NEAHER, District Judge.

This action for an injunction against alleged infringement of plaintiff's trademark "TEFLON" was tried by the court upon the facts. Jurisdiction is grounded upon 15 U.S.C. § 1121 and 28 U.S.C. § 1332 and 1338. The facts and legal discussion which follow constitute the court's findings and conclusions as required by Rule 52, F.R.Civ.P.

I.

Plaintiff, E.I. Du Pont de Nemours and Company ("DuPont") is a Delaware corporation having its principal place of business in Wilmington, Delaware. Since 1946 DuPont has been the registered owner of the trademark TEFLON and holder of eight United States registrations of that trademark granted between 1946 and 1967. DuPont is also the owner and holder of seven registrations, granted between 1964 and 1970, for a number of certification marks of fanciful design embodying TEFLON, TEFLON-S or TEFLON II. All of these registrations are still in full force and effect and DuPont has given the required notice thereof to the trade and to the public. 15 U.S.C. § 1111.

Defendant Yoshida International, Inc., now known as YKK Zipper (U.S.A.), Inc., is a New York corporation having its principal place of business in Lyndhurst, New Jersey. It is a wholly-owned subsidiary of defendant Yoshida Kogyo K.K., a Japanese corporation having its principal place of business in Tokyo, Japan. Both defendants will hereinafter be referred to collectively as "YKK", a designation used by them in their business of manufacturing, importing and selling zippers in the United States and other countries. YKK has manufactured zippers for many years and is believed to be the world's largest manufacturer of such products.

The controversy between the parties culminating in this litigation dates back to 1969. In that year, YKK first introduced to the American market a zipper made of nylon, which it called the "EFLON" zipper. When YKK thereafter sought to register the name EFLON in the United States Patent Office, DuPont promptly filed opposition proceedings, viewing EFLON as an infringement and a source of confusion with DuPont's TEFLON. Those proceedings were suspended, however, when the present suit was instituted in January 1971.

Origin and Use of TEFLON

The name TEFLON was created by DuPont prior to 1946. It is a coined or invented term having no meaning in the English language except as a trademark denoting DuPont resins made from a chemical substance known as polytetra-fluoroethylene. That substance, a resinous solid material (hereinafter "TFE resin"), since its discovery by DuPont, has given rise to a family of chemical products. Continuously since 1944, DuPont has used TEFLON as a trademark in selling TFE dispersions and granular resins to a wide variety of industrial users and also within the DuPont Company for further processing and sale as non-stick TFE finishes for cookware and other products.

Since the end of World War II DuPont's accumulated sales of TEFLON products have amounted to approximately $800,000,000. Sales of TEFLON TFE resins to industrial users, the major portion of which are to buyers outside the company, account for more than 50% of the total TEFLON business. With one exception, DuPont itself does not sell products containing TEFLON directly to the consumer trade. That exception is "Lucite" paint containing TEFLON-E additives. DuPont has, however, in the various ways described below, exerted considerable effort to bring the TEFLON name and trademark to the attention of the consuming public as well as industrial users. Aside from TEFLON labeling on packaging and containers, DuPont's total advertising expenditures on its TEFLON products have been in excess of $32,000,000 over the past 25 years; and since 1965 have exceeded $3,000,000 annually in all but one year.

Although DuPont's TEFLON advertising aimed at the industrial market is confined to industrial trade magazines and publications, chemical journals and electronic magazines, most of the industrial buyers themselves advertise their products to consumers in association with the TEFLON trademark and have been doing so for years. From the standpoint of consumer recognition, TEFLON is most widely known in the cookware market in connection with non-stick coating on cooking utensils.

To promote consumer interest in non-stick cooking utensils, DuPont, in the early 1960's, inaugurated the first of its TEFLON certification programs, which included the development of technical standards for applying TFE coatings to cookware. Through extensive television advertising by DuPont, in which the TEFLON certification seal played an integral part, the consuming public was informed that the non-stick finish on cookware products was a "DuPont approved finish" which had been applied in accordance with DuPont standards. Although not required to do so, many cookware manufacturers have participated in the TEFLON certification program and have themselves advertised their TFE-coated products under the TEFLON mark. Since approximately 1963, shortly after the first TEFLON certification program, sales of TEFLON-coated cookware have accounted for between 35% and 50% of the total cookware market.

Beginning in 1965 DuPont began to expand the range of possible applications for TFE resins and finishes in industrial and consumer products. Its second TEFLON certification program, TEFLON II, followed upon improvements in the art of applying TFE to a metal surface so that the finished product became more scratch-resistant. A subsequent certification program featuring the TEFLON-S seal was based upon the application of TFE to produce a harder but less heat-resistant surface for use in the hardware field on such consumer items as saws, pruning shears, snow shovels and the like.

DuPont, of course, does not stand alone in the manufacture, promotion and sale of TFE resins and finishes. Two principal competitors in this field are ICI America and Allied Chemical, which also advertise and sell a broad range of TFE products under their own brand names. Like DuPont, each has its own trademark, ICI selling its products under the name "FLUON" and Allied under the trademark "HALON."

In order to maintain the recognition and value of its TEFLON trademarks and seals, DuPont has over the years conducted a vigilant trademark education and protection program. Since most of the industrial buyers of DuPont's TFE resins use the TEFLON mark in advertising their own products, such customers receive guidance on correct trademark usage, not only with respect to publication and display advertising, but also regarding references to TEFLON in general business correspondence. DuPont salesmen and others who deal with buyers of TEFLON resins, finishes and fibers are instructed concerning trademark matters; a DuPont publication (Pl.Exhs. 5 and 6) containing instructions on proper usage of the TEFLON trademark has been widely distributed among customers in connection with such efforts.

DuPont's trademark protection program also includes extensive surveillance efforts by its legal and advertising departments as well as the outside advertising agency responsible for advertising TEFLON products. Whenever misuse of the mark is detected, it is promptly called to the misuser's attention. In the cookware trade particularly, DuPont's advertising agency, N.W. Ayer Son, Inc., has maintained continuous watch over TEFLON usage in cookware advertising and has from time to time inserted protective trademark advertising in retailer-oriented publications such as Women's Wear Daily, Home Furnishings Daily, Discount Store News and similar journals, aimed principally at buyers of houseware and home furnishings. The main thrust of these is to impart to the general public the understanding that TEFLON symbolizes DuPont's non-stick finish.

Origin and Use of EFLON

YKK advertises and sells four types of zippers in the United States: a zipper made from DuPont's DELRIN acetal resin; a polyester zipper called "CONCEAL"; a zipper designed for sports clothing called "ZIPLON"; and the nylon zipper called "EFLON", whose tradename is the subject of this action. The component parts of the EFLON zipper are manufactured at YKK's production facility in Japan, the construction having been licensed by YKK from West German companies. YKK has facilities here for assembling the component parts and has plans to manufacture the entire zipper in the United States.

The name EFLON is also a coined fanciful term having no meaning in the English language. It seems to have been conceived in a somewhat ambivalent manner. According to one explanation, the name was in part derived from YKK's manufacturing practice of designating each type of zipper produced by consecutive alphabetical letters. Thus the first YKK zipper, made of aluminum, was coded "A F" ("F" designating "fastener"). A subsequent brass zipper was designated "B F", and so on. The nylon zipper being the fifth in the YKK line was designated "E F".

Another explanation was that prior to commercial production, when a name was being sought which might indicate the nylon zipper's usefulness in women's garments, the name "Elegant Fastener" was suggested, which happened to coincide with the acronym "E F". Thereafter, at a meeting of YKK directors on July 4, 1967, one of them suggested the addition of "lon", already being used by YKK in its trademark "Ziplon", to form "Eflon", which name was then adopted.

Despite early efforts by DuPont Far East representatives to dissuade YKK from registering EFLON as a trademark because of the similarity to TEFLON (Pl.Exhs. 8-10; Def.Exhs. A, B), YKK, beginning in January 1968 and up to September 3, 1972, has obtained such registrations in some 43 countries other than the United States. Def.Exh. S. Although DuPont initially considered opposing EFLON registration, it decided to take no action to prevent it in clothing accessories classes, on the advice of foreign counsel that the different product classes involved precluded successful opposition under foreign trademark laws. Pl.Exhs. 11-15; Def.Exhs. A, B.

In 1969, when YKK introduced the EFLON zipper in the United States, it was sold only to garment makers and not at retail. Sales in that year amounted to $191,433. In 1970 sales were again to garment makers only and amounted to $745,814. In 1971, YKK expanded its EFLON zipper business into the retail home sewing market. An exclusive distribution agreement was signed in the fall of 1971 with Belding Heminway Company, Inc., a retail distributor of articles in the notions and sewing accessories field. As YKK's exclusive retail distributor, Belding and its affiliate, Lily Mills Company, began to advertise and sell the EFLON zipper to department stores, discount houses, fabric and notion shops and the like, all of which sell to women who purchase zippers for sewing at home.

Lily Mills packages the zippers both for itself and Belding. Each zipper is singly packaged in a slim cardboard sheath with a glassine window. Those used by Belding bear the name "Belding Corticelli", beneath which are the words "EFLON Coil." Below the glassine window appears the legend "Superior Quality Nylon Zipper Eflon" in relatively large script. Def.Exh. K. The Lily package varies slightly in legend. Following the name "Lily" are the words "Eflon Coil Zipper." Below the glassine window appear the words "the flexible Eflon zipper of superior quality", the word "Eflon" being printed in script. Def.Exhs. L, M. "EFLON" is also imprinted in tiny letters in the pull tab of each zipper slide.

EFLON zippers sold to garment makers are not individually packaged. "EFLON" is imprinted as above described only on the pull tab of the zipper slide. Def.Exhs. D through H. YKK, however, will on request provide garment makers with hang tags for the zipper. These tags prominently bear on their face the legend "YKK The World's Largest Zipper Company EFLON ZIPPER." Pl.Exh. 52; Def.Exhs. I, J.

In 1971 YKK's sales of EFLON zippers to garment makers amounted to $1,196,519 and to Belding, $218,990, representing 85% and 15%, respectively, of the total of such zipper sales in the United States for that year. In 1972 sales increased to $2,254,360 (79%) for garment makers and $690,597 (21%) for Belding. In the first quarter of 1973 sales to garment makers amounted to $928,689 (91%) and to Belding $95,355 (9%).

Although the foregoing sales figures are indicative of the EFLON zipper's greater use among sophisticated garment manufacturers, YKK's zipper advertising has been unmistakably pitched to the retail garment trade and the consumer. In addition to trade advertising in Women's Wear Daily and the Daily News Record, extensive billboard advertising and full-page advertisements in The New York Times Magazine have been utilized for that purpose. Pl.Exhs. 38, 39. The advertisements also refer to EFLON as the zipper that is "slippery, slide-y" and "runs smoothly" (Pl.Exhs. 56, 57), characteristics which cannot be said to be wholly incapable of association with the widely advertised "no stick" attributes of DuPont's TEFLON. YKK's advertising, moreover, conveys the added impression that EFLON zippers, like "Zippers of Delrin" — another acknowledged DuPont trademarked product mentioned in the same ads — contain or are manufactured of a substance called EFLON. The Claims in Contention

DuPont concedes it has no evidence of actual confusion and the court finds no evidence of actual competition between the parties. Nevertheless, DuPont contends it has made a sufficiently strong showing of a likelihood of confusion among ordinary purchasers as to the source of origin of defendants' zipper. It emphasizes the virtual identity in sound and appearance of TEFLON and EFLON, the peculiar circumstances of defendants' adoption of EFLON and the tenor of their advertising, the functional interrelatedness of the non-stick quality of the respective products, and the substantial possibility that DuPont TFE resins will be used as lubricants or coatings in the construction of zippers or their components.

Defendants, denying any infringement or bad faith in their adoption and use of EFLON, stress the dissimilarity of the products and the customers, and the absence of evidence of either actual confusion or the likelihood thereof. They also argue that TEFLON is at best a weak mark which has in fact become generic and not entitled to trademark protection in any event.

The starting point for an evaluation of these claims is the observation that there is no evidence or suggestion in the record that DuPont and YKK are competitors in any product relevant to this case. The court is thus faced with the question of the scope of protection of a trademark against allegedly confusingly similar use on dissimilar products — a question which "does not become easier of solution with the years." King Research, Inc. v. Shulton, Inc., 454 F.2d 66 (2 Cir. 1972), quoting Polaroid Corp. v. Polarad Electronics Corp., 287 F.2d 492, 495 (2 Cir.), cert. denied, 368 U.S. 820, 82 S.Ct. 36, 7 L.Ed.2d 25 (1961). To the discussion of that question we now turn.

II.

The gravamen of an action for trademark infringement where the parties are not competitors is that the defendant's use of a mark similar to plaintiff's is likely to cause confusion, or mistake, or deception of purchasers as to the source of origin of defendant's goods or services. 15 U.S.C. § 1114(1)(a); King Research, Inc. v. Shulton, Inc., 324 F. Supp. 631, 636 (S.D.N Y 1971), aff'd, 454 F.2d 66 (2 Cir. 1972); see Societe Comptoir De L'Industrie Cotonniere Etablissements Boussac v. Alexander's Department Stores, Inc., 299 F.2d 33, 36 (2 Cir. 1962); Avon Shoe Co. v. David Crystal, Inc., 279 F.2d 607, 612 (2 Cir.), cert. denied, 364 U.S. 909, 81 S.Ct. 271, 5 L.Ed.2d 224 (1960).

It should be noted at the outset that because of the speculative nature of any effort to weigh the probabilities arising from confusing similarity and the unique set of variables each case presents, strict general rules and precedents are of little value. But there are some general principles that do transcend the particulars. In weighing the evidence of likelihood of confusion, the court must strive to place itself in the shoes of a prospective purchaser. In this role, the court does not act as an enlightened educator of the public but takes into account the mythical ordinary prospective purchaser's capacity to discriminate as well as his propensity for carelessness. Accordingly, an overly analytical approach with close attention to specific differences is less important than the overall impression of general similarity. See 3 R. Calmann, Unfair Competition Trademarks and Monopolies § 81.1 (3d ed. 1969) (hereinafter Calmann); 1 A. Seidel, S. Dubroff and E. Gonda, Trademark Law and Practice § 16.01 (1963) (hereinafter Seidel). Finally, where, as is often enough the case, evaluation of all pertinent factors still leaves the matter in some doubt, such doubt should be resolved against the newcomer. J.R. Wood and Sons, Inc. v. Reese Jewelry Corporation, 278 F.2d 157, 160 (2 Cir. 1960) (dissenting opinion); Seidel, supra, § 16.03. The rationale for this rule was aptly stated by Judge Friendly in the Wood case:

"When middle-aged judges are obliged to determine the `likelihood of confusion' in the purchase of engagement and wedding rings by youthful swains not enjoying our advantage of knowing the answer in advance, I should suppose the most resolute mind must entertain some doubts. I prefer to resolve mine in favor of a plaintiff who has spent money and effort in exploiting its mark for nearly a score of years rather than of a defendant who, with the world of possible names before him, has chosen to inch as close to the plaintiff's mark as he believes he safely can, even if he has done this in a `good faith' belief that he has succeeded." 278 F.2d 157 at 160.

Turning to the factors to be considered in assessing a prior owner's claims that a non-competitor's use of a mark constitutes an infringement, we again rely upon Judge Friendly's oft-cited listing of evaluative criteria in Polaroid Corp., supra, 287 F.2d at 495, which has provided the standard method of approach in this Circuit. As he pointed out:

"Where the products are different, the prior owner's chance of success is a function of many variables: the strength of his mark, the degree of similarity between the two marks, the proximity of the products, the likelihood that the prior owner will bridge the gap, actual confusion, and the reciprocal of defendant's good faith in adopting its own mark, the quality of defendant's product, and the sophistication of the buyers. Even this extensive catalogue does not exhaust the possibilities — the court may have to take still other variables into account."

In applying these factors here, however, the court must keep in mind that they

"are variable and relative and no single one, because of its presence or absence, is, in itself, determinative of a case. Rather, the method of approach requires the trial court to consider and weigh the evidence relative to each of these points and such other points as, in the particular circumstances before it, the court finds applicable; then, from a balancing of the conclusions reached on all of these factors, the court decides whether or not the parties are entitled to the relief or protection sought." Kiki Undies Corp., supra, 411 F.2d at 1099.

1. Degree of Similarity

In this case, TEFLON and EFLON are both simple, single words. There is no evidence to suggest that they are used with any special or unique stylistic adornment, so there can be no doubt that EFLON, formed by elimination of the first letter of TEFLON, leaves the two words very similar in sound and appearance. Since, as has been noted, both words are coined, there is no question of similarity of meaning. It is generally thought that use of similar coined words renders confusion more likely. Calmann, supra, § 82.1(a) at 604 n. 78.

To the extent that "flon" or "lon" might be thought to be a free or common suffix, DuPont's use of it cannot preempt the field nor exclude all others from their use. Defendants make much of this point with Exh. R, which contains no fewer than 19 trademark registrations ending in "flon" (with "TUFFLON" at the top of their list) and 98 others ending in "lon." A good many of these reflect commercial adaptation of various synthetic chemical compounds.

DuPont does not directly dispute YKK's contention that "flon" is a common suffix in the plastics field. It is unnecessary, however, to be drawn into an overly analytical discussion of whether the descriptive or fanciful syllable in TEFLON is dominant, see Calmann, supra § 82.1(g)(2) and (3). The court agrees with the general rule that even where a common prefix, suffix or other portion of a prior mark is used, it is the duty of the newcomer to see that the remainder of the mark is sufficiently different from the prior mark that there will be no likelihood of confusion between the marks when considered as a whole. Seidel, supra, § 16.04[5] at 354. See, e.g., W.E. Kautenberg Co. v. Ekco Products Company, 251 F.2d 628, 45 C.C.P.A. 761 (1958) (sustaining opposition by "EKCO", for kitchen utensils, of registration of "WEKCO" for mops and mophandles).

On the other hand, even the close similarity between EFLON and TEFLON is hardly dispositive on the issue of likelihood of confusion. The court's review of cases involving the alteration, addition or elimination of only a single letter from the old mark to the new reveals frequently divergent results, no doubt attributable to the many other factors which must also be considered in each case. See numerous examples listed in Calmann, supra, § 82.1(a).

2. Strength of the TEFLON Mark

The relative terms "strong" and "weak" have frequently been used to characterize trademarks, and it has often been said that when the goods of the parties are not in competition, the strong mark will be protected whereas the weak one will not. W.E. Kautenberg Co., supra, 251 F.2d at 631; Calmann, supra, § 82.1(1) at 757-58; Seidel, supra, § 16.01 at 318. Strength or weakness is primarily a question of assessment of a mark's distinctiveness or popularity. Where the public has been educated to recognize and accept a particular mark as the hallmark for a particular source of that product, or the mark itself is inherently unique or has been the subject of wide advertisement, it is a strong trademark. Similarly, coined or fanciful marks are generally thought to have a much more unique or distinctive appeal than words in common use, and are frequently found to be strong marks deserving a correspondingly broader degree of protection. Arrow Distilleries v. Globe Brewing Co., 117 F.2d 347, 351 (4 Cir. 1941); Intercontinental Mfg. Co. v. Continental Motors Corporation, 230 F.2d 621, 623, 43 CCPA 841 (1956); Calmann, supra, § 82.1(e) at 756; Seidel, supra, § 16.01 at 318; Schechter, The Rational Basis of Trade Mark Protection, 40 Harv.L.Rev. 813, 829 (1927).

Apart from these general rules, strength is primarily a question of degree, an amorphous concept with little shape or substance when divorced from the mark's commercial context, including an appraisal of the owner's policing efforts to ensure that whatever distinctiveness or exclusivity has been achieved is not lost through neglect, inattention, or consent to infringing use. See Calmann, supra, § 82.1(1) at 761.

Applying these principles to the facts as found above, the court concludes that TEFLON is a strong enough mark so that this factor weighs considerably in DuPont's favor. Defendants dispute such a conclusion by arguing essentially that few consumers are aware DuPont is the manufacturer of TEFLON and that the common suffix "flon" precludes a finding that the mark is coined or fanciful.

The court must agree that defendants' evidence does show the "flon" and "lon" suffixes are frequently used on products not dissimilar to TFE resins. But defendants have not established that such suffixes have acquired any independent meaning in the English language nor have they demonstrated that, when considered as a whole, TEFLON inevitably brings to mind this related group of products. Thus while the mark cannot be considered wholly arbitrary or fanciful, neither can it be thought to be the equivalent of a word in common use.

As to defendants' other point, it is the distinctiveness of the mark itself, serving as an indicator of a single source of origin, which is important, and not the extent of correlation in the public mind between the goods and their manufacturer or source of origin. The court must therefore agree with plaintiff's observation that a mark's strength is not diminished because prospective purchasers are unaware of the manufacturer's identity.

3. Sophistication of the Buyers

The parties do not agree on the identity of the measuring group for analysis of the likelihood of confusion among prospective purchasers. This is because EFLON zipper purchasers are of two distinct types: (1) the discriminating or expert purchaser, such as the purchasing agents of various garment manufacturers interviewed by defendants in connection with this case; and (2) the casual or ordinary purchaser — a home sewer, i.e., a homemaker who buys a zipper to be sewn into a garment at home. Defendants argue that, on the facts, the latter group is but a small minority of purchasers and therefore the former group should be the appropriate measuring group. Their position is underscored by defendants' survey evidence which strongly tends to show that not only is there no evidence of actual confusion among the more discriminating group, but also that there is virtually no likelihood of confusion among them. Def.Exh. O, Reactions to the EFLON Zipper by Garment Industry Purchasing Executives.

But this evidence is to be expected and is beside the point in any event. The court agrees with DuPont's contention that the home sewers are the appropriate measuring group on the well-settled principle that when "prospective purchasers are in part discriminating and in part casual the latter group should be the measuring rod." Calmann, supra, § 81.2 at 579. Indeed, defendants may not shield themselves from responsibility for any dilution of plaintiff's trademark in one sector of the public by confining the inquiry to another.

Moreover, defendants' de minimus argument ignores their own proof that, when measured on an annual basis, the retail share of the EFLON zipper market has been expanding more rapidly than the wholesale share, and by the end of 1972 was already 21% of the EFLON market. It also ignores the consumer-oriented thrust of defendants' advertising and the possibility that there might be a substantial change in defendants' sales pattern which would increase consumer exposure of EFLON and the harm to plaintiff. Under these circumstances, the court finds the appropriate measuring group to be the casual purchasers, a group which, it has often been said,

"includes the ignorant, the unthinking and the credulous, who, in making purchases, do not stop to analyze, but are governed by appearances and general impressions."

To that should be added the caveat that the "purchasing public must be credited with at least a modicum of intelligence," or "with a minimum capacity for discrimination." The controlling objective standard of consumer prudence is the ordinary purchaser's general impression when "buying under the normally prevalent conditions of the market and giving the attention such purchasers usually give in buying that class of goods. . . ." Calmann, supra, § 81.2 at 577. Application of this standard is discussed in section 7 infra.

4. Actual Confusion

The parties are in agreement that there was no evidence of actual confusion on the part of either of the two principal classes of purchasers of EFLON zippers, although, as defendants point out, YKK has been selling its EFLON zipper in the United States for close to five years. Plaintiff argues that evidence of actual confusion is not essential to a finding of infringement, while defendants point to cases where its absence was considered strong evidence that there is no likelihood of confusion, and hence, no infringement. But as noted above, actual confusion or its absence is only one factor in the analysis of likelihood of confusion, and cases subsequent to Polaroid Corp., supra, have found its absence to be either important or insignificant, depending on the evaluation of all the other factors. The significance of this factor will therefore have to be judged in light of the evidence with respect to the other factors.

5. Quality of EFLON Zippers

There is no doubt that the EFLON zipper is a high quality product. Although the zipper is made of chemically derived nylon monofilament coil, this case involves no claims of disparagement or passing off of defendants' goods for plaintiff's similar goods — neither DuPont nor any of its users of TEFLON certification marks make zippers. One who produces a high quality product, especially when dissimilar to an established one, can ordinarily be expected to trade on the good will thereby engendered, and is a less likely candidate for reaping benefit from confusion between the two sources of origin.

However, a high quality product is not necessarily an expensive one. As has been suggested above, the care with which a consumer may approach a purchase can be expected to differ markedly from case to case depending upon a number of highly variable factors, including the purchasing environment and the price of the article. In this case, defendants' goods are relatively common and inexpensive, despite their high quality. Courts have frequently recognized that this factor may warrant the conclusion that no particularly significant amount of thought or care would enter the purchaser's mind, and hence increase the likelihood of confusion. See Blue Bell, Inc., supra, 497 F.2d at 435-36 n. 5; Callmann, supra, § 81.2 at 580 n. 42.

Accordingly, the court concludes that while the quality of defendants' product may preclude a finding of disparagement or passing off, and militate against a finding of intent to deceive or confuse, the nature and advertised characteristics of the product are such as to increase the likelihood of confusion.

6. Defendants' Intent in Adopting the EFLON Mark

This factor arises out of the natural inference to be drawn from imitation of a prior mark. On the assumption that a businessman will ordinarily act to his commercial advantage, and that the attraction of an established business' good will to the newcomer's product is such an advantage, the inference to be drawn from imitation is the imitator's own expectation of confusion as to the source of origin of his product. See Time, Inc. v. Ultem Publications, 96 F.2d 164, 165 (2 Cir. 1938); Callmann, supra, § 82.2(b)(1). Where, as here, there is little to distinguish the marks themselves and the prior mark is a long-established one of which the newcomer was aware, doubts about intent are resolved against the newcomer, and a reasonable explanation of its choice is essential to establish lack of intent to deceive, especially where the prior mark is a coined or fanciful one. See Kiki Undies Corp., supra, 411 F.2d at 1101; Callmann, supra, § 82.2(b)(2).

As has been indicated, defendants' proffered explanations for their adoption of a name comprising five-sixths of TEFLON is hardly satisfying in light of the evidence that they were aware of DuPont's objection based upon the similarity of the names at the time defendants first registered EFLON in 1968. While plantiff has not materially contradicted defendants' explanation, it points to several factors raising serious questions as to its credibility. These are (1) defendants' failure to explain why English language words were used to express the creativity of a Japanese company; (2) evidence which shows that none of the other zippers introduced prior to EFLON bears either the alphabetical designation or an acronym for its own attributes or construction characteristics; and (3) the inadequacy of the "elegant fastener" description as being neither implicit in the word "EFLON" nor representative of the EFLON zipper's real consumer appeal. In fact, defendants' own advertising reveals that one of the principal selling points of the EFLON zipper is its "slippery", "slidey", "runs smoothly" feature. Pl.Exhs. 56, 58. And, as already noted, YKK's advertisements also serve to enhance rather than diminish the likelihood of consumer confusion in suggesting that the zipper is made of a substance known as EFLON rather than nylon. Pl.Exhs. 38, 39, 54, 55, 58 and 59.

Taking into account all the evidence on this factor, the court is not persuaded that defendants have come forward with a sufficiently credible explanation of their intent in adopting a name so similar to TEFLON.

7. Proximity of the Products

Likelihood of confusion must take into account the effect of the circumstances under which the products are sold and the nature of the products themselves. There are few general principles, but an impression of common origin may arise from any of a number of factors which might point to a similarity of product or purchasing environment.

Before analyzing the evidence on that point, however, the court is confronted with a rather unique equation. Given that defendants' "product" is zippers, what is plaintiff's "product" for purposes of gauging proximity? Or stated another way, in estimating the likelihood of confusion of source of origin of the zippers, what "source" are we talking about — the source of the TFE chemical substances (DuPont), or the source of varied consumer products which incorporate DuPont's TFE compounds in them (e.g., manufacturers of pots and pans which use the TEFLON certification mark)? As already noted, virtually all consumer purchases of TEFLON-marked products are from manufacturers or sellers other than DuPont. Therefore, the purchasing environments are totally dissimilar unless the court may consider the similarity of the zipper purchasing environment to that for the consumer products incorporating TEFLON under plaintiff's certification mark program.

Whether the owner of a certification mark may claim protection from infringement in this manner is a question on which prior cases provide little guidance. There is no doubt plaintiff seeks in its complaint to act as representative for its "customers" on the infringement cause of action, especially for those who have used its various TEFLON certification marks. Compl. ¶¶ 9, 19.

At the outset, it may be noted that certification marks are generally subject to the same provisions as trademarks, 15 U.S.C. § 1054; In re Professional Photographers of Ohio, Inc., 149 U.S.P.Q. 857, 859 (Pat.Off.App. 1966). The leading case in this circuit, Community of Roquefort v. William Faehndrich, Inc., 303 F.2d 494 (2 Cir. 1962), upheld the grant of summary judgment to the owner of the geographical name and certification mark "Roquefort" for a certain method of production of sheep's milk cheese in the natural caves of Roquefort, Aveyron, France. Id. at 496. In so doing, the court recognized the right of a certification mark owner to prosecute an action for infringement of the mark by defendant, a New York cheese importer. Id. at 498. More specifically, other authorities recognize that the proximity of products need not be measured against the product of the certification mark owner — which in many cases may be nonexistent, as the owner may not itself use the certification mark. Instead, proximity may be measured against that of the certification mark user, for whom the owner acts as a representative in an infringement action. See Jos. S. Cohen Sons Co. v. Hearst Magazines, 220 F.2d 763, 764-65, 42 CCPA 836 (1955); Callmann, supra, § 80.3 at 554. But cf. Florists Telegraph Delivery Ass'n v. Amling's of California, 174 F.2d 142, 36 C.C.P.A. 1071 (1949) (collective marks — discussed and criticized in Callmann, supra).

In this case neither side has made much of a point of the similarity or dissimilarity of purchasing environments as distinguished from the dissimilarity of products. What DuPont stresses is the pattern of expansion of TEFLON TFE applications to highly diversified articles which become more desirable and marketable with a "non-stick" quality, and evidence suggesting that a zipper is just such an item. From this DuPont argues that "it is the similarity in those characteristics which consumers find desirable in the two products which is of controlling significance and which here gives rise to the likelihood that defendants' EFLON zipper `might naturally be supposed to come from' the source responsible for TEFLON coatings", citing L.E. Waterman Co. v. Gordon, 72 F.2d 272 (2 Cir. 1934).

As noted above, YKK advertising of the EFLON zipper has indirectly alluded to the "non-stick" feature which is at the core of many of the TEFLON consumer applications (e.g., frypans). Moreover, the record is clear that sticking of nylon monofilament zippers has been a continuing problem in the zipper and garment industries, partly owing to destruction of the zipper's natural and synthetic lubricity in the dyeing process, and partly owing to structural damage caused by heat deformation when a garment incorporating the zipper has been subjected to ironing or dry-cleaning.

The court also notes again that, as discussed in section 5, supra, a zipper is a relatively inexpensive product whose purchase may be thought to involve no great amount of scrutiny by the consumer. Nor is it a frequently purchased product, at least in comparison to ordinary consumables such as food and other everyday non-durable consumer goods. This would also appear to increase the likelihood of confusion, Seidel, supra, § 16.08[3].

As the next section makes clear, there is no evidence of a present commercial retail use of TEFLON in zippers. See section 8, infra. In that strict sense, the EFLON zipper could not possibly be considered "proximate" to any existing retail product sold under the TEFLON certification mark program. But that is not the end of the inquiry, since "proximity" takes into account not only the nature of the products themselves, but the circumstances under which they are sold. Viewing all these marketplace components together, the court finds persuasive DuPont's argument that an ordinary consumer, confronted with a "smooth running" EFLON zipper, and perhaps dimly aware of the many varied applications of TEFLON to improve the lubricity of consumer products, might naturally and easily assume that the source of origin of the EFLON zipper was the same as or related to the source of TEFLON. That assumption would seem all the more natural when YKK's advertising for EFLON also refers to zippers made of DuPont's Delrin.

8. Bridging the Gap

This factor, closely related to the last, has been otherwise stated in this circuit as "the prospect that the [trademark] owner will want to extend his activities into the disputed area. . . ." Triumph Hosiery Mills, Inc., supra, 308 F.2d at 198. But while it is this prospect — among other rights incident to a trademark — that is protected under federal trademark law, the law is well established that in order to prevail the owner need not show he has in fact or is about to bridge the gap between his and the newcomer's presently noncompeting products. The ultimate test remains one of likelihood of confusion among prospective purchasers as to source of origin; the expansion of business factor turns not so much on objective evidence of actual expansion or potential therefor as it does on whether it is reasonable for the ordinary purchaser to assume such expansion. See Callmann, supra, § 82.2(c) at 808. And it has been said that such an assumption is all the more reasonable "the more a business has already expanded to include various other articles. . . ." Id. at 808-09.

Here the record shows that TEFLON has properties which ought to be useful in helping to solve zipper sticking problems. A single, apparently highly technical and non-commercial use of TEFLON for space suit zipper slides reduced pull friction by approximately 50%. Pl.Exh. 29. YKK and others in the zipper industry have also recognized the potential value of TEFLON as a chemical alternative or complement to design changes in the zipper to solve the friction problem, despite encountering some continuing difficulties in ease and economy of application. The Talon Division of Textron, for example, has made a commercial application of non-TEFLON TFE resin to a nylon zipper it manufactured. Additionally, Elmer's Slide-All, a Borden Company aerosol spray containing TEFLON, is sold as a product which "UNSTICKS: Drawers, windows, doors, zippers." Pl.Exh. 21. But there is no evidence of commercial use of TEFLON in zippers, although DuPont personnel have apparently long recommended its use as a lubricant to reduce friction.

Defendants have been quick to point out that DuPont neither now produces nor has plans to expand into the field of zipper production. If DuPont's past history is any guide, however, the record would support the conclusion that a reasonable possibility exists that the many uses of TEFLON in widely varied commercial applications may be expanded to include a TEFLON certification mark program for commercial zipper manufacturers. At the very least, the evidence is clear that it has not yet been established in the zipper industry that the many advantages a TEFLON coating might add to a zipper are either technically unobtainable or commercially unfeasible. And in such circumstances, it is likewise appropriate to find that the ordinary consumer, knowing little if anything of these details, might already have jumped to the reasonable conclusion that that gap has in fact been bridged by the EFLON zipper.

9. Public Reaction

To Judge Friendly's list the court would add this factor in this case, as extensive survey evidence on public reaction has been submitted by both sides. Such surveys might usually be thought to be evidence of actual confusion vel non, and considered in section 4, supra. But since plaintiff does not appear to dispute defendants' observation that there is no evidence of actual confusion, and also appears to have offered some of its survey evidence on the more general question of likelihood of confusion, it seems appropriate to consider separately the effect of public reaction to EFLON/TEFLON as reflected in those surveys.

It has been suggested that survey evidence may be the most practical manner of approaching the evaluation of public reaction. Calmann, supra, § 82.3(c) at 855. In evaluating such surveys, especially when they are conducted by reputable professionals, as they were in this case, it is generally thought that any technical deficiencies ought to go to the weight accorded them and not to their admissibility in evidence. See id. at 856 (collecting cases). In assessing that weight, however, a good principle to keep in mind is an observation of Judge Wyzanski's on survey evidence of likelihood of confusion:

"If the interviewee is not in a buying mood but is just in a friendly mood answering a pollster, his degree of attention is quite different from what it would be had he his wallet in his hand. Many men do not take the same trouble to avoid confusion when they are responding to sociological investigators as when they spend their cash."

DuPont's two surveys were conducted by Burke Marketing Research, Inc., a nationally-known professional research organization. Only the first, Pl.Exh. 35, Trademark Identification Study, March 20, 1972 (Burke Survey I), is relevant on the issue of likelihood of confusion. DuPont contends that Burke Survey I demonstrates that the strength of the TEFLON trademark is such that when a consumer is confronted with the EFLON mark it either reminds her of TEFLON or immediately brings to mind TEFLON's non-stick characteristics.

Defendants attack the survey on numerous technical and other grounds, the only substantial ones being their claims that (1) the responses obtained were both insufficient and "associational in nature", and (2) the interviews were not conducted in a purchasing environment.

Briefly stated and simplified, the Burke Survey I critical findings were based upon interviews in early 1972 with 801 women in their homes in 10 selected cities around the country. Before being asked about zipper purchases and other demographic information, each respondent who had done some home sewing was shown a card which read as follows:

"LAVORIS EFLON DRISTAN"

For each name the respondent was asked to "describe that product." These answers were recorded verbatim, but if no response were given, whether correct or incorrect, a second question would be asked, "What does it remind you of?", and in the case of EFLON, even a third opportunity to respond, "Well, what do you think it would be like?" The entire study was repeated twice more, using 1600 other women, the only change being in the words on the card.

At the first level of EFLON questioning (describe the product), only 6.6% of respondents made what might liberally be called a TEFLON brand name or product-related identification. At the second level of questioning, the figure jumped by an additional 26.5% of respondents, and at the third level a final 2.2% were added, for a total of 35.3%. On all three levels fully 25.3% gave no response, only 2.1% gave a zipper-related response, and no one identified EFLON as a brand name at any level.

In view of the results, it is understandable that DuPont does not offer Burke Survey I as evidence of actual confusion. Had it done so, the results would have been vulnerable to both of defendants' criticisms supra. The survey's failure to produce evidence of actual confusion, however, must be viewed in the light of its own findings concerning brand name recognition of EFLON at the time, and the court's findings concerning the relatively insignificant retail market position of the EFLON zipper in 1971, just before the survey was conducted.

Under all the circumstances, the court's view of Burke Survey I is that (1) while it appears to demonstrate little or no actual confusion, its failure in that respect does not compel an inference adverse to plaintiff; and (2) it reflects that the TEFLON mark or its non-stick features is fairly well known among prospective zipper purchasers and they are brought to mind in a substantial minority of those purchasers when they are confronted with the EFLON mark, at least in a non-purchasing environment. In this respect, and only in this respect, does the survey materially advance plaintiff's ultimate claim. But as defendants point out, the "fact that one mark may bring another mark to mind does not in itself establish likelihood of confusion as to source." Application of Ferrero, 479 F.2d 1395, 1397 (C.C.P.A. 1973).

10. Foreign Registrations of EFLON

Another factor in this case is the effect of YKK's EFLON trademark registration in over 40 foreign countries, and DuPont's alleged admissions regarding them. DuPont initially argues that foreign registrations should be completely disregarded, citing George W. Luft Co. v. Zande Cosmetic Co., 48 F. Supp. 602, 606 (S.D.N.Y. 1942). But this contention, and the complexities of the interrelationship of domestic and foreign laws in international commerce, may be entirely avoided. Defendants do not seek to prove by evidence of foreign EFLON registrations, or even by evidence of non-opposition by DuPont in over 30 of them, that there is no likelihood of confusion. Nor do they seek to advance thereby any of the imputative factors previously discussed. What they do argue, with the registrations as background therefor, is that DuPont's decision not to oppose was based upon its own conclusion and admission that there was no likelihood of confusion between TEFLON and EFLON.

The court's understanding of a somewhat complex situation is that when YKK first started obtaining foreign registrations for EFLON, DuPont considered opposition thereto, and many individuals in DuPont, representing the many disparate interests of a large, multifaceted corporation, expressed equally varied opinions about the legal and business considerations involved in opposing such registrations. One of those opinions, expressed by W. Alfred Smith, of the DuPont International Department, as part of a memorandum of August 22, 1968, was as follows:

"Even though EFLON is very close to our TEFLON trademark, it is the writer's opinion that no substantial confusion will result as long as EFLON is used to identify slide fasteners made from coiled nylon monofilament. Furthermore, since there is no assurance that we would be successful in opposing this mark, it seems preferable that no action having doubtful outcome be initiated that could upset unnecessarily the existing relationship with YKK. It behooves all concerned however to be alert to possible conflict and to new uses of the EFLON mark that might lead to confusion."

While this is certainly an arguably damaging admission by one of plaintiff's agents — and one on which defendants understandably seek to capitalize — DuPont suggests that the surrounding circumstances reveal that Smith was expressing only his own "opinion." The last paragraph of the Smith memorandum would, however, suggest the opposite conclusion.

But the conflict need not be resolved by more detailed clarification of the agency relationship, for the court is satisfied that under the circumstances in which the memorandum was written, the writer's conclusion was hardly the equivalent of a distillation of DuPont's evaluation of all the criteria that go into the esoteric legal concept of "likelihood of confusion." Particularly important is the fact that the memorandum substantially predated any use of the EFLON mark on zippers in the retail market in this country, which, as the foregoing discussion makes clear, is at the heart of this lawsuit. Thus, the court concludes that it has no significance to the issue of likelihood of confusion as it is raised in this case.

11. Impact of Equitable Relief

This last factor, the question of harm to YKK an injunction would cause, as measured against its benefit to DuPont, has not been particularly emphasized by either of the parties. But what does seem clear is that about two-thirds of DuPont's annual $3 million advertising budget for TEFLON products is aimed at the consumer under the TEFLON certification mark programs. Pl.Exh. 7. It also is fairly clear that YKK's advertising expenditures for the EFLON zipper in the United States from about the time of its introduction until the time of trial were slightly in excess of $200,000. Although, as already noted, YKK's advertising has been consumer-oriented, there is no evidence to suggest what proportion of its investment was expended directly for the retail market. Nor is there any evidence which establishes the approximate expense defendants would encounter in relabeling existing EFLON packaging or in re-educating the public or garment manufacturers about a new name for their nylon coil monofilament zipper.

Contrariwise, there is no evidence, in monetary terms, of the harm either retail or industrial use of the EFLON mark has or is likely to cause plaintiff. However, as the law of this circuit stands, one of the interests a trademark owner may seek to protect from harm by infringement is "the hazard that the owner's reputation may be tarnished by the use by another. . . ." Triumph Hosiery Mills, Inc., supra, 308 F.2d at 198; S.C. Johnson Son, supra, 175 F.2d at 180; L.E. Waterman Co., supra, 72 F.2d at 273-74. But there is also increasing judicial recognition that federal trademark law protects the trademark owner not only from injury to business reputation, but also from dilution of the distinctive quality of his mark. See Callmann, supra, §§ 80.3, 84.2. As Judge Learned Hand said long ago:

"[The trademark owner's] mark is his authentic seal; by it he vouches for the goods which bear it; it carries his name for good or ill. If another uses it, he borrows the owner's reputation, whose quality no longer lies within his own control. This is an injury, even though the borrower does not tarnish it, or divert any sales by its use; for a reputation, like a face, is the symbol of its possessor and creator, and another can use it only as a mask. And so it has come to be recognized that, unless the borrower's use is so foreign to the owner's as to insure against any identification of the two, it is unlawful." Yale Electric Corporation v. Robertson, 26 F.2d 972, 974 (2 Cir. 1928).

The likelihood of injury to DuPont's business reputation seems slight in view of the undisputed quality of the EFLON zipper. On the other hand, TEFLON is unquestionably a symbol of DuPont quality in the composition or desirable attributes of widely distributed and advertised varied products. The likelihood of loss of distinctiveness in the TEFLON mark from continued use of EFLON is, on this record, very difficult to quantify. But that very fact is implicitly recognized by the granting of injunctive relief in most cases where a trademark infringement is established. Certainly, if long prior use of the mark and substantially greater investment in its propagation and protection by DuPont, as well as the many DuPont customer interests involved, are placed on the scale, the balance of possible harm would seen to lean more heavily in favor of DuPont.

III.

There remains the question of an overall assessment of the foregoing factors to determine who shall prevail in this case. We start by attaching little or no significance to the one factor which really does not seem to weigh heavily either way — the foreign EFLON registrations. YKK's most telling point is the absence of evidence of actual confusion, followed by the fact that the EFLON zipper, like the products employing the TEFLON certification mark, is of high quality. Also important is the fact that no zippers are presently sold at retail under the TEFLON certification program.

Against this DuPont can point to several persuasive factors. The marks themselves have a high degree of similarity; TEFLON is a strong trademark deserving of broad protection from infringement; the EFLON zipper is an inexpensive and probably infrequently purchased product bought by homemakers who can be expected to approach such a purchase with relatively little care or scrutiny; survey and other evidence which show that the circumstances under which non-stick products bearing the TEFLON certification mark are sold are such that an ordinary purchaser might either naturally and easily assume that the source of origin was the same as or related to the source of TEFLON, or reasonably conclude that DuPont's TEFLON coatings had entered the zipper field; defendants' failure to convincingly explain their intent in adopting the EFLON mark; and, finally, the protection injunctive relief would afford DuPont as measured against YKK's failure to particularize what harm such relief would cause.

The ultimate issue is not resolved by any sort of numerical weighing of these factors, since one or two could easily be of overriding importance. And we recognize that these conclusions are "imprecise" and only "rough judgments." King Research, Inc., supra, 454 F.2d at 69. But putting all of the court's impressions together, and, as seems appropriate on this record, giving the greatest weight of all to the similarity of the marks and the strength of the TEFLON mark, the court is drawn to the conclusion that DuPont has prevailed on its claim of trademark infringement by a fair preponderance of the evidence.

IV.

YKK's last line of defense in this case is its claim that the TEFLON mark has became generic. DuPont does not dispute YKK's contention that if proved, its claim is a complete defense to the trademark infringement cause of action, but is quick to point out that as a defense to the action, the burden of showing genericness rests squarely on defendants.

Before reviewing the evidence on this point, some further discussion of general principles will serve to sharpen the inquiry. The generic defense, often referred to as "the Aspirin and Cellophane doctrine", has been recognized as placing a penalty on the trademark owner "who has made skillful use of advertising and has popularized his product." King-Seeley Thermos Co., supra, 321 F.2d at 581; Calmann, supra, § 74.2 at 232-33, 244. One commentator has suggested that there must be "conclusive evidence" of the changed character of the mark. Id. In this circuit, however, the most recent statement of the standard of review of a trademark which both identifies the class of product as well as its source, is the public's understanding of the term or terms employed:

"`[A] mark is not generic merely because it has some significance to the public as an indication of the nature or class of an article. . . . In order to become generic the principal significance of the word must be its indication of the nature or class of an article, rather than an indication of its origin.'" King-Seeley Thermos Co., supra, 321 F.2d at 580, quoting from Feathercombs, Inc., supra, 306 F.2d at 256 (emphasis in original).

To that test might be added the observation that because the defendants' burden is a heavy one, Calmann, supra, § 74.2 at 244, doubts should be resolved in favor of the trademark owner, especially if he can demonstrate having taken appropriate action to counteract or resist indiscriminate use of the mark by the public. As plaintiff discovered in its loss of the Cellophane trademark, the basic question is ultimately one of the fact of transition — the public's reaction to a word — not one of the owner's intent. Thus, whether the transition is voluntary or involuntary on DuPont's part is not controlling. See DuPont Cellophane Co., supra, 85 F.2d at 77; Calmann, supra, § 74.2 at 238-39.

I. Evidentiary Examples of Generic Use

Citing them as "dispositive evidence" that TEFLON has become generic, YKK introduced depositions of two General Electric Company employees which reveal that General Electric ceased using the TEFLON mark on April 26, 1966, and has since used the term "GE Double Non-Stick Coating" on its small appliances coated with TEFLON. Def.Exh. U-3, tab 6. Also introduced were numerous 1972 letters from General Electric's files of consumer correspondence on problems its customers had with the non-stick coatings. To the extent they are representative, the letters demonstrate that General Electric customers have continued to associate the word TEFLON with non-stick coatings long after such association might have been expected to decline. But association per se is not evidence of genericness. It cannot be thought unusual for a generic term describing the nature or class of an article (non-stick coatings and finishes) to be associated with its major indicator of source (TEFLON), but it does not necessarily follow that the public has lost sight of the distinction. In fact, the letters are very inconclusive on this point.

Another source explored by YKK was court decisions and U.S. Patents which use TEFLON in a generic manner. Def.Exh. X. But DuPont's counsel has expressed what the court views as the correct reasons for attaching no significance to this particular evidence.

YKK also introduced various advertisements and other published material as showing examples of generic use. Def.Exhs. W and Y. These do show many such examples, although "Teflon" is often capitalized, reflecting some minimal understanding of TEFLON as a trademark despite a failure to include a generic term like "coating" or "coated" with TEFLON. However, most of these individual instances of advertiser or industrial use have little probative value on overall public understanding. Moreover, most are accompanied by letters, introduced by DuPont as part of Exh. Y, which underscore its vigilant efforts to correct generic use of TEFLON whenever it came to DuPont's attention.

In summary, the isolated examples of generic use of TEFLON by the public fall far short of the burden placed on YKK to show that the mark has become generic.

2. Survey Evidence of Generic Use

YKK's principal proof of generic usage by the public comes from a second survey introduced into evidence, Def.Exh. P. That survey, actually two nationwide studies conducted by a reputable professional survey research organization, The Sorenson Group, Inc., in September 1972, is entitled "Awareness and Name Identification of Non-Stick Coating Concept."

The first study ("Survey I") was conducted among adult women, 90.6% of whom expressed awareness of "kitchen pots and pans that have their inside surfaces coated by chemical substances to keep grease or food from sticking to them." Def.Exh. P, pp. 8, 11 Table 1. Of the aware respondents, 86.1% apparently mentioned only "TEFLON" or "TEFLON II" as their sole answer when asked, "What is the name . . . or names of these pots and pans . . .?" Id. at 8, 12 Table 4. Further, 71.7% of the aware women gave only "TEFLON" or "TEFLON II" as the name they would use to describe the pots and pans to a store clerk or friend. Id. at 15 Table 6. The figure was 79.3% counting those who gave responses in addition to TEFLON or TEFLON II. Id. Only 7.3% of the aware women identified DuPont as the manufacturer of the pots and pans. Id., p. 16 Tables 7, 8.

The second study ("Survey II") was conducted among adult women, 89.4% of whom expressed some sort of awareness of "substances that manufacturers sometimes apply to the surfaces of certain products in order to prevent things from sticking to them." Id. at 10, 18 Table 10. Of the aware respondents, 81.4% apparently mentioned only "TEFLON" or "TEFLON II" as their sole answer when asked, "What name or names are these substances called . . .?" Id. at 10, 19 Table 13. Further, 60% of the aware women gave only "TEFLON" or "TEFLON II" as the name they would use to describe the pots and pans to a store clerk or friend. Id. at 21 Table 15. The figure was 70% counting those who gave other responses in addition to TEFLON or TEFLON II. Id. Only 9.1% of the aware women identified DuPont as the manufacturer of the substances. Id., at 22 Tables 16, 17. Understandably enough, no one in either survey apparently ever mentioned polytetrafluoroethylene.

In response to these two surveys, DuPont introduced a second Burke survey, Pl.Exh. 61, conducted telephonically by Burke Marketing Research, Inc., virtually on the eve of trial. It also contained two separate studies. In the first of these ("Survey A"), respondents of either sex who represented over the telephone to be over 18 years of age were told: "Protective coatings are sometimes applied by manufacturers to the inside of household utensils in order to prevent food and grease from sticking." They were then asked, "Do you know a brand name or trademark for one of these coatings?" Pl.Exh. 61, Questionnaire entitled "Protective Coating Study-A." Pursuing affirmative answers only, respondents were then asked, "what is that brand name or trademark?" Id.

The results of the survey were that of the 60% of the respondents who reached the latter question, 80% of them — or 48% of the entire sample — answered "TEFLON." Id., Table 1a 1c. All respondents were then asked, "Can you think of any other words or terms to describe these coatings?" Among the TEFLON respondents to the prior question, 32% were able to supply an additional term, the most frequently expressed of which was "Non-Stick." Id., Table 1b. Sixty-eight percent knew no other words. Id.

The second study ("Survey B") was conducted among adults in a similar fashion. By using the example of "Chevrolet — automobile" the interviewer first explained the difference between a brand name and a common name, and then asked whether each of eight names, including TEFLON, was a brand name or a common name. The results, as shown in more detail in the margin, were that 68% of the respondents identified TEFLON as a brand name and 31% as a common name.

Naturally enough, the parties are in some conflict as to the import for this case of this wealth of statistics. Briefly stated, YKK contends that DuPont's Survey A confirms its own findings, because 68% of those who identified "TEFLON" as a brand name or trademark had no other word or term to describe the coatings. Accordingly, say defendants, this 68% regards "TEFLON" as the descriptive name of the non-stick substances, a figure that compares closely with their own. In response, DuPont argues that YKK's own studies are invalid because the way Survey I and Survey II questions were worded, it was not possible to determine "what percentage of the women who gave a TEFLON response did so in the belief or on the assumption that they were being asked for a brand name." DuPont also argues that the coding of "TEFLON only" responses in both of defendants' surveys were such that they included responses which clearly reflected proper, non-generic use of the TEFLON mark, e.g., "TEFLON non-stick finish."

On a review of the exhibits, and especially the cross-examination of the author of YKK's surveys, Dr. Robert C. Sorensen, the court is satisfied that those surveys are ambiguous on the question of whether the responses truly reflect generic use of the TEFLON mark to the extent indicated. The fact is that the surveys do not really focus on the issue of the absence of trademark significance in public use of the word TEFLON. This is made clear by DuPont's own Survey A, which asked essentially the same question as did Survey II — the name for non-stick coatings — but used the expression "brand name or trademark" rather than "name or names." It seems clear from the results that in all three surveys (I, II and A), respondents were, by the design of the questions, more often than not focusing on supplying the inquirer a "name", without regard to whether the principal significance of the name supplied was "its indication of the nature or class of an article, rather than an indication of its origin." King-Seeley Thermos Co., supra, 321 F.2d at 580.

The only survey which really gets down to this critical element of the case is Survey B. It stands unrebutted as evidence that, to the extent it accurately reflects public opinion, a substantial majority of the public continues to believe that TEFLON is a brand name. YKK criticizes the survey not for being unrepresentative, but for its failure to be tied to a particular product, and argues it is thus no evidence that TEFLON has trademark significance. That contention is without merit. In fact, the responses of the survey reveal that the public is quite good at sorting out brand names from common names, and, for TEFLON, answers the critical question left unanswered by the ambiguities inherent in Surveys I, II and A — that of the principal significance of the TEFLON mark to the public. Not only have defendants failed to show that TEFLON's principal significance is as a common noun, plaintiff has succeeded in showing it to be a "brand name" — an indicator, in the words of DuPont's questionnaire, of a product "made by one company." Pl.Exh. 61, Questionnaire entitled "Protective Coating Study — B."

3. DuPont's Affirmative Efforts

The weakness of YKK's generic defense on the proof is further heightened by DuPont's extensive efforts to protect the distinctiveness of the TEFLON mark, reviewed in part I supra. Plaintiff contends that these "enormous" efforts serve to distinguish the results reached in the Bayer and Thermos cases and underscore the lesson it learned in the Cellophane case. YKK does not dispute the extent of DuPont's vigilance but argues that "it makes no difference." While it is true that DuPont's efforts are not controlling and form no basis for disregarding the general Aspirin and Cellophane doctrine, they do serve to emphasize, as has been noted above, the gravity of defendants' burden, and point up some critical factors which help partly to explain the metamorphosis when it does occur.

One important factor in cases finding the transformation to generic term complete has been the absence of the generic terms in the owner's own use of the trademark. See, e.g., King-Seeley Thermos Co., supra, 321 F.2d at 578; DuPont Cellophane Co., supra, 85 F.2d at 77. DuPont argues that, unlike the Thermos and Cellophane cases, DuPont has always tried to ensure that generic terms accompanied any use of the TEFLON mark, whether in its own advertising or elsewhere. The record of DuPont's protective efforts and YKK's examples of generic use lead the court to no contrary conclusion, and is conclusive against defendants on that point.

Another, somewhat overlapping, factor is a considerable period of acquiescence by the trademark owner in generic use, possibly coupled with a failure to take adequate affirmative action to correct generic use thereafter. See, e.g., King-Seeley Thermos Co., supra, 321 F.2d at 578-79; DuPont Cellophane Co., supra, 85 F.2d at 78-80. In this case, there is no evidence that DuPont has been anything less than diligent in its efforts to protect the trademark significance of TEFLON from the outset, and continuously thereafter.

The foregoing makes clear that, on the facts of this case, YKK was required to make a rather clear and convincing showing that the principal significance of the word TEFLON to the public is as a term for non-stick coatings and finishes rather than its trademark significance. But no matter how defendants' burden is described, it simply has not been met on this record and the defense must fail.

V.

Having concluded that plaintiff has sustained its trademark infringement claim and that the defense of genericness has not been sustained, there remains only the question of equitable relief under the principles of 15 U.S.C. § 1116. It is clear that in framing a decree which will protect plaintiff's mark, the court should be careful to minimize the injury to defendants to the extent practicable, where, as here, there is no evidence of bad faith.

While the record provides a basis for granting injunctive relief, it seems appropriate to require a further hearing, at the earliest practicable date, on the scope and form of the judgment to be entered herein. In view of defendants' permitted use of EFLON in foreign countries, it may be appropriate to provide for a grace period rather than an immediately effective injunction. See Calmann, supra, § 86.1(c).

The parties will, therefore, submit proposed judgments consistent with this opinion, and be prepared to place on the record whatever further evidence they deem necessary to support their proposals. Counsel will advise the court within ten (10) business days from the date hereof of their suggestions for a hearing date.

So ordered.