Drug abuse has always been a world problem. Recently people abuse both controlled and prescribed drugs. Opioids, cocaine, cannabis and amphetamines are the most widely abused drugs. The picture of the drug abuse problem in Scotland can be understood by an extensive study of drug prevalence and characteristics of drug use in the region. In drug-related deaths, post-mortem analysis, which includes autopsy and collecting samples for histological and toxicological analysis, is necessary to be carried out to investigate whether any drug has contributed to the cause of death. The samples which are commonly collected for toxicological analysis are blood and urine. However, when those fluids are not available, body tissues may be taken as alternative samples, such as liver and skeletal muscle. In this case, it is necessary to understand how drugs move and diffuse to these tissues after death. This phenomenon, which is known as post-mortem redistribution, may cause difficulties in the interpretation of post-mortem drug concentrations. Several studies have tried to investigate post- mortem redistribution including how drugs diffuse in the body after death. However, post-mortem redistribution is still not completely understood. This study proceeded by interrogating post-mortem data within the period of 2011-2016 held by Forensic Medicine and Science (FMS), University of Glasgow to review trends of drug-related death in the West of Scotland in which amphetamine-type stimulants (ATS), cocaine and opioids were detected. Opioids were most commonly detected (81.9%) in drug-related deaths in the West of Scotland, followed by cocaine (21.6%) and ATS (9.4%). The interrogation of post- mortem data within the period of 2007-2016 was also carried out to understand certain pathological conditions which are caused by drug abuse. From the results, it is clear that in the West of Scotland people tend to abuse multiple drugs. This trend may apply in the other part of the country and around the world. It is also clear that, even though methadone was prescribed to assist users to stop from drug addiction, especially heroin, many methadone users still abuse other drugs, as methadone was found in most of the cases in addition to other drugs. For this reason, it is important to investigate the results of drug addiction therapy and educate potential users. Subsequently, methods were adapted for analysing liver and muscle samples from the FMS in-house methods for analysing ATS and basic drugs in autopsy blood and validated according to the standard practices for method validation in forensic toxicology (SWGTOX, May 2013). All ATS drugs (amphetamine, methamphetamine, MDA, PMA, PMMA, MDMA and MDEA) and basic drugs (amitriptyline, citalopram, methadone, mirtazapine, sertraline and tramadol) gave acceptable bias, precision, linearity, recovery and stability for analysing liver and muscle samples. An experimental model for drug diffusion in tissues was studied to simulate and understand drug diffusion in humans. The diffusion rate that was used in this model is in accordance with the volume of distribution of each drug. This model is easy and simple to be carried out in any small laboratory. Blood, liver and muscle samples were analysed from 10 cases collected during the period from August 2016 to April 2017 after the next of kin signed the informed consent forms. Four basic drugs (amitriptyline, methadone, mirtazapine and sertraline) were found in 9 cases and analysed to investigate the ratios between blood, muscle, right liver and left liver. The ratios of drug concentrations of muscle:blood, left liver:right liver were found to be lower than 2. As a result, drug concentrations in muscle can be reliable for toxicological interpretation when blood is not available. The ratio of drug concentration in liver and blood has been suggested as a marker of post-mortem redistribution(1) and this study has shown that the ratio of drug concentration in liver and muscle can also be diagnostic in cases where blood is not available.

Between 1991 and 1999 war broke out across Former Yugoslavia. Thousands of people are believed to have been killed and many more were internally displaced or forcibly expelled from their countries. In 1993 the United Nations established the International Criminal Tribunal for Former Yugoslavia (ICTY) to investigate war crimes allegedly committed in the region. Its work is still ongoing. This research comprises an anthropological study of the children in Kosovo and Bosnia-Herzegovina who were killed as a direct result of war crimes perpetrated during the conflicts of the 1990s. It is based on primary forensic data collected by investigators and scientists on behalf of ICTY between the dates of 1996 and 2000. From this data, a single integrated database was created which allowed the numbers of child deaths, causes of death, demographic profiles of the deceased, and post-mortem treatment of their remains to be analysed. As well as examining these factors within each country a significant aspect of the research included comparative analysis between the crimes committed against children in Kosovo and those in Bosnia-Herzegovina. Broad comparisons were also made between adult and child data in both countries. The findings from the research were analysed within their wider socio-political context and an assessment was made of how closely the forensic evidence supported accounts from other literary sources. In its current form, the research can be used as a historical and scientific resource by those wishing to study both the events surrounding the wars in Kosovo and Bosnia-Herzegovina, and the scientific methods used by experts in the field to investigate the crimes. The methodology employed during the research, including the creation of the database, is described in detail and is directly transferrable to other studies of a similar nature. Solutions employed to address the considerable problems encountered during the construction of the database can be applied to other similarly large and unmanageable datasets. The database itself can be expanded to include the forensic evidence collected in Bosnia-Herzegovina and Kosovo since 2001, when ICTY handed over responsibility for the exhumations to local government agencies. It can also be used to examine other aspects of the wars, and adapted to analyse data from other countries. Ultimately it is hoped that this research will be of use in formulating pro-active strategies which might assist in protecting children involved in future conflicts.

This thesis studied opioids and alcohol in forensic toxicology by LC-MS/MS, which avoids time-consuming procedures involving hydrolysis, extraction and derivatisation. Initially, a method was validated for quantification of opioids and unhydrolysed polar metabolites in autopsy specimens and was used to develop procedures for interpretation of forensic toxicology results. The LC-MS/MS method developed has been validated for the simultaneous determination of 24 opioids in human whole blood, including, for the first time in human whole blood, naloxone-3-glucuronide. Although a large number of drugs of interest were included in the method, acceptance criteria for linearity, precision, and recovery for all analytes were achieved. The method was found useful for differentiating between users of heroin and other opioids, such as codeine and morphine, and for determining the survival time in deaths attributed to heroin use. Subsequently, the efficiencies of hydrolytic and non-hydrolytic methods for opioid analysis were compared for buprenorphine (BUP) analysis. The aims were to develop and validate a method for the direct determination (DM) of buprenorphine (BUP), norbuprenorphine (NBUB), buprenorphine-3-glucuronide (B3G) and norbuprenorphine-3-glucuronide (NBUP3G). This method was compared with an in house enzymatic hydrolysis method (HM) for the determination of total buprenorphine (TBUP) and norbuprenorphine (TNBUP), using real positive BUP urine case samples. A comparison between the drug and metabolite concentrations obtained by direct and hydrolysis methods was reported for the first time in this work. LC-MS analysis was also applied to paediatric plasma specimens obtained from a clinical pharmacokinetic study of intravenous and intranasal administration of diamorphine. This work was aimed at obtaining pharmacokinetic data for diamorphine and its metabolites in children following intravenous (IVDIM) and intranasal (INDIM) administration in a blind study. It was intended that the concentrations of active metabolites would be used to evaluate whether or not IN-DIM can deliver rapid and efficient analgesia in children comparable to that obtained with IV-DIM. The pharmacokinetics of DIM and its metabolites following INDIM and IVDIM administration in children have been compared for the first time in this study, which confirmed that INDIM can achieve therapeutic plasma concentrations of active metabolites, although these were lower than those obtained with IVDIM and occur at later times after administration. In Scotland, the number of prescriptions for oxycodone has risen by 430% since prescribing began in 2002. Blood samples from fatalities in the West of Scotland involving oxycodone were analysed using an LC-ESI-MS/MS method developed for the determination of oxycodone and its metabolites in post-mortem specimens. To the author’s knowledge, this is the first report of blood and urine concentrations of noroxycodone and oxymorphone in acute oxycodone overdoses. Also, it is the first LC-MS/MS application to be reported with oxycodone related fatalities cases in forensic toxicology as most of previous reports used GC or HPLC applications. Moreover, this work reported for the first time vitreous humour levels of noroxycodone following oxycodone intoxication. Ten oxycodone-related deaths were identified in the short period of this study in the Strathclyde region of Scotland alone, highlighting the importance of including this drug in routine laboratory screening and confirmation procedures. Polar alcohol metabolites ethyl glucuronide and ethyl sulfate are biomarkers of ante-mortem alcohol consumption and are used to test for post-mortem artefactual formation of alcohol. An LC-MS method for these metabolites using a novel hydrophilic interaction liquid chromatography column was validated and applied to routine forensic casework. Ninety urine case samples were divided into three groups depending on the ethanol concentration found in blood and analysed by the developed method: group A with post-mortem blood ethanol higher than 200 mg/100 mL; group B with ethanol concentration in the range 80 to 200 mg/100 mL and group C with ethanol concentration less than 80 mg/100 mL. It was concluded that the risk of false positive ethanol results increased in the low ethanol concentration group as several cases tested negative for both biomarkers. ETG was detected at low concentrations in some cases for which ETS tested negative, suggesting that either ETG may have a longer half-life in urine or else ETS is unstable. The data was compared with previous studies and confirmed that both ethanol biomarkers should be determined in heavily putrefied cases and when the ethanol level in post-mortem blood is low, suggesting the production of ethanol after death. To the authors’ knowledge, this is the first report of the determination of ETS using an LC-ESI-ion trap-MS/MS method, and of a HILIC-ESI-ion trap-MS/MS method for the simultaneous determination of ETG and ETS in post-mortem urine samples.

The purpose of this study was to develop a single method to analyse drugs in hair and to apply this to case samples received from The Security Forces Hospital, Saudi Arabia. First of all, mass fragments and retention times for amfetamines and cannabinoids were identified by GC-MS using derivatization agents, PFPA/ethyl acetate (2:1 v/v) and PFPA / PFPOH (1:0.75 v/v) for amphetamines and cannabinoids, respectively. The results showed good peak shape, good chromatographic resolution and good sensitivity for amfetamine (AF), methamphetamine (MA), 3,4-Methylenedioxyamfetamine (MDA), 3,4-Methylenedioxymethamfetamine (MDMA), 3,4-Methylenedioxyethylamfetamine (MDEA), Δ9-tetrahydrocannabinol (Δ9-THC) and 11-nor-9-carbody- Δ9-tetrahydrocannabinol (Δ9-THC-COOH) compounds. The comparison of the efficiency of four different pre-treatment methods (enzymatic, alkaline, acid and methanol) to extract AF, MA, MDA, MDMA, MDEA, Δ9THC and Δ9-THC-COOH from hair samples obtained from known amfetamines and cannabinoid abusers was investigated. The preliminary results demonstrated difficulty with the cannabinoids recovery and the lower concentration of standards were not detected using any of the pre-treatment methods. As a result of the poor cannabinoids recovery, only the amfetamines were investigated. For the comparison study, only one hair sample positive for amphetamine was available so the pre-treatment comparison study was based on the recovery of AF using the four pre-treatment methods. The positive hair sample was separated into portions and pre-treatment methods, alkaline (1M NaOH), β-Glucuronidase (helix pomatia), methanol (MeOH) and acid (0.1M HCI) were used on these and compared. The best recovery for amphetamine was obtained using the β-glucuronidase pre-treatment method and this extract was also found to be cleaner than the alkaline and methanol pre-treatments. Β-glucuronidase pre-treatment was selected as the method of choice for the extraction of amphetamine content in hair. The method was validated to include linearity, recovery, intra- and inter-day precision, limit of quantitation (LOQ) and limit of detection (LOD) for all five amphetamine compounds. The method was shown to be reliable and robust for these substances. The stability of AF in hair was investigated to assess the validity of analysing hair samples for the presence of AF in victims of drowning. Ten amphetamine positive hair samples were submerged in fresh and sea water for different periods of time. The drug concentrations in the samples were monitored over a period of 8 weeks. Hair samples were analysed using the validated method. The results showed a significant decrease of amphetamine in hair with the time submerged in sea water. Fresh water had a much less significant effect over the study period. The validated method was successfully applied to 16 case samples obtained for living volunteers with a known history of fenethylline (AF precursor) abuse and 6 post-mortem case samples where amfetamines had been detected in the post-mortem blood.

An attempt was made to evaluate the importance of smoke and toxic gas inhalation in fires by comparing the volatile constituents in blood taken from fire fatalities with those of normal healthy and post-mortem controls. At the initial stage of the project, particular attention was given to the measurement of carbonyl compounds which represent a series of toxic and strongly irritant thermal degradation products from many polymeric materials. Three analytical methods, namely, i) gas chromatographic analysis of carbonyls after conversion to their corresponding 2,4-dinitrophenyl hydrazone derivatives, ii) direct static headspace gas chromatography, and iii) dynamic headspace gas chromatography, were compared for their applicability to the measurement of carbonyls in blood. Problems associated with binding of carbonyls to blood proteins were experienced in the first two methods which precluded their use for this application. The latter method was found to be most suitable in this respect since the bound carbonyls were released during the purging process. The method was also the most sensitive and when used in conjunction with a mass spectrometer, a detection limit in the nanogram per millilitre range was obtained. Volatiles in blood were extracted by purging the samples (1 ml) with 0.6 litre of helium and collecting the components on a small Tenax-GC column. The volatiles were then thermally desorbed and analysed by gas chromatography-mass spectrometry using a 100 m x 0.5 mm i.d. Carbowax 20M SCOT column. The sorption technique was found to extract a wide range of volatile components from blood enabling a comparison of volatile profiles to be made. Thus initially organic nitriles as well as the carbonyl compounds were quantified. Volatile components were identified by comparison of their retention indices and mass spectra with those of authentic standards. Where the latter were not available, tentative identifications were made 9n the basis of their mass spectral data only. Quantification of carbonyls and nitriles was achieved by comparing their response ratios to those from an external standard under identical conditions. During the period from August 1981 to May 1982, thirty-one blood volatile profiles were studied. These included four normal healthy controls and six post-mortem controls. In general, more complex profiles were found in fire fatalities than those of the controls. Over 140 chemical species have been identified in these! profiles and these include series of carbonyls, nitriles, alcohols, esters, aliphatic and aromatic hydrocarbons, halogenated hydrocarbons, heterocyclic compounds and sulphur-containing compounds. A detailed examination of the profiles has indicated that those compounds which might be of significance fell into two main catagories: those which were strong sensory and respiratory irritants, and those which were depressants of the central nervous system. Quantitative measurements of carbonyls in blood have shown that the mean levels of 2-butanone, butandione, 2-pentanone, cyclopentanone, cyclohexanone and hexanal were higher in fire fatalities than those in the controls. Of particular concern were the very high levels of acetonitrile and the presence of acrolein Ca highly toxic and strong sensory irritant) in the blood of some of the fire fatalities. Although the toxicological significance of these gaseous toxicants in causing fire fatalities has yet to be established, the results have clearly demonstrated that most of the fire deaths included in this exposed to a wide range of toxicants. these compounds may have played a vital incapacitation during the fire.

Mitochondrial DNA analysis was undertaken on samples collected from two populations in Peninsular Malaysia, the Modern Malay (102 samples) and Orang Asli (59 samples from Jahai and Kinsiu sub-groups). The hypervariable region 1 (HV1) of the mtDNA control region was amplified and sequenced. Polymorphisms were reported by aligning each sequence to the Cambridge Reference Sequence (CRS). A total of 94 polymorphisms were observed in the Modern Malay samples, which formed 75 different haplotypes. The Orang Asli showed notably lower number of the HV1 region variations, with only 28 polymorphisms and 13 haplotypes observed. Genetic diversity calculated for the Modern Malays and Orang Asli were 0.989 and 0.818, respectively. Probability of random match calculated was 0.0202 for the Modern Malays and 0.1962 for the Orang Asli. The mtDNA coding region variations was examined using RFLP analysis. Combination of both RFLP and HV1 sequence data had placed the Modern Malays into three major Southeast Asian haplogroups, M, B and F. These findings had initially suggested that the Modern Malays shared a common lineage with other populations within this region. Two novel sub-clusters, M21a and R21 were found at a high frequency within the Orang Asli samples. These sub-clusters, which have also been found in other Semang sub-groups appear to be indigenous Semang haplogroups. The limited number of mtDNA haplotypes shared between the Modern Malays and Orang Asli suggested discontinuity of mtDNA between these populations. Even though both populations were believed to be among the earliest populations of Peninsular Malaysia, this result indicates that the Modern Malays were not direct descendants of the Orang Asli. Minisequencing analysis was carried for further interrogation of the mtDNA coding region polymorphisms. Besides mtDNA analysis, the autosomal STR markers were also examined using PowerPlex® 16 system for both populations. These data could provide more information when added to the available STR database for Malaysian populations.

This thesis examines the use of bile and nail as alternative biological specimens in forensic toxicology. In chapter 1, a brief overview of the mechanisms of substance passage into and out of membranes and fluids and matrices was presented followed by a review of the use of alternative biological specimens for analytical purposes. The experimental part of this thesis begins in chapter 2 where a simple and rapid method for the detection and quantification of dextropropoxyphene in bile and its major metabolite, norpropoxyphene, is presented. Following this, nail is examined as an alternative biological specimen in cases of medico-legal interest. Firstly, two major cannabinoids, 9-tetrahydrocannabinol and 11-nor-9-tetrahydrocannabinol-9-carboxylic acid were identified in the nail (chapter 3) by means of RIA and GC-MS. Cannabinoids were found to be present in all six cases that were analysed for cannabinoids by RIA and their concentration ranged from 0.23 ng/mg to 2.80 ng/mg (average: 1.03 ng/mg). Using GC-MS, 9-tetrahydrocannabinol was detected in 11 out of the 14 nail clipping hydrolysates after basic extraction with concentrations ranging from 0.13 ng/mg to 6.97 ng/mg (average: 1.44 ng/mg). 11-nor-9-tetrahydrocannabinol-9-carboxylic acid was not detected in any of these nail clipping hydrolysates but was detected in two of the three fingernail clipping hydrolysates extracted under acidic pH with concentrations ranging from 9.82 ng/mg to 29.7 ng/mg (average: 19.8 ng/mg). In chapter 4, the detection and quantification of diazepam, the most commonly encountered benzodiazepine, were described. Using ECD-GC, diazepam was detected in hydrolysates of all 6 sets of nail clippings in concentrations ranging from 4.37 to 87.8 ng/mg (average: 25.7 ng/mg). In chapter 5, an analytical protocol for the detection and quantification of morphine in nail clippings of known heroin abusers was presented. Finally, a protocol for the detection and quantification of methadone in nail clippings of patients attending a methadone maintenance program was presented.

Experimental mixed-mode solid-phase extraction columns of differing carbon number and carbon loading were investigated for the efficient extraction of drugs of abuse from biological matrices. Methadone and its two major metabolites (EDDP and EMDP) were chosen due to the increase in methadone drug-related deaths in the West of Scotland. Amphetamine and related compounds (methamphetamine, 3,4-methylenedioxyamphetamine (MDA), 3,4-methyleneidoxymethamphetamine (MDMA) and 3,4-methylenedioxyethylamphetamine (MDEA)) were chosen because of the prevalence of use of "speed" and "ecstasy" throughout the United Kingdom and the challenge these volatile drugs pose to the analyst. Improved methods were developed for the efficient extraction of methadone, EDDP and EMDP from whole blood and for amphetamine and related compounds from whole blood and hair. These methods were successfully applied to the analysis of postmortem samples. The stability of methadone, EDDP and EMDP in whole blood was investigated using the developed extraction method after sorting at various temperatures for a period of six months. Methadone remained stable in blood after six months. This was confirmed by the analysis of postmortem cases samples which when re-analysed within six months of the initial analysis, correlated well. Postmortem hair samples proved useful for determining antecedent drug use history whether extracted by solid-phase extraction or after screening with a Cohort enzyme immunoassay kit, adapted for hair analysis. Solid-phase extraction provides a clean and efficient means of extracting drugs of abuse from whole blood and hair. Simple manipulation of the extraction system (sample pretreatment, pH, derivatisation) provides a versatile alternative to lengthy liquid-liquid extraction techniques and has the added potential of automation.

The use of supercritical fluid extraction (SFE) was evaluated for the determination of drugs of Forensic interest. Three matrices were investigated. The first two (blood and vitreous humor) were compared to SPE and LLE methodology currently in use at the Department of Forensic Science and Medicine. The third matrix, hair, was assessed to determine its usefulness as a marker of past drug use. Three types of drugs were investigated by SFE namely, Benzodiazepines, morphine and methadone. Successful methodology was developed for all three matrices and all three drug types, providing an efficient, reproducible alternative method to SPE and LLE, which reduced the environmental risks from organic solvents. The developed methods were applied to the analysis of authentic forensic samples. In addition to comparing well with the results obtained with the conventional techniques, good correlation was obtained between blood and vitreous humor results for temazepam, diazepam, methadone and morphine. Thus, in cases where a body is badly decomposed or burned, vitreous humor can be used as an alternative post-mortem sample. In addition to the determination of morphine, 6-monoacetyl morphine (6MAM) was used as marker of heroin abuse. 6MAM was detected in all three samples matrices, thus confirming the use of heroin prior to death. Hair analysis for all three types of drugs was carried out using a single extraction method. A wide range of concentrations was found for all drug types. As with blood and vitreous humor, 6MAM was detected and used as a marker of heroin abuse. From this, 61% of known heroin users were confirmed.

The work described in this thesis was in the field of Forensic Toxicology, which is the study and practice of toxicology for legal purposes. Three different application areas were considered: (a) the analysis of drugs in blood by gas chromatography-mass spectrometry (GC-MS) following isolation from the biological matrix by solid phase extraction procedures, (b) the analysis of paint solvents in blood by gas chromatography (GC) and GC-MS following dynamic headspace elution from blood, and (c) an evaluation of Thermospray/Plasmaspray (TSP/PSP LC-MS) liquid chromatography-mass spectrometry as the basis of drug screening techniques in forensic toxicology. (a) In the first of these areas, a sensitive, specific and reliable method was developed for the analysis of basic drugs in blood, using morphine and buprenorphine as model compounds which are also drugs commonly abused in the Glasgow area. A novel extraction procedure was used, based on a commercially available chemically-modified silica containing surface-bound benzenesulphonylpropyl groups, which served as a cation exchange resin. Several methods for the initial treatment of the biological matrix wre examined and the one selected involved absorption of the blood sample on diatomaceous earth and solvent elution of the crude extract containing any drugs present in the sample. This extract was then purified using the cation exchange resin: conditions suitable for the efficient retention of basic drugs and subsequent elution were also examined and developed. The extraction efficiencies for morphine were 925% and 954% at concentrations of 35 and 560ng/ml blood, respectively, for buprenorphine 836% and 875% , at concentration of 0.5 and 8ng/ml respectively, and for other drugs were better than 85% . The end-step analytical technique chosen for this method was GC-MS, because of analytical and legal requirements with respect to sensitivity and specificity. The polar nature of the model compounds, and of many other drugs likely to be encountered in this field, required chemical modification of the substances prior to gas chromatography. A comparison was made of three silyl ether derivatives - the trimethylsily (TMS), ethyldimethylsilyl (EMS) and tert-butyldimethylsilyl (MTBS) ethers. The conditions required for derivative formation were examined, including the choice of silyl donor reagent, solvents, temperature of reaction and time to completion. The EDMS donor, diethyltetramethyldisilazane, was selected on the basis of the midl reaction conditions required for the test compounds and also because this reagent caused less interference during GC-MS analysis. EDMS ethers gave satisfactory gas chromatographic behaviour and the presence of prominent ions at high mass was shown to be advantageous for specific and sensitive detection by SIR-MS with little background interference. The developed method was considered to be useful for both clinical and post-mortem blood samples containing morphine and buprenorphine down to the low picogram/millilitre level, and therefore adequate for the intended purpose. The method was subsequently applied to 13 samples submitted for analysis to the Department of Forensic Medicine and Science and found to be satisfactory. (b) In the second area of application, dynamic head space (DHS) elution was used for the analysis of paint solvents in blood. Volatilised solvents were trapped on a Tenax-GC cartridge and were subsequently analysed by GC-MS. The extraction efficiency of this method for C18-C12 n-alkane hydrocarbons, which were present in the paint materials, was better than 90% and sensitivity down to pg/ml levels was obtained. Solvent extraction procedures for these hydrocarbons were evaluated using high purity solvent. The extraction efficiency was better than 85% . Analysis of extracts by GC or GC-MS suffered from interference from the solvent front which reduced the sensitivity. The DHS method was applied to a pilot study for occupational monitoring of a group of painters to assess the presence of paint solvents in their blood. Two venous blood samples were collected at the beginning and at the end of a working week from each subject. They showed the presence of several solvents similar to those present in paint material and the levels in the second series of samples were higher than those of the first series. The differences between the levels in the two series were statistically significant for n-nonane, n-undecane and alkylbenzenes. The levels in the first sample indicated incomplete clearance of these solvents from the body during the weekend, and the second samples indicated solvent uptake during the working week. Solvent contamination in the extraction system was tackled by several approaches but still hindered the accurate estimation of solvent levels in blood. (c) In the third area of application, the operating parameters which control the sensitivity of the mass spectrometer using the TSP/PSP LC-MS interface were evaluated. These included the effects of the probe temperature and discharge voltage on sensitivity and mass spectral fragmentation pattern and the effects of the mobile phase constituents on sensitivity and mass spectral peak stability. Solvent systems containing ammonium acetate buffer and an organic modifier such as acetonitrile produced the best results in plasmaspray LC-MS. Three model HPLC-MS analyses were developed for mixtures of basic drugs, barbiturates and opiates using both the plasmaspray positive and negative ion modes. During the development of the mobile phases, the optimization of chromatography by organic modifiers was assessed. The quality of chromatography obtained was not always as good as expected in conventional HPLC, but the combination of chromatographic and mass spectral data could be used for identification and quantification purposes. A compilation of PSP mass spectra of drugs commonly encountered in forensic toxicology was produced. These mass spectra provided mostly molecular weight information with little structural information.