CUMMINGS, Circuit Judge.

This case arises on the petition of National Dairy Products Corporation (“National”) to set aside an order of the Federal Trade Commission applicable to its Sealtest Foods Division.

National is a Delaware corporation with its principal office and place of business in New York City. It is engaged in the business of purchasing, manufacturing, processing, distributing and selling dairy and other products throughout the United States. It is the nation’s largest dairy product distributor. Its Sealtest Division has general supervision over National’s food, milk and ice cream divisions and subsidiaries. The Sealtest divisions sell a diversified line of food products, including milk and ice cream. In 1956, National’s net sales were approximately $1,352,878,000, increasing to $1,790,834,000 in 1961.

At the close of 1957, the Federal Trade Commission issued a complaint charging that National had violated Sections 2(a) and (d) of the Clayton Act, as amended by the Robinson-Patman Act (15 U.S.C. §§ 13(a) and 13(d)), in the course of sales of milk and other dairy products through its Sealtest Foods Division. In July 1963, a hearing examiner held that National had violated both statutory provisions and accordingly recommended a *521 cease and desist order after rejecting National's defenses of cost justification and good-faith meeting of competition.

National appealed to the Commission from the Section 2(a) portion of the examiner’s order. The Commission granted the appeal in part in a 2-1 decision culminating in the following order:

“IT is ordered that respondent National Dairy Products Corporation, a corporation, and its officers, employees, agents and representatives, directly or through any corporate or other device in or in connection with the offering for sale, sale or distribution of any of the items in the product line of its Seal-test Foods Division, including but not limited to fluid milk, dairy products, ice cream and other food products, in commerce, as ‘commerce’ is defined in the Clayton Act, do forthwith cease and desist from:

“1. Discriminating, directly or indirectly, in the price of such products of like grade and quality by selling to any purchaser at net prices higher than the net prices charged any other purchaser who competes with the purchaser paying the higher price.” 1

In addition to arguing that the Commission’s findings and conclusions are unsupported by substantial evidence, National asserts that the Commission majority misinterpreted the clauses in Sections 2(a) and (b) of the Clayton Act dealing with cost justification, good-faith meeting of competition and competitive effects. Commissioner Elman’s dissent, on which National relies, deals only with the standards to be applied to the defense of meeting competition under Section 2(b) of the Act. The Commission’s order was based on National’s discriminatory pricing in the following areas: (1) Jackson-Lansing-Battle Creek, Michigan; (2) Toledo, Ohio — Monroe, Michigan; (3) Memphis, and (4) New Orleans. As to the first area, the Commission held that the evidence was insufficient to find that National’s sales were in interstate commerce, so that only the other three area are presently involved. In this opinion, each area and the legal issues pertaining thereto will be covered separately.

Toledo-Monroe Area

Here in 1958 National granted 13 customers a fluid milk discount of 12%, 8 customers 10%, and one customer 7%. These 22 discounts were in excess of those received by National’s other retail store customers. One hundred fifty-eight of them received no discounts and 112 received discounts ranging from 2% to 6%.

Although rejecting the examiner’s finding of primary line injury, the Commission sustained his finding of potential secondary line injury to retail grocers selling Sealtest milk. The competitive effects clause of the statute of course does not require showing that injury has actually occurred, but merely that the effect of the discrimination “may be substantially to lessen competition” (15 U.S.C. § 13(a)). As here, any substantial, sustained differential between competing resellers is prima facie injurious. “Mini-injury” is the test. Rowe, “Section 2(a) of the Robinson-Patman Act New Dimensions in the Competitive Injury Concept,” 37 ABA Antitrust Law Journal 14, 16 (1968).

Only six independent grocery stores received more than a 6% discount from National, whereas most of its chain and group store customers were receiving a 12% discount. These high discounts enabled them to sell Sealtest milk at a price lower than the price paid to National for such milk by all but six of its *522 independent store customers. The evidence clearly shows that National’s discounts resulted “in price differentials between competing purchasers sufficient in amount to influence their resale prices” of milk. Under Federal Trade Commission v. Morton Salt Company, 334 U.S. 37, 47, 68 S.Ct. 822, 92 L.Ed. 1196, this showing is adequate to support the Commission’s finding that the effect of National’s price discriminations might be substantially to injure competition among retail stores in the Toledo-Monroe area.

In E. Edelmann & Company v. Federal Trade Commission, 239 F.2d 152, 154 (7th Cir. .1956), certiorari denied, 355 U.S. 941, 78 S.Ct. 426, 2 L.Ed.2d 422, we held the competitive effects clause of Section 2(a) of the Clayton Act satisfied in the following circumstances :

“We therefore turn to the record which shows substantial discrimina-tions in price; that the purchasers of petitioner’s products sold in a market where competition was keen; that these purchasers operated on small profit margins; that many of the purchasers found it expedient to enter into group buying arrangements for the purpose of aggregating their purchases and thereby obtaining higher discounts than they would otherwise receive as ordinary jobbers in contrast to the warehouse distributor.”

Since all these factors were present here, the Commission’s finding of probable competitive injury must stand. This is true even if there had been direct testimony by non-favored customers that the price discriminations had not injured their businesses. Foremost Dairies, Inc. v. Federal Trade Commission, 348 F.2d 674, 680 (5th Cir. 1965), certiorari denied, 382 U.S. 959, 86 S.Ct. 435, 15 L.Ed.2d 362. As there pointed out, injury may be inferred even if the favored customer did not undersell his rivals, for a substantial price advantage can enlarge the favored buyer’s profit margin or enable him to offer attractive services to his customers.

As in United Biscuit Company of America v. Federal Trade Commission, 350 F.2d 615, 621 (7th Cir. 1965), cer-tiorari denied, 383 U.S. 926, 86 S.Ct. 930, 15 L.Ed.2d 845, the Commission did not apply a per se test in finding that the competitive effect of National’s pricing practices would be substantial. Instead, it relied on the following factors and testimony of independent store owners (Trade Reg.Rep. (65-67 Transfer Binder) jf 17,656, pp. 22,917-22,918) (1966)):

“They received either no discount or a discount of 2 or 3 percent on their purchases of Sealtest milk. Their retail prices to the public generally were from 41 cents to 43 cents a half gallon. The generally prevailing retail price of Sealtest milk at the chain stores was 37 cents. At regular intervals, either on weekends or each month, the chains offered Sealtest milk at three half gallons for one dollar.

“The independent store owners named chain stores selling Sealtest milk as their competitors. All of them stated that they had lost business as a result of the chain stores’ prices, one of them characterizing the effect of such pricing as ‘devastating’ on his sales volume. Other owners testified that their customers were not willing to pay the price their stores charged for milk and would drive a substantial distance to take advantage of low milk prices. One Toledo grocer, in answer to a question as to whether his store was in competition with other stores selling Sealtest milk stated: ‘We can’t be in competition with them [at] the price we pay for it. We do try to run a special oceassionally at a loss to ourselves to try to be competitive.’ A Kroger store, receiving a 12 per cent discount, is located four blocks from his store.”

This evidence satisfies the Morton Salt and United Biscuit tests. Furthermore, unless minuscule, the por *523 tion of the market that might be affected by the charging of these discriminatory prices is immaterial. Whitaker Cable Corp. v. Federal Trade Commission, 239 F.2d 253, 255-256 (7th Cir. 1956), certiorari denied, 353 U.S. 938, 77 S.Ct. 813, 1 L.Ed.2d 761. Nor is competitive injury negated by the pricing policies of National’s competitors. Federal Trade Commission v. A. E. Staley Manufacturing Co., 324 U.S. 746, 751, 753-754, 65 S.Ct. 971, 89 L.Ed. 1338; Whitaker Cable Corp. v. Federal Trade Commission, ibid, at pp. 254-256; Foremost Dairies, Inc. v. Federal Trade Commission, 348 F.2d 674, 679 (5th Cir. 1965), certiorari denied, 382 U.S. 959, 86 S.Ct. 435, 15 L.Ed.2d 362. We also cannot accept National’s argument that the disfavored independents should have joined voluntary or cooperative groups and thus obtained higher discounts. The Robinson-Patman Act does not force them to sacrifice their independence. One of the reasons for its enactment was to protect the independents from the chains and other large buying groups. Federal Trade Commission v. Anheuser-Busch, Inc., 363 U.S. 536, 543-544, 80 S.Ct. 1267, 4 L.Ed.2d 1385; Rowe, Price Discrimination Under the Robinson-Patman Act (1962), pp. 11-23.

Even though National’s discounts violated Section 2(a) of the Act, National would have a good defense by showing that its lower price “was made in good faith to meet an equally low price of a competitor” (15 U.S.C. § 13(b)). As to its 22 customers receiving more than the customary discounts, National asserts that discounts to 19 were granted in good faith to meet offers of competitors. The principal reason given by the Commission for rejecting the good-faith meeting of competition defense was as follows (op. cit. p. 22,922) :

“ [R] espondent has failed to demonstrate any reason to believe that the wholesale list prices of competitors whose prices it claims to have met were the same as its own. Not only has respondent failed to make this showing, but the evidence clearly and convincingly discloses that the wholesale list prices of' certain of these competitors were higher than respondent’s.”

The Commission apparently considered that National had not made the requisite showing because National did not prove that its competitors’ list prices were the same or lower, but that requirement is inconsistent with Federal Trade Commission v. A. E. Staley Manufacturing Co., 324 U.S. 746, 65 S.Ct. 971, 89 L.Ed. 1338; Callaway Mills Co. v. Federal Trade Commission, 362 F.2d 435, 443-444 (5th Cir. 1966), and Forster Manufacturing Co. v. Federal Trade Commission, 335 F.2d 47, 55-56 (1st Cir. 1964), certiorari denied, 380 U.S. 906, 85 S.Ct. 887, 13 L.Ed.2d 794. We agree with Commissioner Elman’s dissenting view that such a burden of proof is too strict and unreasonable and is not imposed by Section 2(b); but even if there were such a burden, National has met it.

Actually the evidence showed that Page’s list price for half gallons was 2 cents below National’s; Driggs’ half gallon prices were the same; Meadow-gold’s, Trilby Farm’s and Cherry Grove’s prices were the same or less; while Wilson Dairy, Independent Dairies and United Dairies had lower Monroe prices than National. The only relevant list price missing from the record is Babcock’s, and the majority opinion concluded that Babcock’s list was higher than National’s. The majority speculated that Babcock must have had a higher list price than Sealtest because Babcock’s discount schedule had lower point requirements. If this were true, Babcock would not be competitive with Page, Driggs, or the lower-priced small dairies. The Commission assumed that Babcock’s list prices must be higher than National’s because National “could not sell milk in this market at net prices which were consistently one or two cents above its competitors’ ” (op. cit. p. 22,923 (1966)). This theory is contradicted by the facts, for the record shows that various National competitors had the same list prices and better discounts than National. Being a well-known national brand, Sealtest could sell *524 at a higher price than local brands. The record shows that Country Store priced Sealtest 2 cents above Driggs, and Seal-test was selling from 2 to 4 cents above Page in Clark’s Market. The record also shows that National had 253 independent store customers in the 0 to 4% discount brackets, accounting for more than 10% of National’s sales, even though Page’s list price was 2 cents less than Sealtest’s. Thus Page, without granting any discount, could undercut National in this group by as much as 2 cents. We conclude that no substantial evidence supports the Commission’s finding that Babcock’s list prices were higher than National’s.

The Commission rejected the good-faith meeting of competition defense with respect to National’s discounts to the Associated Grocers and Saveway group purchasers, on the ground that those concerns were already planning to replace Meadowgold as their milk supplier. National’s representatives were seasoned dairy men and were familiar with the competitive situation, so that they should have been aware of AG and Saveway dissatisfaction with Meadowgold. It was, we believe, permissible for the Commission to conclude that National did not in good faith believe it was necessary to compete with Meadowgold’s prices in order to recapture AG and Safeway as customers.

National has not endeavored to support its discounts to Joseph’s, National Food or Wrigley’s corporate chains by the defense of meeting competition. Joseph’s and National Food were both large customers, and even Wrigley’s purchased $4300 worth of milk per month from National. National’s failure to support the meeting competition defense to these three customers and AG and Saveway justified the Commission’s view that National’s discounts in this area might substantially injure competition.

Another reason advanced by the Commission majority for rejecting the defense of meeting competition was that National was meeting unlawful discounts with respect to the AG and Saveway accounts. However, as held in Standard Oil Co. v. Brown, 238 F.2d 54, 58 (5th Cir. 1956), Section 2(b) may be satisfied unless the seller is meeting prices “that he knows to be illegal or that are of such a nature as are inherently illegal.” That standard was met here, for Meadowgold’s prices to AG and Saveway were not plainly illegal. Meadowgold’s discounts may have been cost justified 2 or made to meet another competitor’s discounts. 3 Moreover, as Commissioner Elman observed, the Commission previously abandoned any requirement that a seller must show it had reason to believe that it was meeting lawful lower prices. See Continental Baking Co. (Trade Reg.Rep. (63-65 Transfer Binder) § 16,720 (1963)); Ponca Wholesale Mercantile Co. (Trade Reg.Rep. (63-65 Transfer Binder) § 16,-814 (1964)), and Beatrice Foods Co. (Trade- Reg.Rep. (65-67 Transfer Binder) j[ 17,311 (1965)).

Since the Commission’s opinion does not discuss the meeting competition defense with respect to 17 other Toledo-Monroe customers of National, this opinion need not consider those off-scale discounts.

Next, National relies on the cost savings defense with respect to 8 of its 22 customers receiving discriminatory discounts. The milk deliveries were to the individual stores and not to any central warehouse. Time studies showed that it took National’s drivers less time per ease to deliver to a large volume store, so that National’s per case distribution costs would be less in serving such stores. The cost savings proviso permits differentials making “only due allowance for differences in the cost of manufacture, sale, or delivery resulting from the delivery *525 methods or quantities” in which commodities are sold or delivered to purchasers (15 U.S.C. § 13(a)). National endeavored to establish this defense both on a “per [chain .or group] purchaser” and on a store-by-store basis.

Under the former method, National totaled the distribution cost to each unit and divided it by the total dollar purchases of rebatable products of all units in the chain or group to obtain the cost per dollar of serving the purchaser/ This averaging method would warrant the same discounts to all stores whose volume qualifies for the particular discount in question, assuming the discount brackets are fairly drawn. Therefore, averaging would be permissible for National with respect to six of its corporate chain customers. 4 /However, in the case of the Red & White Stores and the Save-way Stores, the averaging method would not be justified in granting a 12% discount to each member store because various of them individually purchased enough milk only to qualify for a much smaller percentage discount, such as received by most of National’s independent store customers. United States v. Borden Co., 370 U.S. 460, 82 S.Ct. 1309, 8 L.Ed. 2d 627, condemns cost justification that would create such artificial disparities. There the Court disapproved of discounts permitting an independent store to receive a lower discount than a chain store of the same size. The following cost savings guidelines were established in Borden (370 U.S. at p. 469, 82 S.Ct. at p. 1314):

“A balance is struck by the use of classes for cost justification which are composed of members of such selfsameness as to make the averaging of the cost of dealing with the group a valid and reasonable indicium of the cost of dealing with any specific group member. High on the list of ‘musts’ in the use of the average cost of customer groupings under the proviso of § 2(a) is a close resemblance of the individual members of each group on the essential point or points which determine the costs considered.”

National’s “per customer” cost savings defense fails because there is no “close resemblance of the individual members” of these two voluntary groups, so that the cost of serving them may not be computed by using the averaging method. 5 Otherwise, as Justice Clark picturesquely put it, one would be “averaging one horse and one rabbit” (at p. 470, 82 S.Ct. at p. 1314), with an entirely different cost of serving each.

National argues that “purchaser” as used in the cost proviso means all the stores in a chain or group taken as a unit. This may be true, but it .does not help National. The crucial question is not whether a chain is a purchaser, but whether the discount is “due allowance” for cost differences. The Commission observed that when a stoi^e, by virtue of being averaged with a larger store, receives a discount it has not earned, it gains an unfair competitive advantage of the sort which the amended Clayton Act was designed to prevent. An averaging system, when it permits some stores to receive a significantly larger discount than they could earn individually, has an anticompetitive effect comparable to that *526 of practices long condemned by the Act. Thus a territorial price discrimination violates the Act because a seller’s high prices in one area subsidize its low prices in another. Moore v. Mead’s Fine Bread Co., 348 U.S. 115, 119, 75 S.Ct. 148, 99 L.Ed. 145. Likewise, the Supreme Court has condemned a basing point system in which the seller absorbs freight charges to some customers and adds “phantom” freight charges to others. Corn Products Refining Co. v. Federal Trade Commission, 324 U.S. 726, 65 S.Ct. 961, 89 L.Ed. 1320. In that case, the Court noted that the prices were discriminatory because they were related “to factors other than actual costs of production or delivery” (324 U.S. at p. 732, 65 S.Ct. at p. 964 emphasis supplied). The same is true here. In the automotive parts industry, discount sales to buying groups have repeatedly been condemned where the group buying does not create cost savings for the seller. Mid-South Distributors v. Federal Trade Commission, 287 F.2d 512, 517 (5th Cir. 1961), certiorari denied, 368 U.S. 838, 82 S.Ct. 36, 7 L.Ed.2d 39; American Motor Specialties Co. v. Federal Trade Commission, 278 F.2d 225, 228 (2d Cir. 1960), certiorari denied, 364 U.S. 884, 81 S.Ct. 169, 5 L.Ed.2d 105; Standard Motor Products, Inc. v. Federal Trade Commission, 265 F.2d 674, 675-676 (2d Cir. 1959), certiorari denied, 361 U.S. 826, 80 S.Ct. 73, 4 L.Ed.2d 69. Because milk is not centrally warehoused, the usual economies of scale associated with chain and co-op buying are absent here. The cost of selling a given volume of milk to one store is the same whether that store is part of a group or chain or an independent. Therefore, when the group or chain receives a greater discount than the independent, the seller is bowing to the size of the buyer and not making “due allowance” for cost savings. In Federal Trade Commission v. Standard Motor Products, Inc., 371 F.2d 613 (2d Cir. 1967), enforcement of a Commission order condemning an annual rebate system was denied in part because the Commission “should have analyzed the competitive injury resulting from Standard’s purchaser classification.” 371 F.2d at p. 621, n. 12. National here asks the Court to fault the Commission for making just such an analysis!

National also argues that abandonment of the averaging system will put it at a competitive disadvantage vis-a-vis intrastate dairies who will continue to grant average discounts to chains. If this happens, the net prices of the competing dairies will either be lower than National’s or they will not. If they are lower, the meeting c-mpetition defense may be available to protect National; if they are not lower, National will not really be at a competitive disadvantage.

Finally, National relies on a number of cases dealing with the adequacy of cost studies. 6 These are inapplicable because the Commission, reversing the examiner, approved the methodology and results of National’s cost study. It is true that if a time-study man has to ride every milk truck on every delivery, the cost justification defense will become a nullity. The Commission has not imposed such an unreasonable requirement here. In fact, it has treated National’s one-week study as a reliable barometer of market conditions. In rejecting the averaging system, the Commission has not impugned any of the data obtained by the study; rather, it has interpreted that data in the light of the purpose of the amended Clayton Act to protect independent merchants and in accord with United States v. Borden Co., 370 U.S. 460, 82 S.Ct. 1309, 8 L.Ed.2d 627.

Borden does permit the grouping together of reasonably homogenous units. For example, if a National cost study showed that a store using 3,000 units *527 justified a certain discount on a cost basis, under Borden National could award such a discount to all other stores similarly situated without doing a time study for each store.

The Commission found that pri- or to the adoption of the averaging schedule in 1960, National had a discount system in which the purchases of all the stores in the chain or group were aggregated. Under such a system a group of stores, none of which was individually entitled to any discount, could obtain the maximum discount by aggregating their purchases. An aggregating system bears less resemblance to the realities of cost savings than does an averaging system. National does not seriously dispute this. Rather, it argues that it did not in fact employ an aggregating system but granted the large discounts to meet competition (except as to Joseph’s, National Food and Wrigley). Because the evidence is in conflict on this question, we may not disturb the Commission’s finding that the prior discount system to chain and group purchasers was based on impermissible aggregating. National did not attempt to cost justify the aggregating practice.

National asserts that it has cost justified its discounts to 6 corporate chains (Big Bear, Joseph’s, National Foods, Kroger, Wrigley’s, and Seaway Foodtown (except for one store)) even on the store-by-store basis demanded by the Commission. 7 The Commission’s brief does not dispute this and our study of the record supports National’s portrayal. Nevertheless, National’s remaining unjustified discounts in this area adequately support the Commission’s competitive injury conclusions.

Memphis Area

In Memphis, 200 independent stores, representing 50% of National’s ice cream customers, received no discount. At the same time, National had the following discount schedule in effect for independent stores:

Gallons Discounts/Gallon

0 — 49 ................... 0

50— 79 ................... 20

80 — 109 ................... 30

110 — 139 ................... 40

140 and over................ 50

Under this schedule most independents received no discount. Chain and group stores were accorded a 70 per gallon off schedule discount.

The sale of ice cream in this area was highly competitive, and the independent stores receiving no discount were located near many of the chain and group members receiving the 70 per gallon discount, thus enabling them to sell ice cream at lower prices than the independents. Numerous chain and group member stores handled smaller ice cream volume than numerous independents and yet received the 70 discount solely due to their affiliation. Concomitantly, many independents received no discount while equally small affiliated stores received the 70 per gallon discount. We agree with the Commission that these facts show that National’s discounts had the probable effect of lessening competition at the secondary level. See Federal Trade Commission v. Morton Salt Co., 334 U.S. 37, 50-51, 68 S.Ct. 822, 92 L.Ed. 1196; United Biscuit Company of America v. Federal Trade Commission, 350 F.2d 615, 621 (7th Cir. 1965), certiorari denied, 383 U.S. 926, 86 S.Ct. 930, 15 L.Ed. 2d 845; Whitaker Cable Corporation v. Federal Trade Commission, 239 F.2d 253, 255 (7th Cir. 1956), certiorari denied, 353 U.S. 938, 77 S.Ct. 813, 1 L.Ed.2d 761. As with the Toledo-Monroe market, such a conclusion is not foreclosed by the availability of these discounts to National’s independent customers deciding to join a voluntary or cooperative group.

As in the Toledo-Monroe area, the Commission concluded that National’s *528 discount schedule was cost justified. However, to cost justify the 70 per gallon discount to chain and group stores, National used an averaging method even though the chain and group stores varied in size. Under United States v. Borden Co., 370 U.S. 460, 82 S.Ct. 1309, 8 L.Ed.2d 627, such a cost justification defense was properly rejected.

National also contends that its 70 per gallon discount granted to Malone & Hyde, Inc., a wholesale grocer sponsoring a voluntary chain of stores, was a legitimate functional discount. On the other hand, the Commission found that National sold to the M & H member stores rather than to M & H. The invoices stated that the ice cream was “sold to” the particular stores, not to M & H. In addition to the 70 per gallon discount, National granted M & H a 2% allowance for assuming the credit risk of the member stores. As the Commission noted, this allowance is consistent with the member stores themselves being the purchasers, for if M & H were the purchaser, National would have no risk as to the credit of the stores. The Commission permitted this 2%. allowance as a payment for services performed by M & H for National. It was not included in the price discrimination that the Commission found National gave to the M & H stores.

The Commission explained its conclusion that the M & H stores were the purchasers from National as follows (op. cit. p. 22,925):

“The circumstances are such that there is very little difference between the manner in which a Malone & Hyde member store is serviced by respondent as compared to a non-member retail customer. In both cases, respondent receives orders directly from the store, delivers the ice cream directly to the store and services the cabinets. Whereas an independent store owner may pay the deliveryman, the Malone & Hyde store is furnished with a copy of the invoice and the original is sent to Malone & Hyde headquarters. Once a week, respondent bills Malone & Hyde for all stores and receives a check drawn on the Malone & Hyde account.”

In our view, M & H is sufficiently unlike the ordinary wholesaler to support the Commission’s decision that the member stores were the actual purchasers. However, under Federal Trade Commission v. Fred Meyer, Inc., 390 U.S. 341, 88 S.Ct. 904, 19 L.Ed.2d 1222, the M & H retail stores could be classified as purchasers from National even if they technically purchase from M & H where, as here, such a classification would further the purposes of the Clayton Act, as amended. Cf. Federal Trade Commission v. Sun Oil Co., 371 U.S. 505, 512-523, 83 S.Ct. 358, 9 L.Ed.2d 466.

New Orleans Area, and Scope of Order

Here National sold a private label milk, “Velva,” to H. G. Hill Stores, Inc., a chain (and its successor, Winn-Dixie Stores, Inc.) at a net price of about 100 per gallon less than to many of National’s store customers receiving no discount. Besides the 20% discount to Hill, National sold milk to 5 other wholesale customers at 5-10% off its wholesale list price. The Commission sustained the defense of meeting competition with respect to all customers receiving discriminatory discounts except the Hill Stores.

To sustain the defense of meeting competition with respect to Hill in 1951, National relied on competitors’ bids to public institutions. However, the bids to such institutions were not subject to the Act and involved different delivery costs and therefore would not show what the competitors’ bids were to Hill.

To sustain the defense of meeting competition with respect to Hill in 1954, National asserts that Hill’s representative told National of a low offer from the Franklinton, Louisiana, Co-operative. The Commission disregarded the Frank-linton offer because National had not verified it 8 and because the cooperative’s *529 offer was suspect by virtue of National’s own cost study. Also, according to the Commission’s interpretation of the testimony of National official Lewis Robinson, National did not learn of the cooperative’s offer until making its bid. The Commission found that National's 1954 price to Hill was based on its own cost study. This conclusion was permissible from the evidence adduced.

National contends that it ceased selling milk to Hill’s successor in 1960 and that a 1958 Louisiana statute bans discounts on milk, so that the requisite competitive effects no longer exist under Section 2(a). But such a longtime discriminatory practice, if followed elsewhere by National, could similarly injure competition with customers receiving no equivalent discounts, so that the Commission was entitled to consider this evidence in framing its order. Federal Trade Commission v. Ruberoid Co., 343 U.S. 470, 72 S.Ct. 800, 96 L.Ed. 1081; Foremost Dairies, Inc. v. Federal Trade Commission, 348 F.2d 674, 681-682 (5th Cir. 1965), certiorari denied, 382 U.S. 959, 86 S.Ct. 435, 15 L.Ed.2d 362. Even though National has stopped granting unjustified discounts to Winn-Dixie in New Orleans, its similar practices in Toledo-Monroe and Memphis, as well as its former practice in New Orleans, justified the nation-wide order entered. In contrast to Dean Milk Co. v. Federal Trade Commission, 395 F.2d 696 (7th Cir. 1968), there are no considerations present here that impel a limitation of the Commission’s order to the specific areas where the price discriminations were proved. National operates in 35 states and it is concededly typical for Sealtest to have volume discount schedules in effect at its various operations. The practices condemned by the Commission occurred in widespread areas over a long period of years. Consequently we cannot say that this order was so broad as to constitute an abuse of discretion. See Federal Trade Commission v. National. Lead Co., 352 U.S. 419, 428, 431, 77 S.Ct. 502, 1 L.Ed.2d 438; Swift & Company v. United States, 393 F.2d 247.256 (7th Cir. 1968), and Lloyd A. Fry Roofing Co. v. Federal Trade Commission, 371 F.2d 277,. 284, 286 (7th Cir. 1966).

We have considered other points raised by the parties but they merit no discussion.

The order is affirmed and will be enforced.

. A second portion of the order concerned National’s violation of Section 2(d). National did not appeal to the Commission from that portion of the order, but in this Court, National objected to the entire order on the ground that it was supported by only two of the Commission’s five members. In view of Federal Trade Commission v. Flotill Products, 389 U.S. 179, 88 S.Ct. 401, 19 L.Ed.2d 398, this point has been abandoned.

. Averaging the cost of serving chain stores was not condemned by the Supreme Court until 1962. United States v. Borden Co., 370 U.S. 460, 82 S.Ct. 1309, 8 L.Ed.2d 627.

. Under Sunshine Biscuits, Inc. v. Federal Trade Commission, 306 F.2d 48 (7tli Cir. 1962), Meadowgold could obtain new customers by meeting competitive prices selectively.

. One Kroger store took only 7,864 points a month and was cost justified up to an 11.47% bracket. Being only 1 of Kroger’s 22 stores and almost qualifying for the 12% discount, this store should be permitted within the averaging. One Seaway Foodtown store took only 5,891 monthly points and was cost justified only up to 9.18%. As to such a borderline store, any injurious effect of the discount would seem so remote that this store could also be considered as sufficiently homogenous for averaging.

. American Motors Corp. v. Federal Trade Commission, 384 F.2d 247 (6th Cir. 1967), certiorari denied, 390 U.S. 1012, 88 S.Ct. 1260, 20 L.Ed.2d 164, does not support National’s use of the average cost of serving the members of the chain and voluntary groups. As the court there observed, American Motors’ cost report “Did not average the volume of sales or the lesser cost of selling to the respective merchandising distributors” (384 F.2d at p. 256). It was chiefly on this basis that American Motors distinguished Borden.

. E. g., American Can Co. v. Russellville Canning Co., 191 F.2d 38 (8th Cir. 1951); Reid v. Harper & Bros., 235 F.2d 420 (2d Cir. 1956), certiorari denied, 352 U. S. 952, 77 S.Ct. 326, 1 L.Ed.2d 242; Federal Trade Commission v. Standard Motor Products, Inc., 371 F.2d 613 (2d Cir. 1967).

. The Commission did hold that National’s regular discount schedule of 0% to 6% to independent stores was cost justified on a store-by-store basis.

. Efforts to verify were present in Forster Manufacturing Co. v. Federal Trade Commission, 335 F.2d 47, 54 (1st Cir. 1964), certiorari denied, 380 U.S. 906, *529 85 S.Ct. 887, 13 L.Ed.2d 794; Beatrice Foods Co., Trade Reg.Rep. (65-67 Transfer Binder) If 17,311, at p. 22,470 (1965), and Continental Baking Co., Trade Reg. Rep. 63-65 Transfer Binder) If 16,720, at p. 21,648 (1963), on which National relies.