American Home Products Corp. v. Barr Laboratories, 656 F. Supp. 1058 (D.N.J. 1987)

*1059 William J. Fiore, Meyner & Landis, Newark, N.J., Marie V. Driscoll, Townley & Updike, Egon E. Berg, American Home Products Corp., New York City, for plaintiff.

John G. Gilfillan, Carella, Byrne, Bain & Gilfillan, Roseland, N.J., for defendant Barr Laboratories Inc.

Joel H. Sterns, Sterns, Herbert & Weinroth, Trenton, N.J., Barry J. Brett, Parker Chapin Flattau & Klimpl, New York City, for defendant L. Perrigo Co.

HAROLD A. ACKERMAN, District Judge.

This is a civil action in which plaintiff alleges that defendants have violated federal and state trademark laws. Specifically, plaintiff claims that defendants have violated § 43(a) of the federal Lanham Act, 15 U.S.C. § 1125(a), and have committed the torts of unprivileged imitation and "passing off" as defined under New Jersey law, by selling the drug ibuprofen in a 200-milligram over-the-counter tablet form in a color "similar or identical" to the color of the 200-milligram over-the-counter ibuprofen tablets sold by plaintiff under the name "Advil."

In September 1986, plaintiff moved for a preliminary injunction against defendants, requesting that defendants be prohibited from manufacturing, selling, or promoting their allegedly infringing tablets, from confusing the buying public regarding the separate commercial identities or sources of the products involved, and from otherwise competing unfairly against plaintiff. Upon completion of the hearings held on plaintiff's motion, I ordered that the trial of this action on the merits be advanced and consolidated *1060 with plaintiff's motion, in accordance with Fed.R.Civ.P. 65(a) (2). I now decide the merits of this action, stating my findings of fact and conclusions of law as required by Fed.R.Civ.P. 52(a).

This action was commenced by the filing of a complaint on April 3, 1986. The complaint initially alleged Lanham Act, unprivileged imitation, and "passing off" violations arising from a resemblance in color between plaintiff's and defendants' ibuprofen tablets, and from defendants' use of the phrase "advanced pain medicine" in the packaging and promotion of its tablets. The complaint was amended on October 9, 1986 to state claims based only on the alleged resemblance in color between plaintiff's and defendants' tablets. Like the original complaint, plaintiff's amended complaint seeks injunctive relief only; specifically, that defendants be preliminarily and permanently enjoined from manufacturing, selling or promoting their allegedly infringing product, from acting to confuse or deceive customers regarding the separate commercial origins of plaintiff's and defendants' products, and from otherwise competing unfairly against plaintiff.

On May 9, 1986, defendant L. Perrigo Company moved to dismiss the complaint. I denied that motion in a decision rendered July 15, 1986.

On September 16, plaintiff applied for an order in part temporarily restraining defendants' sales, to be heard on short notice. I denied plaintiff's TRO application on September 19, 1986 on the ground that plaintiff had failed to demonstrate a likelihood of success on the merits. This court specifically found that plaintiff had not made a clear showing of secondary meaning or likelihood of confusion. Finally, this court found that plaintiff failed to meet its burden of establishing irreparable harm. This court also found that defendants would have suffered a substantial hardship if restrained from shipping on September 24.

On September 30, October 10, October 20-23, and November 7, 1986, I held hearings on the issues of fact raised by plaintiff's preliminary injunction request. At the hearings, plaintiff presented four witnesses, L. Randall Yates, President of Whitehall, Ruth Elkins, a survey taker in the field, Dr. Jacob Jacoby, plaintiff's expert, and Norman Passman, principal of Guideline Research Corp., and read into the record portions of the deposition transcripts of Frederick Radford, Perrigo's Director of Regulatory Affairs, Margaret Murphy, a product manager for Perrigo, and Alexander Cossin, in-house counsel for Barr. Plaintiff introduced 147 exhibits into evidence. Barr produced its president, Edwin A. Cohen, and Perrigo offered the testimony of Mark Olesnavage, Director of Marketing of Perrigo, and Dr. William Lazer, defendants' expert. Defendants introduced 49 exhibits into evidence. After the hearings, I ordered trial of this action on the merits advanced and consolidated with plaintiff's motion for preliminary injunctive relief. I now decide the merits of this action based on all the submissions of the parties and the evidence received at trial.

Plaintiff is a Delaware corporation with its principal place of business in New York City. Through its Whitehall division, plaintiff distributes and sells "Advil," a 200-milligram analgesic containing the active ingredient ibuprofen, available over the counter in tablet or caplet form.

Defendant Barr Laboratories is a New York corporation with its principal place of business in Northvale, New Jersey. It manufactures as one of its products the 200-milligram over-the-counter ibuprofen analgesic tablets about which plaintiff complains for distribution by defendant Perrigo. Defendant L. Perrigo Company is a Michigan corporation which does business in New Jersey. It markets and sells the 200-milligram over-the-counter ibuprofen tablets which defendant Barr manufactures, and about which plaintiff complains, for resale under its own label and the labels of other retailers.

Because plaintiff alleges a violation of 15 U.S.C. § 1125(a), this court has subject *1061 matter jurisdiction over the claims brought in this action under 28 U.S.C. §§ 1331 and 1338, 15 U.S.C. § 1121, and the doctrine of pendent jurisdiction. See United Mine Workers v. Gibbs, 383 U.S. 715, 86 S. Ct. 1130, 16 L. Ed. 2d 218 (1966). The parties do not dispute that the court has personal jurisdiction over the defendants and that venue is properly laid in this district.

Section 43(a) of the Lanham Act, 15 U.S.C. § 1125(a), states in full:

Any person who shall affix, apply, or annex, or use in connection with any goods or services, or any container or containers for goods, a false designation of origin, or any false description or representation, including words or other symbols tending falsely to describe or represent the same and shall cause such goods or services to enter into commerce, and any person who shall with knowledge of the falsity of such designation of origin or description or representation cause or procure the same to be transported or used in commerce or deliver the same to any carrier to be transported or used, shall be liable to a civil action by any person doing business in the locality falsely indicated as that of origin or in the region in which said locality is situated, or by any person who believes that he is or is likely to be damaged by the use of any such false description or representation.

Section 43(a) creates a federal cause of action for unfair competition. American Greetings Corp. v. Dan-Dee Imports, Inc., 807 F.2d 1136, 1140 (3d Cir. 1986); Williams v. Curtiss-Wright Corp., 691 F.2d 168, 172 (3d Cir.1982).

Unfair competition under § 43(a) includes acts of unprivileged imitation. American Greetings Corp., 807 F.2d at 1140; Ciba-Geigy Corp. v. Bolar Pharmaceutical Co., 747 F.2d 844, 854 (3d Cir. 1984), cert. denied, 471 U.S. 1137, 105 S. Ct. 2678, 86 L. Ed. 2d 696 (1985); SK & F, Co. v. Premo Pharmaceutical Laboratories, Inc., 625 F.2d 1055, 1065 (3d Cir.1980). A plaintiff shows unprivileged imitation in violation of § 43(a) by showing

that the feature or overall combination of features imitated is non-functional, that it has acquired secondary meaning, and that members of the consuming public are likely to confuse the source of the product bearing the imitating feature or combination with the source of the product bearing the imitated feature or combination.... When a court is satisfied of these three elements it may properly enjoin the defendant from copying the claimed feature or combination of features.

American Greetings Corp., 807 F.2d at 1141 (citations omitted); see Ciba-Geigy, 747 F.2d at 850 and 854; SK & F, 625 F.2d at 1063, 1065.

The Third Circuit has made clear that the federal law of unprivileged imitation under § 43(a) is equivalent to the New Jersey law of unprivileged imitation. American Greetings Corp., 807 F.2d at 1141. Accord SK & F, 625 F.2d at 1065; United States Golf Association v. St. Andrews Systems, Data-Max, Inc., 749 F.2d 1028, 1032 n. 6 (3d Cir.1984); Ciba-Geigy, 747 F.2d at 854 n. 7; Freixenet S.A. v. Admiral Wine & Liquor Co., 731 F.2d 148, 151 n. 5 (3d Cir.1984); Ideal Toy Corp. v. Plawner Toy Mfg. Corp., 685 F.2d 78, 80 n. 1 (3d Cir. 1982.) Therefore, I shall apply the test presented above to determine whether defendants have violated both federal and state provisions prohibiting unprivileged imitation, or whether defendants have violated neither.

Unfair competition under § 43(a) also includes acts of "passing off" one's goods as the goods of another. Ciba-Geigy, 747 F.2d at 854; SK & F, 625 F.2d at 1065-66. Under New Jersey law, it is "passing off" for a party to fraudulently market his goods as the goods of another, or to put his goods in the hands of another in a form in which the party can reasonably anticipate that they may be passed off as the goods of another. Ciba-Geigy, 747 F.2d at 854-53; SK & F, 625 F.2d at 1062. I note, without further discussion at this point, that the test for passing off under *1062 § 43(a) may diverge from the test for passing off under New Jersey state law. In Ciba-Geigy, the Third Circuit suggested that the United States Supreme Court in Inwood Laboratories v. Ives Laboratories, 456 U.S. 844, 102 S. Ct. 2182, 72 L. Ed. 2d 606 (1982), may have held that a Lanham Act action for passing off will not stand unless a defendant fraudulently markets his goods as the goods of another or puts his goods in the hands of another with the intent that they be passed off, or continues to supply another when the party knows or should know that they are being passed off. 747 F.2d at 854-55; see Ives, 456 U.S. at 854 & n. 13, 102 S. Ct. at 2188 & n. 13; see also 456 U.S. at 860-62, 102 S. Ct. at 2191-92 (White, J., concurring). The Ciba-Geigy court noted that such a test diverges materially from the test for passing off under New Jersey law, 747 F.2d at 854-55, but found it unnecessary to decide whether the Ives opinion established the test as the federal law of passing off under the Lanham Act, or merely discussed the test in dicta. 747 F.2d at 855.

As part of its proof under the legal formulas described above, plaintiff has offered as evidence the results of a consumer survey relating to the secondary meaning of the color used in Advil tablets and the likelihood of confusing the commercial sources of plaintiff's and defendants' tablets. Properly conducted surveys are admissible evidence. The Third Circuit has identified a properly conducted survey by noting the following characteristics:

A proper universe must be examined and a representative sample must be chosen; the persons conducting the survey must be experts; the data must be properly gathered and accurately reported. It is essential that the sample design, the questionnaires and the manner of interviewing meet the standards of objective surveying and statistical techniques. Just as important, the survey must be conducted independently of the attorneys involved in the litigation. The interviewers or sample designers should, of course, be trained, and ideally should be unaware of the purpose of the survey or the litigation.

Pittsburgh Press Club v. United States, 579 F.2d 751, 758 (3d Cir.1978); see Boehringer Ingelheim G.m.b.H. v. Pharmadyne Laboratories, 532 F. Supp. 1040, 1055-56 (D.N.J.1980). See also National Football League Properties v. N.Y. Giants, 637 F. Supp. 507, 513-14 (D.N.J.1986), where the district court identified substantially identical criteria for admissible survey evidence by looking to the Federal Judicial Center's Manual for Complex Litigation, 116 (5th ed. 1981). Although a survey may not meet all the Pittsburgh Press Club criteria, the results may still be admitted and given lesser weight or applied to narrower issues of factual dispute. Boehringer Ingelheim, 532 F. Supp. at 1053-54 at n. 17. It is the burden of the proffering party to prove the quality of survey evidence. National Football League Properties, 637 F. Supp. at 513-14.

Should plaintiff succeed on the merits of at least one of its claims, the court may fashion an appropriate, permanent injunctive remedy so long as the balance of equities favors such a remedy. Ciba-Geigy, 747 F.2d at 850. In cases where plaintiffs have requested permanent injunctive relief to remedy violations of § 43(a) of the Lanham Act, courts have held that a finding of likelihood of confusion is sufficient to establish the plaintiff's irreparable harm and the public's interest in permanent injunctive relief.

Ibuprofen is an analgesic which is used for the relief of headache, arthritis pain, pain of menstrual cramps, backache, muscle ache, fever and toothache. Ibuprofen is the active ingredient in Advil, and as such has been marketed as a "third generation" pain reliever, succeeding aspirin and acetaminophen, which is used in Tylenol. Ibuprofen has been portrayed as the product of a long effort to develop an anti-inflammatory non-steroidal synthetic drug without the side effects of aspirin. Ibuprofen competes with aspirin, buffered aspirin, acetaminophen and other analgesics. Ibuprofen was developed by the Boots Company, of the United Kingdom, which held the *1063 patent for the product. Plaintiff does not claim to have played any role in inventing ibuprofen or establishing its utility or safety as an analgesic.

Ibuprofen was introduced and marketed in the United States as a prescription medication by the Upjohn Company under the brand name Motrin pursuant to a patent license from Boots. Motrin is established as a successful and widely accepted prescription drug. Motrin is available in tablets of 300 mg., 400 mg., 600 mg., and 800 mg., strengths. The 300 mg. tablets are white; 400 mg. Motrin tablets are circular, sugar-coated, orange tablets with rounded edges. The 600 mg. tablets are peach-colored and the 800 mg. tablets are orange. Ibuprofen is also available in prescription strength under the name Rufen, produced by the Boots Company, and in a white tablet produced by defendant Barr.

On May 18, 1984, FDA approval was granted for selling 200 mg. ibuprofen tablets over the counter without prescription. On May 21, 1984, pursuant to a license from Boots, plaintiff shipped the first of its own 200-milligram over-the-counter ibuprofen tablets under the registered trademark "Advil." From the beginning, plaintiff has manufactured its tablets with a distinctive appearance that has never been changed. Advil is made in a red-brown color which resembles the color of common clay. The parties have referred to this color as, among other descriptive names, "rust," "brownish-orange" and "terra cotta." It strikes one as an unusual color for a medicine tablet. Without prejudice to its distinctiveness, I shall refer to the color conveniently as "brown." The tablets are made in a rounded, streamlined shape, similar in shape and size to the commonly-known "M & M" candies. The tablets have a shiny surface. On the face of each tablet, the name "Advil" is printed in small, delicate, black upper- and lower-case letters, in a typeface identical to the typeface used on the Advil box and in Advil advertising. The name is not impressed into the surface of the tablet, but is instead embossed across its surface, as if by some printing technique.

With its color, shape, sheen, and label, Advil presents overall a "high style," "high design," perhaps even "high tech," look the look of a tablet for an age in which the aesthetic design of consumer products appears to attract a heightened degree of attention. Advil bears an overall modern appearance, quite different from for example the old, familiar aspirin tablet. This is a conclusion I seem to share with plaintiff itself. Plaintiff introduced as evidence a number of different promotional signs which bore large color pictures of an Advil tablet, a Tylenol tablet, and a generic aspirin tablet. The Advil tablet sits on the right of the sign, over the word "today." The Tylenol tablet sits slightly above and behind the Advil tablet, over the designation "1955." The aspirin tablet sits above and behind the Tylenol tablet, and is labelled "1899." The sign invites consumers to contrast the pills visually and as successive steps in pharmacological progress, and to connect appearance with advancement. The aspirin and Tylenol tablets pictured are white and hard-edged, have no surface sheen, and have their names pressed into the faces of the tablets. Advil, as pictured, presents a vivid visual contrast to the other two pills. It is brown, round, shiny, and embossed with the name "Advil" in delicate black letters. Overall, as a combination of color, shape, sheen, and label, Advil looks wholly new and different.

Beginning in May 1984, ibuprofen was also available in a 200-milligram over-the-counter form, in a tablet called "Nuprin." Nuprin was made by the Upjohn Company under a license from Boots, and sold for Upjohn by Bristol Meyers. From May 1984 until September 1986, only plaintiff and Upjohn produced 200-milligram over-the-counter ibuprofen tablets for American consumers.

Plaintiff's sales of Advil currently amount to about $180 million per year. As of September 24, 1986, Advil brand ibuprofen had approximately 67-75% of the over-the-counter ibuprofen market by far the largest share of that market and over 9% of the entire $2 billion over-the-counter analgesic market. Advil remains today one of the highest priced analgesics or ibuprofen *1064 products available in the over-the-counter market.

Pursuant to the provisions of the Federal Drug Price Competition and Patent Term Restoration Act of 1984, patent protection for ibuprofen was maintained until September 24, 1986, and thereafter ended. After that date, companies were free to produce over-the-counter ibuprofen without license from former patent-holders. New ibuprofen manufacturers could also take advantage of an abbreviated drug approval process to speed manufacturing applications through the Food and Drug Administration. On September 25, 1986, the FDA granted defendant Barr's abbreviated application to manufacture and sell 200-milligram over-the-counter ibuprofen tablets. On September 26, 1986, defendant Perrigo began selling defendant Barr's over-the-counter ibuprofen tablets throughout the country. Perrigo packages them under at least 26 different retail brand labels, many of which are the "house brand" labels of various retailers, and one of which is Perrigo's own label. The tablets are sold to consumers at a lower price than Advil, typically half the Advil price. Defendant Perrigo has already shipped at least 50 million tablets.

Defendants' tablet is manufactured in a color similar to the brown color of Advil. When the tablets are examined side-by-side, the defendants' tablet appears to bear the same hue as Advil, in a slightly lighter shade. When defendants' tablet and Advil are examined at different times, the color of the defendants' tablet strikes one as being very similar, and perhaps identical, to the color of Advil. Defendants' tablet does, however, have a distinctly different shape. It is shaped like an old-fashioned aspirin tablet; that is, like a thick slice from a cylinder, with the round face and back of the tablet bowing out convexly. The tablet does not have a notably shiny surface, as does Advil. The characters "I-2" are impressed into the face of the tablet. The characters, in a typeface reminiscent of the typeface on an old manual typewriter, are large enough to virtually fill the face of the tablet. Apparently, no sort of printing process is used to label the tablets, as appears to be the case with Advil.

There is nothing "high style" about defendants' tablet. Overall, defendants' tablet looks most like an old aspirin tablet painted Advil brown. As I noted before, and as plaintiff's promotional signs point out, there is a striking difference in appearance between Advil and the old, familiar aspirin tablet. Painting that old tablet brown is not enough to make it look like sleek, modern Advil. Defendants' tablet and plaintiff's tablet are clearly dissimilar in overall appearance.

Plaintiff adduced as evidence at trial a number of letters sent between defendants Perrigo and Barr, as well as other memoranda and letters of Perrigo's, which clearly show defendants' intent to produce their ibuprofen tablets in a color similar to Advil. Plaintiff's evidence also shows that defendants were concerned that too close a resemblance between Advil and defendants' tablet might implicate legal prohibitions on "tradedress" infringement. See the March 24, 1986 Perrigo memorandum attached as Exhibit 7 to the affidavit of Marie Driscoll, Esq., dated August 25, 1986. Overall, these documents show an intent on the part of defendants to create a tablet which resembled Advil in one respect, color, and did not resemble Advil in other respects. Defendants felt that the tablets were close in color, but different in shape, surface finish, and labelling. Defendants clearly wanted some physical resemblance between the tablets, but not too much.

Plaintiff, of course, claims that the extent to which defendants did copy the Advil appearance did indeed violate legal prohibitions, specifically, the laws prohibiting unprivileged imitation and passing off. In support of its claims of unprivileged imitation, and specifically its contentions that Advil's tablet color possesses secondary meaning and that a likelihood of confusion exists regarding the commercial origins of plaintiff's and defendants' tablets, plaintiff has proffered as evidence the results of a consumer survey designed and supervised by Dr. Jacob Jacoby and executed by Norman *1065 Passman and Ronald Silver of Guildline Research Corporation.

Dr. Jacoby instituted the survey after being told by plaintiff that plaintiff expected a competitor to soon issue an over-the-counter 200-milligram ibuprofen tablet in a shape and color similar to Advil's. Dr. Jacoby designed his survey to determine (1) the extent to which Advil users "correctly associate" an Advil tablet without the name Advil on it with the brand name Advil, and (2) the extent to which Advil users "incorrectly associate" the competitor's tablet with Advil. Jacoby, Final Report, plaintiff's Exhibit 126 at 3. The first goal of the survey was, in effect, plaintiff's attempt to measure secondary meaning of the Advil tablet color, and the second goal was plaintiff's attempt to measure the likelihood of confusion regarding the sources of plaintiff's and defendants' products. Dr. Jacoby conferred with counsel for plaintiff to ensure that such a survey would be, as Dr. Jacoby testified, "responsive" to the legal issues about which plaintiff was concerned in this case.

Dr. Jacoby designed the survey as follows. First, he defined the proper "universe" of people about whom conclusions would be drawn to include people over eighteen years of age who had used Advil in the last three months. He then drew up a plan for taking a non-probability sampling from that universe. Non-probability samples are commonly used in market research to draw conclusions about customer universes. Under the plan, the country was divided into 4 census regions. Location cities in each region were chosen. In each city, interviewers at shopping malls approached people to solicit their participation in the survey. Because, as Dr. Jacoby testified, "there was no basis to know what the user population of analgesics in general, Advil in particular, is like," people were selected so as to duplicate the distribution of age groups and sexes present in the general American population, not the survey universe. Interviewers approached people who, by their appearance, fit the pre-determined quotas for age-group and gender distributions. These people were asked to indicate which items on a list of varied consumer products they had used in the last three months. People who indicated Advil among all the items identified were asked to consent to further questioning. This further questioning involved, first, dividing respondents into one of two sub-groups, the "Advil" group or the "generic" group, showing each respondent four tablets sealed in separate transparent bags, and then asking each respondent to indicate whether they "associated" each of the four tablets with any of thirteen brand names or ingredient names for drugs, including Advil. Respondents in the "Advil" group saw a blue Excedrin PM pill, which bore a "PM" stamped into it, a yellow Nuprin pill with brand name removed, a white Trendar pill, and an Advil tablet with the name "Advil" removed. The name Advil was removed, as Dr. Jacoby's final survey report states, "to avoid bias" in respondents' answers, and because, as Dr. Jacoby testified, the brand name "Advil" printed on Advil tablets was an "important" visual aspect of the tablets for which recognition was not being tested. The "generic" group saw the Excedrin, Nuprin, and Trendar pills, plus defendants' ibuprofen tablet, unaltered. The "Advil" group's answers were meant to establish a finding regarding secondary meaning, while the "generic" group's answers were meant to establish a finding regarding confusion. Dr. Jacoby had consulted with counsel for plaintiff before finalizing the form of the "association" questioning.

Dr. Jacoby submitted his final report on the survey and its results on September 10, 1986. In the end, 363 responses, 178 from "Advil" group members and 185 from "generic" group members, were used to draw conclusions. About 24% of these finally counted respondents came from San Francisco, one of the 8 cities where the survey was conducted; in contrast, only about 3% came from another city, Philadelphia. Dr. Jacoby stated his findings as follows. First, 71.9% of those people who claim to have used Advil within the last three months "correctly associate" an unlabelled Advil tablet with the name Advil, and 22.4% make no association with the unlabelled *1066 Advil tablet. Second, 18.4% of these people who claim to have used Advil within the last three months "incorrectly associate" defendants' ibuprofen tablet with the name Advil. He concluded that "the Advil tablet possesses a high degree of visual identity" and that defendants' ibuprofen tablet will be mistakenly identified by, as Dr. Jacoby's final survey report stated, a "substantial number" of people. In testifying regarding the significance of his findings, Dr. Jacoby emphasized in his testimony that the survey was not intended or designed to duplicate the conditions attendant when a consumer actually decides to purchase a product, nor could his findings be construed to relate to the issue of consumer confusion before the initial purchase of a product.

Aside from the appearance of the pills themselves, packaging and promotion also contribute to consumer perceptions of Advil and defendants' ibuprofen. Defendant Perrigo provides its retailer customers with promotional support in selling defendants' tablets. This support includes advising retailers on the design of their house brand packages, executing the packaging, and providing shelf signs and aisle displays for the tablets in the retailers' stores. Perrigo's advice to retailers is embodied in what it calls its "graphic guidelines" on the promotion of defendants' tablet. The graphic guidelines are intended to suggest how defendants' tablet can be presented to the buying public as a generic, lower-priced alternative to Advil. In some aspects, the graphic guidelines attempt to create a very limited or selective physical resemblance between the packages for Advil and each generic brand. For example, defendant Perrigo recommends the use of the colors blue and yellow in tablet packages and promotional signs, which are the dominant colors in Advil packaging. But Perrigo also recommends that the name brand of the generic tablet never be printed in yellow, as the name "Advil" is in plaintiff's packaging, and that generic brands avoid the use of a red stripe to highlight their packages. A bright red stripe provides a strong visual accent to the Advil package. Other aspects of Perrigo's promotional efforts attempt to distinguish Advil and the generic brands as comparable products selling at different prices. For example, defendant Perrigo supplies retailers with promotional signs colored a deep sky blue with yellow and black printing. The signs say either "compare to Advil and Nuprin" or "compare and save" printed above pictures of the Advil box and the box of the private label for which the sign was made.

Defendant Perrigo does not contractually impose its graphic guidelines or promotional signs on its retailers. Thus, it cannot say that the packaging and promotion of defendants' tablets always follows a scheme which it would itself consider wholly acceptable under its own guidelines. At trial, plaintiff showed three instances in which Perrigo's graphic guidelines were in fact violated. One house brand, K-Fen, runs the name K-Fen in large yellow letters on its package. Two other brands, Eckerd and Waldbaum's, use single red stripes on their packages. There was no other evidence shown regarding deviations from defendant Perrigo's guidelines.

In general, the packaging used for Advil bears no material similarity to the packaging used by defendant Perrigo for its retailer customers. Plaintiff and Perrigo both sell their tablets in bottles marked with paper labels, which in turn are packaged in cardboard boxes. Plaintiff's bottle and Perrigo's bottle are both squat, white, plastic containers, which by themselves appear very similar. The label on each bottle bears the applicable brand name, and is printed in colors and designs which match the cardboard box in which the bottle comes. The colors and designs used by each of the private labels supplied by defendant Perrigo vary considerably. Many of them use some shade of blue and some shade of yellow in their packaging. Most carry the brand name "Ibuprofen," or the name Ibuprofen in conjunction with some retail trade name, such as "Waldbaum's Ibuprofen." One brand is named "Medi-Profen" and one is named "Modern Pain Medicine." The K-Fen brand, mentioned earlier, comes in a box colored a robin's egg blue, with the name "K-Fen" printed *1067 in yellow over the blue field and a broad yellow band running diagonally across a corner of the box. Eckerd's box, mentioned earlier, comes in an electric blue, with the name "Ibuprofen" printed in white above a red band. No yellow is used. Waldbaum's brand, mentioned earlier, comes in a box half white and half a dark grey-green, with a red band running down the middle. The Waldbaum's mark and the name "Ibuprofen" fill the white field. Defendant Perrigo's own brand, called "Ibuprofen," comes in a box which is colored a robin's egg blue, with thin yellow stripes and a wide yellow band within which the name "Ibuprofen" is printed in large black letters. In contrast, Advil's box and bottle label is colored a deep blue, unlike any of the blues used on the Perrigo boxes, with light blue touches. The name "Advil" is printed over the blue in bright, large, yellow letters which almost fill the side of the box or the bottle label. A red stripe runs diagonally over the lower-right corner of the sides of the box bearing the Advil name.

Mr. Lawrence Randell Yates, president of Whitehall Laboratories, testified that he found the Advil box and many of defendant Perrigo's private label boxes similar enough in appearance to exacerbate whatever confusion might flow from the fact that plaintiff's and defendants' tablets were similarly colored, or to create a likelihood of confusion when considered along with the similarity of tablet colors. I find his contentions to be clearly incorrect, and I reject them.

Nowhere in defendant Perrigo's packaging or promotional materials is defendants' tablet itself actually displayed. Although plaintiff does not display its Advil tablet on the Advil box, plaintiff does depict the tablet round, brown, shiny, and with its name "Advil" printed on the tablets in black in its television commercials and in most of its print advertisements. Plaintiff has engaged in extensive television and print advertising of its tablet, airing at least 2,000 television commercials by June 1986, creating over one billion "adult print impressions" in its periodical, advertising, and printing over 6 million Advil pamphlets for doctors to give patients. Mr. Yates testified that at the time of trial over $100 million had been spent on advertising Advil. Clearly, one of plaintiff's goals in this significant marketing effort has been to impose the image of the Advil tablet upon the minds of the public as a recognizable symbol of a product they ought to buy.

Defendant Perrigo sells defendants' tablet in packages of 24, 50, and 100. Plaintiff sells its tablet in packages of various sizes, up to 1,000 tablets per package. Plaintiff claims that institutions buy its product in 500-, and 1,000- tablet sizes, for dispensing to patients, employees, and others who see no other packaging but the tablet itself. Although plaintiff has produced no evidence that defendant's tablets are dispensed the same way, plaintiff does claim that defendant Perrigo's 25-, 50-, and 100- tablet packages could be put to such use by institutional buyers, so that some users of defendants' tablet would first be exposed to the tablet without also seeing any of defendants' promotional or packaging materials.

As noted above, the case at hand was commenced by plaintiff on April 3, 1986. Plaintiff did not move before this court for injunctive relief until late August. During the span of time commencing with the filing of plaintiff's first complaint, defendants lost at least one customer for its brown ibuprofen tablets because of concern that plaintiff would be able to enjoin defendant Perrigo from making ibuprofen deliveries. Defendants in fact produced a small quantity of white ibuprofen tablets for customers who would not accept defendants' brown tablets. White ibuprofen was eventually sold to 4 or 5 of defendant Perrigo's accounts.

It is clear from the facts in this case that defendants have followed a strategy common to the manufacture and marketing of generic products, including drug products. They have attempted to create some limited degree of physical resemblance between their product and the name-brand product against which they compete. Such a resemblance *1068 can produce at least two effects, one of which is legitimate, the other of which is not. The resemblance between two products can alert consumers to the functional or utilitarian equivalence between them, to the fact that one product may be substituted for the other in the ultimate uses for which the products are intended. The free flow of information regarding the substitutability of products is valuable to individual consumers and to society collectively, and by providing it a supplier engages in fair competition based on those aspects for example, price in which the products differ. See, e.g., Virginia State Board of Pharmacy v. Virginia Citizens Consumer Council, 425 U.S. 748, 763-65, 96 S. Ct. 1817, 1826-27, 48 L. Ed. 2d 346 (1976). Product resemblance can also have an illegitimate effect: confusing consumers as to whether the products are not only substitutable, but are in fact the same suggesting, in effect, that one commercial entity is responsible for producing both products. Such a misunderstanding can do great harm to the maker of the imitated product, saddling the maker with undeserved blame from consumers dissatisfied with the imitator's product and an undeserved loss of sales to consumers satisfied with the imitator's product. The federal and state laws against unprivileged imitation and passing off serve to specify when the latter, illegitimate effect is sufficiently manifested to invoke legal sanction.

As I noted above, there are three conjunctive parts to the New Jersey and Lanham Act tests for unprivileged imitation: the non-functionality of the imitated feature, the secondary meaning of the imitated feature, and the likelihood of confusing the source of the imitative product with the source of the imitated product. See American Greetings Corp., 807 F.2d at 1141; Ciba-Geigy, 747 F.2d at 850 and 854; SK & F, 625 F.2d at 1063 and 1065.

The dispositive issue in this case is the possibility of consumer confusion. As the Third Circuit has said, "[r]egardless of how much secondary meaning it possesses, a product's trade dress will not be protected from an imitator that is sufficiently different in its features to avoid such confusion." Freixenet, 731 F.2d at 151. Because the test for unprivileged imitation is conjunctive, the same could be said about the non-functionality of a product's imitated feature; it is irrelevant unless confusion also is shown.

In deciding whether a likelihood of confusion exists between competing goods, the relevant factors to consider are:

(1) the degree of similarity between the owner's mark and the alleged infringing mark;

(2) the strength of the owner's mark;

(3) the price of the goods and other factors indicative of the care and attention expected of consumers when making a purchase;

(4) the length of time the defendant has used the mark without evidence of actual confusion;

(5) the intent of the defendant in adopting the mark;

(6) the evidence of actual confusion.

National Football League Properties, 637 F. Supp. at 516 (quoting Scott Paper Co. v. Scott's Liquid Gold, Inc., 439 F. Supp. 1022, 1036-37 (D.Del.1977), rev'd on other grounds, 589 F.2d 1225 (3d Cir.1978); Caesars World, Inc. v. Caesar's Palace, 490 F. Supp. 818, 823-24 (D.N.J.1980)). See also Scott Paper Co. v. Scott's Liquid Gold, Inc., 589 F.2d 1225, 1229 (3d Cir. 1978), where the Third Circuit ratified the consideration of four additional factors when the products in question were not competing products.

Considering those of the six factors which are relevant to the case at hand, I conclude that plaintiff has failed to prove that defendants' tablets are likely to be confused with Advil under any circumstances. On their own, the boxes and bottles used by plaintiff and defendant Perrigo to package their tablets contribute nothing to the possibility of confusion. As I have found plaintiff's Advil box and bottle, as labelled, do not look like any of defendants' house brand boxes and bottles, as labelled. The house brands clearly announce their brand names, none of which *1069 resemble the name "Advil" in any way. I reject as clearly incorrect any statement made by witness Yates that the parties' packaging may bolster or create confusion between the products. On their own, the parties' promotional materials contribute nothing to the possibility of confusion. Defendant Perrigo's promotional signs and floor displays do not resemble plaintiff's signs, displays, or advertisements. Defendant Perrigo's signs essentially beg consumers to distinguish defendants' generic ibuprofen tablets from Advil. Although plaintiff routinely pictures its tablet in its promotional material, defendant Perrigo never pictures defendants' tablet. If there is any chance that a consumer could be confused as to the source of plaintiff's and defendants' tablets, it must come after the first time that the consumer sees defendants' tablet in person and the first time the consumer sees plaintiff's Advil tablet in person or in plaintiff's promotions.

Plaintiff essentially acknowledges this point by confining its claims of unfair competition to the similarity in color between plaintiff's and defendants' tablets. By so premising its claims, plaintiff implicitly contends that the confusion of which it complains will not take effect until a consumer has seen both pills. In the meantime, plaintiff admits, such a consumer might regularly shop for ibuprofen tablets, view the parties' packaging and promotional material, and without knowing the color of each tablet, select between Advil and any Perrigo generic brand without confusing them at all. Confusion, plaintiff claims, would taint only those sales which took place after the consumer had seen both tablets and had been struck by their resemblance in color.

Were such confusion in fact likely to occur, one of the three elements of unprivileged imitation would be met. Defendants have argued that such likelihood of confusion amounts to "post-purchase confusion," and that such confusion is not actionable. See Smithkline Beckman Corp. v. Pennex Products Co., 605 F. Supp. 746, 751-52 (E.D.Pa.1985). I disagree. First, I read the Smithkline Beckman case merely to forbid actions based on "post-purchase confusion which is not the direct consequence of defendants' action." The district court in Smithkline Beckman was simply refusing to hold defendants liable for post-purchase confusion which plaintiff attributed first to an alleged widespread perception among consumers that name-brand drug manufacturers manufactured their own generic competitors, and second to a similarity in color between the drugs in issue. In the case at hand, however, plaintiff blames any and all confusion on defendants' choice of tablet color. Second, I can see no reason in law or fact why, in an age of skilled and subtle techniques for marketing consumer products and winning new customers, courts should refuse to take cognizance of unfair competitive practices which work themselves out over two, three or more experiences with a product, rather than quickly, at the moment a consumer first sees two opposing products on a shelf and reaches for only one of them. See Lois Sportswear U.S.A., Inc. v. Levi Strauss & Co., 799 F.2d 867, 872 (2d Cir. 1986); Levi Strauss & Co. v. Blue Bell, Inc., 632 F.2d 817, 822 (9th Cir.1980).

Despite the legal viability of an action for post-purchase confusion based on the color of the parties' tablets, I conclude that in this case there is no likelihood that consumers will be confused by the similar color of the parties' tablets into believing that the tablets share a common source. It is clear as a matter of law that a product whose trade dress contains elements of the trade dress of another product does not unlawfully imitate that product if the overall trade dress of the products does not foster confusion. See Freixenet, 731 F.2d at 151, where the makers of a Spanish wine sold in a black frosted bottle were denied a preliminary injunction against a competitor selling Spanish wine in an identical black frosted bottle which bore different labels. See also American Greetings Corp., 807 F.2d at 1140-41; Ciba-Geigy, 747 F.2d at 851; SK & F, Co. v. Premo Pharmaceutical Laboratories, 481 F. Supp. 1184, 1187 (D.N.J.1979), aff'd, 625 F.2d 1055 (3d Cir. 1980). The color which plaintiff's tablet and defendants' tablet share is only one *1070 aspect of the overall appearance, or trade dress, of the tablets themselves. And overall, the tablets are quite dissimilar; so much so that no confusion regarding their separate commercial sources is likely to arise.

I reach this conclusion by considering the three factors from National Football League Properties which are relevant here. First, there is no evidence of actual confusion between the two tablets. Although plaintiff's apparent inability to garner such evidence may be understandable, given the brief five months since defendants' tablet has been sold, its absence does not help plaintiff's case. Second, I have found that defendants' intent was to produce a tablet which bore some resemblance to plaintiff's, but which was at the same time distinguishable from plaintiff's tablet.

Third, as I described above, the degree of similarity between plaintiff's tablet and defendants' tablet is quite low, despite their strong similarity in color. To my eyes, plaintiff's and defendants' tablets are readily distinguishable when compared side by side or when viewed separately. One pill does not confusingly suggest the other. As the finder of fact in this case, my personal observation of the appearance of the tablets may clearly serve as a basis for my conclusions regarding confusion. See Ciba-Geigy, 747 F.2d at 851; Freixenet, 731 F.2d at 151; SK & F, 625 F.2d at 1060-61.

The results of plaintiff's consumer survey do not refute my conclusion regarding the basic dissimilarity in appearance between the parties' tablets. A number of flaws in the design of the survey lead me to accord very little weight to its results. The survey was designed to test the responses of a universe comprised of people over eighteen years of age who had used Advil in the last three months. The proper universe to test would have included other potential users of generic ibuprofen also, for example, users of non-ibuprofen analgesics which undercut Advil in price, so that the full range of potential customers for whom plaintiff and defendants would compete could be studied. The survey addressed its "association" questions to groups drawn from the test universe based on a non-probability sampling plan. While non-probability survey results may be admissible, they are weak evidence of behavior patterns in the test universe. See Boehringer Ingelheim, 532 F. Supp. at 1053-54, 1058. The skewed geographic distribution of the final 363 respondents, with 24% of them living in San Francisco, underscores the absence of probability sampling in the survey. Dr. Jacoby, who designed the survey, consulted with plaintiff's attorneys concerning the definition of issues on which to test and the propriety of the survey's scheme for questioning respondents, a fact which further detracts from the weight which should be accorded the survey results. See Pittsburgh Press, 579 F.2d at 758; Boehringer Ingelheim, 532 F. Supp. at 1057-58.

In addition, I have grave doubts as to whether a response from a member of the "generic" group that he or she "associates" defendants' tablet with the name Advil means that the respondent believes that both products come from the same source. The former U.S. Court of Customs and Patent Appeals has noted:

The fact that one mark may bring another mark to mind does not in itself establish likelihood of confusion as to source. Compare Lever Brothers Co. v. Producers Chemical Service, 283 F.2d 879, 48 CCPA 744 (1960). The very fact of calling to mind may indicate that the mind is distinguishing, rather than being confused by, two marks.

Application of Ferrero, 479 F.2d 1395, 1397 (C.C.P.A.1973). Dr. Jacoby himself, who designed the survey, testified that, in his opinion, the words "associate" and "identify" may have overlapping, but not necessarily congruent, meanings. Given my findings earlier that the parties' tablets share the same color, but are overall readily distinguishable, I am loathe to conclude that by "associating" defendants' tablet and the name Advil, the survey respondents also meant that they could not tell defendants' tablet and Advil apart, or believed that they share a common commercial source. Contrast American Greetings *1071 Corp. v. Dan-Dee Imports, Inc., 619 F. Supp. 1204 (D.N.J.1985), vacated in part on other grounds, 807 F.2d 1136 (3d Cir. 1986), where the district court took cognizance of survey, results showing an "association" between the products in issue, but the association rate was 42%, not merely 18.4% as in the case at hand, and the district court had already found a likelihood of confusion based on its own observations of the products. 619 F. Supp. at 1224.

I note that the survey results were submitted in final form by Dr. Jacoby on September 10, 1986. Defendants' tablets were not available for marketing until September 26, 1986 at the earliest. The survey respondents had thus been asked whether they could associate the name "Advil," among other names, with a pill whose proper identity they could never have known. Respondents may have led themselves to "associate" defendants' tablet with Advil simply because at that time they knew of no other product on the market whose name fit better than Advil. Certainly, they would have been hard pressed to identify any drug as "generic ibuprofen," since generic ibuprofen in general was not available over the counter before September 24, 1986 at the earliest. I also note that although defendants have challenged Dr. Jacoby's impartiality and credibility as a result of his part associations with plaintiff and plaintiff's counsel, I do not challenge plaintiff's survey evidence on that basis.

For all these reasons, I find little probative value in plaintiff's survey evidence. Certainly, plaintiff's survey evidence does not undercut a finding that no likelihood of confusion exists, based on all the evidence considered.

In concluding that no likelihood of confusion exists, I expressly refuse to infer a presumption of confusion from the fact that defendants intentionally copied the Advil color brown for their tablets. In National Football League Properties, the district court found that defendant's "intentional, willful, and admitted adoption of a mark closely similar to" plaintiff's marks raised a presumption of confusion. In contrast, in the case at hand, defendants' intent was to copy Advil's color, but otherwise to create a tablet whose overall appearance was distinguishable. It cannot be said as a matter of law that an intent to create a partial, yet distinguishable, resemblance will presumptively lead to a confusing resemblance. In addition, I expressly distinguish the facts in the case at hand from the facts in Ciba-Geigy, 747 F.2d at 851, and SK & F, 625 F.2d at 1061, where the Third Circuit found that a likelihood of confusion existed between name brand prescription pills and their generic equivalents, where the generic equivalents were precisely identical in appearance to the name-brand drugs, except for small, hard-to-read logos printed on each generic pill. See also Boehringer Ingelheim, 532 F. Supp. at 1045; Biocraft Laboratories, Inc. v. Merck & Co., 532 F. Supp. 1068, 1069 (D.N.J.1980).

In deciding that there is no likelihood of confusion between plaintiff's Advil tablet and defendants' ibuprofen tablet, I have decided that, in those instances where consumers have nothing more with which to identify the parties' tablets than the appearance of the tablets themselves, they will not likely be confused as to source. Defendants had argued that, in making my decision, I should consider the parties' trade dress to include the appearance of their tablets and packages together, as if consumers would never be exposed to the tablets without seeing their packages also. I find it more plausible to conclude, however, that, inevitably, some consumers will first see one or the other tablet without its package, so that the trade dress for the product in that instance reduces to the appearance of the tablet itself. For example, a person who has purchased Advil in the past and has seen Advil tablets may see defendants' tablet for the first time when one day a friend offers one of defendants' tablets without showing the package in which the tablets came. According to my conclusion above, such an exposure to the parties' products will not likely generate confusion as to product source. At such times, without boxes or signs to tell them apart, the parties' pills will remain clearly distinguishable.

*1072 It is reasonable to assume, however, that in the vast majority of instances, consumers exposed to plaintiff's or defendants' tablets will also see the bottles, boxes, and occasionally, promotional signs which go along with the pills. At such times, the parties' clearly distinguishable packaging and signs will further highlight the separate commercial identities of Advil and defendants' ibuprofen. In direct opposition to Mr. Yates' suggestion at trial, I conclude that defendant Perrigo's packaging and promotional materials enhance the initial trade dress differences between plaintiff's and defendants' tablets. Thus, most consumers will observe the partial resemblance between the parties' tablets within a promotional context which further insures that no confusion as to source will arise.

The existence of significant promotional efforts aimed directly at consumers is presumably a concomitant of the parties' over-the-counter business. In the prescription-drug cases cited earlier, Ciba-Geigy, SK & F, Bochringer Ingelheim, and Biocraft Laboratories, there was apparently no use of promotion to distinguish separate products in the eyes of consumers. Thus, those cases are further distinguished from the case at hand by the presence in this case of promotional efforts which additionally distinguish the parties' pills for most consumers. See Smithkline Beckman, 532 F. Supp. at 753.

Given my finding that no likelihood of confusion exists, I need not consider whether the Advil color brown has achieved secondary meaning or whether the color itself is a functional or non-functional aspect of the tablets. I note, however, that plaintiff's consumer survey presents questionable evidence in support of plaintiff's claim of secondary meaning. First, to confine the survey universe to recent Advil users must certainly inflate any finding of secondary meaning, since the universe excludes potential users of defendants' tablet who do not use Advil for example, users of non-ibuprofen analgesics who avoid Advil because of its high price. Second, the secondary-meaning findings suffer from an absence of probability sampling and Dr. Jacoby's consultations with plaintiff's counsel, flaws which I discussed earlier when I reviewed the survey findings on confusion. Finally, the survey tested secondary meaning with regard not only to color, but also with regard to tablet shape and tablet sheen. I cited all three of these visual qualities in finding that the Advil tablet bears an overall "high-design" appearance. As I also noted before, defendants' tablets do not share Advil's shape or sheen, so findings regarding the secondary meaning of Advil's color, shape, and sheen would shed little light on a comparison of color only between plaintiff's and defendants' tablets.

For the reasons stated above, I conclude that plaintiff has failed to show that defendants' have committed unprivileged imitation in violation of the federal Lanham Act and New Jersey state law.

In its responses to the proposed findings of fact and conclusions of law submitted by defendants after trial, plaintiff admitted that "defendants apparently are not currently engaged in acts of palming off." Thus, plaintiff appears to have abandoned its passing-off claims completely. In its responses to defendants' proposed findings of fact and conclusions of law, however, plaintiff also stated that defendants are "furnishing to customers a vehicle for palming off at the retail level should the retailer be so inclined." Reading these statements in conjunction with the admissions cited above, it may be that plaintiff meant only to concede the absence of intentional passing off by defendants directly, and to preserve claims of passing off based on knowledge or reasonable anticipation by defendants of retail passing off by defendants' retailer customers. See Ives, 456 U.S. at 854 and note 13, 102 S. Ct. at 2188 and note 13; Ciba-Geigy, 747 F.2d at 854; SK & F, 625 F.2d at 1062.

Plaintiff has offered no evidence that defendants had any knowledge that retailers were selling their tablets as Advil tablets. Given my finding above that there is no likelihood of confusion between plaintiff's and defendants' pills, whether packaged *1073 and promoted or seen standing alone, plaintiff has failed to show that defendants could reasonably anticipate that retailers would pass off defendants' tablets as Advil. As a result, under the New Jersey state law test for passing off, as well as the test for passing off discussed in the Ives case, defendants are not liable. Defendants have violated neither New Jersey state law or federal Lanham Act prohibitions on passing off. In reaching this conclusion, I have found it unnecessary to decide whether the Supreme Court in Ives created a test for passing off under the Lanham Act which differs from the New Jersey state law test for passing off.

Because I have found that plaintiff has failed to prevail on any of its claims, I need not consider plaintiff's eligibility for permanent injunctive relief. Judgment is for defendants on each of plaintiff's claims.

Defendant Perrigo has moved for sanctions against plaintiff under Fed.R. Civ.P. 11. Defendant Perrigo argues first that plaintiff failed to make a sufficient inquiry into the factual basis of its claims before filing suit. I cannot agree. At the time plaintiff filed its complaint, plaintiff was aware that defendant Perrigo planned to market an over-the-counter ibuprofen tablet similar to Advil in formula, dosage, and color. I cannot say that such facts fail to provide an objectively reasonable basis for plaintiff to bring suit.

Defendant Perrigo also argues that plaintiff has litigated its case in such a manner as to cause harassment to defendants and undue delay in the disposition of this action. It is clear from the law in this circuit that rule 11 sanctions may be levied against a party who litigates in furtherance of improper purposes, including harassment or undue delay, whether or not the party acted with subjective bad faith. Conduct which is merely objectively unreasonable, but which causes harassment, undue delay, or needless increase in litigation expenses, may be sanctioned under rule 11. Lieb v. Topstone Industries, Inc., 788 F.2d 151, 157 (3d Cir.1986).

I find, however, that there is insufficient evidence from which to conclude that plaintiff acted to harass defendants or delay the litigation of this action. Plaintiff filed its complaint in April, after it had heard about defendants' plans but before it had a chance to see the ibuprofen tablet defendants were developing. Plaintiff conducted expedited discovery through the beginning of August. When discovery ended, plaintiff moved for preliminary injunctive relief. I find it notable and puzzling that plaintiff's motion was not brought until the last regularly scheduled motion day before defendants were due to make their first shipment of tablets. But I see no factual basis in this case sufficient to support a conclusion that plaintiff acted unreasonably or in bad faith in bringing its action when it did or conducting its litigation as it did.

In short, I find that there exists no likelihood of confusion as to the commercial source of plaintiff's "Advil" over-the-counter ibuprofen tablets and defendants' generic over-the-counter ibuprofen tablets. I grant judgment for defendants on each of plaintiff's claims. I deny defendant Perrigo's request for sanctions against plaintiff under Fed.R.Civ.P. 11.