Comparative advertising in which the competing product is explicitly named is a relatively new weapon in the Madison Avenue arsenal.1 These cross-appeals by two of the leading manufacturers of analgesics— pain relief tablets — raise questions regarding the permissible boundaries of this novel approach to consumer persuasion. The parties appeal from an order of the United States District Court for the Southern District of New York, Charles E. Stewart, Judge, enjoining the use in television and printed advertising of certain product superiority claims of “Anacin” over “Tylenol.” American Home Products Corp. v. Johnson & Johnson, 436 F.Supp. 785 (S.D.N.Y.1977). The order was issued in an action for a declaratory judgment initiated by appellant American Home Products Corp. (AHP), the manufacturer of Anacin, against McNeil Laboratories, Inc., the manufacturer of Tylenol, and its parent corporation, Johnson & Johnson (collectively McNeil), seeking a ruling that the advertising is not false. McNeil counterclaimed, alleging, inter alia, that the advertisements were false and misleading under Section 43(a) of the Lanham Act, 15 U.S.C. § 1125(a). The district court found that the advertising violated Section 43(a), and accordingly the judgment enjoined AHP from representing that Anacin provides superior analgesia to Tylenol “in the context of a representation as to any anti-inflammatory property of Anacin. tt
AHP argues on appeal that (1) even on the basis of thexdistrict court’s findings, there is no ground for relief under Section 43(a); (2) the findings are clearly erroneous in concluding that (a) the Anacin advertising claims greater pain relief (Claims One and Two); (b) the disputed claim of superiority in reducing pain from inflammatory conditions (Claim Two) is false; and (c) the anti-inflammatory claim for the conditions listed in the advertising (Claim Three) is unsubstantiated; and (3) the terms of the injunction are too indefinite to comply with Fed.R.Civ.P. 65(d).2 McNeil appeals from the district court’s refusal to enjoin those portions of the Anacin advertising which, McNeil asserts, present misleading claims of Anacin’s faster onset of analgesia and harmlessness to the stomach. Since we are of the view that Judge Stewart’s order is based upon sound legal principles, that his findings are not clearly erroneous, and that the injunction as framed is proper in scope and specificity, we affirm.
1. THE FACTS
AHP’s product, Anacin, is a compound of aspirin (ASA), its analgesic component, and caffeine. McNeil’s product, Tylenol, affords analgesia through the ingredient acetaminophen (APAP). Anacin advertises more heavily than the other leading aspirin brands. It took over the Number One pain reliever spot from another aspirin-based product, Bayer Aspirin, a few years ago. Since the summer of 1976, however, Tylenol has replaced Anacin as the largest selling over-the-counter (OTC) internal analgesic product. Anacin remains the largest selling aspirin-based analgesic.
The lawsuit arose out of two Anacin advertisements initiated shortly after Tylenol became market leader. The first is a thirty-second television commercial initially aired by CBS in late November, 1976, and by NBC in early December, 1976. It commences with the phrase: “Your body knows the difference between these pain relievers . and Adult Strength Anacin,” and *163asserts its superiority to Datril, Tylenol and Extra-Strength Tylenol.3 The second advertisement was introduced in national magazines in late January, 1977. It carries a similar theme, stating that “Anacin can reduce inflammation that comes with most pain,” “Tylenol cannot.”4
The controversy began when McNeil protested the television commercial to the networks and the magazine advertisement to the print media on the ground that they were deceptive and misleading.5 McNeil also complained to the National Advertising Division of the Better Business Bureau. These protests were, for the most part, unsuccessful. CBS, NBC and the print media, continued to carry the two advertisements without alteration.6 As a result of the protests, AHP filed a declaratory judgment action under 28 U.S.C. § 2201 and sought to enjoin McNeil from interfering with the dissemination of the commercial and the printed advertisement. McNeil eounter-claimed under Section 43(a) of the Lanham Act, 15 U.S.C. § 1125(a),7 urging that the following claims contained in AHP’s advertisements were false: (A) that Anacin is a superior analgesic to Tylenol, (B) that Anacin is an efficacious anti-inflammatory drug for the conditions listed in the advertisements,8 (C) that Anacin provides faster relief than Tylenol, and (D) that Anacin does not harm the stomach. It sought declaratory relief and an injunction prohibiting AHP from continuing to make false claims which disparaged Tylenol.
After denying McNeil’s motion for a preliminary injunction, Judge Stewart held an expedited trial on the merits. Principally on the basis of consumer reaction surveys, he concluded that the advertisements made the following representations: (1) the television commercial represented that “Anacin is a superior analgesic generally, and not only with reference to particular conditions such as those enumerated in the ad [see *164note 3 supra ] or to Anacin’s alleged ability to reduce inflammation,” id. at 796 (Claim One); (2) the print advertisement claimed that “Anacin is a superior analgesic for certain kinds of pain because Anacin can reduce inflammation,” id. (Claim Two); and (3) both advertisements represented that Anacin reduces inflammation associated with the conditions specified in the advertisements. Id. (Claim Three).9 The district court then concluded that the preponderance of the evidence indicated that Claims One and Two — that Anacin is a superior analgesic in general to Tylenol and a superior analgesic for conditions which have an inflammatory component — were false. Id. at 801-03. The court further held that it could not be determined on the basis of the evidence presented whether OTC dosages of Anacin reduce inflammation to a clinically significant extent in the conditions specified by the advertisements. Thus, Judge Stewart could not reach a definitive conclusion on the truth or falsity of the third claim. Id. at 801, 803. Nevertheless, he determined that because the three claims are “integral and inseparable,” id. at 803, “the advertisements as a whole make false representations for Anacin and falsely disparages [sic] Tylenol in violation of the Lanham Act.” Id. at 803. Accordingly, he held that McNeil was entitled to an injunction against AHP, given the “substantial evidence that consumers have been and will continue to be deceived as to the relative efficacy of the two products and that this deception is injuring, and will continue to injure, Tylenol’s reputation among consumers.” Id. The injunction prohibits AHP from publishing or inducing television, radio or print media to publish
any advertisement or promotional material which contains, in the context of a representation as to any anti-inflammatory property of Anacin or aspirin sold by AHP, any representation that at over-the-counter levels Anacin or aspirin provides superior analgesia to acetaminophen including Tylenol either (1) generally, or (2) for conditions which are associated with inflammation or have inflammatory components, or (3) because Anacin or aspirin reduces inflammation. .
II. THE PRINCIPAL APPEAL
A. Whether Relief May be Afforded under Section 43(a) on the Basis of the District Court’s Findings.
AHP’s first contention is based on the following premises. The advertisements contain no express claim for greater analgesia; they merely assert Anacin’s superiority to Tylenol in reducing inflammation, a claim which the district court found not to be false. AHP further assumes that the anti-inflammatory claim was held to be true and is in fact unambiguous. The argument is that a truthful and unambiguous product claim cannot be barred under Section 43(a) even though consumers mistakenly perceive a different and incorrect meaning. Thus, appellants urge, the court erred in finding a violation of Section 43(a) by not relying on express claims of superior pain relief, but by interpreting consumer reaction tests — one of which, incidentally, was introduced by AHP and the other of which was not objected to by AHP — as indicating that consumers derive a message of greater pain relief from the explicit “truthful” claim.
Whatever abstract validity AHP’s argument may have, an issue we need not decide, it is clear that in this case the language of the advertisements is not unambiguous. The “truthfulness” of the claims, therefore, cannot be established until the ambiguity is resolved. Moreover, Judge Stewart expressly refused to characterize the anti-inflammatory claims as truthful.10 *165But even assuming the literal truthfulness of the anti-inflammatory claims, appellant’s position is no stronger because of the ambiguity of the total message which, as the district court found, conveys additional claims (Claims One and Two).11
That Section 43(a) of the Lanham Act encompasses more than literal falsehoods cannot be questioned. American Brands, Inc. v. R. J. Reynolds Tobacco Co., 413 F.Supp. 1352, 1357 (S.D.N.Y.1976); 1 R. Callmann, The Law of Unfair Competition, Trademarks and Monopolies §§ 19.1(a), at 631; 19.1(e), at 644; 19.2(a)(1), at 656; 19.-2(b)(1), at 665-67 (3d ed. 1967); cf. FTC v. Sterling Drug, Inc., 317 F.2d 669, 674-75 (2d Cir. 1963) (Federal Trade Commission suit). Were it otherwise, clever use of innuendo, indirect intimations, and ambiguous suggestions could shield the advertisement from scrutiny precisely when protection against such sophisticated deception is most needed. It is equally well established that the truth or falsity of the advertisement usually should be tested by the reactions of the public. 1 R. Callmann, The Law of Unfair Competition, Trademarks and Monopolies, supra, §§ 18.2(b), at 623; 19.2(a), at 655; 19.2(a)(1), at 656; 19.2(b)(1), at 665-69. Judge Lasker astutely articulated these legal principles in American Brands, Inc. v. R. J. Reynolds Co., supra, 413 F.Supp. at 1356-57 (emphasis added), a false advertising case involving ambiguous claims:
Deceptive advertising or merchandising statements may be judged in various ways. If a statement is actually false, relief can be granted on the court’s own findings without reference to the reaction of the buyer or consumer of the product.
The subject matter here is different. We are dealing not with statements which are literally or grammatically untrue .....Rather, we are asked to determine whether a statement acknowledged to be literally true and grammatically correct nevertheless has a tendency to mislead, confuse or deceive. As to such a proposition “the public’s reaction to [the] advertisement will be the starting point in any discussion of the likelihood of deception. . . . If an advertisement is designed to impress . customers, . . . the reaction of [that] group[s] [sic] will be determinative.” 1 Callmann: Unfair Competition, Trademarks & Monopolies at 19.2(a)(1) (3rd ed. 1967). A court may, of course, construe and parse the language of the advertisement. It may have personal reactions as to the defensibility or indefen-sibility of the deliberately manipulated words. It may conclude that the language is far from candid and would never pass muster under tests otherwise applied — for example, the Securities Acts’ injunction that “thou shalt disclose”; but *166the court’s reaction is at best not determinative and at worst irrelevant. The question in such cases is — what does the person to whom the advertisement is addressed find to be the message?
Applying these principles to the facts of this case, we are convinced that the district court’s use of consumer response data was proper. We believe that the claims of both the television commercial and the print advertisement are ambiguous. This obscurity is produced by several references to “pain” and body sensation accompanying the assertions that Anacin reduces inflammation.12 A reader of or listener to these advertisements could reasonably infer that Anacin is superior to Tylenol in reducing pain generally (Claim One) and in reducing certain kinds of pain (Claim Two). Given this rather obvious ambiguity, Judge Stewart was warranted in examining, and may have been compelled to examine, consumer data to determine first the messages conveyed13 in order to determine ultimately the truth or falsity of the messages.
It was also proper to construe Section 43(a) to prohibit the representations, assuming for the moment the correctness of Judge Stewart’s findings on consumer reaction, see Part II, B, 1, infra, and on falsity. See Part II, B, 2, infra. Contrary to appellant’s contentions, Judge Stewart did not use Section 43(a) to prohibit truthful representations mistakenly construed by consumers. See notes 10-11 supra. We have here deliberate ambiguity and unsubstantiated anti-inflammatory claims determined by the district court to be understood by the public as proclaiming superior analgesic results.14 Given the audience reaction to the *167advertisements, as found by the district court, the statute’s proscription of “words or other symbols tending falsely to describe or represent [the goods]” was violated.
B. Whether the Findings Are Supported by the Evidence.
1. The findings that the advertising makes claims of greater pain relief generally and for specific conditions.
AHP argues that even if Judge Stewart was correct in evaluating consumer perceptions, he arrived at the conclusion that the advertisements claim greater pain relief (Claims One and Two) by misinterpreting the consumer reaction data and erroneously affording no weight to the testimony of market research experts concerning the messages conveyed by the advertising. Initially, we note that the district court did take into account the expert testimony offered by both sides on the meaning of the advertisements. Recognizing that in some circumstances such testimony should be given substantial weight, Judge Stewart refrained from doing so here largely because it was neither reliable nor helpful to an understanding of the test data which it purported to interpret. The testimony, however, was used to corroborate the surveys.
The test data itself has some inherent weaknesses.15 However, it was in the district court’s province as trier of fact to weigh the evidence, and in particular the opinion research. See Grotrian, Helfferich, Schulz, Th. Steinweg Nachf. v. Steinway & Sons, 523 F.2d 1331, 1341 (2d Cir. 1975) (trademark infringement case). Additionally, the district court’s findings, based not only on its analysis of the surveys but also on the corroborating testimony of the market research experts, are entitled to the usual clearly erroneous standard of review under Fed.R.Civ.P. 52(a). See, e. g.; Grotri-an, Helfferich, Schulz, Th. Steinweg Nachf. v. Steinway & Sons, supra, 523 F.2d at 1339 n. 16. And our review of the record reveals that the findings on consumer interpretation of the advertising are not erroneous.
The analysis of its test data by ASI Market Research, Inc. (ASI) which was made for and at the request of AHP and was heavily relied on by Judge' Stewart, completely supports his finding that the television commercial made a superiority claim which was not limited to pain associated with inflammation. It states that the commercial
produced a reasonable level of “competitive superiority” type of recall. It is fairly clear from the qualitative group discussions that the side by side comparison to other named brands was being translated into a general “better than” rather than a specific “better for inflammation.”
It also reveals: “[H]eavy symptom-relief response was occurring, predominantly in the area of relief of ‘pain.’ Relief of headaches was secondary, somewhat more visible for ‘Muscle’ [another commercial] than for [this commercial].” Translated from survey jargon, the ASI survey concludes that the “Your Body Knows” television commercial produced a recollection in the selected audience of 250 members that Anacin is a superior pain reliever generally, even though the advertisement may have *168been phrased in terms of comparing inflammation relief.16
AHP’s advertising executives may, indeed, have intended to communicate that Anacin is better for relieving pain. They noted that 74% of the specifically selected audience thought of the commercial in terms of pain relief, while none thought of inflammation as such unless attention was directed specifically to inflammation. Nevertheless, they pointed out that because the references to inflammation triggered pain association, it was not necessary that inflammation relief be recalled.17
What the ASI test shows, then, is the powerful “subliminal” influence of modern. advertisements. See generally M. McLuhan & Q. Fiore, The Medium Is the Massage (1967). The survey reveals that the word “inflammation” triggers pain association, and pain association is what both advertisements are all about. The district court properly relied on these conclusions in finding that the commercial claimed general analgesic superiority.18
Judge Stewart placed less reliance on the Gallup & Robinson, Inc. (G&R) statistical analysis of the television commercial, largely because the breakdown of responses did not differentiate anti-inflammation messages from general pain relief messages. Instead, he reviewed the G&R verbatim interviews himself and found that they “revealed that the symptom relief reported was general pain relief with relatively few references to the specific conditions listed in the ad or to the reduction of inflammation.” 436 F.Supp. at 793. AHP objects to the use of this data, asserting principally its unreliability. See note 15 sqpra. The district court, however, was well aware of this survey’s limitations, as indicated by the detailed discussion of the test’s inadequacies. 436 F.Supp. at 793-94. Judge Stewart therefore gave little weight to the conclusions he drew from the responses because he was “not confident that the G&R technique really tested what message the consumer took from the language used as opposed to the broader issue of what mental processes viewers went through when they noticed an Anacin ad appear on the screen.” Id. at 794. In our view, appellant’s complaint is groundless.
Appellant attacks the district court’s conclusion on the message derived from the print advertisement — “that Anacin was better for relieving pain for some conditions because it reduces inflammation,” id. at 795 —on two grounds. First, AHP argues that Judge Stewart simply misinterpreted the G&R verbatim responses. Appellant contends that while 50% may have reported the message that Anacin reduces inflammation, *169only about 10% stated that Anacin provides superior analgesia. Moreover, the argument continues, the district court’s conclusion that 75% reported superiority in relieving pain for some conditions cannot logically be drawn from the district court’s findings that 50% reported the message of reduced inflammation and that 25% reported the message of Anacin’s analgesic superiority for certain listed conditions. We think that Judge Stewart’s conclusion is fully supported by the responses. The percentage computations may not be precise, but the majority of those who recalled any message reported, either explicitly or by implication, Anacin’s superiority as a pain reliever.
Second, AHP urges that since the television commercial and printed advertisement say the same thing, the district court’s holding that the printed advertisement made a qualified claim of superior pain relief must be wrong. We refuse to engage in this sort of speculation, given Judge Stewart’s careful scrutiny of the consumer surveys. Moreover, the differences in language and in medium of expression likely account for the slightly diverse consumer responses.
We conclude, therefore, that the judge correctly held, principally on the basis of consumer reaction tests, that both advertisements claim greater pain relief. On the assumption that this claim is false, see Part II, B, 2, infra, as the district court held, it was perfectly proper for the trial court to enjoin future advertisements containing superior analgesic claims.
2. The findings of falsity and inconclusiveness.
Appellant apparently does not attack the district court’s finding that Claim One (Anacin is a superior pain reliever in general to Tylenol, see ante at 164) is false. It does dispute findings on Claims Two and Three. See id. It first argues: “Even assuming that AHP’s advertising claims greater pain relief for inflammatory conditions [Claim Two, we believe], the advertising is truthful. . . . ” Brief for Appellant at 17; see id. at 46. Second, appellant asserts: “The District Court found that . . . the anti-inflammatory claim for the conditions listed in AHP’s advertising [Claim Three] is not false. However, the District Court erred in failing to find expressly that Anacin does reduce inflammation for the listed conditions.” Id. at 45.
We conclude that Judge Stewart’s determination that ASA and APAP are equipo-tent as pain relievers for inflammatory conditions is not clearly erroneous. It was based on his careful consideration of medical studies, medical literature and expert testimony. The studies, which we have also examined, reveal that with one exception-in the case of rheumatoid arthritis19 — there is little to indicate that in OTC dosages ASA is more effective in reducing pain from inflammatory conditions20 than APAP.21
*170The district court recognized that the medical experts’ testimony as well as the scientifically unsupported opinions of certain members of the medical community were more conflicting. Judge Stewart was unwilling to give much weight to AHP’s experts, who hypothesized that Anacin reduces pain associated with inflammation better than Tylenol, because much of this evidence admittedly consisted of mere speculation. In sum, Judge Stewart found most of the studies on the comparative analgesic effectiveness of ASA and APAP more reliable and convincing. Given the conflicting testimony of each party’s experts and the sound scientific basis underlying the studies, we cannot say that Judge Stewart erred in his evaluation and weighing of the evidence.
We also conclude that on the basis of the evidence introduced at trial, the judge was justified in holding that he could not determine whether Anacin, at OTC dosages, reduces inflammation in the conditions listed in the advertisements.22 We think it unnecessary to detail the conflicting evidence presented on each of the conditions. For determining the validity of the district court’s holding of lack of substantiation, it is sufficient to note that for each condition mentioned in the advertisements there was credible evidence suggesting that Anacin does not reduce inflammation.23
*171There was testimony that aspirin is not used to control inflammation following tooth extraction because of its well-documented propensity to increase post-operative hemorrhaging. With respect to neuritis, perhaps an out-moded term, there was evidence that all but two neuropathies are noninflammatory in nature, and that the only two which are inherently inflammatory — leprosy and shingles — are not treated by aspirin. The evidence regarding sinusitis revealed that while aspirin can make the patient more comfortable by reducing fever and relieving headache and muscle ache, it has no effect whatsoever on the underlying sinus infection. There was evidence that aspirin is not used for reduction of inflammation from sprains and strains; indeed it can be counterproductive by prolonging and increasing hemorrhaging at the site of the injury. There was basic disagreement on whether tendonitis, to be distinguished from tenosynovitis, is a degenerative or an inherently inflammatory disease. Thus, it could not be determined whether aspirin at OTC dosages provides any therapeutic effect on tendonitis other than pain relief. Backaches have so many different causes, so many of which are not inherently inflammatory in origin, that the advertising simply cannot be accepted as true.
Moreover, Judge Stewart painstakingly considered the evidence most favorable to appellant’s position — studies on aspirin’s effectiveness in reducing inflammation from rheumatoid arthritis. However, he concluded that these studies did not sufficiently resolve the question whether at OTC levels Anacin reduces inflammation in the specified conditions,24 particularly in light of the conflicting medical literature and expert testimony on the subject.
Our discussion of the evidence relied on by the district court is an overall review of the thousands of pages of transcripts and documentary evidence. Nevertheless, we are confident that Judge Stewart’s findings, based on his thorough review of the evidence, are not “clearly erroneous.”
C. Whether the Injunction is Sufficiently Specific.
AHP argues that the injunctive order, see ante at 164, 167, does not comply with Fed.R.Civ.P. 65(d) which requires specificity in terms and description in reasonable detail of the acts sought to be restrained.25 In particular, appellant asserts that the injunction as framed effectively bars AHP from discussing Anacin’s anti-inflammatory benefits in a comparative advertisement because such claims may also contain an implicit representation of superi- or analgesia. Apparently, AHP objects to the fact that the injunction is not limited to express claims for superior analgesia.
We think the district court did a commendable job of drafting an order which specifies, as clearly as possible under the circumstances, the acts to be enjoined. Representations as to any anti-inflammatory properties of Anacin are enjoined only to the extent that they contain or imply three clearly described claims of analgesic superiority at OTC levels. In our view, Rule 65(d) is satisfied. More explicit language, if this is possible, will not diminish AHP’s asserted difficulty in determining whether proposed advertising conveys a message of superior pain relief. We note, moreover, that the district court has retained jurisdiction to enable the parties “to apply to this court at any time for such other orders and directions as may be necessary or appropriate for the modification, construction or carrying out of [the] judgment.” American Home Products Corp. v. Johnson & Johnson, supra, 436 F.Supp. 785. If AHP encounters difficulties under the decree, it can apply to the court for guidance at such time.
*172III. THE CROSS-APPEAL
The district court rejected McNeil’s contentions that the printed advertisement contains claims of Anacin’s faster onset of analgesia and harmlessness to the stomach. See note 9 supra. We have some difficulty with this conclusion because the advertisement suggests to us that Anacin produces analgesia faster than Tylenol and that Anacin is gentle to the stomach.26 But we agree with Judge Lasker’s statement in American Brands, Inc. v. R. J. Reynolds Tobacco Co., supra, 413 F.Supp. at 1357, that “the court’s reaction [to language of an advertisement] is at best not determinative and at worst irrelevant.”27 A court should turn to consumers’ reactions to determine whether the advertising in question makes false representations,28 see id.; 1 R. Callmann, The Law of Unfair Competition, Trademarks and Monopolies, supra, §§ 18.-2(b), at 623; 19.2(a), at 655; 19.2(a)(1), at 656; 19.2(b)(1), at 665-69, as Judge Stewart did. While the evidence could have supported a contrary result, we cannot hold that his findings of insufficient evidence to sustain either of McNeil’s claims are clearly erroneous. The consumer surveys did not sufficiently buttress the contention that the literally true statement of the print advertisement, “[millions take Anacin with no stomach upset,” produces the message that Anacin causes no harm to the stomach. Only three out of the forty-three women questioned in the G&R 'survey reported such a claim. This surely was a proper basis for the district court’s finding that this number is too small to support, by a preponderance of the evidence, the existence of the representation.
The surveys were contradictory on whether the commercials convey a claim of faster relief. The G&R print advertisement test showed '49% reporting “fast acting.” Judge Stewart’s review of the verbatims revealed that most of the 49%, about one-third of the test audience, reported a “faster” claim. The ASI data, however, indicated that only 1% of the television viewers reported this message. Judge Stewart rejected the G&R results because he considered the ASI data more reliable,29 it “so strongly undercut” the G&R data, and the expert testimony supporting the McNeil position simply did not take the ASI study sufficiently into account. 436 F.Supp. at 794-95. We are persuaded by the district court’s clearly articulated reasoning that its weighing of the evidence was reasonable.
The judgment is affirmed.
. Note, The Law of Comparative Advertising: How Much Worse is “Better” Than “Great, ” 76 Colum.L.Rev. 80, 80 (1976).
. AHP also makes a First Amendment claim which was not raised below. Accordingly, we will not consider it on appeal. See, e. g., Jhirad v. Ferrandina, 536 F.2d 478, 486 (2d Cir.) (denial of petition for rehearing), cert. denied, 429 U.S. 833, 97 S.Ct. 97 (1976); Wilkerson v. Meskill, 501 F.2d 297, 298 (2d Cir. 1974) (per curiam).
. The “story board” of the television commercial is printed in Judge Stewart’s opinion below, 436 F.Supp. 785, 788 (S.D.N.Y.1977). The script for the commercial reads as follows:
SPOKESMAN: Your body knows the difference between these pain relievers [showing other products] and Adult Strength Anacin. For pain other than headache Anacin reduces the inflammation that often comes with pain. These do not. (SFX: MUTED KETTLE DRUM.) Specifically, inflammation of tooth extraction!,] muscle strain (SFX BUILDS) [,] backache (SFX BUILDS) [,] or if your doctor diagnoses tendonitis [,] neuritis. (SFX FADES.) Anacin reduces that inflammation (SFX OUT) as Anacin relieves pain fast. These do not. Take Adult Strength Anacin.
. The print advertisement is reproduced in Judge Stewart’s opinion, 436 F.Supp. at 789. It reads in substantial part as follows:
Anacin can reduce inflammation that comes with most pain. Tylenol cannot.
With any of these pains, your body knows the difference between the pain reliever in Adult-Strength Anacin and other pain relievers like Tylenol. Anacin can reduce the inflammation that often comes with these pains.
Tylenol cannot. Even Extra-Strength Tylenol cannot. And Anacin relieves pain fast as it reduces inflammation.
Get fast relief. Take Adult-Strength Anacin. Millions take Anacin with no stomach upset. Anacin.
“These pains” referred to in the text are depicted as located on the human body by spots and include sinusitis, tooth extraction, neuritis, tendonitis, muscular backache, muscle strain, and sprains.
. At least two of the television networks have their own guidelines for comparative advertisements. See Note, The Law of Comparative Advertising, supra, note 1, 76 Colum.L.Rev. at 81 n. 7.
. Before ABC would broadcast the commercials it required AHP to change certain language used in the CBS and NBC commercials. “Anacin relieves both pain and its inflammation fast. These do nothing for inflammation” was substituted for “Anacin reduces that inflammation as Anacin relieves pain fast. These do not.”
. Section 43(a) provides in pertinent part:
Any person who shall . . . use in connection with goods . . . any false description or representation, including words or other symbols tending falsely to describe or represent the same, and shall cause such goods ... to enter into commerce, . - . shall be liable to a civil action . by any person who believes that he is or is likely to be damaged by the use of any such false description or representation.
15 U.S.C. § 1125(a).
. See notes 3-4 supra.
. The judge also held that the advertising does not represent that Anacin provides faster analgesic action or that Anacin is harmless to the stomach, again on the basis of consumer reaction. American Home Prods. Corp. v. Johnson & Johnson, 436 F.Supp. 785, 796 (S.D.N.Y. 1977). See Part III infra. Accordingly, no findings on the truth or falsity of these alleged representations were made.
. The flaw in appellant’s logic is its conclusion of truthfulness from the district court’s candid recognition that it could not determine whether *165the anti-inflammatory claim was true or false. Judge Stewart did not find this claim truthful:
Accordingly, we find that there is no reliable evidence showing that ASA reduces inflammation to a clinically significant extent in the conditions listed in the advertisements at OTC dosages. On the other hand, we think that McNeil has not proved by a preponderance of the evidence that this claim is, or tends to be, false.
436 F.Supp. at 801.
. Appellants apparently solely refer to the claims of superior anti-inflammatory effect as the “truthful and unambiguous product claim.” We have already pointed out, note 10 supra, the fallacy of interpreting the district court’s finding of unsubstantiation as a finding of truthfulness. Additionally, appellants err in their failure to recognize that the advertisement must be viewed in its entirety. As stated by Judge Kaufman in FTC v. Sterling Drug, Inc., 317 F.2d 669, 674 (2d Cir. 1963):
It is therefore necessary ... to consider the advertisement in its entirety and not to engage in disputatious dissection. The entire mosaic should be viewed rather than each tile separately. “(T]he buying public does not ordinarily carefully study or weigh each word in an advertisement. The ultimate impression upon the mind of the reader arises from the sum total of not only what is said but also of all that is reasonably implied.” Aronberg v. Federal Trade Commission, 132 F.2d 165, 167 (7th Cir. 1942).
By paraphrasing the language of the advertisements simply to state that Anacin reduces inflammation while Tylenol cannot, they overlook the obvious ambiguous nature of the advertisements. See note 12 & accompanying text infra.
. The television commercial initially states that “[y]our body knows the difference between these pain relievers . . . and Adult Strength Anacin,” thereby suggesting that the individual is going to feel the difference in pain reduction from use of Anacin. The reference to superior analgesic properties without mention of inflammation ‘is quite apparent. The commercial then goes on to say that “[fjor pain other than headache Anacin reduces the inflammation that often comes with pain,” while the competition does not. This statement is ambiguous because it does not state that “for conditions other than headache, Anacin reduces inflammation.” It says “for pain other than headache,” implying that by reducing the inflammation that often comes with pain, the pain is itself reduced. Again, after listing specific ostensibly inflammatory conditions, the television commercial states that “Anacin reduces that inflammation as Anacin relieves pain . . . fast,” while the competition does not. Meanwhile, pulsating spots of inflammation of a male body are eliminated as Anacin works by “relievfing] pain.” Quite clearly the television commercial is purposely ambiguous. Quite clearly a claim for superior analgesic effect is intended.
The printed advertisement is similar. The lead statement that “Anacin can reduce the inflammation that comes with most pain, Tylenol cannot” implies superior reduction of pain. The smaller print states .that “[w]ith any of these pains [rather than any of these conditions], your body knows the difference between the pain reliever in Adult-Strength Anacin and other pain relievers like Tylenol.” In other words, your body can tell the difference between pain relievers, again strongly suggesting that Adult Strength Anacin provides greater pain relief than Tylenol. After comparing Anacin with Tylenol, the advertisement continues: “And Anacin relieves pain fast as it reduces inflammation,” implying that Anacin is superi- or as a pain reliever. At the very least, there is sufficient ambiguity in the advertising in question to justify the district court’s examination of consumer data.
. AHP concedes that consumer data may be used to determine whether a particular representation has been made, that is, to resolve ambiguity. Brief for Appellant at 22-23. Appellant simply disagrees on whether the advertising statement is unambiguous and truthful.
. AHP relies upon Alfred Dunhill, Ltd. v. Interstate Cigar Co., 499 F.2d 232 (2d Cir. 1974), in which we refused to extend Section 43(a) to preclude sales of damaged goods bearing the Dunhiil brand name, notwithstanding the possibility that consumers would receive the false impression that the goods sold were Dunhill’s usual high-quality product. Dunhiil, however, involved a failure affirmatively to disclose material facts which would have been necessary to prevent purchaser deception. There was no false representation in the sense contemplated by the Lanham Act. The goods sold were Dun-hill goods. Here, however, by innuendo if not by direct statement, Anacin is represented as being a superior general analgesic. Appellant’s misplaced reliance on the Dunhiil case, analogizing the truthfulness of the DunhilLname to the “truthfulness” of the anti-inflammatory claims, is caused by its incorrect characterization of the anti-inflammatory claims as truthful.
. Questions of reliability are raised because the two television commercial tests upon which the district court relied focused, to a considerable extent, on audience recall rather than on immediate impressions. Delayed recall measures consumer interest and advertising persuasiveness as well as message content. The Gallup and Robinson, Inc. (G&R) test was conducted approximately 24 hours after the commercial was aired by the use of telephone interviews with persons who claimed to have watched the program accompanying the commercial. The ASI Market Research, Inc. (ASI) testing was performed at special screenings for a specially selected audience. The viewers were questioned about their reactions both during and one hour after the screening. Some of the people tested by both surveys either had bad memories or paid little attention to the television commercial, resulting in inaccurate descriptions, not only of the claims made but even of the products discussed. But such inaccuracies, we suppose, are to be expected in advertising research.
. AHP contends that because there was insufficient evidence on the methods of coding the consumer responses, the district court could not have interpreted the data meaningfully. More particularly, appellant argues that some of the percentage figures espoused by the court are inaccurate in that non-exclusive categories were treated as exclusive, thereby increasing erroneously the percentage of people who mentioned pain relief generally. McNeil argues that the interpretation of the ASI figures was accurate in all respects. We find it unnecessary, however, to delve into the murky area of statistics evaluation. Both the AHP marketing executives’ evaluation of the commercial, see note 17 & accompanying text infra, and the ASI analysis of its own data conclude that the principal message received by consumers was general analgesic superiority.
. The internal memorandum states:
“Body” was somewhat stronger [than another Anacin commercial] on pain mention (74% of those playing back pain vs. 57% for [the other commercial]) as a percentage of total symptom relief.
It continues:
In the “Body” commercial, inflammation, per se, was not played back in copy point recall and had to be probed for in the focus group session. It may well be that recall and/or complete understanding of inñammation is not essential since 74% of those recalling symptom relief in “Body” mentioned “pain” and 20% mentioned “headache pain” despite the phrase “For pain other than headache” preceding the list of inflammatory symptoms within that commercial. In other words, infíammation is a word that triggers pain association.
(Emphasis added.) The subliminal influence is present even though — incredibly enough — the survey showed 46% of the people tested found something in the commercial “hard to believe.”
. Moreover, the direct testimony of the AHP expert on consumer reaction data, aptly named Dr. Payne, supports Judge Stewart’s finding that a high percentage of viewers reported a message of competitive superiority.
. Two of many studies on rheumatoid arthritis found ASA superior to APAP in pain reduction. With regard to the 1974 test, which found ASA slightly more effective than APAP in reducing pain from rheumatoid arthritis, it repeatedly points out the lack of statistical significance of the subjects’ responses. The district court was entitled to rely on this caveat, see 436 F.Supp. at 802, in discounting the weight to be given to the study. Moreover, it appears from our review of the study that ASA and APAP were never directly compared in the same trial group, another factor undermining the test’s reliability.
Judge Stewart found the second study, nearly 20 years old, to be methodologically unsound in comparison to the numerous other scientific reports introduced. Additionally, the study does not seem very helpful in resolving the issues below because it measured pain relief by noting differences in grip strength and joint stiffness. We have difficulty, therefore, understanding whether and when its conclusions are directed to pain relief (relevant to Claim Two) or to inflammation relief (relevant to Claim Three).
. The conditions specifically examined include postpartum pain, postoperative pain, tonsillectomy and oral surgery.
. AHP heavily relies on one statement from a report of the Advisory Review Panel on OTC Internal Analgesic and Antirheumatic Products *170[hereinafter IAP Report], appointed by the Food and Drug Administration, which states:
The Panel concludes that acetaminophen is effective in relieving the pain of headache, and that it is a general analgesic of proven efficacy as shown by clinical testing. Thus, acetaminophen is considered to be equivalent to aspirin in its analgesic effects, although the lack of anti-inflammatory action might make it less useful in-conditions having an inflammatory component.
42 Fed.Reg.No.131, Book 2, at 35413 (1977) (emphasis added). We do not disagree with Judge Stewart’s characterization of the inference to be drawn from the last clause in this statement as “reputable . . . speculations.” 436 F.Supp. at 803.
. As far as we can tell, even if we were to disagree with the district court’s conclusion, the outcome of this case would not be altered because the district court’s holding that the advertisements carry the messages that Anacin provides greater pain relief in general and in the specified conditions associated with inflammation would still stand. Appellant seems to concede as much in its Reply Brief at 33 n. 1. Judge Stewart found that even though it could not be determined whether or not OTC dosages of Anacin could or did reduce inflammation for the specified conditions, to the extent that the advertisements claim superior analgesia for the specified inflammatory conditions, they are false. He further determined that the claim that Anacin reduces inflammation and Tylenol does not is so integrally connected to the two false analgesic representations — obviously because anti-inflammation claims trigger pain claims, see notes 16-18 & accompanying text supra —that the false claims cannot be avoided even though they may stem from one of unde-terminable truth. Declaring the anti-inflammation claims to be truthful would affect neither the existence nor the falsity of the two analgesic claims which consumers report from the ambiguous commercials. Accordingly, enjoining any advertisement “which contains, in the context of a representation as to any anti-inflammatory property of ANACIN,” see ante at 164, the representation that Anacin is a superior analgesic generally or a superior analgesic for conditions which are associated with inflammation or have inflammatory components, would in all probability be proper. See Part II, A, supra.
. At larger than OTC levels, it is less disputed that ASA may reduce inflammation accompanying some of the stated conditions. But the anti-inflammatory claims are based, as they must be, solely on OTC dosages. Otherwise, the advertising would invite the consumer to guess the appropriate levels of intake, a dangerous and potentially fatal practice. See generally Morgenstern Chemical Co. v. G. D. Searle & Co., 253 F.2d 390, 393-94 (3d Cir.) (possibility of confusion between medical products’ similar trademarks warrants injunction where health and safety factors are involved), cert. denied, 358 U.S. 816, 79 S.Ct. 25, 3 L.Ed.2d 58 (1958); Syntex Laboratories, Inc. v. Norwich Pharmacal Co., 315 F.Supp. 45, 53 (S.D.N.Y.1970) (importance of public health and potential physical harm to consumers emphasized in trademark confusion case), aff’d, 437 F.2d 566 (2d Cir. 1971).
The IAP Report emphasizes the inefficiency and danger of self-medication where anti-inflammatory therapy is desired. See, e. g., 42 Fed.Reg.No.131, Book 2, at 35381-82; 35459; 35355.
Judge Stewart expressed similar concerns. 436 F.Supp. at 798.
. The district court opinion, 436 F.Supp. at 799-800, indicates very careful consideration given to the studies.
. The purposes of Fed.R.Civ.P. 65(d) are discussed in Schmidt v. Lessard, 414 U.S. 473, 476-77, 94 S.Ct. 713, 38 L.Ed.2d 661 (1974) (per curiam).
. Judge Stewart did not address the issue of truth or falsity of such claims.
. In short, we do not have the same expertise as the Federal Trade Commission when it interprets the language of an advertisement to determine whether an advertisement is an unfair or deceptive practice in violation of 15 U.S.C. §§ 45 and 52. See, e. g., Resort Car Rental System, Inc. v. FTC, 518 F.2d 962, 964 (9th Cir.) (per curiam), cert. denied sub nom. MacKenzie v. United States, 423 U.S. 827, 96 S.Ct. 41, 46 L.Ed.2d 42 (1975). Accordingly, we, as judges, must rely more heavily on the reactions of consumers, as found by the finder of fact.
. The law of unfair trade practices has come a long way since the days of Triangle Publications, Inc. v. Rohrlich, 167 F.2d 969 (2d Cir. 1948), overruled in part in Monsanto Chem. Co. v. Perfect Fit Prods. Mfg. Co., 349 F.2d 389 (2d Cir. 1965), cert. denied, 383 U.S. 942, 86 S.Ct. 1195, 16 L.Ed.2d 206 (1966). Judge Frank, dissenting in Triangle Publications, took “judicial notice” of the views of adolescent girls, their mothers and sisters, persons he chose at random, on the trade-name association between “Seventeen” magazine and girdles sold under the name “Miss Seventeen.” Id. at 976. He also suggested the use of psychological testing to determine confusion. Id. at 976-77.
. See ante at 168-169; 436 F.Supp. at 794, 795.