This is an appeal from the Tax Court's ascertainment of deficiencies totaling $23,547.07 in petitioner's income taxes for the years 1947 through 1950 and from the assertion of penalties of $1,070 for 1947 under § 293(b), I.R.C.1939, 26 U.S.C. § 293(b),1 and of $2,246.92 under §§ 294(d) (1) (A) and (d) (2), I.R.C. 1939, 26 U.S.C. § 294(d) (1) (A), (d) (2).2

During the critical period petitioner was the owner of four farms in Lancaster County, Pennsylvania; two of these he operated himself and the other two were rented on an equal shares arrangement. He also engaged in wholesaling potatoes, a business which prospered and which required most of his time. His financial records are available only from August, 1949 on, those prior to that time having been destroyed under what the Tax Court justifiably found were innocent circumstances. Bank records were in existence, however, and since petitioner transacted virtually all of his business through his checking account, reliance on the bank records by both parties is well placed.

Several questions are presented, the one most fundamental to the case being whether the Commissioner could and did properly resort to the net worth method for asserting the deficiencies. § 41, I.R. C.1939, 26 U.S.C. § 41 states:

"The net income shall be computed * * * in accordance with the method of accounting regularly employed in keeping the books of such taxpayer; but if no such method of accounting has been so employed, or if the method employed does not clearly reflect the income, the computation shall be made in accordance with such method as in the opinion of the Commissioner does clearly reflect the income. * * *"

Since no records were available for the period from January 1, 1947 until August 9, 1949 the Commissioner was clearly entitled to resort to the net worth method to determine petitioner's income during that interval. Hooper v. United States, 10 Cir., 1954, 216 F.2d 684. For the remainder of the time in question there were records, but it is equally clear that the government may resort to the net worth method for testing their accuracy. Davis v. C. I. R., 7 Cir., 1956, 239 F.2d 187; Jacobs v. United States, 1954, 126 F. Supp. 154, 131 Ct. Cl. 1. If glaring discrepancies are found, as in this instance, the net worth computation may be used as prima facie evidence of the actual amounts of income. Holland v. United States, 1954, 348 U.S. 121, 75 S. Ct. 127, 99 L. Ed. 150; Kite v. C. I. R., 5 Cir., 1955, 217 F.2d 585; United States v. Ridley, D.C.Ga. 1954, 120 F. Supp. 530. In the absence of serious challenge to the Commissioner's use of the net worth system of computation his determination is presumptively correct. Rubino v. C. I. R., 6 Cir., 1955, 226 F.2d 291. See Helvering v. Taylor, 1935, 293 U.S. 507, 515, 55 S. Ct. 287, 79 L. Ed. 623. Having decided that net worth can be applied, we come to the questions raised by petitioner concerning the correctness of the mechanics employed in so doing.

As a formula the net worth method might be stated: increase in net worth plus non-deductible disbursements minus non-taxable receipts equals taxable net income. "Increase in net worth" depends, of course, upon careful measurement of net worth at the beginning and at the end of the year. Though the initial determination of net worth — here petitioner's net worth as of January 1, 1947 — must be fixed carefully, it does not have to be done to a mathematical certainty. Gariepy v. United States, 6 Cir., 1951, 189 F.2d 459; United States v. Glazer, D.C.E.D.Mo.1952, 14 F.R.D. 86.

Petitioner complains that the Commissioner's treatment of checks drawn and presumably delivered but still outstanding at the end of the tax year was incorrect. Petitioner asserts that his bank balance should have been reduced by the aggregate amount of such outstanding checks, thereby decreasing the figure for net worth at the end of each taxable year. The Commissioner counters that the bank balance was not so reduced because the petitioner had been on a cash basis rather than an accrual basis for the years in question, and to have reduced the bank balance for checks outstanding would be a distortion of the income figures under that plan.

Most of the outstanding checks in each year represented disbursements for business expenses, and were deductible. For a taxpayer on a cash basis, as petitioner was, the year in which a business expense can be deducted turns on when it was paid. The question is whether the expenses were paid if the checks were still outstanding. As a general proposition delivery of a check will establish the same right to a deduction as would delivery of cash. It does not matter that the check was not cashed or deposited or the drawer's account charged until the following year. The check is regarded as payment on a condition subsequent, and if the condition of honor on presentment is met the payment is regarded as absolute from the time the check was delivered. C. I. R. v. Bradley, 6 Cir., 1932, 56 F.2d 728, affirming 1930, 19 B.T.A. 49; Field, 15 T. C.M. 631 (1956). And see Broussard, 1951, 16 T.C. 23; Estate of Spiegel, 1949, 12 T.C. 524. Where the check is given with the understanding then or later that it will not be cashed in the current tax year, a cash basis taxpayer is not allowed a deduction. Eagleton v. C. I. R., 8 Cir., 1938, 97 F.2d 62; Fischer, 1950, 14 T.C. 792. It has been held that the date of mailing the check to a charity was the date of payment for the cash-basis drawer. Witt's Estate v. Fahs, (D.C.Fla.1956), 160 F. Supp. 521. Consequently it would seem that the petitioner might have deducted the amounts of the outstanding checks representing payments for business expenses, absent any evidence that there was an understanding with each payee the checks would not be cashed until a new tax year had begun. Since that is so it would appear that the cash-basis taxpayer subjected to a net worth computation of his tax liability would be entitled to have his year-end bank balance reduced by the amount of all outstanding checks. The presumption conforming with experience would be that outstanding checks have been delivered; the burden of proof would then be on the government to show that they had not been delivered. Those checks representing non-deductible disbursements would also have their tax effect in the earlier year, being added into the non-deductible expenditures element of the determination of taxable income.

The Tax Court in its opinion did not allow the amounts of the checks outstanding to reduce the bank balance at year's end because it found that all of them except insignificant amounts were for the acquisition of assets which it assumed were still on hand at the end of the tax year. Hence the change in net worth would not be affected by an accounting formality crediting one asset and debiting another, leaving the total unchanged. This reasoning is acceptable as to the checks representing purchases of livestock for reasons to be later set out, but it has no support in the evidence with reference to potatoes and feed. The feed may already have been consumed, the potatoes sold and payment received, in which case petitioner's increase in net worth and consequent tax liability would be overstated.

What petitioner's eventual tax liability will be is for the Tax Court to say after a redetermination of the figure for "Cash on hand" represented by bank balances at the end of each year. The checks outstanding at the end of the years 1946 through 1950 will be given effect in the year they were delivered in such a redetermination.

Another item in the annual computation of net worth which petitioner asserts is represented by erroneous figures is livestock on hand. It seems to be agreed that the Commissioner's figures for 1947, 1948 and 1949 are correct, leaving those for 1946 and 1950 in dispute. Petitioner told the investigating agents and testified before the Tax Court that it was his practice to buy cattle in the fall, to keep them over the winter, and to sell them in the spring. There is no way of ascertaining the identity of animals sold or in fact how many were bought and sold or remained on hand at any one time. The Commissioner therefore deemed it reasonable to take the total of disbursements for livestock made between October 1 and December 31 each year as representing the value of livestock bought each fall and on hand the beginning of each succeeding tax year. Petitioner claims this is arbitrary. He suggests a figure of some $16,000 rather than the $12,000 figure of the Commissioner for December 31, 1946 and $11,443 rather than $12,592 for December 31, 1950. Petitioner complains that according to the evidence some cattle purchases were made for cash and that some of the bank records showed purchases before October 1st. There is insufficient foundation in the record as to sales of cattle and as to the amounts of cash disbursements for livestock and the years in which they were made to warrant replacing the Commissioner's figures with petitioner's. It is for this reason that outstanding checks at year's end which represent purchases of livestock need not be subtracted from the year-end bank balances; the amounts would simply require transfer to another category of assets. We assume that the Commissioner in his original determination did not include in livestock inventory any animals purchased by checks still outstanding at year's end, as that would have had the effect of twice including their value. The judgment of the Tax Court with respect to its treatment of livestock will be affirmed.

It is further claimed that the Commissioner erred in not allowing some $3,000 in his initial calculation for the end of 1946 for trucks. The Tax Court was justified in sustaining the Commissioner's figure by finding that the trucks then on hand had already been fully depreciated, and its judgment with respect thereto will be affirmed.

Petitioner disputes the figure of $2,000 used throughout for value of farm equipment but his accountants used that figure also, there simply being inadequate evidence as to the value of the equipment. Petitioner did himself testify as to equipment owned at the end of 1947 and gave an estimate of valuation at between seven and nine thousand dollars. Nevertheless there is nothing to show whether this valuation was for the equipment new or in its 1946 condition, or to indicate its expectancy of continued usefulness. The burden of proving the invalidity of the Commissioner's figures has not been carried on this point.

The Tax Court held that there had been no arbitrary disregard by the Commissioner for costs in the purchase and improvements of petitioner's warehouse. It found the evidence lacking in support for the figures advanced by petitioner through his accountants; our own examination of the record forces agreement. Much the same must be said for petitioner's contention that he is entitled to greater reserves for depreciation than were allowed; the record fails to bear out the larger figures of petitioner's accountants. In his brief petitioner mentions that the age, type of construction, and mode of use of the assets called for a greater depreciation percentage though there is no evidence respecting these items. The Tax Court's treatment of depreciation, therefore, was proper.

A controversy remains as to what the Commissioner denominated "Cash drawings and checks" in the following amounts:

     1947         1948          1949          1950

  $3,660.32     $10,016.05     $6,059.21     $5,709.16

These are the totals of cash received at the time checks were deposited by petitioner and of checks drawn by him for purposes he could not identify; these sums were all assumed to be for non-deductible expenses. This treatment of the cash and unidentifiable checks well illustrates the fact that a taxpayer confronted with an investigation is presented with a dilemma of how much explaining he ought to do. "* * * (T)he prosecution may pick and choose from the taxpayer's statement, relying on the favorable portion and throwing aside that which does not bolster its position * * *."3 The record reveals that the taxpayer did cooperate; but his statements that at least some of his deductible disbursements were made in cash were rejected as self-serving. The Tax Court also rejected the evidence as to petitioner's living expenses, enough about which appears to indicate that they were minimal. Petitioner testified that his family raised practically all of their own food and that they bought only sugar, salt, coffee, and other minor staples; he estimated that the weekly grocery bill was three to five dollars per week which seems not unreasonable for a frugal family of five that produced much of its own food. Petitioner does not use tobacco; it is stated in the brief that he does not use intoxicating liquors either, but the transcript contains nothing as to this. There was further testimony that he hauled his own coal which cost him about $100 a year. He calculated clothing expenses for his family to have been about $200 annually. He said he gave between $250 to $500 each year to his church. During the disputed years he had not yet ever attended a movie. In 1949 petitioner and his wife drove to Florida in petitioner's car with another couple who paid for gas and oil; once there, they stayed with relatives. In 1950 they went to California, again with another couple who shared expenses; petitioner recalled that the trip had cost him some $600. Plainly petitioner's habits were not extravagant. In view of that conclusion and giving some credence to the testimony of cash disbursements for deductible expenses, it would follow that the "cash drawings and unidentified checks" entry has, at least for some years, been unnecessarily overstated. How much of the language of the Holland case in 348 U.S. at pages 135-136, 75 S.Ct. at pages 135-136, requiring the government to some extent to corroborate its proofs by negating reasonable explanations furnished by the taxpayer would necessarily be applicable to a civil case relying on net worth computations is debatable. It is proper, however, that the rule of Cohan v. Commissioner, 2 Cir., 1930, 39 F.2d 540 be applied to the facts before us. Some estimation of the amount of deductible items involved in the cash withdrawals and unidentified checks is called for. Cf. Potson v. C. I. R., 1954, 22 T.C. 912. The refusal to make any such estimation was arbitrary.

Assessment of the fraud penalty for 1947 of $1,070 under § 293 (b), Internal Revenue Code of 1939, was rested by the Tax Court on its finding that petitioner had reported only $10,593.63 from the sale of tobacco which brought $24,500. Petitioner urges that the Commissioner did not attempt to assess the additional taxes that would have been due, ostensibly because the net worth computation did not sustain the proposition that petitioner had actually received for himself the $24,500. Petitioner testified that his practice upon receiving a check in payment for products marketed by him in the joint interest of himself and his tenant farmers was to deposit the check in his account and then to draw his own check on the account in favor of the tenant farmers with whom he had the shares agreement. He, in those circumstances, would not have had to report as gross income more than half of the amount received. The discrepancy between that sum and what petitioner did report is not explained by proof adduced by either side, but it is a fair assumption that the portion of the proceeds of sale attributable to the costs of production would have been remitted to the tenant farmers. Petitioner's share would have been in the profits. It is concluded therefore that the burden of proving fraud, § 112, Internal Revenue Code 1939, 26 U.S.C. § 112, was not met by the Commissioner and that the evidence was insufficient to support the fraud assessment. It will be disallowed.

Petitioner asserts that the collection of taxes for the years 1947 and 1948 is barred by the statute of limitations. Assessment of a deficiency must be made within three years after the filing of the return. § 275(a)-1 Internal Revenue Code, 1939, 26 U.S.C. § 275(a). But if a taxpayer has omitted from his return gross income amounting to more than 25% of the gross income stated in the return, assessment may be made within five years. § 275(c) Internal Revenue Code 1939. The income tax returns for the years 1947 and 1948 were filed on March 10, 1948 and March 7, 1949, respectively. The taxpayer had executed consents extending to June 30, 1954, the period for assessment of taxes for 1947 and 1948 with the provision that the consent applicable to 1947 was "not intended to extend the statutory 3 year period and will be operative only if it can be ultimately established that Sec. 275 (c) of the Internal Revenue Code is applicable." § 276(b) Internal Revenue Code 1939, 26 U.S.C. § 276(b). Consents for 1949 and 1950 were also entered into. The deficiency notice for the taxable years was mailed on February 24, 1954. It can be seen that the petitioner's contentions as to the statute of limitations are without merit except possibly as to 1947.4 There only the five year limitation was extended to June 30, 1954. Since a recomputation will be necessary it is not possible for us to pass definitively on this question. It does appear, however, that petitioner's gross income was reported by him as $35,496.87; this means that if the outcome of the recomputation does not give a gross income in excess of this figure by 25% thereof — namely $44,371 — the assessment will have been out of time.

One further point remains. The Commissioner imposed an addition of $1,404.32 under § 294(d) (1) (A), Internal Revenue Code 1939, for failure to file a declaration of estimated 1950 tax on time, and an addition of $842.60 under § 294(d) (2), Internal Revenue Code 1939 for a substantial understatement of estimated 1950 tax.5 The assessment for failure to file in time may be avoided by proof that the failure was due to reasonable cause and not to willful neglect. No such defense is provided in § 294(d) (2). Petitioner argues that the penalty assessments should not be imposed because he relied completely on the accountant he engaged to handle tax matters. The situation is not distinguishable from Bouche v. Commissioner, 1952, 18 T.C. 144, which found a similar defense insufficient because there was no proof as to the reasonableness of so great reliance.

The decision of the Tax Court will be affirmed in the respects above noted. It will be reversed as to the determination of fraud in the 1947 return and as to the fraud penalty assessed for that year. It will be vacated and returned to the Tax Court for a redetermination of the figure for "Cash on hand" represented by bank balances at the end of each year concerned; for an estimation of the amount of deductible items involved in the cash withdrawals and unidentified checks and for a resolving of petitioner's claim that the statute of limitations bars the collection of taxes for 1947, taking into consideration the above referred to recomputations.