JUDGMENT Gyanendra Kumar, J. - In the present criminal revision against conviction u/s 60(a) of the UP Excise Act, B.D. Gupta, referred the following question for consideration by a larger Bench: Can an accused person be properly convicted for the offence of having bean found in possession of Ganja on the testimony of an Excise officer, on the basis of sight and smell, that the recovered article was Ganja, or is it necessary for the prosecution to lead evidence based on chemical analysis to establish that the article alleged to have been recovered was Ganja? 2. The definition of Ganja is given in Rule 15 of the UP Excise Manual, Part II, Ch. I, which runs thus: Ganja means the dried flowering tops of the female hemp plant which have become coated with resin in consequence of being unimpregnated and therefore, unable to set seeds freely. 3. The facts of the case are that Smt. Manjauka and Ramjas Applicants are mother and son, living in village Dharampur, Distt. Jaunpur. On 29-12-1967 1957, at about 6 p.m. their house was searched by P.W. R.N. Tripathi, Excise Inspector Circle II, Jaunpur in the presence of Manjauka accused and certain search witnesses, with the result that 6-1/4 seers of alleged Nepali ganja was recovered from the house of these accused from two places in the same room. The Excise Inspector considered the recovered commodity to be ganja by sight and smell. In his test report (Ex-ka-2) the Excise Inspector mentioned that he considered the recovered commodity as contraband Nepali ganja as its colour was brownish green and it had characteristic texture of ganja. He added that "this is the flowering top of female hemp plant." He further noted that the commodity had the special type of ganja smell. 4. The necessity to make the instant reference arose because of conflict of views expressed in two Single Judge decisions of this Court, viz. in Chungur v. The State 1952 AWR 112 and Mohan Lal v. State 1968 AWR 456 as well as certain observations made by the Supreme Court in State of Andhra Pradesh Vs. Madiga Boosena and Others, AIR 1967 SC 1550 . 5. in Chungur v. The State 1952 AWR 112 and Mohan Lal v. State 1968 AWR 456 as well as certain observations made by the Supreme Court in State of Andhra Pradesh Vs. Madiga Boosena and Others, AIR 1967 SC 1550 . 5. In Chungur's case (supra) Kidwai, J. made reference to Rule 278 framed under the UP Excise Act which showed that it was not essential that every intoxicant produced in s case should be analysed by the Chemical Examiner and that it was only when the Excise Inspector or other officer prosecuting considered it necessary, that he is to move the Magistrate to send the intoxicant to the Chemical Examiner. In Chungur's case the Excise Inspector had found the article in question to be contraband Nepali ganja which was, in his opinion, different from the ganja produced in Government godowns. In that case Kidwai, J. had accepted the opinion of the Excise Inspector as good evidence to maintain the conviction. 6. There is yet another Division Bench case of this Court to which Kidwai, J. was a party, viz. The State Vs. Kaptan Singh, AIR 1952 All 118 . In this decision Chungur's case was not referred to but it was held that-- Opium in the form of coagulated juice is now well know a in this country, being widely used for medicinal and other purposes, that any one can identify it and it is unnecessary to call in an expert to establish its identity. The testimony of the Excise Inspector cannot be looked upon in this case as merely an opinion of an expert as lie did not claim to be one nor did he give any testimony as an expert. Yet the testimony of the excise Inspector was found entitled to weight as good evidence even if no reasons were given by him for his opinion. 7. As against the above view, Roy, J. in Ram Piyari v. State Cr. Revs. Nos. 1563 and 1564 of 1954 D/- 7-5-56 held that the report of the Excise Inspector should not be relied upon without chemical analysis. Likewise it was held by A.P. Srivastava, J., in Cr. Rev. No. 350 of 1956, which related to liquor, that the test of Excise Inspector by smell, touch and burning is no test, without the evidence of an expert and should not be accepted. 8. Likewise it was held by A.P. Srivastava, J., in Cr. Rev. No. 350 of 1956, which related to liquor, that the test of Excise Inspector by smell, touch and burning is no test, without the evidence of an expert and should not be accepted. 8. In Boosenna's case (supra), which related to recovery of illicit liquor, it was observed by the Supreme Court in paragraph 10: Except for a general statement contained in the evidence of the witnesses, particularly P.Ws. 1 and 4 that there was a strong small of alcohol, emanating from the tins, which were pierced open, there is no other satisfactory evidence to establish that the article is one coming within the definition of the expression 'liquor'. Merely trusting to the smelling' sense of the Prohibition officers and basing a conviction, on an opinion expressed by those officers, under the circumstances, cannot justify the conviction of the Respondents. In our opinion, better proof by a technical person, who has considered the matter from a scientific point of view, is not only desirable, but even necessary, to establish that the article seized is one coming within the definition of 'liquor'. 9. Following the principle laid down in the above decision, S.D. Singh, J. held in Mohan Lal's case (supra), which was a case relating to recovery of ganja, that conviction u/s 60(a) of the U.P. Excise Act was not sustainable on the evidence of the Excise Inspector, who made no chemical analysis of the article yet considered it to be ganja merely by its sight and smell. 10. It is true that in order to bring an article within the definition of liquor it is necessary to prove that it contained a particular percentage of alcohol, which can only be determined by chemical analysis. Yet from the above quoted observations of their Lordships of the Supreme Court in Boosenna's case, which though related to liquor, it could safely be inferred as a general rule that in cases where scientific investigation was possible and desirable, the mere opinion of an Excise Inspector that an article was contraband ganja or some other contraband article, should not be deemed sufficient to sustain conviction, because the prosecution has to establish; case beyond all reasonable doubt. Such doubt can obviously be eliminated only by scientific investigation. 11. Such doubt can obviously be eliminated only by scientific investigation. 11. In the present case in order to determine whether the Excise Inspector P.W. 1 R.N. Tripathi, could be considered to be an expert, as claimed by the prosecution, or whether his evidence, even as an ordinary witness, could be treated sufficient for maintaining the conviction of the Applicants we, on 23-12-1969, examined Sri K.D. Misra Excise Inspector, Excise Intelligence Bureau, UP, Allahabad as HCW 1. Our anxiety was to find out whether the Excise Inspectors in Uttar Pradesh received any special or technical training in identifying ganja, which might give them the status of an expert. This witness deposed that Excise Inspectors were selected by the UP Public Service Commission and were straightaway posted to various stations after three months training at the Harcourt Butler Technological Institute, Kanpur. He said that after receiving the above training at Kanpur an Excise Inspector was first attached to a senior Excise Inspector for field training, lasting for a month. Thereafter he was posted as an independent Excise Inspector. Sri Misra clarified that the field training consisted of matters like inspection of excise shops, detection of excise cases and arrest of offenders under the Excise Act. He added that at the Harcourt Butler Technological Institute, the selected candidates were given training in alcohol technology and identification of various intoxicating drugs such as bhang, ganja, charas and opium. At the end of three months training a test was held by the Principal of that College. The Examination consisted of written, oral and practical tests. The written test paper related to alcohol technology. Similar questions were asked in oral test. In practical test, samples were given to test the strength of alcohol. He conceded that at the Harcourt Butler Technological Institute, the emphasis was on alcohol test. He also admitted that during his training at the Harcourt Butler Technological Institute, Kanpur, he was asked to test ganja only on two or three occasions, "though alcohol was tested by him many a time. He went on to say that various varieties of ganja are known after the places where the yielding ganja plants are generally grown, for example, Rajshahi, Khandwas and Nepali ganja. According to this witness there was no difference between these three varieties of ganja, except that according to the fertility and nature of the soil the size of the ganja buds was different. According to this witness there was no difference between these three varieties of ganja, except that according to the fertility and nature of the soil the size of the ganja buds was different. The larger the bud the stronger the smell. The witness said that in fresh buds the smell was more pronounced. But after they are dried, the smell diminishes by and by, till after five years it altogether disappears. However, it still retains the intoxicating property, though in lesser degree. The witness conceded that after a long sized bud is dried and crushed to pieces it is difficult to say whether those pieces are of a long bud or a small bud. He added that in order to find out whether a particular article is ganja they apply optical test and also smell it. But if it is old then they burn it and smell the incense. According to this witness, ganja is identified by mere sight and smell and in case of doubt by its incense. He further conceded that the buds of Caledium and other similar plants when dried and pressed will have an appearance of ganja and that there were several other varieties of plants (especially their leaves and buds) such as Arhar, which will have similar look, when dried and pressed. He went on to say that commonly ganja is not only the bud part of a female hemp plant but it also includes the soft and tender leaves thereof. He said that ganja contains some chemical properties which produce intoxication when taken orally or when smoked. 12. From the above statement of Sri Misra it is quite clear that hardly any specialised training is given to Excise Inspectors at the Harcourt Bulter Technological Institute, Kanpur for identifying ganja and that the optical test used by them may be illusory or misleading, particularly when the dried buds of certain other plants may look very similar in appearance. As regards test by smell, it is, in the first place, a somewhat uncertain type of test, varying from man to man, according to the development of his sense of smell. Secondly, if some other similar looking dried and crushed buds and tender leaves are placed with or under real ganja for some time, they would catch the smell of ganja and would themselves start smelling as such. Secondly, if some other similar looking dried and crushed buds and tender leaves are placed with or under real ganja for some time, they would catch the smell of ganja and would themselves start smelling as such. In this age of large scale production of synthetic goods and fake product such a possibility cannot be ruled out. That being so, the detecting of ganja by mere look and smell is a very uncertain and unsatisfactory test. 13. We, therefore, considered it proper to examine two experts as Court witnesses. HCW 2 Dr. (Mrs.) Radha Pant is an M.Sc. and PHD of Bombay University and is now the Head of the Department of Home Science and Biochemistry in the University of Allahabad. She has 25 years teaching experience in Organic and Biochemistry. She deposed that it was possible to chemically analyse the dried flowering top of the female hemp plant coated with resin in consequence of being unimpregnated and therefore unable to set seeds freely. The flowering top of the female hemp plant itself exudes resin which gets coated thereon, preventing impregnation. But if the flower of a female hemp plant is fertilized or impregnated at the outset, the resin coating is not formed thereon. According to Dr. Radha Pant, the resin of female hemp plant has two alkaloids, cannabiol and tetrahydrexy cannabiol. These alkaloids are narcotic. It is solely 4 due to the presence of these narcotic alkaloids that ganja produces intoxication. On chemical examination it can be determined whether a particular flowering top is coated with resin which will respond to the tests for the two alkaloids mentioned above. It is possible to detect the presence of these materials by determining its optical activity by means of an instrument known as polarimeter, after resin is dissolved in a suitable solvent. However, Dr. Radha Pant conceded that flowering tops of certain other plants also have resin which gets deposited thereon She added that the look of the flowering top pf dried female hemp plant after a lapse of three years would definitely change and its smell would also become much fainter. The Government Advocate showed a sample to the witness, who said that it looked and smelt like ganja, but she could not definitely say whether it was really ganjal. 14. Coming to the testimony of HCW 3 Dr. The Government Advocate showed a sample to the witness, who said that it looked and smelt like ganja, but she could not definitely say whether it was really ganjal. 14. Coming to the testimony of HCW 3 Dr. D.D. Pant, Professor and Head of Botany Department in the University of Allahabad, we find that he is a specialist in the form and structure of plants. He describes ganja as pressed female flowers of Cannabis sativa, popularly known as hemp or bhang plants. He deposes that there are hairs fiver the female flowers, which are glandular and secrete resin. The resin so secreted is deposited on the surface of the flower, which has yellowish shining appearance. It is these dried flower buds of the female hemp plants with resin deposits which are called ganja. Dr. Pant goes on to say that the structure of female flowers, even though pressed, crushed and crumbled, can be determined by their epidermal structure under a microscope but not by the naked eye, because after the flowers are crushed and crumbled, the hairs cannot be seen with the naked eye. Dr. Pant is not in a position to say whether dried and crushed tops of Humulus would look like those of Cannabis sativa. But he admits that there are other plants, not of the same family which may look similar to those of cannabis sativa, but they can always be distinguished under the microscope. Describing the flowering top of a female plant of Cannabis sativa, Dr. Pant deposes that it has flowers in closely crowded clusters and each flower bears two thread like stigmas and there is an ovary at the base, with a single pendulous ovule and a ring like perianth. He says that if such female flowering tops are pressed, all the above distinctive qualities of the flowers will not be obvious to the naked eye but the tops can be soaked in water and the flowers carefully dissected put under a low magnifying microscope and thereafter each part will be obvious. Dr. Pant admit J that even the tender leaves of Cannabis sativa when dried, crushed and crumbled may look similar to ganja, but they can always be differentiated under the microscope. In the end Dr. Dr. Pant admit J that even the tender leaves of Cannabis sativa when dried, crushed and crumbled may look similar to ganja, but they can always be differentiated under the microscope. In the end Dr. Pant stated that for a trained eye it may be possible to distinguish bhang from ganja, but it would be necessary to confirm the same by microscopic examination. He conceded that he will not be able to distinguish the incense of ganja from that of other dry flowering tops. 15. From the above authoritative observations of Dr. D.D. Pant and Dr. (Mrs.) Radha Pant it is abundantly clear that in order to find out definitely whether a particular article is really ganja it must be chemically analysed. According to the two experts, there are flowering tops of certain other plants which, may very much, resemble the tops of female hemp plant and there may also be resin deposit thereon. It may be that an examination of the flowering top after soaking it' in water may reveal whether it relates to a female hemp plant or not. But the presence of the intoxicant resin deposit containing the two alkaloids which alone causes intoxication cannot be detected even under microscopic test. Hence chemical examination, is the only sure scientific test to resolve the doubt whether a particular article is really ganja or anything else; just looking or smelling similarly and that a test by mere sight and smell is both vague and uncertain. 16. That being so, the opinion of the Excise Inspector PW 1 R.N. Tripathi that the article recovered from the possession of the two accused was Nepali ganja is a mere opinion of a person, who can by no means be called an expert. Needless to repeat that in a criminal case, the prosecution has to establish its case beyond all reasonable doubt. 17. In the present case we wanted to send for the samples of the alleged contraband ganja for chemical examination, in order to come to a definite conclusion whether it was (really ganja, for the possession of which, the conviction of the accused persons could be maintained. However, on 17-8-1970 the Govt. Advocate made a statement that the entire lot of the alleged contraband ganja had been destroyed, after the hearing of the appeal by the learned Sessions Judge. However, on 17-8-1970 the Govt. Advocate made a statement that the entire lot of the alleged contraband ganja had been destroyed, after the hearing of the appeal by the learned Sessions Judge. The result is that the alleged incriminating article is not now available for scientific analysis. 18. Under the circumstances, our reply to the question referred to us is that an accused person cannot be properly convicted for the offence of having been found in possession of alleged ganja on the testimony of Excise Inspector on the basis of sight and smell test. It is necessary for the prosecution to lead evidence based on chemical analysis in order to establish that the article alleged to have been recovered from the accused was really ganja. 19. Let this case be now listed before the learned Single Judge for further orders.