This appeal from a Tax Court decision for the government raises the question whether one of several financial transactions between a parent corporation and a wholly owned subsidiary, spanning the years 1960 to 1962, constituted a taxable dividend to the parent. The facts, while clear and largely undisputed, present the problem of determining under which of three transactional shells is lodged the taxable pea.
The taxpayer (hereinafter referred to as Ogden) is a Rhode Island corporation which was organized in 1960 by the four Salmanson brothers for the purpose of acquiring the stock of the National Ring Traveler Company (Ring). The total purchase price of the Ring stock was $746,820, or $174. per share. Ogden having had only $20,000 in capital, financing arrangements were necessary in order to complete the Ring purchase. Ogden borrowed $495,000 from the Rhode Island Hospital Trust Company and additional amounts from two corporations owned by the Salmanson brothers.
After the acquisition of its stock by Ogden, Ring advanced to Ogden the sum of $495,000 which was used to pay off the Hospital Trust loan. Thereafter, Ring advanced an additional $120,000 to Ogden on open account. As of December 31, 1960, Ring’s balance sheet showed $615,000 as due from Ogden, and Ogden carried the $615,000 advance on its balance sheet as a liability owing to Ring. As evidence of the alleged liability Ogden issued its demand promissory note to Ring.
On November 27, 1961, Ogden borrowed $615,000 from Hospital Trust and used the loan proceeds to discharge its debt to Ring. Ring then used the $615,000 to purchase short-term Treasury bills which were pledged to Hospital Trust as security for the Trust Company loan to Ogden.1
On January 4, 1962, on instructions from Ring the Trust Company sold the Treasury bills and applied the proceeds to discharge Ogden’s loan obligation. The interest due on the loan was paid by Ring.2 On the same day Odgen delivered its unsecured, non-interest-bearing, demand note in the amount of $615,000 to Ring.
The Commissioner assessed a deficiency against Ogden for $271,032.80 in 1960, $282, 607.60 in 1961, and $282,586.-37 in 1962.3 These assessments were alternative, so that ultimately the taxpayer would be liable for a deficiency in one year only.
The basis for the assessments is the Commissioner’s contention that the $615,000 advanced from Ring to Ogden was a dividend rather than a loan. In the Commissioner’s view there was a dividend either: (1) in 1960 when Ring *225advanced $615,000 to Ogden; (2) in 1961 when Ring pledged the $615,000 in Treasury bills as security for the Trust Company loan to Ogden; or (3) in 1962 when the Treasury bills were sold by the Trust Company in discharge of Ogden’s debt.
The tax court held that a dividend occurred in 1962 when the Treasury bills were sold. The taxpayer makes two arguments. The first is that the entire series of transactions were mere mutations of the same bona fide debt. As to this contention, taxpayer cannot overcome the tax court’s finding that the 1962 transaction was not a loan in view of the very substantial obstacle of the clearly erroneous rule. Rule 52, Fed.R.Civ.P.; Estate of Broadhead v. Commissioner of Internal Revenue, 391 F.2d 841 (5th Cir.1968); Biritz Construction Co. v. Commissioner of Internal Revenue, 387 F.2d 451 (8th Cir. 1967); Goldman v. Commissioner of Internal Revenue, 388 F.2d 476 (6th Cir. 1967); Lamont v. Commissioner of Internal Revenue, 339 F.2d 377 (2d Cir. 1964).4
The taxpayer’s second argument is a closely reasoned one. It contends that if there was no loan in 1962, there was no loan in 1960 — and it refers to several references in the tax court’s opinion to the transactions in 1960 and 1961 as dividends. But, says taxpayer, if the 1960 advance of $615,000 was a dividend, it more than used up the $315,889 of earned surplus then shown on the balance sheet. And, since a stipulation by the parties referred to the balance sheets as “purporting to reflect [Ring’s] financial condition”, this figure represented all earnings and profits whenever accumulated. The 1962 distribution, therefore, could not possibly have been made out of earnings and profits accumulated after February 28, 1913 or current earnings, as required by § 316 of the Internal Revenue Code of 1954.5
The flaw in taxpayer’s logic lies in refusing to recognize the options available to the government. That is, taxpayer assumes that because the government asserted, as one of its three mutually exclusive alternatives, a tax liability for 1960, based on the advance made in that year, it should be bound by its theory that a dividend had been paid in that year, exhausting all the earned surplus. Had there been no transactions subsequent to 1960 and no evidence bearing on any changed intent or ability to repay the advance, the government might well have been confined to the theory of a dividend as occurring solely in 1960. Subsequent transactions, however, did occur — a purchase, pledge, and sale of securities, and a new note — as well as sufficient change in circumstances to justify the tax court in saying that *226while there were plans in 1960, admittedly nebulous, to secure funds through merger for the repayment of the advances, they had apparently been abandoned by 1962. Under these circumstances, the government cannot be forced into the straight] acket urged by taxpayer. While it could pursue the tax consequences of 1960 on the theory that a dividend had been paid (under the perhaps critical disadvantage of having the burden of proof, see n. 5, supra) it could also pursue the tax consequences of 1962 (under the perhaps critical advantage of not having the burden of proof).
Admittedly, some language in the tax court’s opinion has given comfort to taxpayer. The court does state at one point that “[h]aving determined that the advances were dividends and not bona fide loans, it is still necessary to determine in which year the distributions are taxable as dividends to Ogden.” This is consistent with taxpayer’s theory that a 1960 dividend would, so to speak, preempt the field. But the statement was preceded by the conclusion that “We think it is quite clear from all the evidence that the series of transactions * * * resulted in a distribution by Ring to Ogden substantially equivalent to a dividend * * And the opinion concluded with the observation, as we have noted, that while ability to repay in 1960 was doubtful, all doubt was removed by 1962. The presence of such ambiguity as there was is not enough to command reversal.
We would add one final observation. Even were there a clear and independent finding that the 1960 advance was a dividend, thus exhausting “earned surplus” for the subsequent years, taxpayer would still not have carried its burden of demonstrating that the 1962 distribution had not been made out of earnings and profits. Even if the stipulation “purporting to reflect [Ring’s] financial condition” were to be taken as prima fa-cie proof of the amount of earned surplus in the respective years, this is not necessarily to be equated with “earnings and profits”. Bittker and Eustace, Federal Income Taxation of Corporations and Shareholders, §§ 5.03, 5.04 (2d ed. 1966).
Since, therefore, we cannot accept the stipulation as constituting prima facie proof that the earned surplus figure of Ring for 1962 — substantially the same as in 1960 — constituted all the earnings and profits since 1913, it follows that, even if the 1960 transaction had been a dividend rather than a loan, this would not demonstrate that such earnings and profits had been exhausted.
We hold, therefore, that the taxpayer did not carry its burden of proof for the 1962 distribution.
Affirmed.