ABSTRACT
Introduction
Testing for controlled drugs
Types of drugs encountered in the United Kingdom
Author: P. J. O'NEIL, G. F PHILLIPS and, T. A. GOUGH
Pages: 17 to 33
Creation Date: 1985/01/01
The Laboratory of the Government Chemist examines most of the drugs that have been seized at the point of entry into the United Kingdom of Great Britain and Northern Ireland and has developed analytical methods for their rapid identification in the field and for more exact analysis and quantitation in the Laboratory. These methods are described for the major types of drugs encountered. Many seizures are examined in greater detail in order to compare samples and to correlate origin with physical and chemical appearance. Information on the procedures necessary to undertake this aspect of work is also presented.
The United Kingdom of Great Britain and Northern Ireland is not an illicit drug producing or exporting country, although Cannabis sativa is cultivated on a small scale and there is occasionally a clandestine amphetamine or lysergide (LSD) laboratory. Virtually all unauthorized drugs in the country that fall under the Misuse of Drugs Act 1971 have been imported.
The United Kingdom has a stringent control system at all airports and seaports the purpose of which is to intercept smuggled drugs at the point of entry into the country. Her Majesty's Customs and Excise is responsible for examining shipments of drugs that are imported legally, seizing materials that are suspected to contain illegal drugs and subsequently prosecuting offenders. The Laboratory of the Government Chemist is responsible for examining nearly all smuggled drugs seized by officers of Her Majesty's Customs and Excise within the United Kingdom and presenting scientific evidence at trials. The Laboratory also examines drugs on behalf of the police if the person in possession of the drugs is a serving member of the armed services. It is not, however, normally involved in the examination of drugs seized by the civil police within the United Kingdom, which is dealt with by the Home Office Forensic Science Service.
Every year the Laboratory examines about 5,000 exhibits for the customs service and 1,000 exhibits for the military services. In addition to the main Laboratory in London, there is a small testing laboratory at Heathrow, the country's major airport near London, which is staffed on demand. Members of the Drugs Section of the Laboratory make about 150 visits a year to Heathrow as well as approximately 200 other journeys throughout the country to examine drugs at the point of seizure. In 1984, approximately 16 t of cannabis products, 300 kg of heroin and smaller quantities of other drugs were examined outside the main Laboratory.
The facility at Heathrow is specifically designed to offer a "same day" analysis service to Her Majesty's Customs and Excise. Small quantities of cannabis or cannabis resin (usually less than 2 kg) can be examined by a scientist on the morning following the seizure. A "scientific witness" statement is prepared and the trial can be held at the local magistrates court the same morning. Such minor offenders are frequently sentenced within 24 hours of their arrest.
Although more than 100 items are listed in the Misuse of Drugs Act 1971 of the United Kingdom [ 1] , there are only five types of controlled drugs whose products are commonly imported into the United Kingdom. These are products of Papaver somniferum (opium, morphine, heroin), Erythroxolon coca (cocaine), amphetamine-related substances, LSD and Cannabis sativa (herbal cannabis, cannabis resin, liquid cannabis). However, a vast range and quantity of commodities enter and leave the United Kingdom, and customs officers require quick sorting tests to indicate the possible presence of illegal drugs. A full range of field tests are commercially available in kit form. Many of the individual tests and the entire test kit methodology were developed at the Laboratory [ 2, 3] . A sequence of simple and quick colour tests gives the customs officer a strong indication of the presence of a controlled drug. While a negative result of all the tests is sufficient to conclude that no commonly encountered controlled drugs are present, field tests alone are insufficient as positive identification for forensic purposes.
Since the country of origin of and route taken by many controlled drugs seized at the port of entry is known, the Laboratory can correlate physical appearance and chemical composition with geographical origin. The Laboratory has an extensive collection of samples taken from items submitted for forensic examination- There are now over 10,000 such items, covering the entire range of controlled drugs, pharmaceuticals and related materials.
The products of Papaver somniferum are collectively referred to as opiates and may be encountered in crude form, for example opium and poppy straw, or as more or less purified chemical substances, such as morphine and heroin.
Opium contains 25 to 30 alkaloids, of which morphine, codeine, thebaine, noscapine, papaverine and narceine are the most important. In Indian opium, the morphine content averages 10 per cent by weight. In a detailed paper, de Faubert Maunder [ 4] describes field and laboratory tests for raw and prepared opium ; a comprehensive account is given of the reactions of opium to common colour tests, and the author presents a general analytical scheme for these materials. Opium is composed of many constituents and varies widely in colour; samples of opium are not readily amenable to any field or colour test. In the field-test scheme for drugs of abuse devised by the Laboratory, the investigating officer is advised to submit suspected opium to a laboratory for examination. However, in order to keep the number of false submissions to a minimum, investigating officers need to be thoroughly acquainted with the physical forms and properties of opium.
Colour tests for opium carried out in a laboratory are also of limited value since they cannot be applied directly to the base material and are difficult to interpret and inherently unreliable. In the laboratory thin layer chromatography (TLC), using authentic opium as a standard, should be adopted as the final unequivocal test for the presence of opium. This is the method used at the Laboratory of the Government Chemist, where the preferred solvent system [ 5] is ethyl acetate-methanol-ammonium hydroxide (85 : 10 : 5). Opium is detected using ultraviolet (UV) irradiation at 254 nm or an iodoplatinate reagent. In order of increasing Rf value, the major components of opium visualized by these means are morphine, codeine, thebaine, papaverine and noscapine.
For most forensic purposes more sophisticated tests than TLC, e.g. combined gas chromatography and mass spectrometry (GC-MS), are unnecessary. They are needed to quantify the morphine content only if the quality of the sample is questioned in court. For morphine quantitation the preferred technique is high performance liquid chromatography (HPLC). The difficulty in using this technique lies not in the chromatography but rather in ensuring that all the morphine has been extracted from the opium in preparing the solution for analysis. The HPLC system developed by Baker and Gough [ 6] is suitable for the quantitation of morphine and codeine.
Although rarely encountered in the United Kingdom, samples of poppy straw can be readily screened for the presence of opium alkaloids using the appropriate chromatographic technique, the most effective being TLC. Since the presence of morphine and codeine has been reported in other species of poppy, e.g. Fructus papaveris, the Laboratory has always sought botanical confirmation of the species from the Royal Botanical Gardens, Kew.
A concentrate of poppy straw can be identified by its alkaloidal constituents and any botanical debris present. The borderline between a concentrate of poppy straw and crude or impure morphine is ill-defined; in borderline cases the levels of the other alkaloids and the morphine content must be determined using the methods described for opium.
In the United Kingdom illicit morphine is less commonly encountered than heroin either by Her Majesty's Customs and Excise at ports of entry or by the police within the country. There have, however, been a number of seizures of kilogram quantities of this drug. It is usually in the form of an off-white or light brown powder. Occasionally, it is seized in the form of small tablets which may also contain codeine and which have presumably been diverted from licit sources. Initial screening within the Laboratory is undertaken using the Marquis test (cherry red becomes purple) and the nitric acid test (intense brick red); the latter distinguishes morphine from heroin. Neither test is, however, an unequivocal proof of identity. Virtually all samples dealt with by the Laboratory have been of high purity (greater than 50 per cent), and thus infra-red spectroscopy offers the most effective proof of the presence of morphine.
The determination of the salt form of morphine is considerably more difficult than the straightforward differentiation between diamorphine as the base or hydrochloride salt in illicit heroin samples. 1 Virtually all illicit morphine samples give both a chloride and a sulphate precipitate. This is probably because, although the morphine is present as the sulphate, chloride ions are also present as a residue from the precipitation (using ammonium chloride) of morphine after its extraction from opium. The infra-red spectra of impure morphine base and impure morphine hydrochloride are similar, and it is not often possible to distinguish between them. The use of X-ray diffraction (XRD) offers a satisfactory solution to this problem, and it has been used successfully on a number of occasions at the Laboratory, although when the concentrations of morphine in the sample are low XRD patterns are often too weak to interpret.
1 Throughout this article the name "diamorphine" is used for pure diacetylmorphine and "heroin" for crude acetylated morphine products.
The most convenient method of determining the purity of illicit morphine is HPLC. Of the common narcotic opiates, morphine is the most polar compound. As a result, it is prone to loss by adsorption in gas chromatography columns, although such losses are reproducible under some conditions [ 7] . In view of this limitation, HPLC offers a more attractive method for morphine quantitation. In the system developed in the Laboratory [ 6] , morphine is eluted in approximately 15 minutes and is quantified by peak area measurement following calibration with standard solutions of pure morphine.
The physical appearance of illicit heroin encountered at the Laboratory varies widely, ranging from almost pure white heroin hydrochloride intended for injecting (and often indistinguishable from pharmaceutical grade diamorphine) to crude and impure heroin, which is often in the base form and is intended for smoking or inhaling. The only type of heroin that has not been found in the United Kingdom is that of Mexican origin; the entire Mexican production appears to remain within North America.
The types of heroin encountered in the United Kingdom listed by source of supply are described below.
South-East Asia. The types of heroin encountered in the United Kingdom are: (a) "Chinese No. 3", a hard granular material (usually 1 to 5 mm in diameter), that does not yield to pressure and often contains only a small amount of powder varying in colour from grey to dirty brown; (b) "Chinese No. 4", a white microfine dry powder, often crystalline; (c) "Penang Pink" a granular material similar to Chinese No. 3 but dirty pink in colour. In Chinese No. 4 alkaloids are always present as the hydrochloride salts; the other two types most often occur as hydrochlorides but may se the free base(s).
India. Heroin from India is in the form of a white microfine powder similar to Chinese No. 4 or, on occasion, a creamy or dirty white powder that may be aggregated. The alkaloids are almost always present as hydrochloride salts.
Iran (Islamic Republic of), Heroin from the Islamic Republic of Iran is very similar to the common type from Pakistan but less variable in both physical appearance and chemistry. It is usually a fine brown powder containing about 70 per cent heroin as the free base-
Lebanon. Heroin from Lebanon is in the form of a white or off-white fine powder; it is rarely encountered in the United Kingdom.
Nigeria. The fine off-white or light brown powder from Nigeria is not distinguishable from the common heroin from Pakistan.
Pakistan. Heroin from Pakistan seized in the United Kingdom varies in colour and consistency. The most common type has been encountered in virtually every shade from beige to dark brown. It is usually a powder, often a fine one, but occasionally it is found as small aggregates that are soft and yield to slight pressure. The product has a characteristic opium-derived odour, the purity is in the range 70 to 80 per cent and the alkaloids are present as the free base. An uncommon type is a white or off-white fine dry powder, with less odour than the common type. The purity is in the range 80 to 90 per cent, and the alkaloids are present as hydrochloride salts.
Turkey. Heroin from Turkey is beige or very light brown, in powder form and most often without aggregation.
Syrian Arab Republic. Heroin from the Syrian Arab Republic, in the form of a pale orange-brown powder, is rarely encountered in the United Kingdom.
Field tests for heroin products are the same as those for morphine. The reaction to the Marquis test is identical, but heroin products can be distinguished by the nitric acid test (slow production of a green colour). The purity of heroin seized in the United Kingdom is rarely below 30 per cent, and thus infra-red spectroscopy is the easiest unequivocal test for the presence of heroin. Using this technique there is usually no difficulty as with morphine in determining the salt form, but results can be confirmed by precipitation using silver chloride and barium sulphate.
The quantitation of heroin is carried out by gas liquid chromatography (GLC) and HPLC; each method has certain advantages. The Laboratory, like other laboratories, employs the GLC system using a stationary phase of 3 per cent OV 17 silicone gum [ 7] . This stationary phase resolves diamorphine from other opiates and common diluents (cutting agents) that are encountered in illicit heroin and produces a satisfactory quantitation of the diamorphine content. However, papaverine, noscapine and morphine are not satisfactorily eluted, and 6-acetylmorphine is unresolved from acetyl-codeine. Thus, OV 17 is of little help in the chemical "fingerprinting" of illicit heroin for comparison or in determining geographical origin. This method was in use in the Laboratory from 1974 to 1978. During that time the only types of illicit heroin frequently encountered in the United Kingdom were from South-East Asia, and the Laboratory was seldom asked to provide chemical fingerprinting of heroin. In any case, papaverine, noscapine, morphine and codeine are absent or at least at very low levels in this type of heroin. However, high levels of caffeine (from 30 per cent to 70 per cent) are found in Chinese No. 3, which on OV 17 is eluted soon after the solvent peak; it can thus be readily determined. In addition, the common synthetic local anaesthetics -lignocaine, benzocaine, and procaine - are resolved from each other and from caffeine and are eluted well before the opiates. OV 17, which is also used in the quantitative analyses of many other controlled drugs (e. g. cocaine, methadone, pethidine and constituents of the products of Cannabis sativa), is thus a very useful stationary phase for quantitative work in the forensic laboratory.
In 1976 heroin from the Islamic Republic of Iran began to be illegally exported to the Western countries. When subjected to the TLC system used in the Laboratory, it was found to contain papaverine and noscapine as well as diamorphine, 6-acetylmorphine and acetylcodeine. Some samples, which either were badly made or had undergone hydrolysis, contained high levels of 6-acetylmorphine and morphine.
A GLC system that could separate and elute as many as possible of the components of heroin from the Islamic Republic of Iran was highly desirable. Although adsorption effects were taken into account when selecting the stationary phase, the primary criterion was the separation of the five narcotics (diamorphine, 6-acetylmorphine, acetylcodeine, codeine and morphine) from each other and from caffeine. A range of phases from completely nonpolar (hydrocarbon greases) to polar-substituted silicone gums was examined [ 7] , the different phases being selected on the basis of the McReynolds values. A systematic study of the stationary phases available showed that silanized OV210 fulfilled the criteria most effectively. Only papaverine and noscapine failed to be eluted by the stationary phase, while separation and reproducible quantitation of all six compounds and common cutting agents were possible. The analysis time was 25 minutes per injection.
The Laboratory has taken part in a collaborative study with the laboratory of the Department of Scientific Services of the Republic of Singapore to evaluate chromatographic methods currently used in the two laboratories for the quantitative analysis of heroin mixtures. Nearly 100 samples, taken from seizures originally analyses forensically, were made available for the evaluation, which was carried out simultaneously in London and Singapore. The samples were representative of the entire range of preparations that had been encountered and were of known geographical origin. Close agreement was achieved in nearly all samples [ 8] . A collaborative study (unpublished), performed m a similar manner, has also been undertaken in co-operation with the Australian Government Analytical Laboratory. Again close agreement for the majority of samples was achieved.
One type of sample known to give different quantitative results by HPLC and GLC analyses contains appreciable levels of 6-acetylmorphine and morphine in addition to diamorphine. The most likely explanation for this phenomenon is the transacetylation of the various morphinans within the GLC column, which results in discrepancies for the concentration of 6-acetylmorphine. A study of the transacetylation process showed that the discrepancy could be reduced by using the lowest possible injection port temperature consistent with good chromatography [ 9] .
The need to overcome the problem of transacetylation and the non-elution of papaverine and noscapine has led to development of a capillary GLC method. The criteria used in this method are:
Separation and elution of the five major components found in most heroin samples (diamorphine, 6-acetylmorphine, acetylcodeine, papaverine and noscapine);
Accurate quantitative analyses of these compounds, i.e. there must be minimal adsorption and transacetylation;
No derivatization of samples prior to chromatography;
As short an analysis time as possible, preferably less than 30 minutes.
A systematic study has been made of the stationary phases chemically bonded on to vitreous silica capillary columns. A splitless injection is used in which the sample is introduced into a cooled zone to remove the solvent prior to eluting the active constituents by temperature programming. All five components are separated on a column of OV 210 within 30 minutes, and the quantitative data obtained using laboratory prepared mixtures of opiates and illicit samples are satisfactory [ 10] .
The HPLC system used by the Laboratory for illicit heroin analysis [ 6] uses amino-propyl bonded silica packing and a mixture of 85 per cent acetonitrile and 15 per cent 0.005 M tetrabutylammonium phosphate as the mobile phase. Caffeine, heroin, acetylcodeine, 6-acetylmorphine, codeine and morphine are separated from each other and can be accurately quantified. Papaverine, noscapine, thebaine and strychnine are also separated from the principal components. The total analysis time is 17 minutes; however, codeine and morphine are absent in the majority of illicit samples, and the time can be reduced to 5 minutes. As an alternative HPLC system, the Laboratory has recently adopted the procedure described by Huizer [ 11] .
The Laboratory's reference collection of heroin contains over 600 items and is representative of approximately 90 per cent of the heroin seized by law enforcement organizations in the United Kingdom in the last 10 years. Each of these heroin samples has been subjected to chemical analysis using the chromatographic systems described above. The chemical data thus generated, together with the colour and physical appearance, have been correlated with the putative geographical origin to produce a typographic description of illicit heroin products. The data can also be used to establish the frequency with which a particular profile occurs. This information is of value in comparing samples of seized heroin. The Laboratory has never encountered a coincidence of chemical composition in unrelated samples, and concomitantly, the chemical profiles of samples known to be related, for example heroin removed from different packages concealed within the same false bottom of a suitcase, have been determined to ascertain the variation of composition. The Laboratory's work on the chemical profiling of heroin has been published [ 12, 13] and, as further samples are collected, it will be updated.
Seizures of coca leaf in the United Kingdom are rare and invariably small. The physical appearance of coca leaves is unique, with two longitudinal indentations on either side of the midrib that are more conspicuous on the grey-green underside than the dark green upper side. Confirmation is simple: the chopped leaf material is triturated with alkaline methanol, and the extract is subjected to TLC or other chromatographic techniques.
The Laboratory has examined many samples of cocaine powder seized at the port of entry. The quality has varied from essentially pure to highly adulterated cocaine. The field test used for cocaine [ 14] consists of cobalt (II) thiocyanate in acidic solution which, under test conditions, produces an intense turquoise colour within 5 seconds in the presence of cocaine. This test is only an indication of the possible presence of cocaine, as some other materials (including certain controlled drugs) give the same reaction within the time limit. The TLC systems in use for cocaine are also less specific than those used for other drugs because the Rf value and colour generated are by no means unique to cocaine. A more positive identification can be made by developing a number of plates, each sprayed with a different reagent, in different solvents. However, for unequivocal identification, infra-red spectroscopy offers the simplest approach. Cocaine brought into the United Kingdom is almost always present as the hydrochloride salt. Infra-red spectroscopy (confirmed if necessary by precipitation tests or XRD) also provides an easy way of determining the salt form. The determination of the purity of cocaine is performed by GLC using an OV 17 stationary phase. A pressure-programmed GLC system with this stationary phase is used if other local anaesthetics are present in order to reduce the analysis and equilibration time substantially [ 15] .
The only amphetamine and structurally related drugs that are encountered with any regularity in the United Kingdom are amphetamine, dexamphetamine and methylamphetamine. These amphetamines, the pro- ducts of illicit laboratories, are in the form of white powders, which are often damp owing to the presence of solvent residues. Such powders usually emit the odour of acetone or other organic solvents. A colour change is sometimes observed as samples are dried. Tentative recognition of these three compounds can be easily made using the Marquis test; all three produce a yellow-orange colour. The large seizures examined in the Laboratory (once the residual solvent has been removed) are usually sufficiently pure to permit use of infra-red spectroscopy without prior "clean-up", but this may not be possible for smaller samples seized at the street level. The simplest method of clean-up is to subject the suspect amphetamine to column chromatography using diethyl ether. The amphetamine may be subsequently eluted by basification, collected in a few drops of hydrochloric acid and, after drying, examined by infra-red spectroscopy. This procedure is the best way to identify the chemical species present; XRD is used in the Laboratory to differentiate between stereoisomeric forms in samples containing reasonable levels of the amphetamine.
TLC of amphetamines is performed using the system developed by Davidow [ 5] ; the detection is by UV absorption at 254 nm or colour development with iodoplatinate reagent or ninhydrin. Phillips and Gardiner [ 16] examined a large number of phenethylamines in two TLC systems: silica gel developed in methanol-ammonium hydroxide (100 : 15) and alkaline silica gel developed in chloroform-methanol (9 : 1). The authors found that considerable discrimination between phenethylamines was possible using these two systems visualized with iodine in methanol spray. Unfortunately there is little difference in the Rf of amphetamine and some chemically related compounds. TLC in this application is useful as an indicator of the class of compound only.
The purity of amphetamine and chemically related compounds can be determined most effectively using GLC with a column consisting of 10 per cent potassium hydroxide and 10 per cent Apiezon "L" on Chromosorb W in a 1.5-m glass column and nitrogen carrier gas at 30 ml min -1. The compounds should be analysed as bases by adding a few drops of strong ammonium solution to the sample powder. The organic solvent must then be quickly added, as the free bases rapidly decompose in the atmosphere. The most satisfactory solvent is diethyl ether.
Few of the hallucinogens, whether as natural products, as extracts or synthesized, have been encountered in large amounts in the United Kingdom. The exception is LSD, which has been encountered in a variety of presentations: in solution, added in solution to other tablets, on sugar lumps, in refilled capsules, as illicitly prepared tablets ("microdots" or "domes") and, increasingly in recent years, within absorbent paper. 2 The small tablet and paper presentations are so distinctive that they are easily recognized by experienced forensic analysts. Customs and other enforcement officers may not recognize them as easily; therefore the Laboratory developed a field test for examining suspect lysergide presentations based on the widely used reagent dimethylbenzaldehyde (DMAB) [ 17] . A drop of the reagent is placed on the suspect material on a filter paper; violet or purple striations on the filter paper indicate the possible presence of an indolic substance. However, at the time the test was developed lysergide was nearly always in tablet form; the response of some paper presentations to the test is not as distinctive. In addition, because LSD decomposes, old samples may give a poor or slow response; the effective response time of this test is up to 5 minutes. The LSD test is incorporated in the comprehensive field testing kit developed by the Laboratory.
Infra-red spectroscopy of lysergide in individual illicit dosage forms is very difficult because of the extremely small quantities involved. Furthermore, the presence of impurities, such as the coloured dyes of tablets or paper patterns that are coextracted with the lysergide, make the infra-red spectrum difficult to interpret. The most satisfactory method of confirmation is by TLC or HPLC; lysergide is too labile for useful GC analysis. Two TLC methods are used, both with acetone as an eluting solvent. A methanol extract of the lysergide is spotted onto silica and alumina plates. Detection is by a diluted form of the field test reagent.
For quantitative determinations, the HPLC method developed by Twitchett [ 18] has been adopted. This will separate lysergide from 17 related compounds, including naturally occurring ergot alkaloids. The very low levels of lysergide within illicit presentations makes the determination of the salt form of this drug difficult. The infra-red spectroscopy and XRD methods are of no value in this instance; however, an ion chromatographic method has been developed [ 19] in which various anions in illicit presentations are separated. The quantities of any anions that form stable salts with lysergide are compared with the amount of lysergide in the sample, from which the salt form of the presentation may be deduced. Detection is by measuring electrical conductivity. No organic ions, with the exception of tartrate, have been found to occur in either paper or tablet presentations of lysergide. However, the majority of samples encountered at the Laboratory have been found to contain lysergide as the free base. Since these were paper presentations, presumably made by dipping the paper into lysergide solution, there would be no advantage in converting the lysergide to tartrate, which is used only when lysergide tartrate needs to be precipitated prior to its addition as a solid to powder excipients.
2 The paper is usually covered with a non-absorbent sheet that bears a repeating pattern: the covering sheet contains no lysergide.
Since January 1985, methylphenobarbitone and 5,5 disubstituted barbituric acids have been subject to control under an amendment to the Misuse of Drugs Act 1971. In preparation for this amendment, Dybowski and Gough [ 20] developed an analytical procedure for the identification of such barbiturates. The procedure involves a simple field test that can be used by non-scientific personnel, TLC, HPLC, GLC and infra-red spectroscopy.
The field test reagent for barbiturates consists of a solution of cobalt (II) thiocyanate dissolved in methanol and 2,6-dimethylmorpholine. One drop of the reagent is added to about 10 mg of the sample on a filter paper. The reagent colour is pale blue; a change in colour to violet or purple within 10 seconds is indicative of the presence of a barbiturate. If any other colour is produced, the test is repeated using two filter papers. The sample is placed on the top paper and a few drops of acetone are slowly added until the bottom paper is impregnated. After drying the bottom paper, the field test reagent is added.
TLC is developed with chloroform-acetone (4 : 1) using silica gel plates. Visualization is by spraying with 2 per cent mercury (II) chloride in ethanol followed by 0.2 per cent 1,5-dipehnylcarbazone in ethanol. GLC is performed on a capillary column of vitreous silica coated with a bonded dimethylsilicone stationary phase.
HPLC is performed with a column containing Spherisorb ODS. The mobile phase is 0.1 M sodium dihydrogen orthophosphate in methanol at pH 8.5. Detection is by UV absorption at 210 nm. The combination of GLC and HPLC has been found to be satisfactory for the identification of most barbiturates, and the TLC system is a useful sorting procedure. Confirmation by mass spectrometry is carried out where necessary.
Approximately 80 per cent of the samples submitted to the Laboratory are of cannabis products, the most common being herbal cannabis, cannabis resin and liquid cannabis (hashish oil). Since the country of origin is usually known, it has been possible for the Laboratory to build up a comprehensive collection of cannabis products from all over the world [ 21] . This collection has been a vital ingredient in the training of new members of the Drugs Section of the Laboratory.
Customs officers often encounter substances that they believe may contain cannabis. Thus, a fast sorting test to indicate the likely presence of cannabis products is needed to minimize the time spent with innocuous materials. De Faubert Maunder developed such a test [ 22] based on the well-known colour reaction between azo dyes and cannabinoids. He tested a number of azo dyes and concluded that Fast Blue B was the best reagent, although care had to be exercised with this substance because of the possibility of carcinogenic impurities being present. Subsequent work at the Laboratory concluded that Fast Blue BB, which is known to be safer than Fast Blue B, is equally satisfactory as a test for cannabis products. Fast Blue BB is now preferred both as a field test reagent and for visualizing TLC plates. Fast Corinth V is used in the commercial version of the test kit.
The field test for cannabis products is carried out using two filter papers folded together into a normal cone used in filtration. Approximately 1 mg of the suspect material is placed in the inner paper and petroleum ether 3 is added slowly until the outer paper is moistened. The two filter papers are then separated, the suspect material set aside and the upper filter paper discarded. The lower filter paper is opened and allowed to air dry; approximately 0.1 mg of the test reagent is placed in the centre of the paper and one drop of water containing 1 per cent sodium hydrogen carbonate is added. A red to violet colour developing as the water expands over the area originally covered by the petroleum ether is indicative of cannabis products. The test reagent consists of 1 per cent Fast Blue BB in powdered anhydrous sodium sulphate. Sodium hydrogen carbonate makes the aqueous solution alkaline thereby facilitating colour development. Approximately 240 botanical substances other than cannabis products have been subjected to the field test [ 23] , and only mace and nutmeg gave a positive response, although they produced a pink rather than red or violet colour.
Two TLC systems are favoured for the examination of cannabis products [ 24] ; both use 100 mm x 100 mm pre-coated silica gel plates with a layer thickness of 0.25 mm. Either system will separate most of the commonly occurring cannabinoids, although the acid precursors run together as a zone near the base of the plate. For routine examinations, a solvent mixture of alcohol-free chloroform (prepared by standing chloroform over anhydrous granular calcium chloride for 24 hours) and 1,1-dichloroethane (15 : 10) is used. More diagnostic information can be obtained using two-dimensional TLC; the plate is developed first in the solvent and then at right angles in xylene-dioxan (19 : 1). The plate should be dried and then lightly sprayed with diethylamine between the two runs. In a modification of this system, the plate is heated after the first run in order to decarboxylate the cannabinoid acids, which may then be separated during the second run. These TLC methods are used extensively for assessing the age of cannabis and as an aid in determining the origins of seizures [ 21] .
GLC analysis of cannabis products, which is carried out on an OV 17 stationary phase [ 25] , results in the decarboxylation of the cannabinoid acids and hence the loss of potentially useful information. For this reason GLC offers no advantage over TLC for qualitative work. GLC analysis of cannabis products is, however, extremely useful in the determination of the total tetrahydrocannabinol (THC) content, which is generally believed to be the most satisfactory indicator of the "quality" of the product [ 26 - 28] . The concentration of THC is used to calculate the quantity of cannabis or cannabis resin used in liquid cannabis (hashish oil). However, the analyst must know both the country of origin of the liquid cannabis exhibit and the average THC content of the resin or herbal material produced in that country. The herbal or resin source can be easily inferred from the presence or absence of cannabidiol in a TLC examination. As far as seizures made in the United Kingdom are concerned, virtually all the "resin-belt" countries export liquid cannabis containing high levels of cannabidiol (e.g. India, Lebanon, Morocco, Pakistan). Of the countries that do not produce cannabis resin, only Jamaica and Kenya have exported liquid cannabis, which in both cases has been found to be devoid of cannabidiol and is thus from a herbal source. Typical concentration factors in liquid cannabis are 2 to 4 times for the "resin" countries and 6 to 10 times for the "herbal" countries; for either variety the optimum THC level is approximately 30 per cent by weight, although the rare Indian or Nepalese sample has been found to contain 50 to 60 per cent.
3 Petroleum ether with a higher boiling range should be used in tropical climates.
HPLC provides quantitative information on the cannabinoid and cannabinoid acid content and is thus much more useful than either TLC or GLC for quantitative work and fingerprinting cannabis, although TLC is entirely satisfactory for identification purposes. The HPLC system [ 25] uses a Spherisorb S5ODS (ODS-silica) column and 0.02 N sulphuric acid-methanol-acetonitrile (7 : 8 : 9) as eluent to separate the major cannabinoids; these are identified either by comparison with authentic samples or by mass spectrometry following preparative HPLC. Extraction of the cannabis products by ultrasonic agitation (25 minutes) into methanol-chloroform (4 :1) is found to be essentially quantitative. Detection by UV absorption at 220 nm is used in order that the cannabinoids of major importance can be detected with adequate sensitivity. This system, which is used to determine up to 12 cannabinoids, has been used to compare cannabis resin samples. Cannabis resins from Lebanon [ 29] , Morocco [ 30] and Pakistan [ 31] have been shown to be homogeneous products with respect to the distribution of cannabinoids within a single slab of resin, although there were minor differences between the surface and the interior of Moroccan and Pakistani slabs. In general, inter-slab coefficients of variation in a given batch are no higher than the corresponding intra-slab values, indicating that resins, at least from these two countries, are relatively uniform products. In 94 unrelated seizures from Pakistan, no two slabs were found to have coincident cannabinoid profiles. In addition, no correlation was found between their cannabinoid content and the "identification" marks embossed on some resins from Pakistan. The chromatographic methods described above are also used for the examination of cannabis grown for official purposes in the United Kingdom as part of a cultivation programme to study the constancy of chemical profiles of successive generations originating with seeds of known foreign origin [ 32 - 34] .
Misuse of Drugs Act 1971 and amendments.
02G. F. Phillips, "The determination and identification of illicit drugs", Chemistry and Industry, 1975, pp. 497 - 498.
03G. F. Phillips, "Die Analyse von Arzneimitteln", Methodicum Chimicum, F. Korte, ed. (London, Academic Press, 1974), Volume 1 B, p. 903.
04M. J. de Faubert Maunder, "Field and laboratory tests for raw and prepared opium", Bulletin on Narcotics (United Nations publication), vol. 27, No. 1 (1975), pp. 71 - 76.
05B. Davidow, N. I. Petri and B. Quame, "A thin-layerchromatographic screening procedure for detecting drug abuse", American Journal of Chemical Pathology, vol. 50, 1968, pp. 714 - 719.
06P. B. Baker and T. A. Gough, "The separation and quautitation of the narcotic components of illicit heroin using reversed-phase high performance liquid chromatography", Journal of Chromatographic Science, vol. 19, 1981, pp. 483 - 489.
07T. A. Gough and P. B. Baker, "The selection of gas chromatographic stationary phases and operating conditions for the separation and quautitation of heroin and structurally related compounds", Journal of Chromatographic Science, vol. 19, 1981, pp. 227 - 234.
08S. T. Chow and others, "A comparison of chromatographic methods for estimation of the diacetylmorphine content of illicit heroin", Journal of Chromatographic Science, vol. 21, 1983, pp. 551 - 554.
09R. Dybowski and T. A. Gough, "A study of transacetylation between 3,6- diacetylmorphine aud morpbine", Journal of Chromatographic Science, vol. 22, 1984, pp. 465 -469.
10G. Sutherland, P. J. O'Neil and T. A. Gough, unpublished data.
11H. Huizer, "Analytical studies on illicit heroin, II: comparison of samples", Journal of Forensic Sciences, vol. 28, 1983, pp. 40 -48.
12P. J. O'Neil, P- B. Baker and T. A. Gough, "Illicitly imported heroin products: some physical and chemical features indicative of their origin", Journal of Forensic Sciences, vol. 20, No. 3 (1984), pp. 889 - 902.
13P. J. O'Neil, P. B. Baker and T. A. Gough, "Illicitly imported heroin products: some physical and chemical features indicative of their origin: Part II", Journal of Forensic Sciences, to be published in July 1985.
14G. V. Alliston and others, "An improved test for cocaine, methaqualone and methadone with modified cobalt (II) thiocyanate reagent", The Analyst, vol. 97, 1972, pp. 263 - 266.
15T.A. Gough and P.B. Baker, "The rapid determination of cocaine and other local anesthetics using field tests and chromatography", Journal of Forensic Sciences, vol. 24, 1979, pp. 847-855
16G. F. Phillips and J. Gardiner, "The chromatographic identification of psychotropic drugs", Journal of Pharmacy and Pharmacology, vol. 21, 1969, pp. 793-807.
17G.V. Alliston and others, "An improved field-test for hallucinogens", Journal of Pharmacy and Pharmacology, vol. 23, 1971, pp. 71-72.
18P.J. Twitchett and others, "Analysis of LSD inhuman body fluids by high performance liquid chromatography, fluorescence spectroscopy and radio-immunoassay", Journal of Chromatography, vol. 150, 1978, pp. 73-84.
19P.A. McDonald and others, "Ananalytical study of illicit lysergide", Journal of Forensic Sciences, vol.29, 1984, pp.120-130.
20R. Dybowski and T.A. Gough, "dentification of 5,5-disubstituted barbiturates", Journal of Chromatographic Science, vol. 22, 1984, pp. 104-110.
21P.B. Baker, T.A. Gough and B.J. Taylor, "Illicitly imported cannabis products: some physical and chemical features indicative of their origin", Bulletin on Narcotics (United Nations publication), vol. 32, No.2 (1980), pp. 31-40.
22M.J. de Faubert Maunder, "An improved field-test for barbiturates and hydantoins with a modified cobalt (II) thiocyanate reagent", The Analyst, vol. 100, 1975, pp. 878-883.
23M. J. de Faubert Maunder, "An improved procedure for the field-testing of cannabis", Bulletin on Narcotics (United Nations publication), vol. 26, No. 4 (1974), pp. 19-26.
24R. Fowler, R.A. Gilhooley and P.B. Baker, "The thin layer chromatography of cannabis", Journal of Chromatography, vol. 171, 1979, pp. 509-511.
25P.B. Baker and R. Fowler, "Analytical aspects of the chemistry of cannabis", Proceedings of the Analytical Division of the Chemical Society, 1978, pp. 347-349.
26P.B. Baker, K.R. Bagon and T.A. Gough, "Variation in the THC content in illicitly imported Cannabis products", Bulletin on Narcotics (United Nations publication), vol. 32, No. 4 (1980), pp. 47-53.
27P.B. Baker and others, "Variation in the THC content in illicitly imported Cannabis products-Part II", Bulletin on Narcotics (United Nations publication), vol.34, Nos. 3&4 (1982), pp. 101-108.
28P.B. Baker, B.J. Taylor and T.A. Gough, "The tetrahydrocannabinol and tetrahydrocaunabinolic acid content of cannabis", Journal of Pharmacy and Pharmacology, vol. 33, 1980, pp. 369-372.
29P.A. McDonald and T.A. Gough, "The determination of the distribution of cannabinoids in cannabis resin from the Lebanon using high performance liquid chromatography", Journal of Chromatographic Science, vol. 22, No. 7 (1984), pp. 282-284.
30P.B. Baker, T.A. Gough and P.J. Wagstaffe, "The determination of the distribution of cannabinoids in cannabis resin from Morocco using high performance liquid chromatography", Journal of Analytical Toxicology, vol.7, 1983,pp.7-10.
31P.B. Baker and others, "The determination of the distribution of cannabinoids in cannabis resin using high performance liquid chromatography", Journal of Analytical Toxicology, vol.4, 1980, pp. 145-152.
32P.B. Baker, T.A. Gough and B.J. Taylor, "The physical and chemical features of cannabis plants grown in the United Kingdom of Great Britain and Northern Ireland from seeds of known origin", Bulletin on Narcotics (United Nations publication), vol. 33, No. 1 (1982), pp. 27-36.
33P.B. Baker, T.A. Gough and B.J. Taylor, "The physical and chemical features of cannabis plants grown in the United Kingdom of Great Britain and Northern Ireland from seeds of known origin: second generation studies", Bulletin on Narcotics (United Nations publication), vol. 35, No. 1 (1983), pp. 51-62.
34B.J. Taylor, J.D. Neal and T.A. Gough, "The physical and chemical features of cannabis plants grown in the United Kingdom of Great Britain and Northern Ireland from seeds of known origin: third generation studies", Bulletin on Narcotics (United Nations publication), forthcoming.