Ives Laboratories, Inc. ("Ives") brought suit in the District Court for the Eastern District of New York against appellees, who are drug manufacturers and wholesalers, claiming that their manufacture and distribution of a generic drug using capsules identical in color, shape, and size to those long used by Ives in its equivalent trademarked product violated §§ 32 and 43(a) of the Lanham Act, 15 U.S.C. §§ 1114 and 1125(a), and New York State unfair competition laws.1 The district court, Nickerson, J., denied Ives' motion for a preliminary injunction, 455 F. Supp. 939. Upon appeal we affirmed, noting that "we find the case more difficult than did the district judge." 601 F.2d 631, 634 (2nd Cir.). On remand the district court, after a bench trial, found for the appellees.2 488 F. Supp. 394. We reverse on the ground that the undisputed evidence establishes a violation of § 32 and remand for further proceedings not inconsistent with this opinion.3 In view of our holding in favor of Ives on the § 32 issue we express no views on the other issues raised by it on this appeal.

Ives has for some years manufactured a prescription drug called CYCLOSPASMOL. CYCLOSPASMOL is a peripheral vasodilator designed to produce an increase in blood flow; it is most commonly prescribed for ingestion by elderly patients who are experiencing progressive circulatory difficulties. Until 1972 Ives held a patent on CYCLOSPASMOL. Since that time it has continued to hold a trademark to the name CYCLOSPASMOL, but several generic drug manufacturers (including some of the appellees) have begun selling bioequivalents of CYCLOSPASMOL under the generic name "cyclandelate."

Ever since it began manufacturing CYCLOSPASMOL, Ives has used distinctive, colorful gelatin capsules as containers for the CYCLOSPASMOL powder. The 200 mg. capsule is pale blue; the 400 mg. capsule is red and blue. When appellees Premo Pharmaceutical Laboratories, Inc. ("Premo"), Inwood Laboratories, Incorporated ("Inwood"), and MD Pharmaceutical Company, Inc. ("MD") (collectively "the defendant manufacturers") began marketing cyclandelate, they intentionally chose to use gelatin capsules which were identical in color, shape, and size to those used by Ives even though scores of other colors, color combinations, and sizes were available.4

Because CYCLOSPASMOL may only be dispensed by prescription, Ives has directed its CYCLOSPASMOL promotional efforts toward physicians, not patients. Some 230 "detail" men make periodic visits to physicians' offices on Ives' behalf, distributing samples and promotional material. Ives also places advertisements for CYCLOSPASMOL in professional journals read by physicians. The defendant manufacturers, on the other hand, have no similar network of detail men, and do not advertise routinely in professional magazines. They instead make their sales primarily through catalogs, which are distributed widely in the wholesale industry. Some of the catalogs contain charts which pair each generic product being offered with the name brand product for which it may be substituted and offer the generic product at a substantially lower price. The defendant wholesalers (Darby Drug Co., Inc. ("Darby"), Rugby Laboratories, Inc. ("Rugby"), and Sherry Pharmaceutical Co., Inc. ("Sherry")) purchase generic cyclandelate in capsules identical in appearance to those made by the defendant manufacturers and sell them at wholesale, primarily to physicians and pharmacies, at lower prices than those charged by Ives for its trademarked CYCLOSPASMOL.

Ives contends that appellees violated § 32 of the Lanham Act because their deliberate use of look-alike capsules and pro-substitution catalogs encouraged retail druggists to substitute generic cyclandelate illegally for CYCLOSPASMOL. In his first opinion, which denied plaintiff's motion for a preliminary injunction, Judge Nickerson reasoned that Ives' § 32 claim would only be viable if it could show "that defendants have conspired with the pharmacists or counseled or suggested that they disregard the doctors' orders."5 455 F.Supp. at 945. Finding that no such showing had been made, he rejected Ives' § 32 argument.

On appeal from this denial of preliminary relief we recognized that, although a classic case of retailer infringement occurs where a druggist illegally substitutes a generic drug for CYCLOSPASMOL prescribed by the doctor, a violation of § 32 for infringement also arises in the "intermediate" case when "the prescription permits substitution and the druggist fills it with defendant's (generic) product but names it Cyclospasmol." 601 F.2d at 636. We went on to question whether the district court's reading of § 32 had been too rigid:

"We fear ... that his (Judge Nickerson's) approach to the issue may have been unduly narrow. He referred to 'the knowing and deliberate instigation of such a practice' ... and said that before 'passage of the Lanham Act the doctrine of contributory trademark infringement went no further than to hold one who actively shared a retailer's infringement, as in the case, for example, of one who places the infringing label on the articles before delivering them to the retailer.' ... While such cases would surely constitute contributory infringement, they do not mark the limits of the doctrine. The authorities later reviewed indicate to us that a manufacturer or wholesaler would be liable under § 32 if he suggested, even if only by implication, that a retailer fill a bottle with the generic capsules and apply Ives' mark to the label, or continued to sell capsules containing the generic drug which facilitated this to a druggist whom he knew or had reason to know was engaging in the practices just described." 601 F.2d at 636 (emphasis supplied).

However, Judge Friendly concluded that on the record then before the court

"the court was justified in concluding that Ives failed to adduce the quantum of proof necessary to require issuance of a temporary injunction for contributory infringement even on what we deem the proper standard of liability under § 32. Ives' proof was limited to 15 instances of improper substitution and defendants conceded only that the identity of color made this easier if a druggist were so inclined." Id.

On remand the district court heard Ives' evidence on the § 32 claim, and concluded that the factual showing necessary to prove plaintiff's case had once again not been made:

"The court finds that defendants did not suggest, even by implication, that pharmacists fill bottles with the generic drug cyclandelate and apply the trademark Cyclospasmol. Nor did Ives show that defendants continued to sell the generic drug to pharmacists who defendants knew or had reason to know were engaging in such a practice." 488 F. Supp. at 397.

According to the district court, the new evidence introduced at trial on this point was limited to the following instances of mislabeling:

"Some Ives test shoppers gave a prescription for Cyclospasmol signed on the line over 'substitution permissible' to forty-two pharmacists in New York selected from a list supplied by Ives and forty-one from a list of drug stores in Hayes Directory. Of the forty-two on the Ives list twenty-four dispensed Cyclospasmol, and eighteen dispensed the generic product. Six of those eighteen included the word Cyclospasmol on the label. Of the forty-one on the Hayes Directory list twenty-four dispensed Cyclospasmol and seventeen dispensed the generic product. Four of those seventeen included the word Cyclospasmol on the label.

"Of the total of ten pharmacists who dispensed generic cyclandelate and also used the word Cyclospasmol on the label, four used the term 'generic Cyclospasmol' or 'Cyclospasmol generic' and one the term 'Cyclospasmol Gen.' Nine of the ten informed the person buying the capsules that the generic product had been dispensed. (6 Only one appears to have charged a higher 'brand' price. This evidence hardly justifies an inference that defendants' catalogs impliedly invited druggists to mislabel." Id.

In the district court's opinion, this frequent mislabeling of cyclandelate as CYCLOSPASMOL was occurring not because appellees had suggested that pharmacists do so, but rather because pharmacists "misunderstood the precise requirements of the New York Drug Substitution Law." Id. "(A)ny mislabeling stemmed from doubt as to the requirements of the generic drug law, not from inducement by defendants." Id. at 398.

Since the governing legal principles have already been set forth in Judge Friendly's opinion upon the earlier appeal, 601 F.2d 631 (2d Cir. 1979), we need not repeat them here. Applying those principles, we believe that the district court, in considering Ives' § 32 claim, erred upon remand in declining to give any weight to additional evidence offered by Ives indicating that appellees' use of identically-colored, look-alike drug capsules assisted druggists in labeling generic cyclandelate as Ives' trademarked product CYCLOSPASMOL, and that they could reasonably anticipate that the use of these capsules would have such an effect.

At the preliminary injunction hearing Ives introduced evidence of some 15 instances of illegal substitution by druggists of the generic drug cyclandelate for CYCLOSPASMOL after the latter had been prescribed, which we found insufficient to warrant reversal of the district court's denial of preliminary injunctive relief even though, taking into consideration the normal reluctance of any druggist to admit illegal conduct, it was surprising that Ives was able to produce as many such instances as it did. Recognizing this deficiency, Ives at trial introduced substantial additional evidence of "intermediate cases" of contributory infringement in which a druggist who was authorized to substitute the generic drug cyclandelate improperly used Ives' trademark CYCLOSPASMOL on the label.

In preparation for trial Ives carried out in a rigorous, impartial manner two series of drug shoppings at pharmacies. In each, the prescription reading "CYCLOSPASMOL Substitution Permissible" was used. The first series consisted of having the prescriptions filled at 42 pharmacies from a list of drug stores believed to carry both CYCLOSPASMOL and cyclandelate. The second involved the submission of the same prescription to 41 pharmacies selected at random from the Hayes Directory of Pharmacies in New York State. The shoppings revealed that in 29% of the cases pharmacists mislabeled the generic product in some form (6 out of 18 pharmacists in the first shopping and 4 out of 17 in the second shopping).7 In addition Ives adduced evidence that in 9 out of 17 instances where patients retained bottles of prescribed generic cyclandelate the bottles had been mislabeled with the trademark name CYCLOSPASMOL in some form.

This pattern of illegal substitution and mislabeling in New York is precisely the sort of showing which we held would be probative of plaintiff's § 32 claim, stating that "a manufacturer or wholesaler would be liable under § 32 if he suggested, even if only by implication, that a retailer fill a bottle with the generic capsules and apply Ives' mark to the label." 601 F.2d at 636.

The additional evidence introduced by Ives at trial was clearly sufficient to establish a § 32 violation. By using capsules of identical color, size and shape, together with a catalog describing their appearance and listing comparative prices of CYCLOSPASMOL and generic cyclandelate, appellees could reasonably anticipate that their generic drug product would by a substantial number of druggists be substituted illegally for Ives' trademarked CYCLOSPASMOL or that bottles of their lower-priced product might be mislabeled as CYCLOSPASMOL, all to the druggists' economic advantage. This amounted to a suggestion, at least by implication, that the druggists take advantage of the opportunity to engage in such misconduct. By using look-alike capsules, appellees also reduced the likelihood that patients who were given a generic substitute would bring that fact to their doctor's (or their druggist's) attention. Similar activity has been enjoined as enabling druggists to defraud by palming off, Upjohn Co. v. Schwartz, 246 F.2d 254 (2d Cir. 1957); Merrell-National Laboratories, Inc. v. Zenith Laboratories, Inc., 194 U.S.P.Q. 157, 159-60 (D.N.J.1977); E. R. Squibb & Sons, Inc. v. Premo Pharmaceutical Labs, Inc., 195 U.S.P.Q. 545 (S.D.N.Y.1977); Smith Kline & French Laboratories v. Broder, 125 U.S.P.Q. 299 (S.D.Tex.1959). We think appellees' activity should be enjoined as well.

The second premise of the district court's decision, that mislabeling occurs because of confusion rather than as a result of appellees' inducement, finds no support in the record. Indeed, it was the district court itself, in a short and casual exchange with a witness, which suggested that the mislabeling might result from pharmacists' thinking that CYCLOSPASMOL was a generic name rather than a trademark. The witness to whom he made the suggestion responded ambiguously, agreeing only that some pharmacists "may not be knowledgeable concerning the intricacies of the law." No evidence in the form of witnesses or surveys supporting the district court's suggestion was ever introduced at trial.

Against the absence of concrete evidence in the record supporting the district court's explanation for mislabeling we are now in possession of additional undisputed evidence, which we may judicially notice, suggesting that illegal substitution and mislabeling in New York are neither de minimis nor inadvertent. During the month of May, 1980, six indictments were handed down in the New York City area alleging illegal substitution and mislabeling of generic cyclandelate in prescriptions requiring CYCLOSPASMOL.8 These indictments, which reflect the grand jurors' views that reasonable grounds exist for charging the druggists with engagement in such illegal activities, is inconsistent with appellees' contention that mislabeling is rare or unintentional.

Appellees have not offered any persuasive evidence of a legitimate reason unrelated to CYCLOSPASMOL for using identically-colored, look-alike cyclandelate capsules instead of capsules of a different color, size or shape, which would minimize the practice of mislabeling, illegal substitution or confusion. Their arguments that look-alike capsules reduce patient confusion or anxiety in taking medication, enable doctors to communicate with patients more easily concerning their medication, and aid in their identification of the drug in an emergency situation, are unconvincing. Although a few doctors and one pharmacist were of the view that patients receiving capsules in a size, shape or color different from the medication they usually receive may at first believe that they have the wrong drug, there was no evidence of any instances of such confusion, much less that it was widespread, or that it could not easily be cured by explaining the change to the patient, either voluntarily in accordance with accepted professional practice or in response to the patient's inquiry. Indeed one witness called by appellees (Lawler) testified that when the doctor or druggist explains that the different color or different shaped capsules are the same generic drug, patients do not refuse to accept and use it. There was no evidence that doctors or druggists refused to furnish such explanations. No patient testified to confusion or anxiety. Nor did appellees offer any survey of patients' attitudes with respect to the matter. On the contrary, there was evidence that many generic drugs (e. g., cyclandelate made by W. E. Hauck and so-called "branded" generics) are successfully marketed in capsules or tablets that do not copy Ives' trademarked product. Thus the inference is inescapable that appellees' use of identically-colored, shaped and sized capsules was in anticipation that they would successfully capitalize upon public acceptance of CYCLOSPASMOL by reason of the fact that at least a substantial number of druggists would either substitute the generic product illegally or mislabel it as CYCLOSPASMOL.

Finally, we note that the current use of look-alike capsules makes detection of mislabeling very unlikely. The doctors who write prescriptions for CYCLOSPASMOL rarely see the capsules which their patients end up receiving at the pharmacy. The patients who have those prescriptions filled will normally be lulled by the look-alike cyclandelate capsules into thinking that they are receiving what their doctor prescribed.

The simplest way to reduce illegal substitution and mislabeling is to require that cyclandelate be sold in capsules which do not resemble those used for CYCLOSPASMOL. This would alert the patient to the possibility that he has been given cyclandelate when his doctor specifically prescribed CYCLOSPASMOL, as well as put him on notice that the lower generic cost of the cyclandelate should have been charged if that is in fact what he has received. Such a change could be made by appellees without any appreciable difficulty or expense. It would also give Ives the protection to which it is entitled for its trademarked product.

The judgment of the district court is reversed and the case remanded for further proceedings consistent with the foregoing.

I would affirm on the opinion of Judge Nickerson on all of the issues presented on this appeal.1 The majority here reverses solely on the ground that "undisputed evidence establishes a violation of § 32 of the Lanham Act." I respectfully disagree.

I have no dispute with the interpretation of § 32 of the Lanham Act which the majority purports to apply. Indeed it was set forth by Judge Friendly for a panel of this court of which I was a member, in Ives I, 601 F.2d 631 (1979). There is and can be no claim that Judge Nickerson misunderstood the law announced in our prior opinion.

With respect to the claimed § 32 Lanham Act violation, we found in Ives I, supra, that Ives had failed to adduce sufficient evidence of contributory infringement. "(T)he evidence of improper substitution was limited to 15 prescriptions. Discovery and trial might show that the practice was much more widespread or, on the other hand, as defendants suggest, that the pharmacies in question were known bad actors and few others engaged in such practices." Id. at 644. (The parties stipulated on trial that there were only 13 instances of such practices.)

Our inquiry then must focus upon whatever additional evidence was adduced by Ives on the trial. Ives submitted the results of some 83 test shoppings. Forty-two of these shoppings were conducted at pharmacies selected from a list prepared by Ives. One of Ives' own witnesses conceded that the selection of druggists by Ives' detail men for this list might have been tainted by knowledge of this litigation. While Ives would object to the characterization of these pharmacies as "bad actors," it would seem apparent that this list might be suspect and within the class of pharmacies to which we referred in our prior opinion. There is admittedly not a scintilla of evidence to suggest that the defendants sold to those pharmacies knowing or having reason to believe that illegal substitution or mislabeling had occurred. I therefore would conclude that under our prior opinion, the further evidence gleaned from such shoppings would hardly be probative of any violation by the defendants of Section 32. In any event the statement by the majority that Ives carried out in an "impartial manner" two series of drug shoppings is not entirely accurate.

Forty-one shoppings were impartially selected from a list contained in Hayes Druggist Directory, 1979 ed. Even taking the results of both series of shoppings, however, the results confirm Judge Nickerson's finding that there was no reasonable inference of an invitation by the defendants to mislabel. The testing hardly supports a finding that the practices condemned were "much more widespread" than those disclosed when this case first came to us in Ives I.

Only 10 out of the entire sample of 83 mislabeled the generic drug as Cyclospasmol. This amounts to some 12% of the sample and not the 29% estimated by the majority. More significantly, five of these ten included the word generic or its abbreviation before or after the word Cyclospasmol. Nine of the ten druggists advised the patient that they were dispensing the generic drug. Only one druggist out of 83 attempted an illegal substitution, charged a "higher" brand price and thus reaped any profit from a color confusion. These facts support rather than discredit the finding of the District Court that druggists were confused with respect to the labeling requirements of the new generic drug law and were not motivated by profit considerations. Any pharmacist who intended to mislabel would simply place the name Cyclospasmol on the label and charge a higher price. He would hardly put generic on the label and inform the customer that he was receiving a generic. In sum, the additional evidence adds nothing of substance to the record previously before this court.2

It is also significant that the defendants did not employ "detail" men who contacted druggists and thus were in a position to encourage substitution or mislabeling.3 This fact completely distinguishes this case from Upjohn Co. v. Schwartz, 246 F.2d 254 (2d Cir. 1957) which is relied upon in the majority opinion. That case involved the use of look alike products but there was also proof that the defendant's salesmen had contacted druggists for the very purpose of encouraging substitution. No such evidence exists here.

The sampling by Ives of druggists on the Sec. 32 issue was in any event inadequate. On the secondary meaning question (§ 43(a)) which involved a survey of doctors, a list of some 12,000 physicians was assembled and a random selection of names was made in twenty randomly selected geographical areas throughout the country. On the Section 32 contributory infringement issue, however, only 83 druggists located in the State of New York and only half randomly selected, were questioned. There are some 3,000 pharmacists in the State of New York alone. The record indicates that Cyclospasmol is prescribed to some 100,000 patients nationwide. A testing of some 83 druggists in one state does not present a sufficient basis after our remand for the injunctive relief sought here particularly in light of the results obtained.

In an effort to bolster what I believe is a weak record, the majority takes judicial notice of the fact that subsequent to the trial in this case six indictments have been handed down in the New York City area alleging illegal substitution and mislabeling of generic cyclandelate in prescriptions requiring Cyclospasmol. This we are told constitutes "additional undisputed evidence suggesting that illegal substitution and mislabeling in New York are neither de minimis nor inadvertent." Of course it is not disputed that the indictments have been filed but the record does not show any convictions. Six indictments are certainly not proof of any illegality and cannot reasonably be described as undisputed evidence of widespread or deliberate substitution and mislabeling even by the druggists much less by the defendant manufacturers here. If pharmacists are violating state laws which provide criminal sanctions, they should be punished. The criminal actions commenced if anything should discourage unethical pharmacists, and if convictions ensue even the de minimis violations adduced on this trial would be further reduced.

Admittedly there has been an imitation of color and capsule and the use of catalogues displaying both the generic and brand name drugs. There was, however, evidence that the catalogue juxtaposition of the drugs was a trade practice.4

The majority also contends that the appellees have not offered any "persuasive" evidence of a legitimate reason unrelated to Cyclospasmol for using identically-colored, look-alike capsules instead of those of a different color, size or shape. However, the trial court which heard the evidence found that the colors were functional in several respects. The court found that many elderly patients associate the appearance of their medication with its therapeutic effects. The court credited testimony that some elderly patients refuse to take equivalent drugs of a different color despite explanations of equivalence by their physicians while others do eventually accept it but experience confusion and anxiety. 488 F. Supp. at 389, 399. The majority does not characterize this finding or other evidence of functionality discussed in the opinion below5 as "clearly erroneous." This finding was based on the testimony of three physicians and the Chief Pharmacist of a hospital. There is no reason to require the direct evidence of such patients as the majority suggests. Judge Nickerson's findings of fact based upon the testimony of witnesses who appeared before him cannot be disregarded by this court. See, e. g., Master Shipping Agency, Inc. v. M. S. Farida, 571 F.2d 131 (2d Cir. 1978).

In sum I am persuaded that Ives has failed on trial to adduce any evidence of substance which would support a finding of a Section 32 violation. I would affirm.