LEONARD P. MOORE, Circuit Judge.
This action was brought by the United States of America (referred to as the “government”) against a group of oil companies to recoup from them the sum of $66,021,257.77 expended by the government pursuant to the provisions of the Economic Cooperation Act of 1948, 22 U.S.C.A. § 1501 et seq., through the Economic Cooperation Administration and the Mutual Security Agency (referred to as “ECA”), to finance the purchase of crude oil produced in the Middle East and sold by defendants Caltex Oceanic, Ltd. (referred to as “Oceanic”) and Mid-East Crude Sales Company (referred to as “Mid-East”) (sometimes collectively referred to as “Caltex”), to various importers in foreign countries. In the alternative the government asks for a lesser amount, $16,472,178.51, the difference between the amount paid and the amount the government claims should have been paid under its interpretation of the law and the facts. The district judge received a vast amount of proof from both sides which he carefully analyzed in a lengthy opinion (D.C., 155 F.Supp. 121). He concluded that the defendants were entitled to judgment on the merits and dismissed the complaint. The government appeals.
[52]*52sponsible for, these errors, was the court’s, failure, so argues the government, to hold that certain particular sales and transactions selected by the government were determinative of noncompliance with these requirements.
Defendants answer by asserting that the “lowest competitive market prices” provision was merely an expression of policy and that even if it be construed as a substantive rule, all of the sales in question were made in full compliance with its terms.3
Shortly after the termination of World War II the government decided to extend financial aid to certain foreign countries to enable them to purchase various commodities and to assist them in their economic recovery. The plan was popularly known as The Marshall Plan. The commodity here involved was Saudi Arabian crude oil. Under the Plan a buyer would apply to his government for authorization to make a purchase, would negotiate with the seller for the best price he could obtain, and would deposit the purchase price in his own currency in a “counterpart fund.” The seller would be paid in ECA dollars loaned or granted by our government to the participating country.
Originally Congress enacted a statute to the effect that sales should not be financed at prices higher than the market price prevailing in the United States adjusted for transportation costs to destination, quality and terms of payment (22 U.S.C.A. § 1510(l), repealed August 26, 1954). Suppliers were required to sign certificates that the price was within this statutory limit. When ECA Regulation 1 was amended to state rules, such as the “comparable sales” test (note 2, supra), which were stated as fixing the point beyond which purchases would not be eligible for reimbursement by ECA, the Supplier's Certificate was amended to state that the price charged was not in excess of that allowed by the applicable price provisions of the Regulation.
The question is: Were defendants’ prices in compliance with the terms of the Regulation? Payments having been made by ECA, the government now claims that its post-audit discloses noncompliance. As required by the amended Regulation, suppliers’ certificates certifying that the purchase prices were no higher than the permitted prices were submitted by the selling companies, Oceanic and Mid-East.
Regulation 1
Searching for some one or more theories of liability which may prove to be legally sufficient and factually supported, the government has industriously and imaginatively pursued every possibility and clue. The first complaint alleged: (A) that defendants charged prices in excess of: (1) prevailing market prices in the United States; (2) comparable export prices; and (3) prices prevailing in the Middle East; (B) that defendants charged lower prices on non-EC A sales; and (C) that the ECA payments had been made by mistake. By amendment the government asserted that because of excessive prices based on the comparable sales theory the suppliers’ certificates were false.
The second amended complaint narrowed the issues. The allegations of fraud and deceit were abandoned and damages only from September 1, 1950 to September 1, 1952 were sought. In support of its “lowest competitive market price” theory the government advanced three arguments: (a) that the prices charged could not exceed the lowest price calculated according to its formula; (b) that a special rule required such a price; and (c) that there was a specific agreement between the parties to charge such a price. The defendants point to these vacillations in theory as a sign of weakness and claim that the government seeks to capitalize on hindsight. This is fair comment but were the government, even after many years of pretrial preparation, able to discover a clear violation of law its earlier misconceptions should not becloud the merits if at last it were able to hit upon a sound theory. After all, [53]*53the defendants required many years of exploration before oil in commercially profitable quantities was found.
Regulation 1 is the keystone of the government’s case. The words are reasonably clear but the parties give them widely divergent interpretations. Thus the Regulation states that “it is the policy of ECA to make payment only for purchases of commodities * * at prices that approximate, as nearly as practicable, lowest competitive market prices.” Defendants stress the word “policy” and would limit its meaning to a broad general statement whereas the government interprets it as a binding declaration of price. The next paragraph of the Regulation refers to “rules set forth in this section” which “are intended as a guide to buyers and sellers in conducting their negotiations.” Policy having been stated, normal procedure would call for immediate reference to specific “rules” but even the “rules” are only to be a guide to buyers and sellers. ECA, as the government properly asserts, was not a price-fixing agency. Yet these “rules” assume more than policy importance because they “fix the point beyond which purchases will not be eligible for reimbursement by ECA.” The next sentence restates this same declaration in the affirmative, i. e., “Compliance with them will make a purchase eligible for financing, * * * ” ECA was vested with the duty “to determine whether there has been compliance” by making its own “post-audit.” Thus far the procedure seems definite enough, “rules,” “compliance” and ECA “post-audit” to determine compliance. But what happens if there be violations? Again the Regulation speaks. “If it appears that the objective of lowest competitive market prices is not being met, ECA will take appropriate action to impose additional limitations.” The procedural cycle is thus complete. Sanctions are to be imposed for violations and ECA is charged with the responsibility of taking “appropriate action.”
The government relies upon several theories. First, it would make the policy statement in the Regulation a specific price rule. Second, it would fix the lowest competitive market price in accordance with the government’s hypothetical formula for calculating an assumed f. o. b. price in the Middle East by converting a c. i. f. price (Canada) into an f. o. b. price (Middle East) or, in the alternative, using the “offtake” price of $1.43. Third, an agreement between the parties to charge no more than a price derived from the alleged formula.
Although there is sharp dispute as to whether Section 201.22 of Regulation I is a price setting rule or merely an expression of policy, it is not necessary to resolve this question. Even were it assumed that the section is a price fixing rule, the government has failed to show that it was violated by the defendants.
The phrase “approximate, as nearly as practicable, lowest competitive market prices” has little meaning except in relation to some specific fact situation. The word “competitive” immediately raises the question “with whom” and “where”; the word “market,” the question “in what commodity” and “where.” Even then the words are so broad and general that interpretation must be sought. In a single business transaction interpretation by the acts and words of the parties frequently may prove to be unsatisfactory in arriving at their mutual understanding of words but if these transactions in relation to the same words are continued over a period of years with a high degree of uniformity of action such mutual interpretation usually attains a satisfactory degree of accuracy. This principle is singularly applicable here as the facts reveal. These facts have been so thoroughly mar-shalled in the district court’s opinion that it will only be necessary to sketch a background against which the errors alleged can be outlined. Particular emphasis, however, must be placed upon the interpretive acts of the parties themselves.
A brief review of the world oil situation discloses the fact that only after World War II did the Middle East join the United States and Venezuela as a [54]*54major producer of crude oil. Various American oil companies had acquired substantial interests in this area. Of this group, the two principal companies named as defendants were Standard Oil Company of California (Socal) and The Texas Company (Texas — sometimes referred to as “Texaco”). Socal and Texas each owned 50% of the stock of the defendants Oceanic and Mid-East (the selling companies). Each also owned a 30% interest in the producing company, Arabian American Oil Co. (Aramco) and a pipeline company, Trans-Arabian Pipe Line Co. (Tapline). The remaining 40% of Aramco and Tapline was owned 30% by Standard Oil Company (New Jersey) and 10% by Socony-Mobil Oil Co. Inc. (Socony). Aramco’s production was shared in proportion to the stock interest, Oceanic (by agreement of Socal and Texas) 60%, Standard Oil (New Jersey) 30% and Socony 10%. Because Aramco was wholly owned by these four companies an arbitrary price was assigned to the oil to which they were entitled of $1.43 per barrel, a figure calculated to finance Aramco’s development and yet permit competition with other Middle East oil. The terms “offtake agreement” and “offtake price” were used in connection with this arrangement. “Offtakers” were the companies entitled to this participation or their nominees.
This “offtake” price ($1.43) was in no sense a competitive or market price. The government in its first complaint had claimed this figure as a market price but abandoned this position when the second amended complaint (on which the case was tried) was filed. Thus Oceanic was the designated “offtaker” of Socal and Texas and it, in turn, nominated Mid-East, Caltex (U. K.) or occasionally other subsidiaries of Socal or Texas to participate in its share. Caltex (Oceanic and Mid-East) sold only in the Eastern Hemisphere and made no sales in the Western.
The two ports of delivery by pipeline were Ras Tanura on the Persian Gulf and, by a long pipeline across the desert, Sidon on the Mediterranean. When the United States was the principal producer of oil shipped from the Gulf Coast prices were determined by reference to Gulf Coast prices. Venezuelan oil then entered the competitive market. While Western Hemisphere oil was still moving to European ports a price for Aramco crude was fixed by taking the delivered price of oil from Jusepin, Venezuela at North European ports of $3.52 per barrel. After deducting the freight rate ($1.45) from Ras Tanura to this port and $.04 for refining value differential an f. o. b. rate of $2.03 per barrel at Ras Tanura was obtained. This was in the Spring of 1948.
As Middle East oil became more plentiful it was shipped in larger quantities to the European market. These deliveries, in turn, had a marked effect on the price structure which no longer was dependent upon the factors which had previously been regarded as determinative. Because of competition from Gulf Oil Corporation in April 1949 Caltex reduced its price from $2.03 to $1.88 and in September 1949 to meet the same competition the price was reduced to $1.75. The price remained at this figure until' July 1953 when it increased it to $1.97. The pipeline to Sidon differential of 66(; is not in dispute. Thus the delivered prices throughout the period in controversy were $1.75 at Ras Tanura and $2.41 at Sidon. Attention must therefore be concentrated upon these prices in relation to their compliance with the terms of Regulation 1.
The government concedes that “The Administrator of ECA * * * advised defendant in writing on or about August 22, 1950, in effect, that the practice of' the defendant, in then charging $1.75 per barrel, f. o. b. Ras Tanura, for Middle East crude oil, ECA financed, conformed to the requirement of Regulation 1” (Sec. Am’d Compl., par. 20). No complaint is made as to lack of compliance prior to September 1, 1950. The-government specifically restricts its claim to the period from September 1, 1950. After September 1, 1950, alleges the government, “freight rates for the trans[55]*55portation of crude oil from Ras Tanura to the Atlantic seaboard of the United States substantially increased over the price of $1.10 per barrel, which $1.10 charge, as above alleged, had been used to determine the competitive price of Middle East crude oil when the $1.75 per barrel, f. o. b. Ras Tanura, was approved by EC A and when such rate was current” (Sec. Am’d Compl., par. 22). Thus the government by the assumption of certain freight rates from the Middle East to the Atlantic seaboard would fix “the competitive market price of Middle East crude oil, determined by subtracting transport costs (and tariff as applicable) from said delivered price.” Basing its claim upon such calculations it asserts that the price should have been “substantially less than $1.75 per barrel, f. o. b. Ras Tanura, paid defendant by ECA” (Sec. Am’d Compl., par. 22).
Proceeding on this assumption the government bases virtually its entire case upon certain intracompany transactions and exchanges and upon a few c. i. f. sales in Canada and in Portland, Maine made by Texaco. These c. i. f. contracts it claims by means of certain assumed freight charges which it deducts from the c. i. f. price can be turned into a competitive f. o. b. market price lower than the $1.75 and the $2.41 charged respectively at Ras Tanura and Sidon.
The sales upon which the government erects its argument require some analysis. In October 1949 Texaco contracted to sell 2,000,000 barrels of Aramco crude to the British-American Oil Company, a Canadian corporation, delivery to be during the calendar year 1950 at Montreal or Portland, Maine. The price was $2.65 at Montreal and $2.60 at Portland. On September 27,1950 a second contract was made between the same parties, the price being $2.75 Montreal and $2.70 Portland for delivery during 1951. Pursuant to these contracts 3,733,819 barrels of Aramco crude were shipped between September 2,1950 and December 1, 1951.
Taking the cost of ocean freight from Ras Tanura to Portland or Montreal between a range of $1.097 per barrel and $2.197 per barrel and from Sidon to Portland or Montreal as between $1.00 and $2.419 per barrel the government would calculate an f. o. b. price at Ras Tanura of a low of 80 per barrel to a high of $1.55 per barrel. From Sidon the corresponding prices would have been 280 per barrel to $1.71 per barrel. Against these fictitious prices, artificially calculated, were the actual prices set in the competitive market of the Eastern Hemisphere of $1.75 and $2.41. The government does not state whether it claims that 80 a barrel f. o. b. Ras Tanura would be a realistic price. In contrast with the volume of oil being sold under competitive conditions in the Eastern Hemisphere by some of the world’s largest oil companies, it would be highly unreasonable to accept a few c. i. f. sales in Canada as fixing either the locale of the “market” or the “lowest competitive price” in the Eastern Hemisphere.
The next Canadian sale relied upon was by Texaco to McColl-Frontenac Oil Co. Ltd. Between December 1950 and December 1951 Texas sold 7,989,854 barrels of Aramco crude at $2.75 per barrel, c. i. f. Montreal and $2.70 or $2.67 per barrel c. i. f. Portland. Between December 18, 1951 and August 31, 1952, 5,580,-951 barrels were sold at $2.85 c. i f. Montreal and $2.80 or $2.77 c. i. f. Portland. Using the same assumed freight rates the government obtains a range on the first contract of from $1.08 per barrel at Ras Tanura to a high of $1.91 per barrel at Sidon. Under the second contract corresponding prices would have been 740 to $2.01 per barrel, both at Sidon.
The fallacy of the government’s theory is that although in April 1948 freight rates had been applied to Western Hemisphere prices to determine the price of Eastern Hemisphere crude at a time when Western Hemisphere crude was moving to the Eastern Hemisphere, the expanded production in the East had reduced the price of Middle East crude to less than that of Western competition. Therefore the price of Middle East crude oil in the Eastern Hemisphere was no [56]*56longer determined by the price paid for Western Hemisphere oil. During the period herein involved there were many sudden shifts in the economic and political situation. Because of international crises, for example, the Korean War, and the closing of the Abadan refinery, the demand for tankers fluctuated violently. There was no fixed or independent market from which a proper freight rate could be obtained so as to have a constant factor to insert in the formula which the government would create. The cost to Texaco of operating the tankers in which it carried the oil sold to Canada depended upon a large number of variables so that it would have been impossible to calculate an f. o. b. price at Ras Tanura from the Canadian sales even if it were assumed (which would be contrary to fact) that the Canadian sales were in a competitive market for Eastern Hemisphere oil.
The wide fluctuation in the theoretical f. o. b. prices derived by the government was due to the use of Texaco’s intracom-pany freight costs in computing a “net-back” price based on the Canadian c. i. f. contract prices. These costs varied widely depending upon whether the particular tankers were company owned or chartered, the method of accounting, i. e., operating or incremental cost, and scarcity or surplus of tankers. However, the cost to Texaco of carrying these shipments to Canada under its contracts over a two-year period was not known to the Caltex group. Nor would Texaco’s costs have been binding upon them. Socal, for example, had lower freight costs which would have resulted in a higher netback.
Fundamentally the weakness of the government’s reliance on the Canadian c. i. f. sales is that a few isolated contracts by a single seller (not Caltex) to two purchasers in Canada amounting in total volume to approximately 1.5% of the total Middle East production do not even establish a market price; much less a competitive market price. It is significant that ECA during the period involved in the suit stated (April 19, 1951) “It is not our position that ECA should go so far as to insist on the price realized by Middle East suppliers on substantial shipments to the Western Hemisphere” (Defts.' Exh. 50). Furthermore little reliance was placed upon the government’s net-back formula theory by ECA’s own economic experts.
The government’s theory would create a different price for every shipment depending upon the cost of freight. The government not only failed to prove that there was an established “market” or “market price” in Canada but advanced no possible basis for assuming that a “competitive market price” for fixing the competitive price of the vast volume of Middle East oil could be found in the three or four isolated Western Hemisphere sales by Texaco.
ECA Knowledge of Prices and Supervision
Both prior to and during the period involved in the suit ECA kept under observation the prices at which goods were moving under the program and forthwith suspended the procurement of commodities found to be overpriced. ECA was not merely a name or a passive agency. On the contrary its activities were in charge of a skilled and experienced Administrator under whom were many divisions staffed with employees assigned to special fields. Thus there was a Petroleum Branch of the Industry Division which reviewed the oil market and ECA transactions in this commodity. A Price Branch also reviewed prices and conducted post-audits. Unlike some situations in which such audits were made years later these audits were completed within four to six weeks after the transactions.
ECA assumed an active role in endeavoring to bring down the price of petroleum from the Near East. Throughout 1949 there were frequent communications and discussions between ECA and Caltex relating to the price of Middle East oil and the effect of competitive forces thereon. To assure himself of independent and disinterested expert advice the ECA Administrator appointed his own consultants (five recognized experts) to advise him in deter[57]*57mining ECA’s future course of action the case of Middle East oil prices. in
In March 1949 the Administrator advised defendants that the experts had approved the $2.03 (Ras Tanura) price but requested a further price study in view of the increasing oil supply. Competition having caused the price to drop to $1.75 even this price was examined for ECA by its experts in November 1949. At this time they were aware that oil was being transported from the Middle East to the Western Hemisphere but they were of the opinion that a Western Hemisphere price would not be the basis for a Middle East price on a net-back formula theory.
To ascertain and check prices while ECA continued to finance these transactions it had the benefit of statistics as to oil transactions from many government departments. Thus it was not unaware of either the Canadian sales or Western Hemisphere shipments generally. In 1949, 1950 and 1951 ECA was constantly considering the Middle East crude price structure. Its letter to Caltex of August 22, 1950 contains a lengthy review of its knowledge of prices charged and the competitive situation. A change in pricing policy with respect to petroleum products was made the subject of special action by ECA which promulgated Amendment 5 to Regulation 1 which fixed a price formula for such products. As a result Caltex elected to cease making ECA-financed sales in this field. With full knowledge of the effect of a net-back formula on products prices ECA’s failure to make a similar amendment as to crude oil assumes more than negative importance. It was tantamount to a ruling that the products formula was not to replace the method of pricing then current for crude. The trial court characterized the inference to be drawn from this failure to amend as to crude as “overwhelming” and so it is.
In 1951 ECA still continued to interest itself in prices. It debated within its ranks whether lower prices should be negotiated, whether the offtake price of $1.43 to subsidiaries should be regarded as a market price or whether the Regulation should be amended. On January 18, 1951 ECA demanded a refund of the price differential between $1.75 (and also $1.88 before the reduction to $1.75) and $1.43, the offtake price. On March 8, 1951 ECA and Caltex discussed this problem. The discussion included the Canadian sales and the offtake “sales.” For more than a year thereafter ECA, although it had the power to suspend or discontinue financing at prices not in conformity with the Regulation, proceeded with its financing at the $1.75 price.
The government misinterprets the legal effect of the discussions between the parties and would turn them into the side road of estoppel. Thus it states “Appel-lees argue that because ECA continued to finance Aramco crude at $1.75 and $2.41 per barrel, the United States is estopped to question whether the prices conformed to the standards set out in the Regulation” (Reply Br., p. 10). It then answers the argument by saying that the doctrine of estoppel does not apply to the United States. ECA’s actions are not significant as proving es-toppel but they are highly significant in establishing what ECA did or did not do under the very regulation which gave it the power to act. As the trial court said “Such contemporaneous construction is entitled to great weight. See Tobin v. Edward S. Wagner Co., 2 Cir., 1951, 187 F.2d 977.”
ECA itself realized that Regulation 1 needed further clarification and amendment when it promulgated Amendment 5. ECA’s letter of August 22, 1950 (Pltf.’s Exh. 27) establishes its knowledge that the controversy then existing as to price obliged it to remedy the situation by “amending ECA Regulation 1.” To limit payments to the lowest competitive price it conceded was “a clear responsibility on the part of ECA.” To this end it “decided to amend ECA Regulation 1” as to petroleum products. The amendment (Pltf.’s Exh. 27, p. 6) as published presented a definite formula which suppliers could accept or reject. The [58]*58prospective need for further action by ECA is recognized in its statement that a special ruling (September 1948) was only to be applicable “until the publication of official regulations in the Federal Register prescribing a different rule” (Pltf.’s Exh. 27, p. 7). No such rule as to crude oil was ever published. Therefore, the best interpretation of the mutual contemporaneous understanding of the parties under which the defendants made and ECA financed crude oil sales is to be found in their words and their acts. With such an understanding of the Regulation defendants have fully complied.
Comparable Sales
Secondly, the government argues that defendants’ sales were ineligible for ECA financing because the prices charged exceeded prices on “comparable” export sales. The section of Regulation 1 relied upon reads, in part (§ 201.22(e) (2)):
“A price for a purchase * * * will be approved for reimbursement if:
“(i) It does not exceed any price charged by the supplier at the time of purchase in a comparable export sale of the same or a similar commodity; * *
The government relies primarily on the Canadian sales and alleged “sales” to subsidiaries Caltex (U. K.) and California Transport.
The Canadian sales were made by Texaco whereas the ECA financed sales complained of were made by Oceanic and Mid-East. The government is forced to argue that because of the part ownership of the selling subsidiaries by Texaco, Texaco’s sales should be regarded as Oceanic and Mid-East sales and be binding upon them. It calls them nothing more than corporate agents of Texas and Socal and then would invoke the legal doctrine that responsibilities cannot be avoided by using subsidiaries which, without independent business of their own, act as agents of the parent corporations. Granting that this principle may be applied to certain factual situations, as in the cases cited by the government, it has no application whatsoever where the trial court has found (amply supported by the evidence) that the selling subsidiaries were organized for legitimate business purposes and operated autonomously. The mere fact that Texas and Socal, both large integrated and competing oil companies, each owned 50% of the stock of Oceanic and Mid-East does not change the character of Oceanic and Mid-East as independent selling companies or foreclose either Texaco or Socal from selling in its own right. The trial court found that “There is no doubt that Socal and Texas were in complete ignorance of each other’s Western Hemisphere operations, and that Caltex was in a similar position” (155 F.Supp. at p. 150). Oceanic and Mid-East could scarcely be accused of making comparable sales at lower prices when (1) they did not make the sales, and (2) they did not even know of the sales or prices.
The government’s “comparable” sales argument fails because the facts do not support it. The sellers responsible for complying with Regulation 1 were Oceanic and Mid-East. The place of the selling operation was the Middle East and the price was an f. o. b. price at the ports there. “Comparable” sales should have some reasonable relationship with the sales with which they are being compared. The trial court correctly held that the Canadian sales by Texaco at c. i. f. prices did not meet this test.
The Caltex (U. K) “sales” were not sales in fact but rather a result of the substitution of Caltex (U. K.) for Oceanic as an off-taker. In this category Cal-tex (U. K.) did not make independent purchases at $1.43 but merely participated in the original arrangement whereby off-takers took at this price. Furthermore the proof established that ECA was aware of the role which Caltex (U. K.) was playing and did not regard its purchases as “comparable sales.” The same situation existed in the case of California Transport, a Liberian subsidiary of So-cal. As it was a designated off-taker of Socal, the transactions were not “sales,” [59]*59nor was the $1.43 figure a “market price.” Upon the trial the government also characterized certain crude oil exchanges between Texas, Socal, Socony, ■Shell and Atlantic Refining as sales. Apparently recognizing the prerogatives of Texas and Socal to 60% of Aramco’s crude, the government does not now rely ■upon these exchanges.
The trial court summarized the comparable sales phase of the case by saying that “none of these transactions constitute comparable sales even if it be assumed that all defendants may be treated as one * * * ” (D.C., 155 F.Supp. at page 159). The facts support the finding that the sales were not comparable; they also require the legal conclusion that the defendants should not be treated as one.
The trial court concluded that the Supplier’s Certificate signed by the suppliers “affords no basis for this lawsuit.” Upon this appeal the government does not press a claim based upon false representations in the certificates. The court found that the suppliers when they signed “acted in complete good faith.” The evidence supports the finding.
BurcL&n of Proof
The government also advances as a point of error the trial court’s supposition that the government had the burden of proving the ineligibility of defendants’ prices for ECA financing.
The doctrine of burden of proof has two aspects, procedural and substantive. As a matter of procedure it requires the party carrying the burden to come forward with its proofs. As a matter of substantive law the term is frequently used to indicate the fact that the burdened party’s proof must outweigh the opposing proof in order to establish the claim asserted. Thus in a controversy in which the quality of the proof is in even balance the party bearing the burden fails to meet the legal duty placed upon him. However, after the trial is concluded and the facts have been so thoroughly established by all parties as to leave no doubt as to the proper factual conclusions, burden of proof, both procedurally and substantively, is no longer of importance in arriving at a decision on the merits.
The government’s statement that “The opinion of the District Court in this case leaves no doubt that its decision against the Government rested upon the assumption that the Government was charged with the burden of proving the ineligibility of the ECA-financed Aramco crude price and the further burden of proving that the appellees had been overpaid” is completely unjustified. Its own record reference (155 F.Supp. at page 145) reveals that the court was dealing only with the difficulty of determining “the precise location of the watershed” (a theoretical economic concept).
There is no indication that on any point material to the decision the trial court was influenced by, or relied upon, the burden of proof doctrine to buttress his conclusions. Without a showing of reliance thereon there can be no reversible error based upon any burden of proof theory.
Since defendants fully sustained any burden of proving the eligibility of their prices for ECA financing there is no need to analyze or distinguish the burden of proof cases cited by the government.
In conclusion the important issue to be resolved aside from all technicalities of law and fact is: Was improper advantage taken of the government by the defendants in their ECA-financed sales? If so, every effort should be made to require reimbursement. After a careful review of the record, the conduct of the parties themselves, their own approaches to the problems then facing them and their own contemporaneous interpretations of the Regulations, the conclusion is clear, as succinctly summarized by the trial court, “that the defendants’ proof showed beyond contradiction that the prices financed by ECA were in fact the lowest competitive market prices.” The fact that “Throughout the entire period ECA continued to finance at such prices, which perhaps more than anything else indicates ECA’s acknowledgment that the prices charged were in fact the lowest [60]*60competitive market prices” gives additional assurance that the payments made were proper. The constant activity of EC A, aided by its own independent experts in keeping informed of the various crude oil prices throughout the world was the government’s best guaranty against any overreaching, intentional or inadvertent. The proof here presented reveals that ECA performed its function well.
The judgment is affirmed.