MEMORANDUM-DECISION AND ORDER

POOLER, District Judge.

INTRODUCTION

Kompan A.S., Kompan, Inc. and Kompan/Big Toys Northeast (collectively Kompan), manufacture and distribute playground equipment. Defendants Park Structures, Inc., Kay and Alan Bayman and Baughman Bros., Inc. (collectively "PSI") compete with Kompan as manufacturers and/or distributors of playground equipment. PSI has created a product line, the "Karavan" line, that closely resembles Kompan's pre-existing early childhood line. Kompan alleges that PSI's hen bouncer, bunny bouncer, and goose bouncer — along with other products in the Karavan line — have bounced over the elusive border that separates imitation as flattery from imitation as unfair competition. Kompan seeks a variety of injunctive relief. Because Kompan has shown a likelihood of success on its trade dress infringement claim, Kompan is entitled to much, but not all, of the injunctive relief it seeks.

BACKGROUND

In the late 1960's, a Danish city commissioned Tom Lindhardt to create an urban sculpture. Olson Decl. I ¶ 8. Lindhardt built his sculptures by cutting plywood into abstract shapes with rounded edges. He painted the sculptures in bright basic colors and covered bolts, hinges and fasteners with round white caps. Id. ¶ 9. Lindhardt received additional commissions for his play sculptures and, in 1970, along with two others, formed Kompan A.S. to produce and market the sculptures commercially. Id. ¶ 11. Kompan maintains that it has consistently used the same design elements and trade dress since Lindhardt's first sculptures. Id. ¶ 9. Kompan began marketing its line in the United States in 1986 or 1987. Compare Am. Compl. ¶ 14 with Bogan Decl. I ¶ 4. However, some of the playground equipment for which Kompan seeks injunctive relief was introduced later, between 1992 and 1994. Bogan Decl. I ¶¶ 5-6.

Kompan's annual world-wide revenues approximate $100 million. Olson Decl. I ¶ 4. Last year Kompan, Inc., Kompan's United States subsidiary, earned approximately $4.8 million and spent approximately $350,000 on advertising. Id.; Olson Decl. II ¶ 12.

In March 1995, Jeffrey L. Olson, the president and general manager of Kompan, Inc., first saw a catalog for PSI's Karavan line. Olson Decl. I ¶ 23. Olson believed that virtually every item in several of Karavan's product groupings closely imitated the shape, color, and composition of playground equipment manufactured and sold by Kompan. Id. Olson also believed that PSI had further attempted to confuse consumers as to the source of its new product line by using the name "Karavan" which has sound and sight similarities to the Kompan name and by using a similar format for its catalog. Id. at ¶¶ 23-31. Kompan therefore commenced this action, which alleges violation of the Lanham Trade-Mark Act, 15 U.S.C. § 1051 et seq. (the "Lanham Act" or the "Act") by false advertising and trade dress infringement, common law unfair competition and unjust enrichment, and violation of various New York statutes on April 10, 1995. On that same date, Kompan's counsel sent a letter to Karavan demanding that Karavan cease all manufacturing, marketing, selling and distribution of 27 items that Kompan believed infringed its trade dress. Holman Aff. ¶ 2 Ex. A. The April 10, 1995, letter demanded that Karavan provide assurance of its intention to cease its infringing activities by April 14, 1995. Id. At the request of PSI's counsel, Kompan extended this deadline to April 21, 1995. Id. ¶¶ 3-4, Ex. B C. On April 18, 1995, PSI's counsel responded by denying infringement. Id. ¶ 5, Ex. D. Within a week, Kompan filed a motion for a preliminary injunction. PSI has filed answers and counterclaims to both the complaint and a subsequent amended complaint and opposes Kompan's motion. I heard oral argument on June 5, 1995, and allowed additional submissions, the last of which were received on June 19, 1995.

DISCUSSION

Although Kompan seeks injunctive relief based on each of its claims against PSI, its proof and argument — and PSI's response — focus on the claim, made under section 43(a) of the Lanham Act, that PSI has infringed upon its trade dress. 15 U.S.C. § 1125(a)(1)(A). I find that Kompan has established both likelihood of success and irreparable harm on its trade dress infringement claim because it has shown that the Kompan early childhood product line — with the exception of the Crazy Scrambler™ — has a distinctive trade dress and that PSI's imitative Karavan line creates a likelihood of confusion. I therefore address Kompan's remaining claims only briefly.

I. The Preliminary Injunction Standard

In order to obtain a preliminary injunction, the movant must show "(a) irreparable harm and (b) either (1) likelihood of success on the merits or (2) sufficiently serious questions going to the merits to make them a fair ground for litigation and a balance of hardships tipping decidedly toward the party requesting the preliminary relief." Jackson Dairy v. H.P. Hood Sons, 596 F.2d 70, 72 (2d Cir. 1979). This standard applies to actions alleging violation of the Lanham Act by false advertising or by trade dress infringement. See Castrol, Inc. v. Quaker State Corp., 977 F.2d 57, 62 (2d Cir. 1992) (false advertising); LeSportsac, Inc. v. K Mart Corp., 754 F.2d 71, 74 (2d Cir. 1985) (trade dress).

II. The Lanham Act Trade Dress Infringement Claim

"Trade dress" can refer either to a product's packaging or to its design. Wallace Int'l Silversmiths v. Godinger Silver Art Co., 916 F.2d 76, 79 (2d Cir. 1990), cert. denied 499 U.S. 976, 111 S.Ct. 1622, 113 L.Ed.2d 720 (1991). The trade dress of a product is its total image and may include features such as size, shape, color or color combinations. Stormy Clime Ltd. v. Pro-Group, 809 F.2d 971, 974 (2d Cir. 1987) (citing John H. Harland Co. v. Clarke Checks, 711 F.2d 966, 980 (11th Cir. 1983)). In order to prevail on its trade dress infringement claim, Kompan must first show that its trade dress is protectable under the Lanham Act because it is inherently distinctive or, if its trade dress is not inherently distinctive, that it is protectable because it has acquired secondary meaning. Villeroy Boch Keramische Werke K.G. v. THC Systems, 999 F.2d 619, 620 (2d Cir. 1993). Kompan must then show that there is a likelihood of confusion between its product and PSI's. Paddington Corp. v. Attiki Importers Distributors, 996 F.2d 577, 582 (2d Cir. 1993). In a Lanham Act case, showing a likelihood of confusion also can establish irreparable harm. Hasbro, Inc. v. Lanard Toys, 858 F.2d 70, 73 (2d Cir. 1988). Kompan urges that it has shown a likelihood of success as to distinctiveness, secondary meaning, and likelihood of confusion. PSI, in addition to denying Kompan's claims, also affirmatively claims that Kompan's trade dress is functional and thus not protected by the Lanham Act. See Wallace, 916 F.2d at 79; Stormy Clime, 809 F.2d at 974. I consider each of these issues — protectability, likelihood of confusion, and functionality — in turn.

A. Protectability. Trade dress is protectable if it is "capable of identifying products or services as coming from a specific source" and thus inherently distinctive. Two Pesos, Inc. v. Taco Cabana, Inc., ___ U.S. ___, ___, 112 S.Ct. 2753, 2760, 120 L.Ed.2d 615 (1992). Because producers have an almost unlimited choice in design, trade dress choices will normally be inherently distinctive. Cf. Paddington, 996 F.2d at 583 (package design). However, if a particular industry customarily uses a certain trade dress — like the packaging of lemon-lime soda in green 12-ounce cans — then the trade dress is generic rather than distinctive and is not entitled to protection. Id. Finally, descriptive trade dress such as the use of a shining car to illustrate a bottle of car wax is not distinctive, id. at 584, but may be protectable if the plaintiff shows that its trade dress has acquired secondary meaning, Two Pesos, ___ U.S. at ___, 112 S.Ct. at 2758.

Kompan claims as its trade dress its configurations of certain design elements such as stylized, soft-edged abstracted but recognizable shapes of animals, vehicles, houses and castles; superimposed flat panels; bright primary colors; patented self-locking shock resistant protective caps known as "DO-NUTS"; and durable springs coated with non-toxic blue paint. See Am. Compl. ¶¶ 17-18; Olson Decl. I ¶ 16. Kompan seeks a determination that the trade dress of its entire early childhood product line is distinctive because the same distinct design elements and configurations of design elements transcend differences among individual products and define the product line. Olson Decl. IV ¶ 3. PSI, on the other hand, argues that no particular design element or configuration of elements (except the blue spring which PSI did not copy) used by Kompan is distinctive either as to an individual product or as to the entire early childhood product line. PSI believes that each of the particular design elements Kompan cites is generic because many manufacturers of playground furniture use one or more of these elements. Bayman Decl. I ¶¶ 7-8; Def.Ex. A-G, I, K-M and DD.

I turn first to PSI's argument that Kompan has no distinctive trade dress because many of the specific elements in Kompan's claimed trade dress are commonly used in the industry and therefore generic. PSI points out that it is common practice in the industry to fabricate playground equipment from flat panels in stylized shapes of animals or objects attractive to children, and that soft corners and protective white and colored caps for protruding components are common safety features. Bayman Decl. I ¶¶ 7-12, 14-18, Def.Ex. A-G, I, K-M. PSI's argument misses the mark in two respects. First, some of Kompan's design elements are truly unique. Although rounded corners and stylized shapes may be common in the industry as PSI urges, Kompan's use of curved shapes goes beyond the functional necessity to protect children from sharp edges to a design choice that is made even when an angle or straight line would not be dangerous. Second, even if each of the design elements in a product's trade dress would not be inherently distinctive, the court must look at the combination of elements and the total impression it gives the observer in order to determine whether the trade dress is distinctive. Paddington, 996 F.2d at 584. A comparison of the products in other manufacturer's lines to Kompan's shows that, as to most of those products, no other manufacturer combines even the generic design elements in ways similar to Kompan. Compare Olson Decl. IV, Ex. W with, Ex. Y — RR. The only exception is Kompan's Crazy Scrambler™. The Crazy Scrambler™ has a generic trade dress used by several manufacturers to convey the idea of a motorcycle; this generic trade dress does not give as much emphasis to the characteristic Kompan design elements as does the remainder of the early childhood line. See Def.Ex. N. Kompan has therefore established a likelihood of success on its claim that all of its products except the Crazy Scrambler™ have a distinctive trade dress that is protectable.

Kompan has also established a likelihood that it will be able to demonstrate that its distinctive trade dress exists throughout its early childhood line. Courts in this Circuit — or applying this circuit's law — have found an entire product line's trade dress to be distinctive. See, e.g., Imagineering, Inc. v. Van Klassens, Inc., 53 F.3d 1260 at 1264 (Fed. Cir. 1995) (applying Second Circuit law); Life Indus. Corp. v. Star Brite Distrib., 31 F.3d 42, 46 (2d Cir. 1994); Saban Entertainment v. 222 World Corp., 865 F. Supp. 1047, 1055 (S.D.N.Y. 1994). Product line trade dress has been found both when the trade dress is virtually identical across the product line, see Life Indus., 31 F.3d at 42, 46, and when certain strong features marking the trade dress are found throughout the product line, see Imagineering, 53 F.3d at 1264, and Saban, 865 F. Supp. at 1049-50, 1055. Careful review of Kompan's catalog and the catalogs of its competitors demonstrates that Kompan has a distinctive trade dress containing design elements and combinations of elements which recur throughout its product line and are unique in the outdoor playground business. See Olson Decl. IV Ex. W and Z-RR. Kompan emphasizes soft curved shapes, including curved openings, repetitive s-curves or reverse s-curves, and its signature hen-head and stylized beak shapes. When function would allow a choice between a curved shape and an angular or straight line, Kompan always chooses the curved line. Even protective slats on various pieces of equipment incorporate curves. Although other manufacturers sometimes use curved lines, they neither use them as consistently as Kompan nor emphasize curves as design elements in the way that Kompan does. In addition, Kompan uses color in a way not replicated by the other manufacturers. Other manufacturers also use the primary colors, red, blue and yellow, emphasized in Kompan's designs, but none do so as predominantly or in the same combinations of shades and hues. A potential consumer familiar with Kompan's use of color might well recognize a Kompan product by its color combinations alone. Finally, Kompan uses its plastic DO-NUTS as design elements in a way that is not replicated by any of the other manufacturers. The white and bright-colored DO-NUTS stand out in a distinctive way on Kompan's large and small products alike. Kompan has established a likelihood of success on its claim that its product line trade dress is distinctive and thus protectable. As noted previously, however, this distinctive trade dress does not extend to Kompan's Crazy Scrambler™ (M130) which appears to have a generic trade dress.

B. Likelihood of Confusion

Having found that Kompan has established a likelihood of success on its claim that its trademark is distinctive and therefore protectable, I turn now to the issue of likelihood of confusion. Likelihood of confusion refers to the likelihood that an appreciable number of consumers will be misled or confused as to the source of the goods in question. Mushroom Makers v. R.G. Barry Corp., 580 F.2d 44, 47 (2d Cir. 1978), cert. denied 439 U.S. 1116, 99 S.Ct. 1022, 59 L.Ed.2d 75 (1979). This confusion may, but need not, occur at the time of purchase; post-sale confusion as to source is also actionable. Lois Sportswear, U.S.A. v. Levi Strauss Co., 799 F.2d 867, 872 (2d Cir. 1986). In addition, the owner of distinctive trade dress merits protection against the possibility that a consumer will believe imitative goods are sponsored by or associated with the originator. Id.

In evaluating likelihood of confusion in the trade dress area, courts in this circuit weigh eight factors first described by Judge Friendly in Polaroid Corp. v. Polarad Elec. Corp., 287 F.2d 492, 495 (2d Cir.), cert. denied 368 U.S. 820, 82 S.Ct. 36, 7 L.Ed.2d 25 (1961). Paddington, 996 F.2d at 584. These factors are: strength of the trade dress, degree of similarity between the trade dress of the senior and junior users, proximity of the products, likelihood that the prior owner will bridge the gap, actual confusion, the defendant's good or bad faith, the quality of the defendant's product, and the sophistication of the buyers. Polaroid, 287 F.2d at 495. The court should examine each Polaroid factor to determine how it bears on the ultimate question of likelihood of confusion, Lois Sportswear, 799 F.2d at 872, and also weigh any additional appropriate factors that are appropriate, Polaroid, 287 F.2d at 495. The evaluation is not mechanical, i.e. the party with the greater number of factors weighing in its favor does not necessarily prevail. Paddington, 996 F.2d at 584 (citing Physicians Formula Cosmetics v. West Cabot Cosmetics, 857 F.2d 80, 85 (2d Cir. 1988)).

1. Strength of Kompan's Mark

A trade dress's strength is its "distinctiveness . . ., or more precisely, its tendency to identify the goods sold under the [trade dress] as emanating from a particular, although possibly anonymous, source." McGregor-Doniger Inc. v. Drizzle Inc., 599 F.2d 1126, 1131 (2d Cir. 1979). Trade dress that "seems to the consumer uniquely intended to indicate a product's source [is] strong whether or not the consumer is familiar with the [trade dress] or knows the source." Paddington, 996 F.2d at 585. Inherent distinctiveness may suffice to establish the strength of a mark. Id. The court may weigh secondary meaning in determining strength but may not find a mark to be weak solely based on lack of secondary meaning. Id. Because Kompan's trade dress is very distinctive and there are no countervailing factors, I find that Kompan's trade dress is also strong.

2. Similarity Between Kompan's and Karavan's Trade Dress

To assess similarity, a court must consider both similarity of detail and similarity of overall impression. Paddington, 996 F.2d at 586. Minimal differences should not detract from a finding of overall similarity. Id. Because Kompan has a distinctive trade dress across its early childhood product line with the exception of the Crazy Scrambler™, examine not just the similarity between individual products but the overall similarity between the group of Karavan products whose production and distribution Kompan seeks to enjoin and the Kompan early childhood product line. The trade dress of most of the products in PSI's Karavan line bears a striking similarity to analogous products in Kompan's product line, and the total impression of overall similarity goes well beyond what would be likely coincidentally.

a. Karavan's train (Locomotive (Model 1100-1), Passenger Car (Model 1100-2), and Flat Car (Model 1100-3)) bears an extraordinary similarity to Kompan's train (Steam Engine (M525), Railway Carriage (M526) and Goods "Waggon" (M527)). The general configurations of the trains are virtually identical. In addition, each is primarily red with a black curved roof, each has similarly (and unusually) shaped doors and windows incorporating the Kompan curves, and each uses the white plastic "DO-NUTS." PSI's attempt to focus attention on minor differences in color and on the different configuration of the cowcatcher and smoke stack does not detract from the overall impression of similarity.

b. Karavan's car (Model 1200-1) is quite different from Kompan's Mini Car (Model M531). The general configuration of Karavan's car is not like Kompan's, and the use of color differs significantly. However, Karavan's car does use certain elements characteristic of the Kompan trade dress, e.g. the prominent use of the white protective coverings as design elements and non-functional repetitive curved shapes. Standing alone, the similarities between the Kompan car and the Karavan car would not lead to confusion. However, in the context of an entire product line consisting of very similar products, these similarities may be sufficient to produce confusion.

c. Karavan's camper (Model 1200-2) has a moderately strong similarity to Kompan's camper (M532). The exterior shapes of both look like they were stamped from the same mold. Both also have similar although not identical doorways and a similar cut out hole toward the rear and top of the camper. Finally, both have non-functional curvilinear design elements and use the white protective coverings as design elements. The campers are, however, made in different colors and have different albeit minor design details. Their overall similarities are likely to cause confusion.

d. Karavan's Sand Ship (Model 1300) is not very similar to Kompan's Sand Ship (M509 and M519) in appearance. However, Karavan's Sand Ship does use non-functional curves and color in a manner similar to Kompan's general trade dress. In addition, Karavan's Sand Ship is very similar to Kompan's Coaster (MS 10). Olson Decl. IV Ex. W at 72. In the context of the entire line, these similarities are at least somewhat likely to cause confusion.

e. Karavan's Sand Factory (Model 2100) bears a moderate resemblance to Kompan's Sandworks (M581). While there are differences in overall configuration and slight differences in the use of color (Kompan uses red and orange while Karavan uses red, yellow and white), Karavan's choice to duplicate three very significant design elements from Kompan's product, i.e. a beak shape at the top of the sand factory, soft curved panels and repetitive, abstract hens' heads contributes to a likelihood of confusion. In addition, some of the apparent differences may result from different camera angles.

f. Karavan's Wide Tot Slide (Model 2800) differs significantly from Kompan's Slide and Cave (M322) in both color and design elements. However, PSI has again chosen to incorporate design elements intrinsic to the Kompan trade dress: these include the soft-edged superimposed curvilinear panels and white protective coverings. In addition the overall configurations of the two slides are similar.

g. Despite minor differences in color and design, Karavan's Sand House (Model 2200) and Pirate's Ship (Model 2300) produce the same overall impression on the observer as Kompan's Sand House (M507) and Sand Ship (M519).

h. Karavan's Slide House (Model 2500) bears a moderate resemblance to Kompan's Toddler's Den and Slide (M617 and M313). The overall configurations of the two are similar, but there are significant differences both in color and design features. However, PSI has again elected to use design elements that are part of the Kompan trade dress, e.g. curved shapes for the openings into its house, decorative slats and staircase and the white protective coverings.

i. Karavan's Magician's House (Model 2600) is quite similar to Kompan's Magic House (M650) in overall configuration. Different color choices and a design that differs in minor details do not detract from the overall impression of similarity.

j. Karavan's elephant slide (Model 2700) and Kompan's Baby Elephant Slide (M304) use distinctly different colors and have a somewhat different overall configuration. However, Karavan's use of overlapping panels, curvilinear shapes and white protective caps as design elements mimics Kompan's trade dress.

k. Karavan's 4-Seat Bouncer (Model 4010) and Kompan's Crazy Daisy™ (M128) have strongly similar overall configurations with some differences in color and minor design elements.

l. Karavan's animal bouncers — the Goose Bouncer (Model 4020), Elephant Bouncer (Model 4030), Turtle Bouncer (Model 4040), Hen Bouncer (Model 4050), Panda Bear Bouncer (Model 4060), Dinosaur Bouncer (Model 4070), Bunny Bouncer (Model 4080), and Seal Bouncer (Model 4120) — have configurations that are almost identical to Kompan's Crazy Gander™ (M106), Crazy Nellie™ (M123), Crazy Turtle™ (M124), Crazy Hen™ (M100), Crazy Bear™ (M127), Springosaurus™ (M113), Crazy Rabbit™ (M117) and Crazy Sea Lion™ (M118). PSI has also used almost identical color choices for its bouncers with the exception of the dinosaur and seal. Overall, the similarity between Karavan's and Kompan's animal bouncers is extraordinary.

m. I have found that Kompan has not shown a likelihood that it can prove that the trade dress of its Crazy Scrambler™ (M130) is distinctive. However, Kompan's Crazy Racer™ (M158) adds a side car that uses distinctive Kompan design elements, making the trade dress of the Crazy Racer distinctive. Karavan's Motorcycle with Side Car Bouncer (Model 4110) has a side car that is virtually identical to the sidecar in Kompan's Crazy Racer.™ The addition of the sidecar makes the overall appearance of Karavan's Motorcycle with Side Car Bouncer quite similar to that of Kompan's Crazy Scrambler™.

n. Karavan's Group Seesaw (Model 4500) is a virtually identical copy of Kompan's Multi Seesaw (M146) with only minor differences in color and design detail.

o. Karavan's two tables, the Round Table (Model 5300) and the Picnic Table (Model 5400), closely resemble Kompan's Round Table (M200) and Garden Table (M230) although there are color and design detail differences.

In addition to emphasizing differences in configuration, color, and design detail between Kompan's individual products and its own, PSI urges that three general distinctions — PSI's use of a small metal plate with its trade name on each product, the plastic composition of PSI's product line as opposed to the wood composition of Kompan's, and PSI's use of a grey enclosure for its springs in contrast to Kompan's blue exposed spring — sufficiently differentiate the two lines.

A prominent, permanent label tends to dispel confusion as to source and is highly relevant to the inquiry on similarity. Bristol-Myers Squibb Co. v. McNeil-P.P.C., Inc., 973 F.2d 1033, 1046 (2d Cir. 1992). However, Karavan's yellow label is approximately three inches by two inches, quite small in light of the size of most of the playground equipment. The Karavan label does not appear in any of Karavan's catalog pictures. Def. Ex. U, V and X. Under these circumstances, the label is unlikely to dispel confusion as to source.

Similarly, viewing the catalog pictures as well as Kompan's Crazy Rabbit™ and Crazy Hen™ and PSI's Bunny Bouncer and Hen Bouncer "in person" convince me that neither the difference between wood and plastic nor the different spring colors are likely to detract from the overall impression of similarity. Although a close inspection reveals that Kompan uses wood and a blue spring and PSI uses plastic and a spring enclosed in a grey plastic sleeve, these details do not stand out from a distance of more than a few feet or in their catalog presentation. Based on the proof offered to date, Kompan has established that the similarities outweigh the differences both as to most of the products examined and as to the two groups of products. Even where there are striking differences between products, the use of design elements intrinsic to Kompan's trade dress contributes to a likelihood of confusion in the context of two lines of playground furniture which are very similar overall. Kompan has also established that the similarities contribute to a likelihood of confusion.

Kompan argues that similarities in trade name — Karavan and Kompan — and in catalog format should also be weighed in assessing similarity. The Karavan catalog format does not by itself closely approximate Kompan's catalog. Any confusion caused by the catalog can be attributed either to the similarities in the products themselves or to the similarities in trade names. The Kompan and Karavan trade names are similar in both configuration and sound. See Grotrian, Helfferich, Schulz, Th. Steinweg Nachf. v. Steinway Sons., 523 F.2d 1331, 1340 (2d Cir. 1975) (holding that "[t]rademarks, like small children, are not only seen but heard.") PSI argues that these similarities are effectively ameliorated because Kompan follows its name with a stick figure of a child while PSI encloses the Karavan logo in the outline of a train. However, PSI adds to the confusion caused by the similar names by failing to indicate anywhere in its catalog that the Karavan line is produced by Park Structures, Inc. Viewed in an overall context that includes similar configuration and sound as well as PSI's decision to leave its own name off the Karavan catalog, the Karavan trade name does contribute to a likelihood of confusion. The remainder of the Karavan catalog format does not contribute to a likelihood of confusion.

3 4. Proximity and Bridging the Gap

The proximity factor assesses the degree to which two products compete with each other. Lang v. Retirement Living Publishing Co., 949 F.2d 576, 582 (2d Cir. 1991). Similarity in trade dress is more likely to cause confusion if the products serve the same purpose, fall within the same general class or are used together. Id. Because Kompan's and PSI's playground furniture compete in the same narrow market, they are highly proximate. See Hasbro, 858 F.2d at 77.

Bridging the gap refers to the likelihood that the senior user of a trade mark or trade dress will enter the market in which the junior user operates. Id. at 78. This factor favors a finding of likelihood of confusion if there is no gap to be bridged, i.e. the junior user competes directly with the senior user. Id. Here there is direct competition between plaintiff and defendant. Therefore this factor, too, increases the likelihood of confusion.

5. Actual Confusion.

The actual confusion factor looks to whether any consumers have actually been confused by the similarities in trade dress. Cf. Centaur Communications v. A/S/M Communications, 830 F.2d 1217, 1227 (2d Cir. 1987). Evidence of actual confusion is not required to show likelihood of confusion. Id. In addition, a plaintiff who moves promptly to stop infringement should not be penalized because she cannot prove actual confusion. Lois Sportswear, 799 F.2d at 875.

PSI's Karavan line has been on the market for only a limited time. Despite this fact, Kompan has introduced some evidence of actual confusion. Kompan attempts to show actual confusion by evidence (1) that Kompan received a phone call from an unnamed consumer who had seen the Karavan line and believed Kompan was introducing a new plastic line, (2) that a customer pulled a bid specifying Kompan equipment in order to be able to consider Karavan equipment because the two lines looked similar and Karavan's equipment was less expensive, and (3) that Kompan received a letter from Candace Brewer, a landscape architect, who upon reading the Karavan catalog first believed that the catalog came from Kompan and later believed that Kompan must have licensed Karavan to market its playground furniture. Olson Decl. III Ex. R. The pulling of the bid does not indicate actual confusion because the consumer apparently realized that the two lines had different sources and simply wanted to comparison shop. However, the other two examples are some, albeit limited, evidence of consumer confusion. Given the limited period of time that the Karavan line has been on the market, this evidence of actual confusion favors plaintiffs.

6. Good or Bad Faith

Kompan has shown that its trade dress existed for a substantial period of time prior to PSI's introduction of the Karavan line. Kompan has also shown that PSI solicited possible representatives by indicating that its Karavan line is intended to compete with Kompan. Compl. Ex. C, Olson Decl. I Ex. K. PSI has not offered any explanation of the similarities between its line and Kompan's.

Actual or constructive knowledge of a prior user's trade dress along with "similarities so strong that it seems plain that deliberate copying has occurred" support a finding of bad faith. Paddington, 996 F.2d at 587. Intentional copying also gives rise to a presumption that the copier has succeeded in causing confusion. Id. at 586. Here because Kompan and PSI compete in the same field, PSI must have been aware of Kompan's trade dress. Therefore, the unexplained and striking similarities between Kompan's trade dress and Karavan's create both a presumption of likelihood of confusion and a strong showing on bad faith.

7. Quality of the Junior User's Product

Kompan argues that PSI's Karavan line has several safety defects. The alleged defects include a probability that the concrete below PSI's bouncers will become exposed, badly positioned handles on certain models, a failure to cover all nuts and bolts, and inadequately protected springs. Statements of plaintiff's counsel at preliminary injunction argument; Olson Decl. I ¶¶ 27-28; Olson Decl. III ¶¶ 7, 9. PSI responds that it has never been the target of a products liability action, that its products comply with Consumer Product Safety Commission Standards while Kompan's products show areas of noncompliance, and that its plastic is safer and sturdier than Kompan's wood. Bayman Decl. I ¶¶ 3, 5; Fedele Decl. ¶ 4. These conflicting claims must await trial for resolution. At this point, Kompan has not established that the quality of PSI's Karavan line would affect the overall analysis of likelihood of confusion.

8. Sophistication of the Relevant Consumer Group

Unsophisticated consumers aggravate the likelihood of confusion. Hasbro, 858 F.2d at 79. However, sophistication may, on occasion, also increase the likelihood of confusion. Centaur Communications, 830 F.2d at 1228 (citing Grotrian, 523 F.2d at 1341-42).

Kompan and PSI's consumers include governmental authorities, schools, day care centers, and private contractors. Bayman Decl. I ¶ 19; Olson Decl. II ¶ 13, Ex. P. PSI emphasizes that governmental consumers operate via a bid process which makes confusion at the time of purchase extremely unlikely. Bogan Decl. I ¶ 14; Bayman Decl. I ¶ 20. Kompan, on the other hand, points out that many of its customers are day care centers and not sophisticated. Olson Decl. II ¶ 13. Kompan also argues that it need not establish that consumers are likely to buy PSI's products because the consumer believes it is made by Kompan. Kompan argues correctly that it can prevail by showing that confusion between the Kompan and Karavan lines and names will mistakenly lead the consumer to believe there is some connection between the two and therefore develop an interest in the Karavan line that it would not otherwise have had. See Mobil Oil Corp. v. Pegasus Petroleum Corp., 818 F.2d 254, 260 (2d Cir. 1987); Grotrian, 523 F.2d at 1342.

Because playground furniture is expensive and buyers often buy after a bid process, the relevant consumer market is, with exceptions, likely to be relatively sophisticated. However, as shown in the letter from Candace Brewer to Kompan, this sophistication cannot protect Kompan against initial confusion by consumers who will assume either that PSI's product is made by Kompan or that Kompan sponsors the Karavan line. Olson Decl. III Ex. R. This initial confusion may create an injury actionable under the Lanham Act. Pegasus Petroleum, 818 F.2d at 260.

9. Overall Likelihood of Confusion

Kompan has made a sufficient showing on the likelihood of confusion between its products and Kompan's. Kompan has strong trade dress, the majority of products in PSI's Karavan line are very similar in trade dress to their analogs in the Kompan line, and the two product lines compete head to head. These three factors — strength, similarity and competitive proximity — are "perhaps the most significant in determining the likelihood of confusion. . . ." Pegasus Petroleum, 818 F.2d at 258 (citing Vitarroz Corp. v. Borden, Inc., 644 F.2d 960, 966 (2d Cir. 1981)). In addition, Kompan has produced persuasive evidence that PSI adopted its trade dress in bad faith. Kompan, of course, need not show that it intends to bridge the gap between its products and PSI's because there is no gap; both companies compete for the same market. Finally, the three areas in which Kompan's showing is less strong — actual confusion, quality of the junior user's product and sophistication of the relevant consumer group — do not outweigh its strong showing in other areas. As noted previously, Kompan has offered some evidence of actual confusion and could not be expected to offer extensive evidence at this stage of the litigation. Because the record does not clearly establish the quality of the PSI line relative to Kompan's line, this factor cuts neither for nor against Kompan. Finally, although the relevant consumer group may be fairly sophisticated, its sophistication cannot protect Kompan against the initial interest PSI may receive because of buyers' assumptions that the two companies are related.

C. Functionality

PSI argues that despite Kompan's showing on distinctiveness and likelihood of confusion, its trade dress claim fails because Kompan's trade dress is functional. Functional trade dress is not protectable under the Lanham Act. L.A. Gear, Inc. v. Thom McAn Shoe Co., 988 F.2d 1117, 1129 (Fed. Cir.) (applying Second Circuit law), cert. denied, ___ U.S. ___, 114 S.Ct. 291, 126 L.Ed.2d 240 (1993). PSI has the burden of proof on the issue of functionality. LeSportsac, 754 F.2d at 75. Trade dress is functional if it is essential to the use or purpose of the article or affects its cost or quality. L.A Gear, 988 F.2d at 1129. The concept of functionality can also apply — on a limited basis — to strictly ornamental features. Wallace, 916 F.2d at 79-80. A particular "design is functional because of its aesthetic value only if it confers significant benefit that cannot practically be duplicated by the use of alternative designs." Restatement (Third) of Unfair Competition § 17 cmt. c (Tentative Draft No. 2, 1990) ( quoted in Villeroy Boch, 999 F.2d at 621).

PSI's functionality claim has two aspects. First, PSI claims that some of the elements featured in Kompan's trade dress, e.g. the flat panels and the safety caps, are functional because they reduce the cost of the product or are essential to the safe use of the product. Second, PSI claims that each of the elements of Kompan's trade dress is aesthetically functional, i.e. necessary to compete in the market place.

Both of PSI's functionality claims fail because they rely on a disaggregation of the design elements in Kompan's total "look." See Def. Mem. at 9-13. PSI argues that because all of the individual design elements in Kompan's look are functional, Kompan cannot combine these elements to make a non-functional overall look. Id. at 9. "[B]reaking [Kompan's] trade dress into its individual elements and then attacking certain of those elements as functional . . . misconceives the scope of the appropriate inquiry." LeSportsac, 754 F.2d at 76. Although Kompan's complaint can be read to seek protection for both its design elements and the particular combinations of those elements it has used, Am.Compl. at ¶¶ 17-19, Kompan concedes that it is entitled to protection only for the combinations it has chosen, not for design elements such as colors or animal shapes, Pl.'s Reply Mem. at 5. PSI must therefore show that it needed to copy Kompan's overall design in order to compete in the market. Villeroy Boch, 999 F.2d at 621. The functionality test must focus "upon hindrances to legitimate competition [in order to] accommodate consumers' somewhat conflicting interests in being assured enough product differentiation to avoid confusion as to source and in being afforded the benefits of competition. . . ." Stormy Clime, 809 F.2d at 978.

PSI has made no attempt to show that it must replicate Kompan's combinations of design elements in order to compete in the market. PSI relies, instead, on its claim that the individual elements of Kompan's trade dress are functional. To show that individual elements of Kompan's trade dress are functional, PSI has offered pictures of products made by some of Kompan's competitors that use some of these elements. See Def.Ex. A-G, I, L-M. Because the competitors use the elements in ways that differ from Kompan's usage, PSI's own exhibits tend to show that it is not necessary to copy Kompan's overall "look" in order to compete in the market place. PSI has therefore failed to counteract Kompan's showing on distinctiveness and likelihood of confusion by showing that Kompan's trade dress is functional. As noted above, a showing of likelihood of success also creates a presumption of irreparable harm. PSI has done nothing to rebut this presumption, so Kompan is entitled to injunctive relief on its trade dress claim.

III. False Advertising Claim

Kompan also claims that PSI violates the Lanham Act by false advertising. Section 43(a) of the Lanham Act protects against two distinct wrongs: false designation of origin and false description or representation (false advertising). 15 U.S.C. § 1125(a)(1)(A) and (B); Agee v. Paramount Communications, 853 F. Supp. 778, 790 (S.D.N.Y. 1994), aff'd in part, rev'd in part on other grounds, 59 F.3d 317 (2d Cir. 1995); cf. Forschner Group, v. Arrow Trading Co., 30 F.3d 348, 357, 360 (2d Cir. 1994) (holding that defendant's use of the phrase "Swiss Army knife" did not constitute false advertising under section 43(a) of the Lanham Act but remanding to consider whether, among other things, use of the same phrase constituted false designation of origin). Kompan blurs the line between these two distinct causes of action by premising its false advertising claim, in part, on the appearance of PSI's products as they appear in its catalog and on the use of color, typeface and layout in the catalog. See Pl.'s Mem. at 17-18. Kompan claims that PSI's similar product depictions and similar color, typeface and layout make an implicit claim that the Karavan line is similar to or superior in quality to Kompan's line. This claim is more properly seen as a false designation of origin claim, and I have addressed it as such in Part II of this opinion. Cf. Forschner Group, 30 F.3d at 357, 360.

Kompan also claims, however, that PSI engaged in false advertising by literal claims that its Karavan Line was superior to Kompan's line. Kompan finds these claims in a letter written by PSI to a dealer. See Compl.Ex. C. The letter states, "Karavan is intended to compete directly with Kompan in the early childhood market with a superior product line and lower prices." Id. Kompan has failed to show a likelihood of success on this aspect of its false advertising claim. When a defendant's advertisement claims that its product is superior to the plaintiff's, the plaintiff must affirmatively prove that the defendant's product is equal or inferior to the plaintiff's product. Castrol, 977 F.2d at 63. Kompan's claims of PSI's inferior quality rest on alleged safety defects. See Olson Decl. I ¶ 27; Olson Decl. III ¶¶ 8-10. PSI responds by denying Kompan's claims of safety defects, pointing out alleged safety problems with Kompan's products, and arguing that its claim of superiority is justified because its line is more wear resistant and safer than Kompan's. Bayman Decl. I ¶¶ 3, 5, 21-22; Bogan Decl. I ¶¶ 9-13; Fedele Decl. ¶ 4, Roehrig Decl. ¶ 4 and Ex. AA; Bayman Decl. III ¶ 4. The claims made on both sides lack independent substantive documentation and tend to cancel each other out. At this juncture, Kompan has not established a likelihood of success or serious issues going to the merits on its false advertising claim.

IV. New York State Common Law and Statutory Claims

Kompan also seeks injunctive relief under New York's common law and a variety of New York State statutes.

In its third cause of action, Kompan alleges unfair competition. Under New York's common law of unfair competition, a producer's trade dress is protected against practices "imbued with an odor of bad faith" including palming off and deliberate copying. Laureyssens v. Idea Group, 964 F.2d 131, 138 (2d Cir. 1992). To support a claim of unfair competition, plaintiff must also establish likelihood of confusion. Stern's Miracle-Gro Products v. Shark Products, 823 F. Supp. 1077, 1093 (S.D.N Y 1993). As explained earlier, Kompan has made a strong showing on both likelihood of confusion and bad faith and therefore is alternatively entitled to preliminary injunctive relief on its common law claim.

Plaintiff's fourth and seventh causes of action alleges deceptive acts and practices and false advertising in violation of sections 349 and 350 of New York's General Business Law. Most trademark and trade dress claims fall outside the ambit of sections 349 and 350. Jaret Int'l v. Promotion in Motion, 826 F. Supp. 69, 78 (E.D.N.Y. 1993) (citing R. Givens, Practice Commentaries on N YGen.Bus. Law § 349 at 567 (McKinney 1988) . . ., quoted in Bristol-Myers Squibb Co. v. McNeil-P.P.C., 786 F. Supp. 182, 215 (E.D.N.Y.), aff'd in part, vacated in part on other grounds, 973 F.2d 1033 (2d Cir. 1992)). In order to state a claim under section 349 or section 350, a competitor must allege the sort of harm to the public interest that would trigger FTC intervention under 15 U.S.C.A. § 45. Cf. Bristol-Myers, 786 F. Supp at 215-216 (showing of intentional copying insufficient for a section 349 violation where there was no showing of substantially different ingredients or any risk to public health in the sense contemplated by the statute). Although plaintiff argues that defendant's product is unsafe, it has not, as detailed above, made a sufficiently strong showing on this point. To the extent that plaintiff's claims of violation of Gen.Bus.L. § 349 or § 350 rest on allegations of false advertising, plaintiff has failed to show a likelihood of success for the same reason that it has failed to show a likelihood of success on its Lanham Act false advertising claim. Cf. Towers Financial Corp. v. Dun Bradstreet, 803 F. Supp. 820, 823 (S.D.N.Y. 1992) (holding that both Lanham Act false advertising claim and a false advertising claim made under section 349 or section 350 require a showing that the advertisement was false or misleading).

In its fifth cause of action, Kompan alleges violation of New York's anti-dilution statute. N.Y.Gen.Bus. Law § 368-d. This statute requires that plaintiff show that it has an exceptionally strong mark. Sally Gee, Inc. v. Myra Hogan, Inc., 699 F.2d 621, 625 (2d Cir. 1983) (citing Allied Maintenance Corp. v. Allied Mechanical Trades, 42 N.Y.2d 538, 545-46, 399 N YS.2d 628, 369 N.E.2d 1162 (1977)). A determination under the Lanham Act that a mark is distinctive does not mean that it should be considered exceptionally strong for purposes of section 368-d. Id. At this early point in the litigation, I do not believe Kompan has shown a likelihood that it can prove an "exceptionally strong" trade dress.

Kompan's sixth cause of action alleges violation of N YGen.Bus.L. § 133. Section 133 governs the use of names or addresses in trade. Because Kompan has not sought to enjoin PSI's use of a trade name or address, section 133 has no apparent relevance to this request for injunctive relief.

Finally, in its eighth cause of action, Kompan seeks recovery for unjust enrichment. Kompan alleges that PSI will be unjustly enriched by its trade dress infringement and false advertising. To prevail on a claim of unjust enrichment in the trade dress area, Kompan must show that PSI has received a benefit at plaintiffs' expense "under such circumstances that in equity and good conscience he ought not retain." Yankee Publishing v. News America Publishing, 809 F. Supp. 267, 282 (S.D.N.Y. 1992) (quoting Miller v. Schloss, 218 N.Y. 400, 113 N.E. 337 (1916). Except as a measure of damages, plaintiff's unjust enrichment claim adds little to its trade dress infringement claim. The injunction entered on the trade dress infringement claim will address any harm plaintiff might suffer on its unjust enrichment claim.

V. Scope of the Injunction

Kompan has requested broad-based injunctive relief including a recall order and an order directing the defendants not to fill pending orders for infringing materials.

Recall is a drastic remedy that may impose a substantial burden on defendants. Nevertheless, it can be an appropriate remedy within the discretion of the district court at least in the context of a permanent injunction. See Nikon Inc. v. Ikon Corp., 987 F.2d 91, 97 (2d Cir. 1993) (citing Perfect Fit Indus. v. Acme Quilting Co., 646 F.2d 800, 805 (2d Cir. 1981), cert. denied, 459 U.S. 832, 103 S.Ct. 73, 74 L.Ed.2d 71 (1982). Some authority also supports the use of a recall provision in a preliminary injunction. See Hukafit Sportswear v. Banff, Ltd., 1985 WL 5943 at *3-*7 (S.D.N.Y. 1985) (and cases collected). Here, Kompan — perhaps due to the early stage of this litigation — has given no real sense of the harm that will ensue if a recall order is not granted. In addition, although PSI has offered no proof on the subject, the very size of the playground equipment at issue in this case suggests that PSI could not recall products without significant expense. A court ordering recall must first weigh the likely burden and expense of a recall. Perfect Fit, 646 F.2d at 807; see also Bausch Lomb v. Nevitt Sales Corp., 810 F. Supp. 466, 478 (W.D.N.Y. 1993) (declining to order recall based on incomplete record and conflicting evidence concerning the number of infringing products in the marketplace). I therefore deny Kompan's request for recall without prejudice to reapplication based on a more complete record.

Kompan is, however, entitled to an injunction against filling currently unfilled orders. This provision merely preserves the status quo. The injunctive relief ordered below includes products that the defendants have in their possession and would otherwise use to fill orders.

Kompan has also requested that defendants be required to file a report detailing their compliance with the injunction and identifying with specificity all new trade dress and advertising proposed for use in lieu of the offending materials. This request is granted.

Finally, Federal Rule of Civil Procedure 65(c) requires the court to condition the granting of an injunction upon the posting of security adequate to protect against damages incurred by reason of a wrongful injunction. Fed.R.Civ.P. 65(c). The parties' submissions on the size of a potential bond have been singularly unhelpful. Although specifically invited to submit additional material on this issue, neither party has made a submission in evidentiary form. Kompan urges that no bond or a minimal bond be set based on the action of other district courts in trademark cases which may or may not have involved a similar risk of loss to the defendant. Pl.'s Letter Br. 6/12/95 at 4. PSI estimates, without supporting detail, that it will lose between $500,000 and $2,000,000 primarily in the form of goodwill if an injunction is entered. Roehrig letter 6/12/95. Because PSI has not supported its estimate of loss, I direct Kompan to post a bond in the amount of $50,000 without prejudice to an application for modification from either party based on a competent showing.

CONCLUSION

Park Structures, Inc., Karavan, Kay and Alan Bayman and Baughman Bros. Inc., their officers, agents, sales representatives, distributors, employees, subsidiaries, divisions, successors and assigns, and all persons in active concert, participation, and combination with defendants are therefore enjoined from manufacturing, producing, distributing, circulating, selling, offering for sale, advertising, promoting or displaying any of the following products: Model # 1100-1 (Locomotive), Model # 1100-2 (Passenger Car), Model # 1100-3 (Flat Car), Model # 1200-1 (Car), Model # 1200-2 (Camper), Model # 1300 (Sand Ship), Model # 2100 (Sand Factory), Model # 2200 (Sand House), Model # 2300 (Pirate Ship), Model # 2500 (Slide House), Model # 2600 (Magician's House), Model # 2700 (Elephant Slide), Model # 2800 (Wide Tot Slide), Model # 4010 (4-Seat Bouncer), Model # 4020 (Goose Bouncer), Model # 4030 (Elephant Bouncer), Model # 4040 (Turtle Bouncer), Model # 4050 (Hen Bouncer), Model # 4060 (Panda Bear Bouncer), Model # 4070 (Dinosaur Bouncer), Model # 4080 (Bunny Bouncer), Model # 4110 (Motorcycle with Side Car Bouncer), Model # 4120 (Seal Bouncer), Model # 4500 (Group See-Saw), Model # 5300 (Round Table), and Model # 5400 (Picnic Table).

In addition, Park Structures, Inc. is directed to file with the court within 30 days a report detailing its compliance with this injunction and, if it wishes to issue some or all of the same products with a different trade dress, identifying with specificity that trade dress.

Finally, Kompan is directed to post a bond in the amount of $50,000.

IT IS SO ORDERED.