This is an action in two counts by the Administrator to enjoin alleged violation of Section 4(a) of the Emergency Price Control Act of 1942, 50 U.S.C.A.Appendix, § 904(a), and of Regulations under that Act. After full hearing on the permanent injunction, the petition was dismissed. Plaintiff appeals.
Appellee is a manufacturer of six varieties of candy bars (Milky Way, Snickers, Three Musketeers, Dr. I. Q., Forever Yours, and Mars) which it sells broadly in interstate commerce. The charge in the first count of the petition is that the Regulations fixed ceiling prices on such products as the highest price charged therefor in March, 1942, to a purchaser of the same class; that these sales were made in boxes (containing 24 bars) placed in cartons holding 12 such boxes; that during March, 1942, “all of the bars of candy of the same kind, and boxes and cartons thereof * * * contained exactly the same ingredients and were exactly the same size and weight during each and every day of the entire month”; that, after the effective date of the Regulations (May 11, 1942), “every candy bar manufactured, sold and delivered * * * was materially reduced in size and weight and as a result thereof all purchasers of said candy bars, and boxes and cartons thereof, have received * * * approximately 11% less candy than they received during the month of March, 1942, without receiving any corresponding reduction in the price”; and that because of such weight reduction, the price has been increased above the highest March, 1942, price, in violation of the Regulations. The second count alleges violation of the Regulation requiring a statement to be made (by July 1, 1942) and kept by appellee showing the highest March price, with “an appropriate description or identification of each of such commodities.” The charged violation under this count is in that appellee “incorrectly listed the weight of each of the commodities sold and delivered during March, 1942, by it, at a lesser weight than the true weight thereof”.
Appellee’s position as shown in its answer is: That it has not and does not sell by weight but by numbers of bars “each having a printed wrapper weight as required by the Federal Food, Drug, and Cosmetic Act [21 U.S.C.A. § 301 et seq.],” and since it pays freight, the consumer is, therefore, not concerned in the weight “if the Defendant has in fact fulfilled its contract to deliver” the required number of bars, each having the weight stated on the wrapper; that during March, 1942 the bars of the same kind, and boxes and cartons “did not contain exactly the same ingredients and were not exactly the same size and weight”; that, just as it changed its formulae in March, also it changed them after the Regulations became effective, such changes being necessitated at various times because of scarcity of materials, weather conditions, efforts at standardization, and to comply with the necessities of business operations — some changes (“particularly with reference to chocolate coating practice”) being made at the instigation of the War Production Board and with the consent of representatives of the Office of Price Administration; that at no *847time did it reduce the net weight of its bars below the net weight shown on the wrappers during March, 1942; that at all times, it sold its bars at the same prices; and that readjustment rulings of the Administrator allowing some of its competitors to reduce weights of candy bars without decrease in prices thereof amount to constructions that the weight of candy bars is not material in determining maximum price violations.
In a memorandum in connection with the temporary injunction, the court stated its view as to the issues as follows: That the commodity is “a five-cent candy bar with a particular trade name and its identity is not changed with every slight fluctuation in weight”; that “to preserve its identity it may be necessary to decrease its weight as one ingredient is necessarily substituted for another and heavier ingredient, not longer obtainable”; that there might be such a decrease in weight as would violate the Regulations; that the test is “Has there been such a decrease in weight beyond a reasonable tolerance justified by the cheapness and nature of the commodity, the method of its manufacture, and the changing character of its ingredients, as evidences a deliberate intent and purpose to obtain more money for the same commodity than was obtained during March ?” ; that “the burden is on the plaintiff to show by competent evidence that the defendant did decrease the weight of its candy bars and to such an extent as exceeded a reasonable tolerance.”
The findings and conclusion were as stated in footnote.1
Contentions Here.
(1) Both parties attack the finding of the court (finding 6) as to the decreases in weights, after the effective date of the *848Regulations, below the base weights. The court found, in definitely stated amounts, such decreases as to four (Milky Way, Snicker, Three Musketeers and Dr. I. Q.) of the six varieties of bars. As to the other two (Forever Yours and Mars), it found increases. Appellant contends that this finding was “contrary to the evidence” as to all kinds of bars, in that the only findings justified by the evidence were that all bars showed decrease and a materially greater decrease than found by the court. Appellee contends that the only justifiable finding was that there had been no competent proof of decrease but, on the contrary, the competent proof showed an increase as to all kinds of bars.
(2) Appellant contends there was no evidence to justify the position that the decrease in weights “might” have been caused by wartime material shortages. Appellee urges that any such decrease might have been caused by variations in formulae (some of which were caused by such shortages) and or methods of manufacture.
(3) Appellant contends the burden of showing that such decreases were caused by wartime shortages or other innocent causes was upon appellee. Appellee urges that negativing such causes is part of plaintiff’s case.
(4) Appellant urges error in the position of the court (stated in the memorandum opinion) that, even though the small weight decreases found were deliberate, no violation resulted. Appellee urges that weight is not a consideration in the mind of the candy bar purchasers and, therefore, small decreases in weight are not violations.
(5) Appellant contends that the action of the appellant in making readjustments by decrease in weights of candy bars sold by other manufacturers is not material in this litigation. Appellee urges that such readjustments should be regarded as administrative rulings showing the lack of or triviality of weight as a price measure in this sort of product.
(6) Appellant attacks and appellee upholds the sufficiency of the evidence to justify finding 9 that the second count was not sustained by proof.
(7) Appellee contends also that appellant departed from his pleadings in the proof.
The above statement of contentions is an attempt to cover concisely the various “points” of the briefs. To get down to more material matters, contentions 5 and 7 may be disposed of shortly. As to contention 5, it is sufficient to state that these readjustments were under regulations authorized by the Act, Title 50 U.S. C.A. Appendix, § 902(c); were upon particular situations; and are really not pertinent here. If they could have any bearing upon the attitude of appellant toward decreases in weights in candy bars, they reveal that such decreases are considered by appellant as influencing price ceilings. As to 7, no departure sufficient to constitute a fatal variance appears. This leaves contentions 1 to 4 as to count one; and contention 6 as to count two.
The contentions as to count one bring before us certain matters which, for convenience in examination, may be regarded as presenting the propositions following:
I. Under what circumstances is a decrease in weight of this kind of product a violation of the Act?
II. In determination of the fact situation here, upon which party is the burden of proof as to whether any decreases were caused by methods of manufacture, by changes in formulae or by both?
III. Do the findings of fact which the court made or findings which the court should have made show such violation ?
I. The Act here involved is the Emergency Price Control Act of 1942, 50 U.S.C.A. Appendix, § 901 et seq.2 The prime purpose of the Act is to prevent undue inflation in prices of commodities and services during the war. One method of reaching this result is by “stabilization” of such prices (Title 50 U.S.C.A. Appendix *849§ 901(a). One way of stabilizing is to fix maximum prices (same, § 902(a). The administrative machinery created to accomplish this stabilization is the Office of Price Administration under direction of a Price Administrator (same, § 921) who is authorized to make regulations.
April 28, 1942, the Administrator promulgated General Maximum Price Regulation (U.S.C. Cong. Service No. 4, p. 375), effective May 11, 1942, establishing maximum prices for commodities as being “The highest price * * * charged for a commodity delivered * * * during March, 1942” (same, p. 376).
The statutory plan — to be worked out by the Administrator — for establishing and applying maximum prices as a means of price stabilization is stated in three provisions of the Act. The first gives authority to establish maximum prices (§ 902(a). The second makes the established prices applicable to the business as it is being conducted — this by forbidding regulations which “compel changes in the business practices, cost practices or methods, or means or aids to distribution, established in any industry, except to prevent circumvention or evasion of any regulation, order, price schedule, or requirement under this Act” (§ 902(h). The third is authority to “provide for such adjustments and reasonable exceptions, as * * * are necessary or proper in order to effectuate the purposes of this Act” (§ 902 (c).
The effect of these three interacting provisions upon each other in making up an integrated plan is as follows. Standing alone, an established maximum price is a level or ceiling which cannot be lawfully exceeded.3 Application of this absolute standard to a particular business is modified to meet actually existing methods of conducting that business, so long as those methods are not used to circumvent or evade the Act. Application of that absolute standard may be modified or avoided in particular instances, by adjustments or exceptions, where the Administrator deems such necessary or proper to effectuate the Act. Thus, the integrated plan is that maximum price levels may not be exceeded except through innocent variations caused by established methods of a business unless the Administrator has allowed a departure by adjustment or by exception.
Applying this construction of the Act to the situation here, there results the following: This company makes six kinds of candy bars from various ingredients which are combined to produce the finished product by established methods of manufacture in its plant. The Administrator contends that it has violated the established maximum prices for its products by the device of decreasing the weights thereof without change in prices. The company contends (1) that any decreases are slight and do not affect the price situation since its products are sold by the piece and not by weight and that the incentive for purchases is the taste; and (2) that any weight decreases are caused by changes in its formulae or by its manufacturing methods or by both.
The circumstance that these products are sold by piece and not by »weight is of no effect. Candy is a food. Weight is one measure of the amount of food. Many kinds of foods are sold by piece or by package. Any decrease in weight of such without a corresponding decrease in price has clearly the same effect, on the purchaser and on the seller, as an increase in price — less is sold for the same money. The degree of decrease in weight with resultant effect of increase in price is not material for reasons hereinbefore stated. That taste is the purchasing incentive is not material — the Act has no concern with why the purchaser buys but only with whether he pays more than the established maximum price. If any decreases are caused by innocent changes in formulae or by innocent methods of manufacture or by both, those decreases are not violations of the Act — by “innocent” is meant such as do not result in circumvention or evasion of the maximum price.
The result as to this issue I is as follows. Since appellee has not obtained from the Administrator any readjustment or exception allowing it to decrease weights, decreases in weights without corresponding decreases in prices violate the Regulation and, therefore, the Act unless *850such decreases are caused by innocent formulae changes, by established manufacturing methods, or by both (innocently employed).
II. This issue is as to which party has the burden of proving the effect on decreases in weights caused by methods of manufacture or and by changes in formulae used in making these candies. Concerning the effect of manufacturing method and of changes in formulae, there are three findings of the court (Nos. 2, 3 and 7), the net effect of which is that the court was unable to determine whether or not the decreases in weights were caused by-manufacturing methods and or changes in formulae. The result is that whichever party had the burden of proving the effects of such method and formulae changes upon such decreases, failed to carry it.
This burden of proof rested upon appellee for two reasons. The first reason is that the general requirement of this remedial Act is strict adherence to established maximum prices and that permissible departures therefrom, caused by manufacturing formulae or methods, are in the nature of exceptions to the general application of this requirement. Hill v. Smith, 260 U.S. 592, 595, 43 S.Ct. 219, 67 L.Ed. 419; Kreitlein v. Ferger, 238 U.S. 21, 26, 35 S.Ct. 685, 59 L.Ed. 1184; Schlemmer v. Buffalo, etc., Ry. Co., 205 U.S. 1, 10, 27 S.Ct. 407, 51 L.Ed. 681; Miller v. United States, 8 Cir., 123 F.2d 715, 718; Manhattan Oil Co. v. Mosby, 8 Cir., 72 F.2d 840, 843. Also see United States v. Denver & R.G.R. Co., 191 U.S. 84, 90, 24 S.Ct. 33, 48 L.Ed. 106. When the Administrator has introduced proof of a decrease in weights with no corresponding price reduction, he has made out a prima facie case of violation of the Regulation and the Act. The defendant can meet this prima facie case by evidence disproving the fact of decrease in weights or by evidence excusing the decrease. If he seeks to excuse the decrease, his position is analogous to confession and avoidance and he has the burden of establishing his excuse to the satisfaction of the court. The second reason is that the evidence to prove the existence of and the effect upon product weights of manufacturing methods and of changes in formulae is peculiarly within the knowledge of the manufacturer. United States v. Denver & R.G.R. Co., 191 U.S. 84, 92, 24 S.Ct. 33, 48 L.Ed. 106; Selma, etc., R. Co. v. United States, 139 U.S. 560, 567, 568, 11 S.Ct. 638, 35 L.Ed. 266; Rossman v. Blunt, 6 Cir., 104 F.2d 877, 880; Cliett v. Scott, 5 Cir., 102 F.2d 725; Miller v. Lykes Bros., etc., Co., 5 Cir., 98 F.2d 185, 186, certiorari denied 305 U.S. 641, 59 S.Ct. 150, 83 L.Ed. 413; Liberty Bell Gold Mining Co. v. Smuggler-Union Mining Co., 8 Cir., 203 F. 795, 804, certiorari denied 231 U.S. 747, 34 S.Ct. 320, 58 L.Ed. 464; The Medea, 9 Cir., 179 F. 781, 786; West v. W. A. McLaughlin & Co.’s Trustee, 6 Cir., 162 F. 124, 128.
III. This point has to do with the sufficiency of the evidence to support three results stated in the findings: That there were decreased weights as to four kinds of bars (Milky Way, Snicker, Three Musketeers and Dr. I. Q.); that there were no decreases (on the contrary, increases) as to two kinds of bars (Forever Yours and Mars); and that the burden of showing that- decreases had been caused by changes in formulae, by manufacturing methods, or by both had. not been carried.
These findings are attacked by both parties. Appellant contends that the evidence required the court to find that there were greater decreases than found as to the four kinds; and that decreases should have been found for the two other kinds. Appellee contends that no decreased should have been found as to any kind of bars; and that, if decreases were found, the court should have found them caused by manufacturing methods and or formulae changes. If any finding is to be overturned, it can be only if we- determine that it is “clearly erroneous.” Federal Rules of Civil Procedure, rule 52(a), 28 U.S.C.A. following section 723c.
We shall separately examine these under the two headings: (a) Weight Variations and (b) Formulae and Manufacturing Methods. For convenience, we consider first the separate and combined effect of the two matters claimed by appellee to excuse any decrease, to-wit, Formulae Changes and Manufacturing Methods.
Formulae.
The evidence of formulae here is contained in 50 exhibits (Nos. 40-89) which are headed “Cost sheet.” Each covers a week. Except the week of June 25-July 5 (when there was no plant operation), one appears for each week in 1942 —beginning January 5 and ending December 26. Each is made up of six sheets, *851one for each brand of candy.4 The purpose of these sheets was to show a “rough approximation of the trend in cost per unit in interim [weekly]” between actual cost figures which were calculated monthly. The sources of the sheets were production figures (which were averages), and, records showing “approximate plan for making the candy.” The poundage was based on “average of experience over certain previous periods of time.” They were made at the end of the week stated on the sheet.
In popular and in dictionary understanding, a formula is “a rule prescribing ingredients, with proportions, for the preparation of a 'compound; a recipe.” Webster’s New International Dictionary, 2nd ed. Irrespective of the cost purpose for which these sheets were made, they do possess the cardinal characteristics of formulae. They state ingredients and quantities thereof. The distinction between them and a true formula is that a formula states a standard for manufacture while these sheets state the ingredients *852and approximate quantities used in previous actual manufacture. In this business, there were no set stated or written formulae but the operation was based upon long practical- experience by experienced operatives in producing the particular kinds of candy. This developed practical, working standards rather than laboratory standards.
As to changes in these practical standards. There were only two changes in ingredient by substitution. The first was in the item “hard butter” used in making the nougat. This had been cocoanut oil until war caused shortages compelled a change to Kelko, largely composed of soybean oil. • The second was through increased use of corn syrup, also in the nougat, to compensate for decreased use of sugar. There was another change — not by substitution but within the same ingredient — in that different kinds of peanuts had to be used in certain brands because of shortage. There is no evidence as to any difference in weights as between cocoanut oil and Kelko, between corn syrup and sugar, or as between these different kinds of peanuts used.
The main changes were in quantities of various ingredients used. Chocolate became scarce, resulting in use of less in coating the kinds of bars where it was used. There were compensating increases in other ingredients, such as caramel. It no where, appears that such shortage had any effect upon the weight of the bars. There is failure to show that any decrease in weights of bars was caused by changes in ingredients or in proportions or weights of ingredients used in the candy.
Manufacturing Method.
While the evidence is clear that the company could and did control the size and weight of its products, yet this control was not absolute.5 Some of its process *853involved the human element which, although by experienced expert workmen, could not be mathematically and constantly exact. However, reasonable exactness was sought and reasonably attained. While the candy was sold by the bar or multiples thereof, yet the influence of size in sales and of size and weight in manufacturing cost is clear. They constitute the prime reasons for the striving for uniformity in manufacture. The results were that, while there might be variation in size or weight of bars (even in the same batch of candy), those variations were only such as were unavoidable and were relatively slight. While a witness for appellant stated they attempted no manufacturing tolerances and merely tried to keep the weight above that printed on the wrappers; that the wrapper weight was an average; and that it would be reasonable to regard a 5% difference therefrom — above or below — as a reasonable manufacturing result; yet he stated that an increase of three-sixty-fourths of an inch in size of the bars would be regarded as a “large” increase. Also, that production directions had been given for reduction in width of one kind of bar by one-thirty-second of an inch, and of another by one-sixty-fourth and by one-thirty-second of an inch. Apparently, these very small dimensional changes could be made on the width and length by the gauges on the sawing machines. The thickness of the bare nougat and also of the caramel coating and of the chocolate coating could be and was changed but not so accurately.
The effect of this evidence as to manufacturing methods is that there could be no absolute, mathematical exactness in the weight and size of bars; that there were no recognized tolerances; that approximate uniformity was striven for by expert, experienced workmen, and that very small changes in the size of bars could be and were made. The only evidence as to a range of variance wás that it might be 5% above or below the wrapper weight, with the effort to keep it above. While it is clear that slight variations in size and, therefore, weight constantly might occur yet we cannot say from all of this evidence that the court was wrong in finding that it had not been shown that the decreases here resulted therefrom— much less that the finding was “clearly erroneous”. Rule 52(a).
Weight Variations.
The court found decreases in four bars (Milky Way, Snickers, Three Musketeers and Dr. I. Q.) and increases as to two bars (Forever Yours and Mars) as between May 12, 1942 (after the Regulations became effective) and March 31, 1942 (the date used by the court for a base price).
The contention of appellant that the evidence required findings of greater decreases as to the four kinds of bars need not be separately examined because these decreases, unexcused by formulae changes or manufacturing methods, are sufficient to necessitate an injunction. The real issues as to this point are presented by appellant’s contention that decreases should have been shown in the two bars (Forever Yours and Mars) ; and by appellee’s contentions that there was no competent evidence as to decreases in any bars and that the competent evidence required finding of increases instead of decreases as to all bars.
The entire evidence has been read and studied. In so far as necessary to determine the above issues, it reveals the situation following briefly stated. The Administrator’s case rests upon the proof of weights shown by invoices and bills of lading of shipments made by the company. The company introduced evidence of increases in the form of the individual weights of 120 separate candy bars selected at random.6
Appellee’s attacks upon the competency of this evidence introduced by appellant must be first determined. They are as follows: “(1) weight not the sole criterion for determining price; (2) if weights relevant, then actual rather than average weights, should have been proved; (3) Plaintiff did not prove correct average weights; (4) If weight sole criterion for determining highest price, then lowest weight must be shown by plaintiff; (5) *854Plaintiff failed to prove alleged reduction in weight had any relation to price of candy bars.”
In order to get down to the more difficult of these contentions, several of them may be disposed of shortly. The contention ( (1) above) that weight is not the sole criterion for determining price is quite true but that does not prevent the obvious conclusion that weight is one criterion in determining violation of regulations establishing maximum prices for food products. Contention (5) is directed at the burden of proof and has been determined hereinbefore.
Argument under contentions (2) and (4) is as to the same matter which is that actual weights of individual candy bars, and not average weights, is the only character of competent evidence to prove weights here. Contention (3) is a challenge of the correctness of the average weights shown in the evidence of the Administrator. We consider those two contentions.
Actual or average weights. This contention is that this action is based upon sale of individual candy bars in excess of the highest price charged in March, 1942, per bar; also that price adjustments granted by the Administrator to Mars’ competitors have been on the bases of individual bars and actual bar weights; and, therefore, it “was incumbent upon Plaintiff to prove what was the lowest weight candy bar of each flavor sold by Defendant in March, 1942, and then to prove the sale of candy bars of less than such weight since the effective date of the General Maximum Price Regulation.” The basis of adjustments as to other candy manufacturers is entirely immaterial. The remaining portion of the contention is that since this suit is based on price raises as to individual bars through weight reductions of individual bars, therefore, the only proper proof of such weight reductions is that secured by separately weighing individual bars and selecting therefrom the lightest weight bar of each variety of candy sold during the month of March and comparing those weights with lighter weights, similarly secured, of individual bars sold after the effective date of the Regulation. It is well to test this contention by the actual situation to which it is to be applied.
The situation is as follows. During March, 1942, and the months after the Regulation became effective, Mars manufactured and shipped daily between ten and eleven thousand cartons of these candy bars. Each carton contained 288 bars. The daily manufacture and shipment was, therefore, between 2,880,000 and 3.168.000 individual bars. During the twenty-six week days of March, 1942, these daily operations would involve between 74.880.000 and 82,368,000 individual candy bars. Usually, the bars were shipped out the same day they were ready for shipment and little stock went over to the next day. Also, this Regulation (promulgated April 28 and effective May 11, 1942) took as a base price period the already past March sales. Long before the Regulation was promulgated and yet longer before the Administrator could have had any reason to investigate compliance by the company, these millions of bars (made and sold in March) had gone to dealers all over the country and many must have been consumed. It is to this situation that appellee insists there should be applied a rule of evidence which is impossible. If this is the applicable rule, then the Act is a failure as to most if not all of the many foods which are sold by piece or by package. This is true because, if the base period selected is before the regulation is effective, there is no way to detemine the base price, while, if a period is selected which is later than the regulation, the seller can sell one piece or package at such a high price as to render maximum price regulation, as to him, an idle gesture. Rules of evidence are to promote, not to thwart, justice. Since the nature of the situation is such as to make the weighing of each bar impossible, some other line of proof is not only proper but is necessary. An average weight, if fairly reached, would seem unobjectionable as a means of determining individual bar weights. This is particularly true when the existence of individual variations in weights here is considered. Such variations could not be escaped. However, they were only such as were unavoidable and they were equally likely to go above or below the desired. There seems no-reason why average weight should not be evidence for calculation of fair individual weights, provided such averages, were determined in a fair and representative manner.
Correctness of These Average-Weights. Appellee contends that these *855averages “were not designed for the purpose for which Plaintiff attempted to use them, and were wholly inaccurate for that purpose.” The purpose for which the average weights were made is of no importance unless it affects the accuracy or representative character of the average weights.
The purpose, the method of reaching, and the use of the average weights, relied on here by the Administrator, were as follows. The Western Weighing and Inspection Bureau is an agency of the railroads, at Chicago, to weigh and inspect freight under agreements with shippers. One purpose of the Bureau is to reach a standard shipping weight (for freight charge purposes) on individual packages where the shipper has many uniform packages to be transported. The practical advantages, to the shipper and to the railroad, of such an arrangement .are obvious. One procedure to secure this result is for representatives of the Bureau to weigh a number of such uniform packages and therefrom strike an average weight. The average weight is then used by the shipper in its bills of lading and other shipping papers and is accepted by the railroads. When “any change is made ■which will affect the weight arrived at by the use of the average, including any change made in package or material used,” the shipper is required promptly to notify .the Bureau. When such notice is given,, :a new average weight is determined. The. Bureau has the right to inspect, at any time, all records of the shipper bearing ■upon weights in order to verify the weights and descriptions used by the shipper in its invoices, shipping orders, bills of lading ■or weight certificates. A contract embodying the foregoing procedure existed between this company and the Bureau covering shipments of cartons of candy bars. .Such average weights were used by the company in its invoices and bills of lading and the company paid freight charges on that basis. It is these average weights, as shown in many invoices and bills of lading made out by the company and coming from its files, which are the bases relied upon by the Administrator for proof of weights of the candy bars shipped at various times here pertinent.
The actual method of reaching these average weights here was as follows: The shipping package was a carton (containing twelve paper boxes, each holding twenty-four bars of the particular brand of candy). An experienced weigher from the Bureau went to the shipping room of Mars where there was a scale (required to be kept accurate by the company) and a number of cartons of candy ready for shipment. He reached an average weight for each brand of candy separately in the manner following: one carton weighed, another added and both weighed and likewise a third, fourth and fifth carton added and cumulatively weighed; these five removed and five new cartons put on and weighed; the second five removed, and ten other cartons weighed; the ten removed, and ten other cartons weighed; these removed and ten others weighed; these removed and ten others weighed. This procedure gave weights of 1, 2, 3 and 4 cartons, two weights of 5 cartons (one of which included the four cartons previously built up) and four weights of ten cartons. The average weight was reached by adding the weights of the two 5 carton lots and of the four 10 carton lot weights and dividing by SO. The results of this weighing of each brand of candy were set out on a “shippers certificate of weight.” This certificate contained, among other matters, the date of weighing, date average weight of last previous test, a description of the package (by name of brand and dimensions), name of manufacturer of the containers (boxes and cartons), signature of a representative of the company, and signed certificate by Bureau weigher as to his participation in the weighing and as to correctness of the weights shown. Where the weighing was the result of a change in the package, the certificate set forth wherein the change was made.7 The weights were checked, at least yearly, by the Bureau; and also whenever changes were made on notification. The company kept an independent check of weights, described as: “We spot check *856one hundred fifty to two hundred cartons of maybe three or four products daily and maybe the next day we will check the other three or four products so we see they conform to the shipping invoice on the bills-of-lading.”
While these weights were for the purpose of shipping, yet they were fairly reached in a manner to represent individual carton weights; were used by both parties as the basis for payment of thousands of dollars of freight by Mars (which paid and made no charge for freight to its customers) ; and were constantly checked by appellee. They are a fair basis of loaded carton weights for any purpose.
These weights were of cartons, each containing twelve paper boxes of twenty-four candy bars each. To reduce those carton weights to candy bar weights, evidence was introduced as to average weights of empty cartons and boxes; these weights deducted from the above shipping carton weights; and the result divided by 288 (the number of bars in a carton). We think this method of determining average weights of bars as correct and fair as could be obtained under the circumstances and that such average weights are fairly representative of and entirely competent evidence of weights of individual candy bars for the purposes of this suit.
Since we determine that the above average weights are competent evidence, the next issue is whether the.findings of the court that there were increases in the two bars (Forever Yours and Mars) are to be set aside on the Administrator’s contention that the evidence compelled the contrary findings that there had been decreases, instead of increases, as to these two bars. Appellant urges that, in reaching the figures which the court found to represent the increases on May 12 over the March base prices of these two kinds of bars, the court (a) used erroneous March base price figures; and (b) went to the “cost sheets” for its bases and that such sheets “furnish merely a theoretical average bar weight derived from an analysis of formula data respecting various candy ingredients in their raw state” and that there was no evidence that bars of such weights had ever been manufactured or shipped.
As to the March Base Price. Appellant urges that the court should have adopted, as the March base price, the smallest average weight of that month, as shown by the company invoices. The company urges that the base price is that shown by average weights taken on March 31 by the Bureau when (it contends) there was a new set of average weights determined which were the lowest for March. Concededly, these March 31 certified weights are lower than any shown by the March invoices. Thus, this issue becomes whether or not the average weights taken and certified by the Bureau on March 31 are to be taken as the base weight.
There is no dispute that the Bureau took and certified these weights on March 31 but that certification is not the test. The Regulation establishing the March base prescribed “the highest price charged by the seller during March, 1942,” to apply tO' commodities “delivered * * * during March, 1942.” Delivery during March, 1942, was defined as being “if during such month it was received by the purchaser or by any carrier * * * for shipment to the purchaser”. Gen.Max.Price Regulation, sec. 20(d). Thus the test is a sale and delivery to the purchaser or to a carrier for him during March. The contest here is over whether there was such sale and delivery during March under the Bureau weights taken and certified that day.
The evidence as to such sale and delivery or not is as follows:
The Administrator introduced unchallenged testimony that customarily there was a lag of a day between the certification of weights by the Bureau and shipments thereunder; that there was a similar lag between the making of invoices and shipments thereon; that there was no invoice or bill of lading of March 31 showing shipment at the new weights but that all such showed shipment at the old weights. The company gave evidence, by the Bureau weigher and by the manager of its shipping department, that the weight tests were made on March 31 and that the cartons weighed were taken at once from the scales and loaded into nearby freight cars; and the manager testified these cars left the plant that day. Standing alone, the above evidence by the parties might present a clear case of conflict. However, the Administrator urges that no sale and delivery was thus shown because there was undisputed evidence that the company had warehouses in Newark, New Jersey, and in Dallas, Texas, to which it shipped for storage and distribution; that there was no evidence as to whether these cartons loaded *857on March 31 were on sales to customers or on shipments to the warehouses; and that it must he shown they were on sales to come within the Regulation. Certainly it is true that delivery to the carrier must, to be within the Regulation, be for shipment to purchasers. There was no evidence as to sales of or as to destinations of these cartons. Therefore, the vital question is upon whom rested the burden of showing whether or not these deliveries to the carrier were on purchases. We think this burden was on the company because it alone had peculiar knowledge of that fact (see citations hereinbefore under Point II). This is particularly true here because no shipping or other related selling records show any such sales on that date but to the contrary. The weights taken and certified on March 31 have not been proven to be within the Regulation and cannot be used as bases for March maximum prices.
As to the Evidence Used and Which should have been Used by the Comt in Determining Decreases or Increases. The court seems to have used different characters of evidence in determining increases or decreases as to the different brands of candy. The evidence by the company as to weights of separately weighed individual bars was rejected. The weights as to Milky Way bars was derived from the weights certified by the Bureau on March 31 and on May 12. Exhibit 118, Tr. 1283, 1284. As to the other five brands, the court used the weights shown in the cost sheets. Exhibits 119-123, Tr. pp. 1292, 1302, 1312, 1323 and 1334.
We think the court erred in selecting the Bureau certified weights, as to Milky Way Bars. These weights were tak,en and certified as of March 31 and May 12, respectively. For reasons stated hereinbefore, there was no evidence of any delivered sales upon those dates of any Milky Way bars at these certified weights.
We think the court erred in selecting the cost sheets weights as to the other five bars. While these weights were some evidence of weights, they were of a nature and so affected by variables as to render them obviously much less reliable than other evidence before the court. They were derived from weights of raw ingredients, took no account of water additions and effect of cooking during subsequent manufacture nor of material adhering to machines nor of “scrap” coming over from previous manufacture or passing over from the week covered by the figures. While the cost sheets purport to show ingredients and weights as to each particular kind of candy, the evidence is clear that different kinds of candy were made the same day; that the same nought was used for all bars; that caramel from the same tanks was used on four kinds of bars; and that chocolate from the same tanks was used on all bars in varying amounts (one kind of bar being entirely coated and other kinds not). It is obvious that the allocation of even these raw ingredients to different kinds of bars could be only a very rough approximation based on no reliable data. Nor was any attempt made to have such allocation closely approximate. The only purpose of these sheets was to keep a weéfcly ingredients cost “trend” check between the regular monthly computations of ingredient consumptions. Appellant regarded this data merely as a check on the weight changes shown by the invoices; and appellee strongly contended that these sheets were not competent to show weights “since they have to do only with the costs and since they are only the plan by which the defendant attempts to make the candy; they are not the accurate records as to what has gone into the candy cnor as to what is to go into the candy, and therefore, neither the records themselves have any bearing upon the weights of the candy nor the weights of any one or all of the candies nor do they have any bearing upon any of the issues in this case.”
We think the court correctly disregarded the evidence as to separate weights of individual bars. While a bar was the unit of sale by the retailer, it was not of sale by appellee. With the exception of a half dozen special instances where a box was sold, its sales were all by the carton. The maintenance of a maximum price here is not affected by nor concerned with a few vagrant individual bars which may happen to be above or below the established price- — such do not affect inflation or price stabilization. The concern is with a conscious decrease in weights of its products generally as a course of business.
The result of the above statements (as to use of certified weights, of cost sheet weights, and of separately weighed bars) is that none of them should have been used as the ground work for determining decreases or increases in weights. The use of such was “clearly erroneous.” The evidence which should have been used is the *858invoices and accepted bills of lading which showed the weights at which the candy was shipped. The company paid large amounts of freight on that basis constantly and made an almost daily check thereon. Those weights are the bases for far more accurate and fair average weights than any other evidence here. They allow to the company and they hold it only to the results shown by the established method of conducting its business. They are fair to the company and to the Administrator.
Count 2.
The court found “there was no sufficient proof of the facts charged in the second count of the complaint.” That finding is challenged by appellant. This count charges violation of a Regulation of the Administrator in failing to prepare (on or before July 1, 1942) and to keep available for examination, a true statement showing its highest price during March, 1942, with an appropriate description or identification of each of such commodities. It is not charged that appellee failed to make a statement and keep it available for examination. The charge is that the statement “incorrectly listed the weight of each of the commodities sold and delivered during March, 1942, by it, at a lesser weight than the true weight thereof.” Appellee insists that the weights so listed were true.
The weights listed in the statement were weights which appellee claimed were those of candy bars sold and delivered by it to a carrier on March 31, 1942. On that day, the weigher for the Bureau had made test average weights for shipping purposes. These tests applied to all six brands. These weights are the same as those rejected hereinbefore because the proof of sales and deliveries at these weights on March 31 was not made. The result is that the weights on the statement made by appellee did not comply with the Regulation.
Conclusion.
The results of the foregoing determinations are that the decree is reversed as to both counts of the petition with directions to set aside the order of dismissal and enter a decree enjoining defendant, its officers, agents and employees from selling, delivering or offering for sale or delivery any candy bars at prices in excess of the highest prices shown on its invoices and bills of lading during March, 1942, and commanding the making of a statement of such highest March prices and maintenance of the same for inspection under the Regulation providing therefor. To prevent disruption of defendant’s business, the decree should be made effective fifteen days after date of entry thereof. Mandate from this Court will issue ten days after entry of order in this Court.
“1. The defendant in March 1942 manufactured and sold candy bars intended to be retailed to ultimate consumers at five cents each. Defendant sold them for 68 cents per box of twenty-four delivered.
“2. The candy bars were of six kinds and names: ‘Milky Way’, ‘Snickers,’ ‘The 3 Musketeers,’ ‘Forever Yours,’ ‘Mars’ and ‘Dr. I. Q.’ Each kind was of a different average weight than the five other kinds. No two bars even of a particular kind were of exactly the same weight. The defendant, however, could determine approximately the average weight of each kind of candy bar.
“3. The evidence does not disclose the precise formula for any of the candy bars. Each contained a number of ingredients. The formula for any bar was variable. One ingredient might be substituted for another. The substituted ingredient might be of the same or a different weight.
“4. The especially distinguishing characteristic of each kind of candy bar was its taste. The candy bar was sold as a piece of candy, not as so much in weight of sugar, syrup, cocoa, etc. It was a bar of candy, in one piece, convenient to the hand, wrapped in a protective covering, to be sold at retail for five cents.
“5. On the wrapper around each candy bar sold by defendant was printed what purported to be the net weight of the candy, as 2% ounces. Generally speaking, the actual weight always exceeded the stated weight.
“6. The average weight of the ‘Milky Way’ candy bar sold on March 31, 1942, was 2.67970 ounces; the average weight of the same bar sold on May 12, 1942, was 2.61389 ounces; a decrease in weight of .06581 ounces. The average weight of the ‘Snicker’ candy bar sold on March 31, 1942, was 2.86477 ounces; the average weight of the same bar sold on May 12, 1942, was 2.68253 ounces; a decrease in weight of .18224 ounces. The average weight of ‘The 3 Musketeers’ candy bar sold on March 31, 1942, was 2.76321 ounces; the average weight of the same bar sold on May 12, 1942, was 2.68329 ounces; a decrease in weight of .07992 ounces. The average weight of the ‘Forever Yours’ candy bar sold on March 31, 1942, was 2.74768 ounces; the average weight of the same bar sold on May 12, 1942, was 2.78969 ounces; an increase of .04201 ounces. The average weight of the ‘Mars’ candy bar sold on March 31, 1942, was 1.75035 ounces; the average weight of the same bar sold on May 12, 1942, was 1.75127 ounces; an increase of .00092 ounces. The average weight of the ‘Dr. I. Q.’ candy bar sold on March 31, 1942, was 2.12826 ounces; the average weight of the same bar sold on May 12, 1942, was 2.11119 ounces; a decrease of .01707 ounces. This decrease (as to most candy bars) after March prevailed until the issuance of the restraining order in December 1942.
“7. The evidence does not disclose from what the decreases set out in Finding No. 6 resulted, whether from the substitution of lighter for heavier ingredients or from a deliberate purpose to decrease or increase weights.
“8. The defendant, unless restrained, will continue to sell candy bars which vary in weight from those sold in March *8481942 by variations as great as those set out in Finding No. 6.
“9. There was no sufficient proof of the facts charged in the Second Count of the Complaint.
“Conclusion of Law.
“If a candy bar is decreased in weight in an amount no greater than that proved in this case and is sold after the decrease at the same price per bar as before the decrease, there has been no increase in price within the meaning of the regulation of April 28, 1942.”
This Act (January 30, 1942, c. 26, 56 Stat. 23) was amended (Oct. 2, 1942, c. 578, 56 Stat. 765) in respects not here material.
This must be true because of the purposes of the Act and because of the nature of a maximum price. How can prices be stabilized if they are variable? How can a price be a maximum if it can be exceeded? This is no matter of degree — successive trickles grow into torrents which sweep away the dam.
An example of the form and character of contents is as follows:
A very skeleton outline of the manufacturing methods -will show the situation. All six brands are made from the same body or nougat. The differences come afterwards in the coatings put on the nougat and in the addition or not of nuts. The first step is a sugar, salt and water mixture in large kettles. A “certain amount of water” is placed in the kettle and sugar run in from storage tanks above. Salt is added and the mixture stirred with a large “paddle stir” by an operator. When this mixture has reached the stage satisfactory to the operator, it is pumped to another receptacle where corn syrup is added and mixed and the mixture pumped into tanks (on the top floor). From there it is brought by pipes, with valve controls, to the “cookers”. After cooking, it is run hot into “beater bowls” where an egg whites mixture is measured' and added and the entire mass whipped into a “frothy substance.” This mixture is conveyed by hand in buckets or bowls to the “dough mixers.” There “scrap” is thrown in. “Scrap” is the leavings of tag ends and imperfect pieces from previous operations on the cutting tables. Also, the fats, malted milk and cocoa are added. All of these are stirred until the mixture reaches a state satisfactory to the operator. This nougat is tipped out on grease covered “tables” which move on rollers below the dough mixers. These tables are from 4 to 4% feet wide by from 8 to 10 feet long. They have iron bars, about one inch high, which are placed along the sides and ends before the table receives the nougat. As dropped on the table, the nougat is a sticky, rather fluid, semi-viscous mass which neither runs freely nor remains stationary. The. next step is to spread this mass over the table inside the rods to secure the desired thickness of the nougat for bars. This is done by three operations: two operators use an iron bar, gauged on the side rods, to spread the mass as uniformly as possible over the table; with hand spreaders, they fill hollows and „ rectify inequalities of surface; that mass is tested for maximum thickness with a spiked hand tool. The table moves then into “bunkers” where the nougat is partially cooled. This completes the nougat. Such bars as are to be caramel coated then pass under tanks from which caramel is released on to the nougat where it is spread to the desired thickness by two operators using an iron bar. Candy not having caramel coating and nougat after receiving caramel coating pass into a "conditioning room” where it remains for some hours to achieve a “particular crystalline set of sugars and fat crystallization” and is cooled for cutting. The cutting of this nougat layer is as follows: it is cut by hand (with long, heavy knives) or by machinery into strips from one-third to one-half the width of the table; these strips are placed, with closely joined ends, on a moving metal conveyor which carries them to an adjustable “sawing” machine which cuts them into the bars. The next operation is the chocolate application. Some bars are “dabbed” by hand and nuts pressed into the chocolate. Milky Way bars are *853passed on a wire mesh conveyor into an “enrober” where chocolate is poured and splashed on them — the amount of chocolate adhering depending upon the temperature of the bars and of the chocolate and upon the speed of the conveyor. The manufacturing process ends with passing all bars into a cooling tunnel which hardens them for wrapping and packing.
Twenty bars were taken from each of the six brands — each on a different date between March 4 and December 31, 1942. There is little harmony in selecting the same - dates for different brands (see Defendant’s Exhibits 29-34).
The transcript contains twelve certificates of change scattered between January 7 and December 22, 1942. Each affects one or more kinds of bars. As to some bars, there is no statement of the reason for change. Where reasons for change are stated, the reason is “change in bar” except as to one kind, on March 6, 1942, and aa to three kinds, on May 12, 1942 — asi to the first of these, the notation was “change made in candy bars also size of carton”; and as to the other three, “change made in bar and carton.”