J.R. MUDHOLKAR, J. (1) THIS is an appeal by special leave from the Order of the government of India, Ministry of Finance (Department of Revenue), rejecting the appellants application preferred under S. 36 of the central Excises and Salt Act, 1944 (1 of 1944). (2) THE appellant is a public limited company and has been carrying on the business inter alia of manufacturing sugar from sugarcane since the year 1933 at its factory-situated at Hargaon in the State of Uttar Pradesh. In May, 1957 an Assistant Chemical Examiner inspected the factory and made a report to the government to the effect that the appellant had not, kept correct accounts of the amount of mixed juice which went into the production of sugar, as required by Rule 83 of the rules framed under the Act taken along with the Form R.G 6 (c) prescribed by the aforesaid rule. The report also stated that for the period between the commencement of the crushing season and 7/05/1957, 11,606 maunds of sugar have been short accounted for. On the basis of this report a show cause notice was issued to the appellant and after hearing its representative the Collector of central Excise, Allahabad, passed an order on 19/02/1958, requiring the appellant to pay duty on 11,606 maunds of sugar at the rate of Rs. 5-10-0 per cwt. In addition, he imposed a penalty of Rs. 2,000.00 on the appellant for not maintaining the accounts correctly and in the manner prescribed by rule 83 read with rule 226 of the central Excise Rules, 1944. Against this order, the appellant preferred an appeal before the central Board of Revenue under S. 35 of the Act and asked for a personal hearing. After granting it and hearing the appellants representation the Board dismiss- ed the appeal. It then preferred an application for Revision before the central Government under S. 36. This application was dismissed on 17/11/1959. After that the appellant sought special leave to appeal from this Court under Art. 136 of the Constitution which having been granted, the appeal has come up before us. (3) WHAT is urged on behalf of the appellant in the appeal is that the Order of the Collector holding that 11,606 maunds of sugar were short accounted for and that the accounts had not been properly kept was based on assumptions for which there was no basis. (3) WHAT is urged on behalf of the appellant in the appeal is that the Order of the Collector holding that 11,606 maunds of sugar were short accounted for and that the accounts had not been properly kept was based on assumptions for which there was no basis. In order to appreciate the point it is necessary to bear in mind, the process of manufacturing sugar employed in the factory. This has been set out in the petition for special leave as well as the statement of the case and it is not contended on behalf of the respon- dent that there is any error in what the appellant has stated in this behalf. The process, in so far as it is material for the purposes of this case is as follows:- SUGARCANE received at the factory is weighed mechanically cut into pieces and fed into the mills called "crusher and mills" where it is crushed. DURING this process, a certain quantity of water as introduced. The mixture of cane guice and water obtained from the crushing operation is called mixed juice. THE mixed juice is led from the crusher and mills to two tanks placed over weighbridges. THE two tanks are attached to weighbridges with lever arrangement which have markings on their beams. The persons incharge of the operation adjust each beam at a certain mark. A movable indicator is also fitted on to the tanks which begins to move up as more and more juice finds its way in the tank. When that indicator comes up to the fixed mark on the beam, the persons in-charge of the operation immediately stop the further in flow of mixed juice. The gross Weight of the tank with its content of mixed juice when the fixed mark is reached is 7.5 tons. The tare weight of the tank being known, the actual weight of mixed juice is ascertained by deducting the tare weight from 7.5 tons. THE mixed juice is then introduced into the manufacturing process where it is clarified, the water is evaporated and, eventually, sugar crystals are formed. The by-products obtained during this process are press cakes and molasses. (4) THE entire operation is thus carried out in aclosed system and the quantity of the sugar as well as of the by-products is known. The percentage of sugar in sugarcane, mixed juice, filter cake and molasses is determined. The by-products obtained during this process are press cakes and molasses. (4) THE entire operation is thus carried out in aclosed system and the quantity of the sugar as well as of the by-products is known. The percentage of sugar in sugarcane, mixed juice, filter cake and molasses is determined. In spite of this, a certain amount of sugar remains undetermined. That is a well accepted fact and is recognised in Form R.T. 7(c) under the heading "Central Excise Series No. 74." This Form is to be used by central Sugar factories for periodical/monthly manufacturing reports under Rule 83, the relevant portion of which runs as follows : * * * *. (5) IT is common ground that under the rules framed under the Act, a Resident Inspector is posted in every central Sugar factory. The sugar is filled in gunny bags after being duly weighed which are kept in the godown. Appropriate entries are made in the godown register. It is also not disputed that nothing is allowed to go out of the factory without the knowledge of the Resident Inspector and without obtaining a gate pass. All the bags going out of the factory are weighed in his presence. (6) IT is, therefore, contended on behalf of the appellant that it cannot be held guilty of short accounting of sugar unless it is established that any part of the sugar manufactured in the factory has been removed from the factory or there were some loopholes in the working of the factory which provide oppor- tunities for clandestine removal of sugar. The only basis for the finding arrived at by the Collector, it is pointed out, was that provided by the calculations made by the Assistant Chemical Examiner. These calculations were attacked on behalf of the appellant upon the ground that they were based not on factual data but on certain assumptions. It is, therefore, necessary now to examine the report of the Assistant Chemical Examiner. The relevent portion of the report runs thus: "ON scrutiny of this register, I was astonished to note that the gross weight of each tank was being taken as 7.5 tons each time. Since in any weighing arrangement, it is difficult to have constant gross weight, it apears to me that there is a short accounting of mixed juice. The relevent portion of the report runs thus: "ON scrutiny of this register, I was astonished to note that the gross weight of each tank was being taken as 7.5 tons each time. Since in any weighing arrangement, it is difficult to have constant gross weight, it apears to me that there is a short accounting of mixed juice. It is evident from the fact that it is rather impossible to have the gross weight as 7.5 tons each time when the pipe through which the juice is received is about 8 in diameter. This short accounting of mixed juice was confirmed by me by the actual weightment in my presence. During a period of about 2 hours of milling test carried out in my presence it was noticed that there was a difference of about 1 % of mixed juice cane between the actual gross mixed juice obtained by me and calculated by the factorys usual method, taking the gross weight as 7.5 tons per tank as per details given below :- Weight of cane 4502.04 mds. Net mixed Crushed %cane Weight of next mixed 4087 mds. 90.78 juice by taking the actual gross weight. Weight of net mixed 4031 mds. 89.54 juice by the factorys usual method of taking the gross weight as 7.5 tons per tank Difference 56 mds. 1.24% crushed juice %cane Weight of next mixed 4087 mds. 90.78 juice by taking the actual gross weight. Weight of net mixed 4031 mds. 89.54 juice by the factorys usual method of taking the gross weight as 7.5 tons per tank ------ 56 mds. Difference 1.24% � THUS from the above data, it appears that there is a short accounting of mixed juice per cent cane at least to the extent of 1%. It is clear from the fact that by the time the juice is filled in the tank up to 7.5 tons and the man on duty closes the valve there is a small time gap. However, small this gap may be, it is definite that some extra mixed juice must drop into the tank. It amount will depend upon the flow of mixed juice, the man on duty and his activeness to close the valve, to weigh the tank and to record in the register. Thus, this extra mixed juice is supposed to remain as unaccounted for during the whole season. It amount will depend upon the flow of mixed juice, the man on duty and his activeness to close the valve, to weigh the tank and to record in the register. Thus, this extra mixed juice is supposed to remain as unaccounted for during the whole season. It is, therefore, certain that the factory has been showing low mixed juice per cent cane through- out this season. CALCULATING on this basis the total mixed juice short accounted from beginning of the season upto 7th May, comes to- WT.OF cane x 1/100 = 10029633 x 1/100 = mds. Taking the net mixed juice per cent came to date as 87.36, this works out equal to 100297 x 100 = 114800 maunds of cane. 87.36 Taking 10.11% as to date recovery % cane this loss is equivalent to. 114800 x 10.11/100 = 11606 mds. of sugar. THUS it appears that there is short accounting of sugar at least to the extent of 11,000.00 maunds. Hence the factory is liable to be charged for submitting wrong data of mixed juice in central Excise record. For this, the factory may beasked to explain and justify on scientific basis the short accounting of 11,000.00 maunds sugar. If no satisfactory explanation is forthcoming then they are liable to be charged for reporting less quantity of recoverable sugar." (7) NOW, these calculations certainly involve some assumptions. The first is that the differenceof56 maunds noticed by the Assistant Chemical Examiner during the two and quarter hours test conducted by him on 7/05/1957 was uniform for every 21 hours working throughout the working hours of the crushing season which began on 4/11/1956 upto May 7, 1957 when the test was conducted. (8) THE second assumption is that the persons in-charge of the opera- tion of letting in mixed juice fill the tanks uniformly upto a level beyond the fixed mark and never below that level or at that level. (9) THE third assumption is that the average of 10.11% of sugar recovery from the mixed juice is a safe basis for calculating the quantity of sugar not accounted for. (10) THE fourth is that the milling performance of the factory has been uniformly efficient throughout the entire period from 4/11/1956 to 7/05/1957. (11) IF any one of these assumptions breaks down, then the ultimate conclusion will have to be rejected as incorrect. (10) THE fourth is that the milling performance of the factory has been uniformly efficient throughout the entire period from 4/11/1956 to 7/05/1957. (11) IF any one of these assumptions breaks down, then the ultimate conclusion will have to be rejected as incorrect. It has to be borne in mind that human element is involved at certain stages of the operation such as time of commencement of the days working, the rapidity or slowness in feeding cut sugarcane into the crusher and mills, accurately adding the same quantity of water in the crusher and mills, stopping the inflow of mixed juice into the tanks at a uniformly higher level than that indicated by the mixded mark, allowing a uniform time lag between the emptying of the tanks and starting inflow into them again and soon. It is also unsafe to rely on the average of 10.11% of sugar recovery for the entire period because it involves the assumption that even during the periods when the gross weight of the tanks was 7.5 tons or less sugar content of the sugarcane crushed in the factory was such as could not have been yielded anything less than 10.11%. For, it is well-known that the sugar content of sugarcane even grown in the same area is not constant but is variable according to the time of the year. Again it has to be borne in mind that the recovery of sugar must necessarily depend upon the milling perfor- mance. It cannot be assumed that even in an ordinarily well run factory the performance would be uniformly good or uniformly the same. (12) WE would also mention one more factor which is mentioned in the letter addressed by the mills to the Resident Inspectoron 27/07/1957, replying to the Assistant Chemical Examiners comment on the working of the factory. In that letter, they have taken exception to the calculations of percentage of mixed juice to the quantity of sugarcane made by the Assistant Chemical Examiner on the ground that it is merely inferential and based upon a factor which was first evolved in 1921 by one Noel Deer by taking average of number of analysis conducted by him. In that letter, they have taken exception to the calculations of percentage of mixed juice to the quantity of sugarcane made by the Assistant Chemical Examiner on the ground that it is merely inferential and based upon a factor which was first evolved in 1921 by one Noel Deer by taking average of number of analysis conducted by him. They point out that as the juice in cane is not uniformly distributed in the pith and the rind, the resultant juice by different milling procedures cannot be the same, or in other words it cannot be presumed to hold a constant and definite ratio to the brix of the first expressed juice. According to Noel Deer, the ratio between the brix of absolute juice and primary juice is 0.975. They point out that in a given case the proportion between the two would depend upon two factors, the fibre percentage of the cane and milling performance. Then they proceed to say : "FIBRE per cent cane in Java used to be about 12.5 while in India it is mostly about 16. Indian cane thus has thicker find and less of pitch. Pitch contains richer juice and the rind contains only protoplasm water. Hence, the absolute juice factor of 0.975 is not absolute for its application under Indian conditions particularly in our case and that is why no reference has been made and no cognisance taken by Sugar Technologists. Association of India regarding its application for inferentially calculating mixed juice per cent in the revised draft for publication of "Methods of Chemical Control in Sugar Factories". (13) AGAIN the value of fibre percentage in cane is very difficult to determine directly and accurately. For calculations, only indirect values are taken which may at the most be only approximate and, therefore, no absolute conclusion should be drawn. They then say: "AFORMULA which does not apply to Indian conditions because of wide diversity in case varieties, milling performance etc. cannot therefore be used to establish under weighment of juice only because our mixed juice does not correspond to it. They then say: "AFORMULA which does not apply to Indian conditions because of wide diversity in case varieties, milling performance etc. cannot therefore be used to establish under weighment of juice only because our mixed juice does not correspond to it. MOREOVER inferential method of calculation merely gives rough idea of the value of data collected by other means and cannot beused to overrule actually found results." Apparently, their contention is that the particular ratio upon which a a part of the calculations of the Assistant Chemical Examiner was based is founded on analysis of Java cane and is, therefore, no guide for determina- tion of the ratio in respect of Indian cane. However, the point raised is a highly technical one and counsel on neither side was able to throw any light upon it. We would, therefore, not base any conclusion upon it but only observe that this argument was not considered by the Collector or by the central Board of Revenue or the central government. In the circumstances, therefore, we must hold that the finding that 11,606 maunds of sugar were not accounted for by the appellant has been arrived at without any tangible evidence and is based only on inferences involving unwarranted assumptions. The finding is thus vitiated by an error of law. (14) THE other finding that the registers were not properly maintained as required by Rule 83 is also an inferential finding based upon the calcula- tions made by the Assistant Chemical Examiner. As we have already held those calculations being based upon unwarranted assumptions cannot form legal basis for a finding that more juice than what was recorded in the register had gone into the production of sugar. (15) IT was, however, contended on behalf of the respondent that at the personal hearing given by the central Board of Revenue, Shri C.M. Dalmia, Assistant secretary of the Appellant and Shri. J.M. Shah, Superintendent Technologist employed by the Appellant had conceded that there had been "some erroneous accounting" on their part in showing the weight of mixed juice on an average of 7.5 tons and that their only plea was that this was not intentional or deliberate nor did it signify any mala fides on their part. No reliance has, however, been placed upon this confession in the Respondents statement of the case. No reliance has, however, been placed upon this confession in the Respondents statement of the case. We may however point out that the gross annual turnover of sugar manufactured in the factory is in the neighbourhood of 12 lakhs of maunds and the amount of excise duty the appellant pays to Government runs into about a crore of rupees per annum. It would therefore be a little far-fetched to infer that what happened at the time of the inspection was something more than an error occasioned perhaps by carelessness nor could it be said that this shows that there was a deli- berate attempt on their part to evade payment of duty on amere 11,606 maunds and amounting to less than a lakh of rupees. We are saying this not because an error due to carelessness in maintaining the registers properly as required by Rule 83 does not amount to a contravention of that rule ; the only reason why we mention this is that in a factory where the turnover of sugar is so considerable and the operations conducted in which the human element plays a significent part it would not be right to base calculations on the surmise that over filling of the tanks was being practised systematically. No doubt, during the test, the tanks were slightly over filled on nine out of ten occasions as pointed out by the Assistant Chemical Examiner. But this could be atributed to a slight failure of the human element resulting from the fact that a special operation was being conducted by the operators in the presence of a government official. We would, however, make it clear that these observations are just incidental and are not the basis of our decision. (16) FOR the reasons already stated, we quash the order of the central Government and the two tribunals below as well as the directions requiring the appellant to pay the additional excise duty of Rs. 90,921-14-0 on 11,606 maunds of sugar and fine of Rs. 2,000.00. If the appellant has paid the duty and the penalty it should be refunded. (17) THE respondent will pay the costs of the appellant in this court.