Warner-Lambert Co.v.Nabisco, Inc.

Trademark Trial and Appeal Board

Apr 7, 1999

No. 91092162 (T.T.A.B. Apr. 7, 1999)

Paper No. 38
DEB
THIS DISPOSITION IS NOT
CITABLE AS PRECEDENT OF THE TTAB APRIL 7,99
U.S. DEPARTMENT OF COMMERCE
PATENT AND TRADEMARK OFFICE
_____
Trademark Trial and Appeal Board
______
Warner-Lambert Company
v.
Nabisco, Inc.
_____
Opposition Nos. 92,062 and 92,162
to application Serial Nos. 74/307,317 and 74/348,595
filed on August 21, 1992 and January 14, 1993
_____
Dolores A. Moro, Anthony F. LoCicero and Laurie J. Gentile
of Amster Rothstein & Ebenstein for Warner-Lambert Company.
Steven H. Hartman for Nabisco, Inc.
______
Before Cissel, Quinn and Bucher, Administrative Trademark
Judges.
Opinion by Bucher, Administrative Trademark Judge:
Applicant, Nabisco, Inc., filed intent-to-use
applications to register the marks BURSTING FLAVOR1 for
chewing gum ("Flavor" is disclaimed) and BURSTING WITH
FLAVOR2 (no disclaimer) for bubble gum and chewing gum.
Both applications for registration have been opposed by

1 Application Serial No. 74/307,317 filed August 21, 1992.
2 Application Serial No. 74/348,595 filed January 14, 1993.
Opposition No. 92,062 & 92,162
2
Warner-Lambert Company on the ground that the marks are
merely descriptive of the goods, in that they each describe
qualities, ingredients or characteristics of applicant’s
bubble gum and/or chewing gum.
On April 17, 1996, in ruling on opposer’s motion for
summary judgment, this Board concluded that summary judgment
was inappropriate in this case. After reviewing carefully
the standards for summary judgment, the Board held that the
evidence of record was insufficient to warrant a finding
that these terms are merely descriptive as a matter of law,
citing, inter alia, Opryland USA Inc. v. Great American
Music Show Inc., 970 F.2d 847, 23 USPQ2d 1471 (Fed. Cir.
1992), and Olde Tyme Foods Inc. v. Roundy’s Inc., 961 F.2d
200, 22 USPQ2d 1542 (Fed. Cir. 1992).
The Board concluded that there was a genuine issue of
material fact in this case as to whether BURSTING FLAVOR and
BURSTING WITH FLAVOR, as applied to bubble gum or chewing
gum, would immediately convey to ordinary purchasers a
characteristic of applicant’s gum. In response to the
Board’s focus on the “perceptions of the average prospective
purchasers for the goods,” the parties have made customer
surveys a major focus of this trial.
In its notice of opposition to BURSTING FLAVOR, filed
on July 15, 1993, Warner-Lambert Company noted its ownership
of a federal trademark registration for CINN-A-BURST for
Opposition No. 92,062 & 92,162
3
chewing gum3 as well as an alleged family of other marks
having *BURST formatives. At the beginning of this
opposition proceeding, Opposer had seven pending, intent-to-
use applications to register marks having *BURST
formatives.4 Only two of these seven applications, for the
marks MINT*A*BURST for chewing gum,5 and FRUIT*A*BURST, also
for chewing gum,6 have since matured into federal
registrations. However, the ground for opposition is not
likelihood of confusion, but rather that BURSTING FLAVOR is
descriptive of a quality, attribute or characteristic of
these goods. Then several days after filing the above
opposition to the application to register the mark, BURSTING
FLAVOR, Warner-Lambert Company, on July 19, 1993, filed a
notice of opposition to the application to register the
mark, BURSTING WITH FLAVOR, on the same ground. After
applicant filed its answers denying the allegations in the
notices of opposition, the two proceedings were consolidated
with the mutual consent of the parties.
Opposer relied upon the above-enumerated registrations,
of which title and status copies were submitted for the

3 Reg. No. 1,652,772, issued July 30, 1991, §8 affidavit
accepted & §15 affidavit received.
4 In addition to Registration No. 1,652,772 for CINN-A-BURST,
Warner-Lambert (opposer) listed a “family” of “BURST” marks
[74/143,729 -- BUBB*A*BURST; 74/347,929 -- BUBB*L*BURST;
74/277,080 (and then later, 74/653,678 -- FRUIT*A*BURST;
74/183,780 -- MINT*A*BURST; 74/277,321 -- PEPP*A*BURST; and
74/277,317 -- SPEAR*A*BURST].
5 Reg. No. 1,865,359, issued November 29, 1994.
Opposition No. 92,062 & 92,162
4
record. The parties jointly stipulated to the admissibility
of the February 1995 declaration of Jill Rand, opposer’s
business manager, as well as the declaration of Hunter
McFarland, applicant’s senior product manager at the time of
opposer’s motion for summary judgment. Furthermore, the
parties stipulated that:
[a]ll deposition testimony and documentary
and/or physical exhibits submitted by the
parties in support of, or in opposition to,
as the case may be, Opposer Warner-Lambert
Company’s Motion for Summary Judgment filed
on February 13, 1995, in these opposition
proceedings, shall be deemed admitted in
evidence for all purposes in these opposition
proceedings.7
Pursuant to this stipulation, we have, inter alia, evidence
of Applicant’s actual usage of these phrases on the trade
dress of its gum packaging.
Opposer filed a notice of reliance on: the discovery
depositions of a number of applicant’s employees -- Ginger
Carol Gay,8 Brad Zabel,9 Kevin Bokeno10 and Robert J Gfeller,

6 Reg. No. 2,011,108, issued October 22, 1996.
7 This joint stipulation was signed by both parties on October
10, 1996
8 Associate Product Manager, Gummy/Chewy Products, Planters
LifeSavers Division, Nabisco Foods Group, RJR Nabisco.
9 Associate Product Manager, Care*Free Gum, Planters
LifeSavers Division, Nabisco Foods Group, RJR Nabisco (1991-
1992), who during this period reported to Bob Gfeller.
10 Applicant’s employee primarily responsible for consumer
research for Care*Free gum. He had extensive interaction with
Kristine A. Woodworth of Directions Research, Inc., during
relevant periods, and produced for his bosses the “topline
reports” made of record in this proceeding.
Opposition No. 92,062 & 92,162
5
Jr.11; the discovery deposition of Applicant’s outside
market research consulting firm, Directions Research, Inc.,
through its designated witness, Kristine A. Woodworth12;
dictionary definitions of the terms “burst” and “flavor”;
and selected responses from applicant to opposer’s first set
of interrogatories.
The record now also contains a consumer survey with
extensive back-up documentation submitted during deposition
testimony on October 8, 1996 by Dr. Donald E. Payne, who
conducted the survey on behalf of opposer (with exhibits)
and the deposition testimony of Dr. Payne again on July 1,
1997, during rebuttal. In turn, applicant submitted the
deposition testimony and “Expert Report” of George Mathis,
its own survey expert.
Following the Board’s denial of its motion for summary
judgment, opposer decided to conduct a survey of consumer
perceptions. Accordingly, in mid-July 1996, opposer’s
counsel, Ms. Gentile and Mr. LoCicero, asked Dr. Donald E.
Payne for his recommendations for consumer research to
determine the consumer perception of the phrases BURSTING
FLAVOR and BURSTING WITH FLAVOR. After conducting the
research agreed to by opposer’s counsel, on August 20, 1996,
Dr. Payne issued his report entitled “Consumer Perception

11 Senior Marketing Manager, Care*Free Gum, Planters LifeSavers
Division, Nabisco Foods Group, RJR Nabisco (1990 – 1992).
Opposition No. 92,062 & 92,162
6
Study.” The Board has reviewed the actual completed survey
instruments, the computer compilations, as well as the Final
Report.
Then, in January 1997, at the request of applicant,
George Mantis issued his own report. Presented as an expert
in consumer surveys, Mr. Mantis was critical of the
reliability and methodologies of the Payne study as well as
what conclusions could fairly be drawn therefrom. In
addition to continuing to emphasize their respective
contentions made at the time of the motion for summary
judgment, much of counsels’ arguments in the trial briefs
are centered on the significance, if any, of the Payne
survey.
Testimony and other non-survey evidence
Before tackling the intricacies of the survey design
and results, we feel compelled to review in some detail the
other evidentiary materials contained within this voluminous
record.
Applicant argues that there is no reason this Board
should make a determination under §2(e)(1) of the Trademark
Act based upon its own employees’ testimony using words like
“describe,” “mean,” or “connote.” On the other hand,
opposer points to multiple examples from applicant’s

12 This deposition was also included in the parties’ joint
stipulation of October 10, 1996.
Opposition No. 92,062 & 92,162
7
(Nabisco / Planters-LifeSavers) employees’ deposition
testimony where these deponents seem to agree with opposer
that these terms constitute descriptive attributes of the
gum:
Brad Zabel’s testimony:
A: “… [BURSTING FLAVOR means] that the
product had a lot of flavor…that there was
some duration component to the BURSTING
FLAVOR…EXTRA seemed to … perform better on
the attributes of long-lasting flavor,
bursting and juicy flavor.” [p.12].
Brad Zabel’s testimony:
Q: “…And bursting with flavor is a product
attribute, is that correct?”
A: “I think so. It’s an attribute of – that
can be used in talking about gum.” [p.13].
Brad Zabel’s testimony on the choice of word
BURSTING over the word BURST:
Q: “… ‘phrasing such as bursting flavor
seemed to describe a gum with a consistent
strong flavor, whereas a burst of flavor
suggested a weaker gum in the middle and
later stages of the chew.’ Does that
correspond to your understanding of the way
those two phrases were perceived by
consumers?”
A: “Yes. That’s the source of this
comment.” [p. 16]
Ginger Carol Gay deposition:
Q: “Do you think that when people chew
Care*Free gum, the perception of its flavor –
that is, what you described as an immediate
hit of flavor and one that continues – is
accurately described as ‘Bursting with
flavor.’?”
A: “Yes.” [p.79]
Kevin Bokeno’s “TopLine Report” dated Sept
26, 1991: “BURSTING FLAVOR” listed in table
under “product attribute.”
Kevin Bokeno’s testimony:
Opposition No. 92,062 & 92,162
8
Q: “OK, what did ‘has BURSTING FLAVOR’
mean?”
A: “A strong flavor that lasted throughout
the duration of the chew. It’s used to
connote a strong flavor.” [p.13]
Kevin Bokeno’s testimony on the choice of
word BURSTING over the word BURST:
‘consumers understood it to describe a gum
with a consistently strong flavor throughout
duration of chew better than BURST, seen as
weaker flavor in middle and end of chew.’
[pp. 19-20]
In making the point that these terms do not immediately
and directly refer to a feature of the gum, applicant argues
from Mr. Gfeller’s testimony that the term “Bursting flavor”
means something more, if not different, than simply a
strong, continuous flavor. Mr. Gfeller testified that
“Bursting flavor” was chosen to impart the image of a flavor
that “tends to grow and grow and evolve and fill or consume
the consumer's mouth, and therefore gives them a feeling of
bursting, long-lasting flavor.”
Opposer responds by pointing out arguably damaging
language elicited from Mr. Gfeller on cross-examination:
Gfeller testimony:
Q: “Would you agree that the list of factors
[in table, including ‘has bursting flavor’]
on the vertical axis of the chart are product
attributes?”
A: “Yeah, they are.” [p.19]
Gfeller testimony:
Q: “..does that portion of the sentence
indicate a characteristic of the Care*Free
Sugarless Bubble Gum?”
A: “Yeah, that it’s bursting with juicy
bubble gum flavor.”
Opposition No. 92,062 & 92,162
9
Applicant points out that the imaging and positioning
studies were tools of market research, done exclusively for
marketing purposes. Applicant argues that there is no basis
for going from these studies (and related testimony)
containing words like “describe,” “mean,” or “connote” to a
legal determination under Sec. 2(e)(1). Yet opposer points
out rather explicit testimony undermining applicant’s
position:
Kristine Woodworth on imaging study (existing
market image in 1991): “… ‘initial burst of
flavor’ was one of ‘the product attributes’
talked about with Kevin Bokeno.” [pp. 20-21]
Applicant also points to the findings of the
Positioning Study, and observes that respondents generated a
diverse set of words having the same connotation as
“bursting” (e.g., “zing,” “flavor surge,” “blast,” “shock
to your mouth,” “ice in your mouth,” “juicy cool,” “minty,”
“minty soft,” “slap in your face,” “cool water on the face,”
etc.). Applicant rightly questions whether such a wide
spectrum of expressions –- some of which are synonymous with
neither the relevant “Bursting” phrase nor each other -- can
provide evidence of descriptiveness under §2(e)(1) of the
Lanham Act. In fact, applicant argues that the cited
marketing studies undertaken by applicant in years past
actually support the opposite conclusion -- that these terms
do not immediately and directly describe one, specific
characteristic of gum. Applicant points out that Ms.
Opposition No. 92,062 & 92,162
10
Woodworth specifically says during her testimony that she
does not really know what these terms are supposed to mean.
Again, however, opposer points to her testimony on
cross-examination that supports its position. Opposer’s
counsel points to the following interview question in the
positioning study: “I’m going to read you several phrases
which could be used to describe bubble gum…” In her
deposition, Kristine Woodworth is asked:
Q: “ What was meant by the phrase ‘… phrases
which could be used to describe bubble gum’?”
A: “We’re just trying to describe to the
respondents what we meant by the product
attributes. Just, ‘here are some words that
could be used to describe gum’.”
Q: “Like ‘having Bursting Flavor’?”
A: “Eh-huh.”
On the question of what the advertising copy reflected
in the record communicates, applicant argues that this
advertising supports a finding of suggestive meaning and not
mere descriptiveness. Opposer points out specific language
from the advertising copy used in the positioning study to
make the opposite point:
CARE*FREE sugarless bubble gum is bursting
with juicy bubble gum flavor. Enjoy the
bursting flavor of CARE*FREE sugarless bubble
gum. When you chew a piece, the juicy bubble
gum flavor fills your mouth and lets you
break away within the world around you.
CARE*FREE sugarless bubble gum’s bursting
juicy flavor lets you break away.
Opposer goes on to point out that based upon data
collected from the respondents, applicant decided that
Opposition No. 92,062 & 92,162
11
“Bursting Flavor” represented a more viable positioning than
“Long-Lasting Flavor” –- yet another alternative,
descriptive phrase.
Applicant stresses that its gum has no crystals.
Applicant has argued throughout this case that nothing in
its gum actually “bursts” in the dictionary sense of this
one word because no “exploding” or “abrupt” action is
characteristic of this gum. In fact, applicant argues that
its marks, “Bursting with flavor” and “Bursting flavor,” are
actually incongruous phrases that by definition require some
mental effort to understand.
On the other hand, opposer argues that “Bursting with
flavor” is an entire expression merely descriptive of an
attribute or characteristic of applicant’s goods. Opposer
counters that these terms must be looked at in their
entireties. The alleged marks are not just “Bursting” in
the abstract, but rather the word “Bursting” within a phrase
where it modifies, or is combined with, the word “flavor.”
Opposer argues that every competitor in this field needs to
be able to use non-proprietary terms like “bursting,”
bursting flavor,” “bursting with flavor,” etc.
On this important point, we agree with opposer.
“Bursting with flavor” and “Bursting flavor,” are not
arbitrary, coined or fanciful terms. Rather, these are
combinations of common English words. Applicant is not
Opposition No. 92,062 & 92,162
12
using them in a novel combination. Consistent with the
dictionary entries made of record by opposer, we find that
these expressions would be readily understood as describing
food items, such as fruits, meats and spices.
The Payne Survey
We now turn to a detailed examination of the Payne
survey, which since 1996 has occupied much of the attention
of the parties to this dispute.
In a nutshell, the Payne survey concluded that three
out of five respondents (60%) thought the term “Bursting
flavor” conveyed to them an attribute of taste or flavor.
Similarly, about four out of five respondents (78%) thought
the term “Bursting with flavor” conveyed to them an
attribute of taste or flavor. Opposer argues that inasmuch
as its survey was properly designed and conducted according
to fair and scientific methodologies, it proves that
consumers perceive applicant’s alleged marks as merely
descriptive.
Applicant contends that the Payne study is “fatally
flawed” and cannot possibly demonstrate in a legal sense the
mere descriptiveness of the terms “BURSTING FLAVOR” or
“BURSTING WITH FLAVOR.” Applicant’s counsel and expert
argue that: these marks mean many different things to
different consumers; Dr. Payne’s own testimony on cross-
examination shows that the word “conveys” could well be used
Opposition No. 92,062 & 92,162
13
to generate free associations from respondents; these two
marks have no readily understood meaning, and they describe
nothing specific about the taste or flavor of the gum; the
ambiguity of the word “convey” in questioning does not
elicit immediate and direct meaning (i.e., vagueness of word
“convey” does not distinguish between suggestive and
descriptive matter); the biased structure of descriptive
terms immediately preceding the test phrases in three of the
four different rotations, the repeated use of probe
questions, along with large but meaningless net percentages
are fatally flawed; and that absent any degree of
“particularity” in alleged descriptiveness, the survey
actually makes the case for suggestiveness.
Opposer’s rebuttal is that the Payne survey was
properly designed; the survey was conducted in accordance
with proper methodologies; applicant has submitted
criticisms of opposer’s survey without proffering a survey
of its own; if the Board finds any weaknesses in the Payne
survey, it merely affects the weight of the survey results,
but does not destroy its probative value; the Board’s
criticisms (in the decision denying summary judgment for
opposer) of attaching legal significance to the words
“connotes” or “describes” in applicant’s earlier market
research is not relevant to the use of the word “conveys” in
a survey specifically designed to test descriptiveness at
Opposition No. 92,062 & 92,162
14
some later date; the accumulated taste/flavor responses are
not distinct, vastly different, remote or tangential, but
rather comprise minor variations on a common theme,
logically and properly combined together into a single set
of like responses; and, that a review of the data
demonstrates a high degree of similarity of meaning among
the combined “taste/flavor” categories.
Opposer and its expert defend the survey’s failure to
permit expressly an answer from respondents of “don’t know,”
arguing the presence of this option would have reduced the
accuracy of this survey. Applicant argues that without the
“don’t know” possibility, Payne was forcing answers.
According to applicant, because this is an opinion survey
rather than testing knowledge of facts, not everyone will
have an opinion, but respondents will still feel compelled
to come up with an answer –- any answer. In rebuttal,
Opposer argues that employing open-ended, unbiased questions
is a much better method for getting consumers’ real
perceptions than are more leading questions. Opposer argues
these were not heavy or intellectual questions -- the
interviewers merely presented a series of everyday words to
young gum-chewers in a shopping mall, and recorded their
reactions.
The Payne survey introduced two “control” terms in the
hope of anchoring points on the generic end of the spectrum
Opposition No. 92,062 & 92,162
15
–- “Sugarless gum” and “Artificially flavored.” Applicant
argues that since the survey screener in each case had
earlier questioned possible respondents about what type of
gum they had bought recently or used regularly (i.e.,
sugarless gum, sugared gum, or sugarless bubble gum, sugared
bubble gum), this was a form of conditioning for
descriptiveness responses that inevitably distorted later
answers as to what was conveyed to the respondent by the
term “sugarless gum.”
The Payne survey also introduced two “control” terms in
the hope of anchoring points on the source-indicating end of
the spectrum – “Double your pleasure” and “The candy coated
gum.”13 However, more respondents put this into the “taste-
flavor category than recognized it as a source indicator.
In fact, the taste/flavor scores for “Double your pleasure”
rivaled those of the two test phrases herein. Applicant
contends that the survey clearly has a high percentage of
“false positives” for taste/flavor, invalidating it as an
instrument to test descriptiveness. In its rebuttal,
opposer argues that while the survey used a phrase it

13 Neither of the parties dwelt on the seemingly flawed choice
of this term as a source indicating control in the Payne survey,
so we will not pursue it at length either. However, this seems
to be a reference to opposer’s 1993 registration on the
Supplemental Register for “THE ORIGINAL CANDY COATED GUM” for
chewing gum [Reg.No. 1,802,989] -- where the generic term “candy
coated gum” is disclaimed. To the extent applicant argues
conditioning for descriptiveness, this is arguably yet another
reason to find bias in the Payne survey’s results.
Opposition No. 92,062 & 92,162
16
thought to be a strong, well-known trademark, perhaps it
proved at present not to be nearly as strong a trademark in
this industry as opposer had assumed.
Dr. Payne testified about the manner in which data
processing persons read verbatim responses and then placed
them in like categories for coding. After several trials,
it became necessary to add new categories for like kinds of
responses not listed in the initial tables. Clearly, prior
to his testimony in this case, Dr. Payne found it necessary
to made some hand-written corrections to the summary tables.
Applicant focuses on several specific examples of hand-
written corrections to fix earlier cases of miscoding. In
short, applicant raises suspicions about the reliability of
the numbers, and argues that any quick synopsis of the Payne
findings that stresses the 60% and 78% numbers would be most
misleading.
Opposer argues in rebuttal that any miscoding resulted
in very small discrepancies that are irrelevant given the
overwhelming thrust of the survey data demonstrating
descriptiveness.
Opposer defends the use of “probe questions” in the
data-gathering phase of the study. Following the initial
question about what each one of the rotating, five phrases
“conveyed” to the respondent, the interviewer would ask two

Opposition No. 92,062 & 92,162
17
more probe questions – i.e., “What else?” and “Anything
else.” Opposer argues that this format is consistent with
standard survey procedure, especially if one is to get
further elaboration on open-ended questions. Opposer noted
that the interviewers used the same five cards (containing
the four “control” phrases discussed above, plus either
“Bursting Flavor” or “Bursting with Flavor”), but rotated
their order to deal with the possibility of “order bias.”
Applicant counters that the responses were clearly the
result of a survey design that fostered extrapolation,
thought, creativity and imaginative thinking. Applicant
contends that the very first probe question encouraged
guessing. Then, according to applicant, the repetition of
probe questions later in the rotations taught the
respondents to employ multi-stage reasoning. Applicant
contends that the nature of the verbatim responses shows
huge mental leaps and lots of free associations by the
survey respondents. Applicant claims that this tended to
increase the chances of “taste/flavor” answers to the probe
questions.
In response, opposer points out that a high percentage
of respondents had no answer to the probes, and therefore
their first answer was direct and immediate. Furthermore,
opposer points out that a careful review of the data shows
that the nature of responses remained the same throughout
Opposition No. 92,062 & 92,162
18
the progression of these fifteen questions (i.e., three
queries -- the initial question and two follow-up probe
questions -- for each of five terms) asked of each survey
subject, i.e., later questions did not result in a higher
percentage of answers to probe questions.
Opposer notes that given the nature of the mall
intercepts, the explicit instructions and the
straightforward nature of the queries, the respondents’
answers were immediate and direct.
By contrast, applicant notes that there is no
indication anywhere of the amount of time respondents took
to answer questions.
However, opposer points out that even applicant’s own
expert agrees that the time it took a respondent to answer a
question is probably irrelevant.
Dr. Payne indicated that if the verbatim answers
revealed no indication of a respondent’s ruminations (e.g.,
‘I believe…,’ ‘I guess I would have to say…’), one has to
assume the answers were direct and immediate. Applicant,
however, contends there is no question but that some complex
thought processes led to these answers. Applicant argues
that even Dr. Payne’s own testimony on cross-examination
shows that the term “conveys” is a term that can be used to
evoke free associations. According to applicant,
associations or perceptions, if not made with sufficient
Opposition No. 92,062 & 92,162
19
immediacy or directness, are useless in making our §2(e)(1)
determination.
Applicant charges that given the expanded scope of the
study’s “taste/flavor” category, the net and total numbers
are meaningless. It contends that if one combines the
vagueness of “conveys” with two follow-up, probe questions
to each phrase’s initial query, and then finally throws
almost anything into the “taste/flavor” category, the study
results are not compelling at all. In reflecting on all the
testimony, including that of Applicant’s employees and
contractors (made part of the record pursuant to the terms
of the parties’ October 10, 1996 joint stipulation),
applicant argues that, like the study, all the answers are
different, vague, and subjective.
Yet viewing the same evidence, opposer reaches the
opposite conclusion -- arguing that the survey results had a
particularity of meaning, not unlike that shown by the
testimony taken of applicant’s employees and outside
marketing specialists.
Opposer notes that three out of ten (29%) respondents
mentioned a specific brand of gum in answering the “conveys”
query. In fact, opposer’s Cinn-a-Burst, Mint*a*Burst,
and/or Fruit*a*Burst were mentioned by 22% of the
respondents while no respondent mentioned applicant’s
Care*Free gum. Applicant argues that in reality, the
Opposition No. 92,062 & 92,162
20
opposer is simply concerned about not being able to use
applicant’s suggestive marks with opposer’s own branded
*A*BURST products.14 Opposer argues that in spite of
applicant’s usage of this term, consumers are making no
brand association with applicant’s CARE*FREE gum precisely
because the term is merely descriptive. (The survey results
and the parties’ respective arguments are almost identical
throughout as to the term “Bursting flavor” and “Bursting
with flavor.”)
Board’s Analysis of Survey
Surveys and survey results make appearances in reported
trademark cases in a number of different contexts: to show
acquired distinctiveness, to demonstrate likelihood of
confusion, to prove implied false advertising claim during
litigation under Section 43(a) of the Lanham Act, and to
determine genericness.
On the issue of genericness, various courts have
applauded the methodology of the survey conducted in the
case of E.I. DuPont de Nemours & Company v. Yoshida
International, Inc., 393 F.Supp. 502, 185 USPQ 597 (E.D.N.Y.
1975) in which the term "TEFLON" was at issue. That survey
offered the examples of "CHEVROLET" to illustrate what it

14 According to applicant, opposer is really afraid these
registrations will “…stymie opposer’s nascent, yet audacious goal
of obtaining a family of BURST gum marks…”
Opposition No. 92,062 & 92,162
21
meant by a brand name and "automobile" to illustrate what it
meant by a product type. Then, respondents were asked to
classify other brand names and product names among which was
“Teflon.” Similarly, the “Thermos” survey asked
participants how they would identify a particular product
given that it performs certain functions. This was also an
effort to determine if the claimed trademark for the product
was generic. See American Thermos Prods. Co. v. Aladdin
Indus., Inc., 207 F.Supp. 9, 134 USPQ 98 (D. Conn. 1962),
aff'd, 321 F.2d 577, 138 USPQ 349 (2nd Cir. 1963).
In making a genericness determination, a consumer
survey has value to the extent it provides the tribunal with
insights into what consumers understand a specific term to
be. WSM, Incorporated v. Hilton et al., supra at 1328, n.3.
The critical question, in many cases, is whether or not the
alleged source-identifier is inherently distinctive to the
prospective purchasers of the article. However, some courts
have cautioned that a Teflon or Thermos survey is helpful
only when the distinction between the brand name and the
product genus is obvious. See A.J. Canfield Company v.
Honickman, 808 F.2d 291, 303, 1 USPQ2d 1364, 1374 (3d Cir.
1986). These courts have expressed doubts that any consumer
survey can be helpful when an arguably descriptive term is
used as a brand name. A.J. Canfield Company at 1373-1374.
Opposition No. 92,062 & 92,162
22
Applicant has raised just such an issue about the Payne
survey herein. Applicant argues that it is incredibly
difficult, if not impossible, to design a survey testing for
descriptiveness because of the critical need to measure
immediacy. In spite of the widespread use of surveys by
parties involved in trademark litigation, there is a real
dearth of reported decisions having surveys providing good
benchmarks against which to measure the Payne survey’s
attempt to determine mere descriptiveness. We do agree with
applicant that it is indeed difficult to calibrate a survey
that reliably determines the descriptiveness of a multi-word
phrase like “Bursting flavor” or “Bursting with flavor.”
Opposer counters that inasmuch as it has done a valid
survey and applicant has not done a survey, the Board should
make inferences adverse to applicant based on that fact.
Opposer argues that most of applicant’s criticisms are not
well taken, and certainly that none are fatal.
To understand the difficult assignment undertaken by
Dr. Payne, it behooves us to look at the spectrum of
increasing distinctiveness of trademarks under the classic
formulation set out by the late Judge Friendly. Terms
purporting to be source-identifiers are often classified in
categories of generally increasing distinctiveness as
follows: (1) generic, (2) descriptive, (3) suggestive, to
(4) arbitrary or fanciful. See Abercrombie & Fitch Co. v.
Opposition No. 92,062 & 92,162
23
Hunting World, Inc., 537 F.2d 4, 9, 189 USPQ 759, 769 (2nd
Cir. 1976). How applicant’s alleged marks have been used,
and how the public understands them, is a question of fact.
However, lest that sounds too easy, these categories, like
colors in a spectrum, tend to blur at the edges and merge
together, making it difficult to apply the appropriate
label. Nonetheless, the correct categorization by any
tribunal of a given term is a factual issue.
This Board is often called upon to make these difficult
determinations when deciding whether a word is, or is not,
merely descriptive. In re Abcor Development Corp., 588 F.2d
811, 814, 200 USPQ 215, 218 (CCPA 1978). In short, terms
that are suggestive as applied to a given product are
naturally understood by the consuming public as designations
of origin. On the other hand, descriptive terms are ill
suited to serve as designations of origin. This is because
such terms are naturally understood by the consuming public
in their ordinary descriptive sense.
We saw above that a Teflon survey is designed to test
terms on opposite ends of the trademark spectrum of degrees
of distinctiveness. However, Dr. Payne’s task of
definitively locating the correct classification of these
terms on the trademark spectrum is much more complicated in
the instant case. Without doubt, “Bursting with flavor” and
“Bursting flavor,” are not arbitrary, coined or fanciful
Opposition No. 92,062 & 92,162
24
terms. On the other hand, no one has argued that they are
incapable of serving a source-indicating function. Hence,
the argument herein is descriptive versus suggestive. These
phrases seem to be located on a continuum where determining
the exact classification is made more challenging by the
blurred edges separating them. However, in a reported
decision from the Eighth Circuit, a somewhat similar survey
was used to make a determination of descriptiveness versus
suggestiveness. See Anheuser-Busch, Incorporated v. The
Stroh Brewery Company, 750 F.2d 631, 224 USPQ 657, 662 (8th
Cir 1984)[“ L A ” for light alcohol beer]. For example, in
comparing the scope of Dr. Payne’s survey with the “ L A ”
case, we note that both surveys were conducted in twelve to
fifteen shopping malls in geographically diverse parts of
the country. In the Payne study, respondents were screened
and gum-chewers were queried further provided they fit into
preset quotas for age and type of gum chewed most often.15
In Anheuser-Busch (the “ L A ”case), respondents were
screened for the proper characteristics and were
subsequently divided into two test cells: those who have
not ever seen, heard of, or tried plaintiff's product, and
those who have seen, heard of, or tried the product.

15 As a matter of fact, the data show that differences among
age groups, between gender groups and by which type of gum the
respondent chewed were generally not significant in the instant
case.
Opposition No. 92,062 & 92,162
25
The parties in the instant case have argued extensively
the design of the Payne survey. They disagree, for example,
over the basic format and wording of the questions. It goes
without saying that in developing a survey questionnaire,
the survey’s designer must pay close attention to question
order, context and wording. For example, in the Payne
study, respondents were told that they would be shown “…
some phrases that are used in connection with chewing gum
and bubble gum, and I’d like you to tell me what they convey
to you.” Respondents were then shown five phrases, one
after the other, and their answers were recorded verbatim.
The order of the presentation of the phrases was
systematically rotated.16 With the main part of the
questionnaire behind them, the respondents were then given a
card listing brands of chewing gum and bubble gum and asked
which one they chewed most often and which others they
chewed occasionally.
The members of this Board panel do not purport to be
experts in survey design and methodology. However, we do
agree with several of applicant’s most damaging criticisms
of the Payne survey. In short, both suggestive terms and
descriptive terms are able to “convey” something, and so the

16 While George Mantis’ criticisms of having the “Burst…” test
phrase immediately following descriptive phrases three of four
rotations may be one more indication of introducing bias into the
results, it does not seem to us to be as critical as other
criticisms to which we have devoted more time.
Opposition No. 92,062 & 92,162
26
form of the “conveys” question is a major weakness of this
survey. We do wonder in this context whether a Teflon-type
survey that first tried to educate respondents to the
differences between a suggestive term serving as a source-
indicators, on the one hand, and descriptive matter, which
does not initially function as a source-indicators, on the
other hand, would have been possible.
In the “ L A ” case, responses were gathered from the
following statements and questions: “Please describe the
product.” “What type of beer is it?” “What is the brand
name of this new product?” “What company produces this
brand?” Query whether opposer could more easily defend the
survey results if it had adopted a similar approach using
applicant’s Care*Free trade dress having thereon the
“Bursting with flavor” phrase? Perhaps including some
closed questions having a quantifiable scale would also have
helped this tribunal determine fine shades of differences
required in making a decision between suggestiveness and
descriptiveness.
We also share applicant’s perception that the survey
design encouraged flavor/taste responses. For example,
among the phrases chosen by Dr. Payne, survey respondents
were asked what the phrase “Double your pleasure” conveyed
to them. Curiously, this registered trademark of the Wm.
Wrigley Jr. Company produced taste/flavor scores similar to
Opposition No. 92,062 & 92,162
27
those for the terms “Bursting flavor” or “Bursting with
flavor.” In spite of opposer’s arguments to the contrary,
we agree with applicant that this alone demonstrates that
the Payne survey is unreliable in that it produces a notable
percentage of “false positives.” Apart from any other
alleged flaws, this result alone indicates that the survey's
basic mechanics and resulting data are subject to
question.17 Yet while finding that the Payne survey has
serious weaknesses, we are not prepared to discard it
entirely as being “fatally flawed.”
Given the cumulative nature of the flaws of this
survey, we definitely agree with applicant that the 60% and
78% numbers are misleading. Results from a reliable
trademark survey having percentages this high would likely
ensure a clear-cut victory for the litigant/survey
proponent. However, it would be most misleading in the
instant case to take these unusually large, bottom-line
percentages as conclusive on the issue of descriptiveness.

17 We conclude that the structure and type of questions
employed by opposer in the Payne survey avoided a bias in one
direction (e.g., by not encouraging false negatives), but this
certainly does not preclude the possibility that the survey
design created a bias in another direction (e.g., false
positives).
As applicant points out, an article placed into the record
by opposer deals with the problem of “false negatives” (where the
respondent has an underlying attitude but expresses no opinion),
while applicant is rightly concerned about the instant study’s
tendency to produce “false positives.” We agree with applicant
that the survey design gave plenty of opportunities for the
respondent who does not have an attitude to express an opinion in
answering these interview questions.
Opposition No. 92,062 & 92,162
28
These numbers exaggerate the sense of particularity and
uniformity of the respondent’s actual answers.
Mr. George Mantis, on behalf of applicant, has
identified a number of other possible flaws in the survey.
For example, he decries the failure explicitly to give
respondents a “don’t know” option in answering the
questions.18 Applicant has also raised concerns about the
reliability of Dr. Payne’s setting up broad categories, in
coding and in tabulating the results.19 However, the record
before us contains the total responses, which can then be
reviewed to ensure the responses were correctly coded
according to the system Dr. Payne designed. In reviewing
boxes of responses and comparing them with computer
summaries, we did not find any major discrepancies with this
part of the process.
We agree with applicant that a number of aspects of
opposer’s survey are rightly criticized. In spite of the
tendency of the Payne survey to generate false positives,

18 One of the articles made part of the record deals with the
availability of explicit “Don’t Know” options. (See also opinion
text, p. 13). It is clear from the literature on this subject
made a part of this record, that in selecting the specific
response options, the survey designer may choose to either
include or exclude a “Don’t Know” box, and should then be able to
defend that choice in the context of the entire survey
methodology and results.
19 Mr. Mantis has criticized Dr. Payne’s arbitrariness in
selecting specific categories, his need to add new categories
part way through the survey (e.g., to count for like-kinds of
responses not listed in the initial tables), in the coding of
Opposition No. 92,062 & 92,162
29
the ambiguity inherent in the word “conveys,” and other less
glaring flaws, on balance, its results still deserve some
weight. The survey results -- even after being accorded
minimal weight -- when coupled with the testimony of
applicant’s employees and outside marketing consultants,
supra, demonstrate that these phrases communicate to
potential customers rather directly a flavor or taste
attribute of these goods.
Even though we are compelled to discount sharply the
results of the Payne survey, one could still fairly conclude
that a plurality of the survey respondents offered a
“taste/flavor” response sometime in their answers to what
the affected terms meant to them. This conclusion is
entirely consistent with other parts of the instant
record.20 Given the overall weaknesses of this survey, the
most we feel comfortable concluding from the Payne survey is
that its results tend to corroborate a finding of
descriptiveness based upon the balance of the record.
Of course, if a term is found to be descriptive without
a showing of acquired distinctiveness, we assume in such a
case that it does not have a distinctive quality. Both of
these applications are intent-to-use applications in which

specific responses, and questions Dr. Payne’s hand-written
corrections to the tables, etc.
20 Discussed above in pp. 6 – 12, supra.
Opposition No. 92,062 & 92,162
30
no Statements of Use have been filed. Applicant was never
asked to prove acquired distinctiveness during the ex parte
prosecution of these two applications. This is a question
on which no evidence was submitted, either before the
Trademark Examining Attorney or before this Board. The
issue was neither tried nor briefed by the parties in these
opposition proceedings.
Decision
BURSTING FLAVOR for chewing gum and BURSTING WITH
FLAVOR for bubble gum and chewing gum are merely descriptive
of applicant’s gum within the meaning of Section 2(e)(1) of
the Lanham Act. Accordingly, the oppositions are sustained
and both applications for registration are hereby refused.
R. F. Cissel
T. J. Quinn
D. E. Bucher
Administrative Trademark
Judges, Trademark Trial and
Appeal Board