Ideal Industries, Inc. v. Gardner Bender, Inc.

612 F.2d 1018

56 A.L.R.Fed. 216, 204 U.S.P.Q. 177

IDEAL INDUSTRIES, INC., Plaintiff-Appellee,
v.
GARDNER BENDER, INC., Defendant-Appellant.

No. 79-1060.

United States Court of Appeals,
Seventh Circuit.

Argued April 24, 1979.
Decided Nov. 19, 1979.
Rehearing Denied Jan. 28, 1980.

George H. Solveson, Andrus, Sceales, Starke & Sawall, Milwaukee, Wis., Richard A. Kranitz, Grafton, Wis., for defendant-appellant.

Alfred H. Plyer, Jr., Chicago, Ill., for plaintiff-appellee.

Before FAIRCHILD, Chief Circuit Judge, MOORE, Senior Circuit Judge,* and WOOD, Circuit Judge.

MOORE, Circuit Judge:

1

This case brings up for decision unique issues in the law of trademarks, especially the question whether numbers may become common law trademarks. Defendant Gardner Bender, Inc. ("Gardner") appeals from an order dated January 10, 1979 of the United States District Court for the Eastern District of Wisconsin, Honorable John W. Reynolds, Chief Judge, which granted plaintiff Ideal Industries, Inc.'s ("Ideal") motion for a preliminary injunction. Gardner was enjoined from selling any electrical connectors bearing the numbers claimed by Ideal to be its trademarks and using the numbers on the labels of cartons or on sales literature. Gardner also was ordered to recall cartons and connectors in the hands of distributors once modified cartons and connectors become available. The district court stayed enforcement of its order on January 15, 1979; this court continued the stay pending resolution of Gardner's appeal.

2

Ideal is a Delaware corporation with its principal place of business in Sycamore, Illinois. Gardner is a Wisconsin corporation which has its principal place of business in Glendale, Wisconsin. Both parties are in the business of making and selling various electrical products. The district court's subject matter jurisdiction rested on the diversity of citizenship of the parties. 28 U.S.C. § 1332 (1976).

I.

3

Understanding of the present controversy requires a brief review of the past. Early in the 1930's Ideal began to market the product now at issue: barrel-shaped, screw-on electrical connectors. These connectors have a hard outer shell made of an electrically-insulating material which is molded around a coiled spring. They connect two or more wires securely when they are screwed onto the bared ends of the grouped wires. Such connectors are usually sold in a range of sizes. Ideal has sold, and continues to sell, a range of six sizes under the trademark "Wire Nut".

4

Beginning in 1936, Ideal adopted an arbitrary series of numbers to designate the different sizes. The number "71" was applied to the smallest, followed by numbers 72, 73, 74 and 76. There are no records which show whether these numbers corresponded to some characteristic of the connectors other than relative size. In 1946, Ideal began to make these connectors with a Bakelite rather than phenolic shell and added a capital "B" to each of the numbers to indicate the new material. The connectors were identified as 71B, 72B, 73B, 74B and 76B (hereinafter referred to as the "71B series"). A sixth, larger size was added in 1966 and was designated "78B". Although Ideal has long used a different material than Bakelite, the 71B series numbers have continued to appear on the connectors, on carton labels, in advertising, and in catalogs since 1946.

5

In 1965 or 1966, Ideal redesigned the cartons, and their labels, in which the connectors were sold. Besides modernizing the look of the label, Ideal increased the size of the 71B series designations so that they became the largest symbol on the labels. Ideal claims that this change was made because the company realized that the 71B series numbers had come to stand for the source of the connector.1 The labels continued to carry the registered trademarks "Ideal", "Wire Nut" and a registered crow's-foot design logo representing Ideal. Underneath the 71B series designation appeared the word "model" in parentheses. In 1969 the connectors and carton labels were color-coded to indicate size differences but this change had no effect on the label design or the relative prominence of the 71B series designations.

6

NOTE: OPINION CONTAINS TABLE OR OTHER DATA THAT IS NOT VIEWABLE

7

The 71B series numbers appear on the tops of the connectors themselves to satisfy the requirements of Underwriters Laboratories, Inc. ("UL") and the Canadian Standards Association ("CSA") that electrical connectors be marked with a type or catalog-number designation. The imprints of UL and CSA also appear on the top of Ideal's approved connectors, along with the trademark "Ideal" and numbers which indicate the range of wire sizes which the connector can accommodate. All these markings satisfy the CSA and UL requirements.

8

Ideal is the dominant firm in the market for barrel-shaped, fixed-spring electrical connectors. Its chief competitors are Hi-Scale Products Corp., Inc. ("Hi-Scale"), ITT Holub Industries ("Holub"), Eagle Electric Mfg. Co. ("Eagle"), and Gardner. Holub sells its connectors under its trademark "Hi" and the size designation series "No. Hi-3", "No. Hi-4" and "No. Hi-6". Beginning around 1972 Holub's carton labels displayed a 71B series number along with the word "size" in a blue circle to indicate the size comparability between Ideal's and Holub's products. Holub's own number series was still the most prominent label designation and continued to appear on the connectors themselves.

9

Prior to 1975, Hi-Scale's labels bore Hi-Scale's own series numbers, such as "HS-18" or "HS-7" as well as the appropriate 71B series number in smaller print adjacent to the word "size". The connectors displayed Hi-Scale's own series designations. In 1976, after Gardner entered the market, the 71B series number began to appear on Hi-Scale labels in approximately the same size print as the Hi-Scale series numbers. The 71B series numbers were still accompanied by the adjacent words "Approx. Size".

10

This case arises from the entry of Gardner into the electrical connector market in 1976. Instead of creating its own series numbers, Gardner adopted the 71B series as the sole type designation for its connectors. In addition, the catalog numbers for each size are similar. While Ideal uses the catalog number "30-073" to represent its 73B connector, Gardner uses "10-073" to represent its own 73B connector. Since Gardner also adopted Ideal's color coding scheme for the connectors, the only way to tell a Gardner connector from one made by Ideal is the presence of a "GB" in place of "IDEAL" on the top of the connectors. All else on the connector is the same. The Gardner carton labels are the same color as Ideal's, although they bear Gardner's "GB" trademark prominently, in addition to Gardner's distinctive logo. The word "style" appears next to the 71B series number on the carton labels.

11

Ideal commenced this action in May 1976, charging that Gardner was engaged in unfair competition and common law trademark infringement. The complaint sought preliminary and permanent injunctive relief and damages. Ideal filed its brief and exhibits in support of its motion for a preliminary injunction in January, 1977. Not until November 30, 1978 did the district court hold a hearing on the motion. Ideal relied only on its claim that Gardner was infringing its alleged common law trademark rights in the 71B series; the unfair competition claim was not pressed as support for the preliminary injunction. The motion was considered on the affidavits, deposition transcripts and exhibits submitted by the parties as well as the pleadings and the arguments made at the hearing. The district court expressly absolved Ideal of fault for the long delay between the filing of the motion and the hearing. The district court rendered its decision and order in favor of Ideal on January 10, 1979; Gardner appeals. We affirm with modifications and remand for further proceedings.II.

12

The scope of review at this stage is quite restricted. An order granting a preliminary injunction in a trademark case "will not be set aside by a Court of Appeals unless it is contrary to the principles of equity or the result of improvident exercise of judicial discretion". Doeskin Products, Inc. v. United Paper Co., 195 F.2d 356 (7th Cir. 1952). This court recently reaffirmed its view that "abuse of discretion" is the proper standard. Helene Curtis Industries v. Church & Dwight Co., 560 F.2d 1325 (7th Cir. 1977), Cert. denied, 434 U.S. 1070, 98 S.Ct. 1252, 55 L.Ed.2d 772 (1978). Furthermore, the district court's findings of fact will not be upset unless they are "clearly erroneous" as provided by Fed.R.Civ.P. 52(a). Fleetwood Co. v. Hazel Bishop, Inc., 352 F.2d 841 (7th Cir. 1965). Therefore, the following discussion of the legal principles governing the alleged trademarks is merely to aid in explaining the court's view that the district court did not abuse its discretion in evaluating the likelihood of Ideal's succeeding on the merits.

13

The basic issue is whether it is likely that Ideal's 71B series numbers are common law trademarks. Gardner's chief argument on this appeal is that numbers which originally were adopted to indicate different sizes of a product rather than its origin cannot be trademarks. Gardner relies heavily on this court's statements in William H. Keller, Inc. v. Chicago Pneumatic Tool Co., 298 F. 52 (7th Cir. 1923), Cert. denied, 265 U.S. 593, 44 S.Ct. 637, 68 L.Ed. 1196 (1924):

14

"There can be no question but what (sic) a number may become a good trade-mark, if its primary adoption be solely to indicate origin. On the other hand, if the figures indicate a grade or a quality only, they may not be the basis for a valid trade-mark." 298 F. at 59.

15

The numbers at issue in the case indicated the length of a piston stroke in a riveting hammer. The quoted statement was a restatement of the Supreme Court's holding in Coats v. Merrick Thread Co., 149 U.S. 562, 13 S.Ct. 966, 37 L.Ed. 847 (1893), that numbers indicating the size of thread cannot be trademarks. Other courts have made similar general statements. See, e. g., Fram Corp. v. Boyd, 230 F.2d 931 (5th Cir. 1956); Dennison Mfg. Co. v. Scharf Tag, Label & Box Co., 135 F. 625 (6th Cir. 1905), Cert. denied, 201 U.S. 648, 26 S.Ct. 762, 50 L.Ed. 904 (1906). Apparently these courts did not have before them the question whether the numbers indicating style or grade had achieved a secondary meaning as symbols of the source of the goods.

16

The general rule with respect to descriptive terms was stated by this court as follows:

17

"A merely descriptive term specifically describes a characteristic or ingredient of an article. It can, by acquiring a secondary meaning, i. e., becoming 'distinctive of the applicant's goods' (15 U.S.C. § 1150(f)), become a valid trademark." Miller Brewing Co. v. G. Heileman Brewing Co., 561 F.2d 75, 79 (7th Cir. 1977), Cert. denied, 434 U.S. 1025, 98 S.Ct. 751, 54 L.Ed.2d 772 (1978).

18

Substantial authority supports the idea that terms which were originally descriptive of style or grade ought to be treated the same way as any merely descriptive term. Clairol, Inc. v. Gillette Co., 389 F.2d 264 (2d Cir. 1968); 1 J.T. McCarthy, Trademarks and Unfair Competition, § 11:15, pp. 371-375 (1973). Thus numbers which describe the size of a product, an aspect of the product's grade, ought to be capable of becoming trademarks if the proponent of the mark can prove secondary meaning. In Armco Steel Co. v. Watson, 188 F.Supp. 554 (D.D.C.1960), the court ordered the Commissioner of Patents to register the numbers 17-4PH and 17-7PH as trademarks designating brands of stainless steel products. The Patent Office had originally denied registration of the marks because the letters "PH" stood for "precipitation hardening", a process used in making the steel, and the numbers described the proportions of metals comprising the stainless steel alloy. The court found that the marks had developed a secondary meaning through long use and advertising, and had thus become trademarks, as well as descriptive terms. In another case, the Trademark Trial and Appeal Board denied registration to a mark which would have infringed the previously registered marks "SK-97", "SK-128", "SK-98", "SK-58" as applied to audio speakers. Although the case did not involve secondary meaning, the Board declared that letter marks are entitled to the same scope of protection as other registered marks. In re Standard Kollsman Industries, Inc., 156 U.S.P.Q. 346 (TTAB 1967). The fact that the 71B series designations are numbers descriptive of size does not prevent them from becoming trademarks if they have acquired secondary meaning.

19

Gardner further argues that the 71B series designations have become generic or common descriptive terms, which by their nature cannot become trademarks. Miller Brewing Co. v. G. Heileman Brewing Co., supra, 561 F.2d at 79-80. In light of Gardner's numerous assertions that the 71B series numbers describe the Size of a connector, this argument is difficult to understand. "A generic or common descriptive term is one which is commonly used as the name or description of a Kind of goods." Miller Brewing Co. v. G. Heileman Brewing Co., supra, 561 F.2d at 79 (emphasis added). If there is any candidate for a generic or common descriptive term among Ideal's marks, it is the registered trademark "Wire Nut", which designates the Kind of electrical connector characterized by a barrel shape and a fixed, coil spring, yet this term has not, as far as we know, passed into common usage. The term "wire connector" is a generic or common descriptive term which probably could not become the exclusive property of any one producer; otherwise, it would be impossible to tell the buyer what the product is.2 Since these characteristics do not apply to the 71B series numbers, they do not fit into the category of generic or common descriptive terms.

20

The 71B series numbers are not "arbitrary" marks in the trademark sense. Although the numbers were chosen arbitrarily in the sense that they do not refer directly to a characteristic of the connectors, the progression of numbers was adopted, and is currently used, to describe the relative sizes of the connectors. Hence, they are merely descriptive, not arbitrary, terms.

21

Ideal concedes that proof of secondary meaning is necessary and argues that there is enough proof in the record to support the district court's finding that the 71B series numbers do have secondary meaning.

22

When evaluating alleged proof of secondary meaning, a court should be chiefly concerned with the attitudes of purchasers towards the mark. Union Carbide Corp. v. Ever-Ready Inc., 531 F.2d 366, 380 (7th Cir.), Cert. denied, 429 U.S. 830, 97 S.Ct. 91, 50 L.Ed.2d 94 (1976). Ideal submitted seven affidavits and eight depositions of its customers from widely separated areas of the country. Twelve of these were from independent wholesalers of electrical equipment, firms which buy connectors from several manufacturers. Three statements came from electrical contractors, the normal end-users of electrical connectors. The district court found that all the persons who gave affidavits or responded to deposition requests stated that they were familiar with the electrical industry, that they regarded Ideal as the source when they saw the 71B series numbers used in relation to electrical connectors, and that they believed that Ideal would be identified as the source of such connectors throughout the industry.3 The only challenges to these proofs were thirteen affidavits of "individual manufacturer's representatives", and two depositions of officers of Ideal's competitors. The gist of these statements was that the 71B series was used commonly throughout the industry to indicate the size of a connector, regardless of its source.

23

The district court quite properly discounted the value of the Gardner affidavits because the affiants were Gardner's sales representatives and thus had an interest in belittling the trademark value of the 71B series numbers. The same is true of the depositions by the officers of Ideal's competitors. Furthermore, the exhibits submitted by Ideal represented the opinions of Purchasers of the product in contrast with Gardner's exhibits. The Union Carbide case stressed the importance of Purchasers' attitudes. When these statements are added to the fact that Ideal has marketed its connectors with the 71B series designations for over 30 years and that during that period Ideal has been the dominant firm (over 80% Of the market in 1976) in the market, we conclude that the district court could properly infer that the 71B series numbers had acquired secondary meaning. See W. E. Bassett Co. v. Revlon, Inc., 435 F.2d 656, 661 (2d Cir. 1970) (properly supported inference of secondary meaning for a merely descriptive term is enough to justify trademark protection).

24

Of course, at trial Ideal must substantiate the inference that the 71B series designations have come to have a secondary meaning according to the standard stated by Judge Denison in G. & C. Merriam Co. v. Saalfield, 198 F. 369, 373 (6th Cir. 1912), Cert. denied, 243 U.S. 651, 37 S.Ct. 478, 61 L.Ed. 947 (1917):

25

"So it was said that the word had come to have a secondary meaning, although this phrase 'secondary meaning' seems not happily chosen, because, in the limited field (the particular trade and branch of the purchasing public), this new meaning is primary rather than secondary; that is to say, it is, in that field, the natural meaning."

26

See also Restatement of Torts § 716, Comment b (1938); Kellogg Co. v. National Biscuit Co., 305 U.S. 111, 59 S.Ct. 109, 83 L.Ed. 73 (1938). We decide now only that the district court had before it enough evidence to conclude that Ideal was likely to succeed on the merits.

27

The last issue which relates to the likelihood of Ideal's succeeding on the merits is the question whether Gardner's use of the 71B series numbers will create a likelihood of confusion as to the source of the goods. Gardner's argument that no proof of actual confusion was submitted is unavailing. No evidence of actual confusion is required in order to prove a likelihood of confusion. Helene Curtis Industries v. Church & Dwight Co., supra, 560 F.2d at 1330; W. E. Bassett Co. v. Revlon, Inc., supra, 435 F.2d at 662. Several factors are important in determining the likelihood of confusion: the similarity of the marks, the similarity of the products, the area and manner of concurrent use, the degree of care likely to be exercised by consumers, the strength of the complainant's mark, actual confusion, and intent on the part of the infringer to palm off his products as those of another. Helene Curtis Industries v. Church & Dwight Co., supra, 560 F.2d at 1330; Union Carbide Corp. v. Ever-Ready, Inc., supra, 531 F.2d at 381-382.

28

Looking at the facts here, we see that the marks are the same, the products are the same, the manner of use of the marks is the same, and the geographical markets overlap or are identical. As if this were not enough, there is evidence that customers are not very careful in ordering the connectors. Two witnesses stated in depositions that sometimes a customer would ask for connectors by the 71B series number alone, without mentioning a producer. Gardner disingenuously argues that these customers are merely indicating indifference as to the source of the goods. The contrary inference is more likely when one remembers that Ideal has been the dominant supplier of connectors using the 71B series numbers for over thirty years. The likelihood of confusion is almost obvious, and is not reduced by Gardner's decision to copy Ideal's color coding scheme in addition to using the 71B series numbers. This issue is subject to further proof at trial but, for present purposes, the likelihood of confusion is strong enough to justify a preliminary injunction.

III.

29

In order to obtain a preliminary injunction, a plaintiff must show, in addition to a likelihood of success on the merits, that it will be irreparably injured unless steps are taken to preserve the status quo. Helene Curtis Industries v. Church & Dwight Co., supra, 560 F.2d at 1330. The trial judge must also weigh the relative hardships to the parties. Id. The district court found that Ideal was in danger of losing business and losing its current identification as the source of the 71B series connectors and that if Ideal should prevail on the merits the burdens of eradicating impermissible uses of its trademarks would be substantially increased were a preliminary injunction not granted. Apparently Gardner failed to produce evidence of the hardships it would suffer if limited, preliminary injunctive relief were granted. The issue is whether the district court abused its discretion in granting the requested preliminary relief. Helene Curtis Industries v. Church & Dwight Co., supra.

30

Gardner argues that Ideal's long delays in commencing and prosecuting this action have vitiated Ideal's claim that it will be irreparably injured unless a preliminary injunction is granted. Ideal filed its complaint seven months after learning of Gardner's intention to enter the market. Eight months later, after accumulating seven affidavits and eight depositions from independent distributors, Ideal filed the brief and supporting papers on its motion for a preliminary injunction. Twenty-two months after this second filing the district court held the hearing on the motion. In its decision and order the district court stated that the twenty-two month delay after the motion papers were filed was "not fairly attributable to the plaintiff". Apparently the delay resulted from the condition of the district court's calendar.

31

Gardner is correct in arguing that the plaintiff's delay in moving for a preliminary injunction has been considered by some courts in assessing the probability of irreparable injury. W. E. Bassett Co. v. Revlon, Inc., supra, 354 F.2d at 874 n.4; Programmed Tax Systems, Inc. v. Raytheon, 419 F.Supp. 1251, 1255 (S.D.N.Y.1976). However, delay is only one among several factors to be considered; these cases do not support a general rule that irreparable injury cannot exist if the plaintiff delays in filing its motion for a preliminary injunction. On the contrary, this court has stated that mere passage of time cannot constitute laches. Helene Curtis Industries v. Church & Dwight Co., supra, 560 F.2d at 1334. In evaluating the defense of laches, the Helene Curtis court looked to whether the defendant had been lulled into a false sense of security or had acted in reliance on the plaintiff's delay. The delay attributable to Ideal in this case neither lulled Gardner nor caused it to act in reliance on the delay. Hence, the district court was within its discretion when it refused to apply the laches defense in its analysis of Ideal's claim of irreparable injury.

32

The existence of irreparable injury is positively supported by the fact that the alleged trademark and the infringing use are identical, that the products are the same, and that the markets are the same. These factors by themselves are indicative of irreparable injury. One commentator has noted:

33

"Also essential to trademark law is the presumption that use of a trademark or trade name identical with that of a competitor, on similar goods and in a similar business, causes deception and confusion of the public." Nims, The Law of Unfair Competition in Trademarks 1078 (4th ed. 1947).

34

And, as this court has already said, "the damage to the goodwill and prominence of the (plaintiff's) trademark through public confusion of it with the (defendant's) trademark is, in itself, an irreparable injury". Helene Curtis Industries v. Church & Dwight Co., supra, 560 F.2d at 1332. See also Omega Importing Corp. v. Petri-Kline Camera Co., 451 F.2d 1190, 1195 (2d Cir. 1971). This readiness to find irreparable injury arises, in part, from the realization "that the most corrosive and irreparable harm attributable to trademark infringement is the inability of the victim to control the nature and quality of the defendant's goods". 4 Callman, Unfair Competition, Trademarks and Monopolies § 88.3(b) at 205 (3d ed. 1970). Monetary damages are likely to be inadequate compensation for such harm.

35

The district court has an obligation to weigh the relative hardship to the parties in relation to its decision to grant or deny a preliminary injunction, even when irreparable injury has been shown. Gardner's brief presents an array of costs which it will have to bear if the preliminary injunction is imposed. Elimination of the 71B series numbers from Gardner's connector molds would shut down its manufacturing operations for at least one month, if Ideal's contentions are accepted, or up to seven months, if Gardner's claims are correct. Gardner's current inventory of connectors, worth approximately $180,000, would be a total loss. This amount is one-half of Gardner's net worth. Ideal's claim that such a high inventory level is the product of mismanagement may be correct; nevertheless, the inventory is a fact which must be considered. Finally, the requirement that all connectors be recalled from the possession of Gardner's distributors would irreparably damage Gardner's relations with its dealers and would create substantial replacement costs.

36

The district court did not fully discuss the relative hardships of the parties but seems to have been handicapped by Gardner's failure to present all the facts prior to the issuance of the injunction. In light of the facts presented to us, we affirm the issuance of an injunction but remand for modification of the order for reasons stated below. The hardships which Gardner will suffer do not warrant denial of all preliminary relief to Ideal. One entering a field already occupied by another has a duty to select a trademark that will avoid confusion. Watkins Products, Inc. v. Sunway Fruit Products, Inc., 311 F.2d 496, 499 (7th Cir. 1962), Cert. denied, 373 U.S. 904, 83 S.Ct. 1291, 10 L.Ed.2d 199 (1963). See also Helene Curtis Industries v. Church & Dwight Co., supra, 560 F.2d at 1333-4. Gardner surely knew that both Hi-Scale and Holub only used the 71B series as size designations and that they used their own series numbers on the connectors and the cartons to indicate the source of the products. In spite of this knowledge, Gardner chose a course which it must have realized would create buyer confusion. Gardner could easily have invented its own series numbers for use on the connectors and thus fulfilled its duty to avoid confusion. Having adopted its course, Gardner cannot now complain that having to mend its ways will be too expensive.

37

Nonetheless, the facts necessitate a different balance in the hardships suffered by the parties. Prior to and during the long delay before a hearing was held on the preliminary injunction motion, Gardner continued to produce the offending connectors to a point where its current inventory constitutes almost half of its net worth. A remedy like the injunction at issue, which renders this inventory a total loss, seems overly harsh. Gardner claims that there is a risk that it will become insolvent. Ideal would be reasonably protected if the district court fashioned a modification to its order, allowing Gardner to make normal sales from its inventory during the time necessary for Gardner to alter its machinery. Ideal claims the changes can be made in one month; Gardner says they will take seven. This is a question for the district court. The district court should exercise its discretion to ensure that Gardner's sales from inventory will be made in good faith and will not subvert our intent by flooding the market with connectors which may confuse buyers. This modification is necessary to maintain Gardner's cash flow and prevent financial disaster to Gardner from loss of its total inventory.

IV.

38

Although the record contains sufficient evidence to support a preliminary injunction with the modification suggested above to protect Ideal's alleged trademarks, additional evidence apparently not considered by the district court calls for a further modification of the injunction order. Gardner has raised the defense that its use of the 71B series designations is a "fair use" of the numbers in their descriptive sense only. In situations where a mark has both a trademark function and a descriptive function, the "fair use" defense allows a junior user of a mark to use the mark in good faith in its descriptive sense, as opposed to its trademark sense. J. T. McCarthy, Trademarks and Unfair Competition, § 11:17 at 377 (1923). The analytical focus of the defense, which exists under both the Lanham Act, 15 U.S.C. § 1115(b)(4) (1976), and the common law, is on how the term is being used. Venetianaire Corp. of America v. A & P Import Co., 429 F.2d 1079 (2d Cir. 1970). The typical case involves a term having a descriptive meaning in popular usage which is adopted by some firm to denominate certain of its products. Even though the term has become a trademark as to those products, it may still be used in its original descriptive sense if the use is in good faith. See Abercrombie & Fitch Co. v. Hunting World, Inc., 537 F.2d 4, 12-13 (2d Cir. 1976) (use of word "safari" with respect to clothing).

39

This case is a slight variant of the typical case. The 71B series numbers had no descriptive meaning prior to their use by Ideal. However, over time they came to have a descriptive meaning which apparently is recognized and used throughout the market. The district court found that it is probable that the 71B series numbers have come to have, in addition, a secondary meaning distinguishing Ideal's products from those of its competitors. These two aspects of the mark can be recognized. Ideal may prevent use of the terms by others in their trademark sense as symbols of the goods of a single producer such that buyers are likely to be confused, but may not claim an exclusive right to use of the marks in their generally accepted descriptive role. See J. T. McCarthy, Trademarks and Unfair Competition, § 11:17 at 379 (1973); Restatement of Torts, § 727, Comment a (1938).

40

Gardner argues that it only uses the 71B series numbers in a descriptive sense, as indicative of size or style. It claims that its packaging and advertising clearly show use of the terms "size" and/or "style" in conjunction with the 71B series designations. An examination of the reproduction of Gardner's packaging and advertising contained in the briefs and appendices reveals that the word "style" appears on the carton labels and in the advertising; nowhere does Gardner show that it is using the 71B series to describe "size". Yet Ideal said that the series in part designates size, Ideal's distributors said it indicates size, and Ideal's competitors only use it to indicate size. Even the affidavits submitted by Gardner declared that the designations indicated the size of the connectors. No one but Gardner thinks the series numbers are descriptive of "style" or type.

41

On the evidence in the record, we conclude that to the extent the 71B series numbers are descriptive, they are descriptive only of size, not of a "style" or "type". When Ideal uses the numbers with the word "model" on Ideal's carton labels, the numbers are being used in their trademark sense. However, in light of the "fair use" doctrine, the district court's order goes too far in prohibiting Gardner from using the 71B series numbers in any way on its cartons. For the purposes of establishing fair preliminary protection for Ideal's alleged trademarks, Gardner should be allowed fairly to use the numbers on Gardner labels, adjacent to the word "size", to inform buyers of the size of the connectors. A final determination of the scope of the "fair use" defense must await the trial.

42

Ideal appears to have implicitly accepted this limited use of the numbers in the course of defending the nature of Holub's use of the 71B series numbers in contrast to Gardner's use. It seems that as long as Holub uses its own series numbers on the connectors and in a dominant manner on the carton labels, Ideal has no strong objection to Holub's discreet use of the numbers on the carton as indicators of the size of the connectors. Ideal, of course, condemns Hi-Scale's expanded use of the 71B series numbers, which began on the heels of Gardner's appropriation of the series numbers. In any event, Holub began using the 71B series numbers as size designations on its cartons in 1972 and Ideal was aware of that use. Not until five years later, after Ideal commenced this suit, did Ideal object to such use of its alleged trademarks. In all fairness, Gardner should not be preliminarily enjoined from engaging in a use which Ideal allowed to Holub for five years. This conclusion is particularly appropriate where the prior use by Holub appears to be within the scope of the "fair use" defense.

43

The district court properly enjoined Gardner's use of the numbers on the tops of the connectors to indicate its type, as required by UL and CSA standards. Gardner's decision to adopt Ideal's color coding scheme as well as Ideal's series numbers increases the likelihood of customer confusion. Gardner will have to create its own series designations for use on the connectors in compliance with the UL and CSA requirements until such time as this case comes to final judgment after trial.

44

In summary, the preliminary injunction should be modified to allow Gardner to sell connectors from its inventory until it can produce connectors without the infringing numbers on top and to use the 71B series numbers on cartons fairly to describe the size of the connectors. We remand the case to the district court for it to exercise its discretion in modifying its order consistent with this opinion and for trial.

*

The Honorable Leonard P. Moore, Senior Circuit Judge of the United States Court of Appeals for the Second Circuit, is sitting by designation

1

We reproduce here the labels of Ideal and Gardner for one size of connector

2

For an excellent presentation on the difference between a merely descriptive term and a common descriptive term, see the example of the "Deep Bowl Spoon" in Abercrombie & Fitch Co. v. Hunting World, Inc., 537 F.2d 4, 10 n.11 (2d Cir. 1976), Quoting Fletcher, Actual Confusion as to Incontestability of Descriptive Marks, 64 Trademark Rep. 252, 260 (1974)

3

Olin Blocker, an electrical distributor from Atlanta, Georgia, described in a deposition his perception of the 71B series as follows:

"Well, I don't recall anybody using these numbers other than Ideal, and it has been, I guess, one of the leaders in the field, and if anybody would say 73B wire nuts, I think that the trade generally thinks of Ideal wire nuts."