Abstract
I. Introduction
II. Experimental
III. Results and discussion
IV. Conclusion
V. Acknowledgements
Author: Lim Han YONG, , Ng Tju LIK
Pages: 45 to 74
Creation Date: 1977/01/01
The urinary excretion patterns of morphine and codeine in a number of individuals following consumption of morphine-based narcotic drugs have been studied. From the data collected the relative amounts of codeine and morphine in urine specimens were compared and certain trends, consistencies and irregularities were revealed. To a certain extent, the ratio of the proportion of codeine to morphine excretion in urine may be used to determine the nature of the drugs consumed. This correlation is useful in the interpretation of the results of urine analysis which is the basis of effective control and rehabilitation of drug addiction.
The abuse of narcotic drugs has become a major problem of our time, incurring great economic and social losses. To combat this serious problem, it is necessary to have a thorough knowledge of not only the physiological and psychological effects of these drugs, but also the transformation of the drugs in the human body and their eventual detoxification and excretion from the body. This knowledge of the fate of the drugs in the body will assist chemists, toxicologists, medical doctors, lawyers and law enforcement officers to understand the circumstances concerning the intake of the drugs, the extent of the abuse of the drugs and in the interpretation of the results of chemical analysis of the blood and urine specimens taken from suspected abusers and addicts.
In Singapore, the problem of narcotic abuse can be roughly classified into the following four categories: ( a) smoking of cannabis (Ganja); ( b) smoking of opium; ( c) injection of morphine; ( d) smoking of heroin (diamorphine). The use of morphine and heroin are relatively recent phenomena and without doubt, they are the most serious forms of drug abuse.
The consumption of morphine naturally leads to the excretion of morphine (mostly in conjugated form) which is perhaps also accompanied by a small quantity of codeine as a result of the methylation of morphine in the metabolic processes (1). The illicit morphine powder peddled by drug pushers and used by morphine addicts ("street morphine") however, often contains a significant amount of codeine. This is because morphine is obtained by the extraction of opium which normally contains about 10 per cent morphine and 0.5 per cent codeine. The consumption of "street morphine" would therefore be expected to result in the excretion of both morphine and codeine in urine. We have studied the urinary excretion pattern of human subjects following the consumption of medicinal morphine for comparison with the urinary excretion pattern of drug addicts.
In the case of the consumption of opium, the main metabolites excreted in the urine are morphine and codeine glucuronides. These can be readily detected in the laboratory and it is rather surprising that frequently a ratio of 1 to 2 can be observed in the proportion of codeine to morphine in urine. We have studied the excretion patterns of morphine and codeine in the urine of human subjects following consumption of medicinal opium.
Codeine is not a common drug of addiction in Singapore, it is however a drug used extensively in medicinal preparations for the treatment of cough and other minor ailments as an analgesic and antitussive. The consumption of codeine naturally leads to the excretion of codeine (conjugated) in urine. However, as O-demethylation is a well-established metabolic pathway, biotransformation of codeine to morphine may be important [ 1] . As a result, morphine may also be detectable in the urine of a person consuming codeine. We have investigated the urinary excretion of morphine and codeine in human subjects after the consumption of codeine contained in medicinal preparations and the results indicated that although both morphine and codeine are excreted in some subjects, it is possible to use the ratio of the amount of codeine and morphine excreted to distinguish a case of "street morphine" from one of codeine consumption.
Heroin (diamorphine or diacetylmorphine) can be easily hydrolysed to monoacetylmorphine (MAM) and then morphine. This hydrolysis which involves de-acetylation is a common metabolic process [ 1] , [ 2] , [ 3] . The use of heroin would therefore cause morphine to be detectable in urine whereas MAM and heroin itself would be difficult to detect due to the almost complete hydrolysis and metabolic conversion. The illicit Asian heroin ("No. 3 Heroin") used by drug addicts in Singapore also contains significant quantity of acetylcodeine which is a by-product in the manufacture of the illicit heroin [ 4] , [ 5] , [ 6] . The manufacture of this heroin uses crude morphine extracted from opium which contains both morphine and codeine and on acetylation both diacetylmorphine (heroin) and acetylcodeine are formed. Acetylcodeine when consumed will also hydrolyse in the body to codeine and will be excreted as such. Therefore it is to be expected that both morphine and codeine would be detectable in the urine of a heroin abuser.
This is illustrated by a study of the urine excretion of a known heroin addict and from the analysis of the urines of thousands of heroin abusers caught by law enforcement officers.
(a) Collection of urine samples -
A fixed dosage of medicine containing morphine or codeine was administered to volunteers usually in the morning after they had discharged their urines. Thereafter, samples were collected at every hour or other convenient intervals and preserved by adding a pinch of boric acid. The volume of the urine collected was recorded and one-tenth of the volume was used for analysis.
The urine excretion pattern of multiple doses of a medicine was obtained by following the urinary excretion of patients undergoing a course of treatment using the medicine.
The urine specimens of a confirmed heroin addict was obtained through the co-operation of the Central Narcotics Bureau.
(b) Chemical analysis -
The procedure for the analysis has been well established [ 7] , [ 8] . Urine sample (one-tenth of total collected) was first hydrolysed to release morphine and codeine from their conjugated glucuronides [ 9] , [ 10] . This was achieved by adding one-fifth of the urine volume of concentrated hydrochloric acid to each sample and autoclaving the solution at 15 lb pressure for 10 minutes.
The solution was cooled and made alkaline with strong ammonia solution and extracted with 15 ml of a mixed solvent of chloroform-isopropanol (9 + 1).
The organic solvent phase was separated, filtered and extracted with 4 ml of N/2 HCl. After separating this aqueous phase, the pH was adjusted by the addition of 5 ml borate buffer (pH 9.2) and then extracted with 15 ml of the chloroform-isopropanol mixed solvent. The organic solvent phase was then separated and evaporated on the water bath. The residue was dissolved in ethanol and spotted completely onto a thin layer chromatography (TLC) plate coated with silica gel G (acc. to Stahl, type 60 Merck).
The TLC plate was then eluted in a tank with a fresh mixture of ethylacetate, methanol and ammonia (17 + 2 + 1). After drying in air, the plate was sprayed with an acidified iodoplatinate spray. The alkaloids were identified by comparison with reference standards also spotted on the TLC plate and by their colour development. Morphine gives a blue reaction while codeine and heroin give violet reactions. These alkaloids are well-separated on the chromatogram. The drugs were quantitated by comparing the intensity of the TLC spots with spots containing known amount of the reference standards prepared separately under the same conditions.
We have found that the efficiency of this extraction procedure was about 70 per cent, i.e. the recovery of the drugs was about 70 per cent. This is comparable with the recovery studies of Mule et al. [ 11] .
The detection limit of this TLC technique was about 1 µg of morphine or codeine and when 10 ml of urine was used, the detection limit of the procedure was around 0.2 µg/ml (0.2 ppm) after taking into account the percentage recovery.
Gas-liquid chromatography (GLC) was also employed, in the analysis of some of the samples for comparison with the TLC results. A Perkin-Elmer 900 Gas Chromatograph with flame ionisation detector was used. The column was a 3 feet, 1/4 inch diameter coiled glass tubing packed with 3 per cent CV-17 on Gas-Chrom Q 80/100 mesh. Oven temperature of 235°C and injection and manifold temperature of 250 °C were employed.
As morphine is adsorbed strongly on the column and showed poor response, prior acetylation of morphine with acetic anhydride and pyridine was necessary for the determination of morphine. The detection limit for morphine (after acetylation) and codeine was around 0.03 µg/ml when 25 ml of urine was extracted.
The results of the analysis are presented graphically in the following sections. It may be noted that the drug determined in each case was the total excreted drug including both the conjugated (glucaronides) and the free drugs. In each case two graphs are presented; in one, the total hourly drug excretion and in the other, the drug concentration, are plotted against time (hour). Whenever circumstances prevented the collection of hourly urine sample, it was collected at intervals of 2 hours, 3 hours or more; in these cases, the total drug excretion found on analysis was equally apportioned for each hour in the interval and plotted on the graphs. These are identified by the dotted lines joining together these points. This is admittedly a simplified technique of apportioning the excreted drug but it greatly facilitated the analysis of the large volume of data collected.
In interpreting these graphs, it must be borne in mind that the points plotted for each hour reflect only the state of the total urine at the end of that hour and should not be taken to represent the instantaneous value of drug excretion at that moment. Therefore, the curve joining these points merely traces out the average hourly change in the drug excretion rather than the instantaneous variation. The total drug excreted over the whole period covered by the collection is, however, still given by the sum of the various hourly values in the left hand side graph.
Using this method of data analysis, it is possible to obtain an overview of the fluctuations of the drug excretion and to observe the initial increase of the drug excretion to maximum (in terms of either total drug excretion or concentration of drug in urine) and the subsequent gradual decrease to non-detectable levels.
It may be noted that because we monitored excretion pattern at hourly intervals, the actual "minute-to-minute" variation of the drug excretion in the body metabolism has been smoothed out to a certain extent. The maximum of excretion therefore tends to be lowered by the averaging over the one hour period and the minimum tends to be correspondingly raised thus resulting in a flattened out curve. Fortunately it appeared that the one-hour interval used was not too long a period and the fluctuation in the drug excretion was still clearly discernable.
(A) Morphine - oral administration of 2 mg medicinal morphine hydrochloride, equivalent to 1.5 mg of morphine base. Considering the normal therapeutic injection dosage of morphine sulphate of 7.5-22.5 mg morphine [ 12] and the therapeutic oral dosage of morphine sulphate tablets of 7.5-15.0 mg morphine [ 13] , this experimental single dose of 1.5 mg morphine is relatively low and would be expected to be well below the usual level of misuse of illicit morphine. Nevertheless, it was already sufficient to give a significant recovery of the drug in the urines of human subjects.
Graphs A (i) to A (v) give the results of these experiments. It can be seen that the time taken to reach the maximum excretion of the drug in urine varies between individuals and can range from 1 hour to 3 hours after administration. In terms of concentration of the drug in urine (see right hand side of graphs), the maximum could be delayed to 5-6 hours after administration of the drug (see graph A (iv)) as a result of the fluctuation in the hourly urine output.
It is noteworthy that from the detection point of view the concentration of the drug in urine is the more important quantity. This is because in carrying out chemical analysis, the laboratory analyst is limited to using a certain convenient quantity of urine (usually 10 ml, about one-half of a test-tube of urine) and only if the drug in urine exceeds the detection limit of the technique he employs will it be detectable. As the detection limit of the TLC technique we employed was 0.2 µg/ml, no drug can be detectable by this technique if present below 0.2 µg/ml in urine. From our results, the maximum concentration of morphine in urine after consumption of 1.5 mg of morphine can be as high as 4.3 µg/ml (see A (i)) and in general would exceed 1.0 µg/ml. Detection at the maximum is therefore easy to achieve. However, it can be seen from one of the graphs that urine collected half an hour after the intake of drug may contain no detectable amount of the drug (A (iii)) while in another case excretion of the drug is rapid and 4 to 5 hours after consumption the urine drug concentration has dropped to below 0.2 µg/ml and cannot normally be detected (A (ii)).
The O-methylation metabolic pathway for morphine has been postulated by Elison and Elliot [ 14] for rats and dogs after their observation of codeine in the animal's urine following administration of morphine. The biotransformation of morphine to codeine in man has been reported [ 15] but was not substantiated by later studies [ 16] . In our studies, we have also not been able to detect any codeine in the urines of human subjects administered with morphine. This is in agreement with Yeh's observations [ 17] . It would therefore appear that the codeine which is often detected in morphine addict's urine arises not from the biotransformation of the morphine consumed but from the codeine impurity which is always present in illicit morphine.
The total recovery of morphine in the 8-hour period after consumption of the morphine-containing medicine was found to be in the range of 20-70 per cent. This wide range is illustrative of the great variation between the metabolism of different individuals.
(B) Opium - oral administration of a single dose of medicinal opium mixture containing 2.5 mg morphine and smaller amounts of codeine and also kaolin.
As can be seen from graphs B (i) to B (iii), the urinary excretion of morphine from the consumption of the opium appears to be a more gradual process than that following the consumption of medicinal morphine. In the case of opium, the maxima in the hourly excretion pattern of morphine were less pronounced and occurred at 2-4 hours after consumption. The drug concentrations in urine were also lower and were generally below 1.0 µg/ml. As expected, a significant amount of codeine was also detected in each urine. The codeine/morphine ratio is in the range of 0.1-0.7.
The total morphine (free and conjugated) excreted during the 8-hour period following consumption of the mixture was found to be in the range of 6-17 per cent. It is seen that although a single dose of opium contains 67 per cent more morphine than a single dose of medicinal morphine, the urinary excretion of morphine, in terms of concentration and total excretion, following the consumption of the former is about 4 times less than in the case of the latter. This clearly indicates that the excipient matters of the drug play an important part in influencing the drug absorption and metabolism.
(C) Codeine - oral administration of a single dose of "Veganin Tablet".
"Veganin Tablet" is a commercial product for analgesic and antipyretic uses. It contains aspirin, phenacetin and codeine phosphate. Each tablet contains 8 mg of codeine phosphate equivalent to 5.9 mg of codeine base.
The urinary excretion of codeine and morphine are presented in the graphs C (i) to C (iii). It can be seen that the maximum hourly excretion of codeine occurred at 2 hours and then decreased only gradually. In two cases (C (i) and C (ii)), morphine was not detected indicating little or no O-demethylation in the metabolism of the individuals. The excretion of codeine in these cases was rather low, and only about 9 per cent of the dose consumed was recovered in 7 hours. In one other case (C (iii)) significant quantity of morphine was detected and the codeine and morphine recoveries were approximately 8 per cent and 1 per cent respectively in 5 hours. Like in the case of opium it would appear that a large proportion of the drug in the preparation (about 90 per cent) is not absorbed when consumed.
(D) Codeine - oral administration of Linotus Codeine (BPC), known also as Linctus Tussis Rubra (LTR).
LTR, a clear red syrup based on codeine has been widely used and prescribed for the treatment of coughs. It contains 15 mg codeine phosphate per 5 ml of syrup. A single dose of 5 ml therefore contains 11 mg codeine base.
A large number of cases of codeine excretion was investigated and a great variety of excretion pattern was recorded. Some of the typical ones are presented in graphs D (i) to D (vi). Most of the cases showed maximum hourly codeine excretion at 1-2 hours but a few exhibited maxima at around 4-5 hours after administration (D (ii), D (vi)). Some were observed to have even two maxima in 8 hours (D (i), D (iii)). Excretion of morphine was significant in some cases but insignificant or nil in others. In the 8 hours period following consumption, the concentration of codeine in urine varied between 0-15 µg/ml with morphine concentration at 0-4 µg/ml. The total excretion of codeine in urine also varied from 7-30 per cent of the consumed doses accompanied by 0-4 per cent of morphine.
It is interesting to note that the pattern of excretion varied not only between individuals but also on different days for the same individual (D (ii), D (iii), D (iv)) reflecting perhaps the different state of health and metabolism of the individual on different days. Therefore, an individual may react differently to an overdose of a drug on different days according to his metabolic detoxification capabilities at different times.
Apart from the detection of morphine and codeine in the urine, we have also observed in some cases another alkaloid which is of similar concentration as the morphine metabolite but with a slightly lower R f value than morphine. This may be norcodeine which is a known metabolite of codeine resulting from N-demethylation 1, [ 17] . It would appear that the detoxification metabolic pathway is largely determined by the individual genetic make-up and is a characteristic factor for an individual.
From a study of the large number of cases presented here, it can be seen that in the consumption of codeine, the ratio of codeine to morphine excreted in urine varied from 3 upwards for different individuals, indicating that theconversion of codeine to morphine does not exceed 25 per cent. Therefore if a person claims consumption of codeine, the ratio of codeine to morphine in his urine should exceed 3.
(E) Codeine - multiple dosage of LTR
The excretion pattern of multiple doses of LTR was investigated by following some clinical cases of patients given the medicine. The typical pattern is shown in graph E (i) in which the administration of LTR (18 mg codeine phosphate per dose)are indicated by arrows. It can be seen that a more complex pattern results and a built up of codeine and morphine can be observed. However, the ratio of codeine to morphine was still well above 3 and the excretion pattern was essentially a superimposition of laterally displaced excretion patterns of single doses of codeine.
(F) Heroin - Smoking of illicit heroin
The smoking of heroin in cigarettes is the most serious current form of drug abuse in Singapore. Pure heroin and heroin hydrochloride are white powder, but impure illicit heroin peddled on the clandestine drug market is usually brownish in colour and contains about 30-50 per cent of heroin hydrochloride with smaller quantities of monoacetylmorphine and acetylcodeine. The rest is essentially caffeine which is the common diluent used by traffickers to "cut" pure heroin into forms suitable for selling.
Illicit heroin available in Singapore is usually packed in plastic phials of about 4 cm long. A full phial contains about 0.8 gram of powder and may contain from 0.2-0.4 gram of pure heroin. The quarter-phial is the usual packing sold to addicts and contains a quarter of the above amount of heroin. A quarter phial may be used for "spiking" 6 to 7 cigarettes, each receiving about 30 mg of illicit heroin (equivalent to around 10 mg of pure heroin). "Spiking" is normally done by placing heroin powder in the groove of a wooden pick and inserting it into the ends of the cigarette. Hardened addicts may use a larger quantity of heroin in their cigarettes.
A very recent form of packing illicit heroin is to heat-seal about 0.1 gram of the powder in a short length (3 cm) of plastic drinking straw. This form of illicit heroin is rapidly gaining popularity among the addicts because of the ease of concealment.
We have made a study of the urinary excretion pattern of a known heroin addict. After having smoked about 60 mg of heroin in 5 cigarettes, the urine excretion of total morphine and codeine are as shown in graphs F (i). It can be seen that the excretion of the alkaloids was much more than those involved in the consumption of single dose of medicinal preparations given in the previous sections. At 5 1/2 hours after consumption the excretion of morphine and codeine was still significant. The codeine to morphine ratio however, remained lower than 0.5 throughout. It is therefore possible to distinguish a case of heroin consumption from one of codeine consumption simply from the ratio of the amounts of codeine to morphine excreted in the urine.
(G) Mass screening of urine
During 1975, our laboratory received some 3,000 samples of urine from various sources for the analysis of drugs of abuse. About 40 per cent of these samples were positively identified to contain morphine and codeine, representing over 1,000 cases of abuse of morphine, heroin and opium. A closer study revealed that about 90 per cent of these involved persons between the age of 15 and 25 who consumed morphine or heroin. The opium addicts were mainly elderly people.
Analysis of the urine samples indicated certain consistencies in the ratio of the amount of codeine to morphine. Abusers and addicts consuming morphine or heroin have ratios falling within the range 0-0.5 while most opium addicts were within the 0-0.8 range. In contrast, our findings for those consuming codeine and morphine medicinal preparations indicated a ratio of > 3.0 and 0 respectively.
Figure G is a graphical presentation of the concentrations of morphine and codeine in the urines of some morphine/heroin abusers and those of volunteers given codeine medicinal preparations. Each point on the graph represents the results of one urine sample. It is clear from this figure that there can be no difficulty in differentiating an abuser who consumed morphine/heroin/opium from a person who takes codeine for medical reasons.
(H) Length of period of detectability following consumption of codeine
Law enforcement authorities are most interested to know how long after the consumption of a drug can it be still detectable from the examination of urine. The answer to this question depends mainly on three factors: (i) the dosage taken, (ii) the sensitivity of the method of detection and (iii) the excretion pattern of each individual.
As can be seen in the graphs presented in the previous sections (A) to (F), the drug concentration in excreted urine generally shows a relatively rapid increase from a low to a high level soon after consumption and it then gradually declines towards the zero baseline. During this period, the drug concentration can range from below 1 µg/ml (i.e. < 1 ppm) to above 10 µg/ml (i.e. > 10 ppm). An analytical method with a detection limit of say, 5 µg/ml, will therefore only be able to detect the drugs in urine during the relatively short period when its concentration in urine exceeds 5 µg/ml but a method with a detection limit of 0.2 µg/ml such as the one we used, would be able to follow the drug excretion for a much longer period of time i.e. until the drug concentration falls below 0.2 µg/ml.
Solomon (18), in a recent publication, claimed that in some patients given codeine treatment, the excretion of codeine and morphine metabolites in their urines can be detected up to 72 and 96 hours after the consumption of codeine and that at 96 hours, only the morphine metabolite was detectable.
To check the validity of these results, we studied the long term urine excretion of a human subject after consumption of 30 mg of codeine phosphate in LTR using both TLC and GLC techniques which have detection limits of 0.08 µg/ml and 0.03 µg/ml respectively when 25 ml of urine samples were analysed.
Our results indicated that 24 hours after the ingestion of the specified quantity of codeine, traces of codeine could still be detected in the urine by GLC but not by TLC technique, implying that the urine drug concentration had dropped to below 0.08 µg/ml in 24 hours. 48 hours and 72 hours later, no codeine or morphine could be detected by either technique.
These results therefore failed to bear out Solomon's observations and it would seem that, at least with the techniques we employed (TLC and GLC), there is no possibility of detecting only morphine metabolite 24 hours, 48 hours or 72 hours after the consumption of therapeutic doses of codeine.
Although we did not carry out similar studies with morphine consumption, it may be expected on consideration of the similar results obtained in sections (A) and (C) for morphine and codeine consumptions that 24 hours after the consumption of morphine, traces of morphine may still be detectable if the detection limit is sufficiently good. However, as routine screening precluded the handling of large samples, the detection of morphine 24 hours after consumption is not easy to achieve. In fact, considering the rapidity of excretion of some of the cases we studied, it is possible that at 8 hours after consumption of a small quantity of the drug, no trace of it can be detected in urine.
This study showed that a number of factors affect the urinary excretion pattern of morphine alkaloids of an individual, these are:
Detoxification efficiency of metabolism - excretion of morphine alkaloids in some individual can be more rapid that others depending on the individual's metabolism.
O-demethylation and N-demethylation metabolic pathways - it appears that possibly due to enzymatic differences, some individuals have more prevalent demethylation pathways than others and thus give rise to significant codeine/morphine/norcodeine conversion before excretion.
Daily fluctuation of metabolic rate - the excretion efficiency for an individual can vary from day to day due to variation in his metabolic mechanism arising from environmental or dietary factors.
Mode of intake of the drug - the route of intake of the drug and the matrix holding the drug has been shown to affect its excretion.
Hence, although the general pattern of excretion is a relatively rapid increase followed by a gradual decrease, individual patterns can show wide variation and it is not impossible to observe two or more maxima in the course of an excretion and that drug excretion at say 6 hours after consumption may be higher than that at 4 hours after consumption. The wide range of possibilities can be seen from the numerous different excretion patterns from a random selection of volunteers given in this report. There is no evidence to suggest that age, sex and ethnic differences have any bearing on the urinary excretion.
A summary of the relation between the types of drug consumed and the excretion of morphine and codeine is given in the following table.
|
Relative intensity of excreted drugs |
||||
|---|---|---|---|---|
|
Type of drug consumed |
Codeine |
Morphine |
Ratio codeine/morphine |
|
|
(
a)
|
Diamorphine
|
nil
|
strong
|
0 |
|
(
b)
|
Illicit heroin
Illicit morphine
|
weak
|
strong
|
0 - 0.5
|
|
(
c)
|
Opium
|
medium
|
strong
|
0 - 0.8
|
|
(
d)
|
Morphine
|
nil
|
strong
|
0 |
|
(
e)
|
Codeine
|
strong
|
weak
|
>3.0
|
It is interesting to note that in the case of the consumption of opium (whether by oral ingestion or by smoking), the ratio of codeine/morphine in urine tends to be higher than would be expected from the proportion of the two alkaloids in opium itself (~ 0.05). This seems to indicate a preferential excretion of codeine over morphine in urine for such cases.
The results given in the above table are valid for consumption of the drugs indicated and it is obvious that if a mixture of the drugs were consumed, the pattern of urine drug excretion would be much more complicated. However, the following observations can be made:
If the ratio is between 1.0 and 3.0, it is safe to say that codeine is involved but it is not the only drug consumed and that one or more of the other drugs listed in the above table must have been also taken. But if the ratio falls below 1.0, any of the listed drugs ( a), ( b), ( c) or ( d) can be implicated as well as an admixture of any of these with ( e).
Further study of the various factors influencing drug excretion pattern is continuing for a better understanding of the problem.
The authors wish to thank Mr. M.C. Dutt, Acting Director of Scientific Services, Singapore for his support of this project and also wish to thank M. John Hanam, Director of Central Narcotics Bureau, Singapore, for his assistance in securing some samples.
A special note of thanks must go to the many volunteers who participated in this programme without whose assistance this work would not have been possible.
We would also like to thank M. Swee Kim Seng for technical assistance and Mr. Tham Keng Leong for the drawings.
E.L. Way and T.K. Adler; Bull. W.H. 25 , 234, 1961; 26, 51, 1962 and 27, 374, 1962.
002E.L. Way, J.M. Young and J.W. Kemp; Bulletin on Narcotics , XVII: 1, 25 , 1965.
003E. G. C. Clark; "Isolation and Identification of Drugs", The Pharmaceutical Press, London 1969, p. 143.
004G. R. Nakamura and T. Ukita; J. Forens. Sci. 1 , 465, 1962.
005G. R. Nakamura; J. Assoc. Off. Anal. Chem. 49, 1086, 1966.
006Lim Han Yong and Chow Shui Tse, "Heroin abuse and a gas chromatographic method for determining illicit heroin in Singapore", submitted for publication.
007S. J. Mule; J. Chromat . 55:255, 1971.
008K.K. Kaistha and J. H. Jaffe; J. Chromat. 60, 83, 1971.
009J. T. Payte, J. E. Wallace and K. Blum; Curr. Ther. Res. 13, 412, 1971.
010F. Fish and T. S. Hayes; J. Forens. Sci. 19 , 676, 1974.
011S. J. Mule, M. L. Basto, D. Jukofsky and E. Saffer; J. Chromat. 63 , 289, 1971.
012Brit. Natl. Formulary 1974-76, p. 232.
013Brit. Pharmacopoeia , 1973, p. 311.
014C. Elison and H. W. Elliot; J. Pharmac. Exp. Ther. 144 , 265.
015U. Beerner and S. Abbot; Experienta, 29 , 180, 1973.
016S. Y. Yeh; Experienta, 30 , 265.
017S. Y. Yeh; J. Pharm. Pharmac. 27 , 214, 1975.
018M.D. Solomon; Clin. Toxicol. 7 , 255, 1974.
