Fifty years of development of opium characterization methods


A short historical account of opium characterization: the United Nations opium research programme, 1951-1967
Basis for opium characterization and origin determination
Classification of opium characteristics and scientific approaches
Evolution of interpretation techniques
Research activities since 1968
Latest approaches using principal component analysis


Pages: 79 to 108
Creation Date: 1994/01/01

Fifty years of development of opium characterization methods

B. A. REMBERG AND NIKIFOROV Institute of Organic Chemistry, University of Vienna, Austria
G. BUCHBAUER Institute of Pharmaceutical Chemistry, University of Vienna, Austria


In view of the recent call by the Sub-Commission on Illicit Drug Traffic and Related Matters in the Near and Middle East, for the "development of mechanisms to identify, with more precision and through laboratory analysis, the sources of opium seized from the illicit traffic" [ 1] , the present paper reviews the rationale and preconditions for any practical and reliable characterization and origin-correlated classification of opium. In that context, the results of the early international efforts under the aegis of the United Nations from 1951 to 1967, as well as the rather sporadic investigations in this direction since 1968, are described. Finally, it is demonstrated that in spite of the application of modern computer-based technology, the main obstacle to comprehensive opium characterization and typology is still the lack of an extensive reference collection of opium samples of known origin.


During the last fifty years, opium manufacturing and trafficking patterns and trends underwent several major changes, while during the same period, cannabis products, for example, have continuously been front runners in terms of quantities manufactured and trafficked globally.

Until the late 1950s and early 1960s, opium was the illicit product upon which international control efforts concentrated, but its leading role was gradually overtaken by heroin and more recently by cocaine. Also, major changes occurred in the number and location of source areas, manufacturing and trafficking patterns, both on the licit and illicit side of the global opium market.

As a result of continued international efforts to adjust global licit production volumes commensurate to the world's medical needs, the number of licit opium-producer countries was gradually reduced in 1953 from 12 to 7 (Bulgaria, Greece, India, Iran (Islamic Republic of), Turkey, the former Union of Soviet Socialist Republics (USSR), the former Yugoslavia) * [ 2] and ultimately to one single country, India, all other major licit producers abandoning the opium manufacturing industry in the early 1970s [ 3] . The downward trend, which lasted until the late 1970s, made the licit opium production scenario much easier to oversee, by eliminating major sources of leakages towards the illicit processing industry and casing tensions and arguments among interested parties over responsibility concerning suspicious opium shipments and production and trading practices. That, in turn, lessened scientific interest in opium characterization studies for drug interdiction purposes.

During the same period, and still continuing today, two main trends are discernable on the illicit side of opium production and processing: a temporary reduction in the number of source countries was quickly followed by an upward trend both in the number of countries and in the global production volumes [ 4] . The emergence of the Golden Crescent (covering part of Afghanistan, Iran (Islamic Republic of) and Pakistan) as a major producer area in the late 1970s and early 1980s and the rapid spreading of opium poppy cultivation in Latin America during the last decade are the main characteristics of that period [ [ 5] , [ 6] ]. As a result, opium not only remained for a long time the second most important item in terms of globally seized quantities (cannabis products showing the highest seizure figures [ 7] , but also the trafficking patterns and routes have changed frequently. Latest drug enforcement reports indicate that total worldwide illicit opium production has exploded since the 1980s, showing an approximately 152 per cent increase between 1985 and 1992 ( [ 6] , p.16). Present global illicit production volumes are constantly estimated at between 3,000 and 4,000 tonnes ( [ 6] , p.107).

Secondly, and parallel to the above, clandestine opium extraction and heroin manufacturing moved closer to the poppy growing areas, invading practically every major opium production area. Variations in clandestine processing were introduced, including the use of new chemicals resulting in new heroin types and subtypes entering the illicit market. The trend is still continuing, with the emergence of a Latin American heroin- processing industry (Colombian heroin [ 5] ) and a situation threatening to spread and diversify in the Near and Middle East and central Asia [ 4] . As a consequence, easy availability of heroin close to the production areas triggered significant shifts in consumption patterns in various parts of Asia where opium smoking was previously the traditional form of consumption (Iran (Islamic Republic of), Myanmar and Pakistan).

Thus, from the late 1950s and early 1960s heroin gradually emerged as the principal illicit opium-based product on the global market attracting focused interdiction efforts, including the gathering of intelligence information [ 8] .

Characterization of seized drug samples through laboratory analysis has long been recognized as a valuable adjunct to the body of information used in interdicting clandestine drug production and trafficking. Such characterization not only provides sufficient proof of identity for judicial purposes, but may also assist in establishing trafficking patterns and eventually the sources of the products. A number of comprehensive. characterization and comparative profiling programmes exist today for amphetamine, cannabis, cocaine, heroin, 3,4-methylenedioxyamphetamine (MDA), 3,4-methylenedioxymethamphetamine (MDMA) as target drugs [9-13], and others are being developed for similar purposes. Increasing efforts are also directed into establishing linkages through laboratory analysis between the final products (amphetamine, cocaine, MDA, MDMA), the intermediate products and the starting materials themselves (coca leaf, phenyl-2-propanone (P2P), safrol, sassafras oil), [ [ 14] , [ 15] ], as well as the processing techniques used in clandestine manufacture. In contrast, the subject of the opium-morphine-heroin nexus seems to be a case of unjustifiable neglect, and it was felt that a fresh look, including a review of previous efforts and an experimental testing of recent technologies available for the purpose of opium characterization, may be justifiable in the light of the above developments.

The recent gradual proliferation of opium production into new producer countries and regions, as outlined above, is not only reflected in the increasing availability of heroin as the end -product on the global drug markets, but also raises a number of technical law enforcement and drug intelligence questions as well. There is an urgent need for factual information on the types and origin of opium poppies cultivated in the new producer areas, on the opium yields and the production and clandestine processing practices, as well as on the specific characteristics (markers) of the products (crude morphine, heroin and opium). Exact characterization of and discrimination among various opium types is becoming, again, a valuable technical tool in tracing the possible origin and in establishing new trafficking routes. In addition, such characterization may also contribute through the identification of specific markers to the establishment of new heroin types. In that context it is not surprising that various regional and international enforcement bodies, such as the Sub-Commission on Illicit Drug Traffic and Related Matters in the Near and Middle East, concerned about the new dynamics of the illicit opiate market, are stressing, again, the need for concerted effort to gather such information, including the "development of mechanisms to identify, with more precision and through laboratory analysis, the sources of opium seized from the illicit traffic" [ 1] .

*Country names listed in the present article reflect current official usage, which may differ from that of the period referred to.

A short historical account of opium characterization: the United Nations opium research programme, 1951-1967

Opium characterization was the first and by all measures one of the most comprehensive research efforts under United Nations aegis in international drug control. The prevailing global situation with respect to opium availability, the multiplicity of source countries and the repeated debates in international forums prompted the specific resolutions of the Commission on Narcotic Drugs and the Economic and Social Council (159 IIC (VIII), of 3 August 1948, and 246 F (IX), of 6 July 1949) on methods of determining the origin of opium and suggesting an international collaboration programme. The declared aim of the programme was to develop methods for determining the origin of opium by chemical and physical means, and Governments which had the necessary experts and laboratory facilities were invited by the Se6retary-General in 1948 to join the programme (Economic and Social Council resolution 159 (VII) IIC of 3 August 1948). Of 29 countries replying to the invitation, nine Governments stated by 1949 their willingness to participate orto cooperate in the scientific part of the programme (Austria, Bulgaria, Canada, India, Japan, Netherlands, Norway, United Kingdom of Great Britain and Northern Ireland and United States of America) [ 16] . In addition, with regard to a statistically sufficient supply of authentic opium samples as a reliable basis for opium origin determination, all opium -producing countries were requested to cooperate in the programme by furnishing representative samples of opium, officially recognized as being produced on their territory, either legally or illegally. By 1958, altogether 268 authentic opium samples were available [ 17] , originating from Afghanistan (3), Greece (1), India (167), Iran (Islamic Republic of) (32), Japan (14), Republic of Korea (1), Lao People's Democratic Republic (5), Mexiko (1), Myanmar (1), Nepal (1), Pakistan (6), Turkey (34), Viet Nam (1) and the former Yugoslavia (1). Nevertheless, it was noted that considerably more authentic samples were required, especially from countries belonging to regions directly affected by or in the vicinity of sources of illicit traffic (Economic and Social Council resolution 11 (XIV) of April/May 1959). As a result of repeated calls by the Commission, by 1960 altogether 500 authentic opium samples were available and served as a basis for the coordinated research programme during the entire period.

Basis for opium characterization and origin determination

Developing a typology for opium as for any other material is based on two opposite phenomena. On the one hand, there have to be common features of the material in question that determine its identity, while, on the other hand, slight differences elucidated by appropriate analytical methods should enable a distinction to be made between two samples. Since opium is a natural product which is produced with very little human intervention, its physical and chemical characteristics and composition are determined by the following factors:

(a) Factors determining common characteristics:

  1. The plant species, which is always Papaver somniferum L. (Papaveraceae), a relatively stable species in its chemistry;

  2. The traditional manufacturing practices contributing to the consistency in cultivation, processing and handling of the opium within any given farmer community;

  3. Geographical and climatic conditions which tend to be the same or very similar in any smaller geographic area;

(b) Factors contributing to variations in opium composition and characteristics:

  1. The existence of a large number of cultivated opium poppy varieties with widely variable chemical compositions;

  2. Smaller variations caused by annual shifts in climatic conditions;

  3. Small variations occurring during the various stages of manufacturing (longer drying process, storage conditions etc.).

Considering the multiplicity of the factors influencing the com- position of opium, it was clear that any analytical approach to origin determination should elucidate relative differences in the composition of opium samples, based on geographical and traditional variations, rather than qualitative differences, which could hardly be expected because the substance involved is always opium.

Other preconditions for the international collaborative programme were:

  1. An extensive collection of authentic opium samples of known origin covering, ideally, all licit and illicit cultivation areas.

  2. A network of collaborating laboratories to guarantee worldwide replicability of results and to enable a rapid exchange of information.

  3. Simple, specific and reliable analytical techniques.

All those conditions were met by the programme, to which more than a dozen national laboratories and research institutions contributed during its existence, with the United Nations Narcotics Laboratory serving as overall coordinator.

Classification of opium characteristics and scientific approaches

Before describing the analytical techniques used in the United Nations research programme (1951 - 1967), a short survey of the amenable opium characteristics is given. In general, opium features can be classified in two different ways, according to either the analytical methods involved (K55) * or the properties of opium constituents [ [ 17] , [ 18] ]. In the present review the latter approach is used [ [ 17] , [ 18] ].

The following opium properties are used for characterization:

  1. General appearance, e.g. colour, shape, odour, texture, coverings/wrappings, inclusions;

  2. Organic constituents, e.g. alkaloids and meconic and other organic acids;

  3. Inorganic constituents and opium ash, e.g. calcium, magnesium, sodium, potassium and phosphate.

During the period of the United Nations research programme, all the above mentioned features of opium were thoroughly investigated, with the clear intention to reveal reliable origin dependent characteristics. The investigations also required basic research on related topics such as the development of suitable methods for alkaloid separation, for their detection or for the estimation of alkaloid content. The collection of comprehensive data of authentic opium samples as well as the analysis of unknowns according to the methods developed (that is, investigations of opium samples that were unknown to the scientists involved, but known to the United Nations) were additional topics of major importance. As a consequence, altogether 148 scientific papers summarizing the various results were contributed to the United Nations document series ST/SOA/SER.K by an international network of laboratories and scientists, and they can be divided into the following five groups:

  1. Studies of analytical techniques (see table 1);

  2. Investigations into origin dependent opium characteristics (see table 2);

  3. Studies of authentic samples for data collection;

  4. Trials of opium "unknowns", i.e. controlled origin determinations (K28, K36, K52, K94, K97);

  5. Reviews (K25, K55).

*All 148 papers entitled "The assay, characterization, Composition and origin of opium" constitute the United Nations document series ST/SOA/SER.K/1-148. In the text, each reference to one of the papers is given as the number in the K series (i.e. K1) instead of the full title and symbol.

The following two tables give an almost chronological overview of the studies concerning the analytical techniques (table 1) and of scientific investigations into origin-dependent opium characteristics (table 2) between 1951 and 1967 under the United Nations programme. The different approaches also reflect the level of scientific progress during that period.

The approaches that have ultimately been developed for opium characterization are arranged according to the classification scheme (general appearance, organic and inorganic constituents) and are discussed in detail below.

The visual examination of the general appearance of an opium sample (macroscopic evaluation) represents the most obvious and simplest method to obtain the first hints concerning its origin. Besides the overall appearance, that is, shape, colour, lustre, odour and obvious coverings of the sample, more detailed examinations of distinct inclusions, the texture and the firmness of the opium are also of great importance (K25, K55).

In most other approaches of the programme, opium classification was based on techniques measuring specific organic constituents, starting with the alkaloids. One of the earliest and most promising studies of opium characterization used the shapes of noscapine crystals in an aqueous sus- pension of opium on microscopic examination, partially under polarized light [ 19] . The results allowed a classification in three groups of origin (China, India, Iran (Islamic Republic of) Republic of Korea, as the first group, Turkey and the former, Yugoslavia as the second group, and Afghanistan, Japan, Mexico and Mongolia, as the third group). Similar classifications were obtained in later investigations, and thus underlined the high diagnostic value of the method, which was regarded as having "probably the greatest value of any single test for origin determinations" (K21). With the application of the technique to 500 authenticated United Nations samples of known origin in 1960 (K104), a comprehensive correlation between a certain crystal shape and a definite origin was established for' the first time, and thus opened access for the origin determination of unknown opium samples.

The application of various colour reactions in combination with photometric evaluation of the colour intensity but without prior separation of opium constituents was also considered a valuable method, especially because of the short and simple procedure required. The resulting colour value was compared with tabulated values of opium samples of known origin, and thus allowed the classification of an unknown sample (K65). The most promising colour reactions with the various opium constituents are summarized below:

Table 1. Analytical techniques investigated within the programme coordinated by the United Nations

Year of first application


References a/

Methods for morphine extraction/separation
K1, K3, K9, K13, K35
Comparison of different photometric detectors
K10, K14
Infrared spectroscopy in alkaloid determination
K24, K79, K119
Paper chromatography
K27, K37, K42, K50, K51, KSS, K110, K139
Value of graphical data comparison
Simultaneous determination of several compounds
K33, K34, K142
Precipitation method for noscapine
K35, K43
Direct ultraviolet absorption spectrophotometry
K48, K54, K75, K84, K98, K103, K113, K122, K125, K148
Ammonium Reineckate for alkaloid detection
Standardization of methods and apparatus
K57, K62, K66, K96, K110
Paper electrophoresis of alkaloids
Colour reactions
K65, K72, K76, K77, KSO, K81, K93, K105, K116, K121, K125, K147
Automatic recording spectrometer
Latex of Papaver orientale as aid in origin determination
K108, K112, K118
Gas-chromatographic fingerprint
K114, K127, K134, K143
Activation analysis of trace elements
Quantitative TLC/ultraviolet determination of the five major alkaloids
a/ Symbol corresponds to paper in the ST/SOA/SER.K/1-148 series.

Table 2. Investigation into origin-dependent opium characteristics

Year of first application


References a/

Crystals (microscopic examination)
K2, K21, K146
Porphyroxine-meconidine content
K4, K19, K141
Noscapine/meconic acid ratio
K5, K6, K7, K123
Codeine content
K8, K20, K32, K46, K132
Codein/porphyroxine ratio
Morphine content
K11, K15, K18, K29, K32, K39, K40, K44, K90, K92, K101, K118, K131, K135, K136
Nitrogen content in different opium fractions
Thebaine content
K17, K91, K131
Papaverine content
Content of non-phenolic alkaloids
K18, 22, 26, 143
Opium ash
K23, K30, K63/Rev.1, K86
Ratios of codeine, noscapine, morphine
Meconic acid content
K64, K90
Amino acid content
K102, K109
Meconic, sulphuric and total acid content
Trace elements
Determination of fatty acids in opium fat
a/ Symbol corresponds to paper in the ST/SOA/SER.K/1-248 series.
  1. The amount of porphyroxine-meconidine is related to the hydrochloric-acid-induced red colouring of the solution after heating;

  2. Meconic acid gives with Ferric sulphate a brownish-purple chelate;

  3. The so-called Soboleva reaction uses sulphuric acid, formaldehyde, bromine water, ammonia and ethanol to give a yellow to red colouring with papaverine;

  4. The intensity of the strong yellow colour from the reaction with phosphoric acid can be used for the estimation of the thebaine content;

  5. The treatment of an acetic acid extract of opium with sodium nitrite and ammonium hydroxide allows a rough estimation of the morphine content, especially in contrast to codeine, its methylated derivative.

Although most of the colour reactions provided useful information for establishing certain opium types, their specificity was relatively low without prior separation of the various constituents, and ultimately had little value for routine application.

All other methods for the determination of opium alkaloids used an additional isolation step prior to the alkaloid estimation. The most promising extraction techniques of the United Nations programme were lime -water extraction of the opium sample with subsequent determination of porphyroxine-meconidine, codeine and morphine ("short method") (K35) and the so-called unified analysis using counter-current solvent extraction as developed by Fulton in 1955 (K33, K34). That procedure led to the determination of morphine, codeine/cryptopine, thebaine, papaverine, noscapine and minor phenolic alkaloids, with the detection and estimation of the separated alkaloids based on either spectrophotometry (colourimetry) or titration. To that time (1957), it was regarded as the most complete analysis of opium alkaloids as described in the United Nations programme.

Using relative levels of morphine as a single criterion (K15), again three groups of opium origin could be distinguished (the Mediterranean region with Greece, Turkey and the former Yugoslavia, the East Asia region represented by China, Republic of Korea and Mongolia, and the third group consisting of India, Indochina, Iran (Islamic Republic of) and Japan).

Further improvement was the use of reading ratios, such as absorption ratios, of either two alkaloids (K8, K38) or one alkaloid and another organic compound, for example, noscapine and meconic acid (K5), also resulting in the same regions of origin as mentioned above. One of the main advantages of such ratios over single analytical values was that they were less sensitive to process errors because they were based on relative instead of absolute data, and thus provided much more reliable and selective criteria for origin determination.

The next large step forward to a simple and unified method for opium origin determination was made in 1957, applying direct ultraviolet absorption spectrophotometry on aqueous opium extracts. By using ultraviolet ratios instead of individual ultraviolet values, the technique was specific enough even to allow distinctions to he made between regions within one country (such as Croatia and Macedonia within the former Yugoslavia (K48) or the various parts of India).

The availability of chromatographic techniques from the 1960s marks the beginning of a new period in chemical analysis due to possible single - step component separations which enormously accelerated comprehensive identification and characterization of any complex mixture. Thus, further studies in the United Nations collaborative programme were focused either on the investigation of appropriate alternatives to the separation of opium alkaloids by extraction, using paper chromatography, paper electrophoresis, glass-capillary gas chromatography and thin-layer chromatraphy-(TLC), or on the determination of other than alkaloidal features of opium.

While simplicity, speed and sensitivity were emphazised as the main characteristics of paper chromatography and paper electrophoresis (K27, K42, K51, K61), the application of glass-capillary gas chromatography resulted in a so-called fingerprint of the alkaloid composition of an opium sample, and thus allowed the estimation of the major opium alkaloids after a single chromatographic separation step (K114). The use of this technique for the determination of the morphine content (K127), the fatty acids in opium fat (K134) and non-phenolic alkaloids (K143) was also reported. With regard to TLC, a comparison of the results of quantitative TLC/ultraviolet spectroscopy of the five major opium alkaloids with those obtained by the unified analysis revealed the superiority of TLC concerning speed and precision (K129).

Besides the many above-mentioned studies concerning opium alkaloids, the potential of further organic opium constituents as characterization criteria was also investigated: In 1953, the quantitative determination of the various forms of nitrogen in the opium, that is the relationship between the total non-volatile water-soluble nitrogen and the nitrogen amount of morphine (calculated from the morphine percentage), was used for the characterization of the drug (K16), and samples from the Republic of Korea, in particular could be distinguished from others.

With regard to a possible relationship between the amino acid composition of opium and the composition of the soil, in 1960, one group providing quantitative data (K109) of 15 opium samples obtained by an ion exchange chromatographic procedure showed a differentiation between opium from India, Japan and the former Yugoslavia. That was followed by a work in 1961 (K106) describing the estimation of the meconic, sulphuric and total acid content to characterize opium samples and enabling a differentiation even of local opium samples (India).

The use of inorganic constituents as the third group of opium features was first described in 1954 (K23). The study dealt with the possible relationship between the origin of an opium sample and the spectrographic analysis of metallic compounds of the corresponding opium ash, its colour and quantity. The occurrence of relatively great amounts of certain metals seemed to be characteristic for opium samples from certain regions. Four further papers on the same topic were published between 1954 and 1959 (K30, K63/Rev.1, K68, K86), but found "little diagnostic value in the determination of the origin of opium".

Finally, the following methods emerging from the programme were considered to be most useful in establishing the probable origin of an opium sample: general appearance; microscopic examination; simple colour reactions; direct ultraviolet absorption spectrophotometry; and alkaloid content.

Parallel to the above, several research groups were actively involved in the investigation of authentic opium samples (that is samples of known origin) to provide a reliable basis for characterization and origin determination of unknown opium samples by establishing comprehensive data collections. The criteria extensively investigated were macroscopic and microscopic features, alkaloid values, meconic acid content, composition of opium ash, as well as the application of simple colour reactions, direct ultraviolet absorption spectroscopy, paper chromatography and paper electrophoresis (K70 and Add.1 and Add.2). The most comprehensive data collections during that period of the programme were reported in 1960 using microscopic tests, in 1961 applying colour reactions, and in 1962 using direct ultraviolet absorption spectrophotometry on approximately 500 authentic samples each (K104, K115, K116, K120), and represented a reliable basis for the classification of unknown opium samples at that time.

Evolution of interpretation techniques

Not only the analytical techniques themselves, but also the methods available for data interpretation and evaluation gradually evolved during the period of the United Nations opium research programme. While the tabulation of samples arranged in order of origin with the determined values in corresponding columns was the most usual way for collecting similar data from different samples at the beginning, it was of limited practical value for origin determination because of the requirement of a very time-consuming manual comparison process. Within the United Nations programme, tabulation was applied for data collection from over 500 authenticated opium samples.

Other data presentation methods under investigation were histograms (bar charts) that led to the corresponding profile for a given class of components (for example, ash profile (K30) or alkaloidal profile (K31)) or scatter diagrams that combined either two individual (for example, alkaloid concentration) values or two corresponding ratios, and subsequently led to two-dimensional graphical presentations that allowed a simple visual correlation between sample and origin (see figure 1). A first application of discriminatory functions in opium origin determination was mentioned in 1957 (K55). The great advantage of that method (provided the appropriate computer technique is available) is the possibility of including numerous opium characteristics (up to 10 in 1957) in the interpretation step, while scatter-diagrams allowed the use of two values or ratios only. That advantage was due to the fact that the mathematical procedure used reduces the data in a way which finally leads to a two-dimensional plot, and thus allows the clear separation of opium samples of different geographical origin (India, Iran (Islamic Republic of), Turkey).

By far the most important evaluation technique developed within the research programme was a punched -card system, described in 1957 (K56). The special edge-punched double-holed card as shown in figure II allowed cross comparisons of as many as 52 characteristics of an unknown sample with the same data from all previously analysed samples of opium, and 95 per cent probability for correct origin determination were reported (K55). But in spite of the excellent accuracy of the punched- card system, the seven days required for the determination of all necessary characteristics were the main disadvantage of the method and prevented its routine application. Thus, the demand for more rapid methods with the same reliability even in those early days was not surprising (K55).

Graphic examples of the above-mentioned presentation methods were given in two reviews (K55) [ 18]

Research activities since 1968

Further development of the opium characterization methodology was discontinued around 1967 when efforts began to focus on heroin globally. Since then only sporadic efforts have been made to develop new analytical or interpretation technologies. In recent decades both technologies have undergone tremendous improvements in selectivity, simplicity and discriminating power, thus opening the way to much simpler and more powerful characterization methods. Not only separation techniques, such as capillary gas chromatography or high-performance liquid chromatography (HPLC), but also valuable coupled techniques, used in many laboratories today, such as gas-chromatography or liquid-chromatography mass spectrometry, gas-chromatography Fourier-transform infrared spectroscopy and liquid-chromatography ultraviolet spectroscopy, have for the first time opened access to analytical investigations to an extent not imaginable in the early days of the United Nations programme.

Figure I. Scatter diagram: separation of opium samples of 10 different geographical origins using ultraviolet absorption ratios

Full size image: 82 kB, Figure I. Scatter diagram: separation of opium samples of 10 different geographical origins using ultraviolet absorption ratios


Note:E xis the ultraviolet absorption at wavelength x.

Figure II. Punched card used for opium origin determination

Full size image: 120 kB, Figure II. Punched card used for opium origin determination

Source: ST/SOA/SER.K/56.

But the principal handicap in developing an efficient modern technology for opium origin determination is the lack of a comprehensive collection of authentic samples. Old collections such as the one still in existence in the United Nations Narcotics Laboratory are outdated because of the normal ageing process, and it is unlikely that a similar new collection encompassing all major licit and illicit producer countries and areas could be created easily without the firm commitment of all Governments concerned.

Table 3 summarizes a selection of the analytical approaches used since 1968 in order to provide rapid and sensitive methods for the determination of the major opium alkaloids in one single run, thus opening access to routine and automated analyses. Since little or no attention was given to data interpretation with regard to opium classification, the investigations have not improved practical opium characterization methodology.

As noted above, there are only very few recent studies concerning the classification of opium samples according to their source of supply or geographical origin. Within the available literature, the following two kinds of papers can be distinguished:

  1. Papers only describing origin-related opium characteristics;

  2. Papers evaluating origin-related characteristics in order to establish a general scheme for the classification of unknown samples.

While publications within the first group mostly reveal the origin dependence of certain features by concluding that, for example, a certain alkaloid concentration ratio is characteristic for opium from a certain country while other countries exhibit other ratios, so far only one paper has described further steps towards a practical application of the data obtained by providing statistical means to handle numerous values of as many known opium samples as possible, thus forming the basis for an origin determination of unknown opium samples [ 68] .

Most studies since 1968 used, however, the alkaloid content of opium, that is, qualitative data reflected by the alkaloid spectra or quantitative data on the main alkaloids, for origin correlated statements [ [ 21] , [ 31] , [ 69] , [ 70] , [ 71] ]. Thus, for example, the contents of the five major alkaloids determined after gas- chromatographic separation were shown to be clearly different in opium samples originating from India and Japan [ 21] . The morphine content on its own, estimated by means of a so-called cerimetric method, also provided origin-dependent differences [ 61] . A study in 1989 recommended derivative ultraviolet spectroscopy [ 72] for the identification and estimation of morphine, emphasizing the possibility of differentiating between genuine and spurious opium samples on the basis of better results for morphine determination than those obtained by calorimetric evaluation.

Table 3. Analytical techniques applied for opium characterization after 1968



References a/

Gas chromatograph
Separation/quantification of major opium alkaloids
Gas chromatograph
Study of detection devices (e.g. thermoaerosol, electron capture)
Gas chromatograph mass spectrometry
Separation/identification/quantification of opium alkaloids/meconic acid
Desorption chemical ionization-mass spectrometry
Single step identification of all characteristic compounds of the sample
Ion pair chromatography
Separation of alkaloids
Study of stationary phases
Study of solvent systems
Study of detection devices
Detection by simple colour reactions
Overpressed TLC
Advantages over TLC or HPLC
Study of stationary phases and importance of reversed phase HPLC
Study of solvent systems (gradient/isocratic technique)
Study of detection devices
Electrochemical methods
Identification/quantification of opium compounds
Quantification of opium alkaloids
Circular dichroism
Simultaneous determination of alkaloids
Subcritical/supercritical fluid chromatography using C0 2 as mobile phase
Separation of opium alkaloids
Atomic absorption spectrometry
Trace determination of inorganic compounds (e.g. copper)
a/ Numbers refer to the list of article references.

During the same time period, considerable progress was achieved worldwide with regard to heroin source determination using the impurity - profiling or chemical signature approach instead of the usual alkaloid or main component determinations. Although the research was concentrated on heroin, a few studies also included crude morphine and opium as source materials. The exact analytical procedure (extraction, derivatization and gas chromatography separation) was described by Neumann/Gloger [ 73] in 1982. It revealed the presence of many fatty acids and other natural non-alkaloid substances (summarized as trace impurities), besides the whole spectrum of opium alkaloids. Subsequent visual comparison of the chromatograms of different opium samples provided the corresponding sample profiles that may serve as origin determining criterion. Further research in 1987 was aimed at achieving better resolutions of the investigated compounds by using a gas chromatography, capillary-to-capillary column-switching system [ 74] .

The first detailed paper on opium origin correlation was published by Narayanaswami and others in 1979 [ 68] , more than 12 years after the end of the United Nations research programme. Altogether 28 opium samples including 18 samples of known origin (nine authenticated United Nations samples and nine known samples from different regions of India) and 10 seized opium samples were investigated, with the separation and estimation of the five major alkaloids achieved by the technique of gas liquid chromatography. A so-called triangular plot was established using the relative percentage of morphine and two percentage ratios of the remaining four major alkaloids (noscapine/papaverine and codeine/thebaine) as coordinates (see figure III). Hence, this plot corresponds to the already-mentioned scatter diagrams of the United Nations research programme, the only difference being in the number of the values included. While scatter diagrams combine two values or ratios only, triangular plots were established by the combination of three single concentrations or concentration ratios. As a result of the study, the graphical differentiation of five well-defined areas correlating with five geographical origins was claimed by the authors analysing nine authenticated United Nations samples, and was thus considered as a basis for the origin determination of unknown opium samples. In addition, the separation of 18 Indian opium samples in two groups, evidently representing a much closer geographical area, was reported.

Figure III. Triangular plot: separation of opium samples of different geographical origins using alkaloid percentage or percentage ratios

Full size image: 72 kB, Figure III. Triangular plot: separation of opium samples of different geographical origins using alkaloid percentage or percentage ratios

Source:K. Narayanaswami, H. C. Golani and R. D. Dua, "Assay of major and minor constituents of opium samples and studies of their origin", Forensic Science International, vol. 14, 1979, pp. 181-190.

Latest approaches using principal component analysis

A comprehensive literature search by the authors in 1993 revealed that there is no reliable, rapid method for opium origin determination. Considering the renewed interest in scientific characterization methods of the drug, the authors started experimental work in 1993 on the topic, especially making use of a suitable computer technique that has already proved its usefulness in many other kinds of multifactorial correlations [ 75] . Two different analytical approaches applying that computer- aided evaluation method (as described below) were adapted, one using HPLC quantitation of the five major opium alkaloids, while the second was based on quantitation of nitrogen-containing volatiles by gas chromatography with a nitrogen phosphorous detector.

For the determination of quantitative HPLC data of the five major opium alkaloids (morphine, codeine, thebaine, papaverine and noscapine), 27 opium samples of known origin from countries and areas in different parts of the world (Afghanistan, Austria, Greece, Hong Kong, Iran (Islamic Republic of), Japan, Mexico, Turkey, former USSR, and the former Yugoslavia) were available. Subsequently, principal component analysis (PCA) was applied to evaluate the data and to correlate alkaloid concentrations to the originis of the samples [ 76] .

PCA is a multivariate evaluation method that allows the optimal presentation of numerous features of a sample (that is, multifactorial data) in a two - dimensional graph after computer - aided data reduction, and is thus superior to the already mentioned discriminatory functions of the United Nations research programme. In the resulting plot (see Figure IV), every point represents a particular sample, and via clustering, distinct groups of similar samples correlated to distinct geographical origins can be visualized. Thus, PCA allows a classification of the investigated samples according to a property (for example, origin) that is not easily accessible from the original, multifactorial data.

Concerning the authors' investigations into opium origin determina- tion, the four clusters of figure IV were proposed to be denoted as follows: India, southern Europe (Greece, western Turkey, former Yugoslavia), Near and Middle East (Afghanistan, eastern Turkey, Iran (Islamic Republic of)) and Far East (Hong Kong, Japan), while the remaining three countries and territories (Austria, Mexico, former USSR) did not match the classification scheme. It is worth mentioning, however, that the classification emerging from the PCA application is quite similar to the correlations obtained by the scientists involved in the United Nations research programme.

Figure IV. PCA projection of 27 opium samples, each characterized by the concentrations of the five major alkaloids, in the plane defined by the two greatest variances

Full size image: 82 kB, Figure IV. PCA projection of 27 opium samples, each characterized by the concentrations of the five major alkaloids, in the plane defined by the two greatest variances

Source:B. Remberg and others, "Principal component analysis (PCA) of opium alkaloid contents for origin determination", Pharmazie,1994, in press.

Note:The closer two asterisks in this projection, the more similar are the corresponding opium samples.

In a further step, the practical value of the achieved origin classification was verified by applying the same computing process to extended data, that is, to on corresponding alkaloid concentrations taken from the literature [ 68] . And in spite of different analytical methods used for the alkaloid separation prior to their estimation (HPLC in the authors' own investigations versus gas liquid chromatography in [ 68] ), any of the opium samples was located in the plot (that is, in the cluster) according to its origin as denoted by Narayanaswami and others [ 68] .

PCA therefore seems to be of value for future origin determinations provided that the appropriate computer technique and the corresponding data from a sufficient number of authentic opium samples are available. In that case, reliable cluster areas for as many countries or regions of origin as possible can be established that may form, in turn, the basis for the classification of unknown opium samples. Moreover, since origin correlations become more reliable if more characteristics are included in the evaluation process, it should be emphasized that PCA allows the handling of as many different features of the investigated sample as available (for example, alkaloid concentrations, colour values and ultraviolet ratios), in other words, it does not necessarily require uniform data, such as the exclusive use of concentration values.

In a related research direction, PCA was used to estimate the value of quantitation data on nitrogen-containing volatiles for the characterization of different opium samples by gas chromatography with a nitrogen phosphorous detector. In general, the authors' studies of the odour of opium [ 77] were motivated by the fact that in spite of the knowledge of clear scent differences among opium samples of different geographical origins (K25) [ 78] , that feature had never been investigated in detail before.

As a result of comprehensive analytical investigations, altogether 15 nitrogen-containing volatiles, mainly alkyl- and alkoxy-substituted pyrazines, were proved to be responsible for the characteristic odour of opium samples [ [ 77] , [ 79] ]. With that identification of the odour relevant opium constituents, a first step towards an opium origin determination based on scent characteristics was made. In a logical next step, gas chromatography with nitrogen-selective detection was applied to opium samples of different geographical origins (Austria, Hong Kong, India, Iran (Islamic Republic of), Malaysia, Thailand) resulting in different, origin -dependent patterns of nitrogen -containing volatiles. That data was then used for the calculation of the concentrations of the corresponding compounds, which, in turn, were evaluated by PCA.

Since the work undertaken is only in its initial stage, further work on the topic is necessary in order to provide the pattern of odour-relevant nitrogen-containing volatiles of opium as an additional tool in opium characterization.



"Situation and trends in drug abuse and the illicit traffic", report on the 21st session of the Sub-Commission on Illicit Drug Traffic and Related Matters in the Near and Middle East, Vienna, 6 February 1987 (E/CN.7/1987/CRP.1).


"United Nations Opium Conference Protocol and Final Act of 23 June 1953", Bulletin on Narcotics, (United Nations publication), vol. 5, No. 3 (1953), pp. 54-69.


Report of the INCB for 1980: Demand and Supply of Opiates for Medicinal and Scientific Needs (United Nations publication, sales No. E.82.XI.4).


"Report of the regional workshop on the consumption of opiate drugs in Asia", held at Penang, Malaysia, from 9 to 12 December 1981.


Colombian Heroin - a Baseline Assessment (Johnstown, Pennsylvania, National Drug Intelligence Center, April 1994).


United States of America, National Drug Control Strategy. Reclaiming Our Communities from Drugs and Violence (Washington, D.C., Office of National Drug Control Policy, Executive Office of the President, February 1994), p. 16.


H. Neumann, "Analysis of opium and crude morphine samples by capillary gas chromatography. Comparison of impurity profiles", Journal of Chromatography, vol. 315, 1984, pp. 401-411.


Report of the 18th session of the Commission on Narcotic Drugs, 29 April-17 May 1965 (E/CN.7/455, para. 181).


J. F. Casale and J. W. Watterson, "A computerized neural network method for pattern recognition of cocaine signatures", Journal of Forensic Sciences, vol. 38, No. 2 (1993), pp. 292-301.


B. A. Perillo, R.F.X. Klein and E. S. Franzosa, "Recent advances by the U.S. drug enforcement administration in drug signature and comparative analysis, International Workshop on Impurity Profiling Analysis of Illicit Drugs, Proceedings of the International Symposium of Forensic Science, Grand Hill Ichigaya, Tokyo, from 21 to 22 October 1993.


I. S. Lurie and others, "Analysis of manufacturing by-products and impurities in illicit cocaine via high-performance liquid chromatography and photodiode array detection", Journal of Chromatography, vol. 405, 1987, pp. 273-281.


M. LeBelle and others, "Comparison of illicit cocaine by determination of minor components", Journal of Forensic Sciences, vol. 36, No. 4 (1991), pp. 1102-1120.


A. Sinnema and A.M.A. Verweij, "Impurities in illicit amphetamine: a review", Bulletin on Narcotics (United Nations publication), vol. 33, No. 3 (1981), pp. 37-54.


J. M. Moore and others, "Determination and in-depth chromatographic analyses of alkaloids in South American and greenhouse- cultivated coca leaves", Journal of Chromatography A, vol. 659, 1994, pp. 163-175.


R. J. Renton, J. S. Cowie and M.C.H. Oon, "A study of the precursors, intermediates and reaction by-products in the synthesis of 3,4-methylenedioxymethylamphetamine and its application to forensic drug analysis", Forensic Science International, vol. 60, 1993, pp. 189-202.


Methods of Determining the Origin of Opium by Chemical and Physical Means (United Nations Document E/CN.7/159, 15 April 1949; Add.1,9 May 1949; and Add.2,31 May 1949).


"Committee of Experts on the United Nations Programme of Opium Research", Bulletin on Narcotics (United Nations publication), vol. 10, No. 3 (1958), pp. 37-44.


S. Pfeiffer, "Der illegale Handel und Mibrauch von Opium und Mohnalkaloiden sowie die Methoden ihrer Bekämpfung", Pharmazie,vol. 15, 1960, pp. 322-334.


C. C. Fulton, "Photomicrographs of opium crystals", United Nations Document E/ CN.7/117,Part IIA, IIC, 14 April 1948.


S. Mandal, A. A. Naqvi and R. S. Thakur, "A gas chromatographic method for the quantitative determination of opium alkaloids from plant", Indian Journal of Pharmaceutical Sciences, vol. 55, No. 1 (1993), pp. 25-27.


M. Ono, M. Shimamine and K. Takahashi, "Gas chromatographic determination of the major alkaloids in opium", Eisei Shikensho Hokoku, vol. 95, 1977, pp. 4-9.


G. Fisher and R. Gillard, "GLC determination of opium alkaloids in Papaveretum", Journal of Pharmaceutical Sciences, vol. 66, No. 3 (1977), pp. 421-423.


D. Furmanec,"Quantitative gaschromatographic determination of the major alkaloids in gum opium", Journal of Chromatography, vol. 89, No. 1 (1974), pp. 76-79.


A. Bechtel, "Gas-chromatographic identification and quantitative determination of morphine, codeine, thebaine, papaverine, and narcotine in opium extract", Chromatographia, vol. 5, No. 7 (1972), pp. 404-407.


A. R. Sperling, "Analysis of alkaloids in opium", Journal of Chroma - tography, vol. 294, 1984, pp. 277-302.


B. A. Rudenko, N. V. Lenchik and D. N. Dzabarov, "Gas chromato- graphic determination of nitrogen-containing anticonvulsant drugs and opium alkaloids using a thermoaerosol detector", Journal of Chromatography, vol. 364, 1986, pp. 369-376.


P. O. Edlund, "Determination of opiates in biological samples by glass capillary gas chromatography with electron -capture detection", Journal of Chromatography, vol. 206, No. 1 (1981), pp. 109-116.


Y. Suzuki and others, "Application of mass spectrometry to the fields of forensic chemistry. H. Determination of opium alkaloids", Eisei Kagaku,vol. 19, No. 4 (1973), pp. 212-214.


W. Arnold, "GC/MS-determination of opium alkaloids in opium preparations", Beitrage zur Gerichtlichen Medizin, vol. 32, 1974, pp. 199-205.


Y. Ohno and S. Kawabata, "Rapid detection of illicit drugs by direct inlet chemical ionization mass spectrometry", Reports of the Central Customs Laboratory, Ministry of Finance. - 27 Matsudo, Japan, 1987, pp. 7-15.


T. Veress, "Determination of opium alkaloids from samples of plant origin by reversed phase ion pair chromatography", Magyar Kemiai Folyoirat (Hungarian Journal of Chemistry), vol. 92, No. 2 (1986), pp. 54-58.


E. F. Matantseva and others, "Quantitative determination of opium alkaloids by liquid chromatographic methods", Khimiya Prirodnykh Soedinenii (Uzbekistan), No. 5, 1980, pp. 730-731.


M. Skrlj, "Kaolin, an adsorbent for thin-layer chromatography. III. Separation of opium alkaloids and antipyretic mixtures on kaolin adsorbent", Arhiv za Farmaciju (Belgrade), vol. 37, No. 6 (1987), pp.265-271.


F. Machovicova, L. Mesarosova and W. Stalmach, "Determination of morphine in opium and opium tincture according to PhBs III method. 2. Comparison of pharmacopeial methods with pharmacopeial methods using column chromatography for isolation of morphine and related alkaloids", Farmaceuticky Obzor (Bratislava), vol. 46, No. 10 (1977), pp. 449-453.


S.Durresi, "Thin-layer chromatographic separation and characterization of the principal alkaloids of opium from Albanian poppies", Buletini Universiteti Shteteror i Tiranes, Fakulteti i Shkencave Mjekesore, vol. 22, No. 3 (1982), pp. 27-33.


L. Iliev, "Thin-layer chromatographic characterization of the more important alkaloids of poppy. II ", Farmatsiya (Sofia), vol. 31, No. 1 (1981), pp. 1-11.


0. M. Salama, and M. I. Walash, "Densitometric determination of papaverine in opium, poppy capsules and certain pharmaceutical dosage forms", Analytical Letters, vol. 24, No. 1 (1991) pp. 69-82.


Z. Blagojevic, M. Skrij and V. Bulajic, "Determination of opium alkaloids by planimetry after separation by thin-layer chromatography", Arhiv za Farmaciju (Belgrade), vol. 22, No. 2 (1972), pp. 97- 102.


V. E. Chichiro, Z. P. Kostennikova and S. D. Mekhtikhanov, "Densitometric determination of opium alkaloids after separation by thin-layer chromatography", Farma tsiya (Moscow), vol. 20, No. 6 (1971), pp. 37-42.


V. Massa and others, "Photodensitometric determination of alkaloids after thin-layer chromatography. 11. Principal alkaloids of opium", Travaux Societe de pharmacy de Montpellier, vol. 30, No. 4 (1970), pp. 273-282.


L. Ling, "TLC-UV spectrophotometric determination of morphine in opium tincture", Yaowu Fenxi Zazhi (Beijing), vol. 5, No. 4 (1985), pp. 229-230.


S. Dong, S. Li and D. An, "Determination of morphine content in opium tincture by a dual-wavelength TLC scanning method", Yaowu Fenxi Zazhi (Beijing), vol. 2, No. 2 (1982), pp. 68-71.


N. R. Ayyangar, S. S. Biswas and S. Tambe, "Separation of opium alkaloids by thin-layer chromatography combined with flame ionization detection using the peak pyrolysis method", Journal of Chromatography, vol. 547, No. 1-2 (1991), pp. 538-543.


K. Roeder, E. Eich and E. Mutschler, "Direct quantitative evaluation of thin-layer chromatograms by reflectance and fluorescence measurements. III. Determination of morphine, codeine, thebaine, papaverine, and narcotine in opium by reflectance measurements", Archiv der Pharmazie (Weinheim, Germany), vol. 304, No. 4 (1971), pp. 297-306.


E. Stahl and H. Jahn, "Determination of opium alkaloids with TLC and HPLC", Pharmaceutica Acta Helvetiae, vol. 60, No. 9-10 (1985), pp. 248-252.


T. R. Baggi, N.V.R. Rao and H.R.K. Murty, "Visualisation of opium alkaloids on TLC plates by Gibbs Reagent Spray" Forensic Science, vol. 8, No. 3 (1976), pp. 265-267.


J. Pothier, N. Galand and C. Viel, "Determination of some narcotic and toxic alkaloidal compounds by overpressured thin-layer chromatography with ethyl acetate as eluent", Journal of Chromato- graphy, vol. 634, No. 2 (1993), pp. 356-359.


M. Popl, J. Faehnrich and Le Duy Ky, "Influence of stationary phase on the separation of opium alkaloids using reversed-phase liquid chromatography", Sbornik Vysoke Skoly Chemicko- Technologicke Praze, Analyticka Chemie, vol. H21, 1986, pp. 43-50.


G. Ruecker, "Application of extrelut columns in alkaloid determina- tion", Deutsche Apotheker-Zeitung, vol. 119, No. 29 (1979), pp. 1163 - 1164.


V. K. Srivastava and L. M. Maheshwari, "Liquid chromatographic determination of major alkaloids in gum opium", Journal of the Association of Official Analytical Chemists, vol. 68, No. 4 (1985), pp. 801-803.


B. Proksa and J. Cerny, "Determination of narcotoline", Pharmazie, vol. 36, No. 5 (1981), pp. 380-381.


K. R. Khanna and S. Shukla, "HPLC investigation of the inheritance of major opium alkaloids", Planta Medica (Germany), No. 2 (1986), pp. 157-158.


N. R. Ayyangar and S. R. Bhide, "Separation of eight alkaloids and meconic acid and quantitation of five principal alkaloids in gum opium by gradient reversed-phase high-performance liquid chromatography", Journal of Chromatography, vol. 436, No. 3 (1988), pp. 455-465.


N. R. Ayyangar and S. R. Bhide, "Simultaneous separation of the principal alkaloids in gum opium by isocratic, reversed-phase high-performance liquid chromatography", Journal of Chromatography, vol. 366, 1986, pp. 435-438.


C. Y. Wu and J. J. Wittick, "Separation of five major alkaloids in gum opium and quantitation of morphine, codeine and thebaine by isocratic reverse phase high performance liquid chromatography", Analytical Chemistry, vol. 49, No. 3 (1977), pp. 359-363.


S. H. Hansen, "HPLC assay of the opiates in opium and cough mixtures using dynamically modified silica and UV absorbance, fluorescence and electrochemical detection", International Journal of Pharmaceutics, vol. 32, No. 1 (1986), pp. 7-11.


H.A.H. Billiet and others, "Separation and identification of illicit heroin samples by liquid chromatography using an aluminia and C18 coupled column system and photodiode array detection", Journal of Chromatography, vol. 368, No. 2 (1986), pp. 351-361.


H. Wang, "Determination of morphine in drugs by catalytic polarography", Zhongguo Yaoxue Zazhi, vol. 26, No. 10 (1991), pp.606-608.


N.V.R. Rao and S. N. Tandon, "Amperometric determination of opium and strychnos alkaloids with potassium iodotri(iodo)thal- late(I)", Talanta,vol. 27, No. 5 (1980), pp. 449-450.


K. Nikolic, R. Popovic and N. Svecenski, "Conductometric determination of some opium alkaloids in dimethyl sulfoxide", Arhiv za Farmaciju (Belgrade), vol. 27, No. 3 (1977), pp. 157-162.


P. Surmann, "Voltammetric behavior of opium alkaloids", Archiv der Pharmazie (Germany), vol. 312, No. 9 (1979), pp. 734-740.


S. S. Chausovskii and M. Brombergiene, "Potentiometric determination of morphine in an opium-benzoin tincture", Farmatsiya(Moscow), vol. 16, No. 4 (1967), pp. 59-61.


N.V.R. Rao, N. J. Singh and S. N. Tandon, "Determination of opium and strychnos alkaloids by radiometric titrations with potassium thallium(I)iodide reagent", Isotopenpraxis, vol. 16, No. 1 (1980), pp. 29-31.


U. Wieczorek and others, "Radioimmunoassay determination of six opium alkaloids and its application to plant screening", Phytochemistry,vol. 25, No. 11 (1986), pp. 2639-2646.


S. M. Han and N. Purdie, "Simultaneous determination of opiates by circular dichroism", Analytical Chemistry, vol. 58, No. 1 (1986), pp. 113-116.


J. L. Janicot, M. Claude and R. Rosset, "Separation of opium alkaloids by carbon dioxide sub- and supercritical fluid chromatography with packed columns", Journal of Chromatography, vol. 437, 1988, pp. 351-364.


B. Salvesen, "Atomic absorption spectroscopy in analytical pharmaceutical chemistry. Trace determination of copper in drugs", Meddelelser Fra Norsk Farmaceutisk Selskap, vol. 32, No. 6-7 (1971), pp. 79-86.


K. Narayanaswami, H. C. Golani and R. D. Dua, "Assay of major and minor constituents of opium samples and studies of their origin", Forensic Science International, vol. 14, 1979, pp. 181-190.


G. Pruner, "Significance of the cerimetric method in the determination of morphine production from opium", Herba Hungarica, vol. 15, No. 1 (1976), pp. 97-105.


S. N. Tewari, "Paper chromatographic detection of very small amounts of opium alkaloids and determination of their origin", Archiv für Kriminologie, vol. 139, No. 5-6 (1967), pp. 125-30.


A. R. Wijesekera, K. D. Henry and P. Ranasinghe, "The detection and estimation of (A) arsenic in opium, and (B) strychnine in opium and heroin, as a means of identification of their respective sources", Forensic Science International, vol. 36, No. 3 - 4 (1988), pp. 193 - 209.


V. Singh, S. L. Shukla and J. S. Mahanwal, "Identification and estimation of morphine in opium samples by derivative UV spectroscopy', Journal of the Indian Academy of Forensic Sciences, vol. 28, No. 2 (1989), pp. 1-5.


H. Neumann and M. Gloger, "Profiling of illicit heroin samples by high-resolution capillary gas chromatography for forensic applica tion", Chromatographia, vol. 16, 1982, pp. 261-264.


H. Neumann and H. P. Meyer, "Application of a gas chromatographic capillary-to-capillary column-switching system to the analysis of complex illicit drug samples", Journal of Chromatography, vol. 391, No. 2 (1987), pp. 442-447.


J. Devillers and W. Karcher, eds., Applied Multivariate Analysis in SAR and Environmental Studies (Dordrecht, The Netherlands, Kluwer Academic Publishers, 1991).


B. Remberg and others, "Principal component analysis (PCA) of opium alkaloid contents for origin determination", Pharmazie, 1994, in press.


B. Remberg, "Untersuchungen zum Geruch von Opium", dissertation, University of Vienna, Austria, 1993.


E. Steinegger and R. Hänsel, Pharmakognosie,5th ed. (Berlin, Springer-Verlag, 1992), p. 529.


G. Buchbauer, B. Remberg and A. Nikiforov, "Headspace constituents of opium", Planta Medica, vol. 60, No. 2 (1994), pp. 181-184.