Full opinion text
FINDINGS OF FACT AND CONCLUSIONS OF LAW MOTLEY, District Judge. INTRODUCTION AND BACKGROUND On December 10, 2001, plaintiffs Cartier and Cartier International, B.V. initiated litigation against defendants Four Star Jewelry Creations, Inc., Crown Jewelry Creations, Inc., and Globe Jewelry, Inc. alleging trade dress infringement under Section 43(a) of the Lanham Act, infringement of a registered mark under Section 32 of the Lanham Act, trademark infringement under Section 349 of the New York General Business Law, patent infringement, false designation of origin, and trademark dilution. These claims relate to Cartier’s Tank Francaise, Tank Ameri-caine, and Panthere lines of watches. On April 30, 2002 defendants filed their answer and counterclaims, denying plaintiffs’ allegations and seeking a declaratory relief ruling that plaintiffs’ alleged trade dress and design patents are invalid, unenforceable and were not infringed. On February 28, 2003, plaintiffs moved by Order to Show Cause to amend their complaint and for a preliminary injunction ordering, inter alia, cessation of further sales of a watch depicted in defendants’ 2002 catalog, which plaintiffs contended infringed upon their alleged trade dress in their Pasha de Cartier watch and a registered trade mark for the associated Grille design. Defendants did not oppose Cartier’s motion for a preliminary injunction as to the Pasha watches. The request was granted on May 8, 2003 and plaintiffs filed their first amended complaint five days later. After extensive discovery, this case proceeded to bench trial beginning May 10, 2004, and closing May 25, 2004. Both parties submitted their proposed findings of fact and conclusions of law on July 26, 2004. The following are the court’s findings of fact and conclusions of law pursuant to Fed.R.Civ.P. 52(a). FINDINGS OF FACT 1. THE PARTIES Plaintiff Cartier is a division of Riche-mont North America (“Richemont”) and the U.S. selling arm of that company. Tr. at 516. Plaintiff Cartier International B.V. is a Dutch company holding all of Cartier’s international intellectual property rights and licenses thereto. Id. Plaintiffs are referenced hereinafter collectively as “Cartier.” They sell luxury watches in exclusive boutiques, high-end department stores, and authorized retailers throughout the United States. Cartier’s watches relevant to this litigation range from $2,000 to $150,000. Defendants Four Star Jewelry Creations, Inc. (“Four Star”), Crown Jewelry Inc. (“Crown”) and Globe Jewelry, Inc. (“Globe”) are New York corporations owned and operated by Abraham Malek and Mike Genuth. Tr. at 617, 624, 626. Although there are three corporate defendants, they are, in effect, the same company; Crown and Four Star send out virtually identical catalogs retailing Globe’s merchandise that they have purchased exclusively from Globe, Tr. at 740-41, 524-25, and all three defendants operate out of the same office on 47th Street in Manhattan. Defehdants manufacture watches, bracelets, and watch attachments that they sell at wholesale prices. Tr. at 624-26. Their typical customers are “mom and pop retail stores,” most of which are jewelry stores and some of which are wholesalers. Tr. at 625-26, 735-36. The average price of defendants’ watches is $1,500 at wholesale. Tr. at 892. II. THE RELEVANT MARKET Ruediger Albers provided expert testimony on behalf of plaintiffs with respect to the luxury watch market. Albers began his employment with the Wempe Corporation-, a German jeweler and watch seller, in 1987; he is now the President of U.S. operations. He has a Masters in watchmaking and is involved in all aspects of the watch-making business, including customer service, sales, and advertising. Tr. at 12-19. ' The court adopts Albers’ testimony that a “luxury” watch generally has a price tag beginning at $600 or $700. Tr. at 25-26. Although one can purchase a watch that performs its time-telling function for approximately $50 or less, luxury watch consumers purchase watches that reflect their lifestyle and express their personality, social' arid financial status and taste. Tr. at 28-29. Most luxury watches are also “status” watches, meaning they have recognizable, distinguished designs that consumers associate with the particular brand; the brand is a further indicator of craftsmanship and quality. Tr. at 31. As a result, the purchaser and wearer of a luxury watch aligns him or herself with the elite status of the watch. The key ingredients to retailing and manufacturing a successful luxury watch are brand name and watch design. Tr. at 156-57. Part and parcel of this endeavor is to maintain limited distribution networks throughout the U.S. such that the watches are only sold to select retailers. Tr. at 61-62. If a brand becomes too commonplace or overexposed, it cheapens the image and lessens customers’ incentive to purchase such an expensive timepiece. Tr. at 62. III. CARTIER’S PANTHERE, TANK FRANCAISE, PASHA, AND TANK AMERICAINE A. The Designs of the Cartier Watches As a preliminary matter, adopting Cartier’s terminology, a watch “family” consists of several watches with the same basic design but often with variations in size, strap, bracelet, dials, metals used, or diamond settings. Even with these variations, the court’s review of the various families reveals that there is a common appearance to all watches within a family. Tr. at 40; PX 73 at 24-31. Each specific variation in the family is known as a “model” and is identified by a specific reference number. Tr. at 24-31. Occasionally, Cartier will market an “animation”; this is a change in the product offered for a period of time to enhance the product’s image and add variety to the family. Tr. at 522. For example, Cartier will offer a family of watches that previously had a white dial around the face with a rose-colored dial framing the face, or a family that previously did not have diamonds with a diamond setting. Id. Cartier charges defendants with infringing the designs of four families of Cartier watches: Panthere, Pasha, Tank Ameri-caine, and Tank Francaise. For each of these families, Cartier’s statement of the elements and combinations thereof that constitute its trade dress are in the Amended Complaint, paragraphs 18, 19, 22, 23, 27, 28, 30, 31, and 34; Tr. at 397. 1. The Panthere The Panthere trade dress consists of the combination of the following design features: a square face with rounded corners; a square bezel or metal frame with rounded corners and screw tips at 1, 2, 4, 7, 8, 10 and 11 o’clock; an outer frame composed of four segments, with the two segments of the outer frame which are horizontal to the frame being rectangular, and the other two pieces of the outer frame which are vertical relative to the case extending horizontally around the corner of the bezel and then vertically away from the case at each corner (both the inner and outer frames may or may not be decorated with jewels); an octagonal winding crown with “cabochon” (i.e., rounded) stone; art deco Roman numerals inclining to conform to the angle-direction of the watch hands at such time as the hands are juxtaposed to that numeral; a “chemin-de-fer” (i.e., railroad) chapter ring or minute guide between the center of the dial and numerals, with every fifth minute indication being thicker and bolder; and a metal bracelet (a watch band made of metal) consisting of five rows of interlocking rectangular links (with the rectangles being approximately three-and-a-half times as long as wide, with the wider side vertical to the watch face), with the links staggered such that the first, third and fifth rows are staggered one-half a rectangle length to the second and fourth rows. The Panthere is offered in gold, steel, and pavee (i.e., paved with diamonds). Tr. at 533. Currently, Cartier does not include a Panthere watch in its line with a leather strap, although ten years ago Cartier offered a gold Panthere watch with a leather band. Tr. at 108. Such Panthere watches may still be sold by Cartier dealers and boutiques who continue to have them in stock, at auctions, trunk shows, and charity sales. Tr. at 555-56. 2. The Pasha - The trade dress for the Pasha watch family consists of the combination of the following design features: a round and thick watch case (i.e., deep) with a thick outer metal bezel surrounding the watch face that slopes downward from the inside of the case to the outside; a removable, screw down cap that covers a central crown, such that in looking at the watch from the front, there are three metals rings of increasing thickness with engraved striations, a fourth thick metal ring, and a cabochon end, with a sapphire or colored stone or rounded stainless-steel dome, all of which is attached to. the watch with a chain link; single horned extensions at the top and bottom of the watch case connecting it to the strap, such that the strap fits between the extensions with only one single extension on each side; a lug that extends through the strap/bracelet and the single horned extension with pyramid-shaped caps; Arabic numerals in curly font. On models with a metal bracelet, the bracelet consist of H-shaped links alternating with rectangular shaped links, the width of the H-shaped link being half that of the rectangular link. Some models within the family contain two smaller crowns positioned above and below the central crown. They are capped with non-removable caps, similarly designed to the central cap, and end in matching cabochon. Some watches in the Pasha family also include a removable grid or “grille.” The Amended Complaint identifies this element as a separate trade dress consisting of two sets of equidistant perpendicular lines forming a tic-tac-toe, three-dimensional pattern that overlays the face of the watch to the edges of the bezel of the watch. The Cartier Pasha Grille Design is registered to plaintiff Cartier. PX 222. The Pasha is offered in stainless steel, chrome, gold, and full pavee with princess cut diamonds. It is also available with a leather strap or metal bracelet. S. The Tank Americaine The Tank Americaine family’s trade dress consists of the following combination of design, features: a rectangular watch face, with the longer side being vertical; the portion of the case that frames the watch face consists of thick vertical and horizontal frames (known as “brancards”) formed of the case metal, with the corner of the vertical brancards elongated beyond the top and bottom of the watch case, ending in rounded, inwardly curving curves (the frames are sometimes decorated with jewels); the elongated brancard corners-partially frame the end links in the bracelet. chain or frame the end of the watch.strap; a slightly curved qase; Art Deco Roman numerals inclined to conform to the angle-direction of the watch hands at such time as the hands are-juxtaposed to that numeral; an octagonal winding crown with a faceted stone; a chemin-de-fer chapter ring or minute guide between the center of the dial and the numerals, with every fifth minute indicator being thicker and bolder. Some models include a metal chain bracelet with five rows of linked rectangular plates (sometimes decorated with jewels), with the width of the third middle row being twice the width of the links in the other rows. The first, third and fifth links are aligned and arranged staggered to the second and fourth row links. The Tank Americaine is offered in stainless steel, gold, and a full pavee. Tr. at 533-34. It is offered with a metal chain bracelet or a leather strap. Three to four years ago, Cartier discontinued the Tank Americaine with a full pavee dial, but such watches may continue to be sold by dealers and jewelers as part of their leftover stock, in trunk shows, or at charity auctions. Tr. at 568-69. k- The Tank Francaise The Tank Francaise’s trade dress is defined by a combination of the following design features: a square watch face; the case framing the watch consists of thin horizontal brancards and thick concave vertical brancards, both formed of the case metal; at each corner the vertical bran-cards are elongated beyond the bottom and top of the watch case, and end in inwardly angled corners; the vertical brancards are beveled to slope downward laterally from the inside to the outside (they are sometimes decorated with jewels); the elongated brancard corners frame the end links in the bracelet chain or the end of the watch strap; Art Deco Roman numerals incline to conform to the angle direction of the watch hands at such time as the hands are juxtaposed to that numeral; an octagonal winding crown with cabochon; a chapter ring/minute guide on the watch face located between the center of the dial and the numerals. Some models include a metal chain bracelet with alternating H-shaped and rectangular links, with the end portions of the H-shaped links mimicking the beveled upper and lower corners of the vertical brancards (the links are sometimes decorated with jewels). Plaintiff Cartier has a registered trademark for the parallel brancards on the Tank Francaise watch case. PX 141, Tr. at 433-34. The Tank Francaise is offered in gold and steel, with or without jewels on the brancards, and with a metal bracelet or leather strap. B. Functionality of Plaintiffs’ Watch Designs Cartier’s director of watch marketing, Kristina Skoczylas, provided credible testimony to the effect that Cartier’s central focus is on the aesthetic value conveyed by the design of the watch; functional superiority is not among Cartier’s objectives. Tr. at 533. Defendants, however, introduced the expert testimony of John T. Houlihan, former Director of Design of the Timex Corporation, who testified on direct at length as to the functional advantages of the discrete elements of each of the watch families. See Tr. at 276-292. Although the court will not restate his testimony in full, the following are a handful of examples Houlihan cited as evidence of the functionality, as opposed to ornamentality, of Cartier’s watch designs: a) with respect to the Tank Francaise, its minute guide allows the watch wearer to determine the time to the minute; the H-shaped links of the metal bracelet keep the bracelet together; the elongated brancard corners protect the end links and link the bracelet chain or watch strap and the watch case; b) with respect to the Tank Americaine, the curvature of the case allows it to conform to the wearer’s wrist; the Roman numerals conform to the angle-direction of the watch hands so that the wearer can see the time; the bracelet is formed of links so that it can comfortably fit on a wearer’s wrist and attach to the watch face; and c) with the Panthere, the bezel holds the crystal and the eight screws connect the bezel to the watch face; the crown is shaped like an octagon to facilitate the user’s ability to turn it; the fifth minute guide indicator is thicker and bolder to allow the wearer to more easily determine the minute. Houli-han’s conclusion, therefore, is that if Cartier were granted exclusive rights to the foregoing features, competitors would be placed at a significant non-reputation related disadvantage and that many features affect the cost or quality of the watches at issue. Tr. 285-292. On cross examination, however, Houli-han conceded that Roman numerals, a cabochon, an octagonal winding stem, a minute track, a particular face shape, a particular shape of a watch case and its extensions connecting the face to the strap, and a particular bracelet link formation or shape are not essential to the function of a watch, nor do they make it any cheaper or easier to manufacture. Tr. at 367-380. The reason that Cartier’s bracelet fits better, for example, is not a function of design, but rather of superior execution. Tr. at 377-378. Finally, Houl-ihan stated that as a watch designer, if he were prevented from using any one of the elements of Cartier’s design, a square face or Roman numerals, for example, this would be a serious limitation; but if he were unable to use all of the various elements in combination, it would not be a serious limitation. Tr. at 381. Adding to the strength of plaintiffs’ case regarding the non-functionality of their watches, Ruediger Albers, who provided expert testimony on behalf of plaintiffs with respect to the luxury watch market, testified that none of the Cartier design features composing its trade dress have any bearing on the watches’ functionality and that there are many other designs available to a watch manufacturer. Tr. at 50, 87-95. In actuality, many of the trade dress features increase the time, difficulty and cost involved in their manufacture. Id. The court agrees with plaintiffs that the watch designs at issue are purely ornamental and do not play a functional, essential, or cost-saving role in the manufacture of watches. Defendants’ expert testimony was unpersuasive in negating the purely aesthetic purpose motivating Cartier’s chosen designs because their expert did not consider the trade dress as a whole but in its discrete parts. Clearly a watch must have a case or a winding stem or numbers or extensions connecting its strap or bracelet to the case; this is a fairly obvious proposition. But what defendants’ expert did not or could not explain is why the particular shapes and combination of design elements are necessary to a watch’s ability to function. The fact that he agreed that a bar on manufacturing a watch with a combination of features composing Cartier’s trade dress as a whole would not seriously limit his options as a watch designer further substantiates the view that the designs are nonfunctional. As such, the court finds that Cartier’s claimed trade dress is not functional. C. Cartier’s Advertising and Promotional Efforts 1. Generally The court credits the testimony of Cartier’s director of watch marketing, Kristina Skoczylas, as to Cartier’s media and advertising efforts. Cartier places print advertising in magazines, newspapers and billboards, both nationally and regionally. Tr. at 536. National advertising is placed and paid for by Cartier itself and is not associated with any particular dealer. Tr. at 536. Regional advertising is specific to a region of the United States and may be associated with a particular dealer. Tr. at 536. Destino also explained that Cartier engages in “co-op” advertising which involves a cost-sharing scheme between Cartier and its dealers, pursuant to which the advertisement includes a short text identifying the particular jeweler or dealer sharing in the cost of the ad. Tr. at 410. For all three types of advertising, the advertising is created by a company hired by Cartier. Cartier’s advertisements adhere to a common theme or design; they exclusively show a Cartier watch in an enlarged size with a dramatic or colored background. Tr. at 40-42. The focus of these advertisements is on the watch; as a general rule, there are no other words or images distracting from the central focus on the timepiece, and the watch itself appears to take up more than 80% of the surface of the advertisement. See PX-202, 245, 412-13. In addition to print advertising, Cartier distributes catalogs via direct mail, through its authorized dealers, and in its boutiques. Tr. at 518; PX 70, 72-75, 78. Cartier also hosts “trunk shows” in which it shows a collection of watches to dealers. Tr. at 538. £ The Panthere Cartier introduced the Panthere in the United States in 1984 with a considerable product-launch event in its Manhattan store featuring two real-life panthers and in its Los Angeles location. Tr. at 416-17. Cartier dealers also replicated Cartier’s launch on a lesser scale in their own locales with local press, clientele and window displays. Tr. at 417-18. Over the next five to six weeks after the initial unveiling, Cartier held cocktail parties in several stores to which Cartier invited its retail clientele. Tr. at 417. The overall launch budget for the Panthere was in excess of $1,000,000 dollars, including considerable resources spent towards advertising in periodicals and magazines. Tr. at 417-18. In the years following Panthere’s introduction into the U.S. market, Cartier spent about $500,000 on media spending to promote the Panthere. Tr. at 419; PX 123, 128. The advertisements appeared in periodicals including the Los Angeles Times, The New York Times, The Wall Street Journal, the Oregonian, Park Cities News, the Palm Beach Post, and the Connecticut Post. PX 128, 131, 135. See, e.g, PX 278, No. 20382 (New York Times Panthere advertisement, November 7, 1984); PX 279, Nos. 20114, 20115 (Wall Street Journal Panthere advertisements, March 13, 1992, published in five different regions in the United States). 8. The Pasha Cartier introduced the Pasha watch family in 1985 as a modern version of a watch created by Louis Cartier in 1944. In the years 2000 and 2001, Cartier spent upwards of $1,000,000 and $2,000,000 dollars respectively in advertising the Pasha. PX 243. Its print advertisements have appeared in publications including the Los Angeles Times, Greenwich Magazine, the Atlanta Journal, the Chicago Tribune, the Denver Post, the Texas Monthly, Providence Journal, the Las Vegas Review Journal, The Aspen Times, The Washington Post, The New York Times, and Boca Raton Magazine. Plaintiff also introduced as exhibits numerous examples of standard Pasha advertisements that have run in a variety of publications over many years. Tr. at 422,-24; PX 245. A The Tank Americaine Cartier introduced the Tank Americaine in 1989. Total media spending for the Tank Americaine averages approximately $500,000 per year. PX 123. Cartier has advertised the Tank Americaine in publications including Architectural Digest, Conde Nast Traveler, GQ Magazine, Hamptons Country Magazine, Wine Spectator, Forbes, and Travel and Leisure. PX 129, 132. Frank Cologni authored a coffee table book entitled “Cartier, the Tank Watch,” that Cartier gives to clients as gifts or distributes in its boutiques. Tr. 542-43; PX-95. The book expands the visibility of both Tank watch families at issue. 5. The Tank Francaise Cartier launched the Tank Francaise in 1996. It hosted a launch party at the Salon Internationale D’Haute Horloger, a major annual watch fair held in Switzerland, and in New York, where its launch party reflected the watch’s design by adopting a theme reflecting Paris’s square architecture. Tr. 528, 570. Since then, total media spending ranges from over $500,000 in 1999, to as much as $1,750,000 in 2000. PX 123. Advertisements for the Tank Francaise have appeared in, inter alia, Town & Country, In Style, various theatrical and musical playbills, the Harvard Crimson, The New York Times, Vogue, The New Yorker, and Vanity Fair. D. Sales The Panthere is Cartier’s fifth best-selling watch. Tr. at 535. At the time of its introduction into the U.S. market, the Pan-there was $1,800 for the stainless steel model, $11,000-14,000 for the gold, and $2,200 to $2,700 for combinations of stainless steel and gold. Tr. at 420-21. At present, the steel versions retail beginning at $2,450, the gold go up to $11,400, and pavee versions sell for approximately $50,000. Tr. at 533. Between the years 1987 and 2002, Cartier and its authorized dealers sold 88,175 watches in the Pan-there family for total sales over $313 million. The Pasha is Cartier’s fourth best-selling watch. Tr. at 535. Currently, steel versions retail for about $3,700, chrome and other versions from $6,000 to $7,000, 18 karat gold versions for about $30,000, and a full pavee Pasha with princess cut diamonds for about $208,000. Tr. at 535. Between 1987 and 2002, Cartier sold over 42,000 units for sales totaling over $180,000,000. PX 242. The Tank Americaine is Cartier’s third best-selling watch. Tr. at 535. Gold Tanks with a leather strap begin at $5,500 and increase to $10,000 to $20,000 for models with metal bracelets, and a full pavee Tank retails for about $88,000. Between 1987 and 2002, Cartier sold 14,225 watches for sales totaling over $102 million. PX 124. The Tank Francaise is Cartier’s top-selling watch family. Between 1987 and 2002, Cartier sold 93,750 Tank Francaise watches, for sales totaling over $266 million. E. Recognition of the Cartier Watches in the Press and Among Consumers 1. Press Commentary Cartier introduced an impressive body of exhibits during trial illustrating the depth of media commentary about the four watches at issue, both in the popular press and in the more specialized watch and luxury goods press. See PX 96A — 114A. Although it is unnecessary to regurgitate the full panoply of press commentary, the following are indicative examples: the New York Post, December 16, 1998, quotes actress Gwyneth Paltrow as stating that her desired wristwatch is a Cartier Tank watch with a leather strap and sapphire winding stem. PX 252. Another article in the New York Post in 1998 states that “the timeless style of the Cartier Tank Watch with an alligator strap still goes as well with a Harris tweed blazer as it does an Armani tuxedo.” Id. In October of 2000, the magazine Retail World identifies the Tank Amerieaine as a “coveted worldwide ... Cartier classic.” Id. In June 2001, Watch Time, whose audience largely consists of watch aficionados, reported that “the world-renowned Cartier brand bases its strategy on the virtue of being instantly recognizable.” Id. In 1999, Watch Time also wrote: “Representing everything a sporty-yet-elegant wristwatch should be, the Cartier Pasha has, in fact, become a sought after treasure among luxury timepieces limited in numbers only because of its high price. ... The watch, [is] designed to be recognizable as a Cartier.... Among the unmistakably Cartier characteristics are the threaded shaft lungs which dominate the slide view.” Id. In International Watch in November of 1991, the Cartier Panthere is identified as a “classic watch.” In its April-June 2001 edition, the Robb Report writes that “the Cartier Tank is arguably one of the most recognizable status watches today.” Id. In 1997, the Robb Report honors the Cartier Tank Francaise in its “Best Awards,” noting that the Tank Francaise is “an interpretation in line with an inimitable century old tradition that both elevates and combines good taste and contemporary styling.” Id. 2. Consumer Recognition: the Expert Reports Defendants and Plaintiff both conducted surveys to test the secondary meaning of the four families of Cartier watches at issue. Simply stated, the parties retained experts to poll the public as to whether they associated the Panthere, Pasha, Tank Amerieaine, and Tank Francaise, or more specifically, their watch designs, with Cartier. a. Defendants’ Expert: Mr. Harry O’Neill Defendants retained Mr. Harry O’Neill, Vice Chairman of Roper ASW. Mr. O’Neill’s report is hereinafter referenced as the “Roper Report.” The Roper Report was created by intercepting shoppers at six shopping malls throughout the country: Atlanta, Boston, Chicago, Dallas, Los Angeles and San Francisco. Tr. at 175; DX-Y6. O’Neill attempted to pick malls with “relatively upscale stores” in order to maximize the likelihood of identifying survey participants who represented the appropriate population. Tr. at 176. O’Neill concluded that a mall that was anchored by Sears or Kmart, for example, would be unlikely to be frequented by consumers in the luxury watch market. Id. At the malls, shoppers were intercepted and screened to determine their eligibility to participate. Tr. 169. Shoppers who were under 18, did not have their glasses or contact lenses available but relied on them, or who worked for an advertising company, market research company, or watch retailer or manufacturer were ineligible to be surveyed. Id. Shoppers were further asked whether or not they owned a watch worth at least $2,500. DX-Y6, Ex. 1, A. If so, they were qualified to answer the survey’s questions. If not, they were asked: “How likely is it that you would consider buying a fine watch — one that would cost at least $2,500 — in the next couple of years — very likely, fairly likely, not very likely or not at all likely?” Id. Those who responded indicated that they were “very likely” or “fairly likely” qualified to participate. Eligible participants were then shown pictures of a Cartier Tank Francaise, a Cartier Tank Americaine, a Cartier Pan-there, and five other watches made by other manufacturers, namely, Chopard, Rolex, Tag Heuer, Movado and Bvlgari. Tr. at 172. With each picture, a participant was asked: “Do you associate this style or design with the watches of one or more than one company?” Tr. at 174. If so, although unnecessary to establish secondary meaning, as an “added extra attraction,” participants were asked a second, follow-up question as to whether they recognized to which particular company the watch belonged. Id. The results of the Roper study are as follows: 38% of the respondents associated the style or design of the Tank Americaine with one company (with 13% correctly identifying Cartier as that company); 34% of the respondents said that they associated the style or design of the Tank Fran-caise with one company (with 13% correctly identifying Cartier as that company); 31% associated the Panthere style or design with one company (with 13% correctly identifying Cartier as that company). Based on these figures in the Roper Report, O’Neill concludes that a significant portion of the purchasing public does not associate the style or design of the watches at issue with Cartier. Tr. at 180. What is noteworthy to the Court, however, is the considerable discrepancy in findings at the Atlanta mall vis a vis the results obtained in surveying shoppers at the other five malls. Of the six malls involved in creating the Roper Report, only the Atlanta mall was anchored by upscale retail establishments. Whereas the Atlanta Mall was anchored by Neiman Marcus and Bloomingdales, the Boston mall was not anchored by any high-end stores, although there was one within five minutes’ walking distance, the Chicago mall was anchored by Marshall Fields and Carson Pirie Scott and the Dallas mall was anchored by a Dillar Folis and a Mervins. Tr. 199. Further, in Atlanta, 69% of survey respondents owned a watch worth at least $2,500, compared to the 41% of respondents at the other locales. For those who did not already own a fine watch, 55% of the participants were “very likely” to purchase one in the near future, compared to 15% of the participants who answered in similar fashion at the other malls. Accordingly, the court concludes that the population of survey respondents at the Atlanta mall was the most representative of the Cartier consumer population. Here, 63% of the participants associated the style and design of the Tank Francaise with one company, 60% of respondents associated the Tank Americaine with one company, and 60% associated the style or design of the Panthere with one company. b. Plaintiffs’ Expert: Dr. Sidney Lirtzman Dr. Lirtzman criticized the Roper Report on the grounds that it surveyed the wrong population insofar as it failed to distinguish between those “very likely” to purchase an expensive, luxury timepiece in the near future, and those who were “fairly likely” to make such a purchase. Tr. 232. He testified that the survey results from Atlanta indicate that if the Roper Report had been conducted exclusively at “high end malls” and included only those persons more resolute about their intentions of buying a fine watch, the numbers of participants identifying the style or design of the three Cartier watches with one company would have been higher. Tr. at 250-51. To support this conclusion, Lirtzman conducted his own survey designed to parallel O’Neill’s, with the exception of two important differences: Lirtzman only interviewed individuals who either already owned a luxury watch or were “very likely” to purchase a watch in the next year, whereas the Roper Report includes respondents who were “very likely” to purchase a watch “in the near future” and persons who were “fairly likely” to purchase such a luxury watch “in the next couple of years.” Further, Lirtzman intercepted individuals while they were shopping not in shopping malls, but in Tourneau Watch Company stores, two in Manhattan and one in the Roosevelt Field Mall on Long Island, NY, one in Costa Mesa, CA, and one in Century City in Los Angeles, CA. Tr. at 321-22. Tourneau is an authorized dealer of Cartier watches. Tr. at 639. In light of this relationship, the Tourneau stores feature prominent posters of Cartier watches as well as display cases with Cartier watches. Tr. at 190, 823-25, 829. The Lirtzman study also included a few less significant alterations from O’Neill’s study. Lirtzman asked if the participants associated the watch’s design with a particular source, as opposed to asking about whether the participant associated the “design or style” with a particular source. Tr. at 236-37. Because it is irrelevant to establishing secondary meaning, Lirtzman also did not ask O’Neill’s second question as to whether the participant could identify which company she or he associated with the watch’s design. The Lirtzman study was also limited to the Tank Francaise and the Panthere because these watches were the least recognized according to the Roper Report. Lirtzman showed participants pictures of the Tag Heuer and Movado watches, like the Roper Report, achieving the same percentages for recognition of these watches among participants, Tr. at 241, but excluded the other controls. Finally, the photographs shown to survey participants in Lirtzman’s study are increasingly clear and more uniform than those shown to participants in the Roper study. The results of Lirtzman’s study are as follows: 61% of the survey respondents associated the Tank Francaise’s design with a particular source and 63% of the survey respondents associated the Pan-there with a particular source. Tr. at 250-51. Lirtzman concludes from this result and the Atlanta results in the Roper Report that surveying individuals who either own or are very likely to purchase a luxury watch establishes consumer recognition of the Cartier watch families at issue in the range of 50 to 60%. Id. Defendants’ principal objection to Lirtz-man’s report is that in light of the Cartier posters at Tourneau and the fact that its watches are among those displayed in Tourneau’s cases, the result of the study are biased. See Def. Proposed Findings of Fact and Conclusions of Law at 18. The court, however, disagrees. There are a panoply of luxury watches prominently featured at Tourneau, both in the display cases and on the walls as posters and murals; Tourneau changes its displays every few months; and there are 110 brands sold at Tourneau, all of which have multi-pie lines or models within them. Tr. at 219, 666, 833. The Cartier case, for example, contains six to a dozen watch models, including the watches at issue. Tr. at 830. As such, while Cartier is sold at Tourneau and is displayed among the many images a consumer perceives while shopping there, the likelihood that a survey participant’s reaction to the Tank Francaise and Pan-there would have been so influenced is so minimal as to have little to no effect on the probative value of Lirtzman’s report. Moreover, the court credits the testimony of Dr. Lirtzman that valid market research does not require a secondary meaning survey to be conducted in a vacuum given the nature of the questions posed to the survey participants. Tr. at 237, 331-32, 338A12. At Tourneau, consumers were asked questions in an environment in which one would actually purchase a luxury timepiece. Images of the products to be sold are customary in such an environment. Had the Lirtzman’s pollsters asked about particular brands of the watches shown to participants surrounded by promotional images, this would raise the specter of potential bias; but here, where the question was simply whether a participant associated the watch with a particular company, without asking which one, no such concern arises. Therefore, in light of a) the results obtained by defendants’ expert in Atlanta, where the survey was undoubtedly taken in a mall where higher-end merchandise is sold, meaning, an environment more consistent with Cartier’s consumer population, and where the respondents were increasingly likely to either own or purchase a luxury time piece in the immediate future; b) plaintiffs’ survey showing that the Atlanta results are more likely to be accurate than those obtained in other fora; and c) the Court’s concerns about the absence of persons within the age - group 18-34 or mistakes in tabulating their survey results in the Roper Report, see Tr. at 214-15, PX-285, the court adopts the testimony of Dr. Sidney Lirtzman, finding that the results obtained in Atlanta and in the Lirtz-man Report are representative of the secondary meaning of the watches at issue. S. Consumer Recognition: Third Party Retailers’ References to Cartier in Promoting their Watches Cartier introduced evidence to the effect that third party sellers invoke Cartier’s name in selling their watches as a means of appealing to consumers. As exhibited by various printouts from web sites selling watches, the practice is especially rampant among internet sellers who often use such terms as “Cartier style,” “Tank style” or .“Panthere style.” PX-241, 283; Tr. at 455-56. Cartier also employed an investigator, Ms. Sharon Round, in the summer of 2003 to canvas 47th Street in Manhattan, during which she posed as a potential watch purchaser of various retailers in the area. PX 232-235. PX 273 is a business card from a retailer named Aviannes that Round frequented; on the back of the card, the salesperson wrote, “Tank Fran-caise style $4,500” and “Panthere with diamond bezel $2,325.” She further testified that at Agosta Watch, Samo’s Sons, Yae-ger Watch Corp. and Niletti Creations, salespersons represented that they sell watches that are copies of the Cartier watches at issue or that their watches are “Cartier style” or look “just like” Cartier’s. See Tr. at 592-684. This evidence further solidifies plaintiffs’ claim that their watches enjoy recognition among the watch-consuming public. The fact that other retailers continuously invoke Cartier’s name in promoting their watches shows that customers attach meaning to the terms “Tank style” or “Panthere style” that in turn heightens their motivation to purchase. 4. Historical Timing of Public Recognition of the Cartier Watches Having concluded that the majority of the luxury watch-consuming population identifies the watches at issue with one source, the question is at what point in time did the Cartier watches attain such recognizable status. The Panthere was introduced in 1984. According to Ralph Destino, former President and Chief Executive Officer from 1976 until 1985 and Chairman from 1985-2000, the response to the Panthere was so overwhelmingly successful that stores kept waiting lists of the watch shortly after its introduction into the market. Tr. at 419-20. Both Destino and Albers presented credible testimony that the watch was ■ well-established and well-recognized by 1987. Tr. at 38, 47, 421. In light of the considerable resources dedicated to the Panthere’s launch and in the years thereafter, the court finds that the Panthere was well recognized by 1987. The Pasha was first launched in 1985. Between the 1999 article in Watch Time identifying the Pasha as a classically recognizable Cartier, see PX 252, Albers’ testimony that all of the watches at issue were publicly recognized within 3 or 4 years of their introduction, Tr. at 48, the considerable sales of the Pasha watches between 1987 and 2002, PX 243, and Skoc-zylas’s testimony that the Pasha has always been a popular Cartier family, Tr. 532, the court finds that the Pasha achieved public recognition by 1989. The Tank Americaine was first launched in 1989. Albers testified that Tank Ameri-caine sales have been steady and rapid since its introduction. Tr. at 38, 530. The court’s review of Cartier’s sales history of the Tank Americaine reveals that by 1994, sales of this watch family were considerable, grossing more than $3.5 million for Cartier that year. In consideration of these evidentiary items coupled with the strength of evidence presented regarding advertising and media expenditures for the Tank Americaine, the court finds that the Tank Americaine achieved public recognition by 1994. The Tank Francaise was launched in 1996. Cartier’s sales history shows that by 1997, the Tank Francaise was an overwhelming success, grossing $29 million for Cartier that year. Both Albers and Skoe-zylas provided credible testimony that the Tank Francaise was an instant success, with Skoczylas further adding that the watch quickly became an “icon with fash-ionistas” and “everyone is wearing it” in the Hamptons. Tr. at 528-29. The court finds that the Tank Francaise enjoyed public recognition by 1996. F. Similarity of Third Party Watch Designs to Cartier’s Defendants spent, a great deal of effort during trial to debunk the view that Cartier’s four watch families have achieved secondary meaning. They introduced evidence designed to show that many watch manufacturers design and retail watches that look identical or substantially similar to Cartier’s on the theory that such widespread use of Cartier’s design negates secondary meaning. The court will examine each of the allegedly similar third party watches in turn. 1. The Tank Francaise First, defendants aver that the ESQ Chronicle manufactured by Movado bears a strong resemblance to the Tank Fran-caise. DX-R5 at 31; DX-P3 at 31. The court agrees that a side-by-side comparison of the Chronicle and the Tank Fran-caise reveals a striking likeness between the two. However, there is no evidence before the court of advertising expenditures for the Chronicle, its sales have been relatively small in the U.S. watch market, and, most importantly, Movado agreed to discontinue selling the Chronicle at Cartier’s request. See Tr. at 872 (testimony of Stuart Diamond, Movado’s Chief Operating Officer). Second, defendants point to Swatch’s Balmain Elysee, DX-S5 at p. 47, as a third-party design bearing strong resemblance to the Tank Francaise. A side-by-side comparison of the two watches does not illustrate a likeness between the two watch designs; the Balmain’s watch face is square, its brancards are not sharp like the Tank’s, the bracelet links have rounded edges compared to the straight, sharp edges of the Tank, and the Balmain does not have a dial at all analogous to Cartier’s. Further, Swatch’s logistic and planning manager, Robert Piusiensky, testified that Balmain sales are small and before an order is placed for a Balmain, Swatch must first receive an order from one of its accounts. Tr. at 797-98. Third, defendants argue that the Concord La Tour features a design that is similar to the Tank Francaise. DX-D1. Comparing the two, Concord’s bracelet is too different from the Tank’s to warrant a finding of substantial similarity between the two watches. The Tank’s bracelet links are H-shaped and imitate the sharp downward angle of the corners of its bran-cards. By contrast, Concord La Tour’s bracelet has relatively small rounded links lined up in rows, four across. They are neither as stark or as bold as the Tank’s. Further, there is no analogous gap between the LaTour watch case and the bracelet like the Tanks’s and its horizontal brancards look nothing like the Tank’s. Finally, defendants identify Guess as manufacturing and selling a watch looking like the Tank Francaise. DX-L3. This assertion is easily disposed with; a side-by-side comparison reveals only scant similarity between the two watch designs. Most notably, the bracelets and vertical brancards are completely dissimilar. 2. The Tank Americaine Here again, defendants point to several watches that they claim look just like the Tank Americaine, namely, the ESQ Venture, DX-P3, and watches manufactured by Tourneau, DX-Q9, Longines, DX-Q5, Bulova, DX-F1, and Balmain, DX-S5 at p. 42. Having compared each of these watches to the Tank Americaine, the court gives scant weight to this argument. The two most salient factors supporting this conclusion are 1) none of the watches defendants point to are as rectangular in shape as the Tank Americaine and 2) none of the watches defendants point to have a remotely similar bracelet design. The Venture has Arabic numbers instead of Roman numerals and bears a circular minute track, whereas the Tank Americaine does not. Their crowns are also noticeably dissimilar. The overall appearance of the Tourneau at DX-Q9 is so completely different than the Tank Americaine as to require no discussion at all. Further, the watches shown in DX-19, a Tourneau 2002 catalog, feature brancards with a multi-step design that is not a similar design feature of the Tank Americaine. The Roman numerals of the Longines watches are unlike Cartier’s, its brancards have square rather than rounded edges and are not as elongated as the Tank Americaine’s, nor does its crown feature a stone like the Tank Americaine’s. Even if this watch is a top-seller for Longines as Thomas Vitu-lano, Longines’ sales and operating manager testified, it does not detract from the distinctiveness of the Tank Americaine’s design. As to the Bulova, given the difference in dial, shape, crown, and bracelet, it is beyond cavil that one would confuse it with a Tank Americaine; the overall impression is totally different. Finally, the Balmain’s brancards are thin; it does not have the long, lean look of the Tank Amer-icaine, the winding stem is not octagonal like the Tank’s, and its dial is unmistakably different than the Tank’s. The court therefore finds that the third party watches defendants point to are not sufficiently similar to Cartier’s Tank Am-ericaine as to detract from the Tank’s uniquely recognizable design. 3. The Panthere Defendant relies most heavily on three watches to negate the Panthere’s secondary meaning: two Concord Versailles watches and one Movado. DX-05, DX-V6 and DX-P5 at 15. With respect to the Versailles shown in DX-05, even one of plaintiffs’ witnesses agreed that they looked alike: Albers stated that the two are in fact confusingly similar. Tr. at 135— 37. Defendants’ witness Houlihan likewise testified that the watches are similar to the Panthere. Tr. at 352-53. However, Dube stated that the proportional distinctions and absence of screw tips in the Versailles rendered them sufficiently distinct. Tr. at 494-501. A side-by-side comparison of the Versaille and the Panthere reveals that the watch shape and dial are indeed quite similar, but their bracelets are decidedly different. Equally as important, having sold only nine Versailles watches annually on average for the years 1998-2001, DX-I11, it is difficult to conceive how the Versailles could detract from consumer recognition of the Panthere watch family. With respect to the Concord displayed at DX-P5, p. 15, Concord and its authorized dealer, Bailey Banks & Biddle, discontinued selling this watch at Cartier’s request, Tr. at 443-45. G. Cartier’s Efforts to Protect Its Trade Dress 1. Generally There are four individuals on Cartier’s staff exclusively dedicated to protecting Cartier’s intellectual property and its related “zero tolerance” policy towards in-fringers. Tr. at 432, 435. Cartier introduced a range of past “cease and desist” letters addressed to third parties that Cartier believed to be infringing on its intellectual property. PX 136-140, 142-80, 181-201, 207-21. In situations when the recipient of such a letter has not complied with Cartier’s request, Cartier has filed suit. PX 138, 142. Moreover, Cartier employs an investigator to canvas 47th Street in Manhattan to detect potential copy-cats. Tr. at 484-86. The success of Cartier’s enforcement efforts is substantiated by the fact that Tourneau agreed to stop selling a watch that allegedly infringed upon Cartier’s Tank Americaine after having received a cease and desist letter. Tr. at 443, 660-62. Similarly, Concord agreed to stop selling watches Cartier believed to infringe upon its Panthere trade dress. Tr. 443-45. Movado likewise agreed to discontinue sales of an ESQ brand “Chronicle” watch Cartier believed to infringe upon its Tank Francaise watch design. Tr. at 443-47; 872. 2. With Respect to Defendants’ Accused Watches The court credits the testimony of Cartier’s counsel, Bharat Dube, that Cartier did not learn of defendants’ activities with respect to the watches in question until late 2000 when it discovered Globe’s catalog. Tr. at 464. The evidence is undisputed that defendants’ offices were not visible on 47th street, that a passerby would have to be buzzed in to visit their offices, and that defendants’ sales of the accused watches are relatively few and scattered throughout the United States. Tr. at 603-04, 623, 632, 905; PX 19. There is, therefore, no reason for the court to believe that plaintiffs should have or did learn of the defendants’ sales and advertisements any sooner than late 2000. II. DEFENDANTS’ WATCHES AND ACTIVITIES RELATING THERETO A. The Accused Watches 1. Defendants’ Watches Accused of Infringing the Tank Francaise’s Trade Dress Plaintiffs charge Globe XW98, GW882, W1984, W1986, W1989, GW863, GW892, GW866, W1994, XGW265, XGW266, GW875 and Crown Classic L803, L901, L1004, L1005, L1006, L1007 with infringing its Tank Francaise Trade Dress. See PX 255. The court finds that the accused watches are indeed strikingly similar in design to the Tank Francaise, if not virtually identical. The brancards on defendants’ watches are the same design, width, and proportion in relation to the rest of the watch. The Roman numerals are identical. Their watches bear the same H-shaped bracelet links. The watch face is square, exactly like the Tank Francaise’s. Like the Tank Francaise, the winding crown is a faceted octagon set with a cabochon stone. While some of defendants’ watches have brancards covered with diamonds or a pavee watch face or sometimes both, these alterations to watches that are otherwise identical to the Tank Francaise do not alter the court’s finding of similarity, especially given that Cartier has offered watches in the Tank Francaise family with pavee brancards, pavee bracelets, and pavee faces. See PX 73 at 26-29, PX 75 at 31; PX 86 at 56 (Cartier catalogs). By the same token, although some of defendants’ accused watches feature a leather strap, every other design feature of the watches is identical to the Tank Francaise and Cartier has offered the Tank Fran-caise with a leather strap. See, e.g., PX 72 at 54-55, 67; PX 73 at 6, 9, 28, 30-31; PX 86 at 56 (Cartier catalogs). 2. Watches Accused of Infringing the Tank Americaine Trade Dress Plaintiffs charge defendants’ watches Globe XW39, XW40, XW41, XW43, XW42, XW46, XW49, XW50, XW51, XW57, XW58, XW59, XW56, XW60, XW61, XW66, XW67, XW68, XW95, XW96, XW97, GW880, GW881, W1982, W1988, W1992, W1993, GW755, W1688, XGW267, XGW268, GW879 and Crown Classic L801, L804, L805, L806, L807, L906, L1001, L1002, and L1003 with infringing the trade dress of their Tank Americaine watch family. See PX 256. Here again, the court has little trouble finding that defendants’ watches are, in some cases, virtually identical, and in others, undeniably similar to Cartier’s Tank Americaine. Like the Tank Americaine, all of defendants’ accused watches have a rectangular face, with the longer side being vertical. Their watch faces consist of thick vertical and horizontal brancards formed of the case metal, and at each corner the vertical brancards, like the Tank Americaine’s, are elongated beyond the top and bottom of the watch case. Defendants’ brancards also frame the end links of their watches’ bracelet chain or leather strap, and their winding crown is the same faceted octagon with a stone. Defendants’ watches that do not feature a pavee face have the identical Roman numerals and, with the exception of the GW880, on defendants’ watch models with a chain bracelet, the bracelet has the same distinctive five-row rectangular design. The fact that some of defendants’ watches have pavee brancards, or part of their bracelet links are covered in diamonds, or have a pavee watch face does not alter the court’s conclusion. These watches still adhere to the overall look and design of the Tank Americaine. Equally as important, these watches are similar, if not identical, to Tank Americaine watches offered by Cartier featuring diamonds on the face, brancards, and/or bracelet. See, e.g., PX 72 at 33; PX 73 at 17; PX 74 at 2, 32; PX 75 at 60; PX 86 at 54-55. Cartier has also offered the Tank Americaine with a leather strap. See, e.g., PX 72 at 33, 58; PX 73 at 17, 19, 21; PX 74 at 32, 45; PX 75 at 61; PX 86 at 54-55. Finally, although defendant Globe’s GW880 has a dissimilar bracelet link design to the Tank Americaine’s, it is nevertheless decidedly similar to plaintiffs’ Tank Americaine; every other feature of the watch is identical to Cartier’s and its overall look and design is the same. 3. Defendants’ Watches Accused of Infringing the Panthere Trade Dress Plaintiffs charge the following watches as violating their Panthere Trade Dress: Globe XW47, XW93, XW94, W2002, W1704, W1703, W1979, W1983, GW610, GW646, W1151, W1262, GW648, W1689, XW1679, and Crown Classic L802, L905, L907. See PX 257. Like Cartier’s Panthere, Globe XW47, W2002, W1704, GW610, W1151, and XW1679 have a square face with rounded corners, surrounded by a square-shaped, one-piece metal frame or bezel, having rounded corners, and featuring eight screw tips set around the bezel at 1, 2, 4, 5, 7, 8, 10, and 11 o’clock. They also all have Roman numerals and bracelets with five rows of links, although some of them have a different link design than the rectangular link design of the Panthere. Some of the watches also have a chemin-de-fer railroad chapter ring or minute guide located between the center of the dial and the numerals, with the same distinctive design of the Panthere, such that every fifth minute is thicker and bolder. In light of these similarities, the court finds these watches to be noticeably similar to the Panthere even in the instances where they slightly deviate from one of the discrete elements comprising Panthere’s overall design. Between the shape of the watch face, the rounded corners of the face, the overall proportions, the Roman numerals and the thickness of the metal bracelet relative to the face, the appearance of Globe XW47, W2002, W1704, GW610, W1151, and XW1679 is the same as the Panthere. Globe XW93, XW94, W1703, W1979, GW646, W1262, XW1678 and Crown Classic L802, L905, and L907 have either a diamond bezel or a diamond watch face, or a diamond bezel in addition to a diamond watch face. They all feature the Panthere square face with rounded corners, surrounded by a square-shaped, one-piece bezel having rounded corners. Except for Crown Classic L802 and L905 featuring pavee faces, these watches all have the same Roman numerals as the Panthere. Globe XW93 and Globe XW94 do not have the same octagonal winding stem with cabochon, but their bracelet has the same five rows of rectangular links, albeit with rounded corners. Globe W1703 has a five-link bracelet link like the Panthere, although with a very different shape to each link, but it also has a chemin-de-fer minute guide like the Panthere’s, and a winding crown with a cabochon like the Panthere, framed in an identical fashion. Globe W1979 has the same rectangular bracelet links, although with rounded edges, and the same winding stem with cabochon. It has the same chemin-de-fer minute track, with every fifth minute being thicker and bolder like the Panthere’s, although it is on the outside of the Roman numerals whereas the Panthere’s is on the inside. Globe GW646 and W1292 have the same rectangular links as the Panthere’s metal bracelet, the same cabochon on its winding stem, and the same chemin-de-fer minute track on the inside of the Roman numerals with every fifth minute being thicker and bolder. Globe XW1678 has a very different bracelet link design than the Panthere. The Globe XW1678 does not have a cabochon on its winding stem, nor does it have a chemin-de-fer minute track. The similarity of the Roman numerals, square face, bezel, and overall proportions nevertheless render the Globe XW1678 very similar to the Panthere. Crown Classic L907 has the same five-link rectangular bracelet design, with the same chemin-de-fer minute track on the inside of the Roman numerals, with the same winding stem with cabochon. Crown Classic L802 and L905 have pavee bezels and faces, although their bracelets have the same rectangular five-link design. Having carefully examined these watches in relation to the Panthere, the court finds them to be strikingly similar to the Panthere. The court is further persuaded by the fact that Cartier has offered the Panthere with a pavee bezel and a pavee bracelet. See PX 73 at 38; PX 75 at 39; PX 86 at 58 (Cartier catalogs). Finally, Globe GW648 and Globe W1689 have leather straps. Both of them, however, have the same chemin-de-fer, square face with rounded bezel, Roman numerals, proportions, and overall look of the Pan-there. Even with a leather strap as opposed to the Panthere’s customary metal link bracelet, the watches have the same overall design and appearance as the Pan-there. 4.. Defendants’ Watches Accused of Infringing the Pasha Trade Dress Plaintiffs claim that Globe XGW245, XGW 246, XGW 247, and XGW 248 infringe its Pasha trade dress. See PX 258. Defendants’ watches have the same round and thick watch case with thick outer metal bezel. They have the same removable screw down cap covering a central crown and the same horned extensions at the top and bottom of the watch case connecting the bracelet. They feature the same large Arabic numerals in curly font. Their metal bracelets bear the same H-shaped links alternating with rectangular links. XGW 245 has the two small crowns above and below the central crown, capped with non-removable caps designed similarly to the main cap with matching cabochons. Globe XGW 247 and 248 also have the same grid or grille with a tic-tac-toe pattern that is virtually identical to Cartier’s. As such, The court has little difficulty in finding that the watches in question are undeniably similar to Cartier’s Pasha. B. Timing of Defendants’ Entry into the Market with Its Watches Defendants’ 1990 catalog includes watches bearing a striking similarity to the Panthere, Tr. 874-76, DX-C4, leading the court to conclude that defendants entered the market with watches looking like the Panthere at that time. Defendants’ 1997 catalog, introduced in late 1996 or early 1997, includes watches bearing a striking similarity to the Tank Americaine, Tr. at 876-77, DX-V3, leading the court to conclude that defendants entered the market with those watches bearing a striking similarity to the Tank Americaine at that time. Defendants’ 2000 catalog, first introduced in the fall of 1999, includes W1986 and GW863, both of which are accused of bearing the same trade dress as plaintiffs’ Tank Francaise. Tr. at 877-78, DX-S3. The court therefore finds that defendants entered the market with watches bearing a striking similarity to the Tank Francaise in 1999. Defendants’ 2002 catalog features watches that bear a striking similarity to the Pasha. Because the evidence is unclear whether defendants disseminated the catalog in late fall of 2001 or early 2002, the court finds that defendants entered the market with watches with the same overall design and appearance as the Pasha in late 2001 or the beginning of 2002. C. Defendants’ Activities With Respect to the Watches at Issue 1. Source of Defendants’ Watches: Purchase Invoices D