Opinions of the United
2008 Decisions                                                                                                             States Court of Appeals
                                                                                                                              for the Third Circuit


8-5-2008

ET Browne Drug Co v. Cococare Prod Inc
Precedential or Non-Precedential: Precedential

Docket No. 06-4543




Follow this and additional works at: http://digitalcommons.law.villanova.edu/thirdcircuit_2008

Recommended Citation
"ET Browne Drug Co v. Cococare Prod Inc" (2008). 2008 Decisions. Paper 603.
http://digitalcommons.law.villanova.edu/thirdcircuit_2008/603


This decision is brought to you for free and open access by the Opinions of the United States Court of Appeals for the Third Circuit at Villanova
University School of Law Digital Repository. It has been accepted for inclusion in 2008 Decisions by an authorized administrator of Villanova
University School of Law Digital Repository. For more information, please contact Benjamin.Carlson@law.villanova.edu.
                                  PRECEDENTIAL

UNITED STATES COURT OF APPEALS
     FOR THE THIRD CIRCUIT


         Nos. 06-4543 & 06-4658




       E.T. BROWNE DRUG CO.,
        a New Jersey corporation,

                          Appellant / Cross-Appellee

                     v.

     COCOCARE PRODUCTS, INC.,
       a New Jersey corporation,

                          Appellee / Cross-Appellant




Appeal from the United States District Court
        for the District of New Jersey
   (D.C. Civil Action No. 03-cv-05442)
District Judge: Honorable Peter G. Sheridan


        Argued December 13, 2007
       Before: SLOVITER and AMBRO, Circuit Judges
                  POLLAK,* District Judge

               Opinion filed: August 5, 2008

Arnold B. Calmann, Esquire
Saiber, Schlesinger, Satz & Goldstein
One Gateway Center, Suite 1300
Newark, NJ 07102-5311

Edward M. Laine, Esquire (Argued)
David A. Prange, Esquire
Oppenheimer, Wolff & Donnelly
45 South Seventh Street
3400 Plaza VII
Minneapolis, MN 55402-0000

       Counsel for Appellant/Cross-Appellee

Robert Mahoney, Esquire (Argued)
Jeanne Hamburg, Esquire
Norris, McLaughlin & Marcus
721 Route 202-206
P.O. Box 1018
Somerville, NJ 08876-1018



   *
    Honorable Louis H. Pollak, Senior United States District
Judge for the Eastern District of Pennsylvania, sitting by
designation.

                              2
       Counsel for Appellee/Cross-Appellant




                  OPINION OF THE COURT


AMBRO, Circuit Judge

       This case involves a dispute between two manufacturers
of personal care and beauty products that contain cocoa butter.
E.T. Browne Drug Co., Inc. (“Browne”) claims that it has a
protected trademark interest under the Lanham Act, 15 U.S.C.
§ 1051 et seq., in the term “Cocoa Butter Formula,” 1 which
features prominently on its products. Cococare Products, Inc.
(“Cococare”) disputes the validity of this asserted trademark.
The District Court entered summary judgment in Cococare’s
favor after concluding that the term is generic and thus may not
receive protection from the trademark laws. We agree that
Browne has not demonstrated that “Cocoa Butter Formula” is a
protectable trademark, but reach that conclusion by a different
path. We believe that a genuine issue of material fact exists as
to whether “Cocoa Butter Formula” is generic. But even


  1
    The term appears with and without capitalization throughout
the parties’ briefs and the District Court’s opinion. It appears in
all capitals on the products depicted in various exhibits and in
Browne’s registrations. We use “Cocoa Butter Formula” for
consistency but attach no significance to this choice.

                                3
assuming it is descriptive, this term must have a secondary
meaning to be protectable. Because Browne failed to identify
sufficient evidence to create a genuine issue of material fact on
that point, we affirm the grant of summary judgment in favor of
Cococare. We remand, however, to allow the District Court to
enter an appropriate order under 15 U.S.C. § 1119.

I.     Background and Procedural History

        Browne, a New Jersey corporation, markets personal care
and beauty products containing cocoa butter under the brand
name “Palmer’s.” The “Palmer’s” line of cocoa butter products
is the sales leader among personal care and beauty products
containing cocoa butter. The packaging containing those
products displays “Palmer’s” and “Cocoa Butter Formula.”
“Palmer’s Cocoa Butter Formula” is on the principal register of
the United States Patent and Trademark Office (“PTO”), and
thus this term is presumptively valid as a trademark. See 15
U.S.C. § 1057(b). In contrast, “Cocoa Butter Formula” is on the
PTO’s supplemental register but not on the principal register.
The statutory presumption of validity accordingly does not
attach to that term. See 

       Cococare, a New Jersey corporation, also sells personal
care and beauty products containing cocoa butter, although its
sales are far smaller than those of Browne. In 1994, it
introduced new products formulated with cocoa butter and
Vitamin E, labeling them “Cococare Cocoa Butter Formula.”

                               4
This use of “Cocoa Butter Formula” gave rise to its dispute with
Browne.

       “[T]he parties agree that Browne knew of its claims
against Cococare since 1993 but did not prosecute them because
Cococare sales were ‘de minimis’; and Browne could only
confirm two ‘sightings’ of Cococare from 1994 - 2000.”
Amended Opinion Granting Summary Judgment at 4, E.T.
Browne Drug Co. v. Cococare Prods., Inc., No. 03–5442 (PGS)
(D.N.J. Sep. 20, 2006) (“Dist. Ct. Op.”) (footnote omitted).
Browne first objected to Cococare’s use of the term “Cocoa
Butter Formula” in 2002 after it became aware of a product flyer
from a seller of Cococare’s products.

        Browne then brought suit in the United States District
Court for the District of New Jersey after a cease-and-desist
letter sent to Cococare failed to cause it to stop using the
contested term. Browne alleged, inter alia, that Cococare had
violated the Lanham Act and equivalent New Jersey law by its
use of the term “Cocoa Butter Formula.”                Cococare
counterclaimed, inter alia, for cancellation of Browne’s
supplemental registration of “Cocoa Butter Formula” and
amendment of Browne’s principal registration of “Palmer’s
Cocoa Butter Formula.” It moved for summary judgment on the
grounds that “Cocoa Butter Formula” is not a protectable
trademark because it is a generic term, that Browne’s claims are
barred by the defenses of laches and unclean hands, and that
those claims should be dismissed because Cococare’s use of

                               5
“Cocoa Butter Formula” was a fair use of a product descriptor.
Browne cross-moved for summary judgment on the genericness
and fair use issues, and on Cococare’s counterclaims.

       The District Court concluded that “Cocoa Butter
Formula” is a generic term and entered summary judgment in
favor of Cococare. It rejected Cococare’s counterclaims,
concluding that no “substantive evidence” supported the
allegations that Browne made deliberately fraudulent statements
to the PTO. Dist. Ct. Op. at 12–13. The Court noted that
Cococare had admitted it suffered no actual damages as a result
of Browne’s alleged actions, and reasoned that, because
“supplemental registration provides no substantive rights, []
there would be no reason to force the cancellation of a
supplementally registered mark.” 

       Browne appeals. It argues that the District Court erred in
concluding that the term “Cocoa Butter Formula” is generic.2
Cococare appeals the District Court’s refusal to direct the PTO
to cancel the supplemental registration of “Cocoa Butter
Formula” or to direct the addition of a disclaimer to that term on
the principal registration of “Palmer’s Cocoa Butter Formula.”
Cococare also renews the arguments that it is entitled to entry of
summary judgment in its favor because “Cocoa Butter Formula”


   2
     Browne does not discuss the claims it brought in District
Court under New Jersey law. Those claims therefore do not
feature in our analysis of this appeal.

                                6
has not acquired secondary meaning, Browne waited too long to
act and made misrepresentations to the PTO such that its action
is barred, respectively, by the defenses of laches and unclean
hands, and its (Cococare’s) use of the term “Cocoa Butter
Formula” amounts to fair use.

II.    Jurisdiction and Standard of Review

       The District Court had jurisdiction over the Lanham Act
claims pursuant to 28 U.S.C. § 1338(a) and pendent jurisdiction
over the state law claims under 28 U.S.C. § 1338(b). We have
jurisdiction pursuant to 28 U.S.C. § 1291. Our review is
plenary. Berner Int’l Corp. v. Mars Sales Co., 

,
978 (3d Cir. 1993). On appeal from a grant of summary
judgment, our Court exercises the same standard of review as
the District Court and considers whether genuine issues of
material fact exist that preclude entry of summary judgment. 

       A.     The Protectability of the Mark “Cocoa Butter
              Formula”

      To establish trademark infringement in violation of the
Lanham Act, 15 U.S.C. § 1114, a plaintiff must prove that (1)
the mark is valid and legally protectable, (2) it owns the mark,
and (3) the defendant’s use of the mark is likely to create
confusion concerning the origin of goods or services. Freedom

                               7
Card, Inc. v. JPMorgan Chase & Co., 

, 469–70 (3d
Cir. 2005). We thus begin our analysis by asking whether
“Cocoa Butter Formula” is a valid, legally protectable
trademark.

       Terms asserted as trademarks may fall in four categories:

       [1] arbitrary (or fanciful) terms, which bear no
       logical or suggestive relation to the actual
       characteristics of the goods; [2] suggestive terms,
       which suggest rather than describe the
       characteristics of the goods; [3] descriptive terms,
       which describe a characteristic or ingredient of
       the article to which it refers[;] and [4] generic
       terms, which function as the common descriptive
       name of a product class.

A.J. Canfield Co. v. Honickman, 

, 296 (3d Cir.
1986) (internal citation and quotation marks omitted). The
Lanham Act protects only some of these categories of terms.
Working backward, it provides no protection for generic terms
because a first-user of a term “cannot deprive competing
manufacturers of the product of the right to call an article by its
name.” 

(quoting Abercrombie & Fitch Co. v. Hunting
World, Inc., 

, 9 (2d Cir. 1976)); see also Park ‘N Fly,
Inc. v. Dollar Park and Fly, Inc., 

, 194 (1985)
(“Generic terms are not registrable and a registered mark may be
cancelled at any time on the ground[] it has become generic.”).

                                8
In contrast, the Lanham Act protects descriptive terms if they
have acquired secondary meaning associating the term with the
claimant. 

–93, 296; see also 

. Finally, trademark law protects suggestive and
arbitrary or fanciful terms without any showing of secondary
meaning. 

(citing 

).

       Browne has the burden in this case of proving the
existence of a protectable mark because “Cocoa Butter
Formula” does not appear on the PTO’s principal register.

. It contends that “Cocoa Butter
Formula” should receive protection from the trademark laws as
a descriptive term that has acquired secondary meaning.
Cococare responds that the term should receive no protection
because it is generic. The parties thus pose a difficult question
of trademark law. See 

(“Courts and commentators
have recognized the difficulties of distinguishing between
suggestive, descriptive, and generic marks.”).

       Whether “Cococare Butter Formula” is generic or
descriptive, and whether that term has acquired secondary
meaning, are questions of fact. See 

n.24 (noting that
“Courts of Appeals have generally held that a designation’s
level of inherent distinctiveness is a question of fact”); Dranoff-
Perlstein Assocs. v. Sklar, 

, 862 (3d Cir. 1992)
(identifying secondary meaning as a question of fact). We
therefore ask whether Browne has presented sufficient evidence

                                9
to create a genuine issue of material fact as to those questions.
See Fed. R. Civ. P. 56(c); 

.

        As noted, we conclude that Browne has produced
evidence sufficient to create a genuine issue of material fact on
its claim that the term “Cocoa Butter Formula” is not generic,
but descriptive. But as we also conclude that Browne has failed
to produce sufficient evidence to create a genuine issue of
material fact as to whether the term has acquired secondary
meaning, we reach the same result as the District Court –
“Cocoa Butter Formula” may not receive the protections of the
Lanham Act.

              1.     Is “Cocoa Butter Formula” Generic?

                     a.     The Primary Significance Test
                            and the Limited Circumstances
                            in Which Canfield’s Alternative
                            Test Applies

       This appeal raises the initial question of the proper test
under which to evaluate whether the term “Cocoa Butter
Formula” is generic and thus not protectable as a trademark.
“The jurisprudence of genericness revolves around the primary
significance test, which inquires whether the primary
significance of a term in the minds of the consuming public is
the product or the producer.” 

–93. We
ask “whether consumers think the term represents the generic

                               10
name of the product [or service] or a mark indicating merely one
source of that product [or service].” Dranoff-Perlstein 

(alterations in original) (quotation marks and
citation omitted). If the term refers to the product (i.e., the
genus), the term is generic. If, on the other hand, it refers to one
source or producer of that product, the term is not generic (i.e.,
it is descriptive, suggestive, or arbitrary or fanciful). To give an
example, “Cola” is generic because it refers to a product,
whereas “Pepsi Cola” is not generic because it refers to the
producer. To repeat, Cococare contends that “Cocoa Butter
Formula” is generic whereas Browne argues it is descriptive.

        The District Court did not apply the primary significance
test. It instead applied an alternative test stated in Canfield.3


     3
     The District Court treated this question in a somewhat
confusing manner. It first stated that it is a “fundamental
question in this case . . . whether ‘cocoa butter’ or ‘cocoa butter
formula’ is the relevant product genus for evaluating
genericness.” Dist. Ct. Op. at 7. It then acknowledged that “the
product genus in this action, however, is not in dispute.” 

with its previous statements that Browne
describes the genus as “skin care products in the cocoa butter
category” and that Cococare describes it as “personal care and
beauty aid products in the cocoa butter market,” 

(quotation marks omitted), and that any distinction between
“skin care products” and “personal care and beauty aid
products” is one without a difference, 

continued in apparent agreement with the

                                11
Neither party disputed that approach. We conclude, however,
that the District Court should not have ventured beyond the
primary significance test to any alternative gloss.

       Canfield addressed situations in which a manufacturer
created a new product and it was not clear if it also had created
a new product genus. It involved a dispute over the term “Diet
Chocolate Fudge Soda.” 4 “[A] fundamental question . . . [was]
whether chocolate soda or chocolate fudge soda is the relevant
product genus for evaluating genericness.” 

The
primary significance test5 could not answer that question, we


parties, reasoning that the products bearing the term at issue “are
part of [a] new genus; consisting of personal care and beauty aid
products, including creams, lotions, skin moisturizers and soaps,
formulated with cocoa butter.” 

It then departed from
the submissions of the parties and its own prior recognition of
their agreement, however, by stating that “cocoa butter formula
is the relevant product class.” 

  4
    We varied in our inclusion and exclusion of the word “Diet”
in Canfield. We use that word throughout for consistency but
attach no significance to that choice.
  5
    Canfield refers to the primary significance test as if it differs
from the consumer understanding test. See 

(referring
to “the primary significance test and its related test of consumer
understanding”). More recently, we have treated the concept of
consumer understanding as included in primary significance.
See 

(referring to the primary

                                 12
reasoned, since it applied “only after we have determined the
relevant genus.” 

       Our Court concluded that the following rule would help
us fill in this gap of identifying the appropriate genus for
analysis: “If a producer introduces a product that [1] differs
from an established product class in a particular characteristic,
and [2] uses a common descriptive term of that characteristic as
the name of the product, then the product should be considered
its own genus.” 

In those circumstances,
“[w]hether the term that identifies the product is generic then
depends on the competitors’ need to use it. At the least, if no
commonly used alternative effectively communicates the same
functional information, the term that denotes the product is
generic.” 

(internal citation omitted). See generally
Genesee Brewing Co. v. Stroh Brewing Co., 

, 145
(2d Cir. 1977) (discussing Canfield and describing its test as a
complement to, rather than a rejection of, the primary
significance test when a court cannot readily determine the
genus of a new product).




significance test as focusing on “consumer understanding”).
Either way, the primary significance test remains the central test
of genericness in our Circuit. See 

.
Yet we see nothing in the record indicating that either party has
asked that we or the District Court use the consumer
understanding test to gauge whether a term is generic.

                               13
       Canfield does not control here for a simple reason: this
case does not pose the question addressed in Canfield. The
“question . . . at the core” of Canfield was whether “the relevant
product category or genus for purposes of evaluating
genericness is chocolate soda or chocolate fudge soda.” 

We do not face a comparable question, as the parties
before us do not dispute whether we should use an existing
genus or a new genus in our analysis. They instead agree, with
only insignificant quibbles over wording, that “Cocoa Butter
Skin Care Products” or an equivalent term defines the category.

       To understand why this distinction matters, we return to
the principles underlying Canfield. It addressed a weakness in
the primary significance test – the presumption that a court
knows the product’s genus. In most cases, that genus will be
obvious, even for new products. A slight change in a
detergent’s formula, for example, likely will not create a new
product genus. Problems may arise, however, if a product
differs from existing products in what Canfield calls a
“particular characteristic.” Examples may include the addition
of a new flavor or a new featured ingredient (such as honey in
the “Honey Brown Ale” at issue in Genesee Brewing Co.). The
manufacturer then likely has created a new type of product.
That manufacturer may well need to use descriptive terms in the
product name to identify the product to consumers. This raises
the question of the proper genus for the Court’s genericness
analysis: the established product class or a new product class
that modifies the established product class with the new

                               14
characteristic.

       Canfield addressed this problem by articulating a test that
supplies the proper genus for a genericness analysis. Its test
applies when a manufacturer uses the following equation: name
of new product = name of the established product class (“Diet
Chocolate Soda” in Canfield) + name of the new characteristic
(“Fudge” in Canfield). See 

       The established product class in our case is “Skin Care
Products” or “Lotion.” The new characteristic is “Cocoa
Butter.” Browne could have called its new products “Cocoa
Butter Skin Care Products” or “Cocoa Butter Skin Care Lotion.”
Use of these terms would have satisfied Canfield’s equation
(name of the new characteristic + name of the established
product class) and triggered its test. Canfield stated that the
primary significance test would not have been useful because
the genericness determination would have depended on the
unresolved threshold definition of the genus (“Cocoa Butter
Skin Care Products” vs. “Skin Care Products”).

        Of course, when it introduced skin-care products
containing cocoa butter (i.e., adding a new characteristic),
Browne did not label those products with the term “Cocoa
Butter Skin Care Products” or “Cocoa Butter Lotion.” Instead,
it used “Cocoa Butter Formula.” This term does not frustrate
the application of the primary significance test because it does
not raise the question whether to use “Cocoa Butter Formula” or

                               15
“Formula” as the proper genus for our analysis. Cococare also
has not suggested that “Formula” identifies the established
product class. Browne does not make baby formula after all, or
sell algorithms or recipes. Nor does any record evidence
suggest that consumers use “Formula” to describe the skin care
product category. Browne’s use of a different equation to name
its product (“Cocoa Butter Formula” = name of the new
characteristic (“Cocoa Butter”) + a term not describing the
established product class (“Formula”)) does not bring into play
the weakness in the primary significance test that Canfield
addressed because it does not raise the question of the proper
genus for our analysis. Applying Canfield here amounts to
attempting to remedy a non-existent problem. We therefore will
evaluate the genericness of the term “Cocoa Butter Formula”
under the primary significance test only.

                      b.     Is the Term “Cocoa Butter
                             Formula” Generic Under the
                             Primary Significance Test?

       “[T]he primary significance test . . . inquires whether the
primary significance of a term in the minds of the consuming
public is the product or the producer.” 

We


   6
    The public need not know the identity of the producer for
the primary significance of the term to be the producer.

(citing S.Rep. No. 98–627, at 5
(1984)).

                               16
applied that test in Berner, asking whether the evidence
demonstrated that the term at issue primarily signified the
product genus to consumers. 

–81. Like
the term under dispute in Berner, the meaning of “Cocoa Butter
Formula” should be “evaluated by examining its meaning to the
relevant consuming public.” 

That evaluation
requires looking at the mark as a whole, not dissecting it into
various parts. Id.7 We therefore inquire whether the consuming
public understands “Cocoa Butter Formula” to refer to a product
genus or to a producer. We ask specifically if the evidence
submitted by Browne creates a genuine issue of material fact as
to its contention that the consuming public does not understand
the term “Cocoa Butter Formula” to refer to a product genus
(i.e., that it is not generic, but descriptive in this case).

       Plaintiffs seeking to establish the descriptiveness of a
mark often use one of two types of survey evidence. J. Thomas
McCarthy, McCarthy on Trademarks and Unfair Competition
(4th ed. 2008) [hereinafter McCarthy on Trademarks] describes
a “Teflon survey” as “essentially a mini-course in the generic
versus trademark distinction, followed by a test.” 2 McCarthy
on Trademarks § 12:16. That survey runs a participant through
a number of terms (such as “washing machine” and


    7
      Under this analysis, the District Court should not have
broken up the term for purposes of its analysis and considered
“Cocoa Butter” separately from “Formula.” See Dist. Ct. Op. at
5, 7.

                              17
“Chevrolet”), asking whether they are common names or brand
names. After the participant grasps the distinction, the survey
asks the participant to categorize a number of terms, including
the term at issue. 

created for E. I. DuPont
de Nemours & Co. v. Yoshida Int’l, Inc., 

(E.D.N.Y. 1975)).

        A “Thermos survey,” on the other hand, asks the
respondent how he or she would ask for the product at issue. If,
to use the term under dispute in the case from which the survey
gets its name, the respondents largely say the brand name
(“Thermos”) rather than the initial product category name
(“Vacuum Bottle”), the survey provides evidence that the brand
name (“Thermos”) has become a generic term for the product
category. 2 McCarthy on Trademarks § 12:15 (discussing
survey used in American Thermos Prods. Co. v. Aladdin Indus.,
Inc., 

(D.Conn. 1962)). To put this in the terms of
the primary significance test, the term would be generic because
the consumers would be using it to refer to the product category
rather than a producer who makes products within that product
category.

       Browne conducted a survey in this case that generally
adheres to the “Thermos survey” model. The survey posed a
number of open-ended questions asking respondents to “identify
or describe the product category” in which its products fall. It
asked each of the 154 valid respondents “[w]hat word or words
would you use to identify or describe a skin care product which

                               18
contains cocoa butter?” and “[i]f you needed to identify or
describe a skin care product containing cocoa butter, what word
or words would you use instead of or in addition to just saying
cocoa butter, if any?” Neither “Cocoa Butter Formula” nor any
form of the word “Formula” appeared among the respondents’
answers.

       The District Court appears to have admitted the survey.
Cococare does not suggest on this appeal that we should exclude
it. Nor do we see a reason to do so. We therefore ask whether
the survey evidence creates a genuine issue of material fact as
to whether “Cocoa Butter Formula” is not generic, but
descriptive.

       The District Court concluded that the survey had “little
or no probative value” and “should be afforded little or no
weight.” We understand the Court to have made that
determination within the summary judgment framework rather
than to have engaged in any inappropriate weighing of the
evidence at the summary judgment stage. See, e.g., Universal
City Studios, Inc. v. Nintendo Co., 

, 118 (2d Cir.
1984) (holding a survey to be “so badly flawed that it cannot be
used to demonstrate the existence of a question of fact”).
Cococare argues that we should review this decision for clear
error. It points to our decision in Johnson & Johnson-Merck
Consumer Pharmaceuticals Co. v. Rhone-Poulenc Rorer
Pharmaceuticals, Inc., 

, 135–36 (3d Cir. 2005),
where we employed that standard of review. Rorer did not

                              19
come to us at the summary judgment stage, however. It
involved a District Court that had weighed survey evidence
during the course of a four-day evidentiary hearing on a motion
for a preliminary injunction.8



     8
       We have explained the importance of the distinction
between the preliminary injunction and summary judgment
stages of litigation:
        In the posture before us–a trademark case in
        which summary judgment proceedings follow a
        grant of a preliminary injunction in the plaintiff’s
        favor–the distinction between the standards for
        summary judgment and preliminary injunction
        become critical. Failure to strictly observe the
        principles governing summary judgment becomes
        particularly significant in a trademark or
        tradename action, where summary judgments are
        the exception. [I]nferences concerning credibility
        that were previously made in ruling on [a] motion
        for a preliminary injunction cannot determine [a]
        Rule 56(c) motion and should not be used to
        support propositions that underpin the decision to
        grant the motion for summary judgment.
Doeblers’ Pennsylvania Hybrids, Inc. v. Doebler, 

,
820 (3d Cir. 2006) (alterations in original) (citation and
quotation marks omitted). The distinction between the two
standards remains as important in the context of weighing the
results of a survey as in making credibility determinations (the
issue in Doebler).

                              20
        Our case involves no such weighing of evidence or
factual findings. The District Court’s role here was to evaluate
the record evidence to determine whether Browne’s claims
could proceed to trial under the summary judgment standard. In
this context, we do not defer to the District Court’s resolution of
that legal question. Instead, we will conduct plenary review as
we have on other occasions when a District Court has granted
summary judgment. See Iberia Foods Corp. v. Romeo, 

, 302 (3d Cir. 1998); see also 

, 118
(showing no deference to District Court opinion).9

       The District Court faulted the survey for what it
perceived as two errors. It criticized the survey for not using the
term “Palmer’s,” believing that this omission “undermine[d]
Browne’s theory of the case” and made the questions “flawed
and misleading.” Dist. Ct. Op. at 9. The Court also believed
that the survey contained leading questions. For example, it
considered the question “What word or words would you use to
identify or describe a skin care product which contains cocoa
butter?” to be highly suggestive in order to evoke a specific
response.


   9
     We reject Browne’s argument that we should use an abuse
of discretion standard of review because we do not agree that
this is a question of the admissibility of the surveys. See In re
Paoli Railroad Yard PCB Litigation, 

, 749 (3d Cir.
1994) (applying a deferential standard of review to rulings on
admissibility of expert opinions).

                                21
        We do not agree these questions were so misplaced. The
survey was intended to reveal whether customers use the word
“Cocoa Butter Formula” to describe cocoa butter skin care
products or lotions. The parties’ genericness dispute turns on
that question. We thus steer away from the criticism that “some
of the survey questions use two of the three words (cocoa butter)
of Browne’s source identifier (cocoa butter formula).” 

to conjure how Browne could have pursued the core
genericness inquiry without doing so.10 As we believe the
questions are not highly suggestive or misleading, this case does
not, as the District Court suggests, find similarities in Novartis
Consumer Health, Inc. v. Johnson & Johnson-Merck Consumer
Pharmaceuticals Co., 

, 592 (3d Cir. 2002), and

.11



   10
    Cococare presents expert evidence indicating that “Cocoa
Butter” may have discouraged the use of those words in the
answers. But this does not deprive the survey of probative
value. It merely creates a question as to how much weight a jury
should give the survey.
  11
     In addition to this substantive difference, those cases differ
procedurally from this case. In both of those cases we reviewed
the weight given to surveys by District Courts considering
motions for entry of preliminary injunctions. See supra note 8
(discussing procedural distinction between preliminary
injunction and summary judgment stages of trademark
litigation).

                                22
       We also do not perceive any reason for Browne to have
included the word “Palmer’s” in its survey. This litigation
focuses on the term “Cocoa Butter Formula,” not on the
registered trademark “Palmer’s Cocoa Butter Formula.” The
inclusion of the word “Palmer’s” in the survey would have
confused matters and would have taken the survey outside the
“Thermos survey” model.12

        Browne’s survey does have non-trivial flaws, however.
Only 30% of valid respondents used a noun identifying the
product genus (e.g., lotion, cream). The majority of respondents
either answered with an adjective describing the product class
(e.g., healing, moisturizing) or did not answer. This suggests
that the questions confused many respondents. The survey may
have caused this confusion by deviating from the standard
“Thermos survey” model by asking respondents for terms
describing the products in addition to asking (as a “Thermos
survey” should) for terms identifying the products. The survey


  12
    Cococare does not distinguish in its opening brief between
“Teflon surveys” and “Thermos surveys.” It implicitly argues,
however, that Browne should have conducted the former rather
than the latter when it contends that “[t]he proper question
would have asked survey respondents in a straightforward
fashion whether the designation ‘cocoa butter formula’
identified a brand, or a common name.” Cococare does not
explain why Browne had to conduct a “Teflon survey” rather
than a “Thermos survey.”

                              23
likely would have been strongest if it had asked respondents, as
the “Thermos survey” also did, how they would ask at a store for
the type of product at issue.

        These flaws nonetheless do not deprive the survey of
probative value. The survey raises a reasonable inference that
“Cocoa Butter Formula” does not describe the product genus in
the opinion of the 46 respondents who described the product
class. A reasonable jury could rely on that inference to conclude
that consumers do not use the words “Cocoa Butter Formula” to
describe the category of skin care products containing cocoa
butter. Cococare could attack the inference at trial, but that does
not stop the survey from creating a genuine issue of material
fact. It is premature for us now to conclude that the survey does
not provide probative evidence that “Cocoa Butter Formula” is
not generic. Browne could have performed a better survey.
Indeed, it might have rued the survey’s design flaws after a trial.
But the survey is strong enough to allow Browne to survive
summary judgment on the genericness issue.

       Browne also points to evidence that competitors use
terms other than “Cocoa Butter Formula.” This evidence tends
to prove that the term is not generic. See 

n.20 (“Courts have long focused on the availability of
commonly used alternatives in deciding whether a term is
generic.” (citing Holzapfel’s Compositions Co. v. Rahtjen’s
American Composition Co., 

(1901))). It indicates
that other competitors (and thus Cococare) did not need to use

                                24
the word “Formula” to communicate to consumers the type of
products that it sold. This bolsters our conclusion that a genuine
issue of material fact exists on the question of genericness,13 and
thus the District Court should not have entered summary
judgment that “Cocoa Butter Formula” is generic.14 Instead, it
should have proceeded to a secondary meaning analysis.



   13
     In contrast, we struggle to perceive the relevance ascribed
by Browne to the fact that dictionaries do not include the term
“Cocoa Butter Formula.” Not all generic terms appear in
dictionaries, a fact so obvious Browne makes no effort to
disprove it.
   14
      The District Court appeared to reason that the registration
of “Cocoa Butter Formula” on the PTO’s supplemental register
rather than its principal register weakens Browne’s claim that
the term is descriptive. We know of no support for that view.
As the District Court correctly explained elsewhere, the validity
of a term that does not appear on the principal register and is not
distinctive (i.e., one that is descriptive rather than suggestive, or
arbitrary or fanciful) depends on its acquisition of secondary
meaning. See 

(explaining that arbitrary
or suggestive terms are distinctive and automatically qualify for
trademark protection, while descriptive terms only receive
trademark protection after a showing of secondary meaning).
        We also note that Browne is not entitled to summary
judgment in its favor on the question of genericness. The
weaknesses in the genericness survey alone create a genuine
issue of material fact on that point.

                                 25
              2.      Assuming “Cocoa Butter Formula” is
                      Descriptive, Has it Acquired Secondary
                      Meaning?

       Because a genuine issue of material fact exists on the
question whether “Cocoa Butter Formula” is generic, the parties
normally would need to proceed to trial to resolve that issue.
Even assuming that Browne prevailed and proved “Cocoa Butter
Formula” to be descriptive, it also would need to show that the
term had acquired secondary meaning which associated it with
Browne.

        Cococare moved for summary judgment on the basis that
Browne had not produced evidence creating a genuine issue of
material fact on the secondary meaning question. The District
Court set out in its opinion how a secondary meaning analysis
would proceed and laid out relevant factors from our case law.
But it did not go further. Instead, the Court concluded that “[i]n
this case [there] are other controlling factors which taken as a
whole show that cocoa butter formula is generic.” Dist. Ct. Op.
at 12. The parties presented the secondary meaning question to
the District Court, however, and we have their arguments before
us. We thus resolve that question because of our interest in
judicial economy, and affirm on this basis the District Court’s
entry of summary judgment. See Nicini v. Morra, 

,
805 (3d Cir. 2000) (stating that we may affirm on any ground
supported by the record).



                               26
       Secondary meaning is a new and additional meaning that
attaches to a word or symbol that is not inherently distinctive.
See generally 2 McCarthy on Trademarks § 15:1. We have
explained:

       Secondary meaning exists when the trademark is
       interpreted by the consuming public to be not only
       an identification of the product, but also a
       representation of the product’s origin. Secondary
       meaning is generally established through
       extensive advertising which creates in the mind of
       consumers an association between different
       products bearing the same mark. This association
       suggests that the products originate from a single
       source. Once a trademark which could not
       otherwise have exclusive appropriation achieves
       secondary meaning, competitors can be prevented
       from using a similar mark. The purpose of this
       rule is to minimize confusion of the public as to
       the origin of the product and to avoid diversion of
       customers misled by a similar mark.

Scott Paper Co. v. Scott’s Liquid Gold, Inc., 

,
1228 (3d Cir. 1978) (citations omitted).

       We have identified an eleven-item, non-exhaustive list of
factors relevant to the factual determination whether a term has
acquired secondary meaning:

                               27
      (1) the extent of sales and advertising leading to
      buyer association; (2) length of use; (3)
      exclusivity of use; (4) the fact of copying; (5)
      customer surveys; (6) customer testimony; (7) the
      use of the mark in trade journals; (8) the size of
      the company; (9) the number of sales; (10) the
      number of customers; and, (11) actual confusion.

Commerce Nat’l Ins. Services, Inc. v. Commerce Ins. Agency,
Inc., 

, 438 (3d Cir. 2000). “[T]he evidentiary bar
must be placed somewhat higher” when the challenged term is
particularly descriptive. 

       Browne’s proffered showing of secondary meaning
includes the following evidence:

      •      its use and promotion of the term “Cocoa
             Butter Formula” continuously for 20 years;

      •      the substantial amounts of money it has
             spent promoting the term “Cocoa Butter
             Formula;”

      •      the nature and quality of the advertising in
             support of the term “Cocoa Butter
             Formula;”

      •      Cococare’s alleged intent to copy the term

                              28
              “Cocoa Butter Formula;” and

       •      the increase in the sales of products
              bearing the term “Cocoa Butter Formula.”

       This evidence may seem, at first blush, to support
Browne’s claim that the term “Cocoa Butter Formula” has
gained secondary meaning. But serious flaws cause it to fail to
create a genuine issue of material fact on the question of
secondary meaning.

        The evidence’s core deficiency is that while it shows
Browne used the term “Cocoa Butter Formula” on many
occasions over a long period of time, it does not show Browne
succeeded in creating secondary meaning in the minds of
consumers. Although the evidence leaves no doubt that Browne
hoped the term would acquire secondary meaning, nothing
shows that it achieved this goal. Jurors would have to make a
leap of faith to conclude that the term gained secondary meaning
because the record fails to provide meaningful support. A jury
could evaluate the quality of the advertising or consider the rise
in product sales, but it would have to guess what lasting
impression the advertising left in the mind of consumers or what
portion of Browne’s revenue growth it caused.

       We indicated in Commerce National Insurance Services
that a plaintiff might establish secondary meaning through



                               29
evidence of advertising and sales growth. See 

A
plaintiff could create a reasonable inference, for example, that
a term had gained secondary meaning by showing that it had
appeared for a long period of time in a prevalent advertising
campaign. Evidence of revenue growth simultaneous with such
marketing would strengthen that inference, particularly if
supported by evidence of other factors among those we listed in
Commerce National.

        This case, however, differs in an important way from
such an example. Browne has introduced no evidence
indicating that it ever used “Cocoa Butter Formula” as a stand-
alone term in marketing or packaging. Instead, it always used
the term connected with the “Palmer’s,” forming the phrase
“Palmer’s Cocoa Butter Formula.” For example, Browne’s
lotion bottles bore logos with “Palmer’s” immediately above the
words “Cocoa Butter Formula,” creating one visual presentation
for the consumer. The marketing and sales evidence thus likely
would raise a reasonable inference that “Palmer’s Cocoa Butter
Formula” has gained secondary meaning in the minds of the




    15
      We do not suggest that a party attempting to establish
secondary meaning always must show that marketing materials
succeeded in creating buyer association or that the term
contributed to sales growth.

                              30
public.16

        But Browne wants to do something more complicated: it
wants to establish that a portion (“Cocoa Butter Formula”) of
the larger term (“Palmer’s Cocoa Butter Formula”) has acquired
an independent secondary meaning. Nothing in the record
would allow a jury to evaluate the strength of the term “Cocoa
Butter Formula” independently from the larger term including
“Palmer’s.” We thus conclude, under the specific circumstances
presented by this case, that the marketing and sales evidence
provided by Browne does not create a reasonable inference that
“Cocoa Butter Formula” has acquired secondary meaning.

        Nor does Browne’s asserted evidence of Cococare’s
intent to copy the term “Cocoa Butter Formula” create a genuine
issue of material fact as to secondary meaning. This evidence
pertains almost exclusively to trade dress (i.e., the overall
appearance of labels, wrappers, and containers used in




  16
     This case does not put that question at issue. The parties do
not dispute that Browne has a valid, registered trademark in the
term “Palmer’s Cocoa Butter Formula.” Cococare does seek
alteration of the principal registration of that term, but only to
the extent it includes the term “Cocoa Butter Formula,” which
it asserts is generic.

                               31
packaging a product),17 an issue not presented by this case.18
Browne only identifies one piece of evidence that conceivably
suggests an intent to copy. Cococare’s founder testified that he
may have known about Browne’s use of “Cocoa Butter
Formula” when Cococare began using that term on its own
products. But he also testified that he decided to use the word
“Formula” because it is a standard descriptor in the cosmetics
industry. He never testified that he copied Browne and nothing
in the record suggests that he did. Even viewing this evidence
in the light most favorable to Browne, we cannot discern how a
reasonable jury could conclude that Cococare intended to copy
Browne’s use of the term “Cocoa Butter Formula.”

       Browne could have overcome these deficiencies in its
evidence by conducting a secondary meaning survey in the same
way it conducted its genericness survey. It could have used
survey evidence to show that “Cocoa Butter Formula” had
acquired secondary meaning in the minds of consumers, thus
creating a genuine issue of material fact on that issue.19 We


  17
    See generally 1 McCarthy on Trademarks § 8:1 (comparing
trade dress and trademarks).
       18
     Browne apparently dropped a trade dress claim because
Cococare presented unrebutted evidence that it adopted the
contested trade dress earlier than Browne.
       19
     Cococare conducted a secondary meaning survey that it
argues tends to prove a lack of secondary meaning in the term

                              32
never have held, and do not hold today, that a party seeking to
establish secondary meaning must submit a survey on that point.
However, Browne’s failure to conduct a secondary meaning
survey leaves it without evidence of any sort in this case of the
secondary meaning of the term “Cocoa Butter Formula.”

       We thus conclude that Browne has failed to identify
evidence creating a genuine issue of material fact on the
question whether “Cocoa Butter Formula” has acquired
secondary meaning. Cococare is entitled to entry of summary
judgment on the basis that Browne lacks a protectable trademark
interest in the term “Cocoa Butter Formula.”

       B.     Cancellation or Alteration of Registrations and
              Remaining Arguments

       Having held “Cocoa Butter Formula” to be generic, the


“Cocoa Butter Formula.” That survey does not support
Browne’s position. Instead, it arguably provides evidence
against Browne and would be used by Cococare to rebut any
evidence of secondary meaning offered by Browne. We need
not attempt to discern, however, whether Cococare’s survey
undercuts any evidence by Browne. That is not our role at the
summary judgment stage. Instead, we ask whether Browne has
identified sufficient evidence to allow a jury to resolve the
secondary meaning question in its favor. We conclude that it
has not. Accordingly, we need not consider here any evidence
offered in support of Cococare’s position.

                               33
District Court briefly considered whether (as Cococare
requested in a counterclaim) it should require a disclaimer of the
“Cocoa Butter Formula” portion of the principal registration of
“Palmer’s Cocoa Butter Formula” and cancel the supplemental
registration of “Cocoa Butter Formula.” See 15 U.S.C. § 1119.
This provision states in relevant part: “In any action involving
a registered mark the court may determine the right to
registration, order the cancellation of registrations, in whole or
in part, restore canceled registrations, and otherwise rectify the
register with respect to the registrations of any party to the
action.” 

explained that “a controversy as to the
validity of or interference with a registered mark must exist
before a district court has jurisdiction to grant the cancellation
remedy.” Ditri v. Coldwell Banker Residential Affiliates, Inc.,

, 873 (3d Cir. 1992). A controversy clearly existed
as to the validity of the term “Cocoa Butter Formula” and its
place on the supplemental register. This raises the question of
the need for a disclaimer on the principal register of that portion
of the larger mark “Palmer’s Cocoa Butter Formula.”
Moreover, Cococare explicitly counterclaimed for modification
of that larger mark. These circumstances meet Ditri’s
requirement that “a controversy as to the validity of or
interference with a registered mark must exist.” 20


   20
     Since this case falls squarely within Ditri’s requirements,
we need not define the exact contours of “controvers[ies] as to
the validity of or interference with a registered mark.” For
example, we do not decide whether, in the absence of an

                                34
       The District Court declined to order relief under § 1119,
however. It apparently believed that a disclaimer of “Cocoa
Butter Formula” on the principal registration of “Palmer’s
Cocoa Butter Formula” would not benefit Cococare. See
generally 15 U.S.C. § 1119; 3 McCarthy on Trademarks § 19:63
(explaining effect of a disclaimer). Yet, even if Cococare will
not benefit from that disclaimer, we should not allow the
absence of a disclaimer on the principal register to confuse a
future business into believing that it may not use the term
“Cocoa Butter Formula.” Cf. 

. As such,
on remand the District Court should order that an appropriate
disclaimer should be entered on the principal register.

       The treatment of the supplemental registration is more
complicated since a descriptive term lacking secondary meaning
may not appear on the principal register, but may appear on the
supplemental register. See, e.g., In re Bush Bros. & Co., 

, 570 (Fed. Cir. 1989). The District Court should not
enter any order regarding the supplemental registration unless
for some reason Cococare proceeds to trial on its counterclaim
and prevails on the issue of the genericness of the term “Cocoa
Butter Formula.” If Cococare does prove “Cocoa Butter
Formula” to be generic, the District Court then could order the


appropriate counterclaim, a District Court may order the
addition of a disclaimer to a term (e.g., “Palmer’s Cocoa Butter
Formula”), including a portion found to be generic (e.g., “Cocoa
Butter Formula”).

                              35
removal of that term from the supplemental register.

       Cococare argues that Browne’s trademark claim must fail
because of Browne’s failure to assert its rights in the term
“Cocoa Butter Formula” in a timely fashion and, because of
alleged misrepresentations to the PTO, Browne has “unclean
hands.” Cococare also argues that it should receive summary
judgment on its affirmative defense that its use of the term
“Cocoa Butter Formula” constitutes fair use. We need not reach
these arguments in light of our conclusion that Browne does not
have a protectable trademark right in the term “Cocoa Butter
Formula.”

IV.    Conclusion

       We affirm the judgment of the District Court in all
respects but one. We remand to allow the Court to enter an
order addressing Cococare’s request for relief under 15 U.S.C.
§ 1119.




                              36