This appeal is by a manufacturer of men's toilet articles who sought to enjoin an imitative competitor in the same field-imitative in reference to "trade dress" or packaging. The United States District Court for the Eastern District of New York, George C. Pratt, Judge, while having "no doubt" that the packaging of Norco Enterprises, Inc. (Norco), was modeled after that of Paco Rabanne Parfums, S.A. (Paco Rabanne), nevertheless refused to issue a preliminary injunction because of the absence of proof of "side-by-side" sales. On this appeal under 28 U.S.C. § 1292(a)(1), we hold that the court erred as a matter of law in finding that Paco had not established a significant possibility of irreparable harm in the absence of proof of side-by-side sales.

Since 1974 Paco Rabanne has manufactured and distributed cologne and grooming aids for men ranging from body lotion and soap to after-shave lotion and deodorant. Its ten products, including a cologne marketed in five different sizes and a cologne spray, are dressed for the trade in a snappy rectangular box coated in glossy acetate. The box is colored light green and gray with a white border on all sides and the words "paco rabanne pour homme" in black lower-case lettering. Above the name on the front of the box appears a distinctive logo: a lower-case "p" in solid black lettering which has in its center a lower-case "r" in the same color as the background.

As seen by the naked eye and as essentially found by the district court, Norco markets its cologne by the tradename "Deron" in boxes with approximately the same color as the Paco Rabanne box (the district judge thought that the green of the Norco box had a somewhat browner cast to it than the grayer green of the Paco box), with the same border design, and with the same logo design except that a "d" is used and the background-colored letter within it is also a "d". The Norco box has the same basic typeface and the same positioning for the logo design and the name of the product as the Paco box, although it uses the words "deron COLOGNE SPRAY FOR MEN" rather than the semi-French of the Paco box. The "deron" is in lower-case type larger in size than the upper-case phrase following it. To the eye, the packages are deceptively similar in appearance. Deron has been sold since 1979.

The district judge indicated that "if there were evidence before (him) that the two products were lined up on the shelves of, say, the Genovese Drugstores side-by-side for sale, (he) probably would have little difficulty in concluding that there was likelihood of confusion, even in the absence of actual confusion." In the absence of evidence that the two products came "head to head anywhere," however, the court held that there was not a substantial possibility of irreparable harm so as to warrant the grant of a preliminary injunction.

The district court reasoned that there is no harm to the plaintiff unless some consumers or merchants are exposed to both products, regardless of the likelihood of confusion were they to be so exposed. It found no evidence of such dual exposure: "What is missing ... is any evidence that the products are sold in the same store, in the same block, in the same city, or to the same kinds of people." We reverse because we find the evidence in this case shows that Deron and Paco were distributed to the same market.

The court also found that Paco is sold in many places around the world, including Haiti and Santo Domingo, but that the only evidence as to Deron was that it is sold in the Caribbean nations. The judge then stated that while Paco was aimed at a prestige market and there was no evidence about what market Deron was aiming at, it was difficult to imagine that Deron would be sold in economically depressed Haiti if it were a prestige product. He considered it unlikely that Deron was being palmed off in Haiti as expensive French perfume because French language was not used on the label.

At this stage of the proceeding, we find that Paco met its burden of showing a likelihood that the two product lines are being sold in the same market. The evidence shows that Norco deliberately copied Paco's design. From this and the fact that the products were sold in the same country (Haiti) and city (Santo Domingo), the court could infer that Norco intends to sell and is selling its product to people who have already been exposed to Paco's advertising, that is, to people who already know the Paco package and buy it.

In particular, the court could infer that Norco intended to pitch its confusingly similar packaging to the tourist trade, which might have been impressed in the States by Paco Rabanne advertising. Some tourists in the market for men's toiletries might not read the small print and discover that Norco is using English and hence must be an American rather than a French product. Others would be unlikely to think that Paco Rabanne is a French product even though it purports to use French in its smaller print; we note that Paco omits the article "l" before the word "homme." Either of these inferences is at least as plausible as the trial judge's belief that if Norco had intended to palm its product off in the free shops as expensive French perfume, then it would have used French in the label and not English. There is thus no basis for the district court's conclusion that Deron is not marketed to the same luxury trade as Paco.

Moreover, while there was no evidence of side-by-side sales of Paco and Deron in any store, in this circuit it is well established that products need not be sold side by side for a plaintiff to show that there is a likelihood of confusion indicating success on the merits. In RJR Foods, Inc. v. White Rock Corp., 603 F.2d 1058, 1060 (2d Cir. 1979), we stated that "defendant's conscious imitation also supports at least a presumption that the similarity will cause customer confusion .... The test of customer confusion is not whether the products can be differentiated when subjected to a side-by-side comparison, but rather whether they create the same general overall impression." Similarly, in American Home Products Corp. v. Johnson Chemical Co., 589 F.2d 103, 107 (2d Cir. 1978), we held that "(t)he test is not whether the customer will know the difference if he sees the competing products on the same shelf. It is whether he will know the difference if 'ROACH INN' is singly presented and he has heard of 'ROACH MOTEL' " (emphasis in original). These cases naturally follow from Judge Learned Hand's opinion in American Chicle Co. v. Topps Chewing Gum, Inc., 208 F.2d 560, 563 (2d Cir. 1953), in which he stated:

Indeed, it is generally true that, as soon as we see that a second comer in a market has, for no reason that he can assign, plagiarized the "make-up" of an earlier comer, we need no more; for he at any rate thinks that any differentia he adds will not, or at least may not, prevent the diversion and we are content to accept his forecast that he is "unlikely" to succeed.

The question, then, becomes whether the district court may put to one side this clear-cut case law on the issue of likelihood of confusion by casting its holding in terms of irreparable injury. We think not. The cases establish that the absence of side-by-side sales is irrelevant to a claim under the Lanham Act or New York unfair-competition law when direct competition for the same market is involved. To paraphrase Judge Hand, the court should have accepted Norco's belief that it would gain, and Paco Rabanne would be harmed, by the imitation of the latter's packaging.

Some of our cases assume that harm to an imitated plaintiff is irreparable. For example, before turning to the merits (the discussion of side-by-side sales) in American Home Products the court stated: "There is no question that irreparable harm would result from a failure to enjoin if the plaintiff ultimately succeeds on the merits." 589 F.2d at 106. Other cases are more explicit. Omega Importing Corp. v. Petri-Kine Camera Co., 451 F.2d 1190 (2d Cir. 1971), held that

Where there is, then, such high probability of confusion, injury irreparable in the sense that it may not be fully compensable in damages almost inevitably follows. While an injured plaintiff would be entitled to recover the profits on the infringing items, this is often difficult to determine; moreover, a defendant may have failed to earn profits because of the poor quality of its product or its own inefficiency. Indeed, confusion may cause purchasers to refrain from buying either product and to turn to those of other competitors. Yet to prove the loss of sales due to infringement is also notoriously difficult.

Id. at 1195. The likelihood of damage to reputation and good will, even where there is no proof of lost sales or side-by-side competition, entitles a plaintiff to preliminary relief. Commerce Foods, Inc. v. PLC Commerce Corp., 504 F. Supp. 190, 196-97 (S.D.N.Y.1980); Menley & James Laboratories, Ltd. v. Approved Pharmaceutical Corp., 438 F. Supp. 1061, 1068-69 (N.D.N.Y.1977).

We note that in Norco's brief and in argument on this appeal it has never really disputed the proposition that side-by-side sales are unnecessary in connection with a preliminary injunction where there is copying of a trade dress by a competitor in the same field. Rather, Norco argues that the district court did not in fact rely on this erroneous view of the law. Norco contends that Judge Pratt actually based his conclusion that Paco would not suffer irreparable harm on such other considerations as marketing procedures, the shapes of the bottles, and his intuitive reactions, and that he did not abuse his discretion in so doing. We disagree with Norco's characterization of the court's conclusions, although we note that each of these matters is mentioned in the course of the court's remarks, which took the place of a written opinion.

The judge did cite, in support of his conclusion of lack of irreparable harm, the manner in which the parties' products are sold-that is, by salespersons presumably available to answer customers' questions, rather than off the shelf. Sales under these circumstances, the judge thought, made it less likely that customers would be confused than if the products were sold side by side from a shelf in a drug store. There is, however, no evidence of this practice in the record. Although Norco submitted a memorandum that referred to "the suggested procedure in marketing cologne," the only evidence of marketing procedure consisted of the cross-examination of Paco Rabanne's witness, John Michael Lamie. Lamie testified simply that many counters display the bottles where shoppers can test the various fragrances and that fragrance will be one of the customer's important considerations, perhaps after consideration of the name of the product.

Moreover, although the court did note the difference between the bottles, these products are usually sold in the package and the court also noted "the fairly close similarity between the defendant's package and the plaintiff's." And finally, it seems to us that the district judge attempted to set aside his intuitive reactions. He disclaimed having a sharp sense of smell or eye for color, or any familiarity with cologne products, and he stated that he would not take into account the reactions of his staff. In short, none of the considerations Norco contends the district court relied on could have been material to the denial of the preliminary injunction; we read the court's decision as turning on the erroneous view that in the absence of side-by-side sales, there could be no irreparable harm.

Judgment declining to award a preliminary injunction reversed; cause remanded for further proceedings not inconsistent with this opinion. Stay continued.