R.L. Gulati, J.
1. This is a reference under Section 11(1) of the U.P. Sales Tax Act submitted by the Additional Judge (Revisions), Sales Tax, Agra, at the instance of the assessee and relates to the assessment year 1964-65. The connected reference relates to the assessment year 1965-66.
2. The assessee carries on business in oil-seeds, foodgrains and gur in the trade name of M/s. Shyam Lal Om Prakash. It has two shops one at Dankaur, which is the head office, and the other at Gulaothi, which is the branch. On 7th December, 1965, a survey of the assessee's business premises at Gulaothi was carried out by the Sales Tax Officer, SIB (Special Investigation Branch). As a result of the survey 27 loose sheets of paper and one exercise book containing some transactions were discovered. The entries recorded on the loose sheets were found to have been entered in the account books. The exercise book contained entries from 18th November, 1965, to 6th December, 1965, i.e. for a period of one month and 17 days. The total of the credit entries amounted to Rs. 1,57,147 while the entries in the debit side totalled to Rs 2,22,714. On verification it was found that transactions worth about Rs. 88,293,99 had not been recorded in the account books and the rest had been recorded. The assessee's explanation was that the exercise book did not contain any transactions of sale or purchase but was a private account for the transactions of loan and their repayment. This explanation of the assessee was not believed and the amount of Rs. 88,000 and odd was held to be suppressed turnover. In the circumstances, the Sales Tax Officer rejected the assessee's account books and made a best judgment assessment for both the years enhancing the turnover by Rs. 6,00,000 for the assessment year 1965-66 and Rs. 3,00,000 for the assessment year 1964-65. On appeal, the appellate authority confirmed the assessments except that it reduced the turnover of gur by Rs. 75,000 for the assessment year 1965-66 and by Rs. 40,000 for the assessment year 1964-65.
3. The assessee then applied in revision in respect of both the years but remained unsuccessful. At the instance of the assessee, the revising authority has submitted this combined statement of the case relating to the two assessment years and has submitted for the opinion of this court the following questions of law :
(1) Whether the survey dated 7th December, 1965, can be held relevant for the year 1964-65 and whether a suppression could be presumed for that year on that basis ?
(2) Whether the facts and circumstances of the case as held in revision corroborate the estimate of turnovers for the years 1964-65 and 1965-66 ?
4. The date of the survey, namely, 7th December, 1965, admittedly falls in the assessment year 1965-66. We are, however, of the opinion that it is not the date of the survey which is material for determining the question as to whether the result of such survey would be relevant for an earlier assessment year but what is material is the dates and the nature of the suppressed transactions which are discovered as a result of the survey. The exercise book which has been held to be a duplicate account book contains transactions relating to the period 18th November, 1965, to 6th December, 1965. This period falls within the assessment year 1965-66. There is no material or finding that any of the transactions recorded in the exercise book related to a period which would fall in the assessment year 1964-65 nor is there any finding or even a suggestion that any entry in the exercise book disclosed a balance carried over from an earlier year. There might be other circumstances which might suggest that the assessee might have practised suppression of turnover even in the earlier year. But no such circumstance has been indicated by the Judge (Revisions) in his revisional order giving rise to the present references. In the circumstances, it is not possible to hold that on the basis of the survey carried out on 7th December, 1965, any suppression of turnover during the year 1964-65 could be presumed. The Judge (Revisions) held the survey to be relevant lor the assessment year 1964-65 on the view that 'it cannot be believed that the applicant started making duplicate set of accounts only in the year 1965-66. There is some continuity of habits in business.' This conclusion of the Judge (Revisions), in our opinion, is based on no material whatsoever and is a pure conjecture or a surmise. Even the best judgment assessment has to be supported by some material. The material may be direct or circumstantial but a mere suspicion or conjecture cannot take the place of evidence or material. We are, therefore, unable to uphold the finding of the Judge (Revisions) so far as the assessment year 1964-65 is concerned. On the facts and circumstances of the case the survey of 7th December, 1965, was not relevant for the assessment year 1964-65 and no suppression could be presumed for that year on the basis of the survey.
5. Sri Ram Manohar Sahai, learned counsel for the department, has placed reliance upon a decision of a Division Bench of this court in Jagannath Baboo Lal, Hatia, Kanpur v. Commissioner of Sales Tax, Sales Tax Reference No. 189 of 1957 decided on 7th February, 1963 in support of the contention that the survey made during one year can be relevant for the assessment of the earlier year. That case, however, can be easily distinguished. No doubt in that case also survey was made on 4th May, 1954, which was outside the assessment year 1953-54, the assessment year in dispute. But what was discovered as a result of the survey was that the assessee had been following a different method of account keeping than what he had disclosed in the assessment year in dispute. It was discovered that the assessee had been maintaining phutkar jakar bahi and loose papers containing record of the sales. It was further found that the assessee maintained a kachchi rokar but some sales were not entered in it. Numbers of all khatas were not given in the kachchi rokar. And it was also found that according to the accounts the turnover of the assessee was of Rs. 1,06,113-11-0 but it had stock of the value of Rs. 62,694 on a certain date. On these materials this court found a logical connection between the rejection of the assessee's accounts for the assessment year in dispute on the basis of the recovery made during the survey on the ground that it could not be said that the assessee had suddenly changed his method of accounting with effect from 1st April, 1954. When an assessee is found to have kept his accounts on a certain method of accounts, different from the one disclosed to the department, the department is entitled to presume that the assessee must have followed the same method of accounting for the immediately preceding year unless the assessee was able to prove that he had adopted a different method from the beginning of the next assessment year. Moreover, there is a well-recognised method of calculating the turnover with reference to the level of stock. The other circumstance which weighed with the Bench was that the time between the end of the assessment year 1953-54 and 14th May, 1954, was not unduly long. In the present case there is no finding that the method of accounting being followed by the assessee at the time of survey was different from the one which it had been disclosed to the department in the preceding year nor is there any finding that the level of stock during the assessment year 1964-65 was disproportionately high as compared to the total turnover disclosed for that year. The time lag between the survey and the end of the assessment year also in the instant case is fairly long.
6. For the reasons stated above, it is not possible for us to hold that the survey made during one assessment year is always relevant for the preceding year. As stated earlier an answer to the question like this always depends upon the facts and circumstances of each case.
7. The learned counsel for the assessee on the other hand relied upon a recent decision of a Division Bench of this court in Abdul Samad, Yarn Dealer v. Commissioner of Sales Tax, U.P., Lucknow, Sales Tax Reference No. 426 of 1968 decided on 9th Feburary, 1970. Although the facts of printed at p. 390 infra that case were slightly different from the facts of the present case, yet that decision does support the views we are taking that the survey made during a subsequent year cannot always be held to be relevant for determining the turnover of an earlier year.
8. Accordingly, we answer question No. (1) in the negative in favour of the assessee and against the department.
9. Question No. (2) relates to the estimate of the turnover for the two assessment years. After having answered question No. (1) in favour of the assessee, it follows that the enhancement made in the turnover of the assessment year 1964-65 based as it was on the survey of 7th December, 1965, could not be upheld. We, therefore, answer question No. (2) so far as it relates to the assessment year 1964-65 in the negative in favour of the assessee and against the department.
10. We are now left with the part of question No. (2) which relates to the assessment year 1965-66. The only question is as to whether the estimate of the turnover for that year is valid in law. Sri Brij Lal Gupta, learned counsel for the assessee contends that the addition of Rs. 6,00,000 is arbitrary for several reasons. In the first place the suppressed turnover of Rs. 88,000 and odd related to a period of one month and 17 days only. The Sales Tax Officer had calculated the suppressed turnover for the whole year in the same proportion which the period of one month and 17 days bears to the whole year. According to him, the assessee was dealing in gur also and as observed by the Appellate Assistant Commissioner, business in gur is seasonal lasting for a period of six months or so. He, therefore, submits that 'the rule of three' could not be strictly applied in estimating the suppressed turnover. There are two answers to this submission. In the first instance, the Sales Tax Officer has not applied strictly 'the rule of three' inasmuch as the total turnover for the year according to 'the rule of three' would come to over Rs. 7 lacs whereas the Sales Tax Officer has fixed the turnover at Rs. 6,00,000 leaving enough margin for the seasonal nature of business in gur. Moreover, the Appellate Assistant Commissioner has taken this aspect into account and has already allowed a reduction of Rs. 75,000 in the turnover of gur. The learned counsel then submitted that the suppression was discovered only at the branch office and the Sales Tax Officer could not legally presume any suppression at the head office. In support of this contention the learned counsel has relied upon a decision of the Supreme Court in the State of Kerala v. C. Velukutty  17 S.T.C. 465 (S.C.). There is a clear fallacy in this argument. The Sales Tax Officer has merely enhanced the total turnover of the assessee by a sum of Rs. 6,00,000 on the ground that the suppression of the turnover of Rs. 88,000 in a period of one month and 17 days would justify such an addition. He has not made a separate enhancement in the turnover of the head office as was the case in the case before the Supreme Court relied upon by the learned counsel. The total turnover, of course, is comprised of the turnover ,of the two shops but there is no warrant for saying that the sum of Rs. 6,00,000 has been added to the total turnover because of the suppression in the head office. The learned counsel then submitted that during some surveys carried out earlier the department was not able to detect any irregularity in the assessee's accounts. It is only during the survey made on 7th December, 1965, that some suppression was discovered which, according to the learned counsel, is only a stray instance of suppression. This argument again is not sound. Mere fact that the department was unable to lay its hands on the material relating to suppression on the part of the assessee during the survey made earlier does not mean that the assessee was not practising evasion earlier. The exercise book does not contain merely a single transaction of suppression. It shows a continuous and methodical day to day suppression for a period of one month and 17 days. It cannot, therefore, be said that it was a stray instance of suppression which was discovered during the survey. From the magnitude of the assessee's business also, it cannot be said that the estimate of the turnover is so excessive as to be called arbitrary. For the assessment year 1965-66, the assessee itself declared the total turnover of purchases at over Rs. 20 lacs, although he admitted the net turnover at about Rs. 7,84,000 only. The Sales Tax Officer had fixed the total turnover at Rs. 19,36,000 and odd out of which a reduction of Rs. 75,000 was allowed by the appellate authority. In the circumstances, it is not possible to accept the contention of the learned counsel that the estimate of the turnover is without any material or is arbitrarily high.
11. Once circumstances exist to justify a best judgment assessment, the estimate of turnover is a question of fact and a finding on that question by the Judge (Revisions) who is the final fact finding authority under the Act, is not open to challenge merely on the ground that the estimate of the turnover is excessive. We have shown above that the estimate of turnover is supported by cogent materials and is in keeping with the magnitude of the assessee's business.
12. We, accordingly, answer the second part of question No. (2) relating to the assessment year 1965-66 in the affirmative in favour of the department and against the assessee.
13. As the assessee and the department have both succeeded in part, we make no order as to costs. The fee of the learned counsel for the department, however, is assessed at Rs. 200