Before us is a petition for review of a decision of the Tax Court involving asserted deficiencies in income tax against the petitioners for the years 1946 and 1947. The jurisdictional requisites appearing to have been complied with, this Court has jurisdiction by virtue of Section 7482(a) of the Internal Revenue Code of 1954, 26 U.S.C.A. § 7482 (a), (Section 1141(a) of the Internal Revenue Code of 1939 as amended 26 U.S. C.A. § 1141(a)).

The Commissioner of Internal Revenue asserted certain deficiencies and fraud penalties against the petitioner Gene Clark for the calendar years 1946 and 1947, and deficiencies against his wife Faye Clark for the calendar year 1947. These deficiencies and penalties were redetermined by the Tax Court, and from this decision the parties have petitioned for review.

The facts as stipulated and found by the Tax Court appear essentially as follows: During the calendar years 1946 and 1947 the petitioners, Gene and Faye Clark, sometimes hereinafter referred to as Gene and Faye, resided in Los Angeles County, California, and all income derived by them during those years was community income. For the calendar years 1946 and 1947 they filed separate income tax returns on the community property basis, and for the calendar year 1948 they filed a joint return.

Prior to April 23, 1946, Gene Clark and Archie Koyl were associated in a business venture known as Gene Clark Plumbing Company, hereinafter referred to as the “company”, consisting of two shops located in El Monte and Bell Gardens, California, and having a labor force of approximately 35 employees. The plumbing company was engaged primarily in selling plumbing supplies and rendering plumbing services to building contractors.

Gene Clark and Archie Koyl organized a California corporation, Gene 0. Clark, Inc., now known as Atlas Pipe and Supply Company (hereinafter sometimes called the “corporation”) on April 23, 1946, to engage in the wholesale plumbing business. Of the 522 shares of $100 par value stock authorized, 365 shares were issued to Gene, president of the corporation, and 157 shares to Koyl, vice president; one qualifying share was issued to another individual not involved herein. Gene acquired his shares at a cost of $36,500. His shares represented an ownership interest in the corporation of approximately 70 per cent.

On or about March 29, 1948, Gene purchased the 157 shares of stock owned by Koyl for $24,714.49. Thereafter, on March 1, 1949, Gene sold all of his stock to the Koyls, 262 shares to Archie, 260 shares to Fawn. For the period involved here then the proportional stock ownership in Gene Clark, Inc., can be summarized as follows:

Gene Clark, Inc., commenced its business operations on or about April 23, 1946, occupying the same premises as the plumbing company. Within a few months after the formation of the corporation, the inventory of the plumbing company and that of the corporation were commingled and were thereafter kept as a single unit. The only employees on the business premises were those of the corporation. No records were kept that could properly reflect any business transactions of any plumbing enterprise other than the corporation. The only book kept in the office that had any connection with the plumbing company was a check book on the Bank of America in El Monte. No federal tax returns were filed on behalf of the plumbing company for any period after April 23, 1946.

The plumbing company, however, existed for an indeterminate period after the organization of the corporation, solely for the purpose of buying and selling plumbing materials in violation of the then existent regulations of the Office of Price Administration (OPA). This was done because the plumbing company did not hold any license to do business which could be revoked if it were found guilty of violating OPA regulations. The plumbing company was to serve as a front in such transactions for the corporation which held a license to do business.

The corporation kept its books on an accrual method and reported its income on a fiscal year basis beginning with the year ending April 30, 1947. After incorporation of the plumbing enterprise, customers would frequently make out checks to Gene Clark, to the corporation, or to the plumbing company. To obviate the resultant confusion, the corporation adopted a rubber stamp showing all three designations in order that it might properly endorse any check. This composite stamp was used throughout the period here in question.

Until he sold his entire interest to the Koyls in March of 1949, and during the years in question here, Gene was in general control of all of the corporate opera tions and dictated its financial policies. Clark was in full charge of the main office in El Monte, Archie Koyl directed the activities at the shop in Bell Gardens. Virtually all other corporate activities, including the maintenance of corporate records and the disposition of receipts, were under the direct control of Clark.

Fred Files, comptroller and office manager of the corporation, worked under the immediate supervision and direction of Clark. Files’ duties consisted primarily of handling receipts and keeping proper office records. He worked at both shops, keeping one set of books for the entire operation, though consecutively numbered duplicate receipt books were maintained in both shops. When cash was received from a customer, the amount thereof was recorded in the receipt book which was, in substance, merely a memorandum that was later transferred to the cash receipts journal in the books of account. Cash sales were sometimes totaled daily and sometimes only several times a week. A single “cash receipts” figure was usually recorded in the journal for the total amount of the separate sales. Deposits of the total cash receipts were made in the corporation’s bank account, and generally recorded weekly in the cash receipts journal. Files, who handled all of the bank deposits of the corporation, regularly deposited all cash receipts of the corporation which were turned over to him for such purpose by Clark. On a number of occasions, however, Clark instructed Files to set aside the cash proceeds from certain sales and to turn such funds over to him without recording the sales on the books. Also, at different times, Clark would give Files checks made out to the corporation by customers for sales, which sales were unrecorded on the corporate books, in exchange for the cash taken by Clark. An undetermined part of such cash proceeds was used to cash checks in relatively small amounts ranging up to $100, as an accommodation for neighborhood stores and workmen.

There was a substantial but undetermined difference in the amount of cash Files recorded in the corporate books or deposited in its bank account and the amount of cash sales actually made by the corporation. This method of handling cash sales was also the general practice of the company, and was not altered by the coming into existence of the corporation during the taxable periods involved herein.

Between May 1946 and December 1946, Clark, on behalf of the corporation, frequently sold and shipped plumbing materials from the El Monte yard without any entry being made for the transactions in the corporate records. Some of such shipments represented “trading transactions” or non-profit exchanges of materials with competitors for mutual convenience.

Subsequent to December 1946, Clark frequently sold and traded plumbing equipment on behalf of the corporation. He also sold and traded used vehicles. Files was not supplied with the appropriate sales slips and proceeds on many of these transactions. Sometimes Clark would simply give the comptroller a check, without adequate details connected with the sale, and instruct him to remove the particular asset from the corporate books.

During 1947, when maximum ceiling price regulations were in effect under the Office of Price Administration, Clark engaged in black market activities. When he dealt in such illicit activities, Clark would generally pay an undisclosed amount of cash for the purchase of materials over the price indicated on the invoice. These cash funds were taken from unreported corporate receipts. The over-ceiling cash payments were not recorded on the corporate books as part of the total cost of the illicit purchases. Neither Clark, the company, nor the corporation reported the profits from such illegal sales transactions.

During each of the taxable years in question Clark also had an arrangement with Keenan Pipe & Supply Company whereby he was able to purchase materials on behalf of the corporation at one-third off for cash. Under this arrange ment, indeterminate amounts of such purchases were made and paid for, usually with receipts obtained from unreported corporate sales, the parties agreeing not to keep any records of their cash transactions.

Apart from the foregoing modus operandi, during each of the years involved herein, Gene Clark, Inc., made numerous purchases of plumbing materials in the normal course of business which were not recorded on the corporate books. The subsequent sales thereof were likewise unrecorded. Also, in many instances, the profits from such sales were neither reported by the corporation on its income tax returns, nor by Clark on his returns for the years in issue. The corporation also rendered plumbing services for building contractors on a number of housing projects during the years in question, and Clark failed to record the full receipts therefor on its books.

In due course the Commissioner of Internal Revenue authorized one of its agents to investigate the income tax returns of Gene Clark, Inc. This investigation revealed that during each of the taxable years in which Clark was an officer and stockholder of Gene Clark, Inc., these substantial but undisclosed and undetermined amounts of receipts from sales made by and on behalf of the corporation were neither recorded on its books nor reported in its tax returns. As the result of this investigation the agent made the following adjustments to the net income of the corporation:

Year Net Income Per Returns Additions to Net Income per Revenue Agent’s Report Total Net Income per Revenue Agent’s Report

1947 ',632.10 $102,050.17 $132,682.27

1948 16,726.40 92,208.62 108,935.02

From these figures of net income the agent made certain adjustments in the figures which he concluded represented amounts which were available as earnings and profits for distribution to shareholders. He then inferred that these amounts of available surplus for the fiscal years 1947 and 1948 had been distributed to the shareholders. Being unable to determine precisely the manner in which these distributions were divided between the shareholders, the agent simply allocated the distributions in accordance with the shareholder’s stock interest. This report, with its allocations of distribution to shareholders, was used as the basis of the Commissioner’s deficiency notice mailed to Gene and Faye Clark.

In the Tax Court hearing the report was introduced not as evidence but to show the basis of the Commissioner’s deficiency notice. The Tax Court found much of the report to be inaccurate, for which it made certain corrections, and using the portions of the report not contested or on which it found that the petitioners had failed to meet their burden of proof to overcome the presumption of correctness, the court made its determination of the petitioners’ income tax deficiency and penalty. The Tax Court reduced the Commissioner’s deficiencies for Gene Clark for the calendar years 1946 and 1947 from $24,151.53 to $20,-893.80, and decreased the 50 per cent fraud penalty from $12,075.77 to $10,-446.91. The court also reduced the deficiencies in income taxes for Faye Clark for the year 1947 from $12,252.87 to $9,-288.48. The Commissioner concedes that these reductions are proper.

On the basis of the petitioners’ request for review, the case, aside from the questions of fraud and statute of limitations which we will consider later, basically involves the following questions:

1. The extent of earnings and profits of the corporation available for distribution as dividends during the years in question;

2. The extent to which distribution was actually made from these earnings and profits to the shareholders during those years;

3. The proper allocation of these distributions in accordance with the difference between the fiscal years of the corporation and the calendar years of the shareholders; and

4. The allocation of these distributions between the shareholders.

In analyzing the voluminous opinion of the Tax Court in what we consider to be the proper perspective, we are not unmindful that our function on review is merely to determine whether there is substantial basis in the evidence to support the findings of the Tax Court, Commissioner of Internal Revenue v. Scottish American Investment Co. Ltd., 1944, 323 U.S. 119, 124 , 65 S.Ct. 169 , 89 L.Ed. 113 ; Dobson v. Commissioner of Internal Revenue, 1943, 320 U.S. 489, 501 , 64 S.Ct. 239 , 88 L.Ed. 248 , and cases cited; Gen singer v. Commissioner of Internal Revenue, 9 Cir., 1953, 208 F.2d 576, 583 , and we must give due regard to the tax court’s opportunity to base its determination in part on its own appraisal of the credibility of witnesses. United States v. United States Gypsum Co., 1948, 333 U.S. 364, 395 , 68 S.Ct. 525 , 92 L.Ed. 746 ; Baumgardner v. Commissioner of Internal Revenue, 9 Cir., 1957, 251 F.2d 311, 313 . It must also be remembered that the Commissioner’s determination contained in a deficiency notice is presumptively correct and the burden of proving otherwise to the tax court is upon the taxpayer. Tax Court Rule 32, 26 U.S.C.A. Section 7453 ; Welch v. Helvering, 1933, 290 U.S. 111, 115 , 54 S.Ct. 8 , 78 L.Ed. 212 ; American Pipe & Steel Corp. v. Commissioner of Internal Revenue, 9 Cir., 1957, 243 F.2d 125, 126 ; Showell v. Commissioner of Internal Revenue, 1956, 238 F.2d 148, 152 ; Hemphill Schools, Inc. v. Commissioner of Internal Revenue, 9 Cir., 1943, 137 F.2d 961, 963 , although this does not mean that the taxpayer must also prove what the deficiency should actually be, Helvering v. Taylor, 3935, 293 U.S. 507, 513 , 55 S.Ct. 287 , 79 L.Ed. 623 . If the taxpayer introduces evidence from which the determination of the Commissioner contained in a deficiency notice could be found inaccurate then the presumption disappears. Gersten v. Commissioner of Internal Revenue, 9 Cir., 1959, 267 F.2d 195, 199 ; Clinton Cotton Mills v. Commissioner of Internal Revenue, 4 Cir., 1945, 78 F.2d 292, 295 ; Russell v. Commissioner of Internal Revenue, 1 Cir., 1939, 45 F.2d 100, 103 . See also Niederkrome, v. Commissioner, 9 Cir., 1958, 266 F.2d 238, 240 ; Lawrence v. Commissioner of Internal Revenue, 9 Cir., 1944, 143 F.2d 456, 457, 459 ; Hemp-hill Schools, Inc. v. Commissioner of Internal Revenue, supra, 137 F.2d at page 964 . Thereafter the Commissioner has the burden of proving the existence and amount of the deficiency. Cohen v. Commissioner of Internal Revenue, 9 Cir., 1959, 266 F.2d 5, 11 . The tax court’s determination must then rest on all of the evidence introduced and its ultimate determination must find support in credible evidence. Union Stock Farms v. Commissioner of Internal Revenue, 9 Cir., 1959, 265 F.2d 712, 724 .

In the instant case the tax court could not sustain the Commissioner’s determinations of the deficiencies due. For the calendar year 1946 the Commissioner assessed a deficiency of $12,052.82 and a fifty per cent penalty of $6,026.41 against Clark, while the tax court determined $11,752.35 and $5,876.18 respectively. For the calendar year 1947 the Commissioner assessed against Gene and Faye the sum of $12,098.71 each, while the tax court’s determination was only $9,141.45 each. As to the penalty against Gene, the Commissioner computed it at $6,049.36, and the tax court at $4,570.73.

The petitioners argue that “The deficiency notices having been shown to be arbitrary and erroneous and evidence having been so produced, the presumption in favor of Respondent’s determination ceased to exist” and “the Tax Court erred in treating the determinations of Respondent as evidence.” In effect, petitioners contend that, having found the deficiency assessments of the Commissioner to be erroneous, the tax court in fixing its deficiency assessments could not rely upon the Commissioner’s detex*minations as to individual items supporting his determinations of deficiency assessments even though as to such items the petitioners failed to sustain their burden.

However, if there is more than one item of unreported income or contested deduction included in the deficiency assessment, the tax court must weigh the petitioners’ evidence as to each item against the presumption of correctness in favor of each item. If from the petitioners’ evidence the Commissioner’s determination as to any particular item could be found inaccurate, the presumption in favor of the Commissioner’s determination of such item disappears. This does not mean, however, that the presumption of correctness also disappears as to the Commissioner’s determinations of other items included in the deficiency assessments as to which the petitioners have not sustained their burden of proof.

Here the Commissioner’s determinations, of which the petitioners complain, were contained in the revenue agent’s report which was introduced only to show the basis used by the Commissioner in arriving at his determinations. Naturally the tax court could not use this report as evidence of the accuracy of the figures contained therein, but the court, if it so desired could use the report as a framework for its own computations of the amount due, so long as the determination as to the individual items adopted by the court was supported by credible evidence independent of the report.

Keeping these principles in mind, we will consider the treatment of each item on the basis of the evidence before the court, to determine, first, whether the items should be considered as income available for distribution as dividends and, second, whether the item or its equivalent amount was actually distributed to the shareholders. We will consider later the question of proper allocation of the distribution of the dividends between the years involved and among the shareholders.

For fiscal 1946 the revenue agent’s report showed a total adjusted gross income of $138,473.43. Included in that amount were the following items:

(1) Substituted Items: In examining the corporate records the revenue agent testified that he employed the so-called “bank deposit” method of determining income not appearing either in the corporate records or in income tax returns. Using this method the agent analyzed the cash received, recorded sales from the sales journal, and cash deposits in the corporate bank account. The individual cash receipts were matched against the individual record of cash, item for item, for the purpose of identifying each item. Each individual cash sale was also matched against cash deposit items included on the bank deposit slips. This analysis revealed that the total deposits equaled the total recorded sales as would be expected. But it also revealed a substantial difference in the amounts of individual items of cash recorded in the corporate books, either in cash receipts or on the sales journal, and the individual items of cash noted in the corporate bank deposit slips. Theoretically, of course, for each item of cash received in a certain manner there should be a corresponding item of cash in that same amount reflected in the cash bank deposit slips, or otherwise accounted for in the records. From the fact that there was no such correlation the agent concluded that each individual amount deposited, for which there was no corresponding cash receipt notation, was a deposit substituted for cash which had been received and noted in the records but diverted to the shareholders before deposit. As testified by the agent, these totals of so-called substituted items, amounting to $14,806.77 in fiscal 1947 and $34,324.84 in fiscal 1948, were treated as unreported sales and included by the Commissioner in computing his deficiency notices as part of the money actually distributed to the shareholders.

In explanation of the discrepancies between the actual amount of cash received and the items deposited, noted in both years, petitioners introduced evidence regarding the corporation’s practice of cashing checks for neighboring businesses and the black market activities of Clark. The checks cashed by the corporation as an “accommodation” were run through the corporation’s bank account without any record of them appearing as cash receipts. The tax court recognized that the corporation was entitled to an allowance on the basis of accommodation checks, but rejected petitioners’ argument that such checks explained all of the substituted deposits which were not explained by the black market operations. In the absence of any showing by the petitioners as to the actual amount of such checks, the tax court was fully justified in resolving all doubts against the petitioners who were responsible for the inadequacy of proof and determining that ten per cent of the amount considered “substituted” represented the accommodation cheeks in both years. In doing so the tax court was merely applying the so-called Cohan rule, which requires the allowance of a deduction from income to be made upon proof that it is available to the taxpayer, but which allows, in the absence of records or other proof by the taxpayer of an exact amount of the allowance, the Commissioner to compute the amount allowable on such evidence and by such methods as the circumstances afford so long as they are not clearly arbitrary or capricious. “Absolute certainty in such matters is usually impossible and is not necessary; the Board (now the tax court) should make as close an approximation as it can, bearing heavily if it chooses upon the taxpayer whose inexactitude is of his own making.” Cohan v. Commissioner of Internal Revenue, 2 Cir., 1930, 39 F.2d 540, 543-544 . See also Union Stock Farms v. Commissioner, 1959, supra; Baumgardner v. Commissioner, supra, 251 F.2d at pages 313-314; Roberts v. Commissioner of Internal Revenue, 9 Cir., 1949, 176 F.2d 221, 226 , 10 A.L.R.2d 186 ; Cohen v. Commissioner of Internal Revenue, supra, 266 F.2d at pages 11 and 12; Chesbro v. Commissioner of Internal Revenue, 21 T.C. 123 , affirmed 2 Cir., 1955, 225 F.2d 674 , 676, certiorari denied 350 U.S. 995 , 76 S.Ct. 544 , 100 L.Ed. 860 .

The petitioners further contend that having applied the Cohan rule to the accommodation checks the tax court should have also applied it to allow a further reduction of the substituted check items by the cost of certain purchases made by Clark on behalf of the corporation. The tax court found that these were purchases of scarce commodities made on the black market, but found there was no other evidence besides Clark’s own testimony that they were for the benefit of the corporation. Refusing to believe Clark’s testimony, and with no other evidence in the record as to the amounts of these purchases or that the profits from the resale were turned over to the corporation or reported on its books or reported in income tax returns, the tax court concluded that the record afforded no basis for making a reasonable estimate of amounts expended on behalf of the corporation which would justify a reduction or eliminate such income. With this finding we agree. Merely because the tax court believed that there were some illicit purchases by a shareholder and an officer of the corporation does not require that it further find that they were made for or on behalf of the corporation. These purchases are in no way similar to the accommodation checks which clearly involved the corporation and which under the Cohan rule had to be allowed for in determining the corporate income. The tax court was clearly within its province as the trier of the facts in refusing to believe the self-serving testimony of Clark that his black market purchases also involved the corporation. Pool v. Commissioner of Internal Revenue, 9 Cir., 1957, 251 F.2d 233, 247 ; Masters v. Commissioner of Internal Revenue, 3 Cir., 1957, 243 F.2d 335, 338 , affirming 25 T.C. 1093 , 1095, 1100; Chesbro v. Commissioner, supra, 21 T.C. at page 129 .

Petitioners further claim that the testimony of their comptroller, Files, a government witness, corroborated Clark’s testimony that the corporation eventually ended up with the benefit of these black market cash purchases, but nothing in Files’ testimony, or those portions specifically cited by the petitioners in their brief or elsewhere compels this conclusion. It also ignores the further testimony of Clark himself that he was sure Files kept no records and even rather doubted if he knew of any amounts actually used to make the cash purchases.

In circumstances very similar to those existing here, it was held that receipt by shareholders of a corporation of the proceeds of sales attributable to the corporation but unreported by it are taxable as dividends to those shareholders to the extent of the corporation’s earnings and profits. Chesbro v. Commissioner of Internal Revenue, supra, 21 T.C. at page 129 . Here we cannot say that the tax court was clearly erroneous in finding that, except for the ten per cent reduction made by the tax court representing the accommodation checks, the petitioners failed to sustain their burden of proof that these substituted check items were not available for distribution and that the shareholders of the corporation had not received the benefit arising from these unreported sales.

(2) Notes Receivable — Officers.

The tax court stated the facts relating to this item essentially as follows:

On the basis of these facts, the revenue agent in his report treated the balance in the notes receivable account, $36,149.29, not only as actually distributed, but as an addition to the amount of adjustments to gross income— in effect, treating it as actual additional income to the corporation as well as a withdrawal. Clearly the tax court was justified in correcting such an obvious error, and not including the item in earnings available for distribution. The court, however, did agree with the agent that this was a distribution of earnings in the form of dividends and not loans as contended by petitioners. Whether a withdrawal is a loan is a factual question to be determined upon consideration of all the circumstances present in a particular case, and depends upon the exist ence of an intent at the time the withdrawal is made that it should be paid back. Wiese v. Commissioner, 8 Cir., 1938, 93 F.2d 921, 923 , certiorari denied 304 U.S. 562 , 58 S.Ct. 944 , 82 L.Ed. 1529 ; Estate of Isadore Benjamin, 1957, 28 T.C. 101 ; Estate of Helene Simmons, 1956, 26 T.C. 409, 423 ; Carl White, 1952, 17 T.C. 1562 , 1568; 1 Mertens, Law of Federal Income Taxation, Sec. 921. Upon all the facts found here, though individually they might not support the conclusion, we find the tax court’s conclusion that the withdrawals were dividends is not clearly erroneous. The tax court had before it the evidence of withdrawals roughly in proportion to the shareholders’ respective interests; evidence of a dividend declared and distributed to Clark when the loan was outstanding but not used to offset the loan; and the loan being paid back in part, as the bookkeeper testified, by checks from the corporation’s regular customers given to him by Clark, which the court inferred was another substitution. This evidence, coupled with the absence of written agreements, notes, interest, or security arrangements of any kind, which, of course, though not controlling are significant, is sufficient to support the finding of the tax court. Petitioners point out that there was an entry in the corporate books designating the withdrawals not as dividends but as “Notes Receivable — ■ Officers”; but book entries alone are not enough to establish the character of a withdrawal. Petitioners also point out that there was an action of the board of directors authorizing the loan, but, of course the board was controlled by the shareholders who were receiving the loans. And finally, the petitioners also point out that part of the consideration for Clark’s purchase of Koyl’s interest in the corporation was Clark’s assumption of Koyl’s portion of the indebtedness owed to the corporation. On its face, this might be significant, but with the purchase of Koyl’s interest Clark was the sole owner of the corporation. The tax court, therefore, was justified in treating this fact as insignificant in determining the true nature of this item.

(3) Other Distributions — In addition to the foregoing items, the Commissioner’s evidence shows that there was distributed to the shareholders and unreported, the sum of $2,378.50 from Y. L. Creed, who credited this amount as partial consideration for the sale of a house to Clark; the sum of $7,123.36 in the form of unidentified deposits in the personal bank accounts of the shareholders neither reported in the corporate books or in its tax returns; and the equivalent of various deductions which were disallowed to the corporation on the grounds that they were for personal expenses of the shareholders and not legitimate business expenses.

Petitioners claim there is no evidence to show they received and retained for their own enjoyment any of the income treated as distributed to them in the agent’s report. They do admit that not all of the cash received by the corporation was included in its books or tax returns, but their evidence relating to accommodation checks and black market sales, as we have pointed out, fails to fully account for all the unreported income. To constitute a distribution taxable as a dividend, the benefit received by the shareholder need not be considered as a dividend either by the corporation or its shareholders, declared by the board of directors, nor other formalities of a dividend declaration need be observed, if on all the evidence there is a distribution of available earnings or profits under a claim of right or without any expectation of repayment. Internal Revenue Code (1939), 26 U.S.C. Section 115 ; Healy v. Commissioner of Inter nal Revenue, 1953, 345 U.S. 278, 281 , 73 S.Ct. 671 , 97 L.Ed. 1007 ; Dawkins v. Commissioner of Internal Revenue, 8 Cir., 1956, 238 F.2d 174, 178 ; Chesbro v. Commissioner, supra, 21 T.C. at page 129 ; Paramount-Riehards Theatres, Inc. v. Commissioner of Internal Revenue, 5 Cir., 1946, 153 F.2d 602, 604 ; 1 Mertens, Law of Federal Income Taxation, Section 907, Section 912. See also Phelps v. Commissioner of Internal Revenue, 9 Cir., 1957, 247 F.2d 156, 158 . We hold that the tax court, on the evidence before it of the substituted deposits, the unrecorded sales, payments of personal expenses by the corporation, and the transfers of money under the guise of loans, did not err on determining that the shareholders had diverted all the corporate funds represented by these items to their personal benefit.

(4) Other “Available” Items—Also included in the $138,473.43 were five items totaling $63,488.47. Although the agent considered these items to be income he did not include them in his report as available for distribution or as distributed to the shareholders. Instead he deducted this amount from the $138,-473.43 and concluded the balance, $74,-984.96 as available for distribution and as having been distributed to the shareholders. The Commissioner apparently adopted the agent’s treatment for purposes of the deficiency notices. However, after the tax court hearing, the Commissioner, for the first time, changed his position and in his brief urged the tax court to include as available for distribution not only these five items but the amount shown as income in the 1947 corporate federal income tax return of $30,632.10, which the revenue agent had not included in the $138,473.43. No presumption of correctness could attach to these figures not included in the deficiency notice and on which the Commissioner had changed his position after the hearing. Tex-Penn. Oil Co. v. Commissioner of Internal Revenue, 3 Cir., 1936, 83 F.2d 518, 524 , affirmed on other grounds; Helvering v. Tex-Penn. Oil Co., 1937, 300 U.S. 481 , 57 S.Ct. 569 , 81 L.Ed. 755 ; J. A. Overman (1953), Op.Dkt. 45318, 12 T.C.M. 1453, 1454 ; 9 Mertens, Law of Federal Income Taxation, Sec. 50.61. The tax court included these items of $63,488.47 and $30,632.10 as earnings and profits available for distribution, but also apparently treated these items as actually having been distributed. Clearly the sum of $30,632.10 was properly classified by the tax court a3 available for distribution. This is also true of all items comprising the sum of $63,488.47 except one. The one exception is an item labeled “H. L. Brittan Account” in the amount of $1,860.40 which the tax court treated as available for distribution and as actually having been distributed. Except for the agent’s report, which was admitted for a limited purpose only, there is nothing in the record to support the court’s determination that this item should be included in income, let alone as available for distribution and as having been distributed to the shareholders.

As to the other four items comprising the sum of $63,488.47, it must be kept in mind that whatever the classification of an individual item might be for purposes of corporate accounting or computing taxable income, the tax court in determining “earnings and profits” as used in I.R.C. (1939) Section 115, had only to determine whether the item was corporate income available as surplus for dividends. In doing so the tax court was not bound to apply the same standard as required by corporate accounting practice, Commissioner of Internal Revenue v. Phipps, 1949, 336 U.S. 410, 420 , 69 S.Ct. 616 , 93 L.Ed. 771 , or as it would in determining taxable income, Commissioner of Internal Revenue v. Wheeler, 1945, 324 U.S. 542, 546 , 65 S.Ct. 799 , 89 L.Ed. 1166 ; Holmquist v. Blair, 8 Cir., 1929, 35 F.2d 10, 13-14 ; 1 Mertens, Law of Federal Income Taxation, Section 9.28. Therefore, income as shown on the tax return and the four remaining items of the sum of $63,488.47, i.e. the $6,000 credit from Truman Johnson which was not recorded in the corporate books; the deduction of merchandise purchases amounting to $2,914.42, which was disallowed for lack of substantiation; the deduction of $3,703.50 in bad debts, which was disallowed for lack of substantiation; and $49,210.15 in deferred income, explained in the corporate income tax return for 1947, which was introduced in evidence, all constitute items of income or its equivalent for purposes of computing the statutory “earnings and profits,” against which any withdrawals, formal or informal, taxable as dividends can be charged even if not considered income by standards of corporate accounting or for purposes of computing corporate taxable income.

The mere fact, however, that these items are includable in “earnings and profits” available for distribution does not mean that they have been actually distributed. Contrary to the Commissioner’s brief and the tax court’s opinion, this is not a case of constructive dividends where a shareholder is charged with receipt of a dividend even though he has not actually received the distribution. To constitute a constructive dividend it must be shown, at least, that the income, though not received by the shareholder, was nevertheless unqualifiedly subject to his demand and withdrawal and set aside by the corporation for that purpose. I.R.C. (1939) Section 42(a), 26 U.S.C.; Treasury Regulations 118, Section 29.42-2 ; Avery v. Commissioner of Internal Revenue, 1934, 292 U.S. 210, 215 , 54 S.Ct. 674 , 78 L.Ed. 1216 ; Lawrence v. Commissioner, supra, 143 F.2d at pages 458-459; George M. Cox v. Commissioner of Internal Revenue, 5 Cir., 1942, 128 F.2d 957, 958, 959 . There is no presumption of correctness in favor of the tax court’s determination that the sum of $30,632.10 and the sum of $63,488.47 were actually distributed to Clark since they were not included in the computation of the Commissioner’s deficiency notices. Therefore, the tax court’s factual determination that they constituted constructive dividends must be supported by credible evidence. The record in no way reveals that these funds were unqualifiedly subject to the demand of Clark and that they had been set aside by the corporation awaiting his demand. The record also fails to reveal any other evidence which would substantiate the tax court’s treatment of these items as additional distributions to shareholders over and above those we have herein-before discussed under sub-headings entitled “Substituted Items,” “Notes Receivable — Officers,” and “Other Distributions”. The items discussed here under subheading “Other Available Items” may well explain the availability of income making the distributions possible, but such “available items” cannot be used as evidence to increase the total amount of the distributions received by the shareholders. We hold that in its determination of the deficiencies due, the tax court erred in concluding these “available items” had been distributed.

We turn now to a consideration of the corporation’s financial picture for fiscal 1948. The tax court found that the corporation had failed to report $87,681.62 of income for that period, consisting of the receipts of unreported contract sales concealed by substituted checks, and revealed, as in fiscal 1947, by the agent’s use of the “bank deposit method”; and disallowance of certain business expenses on the grounds they were actually personal expenses of the shareholders. On appeal petitioners contest only the inclusion of the substituted check items on the same grounds they contested inclusion of similar items in the fiscal 1947 income; For the same reasons as we held the similar items includable for fiscal 1947, we hold the tax court was fully justified in including them for fiscal 1948.

In computing the earnings and profits available for distribution, the tax court also included the $16,726.40 shown as income in the corporate fiscal 1948 federal income tax return, an item which the revenue agent had considered as not available for distribution. In treating this item in the same manner as it did the similar item contained in the return for fiscal 1947, the tax court’s determination that this item was available for distribution was proper.

The Commissioner’s contention that another item consisting of deferred sales amounting to $6,381.15 was likewise available for distribution but was inadvertently omitted by the tax court is without merit. He seems to have overlooked that this item, unlike the figure contained in the tax return, was included in the unreported sales figure as computed by the tax court and treated as available for distribution.

We find no error in the tax court’s computation of actual distribution to shareholders for fiscal 1948. Unlike its treatment of items reflecting deferred income and income per return for fiscal 1947, the tax court found for 1948 that “While it was appropriate for us to reflect these items in determining earnings available for distribution as ordinary dividends, we cannot use them to increase actual distributions as reflected in respondent’s final determination.” As we have discussed above, in our consideration of fiscal 1947, this was the proper treatment for such items.

In computing the fiscal 1947 and fiscal 1948 “earnings and profits” the tax court properly reduced the amount by the taxes and additions to tax under I.R. C. (1939) Sec. 293(b), 26 U.S.C., accrued at the end of the respective taxable years. Estate of Stein v. Commissioner, 1956, 25 T.C. 940, 965-966 ; Drybrough v. Commissioner of Internal Revenue, 6 Cir., 1956, 238 F.2d 735, 740 , affirming 23 T.C. 1105 . However, the earnings and profits should be further reduced by the interest accrued on these taxes and additions to taxes at the end of the taxable year which the tax court in its computation apparently did not do. Sydney Stark, 1957, 29 T.C. 122 .

We turn now to a consideration of the proper allocation of distributed earnings, first, as between the years in question, and second, as between the shareholders themselves.

The revenue agent, after computing the total amount he considered as actually distributed to the shareholders as a group, allocated 84.259 per cent of the distributions for the fiscal year ending April 30, 1947 as income to shareholders in calendar 1946, and 15.741 per cent as income in 1947. The tax court adopted the agent’s allocation for fiscal 1947, which had been reflected in the Commissioner’s deficiency assessments. The burden was upon the petitioners to show the allocation was arbitrary or erroneous although they were not required to show what the allocation should actually have been. Helvering v. Taylor, supra, 293 U.S. at page 515 , 55 S.Ct. at page 290 ; Greenwood v. Commissioner of Internal Revenue, 9 Cir., 1943, 134 F.2d 915, 919 . However, this burden can be discharged with evidence from which it could be found that the deficiency was incorrect. With such evidence the presumption is no longer available and the Commissioner then has the burden of proving the allocation. Cohen v. Commissioner of Internal Revenue, supra, 266 F.2d at page 11 . If the Commissioner’s allocation is arbitrary, he has the further burden of showing a justification for his use of an arbitrary allocation and the methods employed in his computation. The tax court’s ultimate determination in such case must, of course, rest on a weighing of all the evidence. Gersten v. Commissioner, supra; Niederkrome v. Commissioner, supra; Lawrence v. Commissioner, supra; Hemphill Schools, Inc. v. Commissioner, supra. The tax court based its adoption of fiscal 1947 allocation primarily on the failure of the petitioners to produce records to show actually when the distributions were made. It is true that when no such records are produced, the Commissioner and the tax court can base their determination on such evidence and with such methods as the circumstances afford. Union Stock Farms v. Commissioner, supra; Baumgardner v. Commissioner, supra, 251 F.2d at pages 313-314; Roberts v. Commissioner, 9 Cir., 1948, 176 F.2d 221, 226 , 10 A.L.R. 2d 186 ; Cohan v. Commissioner, supra, 39 F.2d at pages 544, 545. But when reliable records are available they must be used in making the determination to the extent they reveal the necessary information. Here the Commissioner and the tax court chose to completely ignore the fact that the records which supported their determinations of unreported income also in many instances revealed the exact day on which unreported sales were made, when personal expenses of shareholders were paid by the corporation, when unreported sales were completed, and when substituted deposits were made. There is no indication that the agent or the tax court took these factors into consideration in allocating distributions from the fiscal years of the corporation to the calendar year of the taxpayers, even though the agent’s report compiles much of the information from the records which were apparently available at the time of the trial. True, the report was not introduced in evidence for the purpose of showing the truth of these figures. It was, however, introduced to show the basis of the Commissioner’s determination of the deficiencies. A correct analysis of the report overcomes the presumption of correctness as to the Commissioner’s allocation for fiscal 1947. The only explanation of the Commissioner’s percentage basis of allocation appears in the testimony of the revenue agent, which we set forth below.

The tax court further justified the fiscal 1947 allocation on the grounds that the petitioners, as a basis for challenging the allocation, failed to make a motion requiring the Commissioner “to file a further and better statement” under Rule 17(c)(1) of the Tax Court, Rules of Practice. But failure to request the better statement under this Rule in no way forfeits the taxpayers’ right to challenge the statement during the hearing, or gives any more evidentiary weight to the unchallenged statement than it would otherwise have had. The purpose of the rule is simply to limit the proof and scope jf inquiry at the time of trial, and prevent surprise. Gutterman Strauss Co., 1924, 1 B.T.A. 243, 245-246 . It certainly was not intended as a trap for the unwary.

We hold that the tax court was in error in relying upon and adopting the Commissioner’s percentage basis of allocation for fiscal 1947, which allocation arbitrarily ignored pertinent and relevant evidence.

From the fiscal year 1948 the revenue agent attributed 41.523 per cent as income to the shareholders in the year 1947 and 58.477 per cent to them in the calendar year 1948, a year not here under consideration. This allocation was adopted by the Commissioner and formed the basis of his deficiency notices for the taxpayers’ calendar year 1947. On the basis of the agent’s report, which showed the actual dates of distribution of many of the items during the fiscal 1948 as it had for fiscal 1947, the court would have been justified in rejecting the Commissioner’s allocation, but without making any reference to its reasons for doing so the tax court made its own allocation, treating all the earnings and profits available in fiscal 1948 as having been distributed to the shareholders in calendar 1947. The Commissioner seeks to explain the treatment by the tax court on the grounds that when the total distributions represent both distributable earnings and a partial return of capital, as the tax court found here, the distribution must first be applied against the availa ble earnings and profits for distribution as ordinary dividends. “Accordingly,” the Commissioner argues, “any reasonable basis of allocation here, would have to attribute both the entire $45,028.91 (ordinary dividends) and some additional portion of remaining liquidating dividend ($40,798.56) to calendar year 1947 were it not for the fact that Clark’s stock basis was so high that no additional portion of the $40,798.56 could here be treated as calendar 1947 ordinary income.” This might well be true but there is no presumption and the Commissioner must point to some evidence to support such an allocation. Here he deals only in the hypothetical. We find no evidence in the record to support the tax court’s allocation.

As to the allocation of distributions between the shareholders, the tax court adopted the Commissioner’s allocations which, for the years in question, merely allocated according to the shareholders then stock interest, to-wit, seventy per cent to Clark and thirty per cent to Koyl. The Commissioner’s allocations were based on the revenue agent’s report. As stated, this report was received by the tax court not as to the truth of its contents but only to show the manner of calculations and basis of the deficiency notices. The revenue agent’s report contains information as to which shareholder actually received a particular benefit as to many of the individual items distributed. No reason is given why this information was ignored or why those items which could be traced to a particular individual shareholder were not so allocated. The action of the Commissioner in this regard was arbitrary, and, therefore, no presumption can arise in support of his determinations. We hold that the tax court was clearly erroneous in adopting (he arbitrary determinations of the Commissioner. The allocations between shareholders must be redetermined by the lax court.

We will now consider the addition of the fraud penalty to the deficiency against the petitioner, Gene Clark. In his deficiency notice the Commissioner determined that a part of the deficiency of Gene Clark was the result of fraud with an intent to evade taxes due, and under I.R.C. (1939) Sec. 293(b) added 50 per cent of the amount of the deficiency as a penalty. As to the petitioner, Faye Clark, the Commissioner concluded she was personally not fraudulent in the preparation of her returns.

No presumption of correctness attaches to the penalty as it does to the deficiency. The burden is on the Commissioner to establish by clear and convincing evidence that at least some part of the deficiency was due to fraud. A mere preponderance of the evidence will not be sufficient to discharge the Commissioner’s burden. I.R.C. (1939) Sec. 1112; Baumgardner, supra, 322 ; Kurnick v. Commissioner, 6 Cir., 1956, 232 F.2d 678, 681 ; Drieborg v. Commissioner, 6 Cir., 1955, 225 F.2d 216, 218 ; Wiseley v. Commissioner, 6 Cir., 1950, 185 F.2d 263, 266 . Whether the understatement of income in the return is due to fraud presents a question of fact, and when there is substantial evidence to support the conclusion of the tax court its judgment must be accepted. Helvering v. Kehoe, 1940, 309 U.S. 277, 279 , 60 S.Ct. 549 , 84 L.Ed. 751 . In this case, following established precedent, the tax court did not limit itself to the Commissioner’s affirmative evidence, but instead considered, as it was entitled to do, the whole record in determining the existence of fraud. Frank Imburgia, 1954, 22 T.C. 1002, 1014 ; Wallace Petit, 1948, 10 T.C. 1253 , 1258; 10 Mertens, Law of Federal Income Taxation, Section 55.18.

In support of the tax court’s finding that Clark had the requisite fraudulent intent to evade taxes, there is evidence that substantial amounts of receipts from sales made by the corporation were neither recorded on its books nor reported in its income tax returns; that Clark, the majority stockholder, by his own testimony maintained control over the books and records during the years in question; that it was his own idea not to record certain sales as a means of withholding cash from the corporation; that many of the unreported over-ceiling purchases and sales of materials were not accounted for or reported in the income tax returns of the corporation or the shareholders; that at different times he had all the cash received from cash sales turned over to him; that on much of the materials sold no records were kept; and that no inventory or sales records were available from the transfer of the assets of the plumbing company to the corporation. Without the benefit of records to prove his statements, Clark sought to explain these transactions with the testimony that they were for the benefit of the corporation and that the cash which he admittedly received was for the purchase of plumbing materials at over-ceiling prices. The tax court was well within its discretion as the trier of facts in refusing to believe his testimony. United States v. United States Gypsum Co., supra, 333 U.S. 394 -395, 68 S.Ct. 541 -542, 92 L.Ed. 746 ; Baumgardner v. Commissioner of Internal Revenue, supra, 251 F.2d at page 313. All of these practices were followed by Clark during both the calendar years 1946 and 1947, for which the fraud penalties were assessed.

We cannot agree with the petitioners that the tax court’s inference of fraud is based on mere suspicion. We hold the tax court’s finding of fraud was justified under the record before it.

Finally, we consider the questions regarding the statutes of limitation as they apply in this case. 'With respect to Gene Clark, our holding that the tax court was justified in concluding that his returns for 1946 and 1947 were false and fraudulent with intent to evade tax brings them within the provisions of I.R.C. (1939) section 276(a) which provides that the Commissioner may at “any time” seek the collection of taxes evaded through fraud. As to the returns of Faye Clark, which the Commissioner does not allege to have been based on fraud, the normal statutory period for collection of the deficiencies would expire three years after the returns were filed, unless the Commissioner has met his bur-dent of proof to sustain his affirmative allegation that the deficiency against Faye was in excess of 25 per cent of the gross income stated in her returns, in which case the Commissioner would have five years after the returns were filed to enforce collection. Faye’s return for 1946 was filed on March 15, 1947. The Commissioner’s statutory notice of deficiency was mailed to her in February 1953, more than five years later. There fore, as the Commissioner admits, he is barred from any action as to 1946. As to 1947, Faye’s return was filed on March 15, 1948, with the notice of deficiency transmitted to her on February 20, 1953, more than three, but less than five years after the return was filed. Whether the applicable limitation statute is three or five years will depend upon the redetermination which the tax court will make on the remand of this cause in the light of the views expressed herein.

This cause is remanded to the tax court for further proceedings, not inconsistent with the views expressed in this opinion, which further proceedings shall include an opportunity for either party to offer additional evidence on the issues which must be redetermined by the tax court as herein stated and for redetermination of any deficiencies due or penalty thereon from either petitioner.

. T.C.Memo. 1957-129; 16 T.C.M. 555 ; 57 P.H.Memo.T.C. Par. 57,129. . “§ 'Ll5. Distributions by corporations. “(a) Definition of dividend. The term “dividend'’ when used in this chapter (ex-cejrt in Section 201(c) (5), section 204 (c) (11) and section 207(a) (2) and (b) (3) (where the reference is to dividends of insurance companies paid to policyholders) ) means any distribution made by a corporation to its shareholders, whether in money or in other property, (1) out of its earnings or profits accumulated after February 28, 1913, or (2) out of the earnings or profits of the taxable year (computed as of the close of the taxable year without diminution by reason of any distributions made during the taxable year), without regard to the amount of the earnings and profits at the timé the distribution was made. * * “(b) Source of distributions. For the purposes of this chapter every distribution is made out of earnings or profits to the extent thereof, and from the most recently accumulated earnings or profits. * * * ” 53 Stat. 46 as Amended June 29, 1939. . Tile corporation’s explanation of this item: . Sec. 39-42-2 Income not reduced to possession Income which is credited to the account of or set apart for a taxpayer and which may be drawn upon by him at any time is subject to tax for the year during which so credited or set apart, although not then actually reduced to possession. To> constitute receipt in such a case the income must be credited or set apart to the taxpayer without any substantial limitation or restriction as to the time or manner of payment or condition upon which payment is to be made, and must be made available to him so that it may be drawn at any time, and its receipt brought within his own control and disposition. A book entry, if made, should indicate an absolute transfer from one account to another. If a corporation contingently credits its employees with bonus stock, but the stock is not available to such employees until some future date, the mere crediting on the books of the corporation does not constitute receipt. See also See. 39.42-3, Examples of constructive receipt. . The revenue agent when questioned concerning the allocation which was the one adopted by the Tax Court was unable to give an adequate explanation of why he adopted the particular allocation. He testified as follows: “Q. You have that entitled Total Distributions Per EAR, $74,984.96. (Exhibit Q, allocation of income between years). Then, reading over to the right under Gene Clark, you have the years 1946 and 1947, and you, apparently, have broken up the amounts of $74,984.96 into two amounts and you have allocated one amount to 1946 and another to 1947. Will you tell the Court now how you arrived at that allocation for each year? A. As far as I can remember at this time, it was based upon the stock ownership of 30-70, and for each year on the amounts available to be applied as constructive dividends. “Q. Well, how did you know what the amount available for 1946 was? How •did you arrive at the figure of $44,237.13 ? A. I don’t seem to have the computation here of how that is done, sir, but this point has been belabored so much that all I can say is that I did it on a stock ownership basis and I do not recall at this time. “Q. Well, I realize that you attributed 70 per cent of the profits to Mr. Clark and 30 per cent to Mr. Koyl. That has been already gone into, but I am asking you, this first year, you realize, do you mot, that the fiscal year ended April 30, 1947? A. I do, sir. “Q. Well, how could you have had any ¿mount available for distribution in 1946 when the taxpayers were on a calendar year basis? A. Well, if this had been a partnership, I would believe that •when the period ended that this would be when I would have distributed it, but it was a corporation, and the moneys were taken outside the records of the corporation and, therefore, I distributed them as they got them, on a cash basis to the individuals, as they got them or as they were made available to them. “Q. Then you can show me that $44,-227.13 was either received or made available to them in 1946? A. Only that it was available to them. “Q. How? A. By the computation of the adjustments to net income in my report, plus the Kansas farms, and less the accounting adjustments to net income I made that I said were not available. “Q. We will go on now, Mr. Phillips, . Rule 17 (c) Amendment Ordered.— (1) Occasion for. — The Court upon its own motion, or upon motion of either party showing good cause filed prior to the setting of the case for trial on the merits, may order a party to file a further and better statement of the nature of his claim, of his defense, or of any matter stated in any pleading. Such a motion filed by a party shall point out the defects complained of and the details desired. . 26 U.S.C. Sec. 293 . “Additions to the tax in case of deficiency “(a) * * * “(b) Fraud. If any part of any deficiency is due to fraud with intent to evade tax, then 50 per centum of the total amount of the deficiency (in addition to such deficiency) shall be so assessed, collected and paid, in lieu of the 50 per centum addition to the tax provided in section 3612(d) (2). 53 Stat. 88 .” . 26 U.S.C. Sec. 1112 . “Burden of proof in fraud eases. “In any proceeding involving the issue whether the petitioner has been guilty of fraud with intent to evade tax, the burden of proof in respect of such issue shall be upon the Commissioner.” 53 Stat. 160 . . 26 U.S.C. § 276 . “(a) False return or no return. In the case of a false or fraudulent return with intent to evade tax or of a failure to file a return the tax may be assessed, or a proceeding in court for the collection of such tax may be begun without assessment, at any time.” 53 Stat. 87 . . 26 U.S.C. § 275 . “Period of limitation upon assessment and collection “Except as provided in section 276— “(a) General rule. The amount of income taxes imposed by this chapter shall be assessed within three years after the return was filed, and no proceeding in court without assessment for the collection of such taxes shall be begun after the expiration of such period. “(c) Omission from gross income. If the taxpayer omits from gross income an amount properly includible therein which is in excess of 25 per centum of the amount of gross income stated in the return, the tax may be assessed, or a proceeding in court for the collection of such tax may be begun without assessment, at any time within 5 years after the return was filed.” 53 Stat. 86 .