indophenol (Tillmanns reagent) Organic acidsp-Dimethylaminobenzaldehyde (Van Urks reagent) Drugs, sulfonamides, pesticidesDragendorffs reagent Drugs and alkaloidsFast Black K salt Amphetamines, adrenergic -blocking drugs, nor-metabolitesFast Blue B salt, Fast Blue BB salt CannabinoidsFluorescamine Amphetamines, amino acids, sulfonamidesForrest reagent Phenothiazines, antidepressantsFPN reagent Phenothiazines, dibenzazepinesFurfuraldehyde Carbamates, phenothiazinesIodoplatinate, acidic Alkaloids, drugs, quaternary ammonium compundsMandelins reagent DrugsMarquis reagent DrugsMercuric chloride-diphenylcarbazone BarbituratesMercurous nitrate BarbituratesNinhydrin Amphetamines, amino acids4-(4-Nitrobenzyl)pyridine-tetraethylenepentamine PesticidesSalkowski reagent (FeCl3#H2SO4) Phenothiazines, thioxanthenes3,3,5,5-Tetramethylbenzidine, o-tolidine, after Cl2 Pesticides, acidic and neutral drugs

shows TLC systems for broad-scale toxicologicalscreening analysis, chosen partly on grounds of theMLL method, while the corresponding RF librariescan be found from the books of de Zeeuw et al. andFried and Sherma. For acidic and neutral drugs, rec-ommended combinations of systems which posseslow mutual correlation are 2 and 3, and 1 and 3, forbasic drugs 5 and 6, and 8 and 10.

RF correction In screening analysis, where RF libra-ries of hundreds of compounds are utilized, the repro-ducibility of the values is an essential factor. TLC isan open technique, and the RF values, and conse-quently the separation, are affected by environmentalfactors, such as humidity, layer activity and temper-ature. In contrast to column chromatography, the useof a single RF standard for compensating the vari-ation, corresponding to the relative retention time,may produce erroneous results. The method, whichuses three to Rve correction standards that are struc-turally close to the analytes and linear interpolationbetween the standards, is now generally accepted toobtain corrected RF values (hRcF). Table 1 indicatesthe correction standards chosen for the screeningsystems listed.

Identi\cation

Migration distance By carefully adjusting the chro-matographic and environmental conditions it is pos-sible to obtain reproducible results with precoatedplates. Reversed-phase (RP) layers show more batch-to-batch variation than silica gel but RP separations,

using aqueous mobile phases, are less dependent onhumidity. The RF and hR

cF values can be determined

manually or by using a scanning densitometer. Thecorrection of RF values makes it possible to obtainreproducible values in varying conditions and allowsthe use of the large hRcF libraries even in interlabora-tory use.

Commercial software are available that utilizes theconcept of hRcF for identiRcation. Chrom TOX(Merck Tox Screening System, Merck, Darmstadt,Germany) is statistical search software that utilizesthe MLL method for identiRcation by RF values anddigitally coded colour reactions, giving a hit list ofcandidates with probability values. The software isalso capable of adding information from other ana-lytical techniques, such as retention indices from GC,molecular weights from MS and UV spectra fromHPLC. A drawback is that the TLC data have to befed manually. CATS software (Camag, Muttenz,Switzerland) combines instrumental densitometricevaluation of chromatographic plates with substanceidentiRcation by hRcF values and in situ UV spectra(see below).

Visualization reagents The possibility of using vis-ualization reagents for the detection and identi-Rcation of fractions is a unique feature of TLC.Post-chromatography derivatization by spraying ordipping has been used more extensively than pre-chromatography derivatization. The limits of detec-tion by colour reactions generally range from 0.1to 1 g per fraction and by Suorescence reactions,

III /FORENSIC TOXICOLOGY: THIN-LAYER (PLANAR) CHROMATOGRAPHY 2881

Method: C:CAMAGDATA}SC3KRIMSPEC.PAMRaw data: C:CAMAGDATA}SC3KY140798.DFSLibrary: C:CAMAGKRIM1.SCL

Track 15, Analysis n: 2561Peak 5, Measured hR Cf : 34, Area: 9682.6

No. Substance name Diff Correlation

1 Clozapine 2 0.9844152 Thiothixene !9 0.9487593 Norchlorprothixene 0 0.9156274 Flupenthixol !1 0.9024115 Olanzapine !1 0.8910466 Norlevomepromazine !3 0.877048

Confirmation: necessary not necessaryHit *** confirmed by A:*** B:***

Method: C:CAMAGDATA}SC3RPSPEC.PAMRaw data: C:CAMAGDATA}SC3RY140798.DFSLibrary: C:CAMAGRP1.SCL

Track 15, Analysis n: 2561Peak 2, Measured hR cf : 46, Area: 13178.0

No. Substance name Diff Correlation

1 Clozapine 3 0.9404162 Clothiapine !9 0.9376393 Molindone !9 0.8938624 Brucine !4 0.8685695 Apomorphine 2 0.8572986 Metoclopramide !6 0.855726

Confirmation: necessary not necessaryHit *** confirmed by A:*** B:***

Figure 1 Identification hit lists produced by CATS software(Camag) for an analyte fraction on two TLC systems in broad-scale screening analysis for drugs in liver. The reports wereobtained by comparing the hR cF values (window$9 units) andin situ UV spectrum correlation against a library of 325 drugsubstances on each system.

especially using pre-chromatography derivatization,20}100 ng per fraction. The visualization reactionscan be divided into class and substance selective.In visualization sequences, several reagents can beoversprayed one after another to amplify the amountof information obtained from a single plate. Anexample of such sequence for basic drugs is nin-hydrin, FPN (FeCl3#HClO4#HNO3) reagent,Dragendorffs reagent and acidiRed iodoplatinate.Table 2 lists reagents that are commonly used inforensic toxicology.

In situ spectra Earlier it was common to scrape offa separated fraction from the TLC plate and submit itto a further spectrometric or spot test analysis. Todayit is possible to measure the in situ UV spectrum ofa fraction and compare this with stored spectrumlibraries for identiRcation. Representative spectra cangenerally be obtained from well-separated fractionswith substance amounts over 0.5 g on standard TLCplates and over 0.1 g on high performance thin-layerchromatography (HPTLC) plates, although theselimits depend on the shape of the fraction and on theabsorption characteristics of the compound in ques-tion. The upper limit is not a problem as the reSec-tance saturates after certain level and the spectraremain practically the same. The CATS softwarefrom Camag allows the complete sequence of instru-mental screening analysis, including RF correction,spectrum measurement, automated search againsthRcF/UV libraries and reporting (Figure 1).

Other spectrometric techniques than UV have beentested for the measurement of TLC fractions. In situdiffuse reSectance Fourier transform infrared (FTIR)measurements have proved to be feasible for the iden-tiRcation of drugs using a spectral region where silicagel has no strong absorption, and a commercial TLC-FTIR interface is available.

Proprietary Drug Screening Schemes

Particularly in North America, a TLC scheme calledToxi-Lab (Ansys, Irvine, CA, USA) has gained popu-larity in analytical toxicology. This product com-prises the extraction, development, visualization andinterpretation steps of analysis. The Toxi-Lab A de-tects basic and neutral substances and the Toxi-LabB detects acidic and neutral substances. The platesconsist of silica, impregnated with a vanadium saltfor detection purposes, in a glass Rbre matrix. TheToxi-GramC8 are octylsilica bonded phase plates forthe conRrmation of basic and neutral drugs. All theplates have holes at the origin for the inoculation offactory-made standard substance discs and discs con-taining the evaporated sample residues. Detection iscarried out by using a standardized four-stage

2882 III /FORENSIC TOXICOLOGY: THIN-LAYER (PLANAR) CHROMATOGRAPHY

Figure 2 Separation of (1) methamphetamine and (2) amphetamine on the TLC system 6 of Table 1.

visualization sequence, and the interpretation of thecolour patterns is performed with a help of the Toxi-Lab Drug Compendium showing indexed colourphotograms for hundreds of compounds. The se-quence consists of formaldehyde vapour#Man-delins reagent, water, Suorescence under 366 nm UVlight and modiRed Dragendorffs reagent. Thereare also procedures and tests available for speciRcsubstances and classes of substances, such as opiatesand cannabis. There is ample literature available onthe applications of Toxi-Lab to clinical and forensictoxicology.

Another TLC screening scheme, Spot Chek (Ana-lytical Bio-Chemistries, PA, USA), relies on a singlemobile phase, a set of visualization reactions, andcomputerized interpretation of the patterns. Acidic/neutral and basic drugs are developed on separateplates, and on each plate the sample is divided intotwo or three equal portions that are developed inparallel to facilitate the use of visualization reactions.The migration distance is divided into Rve RF zoneswith reference compounds. The computer programdatabase is based on nine colour reaction responsesand the plate zone locations for 243 drug substancesbut requires entry of only one TLC property to gener-ate a matching list.

Automated Multiple Development

There are currently two alternative instrumentalmeans to improve the Separation number (SN) inTLC: automated multiple development (AMD) andoverpressured layer chromatography (OPLC). InAMD, the plate is developed repeatedly in the samedirection, and each partial run goes over a longersolvent migration distance than the previous one.Each partial run uses a solvent of lower elutionstrength than the previous one and in this way a step-wise gradient is formed. SN values of up to 40}50 canbe obtained by AMD but a disadvantage is the time

required for the analysis, which may be several hours.There are no strictly forensic toxicological applica-tions of AMD in the literature but the technique hasan established position in the broad-scale screeningfor pesticides in the environment and has great poten-tial in toxicology.

Overpressured Layer Chromatography

OPLC is based on the forced Sow of the mobile phaseagainst an external pressure, which results in shortdevelopment times and decreased diffusion of theanalyte fractions, making it possible to take advant-age of longer developing distances. Silica gel platesare exclusively used in OPLC as reversed-phasechromatography has become complicated. Methoddevelopment may be laborious due to disturbing ad-sorption zones, often obtained with multicomponentmobile phases. A commercial OPLC instrument isavailable from OPLC-NIT (Budapest, Hungary).OPLC has found use in the separation of closelyrelated compounds in a particular pharmacologicalcategory or compounds originating from a particularbotanical source. Two complementary OPLC systemshave been developed for broad-scale screeningfor basic and neutral drugs with SN values close to30, which is more than twice the values obtainedtypically with ordinary TLC: trichloroethylene}methylethylketone}n-butanol}acetic acid}water17#8#25#6#4, and butyl acetate}ethanol}tripropylamine}water 85#9.25#5#0.75, withlayer pre-saturation.

Target Analysis

Drugs of Abuse

Amphetamines and related stimulants, especiallyamphetamine, methamphetamine and 3,4-methyl-enedioxymethamphetamine (MDMA, an Ecstasy

III /FORENSIC TOXICOLOGY: THIN-LAYER (PLANAR) CHROMATOGRAPHY 2883

Figure 3 The in situ UV spectra of (A) amphetamine,(B) methamphetamine and (C) methylenedioxymethamph-etamine (MDMA). Amphetamine and methamphetamine havesimilar spectra but they can be differentiated, e.g. by using theFast Black K reagent.

component), can be separated by a variety of TLCsystems, such as those in Table 1 (Figure 2), andsensitively detected as Suorescent derivatives of, forexample, Suorescamine or NBD-Cl (4-chloro-7-nitro-2,1,3-benzoxadiazole). Ninhydrin is a tradi-tional reagent for amphetamines. Fast Black K saltreagent is capable of differentiating aliphatic primaryand secondary amines, giving violet and orange-redcolours, respectively, while tertiary amines donot react. Thus amphetamine and methamphetamine,or MDMA and 3,4-methylenedioxyamphetamine,are readily separated. Amphetamines possess poorUV characteristics, so they cannot be analysed byUV densitometric methods at low levels withoutderivatization. However, the methylenedioxy-derivatives can be easily recognized by their UVspectra (Figure 3). Despite the low limits of detectionobtained with pure amphetamine-like substances, thelimits of detection in urine are of the order of250}500 ng mL1 which is close to the standardimmunoassay cutoff value (300 ng mL1).

The analysis of the main urinary cannabinoid, 11-nor-delta9-tetrahydrocannabinol-9-carboxylic acid(THCA), is usually carried out with dedicated TLCsystems, such as ethyl acetate}methanol}water}conc.ammonia 12#5#0.5#1. The detection of THCAis performed with various diazonium salts, such asFast Blue B salt, Fast Blue BB salt and Fast Blue RRsalt. Detection limits of 2}10 ng mL1 can be ob-tained for THCA, and these concentrations comparefavourably with the standard immunoassay cutofflevel of 20 ng mL1.

Screening for cocaine is usually based on the detec-tion of its metabolite benzoylecgonine (BE) in urine.The combination of the following two mobilephases can be applied to the separation:methanol}chloroform}ammonia 60#60#1, andethyl acetate}methanol}water}ammonia 85#13.5#1#0.5. The detection of BE is performedwithDragendorffs reagent or Ludy Tenger reagent, fol-lowed by sulfuric acid, with the limit of detection of200 ng mL1 in urine. The standard immunoassaycutoff level is 300 ng mL1

The opiates of interest in drug abuse testing pro-grammes include the heroin metabolites, 6-mono-acetylmorphine and morphine, and codeine. Thecombination of the following two mobile phases canbe applied to the separation: ethyl acetate}isopropylalcohol}methanol}ammonia 80#15#3#8, and1,2-dichloroethane}isopropyl alcohol}methanol}ammonia 20#20#20#7. The limit of detectionfor opiates with iodoplatinate ranges from 100 to500 ng mL1 in urine, depending on the compound,while the standard immunoassay cutoff level is300 ng mL1.

Other Substances

Toxicological Analysis byMuK ller lists 1453 publishedTLC systems for potentially toxic compounds. A TLCbibliography is available from Camag on CD-ROM,listing 5500 abstracts of papers from 1982 to 1996.The biennial TLC reviews by Sherma in AnalyticalChemistry provide a wealth of information on sys-tems for individual substances in forensic toxicology.

2884 III /FORENSIC TOXICOLOGY: THIN-LAYER (PLANAR) CHROMATOGRAPHY

Status of TLC in Forensic ToxicologyLaboratory

In forensic toxicology, the unique features of TLC arebest utilized in the broad-scale screening analysis fordrugs and poisons in urine or liver samples. For thisapplication, there are equipment, dedicated softwareand reference libraries available from several manu-facturers. Compared to HPLC or capillary elec-trophoresis, TLC allows the detection of even poorlyUV-absorbing compounds using selective visualiz-ation reactions. Compared to GC or GC-MS, TLCallows the chromatography of polar compoundswithout prior derivatization. Another important ap-plication of TLC is the screening or conRrmation ofdrugs of abuse, although the supremacy of the combi-nation of immunoassay and GC-MS in this area hashindered the development of modern dedicated TLCmethods. Immunoassay screening, however, is vul-nerable to sample adulteration and high backgroundnoise. In larger, broad-service laboratories, the vari-ous techniques available today, including TLC, areconsidered complementary rather than exclusive.

See also: II/Chromatography: Thin-Layer (Planar):Modes of Development: Conventional; Modes of Develop-ment: Forced Flow, Over Pressured Layer Chromatogra-phy and Centrifugal; Spray Reagents. III/Alcohol andBiological Markers of Alcohol Abuse: Gas Chrom-atography. Clinical Chemistry: Thin-Layer (Planar)Chromatography. Clinical Diagnosis: Chromatogra-phy. Forensic Sciences: Capillary Electrophoresis.

Further Reading

Adamovics JA (ed.) (1995) Analysis of Addictive and Mis-used Drugs. New York: Marcel Dekker.

De Zeeuw RA, Franke JP, Degel F et al. (eds) (1992)Thin-layer Chromatographic RF Values of Toxi-cologically Relevant Substances on StandardizedSystems, 2nd edn. Weinheim: DFG/TIAFT,VCH.

Fried B and Sherma J (eds) (1996) Practical Thin-layerChromatography. Boca Raton: CRC Press.

Gough TA (ed.) (1991) The Analysis of Drugs of Abuse.Chichester: John Wiley.

Jork H, Funk W, and Wimmer H (1990) Thin-layerChromatography, Reagents and Detection Methods,vol. Ia. Weinheim: VCH.

Jork H, Funk W, Fischer W and Wimmer H (1994) Thin-layer Chromatography, Reagents and DetectionMethods, vol. Ib. Weinheim: VCH.

Moffat AC (ed.) (1986) Clarkes Isolation and Identi-Tcation of Drugs, 2nd edn. London: PharmaceuticalPress.

MuK ller RK (ed.) (1995) Toxicological Analysis. Leipzig:Edition Molina Press.

OjanperaK I and JaK nchen P (1994) The application ofinstrumental qualitative thin-layer chromato-graphy to drug screening. LC-GC International 7:164.

OjanperaK I, Goebel K and Vuori E (1999) Toxicologicaldrug screening by overpressured layer chromatography.Journal of Liquid Chromatography & Related Tech-nologies 22: 161.

Siek TJ, Stradling CW, McCain MW and Mehary TC(1997) Computer-aided identiRcations of thin-layerchromatographic patterns in broad spectrum drugscreening. Clinical Chemistry 43: 619.

Stahl E (1967) Du( nnschicht-Chromatographie, 2nd edn.Berlin: Springer-Verlag.

Stead AH, Gill R, Wright T, Gibbs JP and Moffat AC(1982) Standardised thin-layer chromatographic sys-tems for the identiRcation of drugs and poisons. Analyst107: 1106.

FRAGRANCES: GAS CHROMATOGRAPHY

E. R. Adlard, Delryn Burton, Wirral, UKM. Cooke, Royal Holloway University of London,Egham, Surrey, UKCopyright^ 2000 Academic Press

Introduction

It is difRcult to distinguish between aromas, Savours,taints and perfumes, because to a large extent theseare artiRcial categories that overlap. An aroma maybe deRned as the smell emanating naturally (possiblyin the process of cooking or other method of prepara-

tion) of a foodstuff or beverage. The aroma fromcoffee beans on roasting is a prime example but thereare many others. The main criterion is that the aromamaterial is essentially all in the vapour phase and thenose is responsible for sensing the aroma. A Savour isintimately related to an aroma but may contain in-volatile compounds that give rise to the sensation oftaste but in practice it is common to have a Savourwith an associated aroma. A tainted foodstuff orbeverage is often unsatisfactory for consumption be-cause there are compounds present that have an un-pleasant smell or taste. Taints may arise from natural

III /FRAGRANCES: GAS CHROMATOGRAPHY 2885