COMBE INTERNATIONAL LLC v DR AUGUST WOLFF GMBH & CO KG ARZNEIMITTEL

Section 48(1) provides, so far as relevant:

“Where the proprietor of an earlier trade mark … has acquiesced for a continuous period of five years in the use of a registered trade mark in the United Kingdom, being aware of that use, there shall cease to be any entitlement on the basis of that earlier trade mark … -

(a)

to apply for a declaration that the registration of the later trade mark is invalid, or

(b)

to oppose the use of the later trade mark in relation to the goods or services in relation to which it has been so used,

unless the registration of the later trade mark was applied for in bad faith.”

While the UK was a Member State of the EU, section 48(1) implemented successively Article 9(1) of First Council Directive 89/104/EEC of 21 December 1988 to approximate the laws of the Member States relating to trade marks, Article 9(1) of European Parliament and Council Directive 2008/95/EC of 22 October 2008 to approximate the laws of the Member States relating to trade marks (codified version) and Articles 9(1) and 18(1) of European Parliament and Council Directive 2015/2436/EU of 16 December 2015 to approximate the laws of the Member States relating to trade marks (recast).

Parallel provisions were contained in Articles 51(1)(b) and (3) of Council Regulation 40/94/EC of 20 December 1993 on the Community trade mark, Article 54(1),(2) of Council Regulation 207/2009/EC of 26 February 2009 on the European Union trade mark trade (codified version) and Articles 16(1),(2) and 61(1),(2) of European Parliament and Council Regulation 2017/1001/EU of 14 June 2017 on the European Union trade mark (codification).

Both because the relevant period commenced while Directive 2008/95 was in force and because neither side suggested that the differences between the relevant provisions in that Directive and those in Directive 2015/2436 were material to the issues on the appeal, it suffices to refer to the former.

Recital (12) stated:

“It is important, for reasons of legal certainty and without inequitably prejudicing the interests of a proprietor of an earlier trade mark, to provide that the latter may no longer request a declaration of invalidity nor may he oppose the use of a trade mark subsequent to his own of which he has knowingly tolerated the use for a substantial length of time, unless the application for the subsequent trade mark was made in bad faith.”

Article 9(1) provided:

“Limitation in consequence of acquiescence

Where, in a Member State, the proprietor of an earlier trade mark as referred to in Article 4(2) has acquiesced, for a period of five successive years, in the use of a later trade mark registered in that Member State while being aware of such use, he shall no longer be entitled on the basis of the earlier trade mark either to apply for a declaration that the later trade mark is invalid or to oppose the use of the later trade mark in respect of the goods or services for which the later trade mark has been used, unless registration of the later trade mark was applied for in bad faith.”

The Court of Justice of the European Union has held that the purpose of Directive 89/104, and Article 9(1) in particular, is generally to strike a balance between the interest of the proprietor of a trade mark to safeguard its essential function (namely, indicating the origin of the goods and/or services for which the trade mark is registered), on the one hand, and the interests of other economic operators in having signs capable of denoting their goods and services, on the other: see Case C-145/05 Levi Strauss & Co v Casucci SpA [2006] ECR I-3703 at [29] and Case C-482/09 Budějovický Budvar np v Anheuser-Busch Inc[2011] ECR I-08701 at [34]. It follows that the protection that the proprietor receives under the Directive is not unconditional, since in order to maintain the balance between those interests that protection is limited in particular to those cases in which the proprietor shows itself to be sufficiently vigilant by opposing the use of signs by other operators likely to infringe its mark: see Levi Strauss at [30].

The Court went on in Levi Strauss at [34] to state that “inactivity” for the purposes of Article 12(2) of Directive 89/104 (revocation where the trade mark has become the common name for a product or service as a result of acts or inactivity of the proprietor), another provision to which it had referred in [29] and involving a concept closely akin to that of “acquiescence”:

“… may also take the form of a failure on the part of the proprietor of a mark to have recourse to Article 5 [Rights conferred by a trade mark] in due time, for the purposes of applying to the competent authority to prevent third parties from using the sign in respect of which there is a likelihood of confusion with that mark, since the purpose of such applications is precisely to preserve the distinctive character of the mark in question.”

In Budvar the Court of Justice held at [37] that “acquiescence” within the meaning of Article 9(1) of Directive 89/104 was a concept of EU law which was to be given an autonomous and uniform interpretation. It went on to hold at [44] that “the characteristic of a person who acquiesces is that he is passive and declines to take measures open to him to remedy a situation of which he is aware and which is not necessarily as he wishes … the concept of ‘acquiescence’ implies that the person who acquiesces remains inactive when faced with a situation which he would be in a position to oppose.” It therefore ruled that the proprietor of an earlier trade mark cannot be held to have acquiesced in the long and well-established honest use, of which he has long been aware, by a third party of a later trade mark identical with that of the proprietor if that proprietor was not in any position to oppose that use.

The Court also stated at [49] that the effect of “any administrative or court action initiated by the proprietor” within the prescribed period of five years is “to interrupt the period of limitation in consequence of acquiescence”. It is not clear either from the judgment or the opinion of Advocate General Trstenjak precisely what kind of administrative or court actions the Court of Justice had in mind.

The Court ruled in Budvar that there are four prerequisites for the running of the period of limitation prescribed in Article 9(1): first, registration of the later trade mark in the Member State concerned; secondly, the application for registration of that mark being made in good faith; thirdly, use of the later trade mark by its proprietor in the Member State where it has been registered; and fourthly, knowledge by the proprietor of the earlier trade mark that the later trade mark has been registered and used after its registration.

In Case C-561/11 Fédération Cynologique Internationale v Federación Canina Internacional de Perros de Pura Raza [EU:C:2013:91] the Court of Justice ruled that Article 9(1) of Regulation 207/2009 must be interpreted as meaning that the exclusive right of the proprietor of a Community (now an EU) trade mark to prohibit all third parties from using, in the course of trade, signs identical with or similar to its trade mark extends to a third-party proprietor of a later registered Community (now EU) trade mark, without the need for that latter mark to have been declared invalid beforehand. In that context, the Court stated at [35] that it was apparent from the wording of Article 54 of Regulation 207/2009 that “before the limitation in consequence of acquiescence takes effect, the proprietor of a Community trade mark is entitled both to apply to OHIM [now EUIPO] for a declaration of invalidity of the later Community [now EU] trade mark and to oppose its use through infringement proceedings before a Community [now EU] trade mark court”.

There is no dispute that the reasoning of the Court of Justice in Levi Strauss, Budvar and Fédération Cynologique is equally applicable to Directive 2008/95. This case law is retained EU law, and neither side suggested that we should depart from it.

Since the judge’s judgment, the Court of Justice has given judgment in Case C-466/20 Heitec AG v Heitech Promotion GmbH [EU:C:2022:400] on a reference from the Bundesgerichtshof (German Federal Court of Justice) of four questions concerning the interpretation of Article 9(1) of Directive 2008/95 and Article 54(1) of Regulation 207/2009. This decision is only of persuasive authority. As will appear, however, the Court’s reasoning is directly relevant to two of the issues on the appeal. It therefore merits close attention.

The Court said that the first two questions were asking, in essence, whether Article 9 of Directive 2008/95 and Article 54 of Regulation No 207/2009 must be interpreted as meaning that an act, such as a warning letter, whereby the proprietor of an earlier mark or other earlier right opposes the use of a later mark without bringing an administrative or judicial action, is capable of ending acquiescence within the meaning of those provisions.

The Court began its consideration of these questions at [46]-[47] by re-iterating the points it had made in Levi Strauss and Budvar which I have set out in paragraph 35 above. Having pointed out at [48] that, as stated in recital (12), the rule on limitation in consequence of acquiescence was intended to safeguard legal certainty, the Court repeated at [49] what it had said about the meaning of “acquiescence” in Budvar at [44]. The Court went on:

“50.

It follows from the foregoing that the proprietor of an earlier mark or other earlier right is time-barred from seeking a declaration of invalidity or opposing the use of a later mark applied for in good faith, where, for a period of five consecutive years, while being aware of such use, it failed to carry out an act that clearly expressed its wish to oppose that use and to remedy the alleged infringement of its rights.

…

52.

As regards the conditions under which the proprietor of the earlier mark or other earlier right may be regarded as having carried out an act that produces the effects referred to in paragraph 50 above and therefore interrupts the period of limitation, the Court has held that, in any event, the bringing of an administrative or court action before the expiry of that period ends acquiescence and consequently prevents limitation (see, to that effect, … Budvar … at paragraph 49).

53.

By bringing such an action, the proprietor of the earlier mark or other earlier right unequivocally expresses its wish to oppose the use of the later mark and to remedy the alleged infringement of its rights.”

It seems clear from this that the “administrative or court action” envisaged by the Court is one brought by the proprietor “to oppose the use of the later mark and to remedy the alleged infringement of its rights [emphases added]”. This understanding is supported by the opinion of Advocate General Pitruzella, which the Court followed, at [48]-[51]. It is also consistent with what the Court had previously said in Fédération Cynologique at [35].

The Court then turned to the effect of a warning letter and held as follows:

“54.

Where, as in the present case, the bringing of that action was preceded by the sending of a warning letter, with which the proprietor of the later mark did not comply, that warning letter may interrupt the period of limitation in consequence of acquiescence provided that, following the unsatisfactory response to that warning letter, the proprietor of the earlier mark or other earlier right continues to express its opposition to the use of the later mark and takes the measures available to it to enforce its rights.

55.

In contrast, if the proprietor of the earlier mark or other earlier right, having expressed its opposition to the use of the later mark by a warning letter, did not, after noting the refusal of the addressee of that letter to comply with it or to enter into negotiations, pursue its efforts within a reasonable period in order to remedy that situation, where appropriate by bringing an administrative or court action, it must be inferred that that proprietor failed to take the measures available to it to put an end to the alleged infringement of its rights.

56.

Any interpretation of Article 9 of Directive 2008/95 and Articles 54 … of Regulation No 207/2009 to the effect that sending a warning letter is sufficient, in itself, to interrupt the period of limitation would allow the proprietor of the earlier mark or other earlier right to circumvent the regime for limitation in consequence of acquiescence by repeatedly sending a warning letter approximately every five years. Such a situation would undermine the objectives of the regime for limitation in consequence of acquiescence, referred to in paragraphs 46 to 48 above, and would deprive that regime of its effectiveness.

57.

In the light of all the foregoing considerations, the answer to the first and second questions is that Article 9 of Directive 2008/95 and Articles 54, 110 and 111 of Regulation No 207/2009 must be interpreted as meaning that an act, such as a warning letter, by which the proprietor of an earlier mark or other earlier right opposes the use of a later mark without taking the necessary steps to obtain a legally binding solution, does not stop acquiescence and, consequently, does not interrupt the period of limitation.”

It is clear from this that merely sending a warning letter is not enough; but if the warning letter is followed within a reasonable period by administrative or court action, then the sending of the warning letter will stop time running. The Court’s language leaves open the possibility that something other than administrative or court action may suffice providing that it is capable of leading to a legally binding solution to the problem of use of the later trade mark. It is far from clear what that could consist of, but invoking an arbitration clause to resolve the dispute might be an example.

The Court said at [58] that, by its third question, the Bundesgerichtshof was asking whether Article 9 of Directive 2008 and Article 54 of Regulation 207/2009:

“… must be interpreted to mean that the limitation in consequence of acquiescence referred to in those provisions may be prevented by the bringing of a court action in which the proprietor of an earlier mark … seeks a declaration of invalidity of a later mark or opposes the use of that mark, where the application initiating proceedings, although filed before the date of expiry of the period of limitation, was not, owing to a lack of diligence on the part of the applicant, served on the defendant until after that date.”

At first blush, this rephrasing of the question actually asked by the Bundesgerichtshof is odd because the latter did not refer to a court action seeking a declaration of invalidity, there having been no such action on the facts of the case. Nor was there any reference to such an action in the section of the Advocate General’s opinion addressing the third question. It appears, however, that the Court included this for completeness. What matters is that, even as rephrased, the question is focussed upon the consequences of an action having been filed before the end of the five year period, but only served after the end of that period due to a lack of diligence on the part of the applicant. Suppose, for example, an action for a declaration of invalidity is filed after four years and nine months’ acquiescence in the use of the later trade mark, but due to a lack of diligence on the part of the part of the applicant is only served after 5 years and one weeks’ acquiescence: is that action barred or not?

The Court began its answer by saying at [59]:

“As noted in paragraph 52 above, the bringing of an administrative or court action before the expiry of that period ends acquiescence and consequently prevents limitation. ”

This might be thought to suggest that bringing an administrative or court action seeking a declaration of invalidity is sufficient to preclude acquiescence in the use of the later mark. Such an understanding would, however, be inconsistent with what the Court had said in its response to the first and second questions. Moreover, given the context that I have explained in paragraph 49 above, I do not think that is what the Court meant. Rather, I think that what the Court meant was that an administrative or court action brought before the end of the five year period is not barred by limitation.

The Court went on to say:

“62.

… as the Advocate General observed in point 53 of his Opinion, the lodging of the application initiating proceedings normally reflects the genuine and unambiguous wish of the applicant to assert its rights, which is sufficient, in principle, to end acquiescence and, consequently, to interrupt the period of limitation.

63.

The conduct of that party may nevertheless, in certain cases, raise doubts as to that wish and the serious nature of the action brought before the court seised. That is the case, in particular, where, owing to a lack of diligence on the part of the applicant, the application initiating proceedings is not rectified in good time when it fails to meet the formal requirements of national law for service on the defendant.”

The requirement for diligence on the part of the proprietor in serving the proceedings reflects the underlying need for vigilance on the part of trade mark proprietors. I shall return to this point below.

The final point to note is that, when answering the fourth question, the Court said at [70]:

“As explained in the context of the examination of the first and second questions, the proprietor of an earlier mark or other earlier right that, while being aware of the use of a later mark applied for in good faith, fails, for a continuous period of five years, to act in a manner that unequivocally expresses its wish to oppose that use and to remedy the alleged infringement of its rights is time-barred from challenging the use of that later mark.”

This seems to confirm that the action taken by the proprietor must be action to oppose the use of the later mark and to remedy the alleged infringement of its rights.

As the judge noted, it was common ground before him that Combe had become aware of Dr Wolff’s registration and use in the UK of the Wolff EUTM by late January 2014. These proceedings were brought well over five years after that, whether one takes the date of issue or the date of service of the claim form as the operative date. The judge nevertheless held that the Defendants did not have a defence under section 48(1) for four reasons:

The Defendants contend that each of the reasons given by the judge for rejecting the defence is wrong in law.

Cancellation in itself. It is common ground that an EUIPO cancellation action is an administrative action: EUIPO is an administrative body, not a court, albeit that it is subject to the supervisory jurisdiction of the General Court (and of the CJEU on appeal from the General Court). In my judgment, however, the bringing by the proprietor of the earlier trade mark of a cancellation action in respect of the later trade mark is not sufficient to preclude acquiescence by the proprietor of the earlier trade mark in the use of the later trade mark. My reasons are as follows.

First, there is the wording of the legislation and its rationale. There is no relevant difference in this respect between Article 9(1) of Directive 2008/95 (or Article 54(1) of Regulation 207/2009) and section 48(1), and for convenience I will focus upon the latter. As is common ground, this specifies a single condition which has two separate consequences. It is important to distinguish the condition from the consequences.

The condition is that “the proprietor of an earlier trade mark … has acquiesced for a continuous period of five years in the use of a registered trade mark in the United Kingdom, being aware of that use [emphases added]”. Thus the acquiescence must be in respect of the use of the later trade mark. Acquiescence in the registration of a later trade mark which is not being used does not give rise to a defence under section 48(1).

The two consequences are that the proprietor of the earlier trade mark ceases to be entitled, first, “to apply for a declaration that the registration of the later trade mark is invalid [emphasis added]”; and secondly, “to oppose the use of the later trade mark [emphasis added]”. As is common ground, a declaration that the registration of the later trade mark is invalid has no impact at all on the use of that trade mark: the proprietor of the later trade mark remains free to continue to use it. This is because registration confers no positive right to use a trade mark, rather it is an exclusionary right to prevent others from using it: see Fédération Cynologique and R (British American Tobacco UK Ltd) v Secretary of State for Health [2016] EWCA Civ 1182, [2018] QB 149 at [46]-[69] (Lewison LJ giving the judgment of the Court of Appeal). If the proprietor of the earlier trade mark wishes to prevent use of the later trade mark, it must take action to oppose that use, that is to say, action to enforce the rights conferred by the earlier trade mark to prevent the use by other parties of conflicting signs i.e. infringing use.

Both the condition and the second consequence concern the use of the later trade mark. The rationale for this is obvious: having acquiesced in the use of later trade mark for five years, the proprietor of the earlier trade mark cannot (bring an action to) oppose further use of the later trade mark. The rationale for the first consequence is perhaps less obvious, but is nevertheless not hard to discern. As counsel for the Defendants pointed out, it is to protect the ability of the proprietor of the later trade mark to enforce the rights conferred by that trade mark to prevent the use by third parties (i.e. other than the proprietor of the earlier trade mark, which is protected from such a claim by section 48(2) implementing Article 9(3) of the Directive) of conflicting signs. The need for this arises because, in general, the grounds for bringing infringement proceedings mirror the relative grounds of objection to validity. Thus, if the proprietor of the earlier trade mark is able to oppose the use of the later trade mark in infringement proceedings, it will normally also be able to invalidate the later trade mark on the same grounds. This is illustrated by the present dispute, where the grounds relied upon by Combe for invalidating the Wolff EUTM are, so far as second and third Combe Trade Marks are concerned, precisely the same as those Combe rely upon as establishing that use of the Wolff EUTM by the Defendants infringes those Trade Marks.

Combe point out that the defence created by section 48(1) only applies to later trade marks which are registered and not to unregistered trade marks. Combe argue that an application for a declaration of invalidity is an attack upon the existence of the registration which founds the defence, and therefore that must suffice to preclude acquiescence in the use of the earlier trade mark. The judge was persuaded by this argument, but I disagree. As I have explained, registration and use are different things. The defence is founded upon acquiescence in use, not acquiescence in registration. A claim for a declaration of invalidity constitutes action to oppose the registration of the later trade mark, but it does not constitute action to oppose the use of the later trade mark. Action to oppose use of the later trade mark requires a claim for infringement. As the Court of Justice put it in Fédération Cynologique at [47]-[48], actions for infringement and actions for declarations of invalidity “are distinguished in terms of their object and effects”.

The difference is well illustrated by the facts of the present case. Although Combe commenced the EUIPO cancellation action on 2 December 2017, that had no effect at all on Dr Wolff’s continued use of the Wolff EUTM. Nor will it make any difference if Combe are ultimately successful in those proceedings. The only way in which Combe have been able to force Dr Wolff to stop using the Wolff EUTM (or now the Wolff UKTM) in the UK is by bringing these infringement proceedings (subject to the outcome of the appeal). Combe did not commence these proceedings until 2 October 2019.

Combe also argue that, if the Defendants are correct, then any cancellation action brought within the five year period would be barred once there was five years’ acquiescence. This does not follow. Like most limitation provisions, section 48(1) does not bar claims brought before the limitation period expires even if the limitation period subsequently expires before the case is decided by the relevant court or tribunal. Again this is illustrated by the facts of the present case. The EUIPO cancellation action was bought less than five years after Combe became aware of Dr Wolff’s use of the Wolff EUTM, and therefore it could not be barred by acquiescence, as the Cancellation Division held in a decision delivered more than five years after Combe became aware of Dr Wolff’s use. By contrast, these proceedings were brought more than five years after Combe became aware of Dr Wolff’s use of the Wolff EUTM, and thus are potentially barred by acquiescence.

Secondly, there is the broader purpose of the legislation as explained by the CJEU. There are three inter-related aspects to this. The first is that it is generally to strike a balance between the interest of the proprietor of a trade mark to safeguard its essential function, on the one hand, and the interests of other economic operators in having signs capable of denoting their goods and services, on the other. As the CJEU indicated in Budvar at [29], the way in which the proprietor of the earlier trade mark safeguards its essential function of indicating origin is by taking action to enforce its rights to prevent the use of conflicting signs. Removing a later trade mark from the register does not safeguard the essential function of the earlier trade mark.

The second aspect is that protection is limited to those cases in which the proprietor shows itself to be sufficiently vigilant by opposing the use of signs by other operators likely to infringe its mark. This principle is not new. As long ago as 1880 James LJ said in R. Johnston & Co v Archibald Orr Ewing & Co (1880) 13 Ch D 434 at 464 that “the very life of a trade mark depends on the promptitude with which it is vindicated”. As Jacob LJ explained in Phones 4u Ltd v Phone4u.co.uk Internet Ltd [2006] EWCA Civ 244, [2007] RPC 5 at [8], “[l]ike gardens trade mark cases always get worse with neglect – even if rights are not actually lost, delay is apt to turn what would be over in a few weeks by a quick application into a mini State Trial”. The present case is yet another illustration of the truth of that statement. Again, what is required is action by the proprietor to combat infringement.

The third aspect is the need for legal certainty. Like any limitation provision, section 48(1) is designed to prevent rights being enforced if the right holder does not take action for a period of time, here five years. In many contexts this is simply in order to forestall stale claims being litigated when memories have faded and documents have been destroyed or mislaid. But for the reasons explained above, in the trade mark context use of a conflicting sign may affect the average consumer’s perception of the earlier trade mark. That has potential consequences not only for the proprietors of the respective trade marks, but also for the consuming public. The central (although not necessarily the only) justification for trade mark protection is to reduce what the academic literature refers to as “consumer search costs”, but coexistence of confusingly similar trade marks is liable to increase such costs. Section 48(1) gives the proprietor of the earlier trade mark an incentive to prevent this happening.

Next, there is the guidance given by the CJEU as to what amounts to acquiescence and as to what is required to interrupt it. The judge thought that Combe’s argument was supported by what the CJEU had said in Budvar at [44] and [49]. I disagree.

The judge said that a person who brought a cancellation action could not be said to be passive. That is true so far as it goes, but what the CJEU was referring to in Budvar at [44] was a failure “to take measures open to him to remedy a situation of which he is aware”, namely the use of the later trade mark. As I have explained, that can only be remedied by an infringement action, not by a cancellation action. The same message is conveyed by the Court of Justice’s reference to the proprietor being inactive “when faced with a situation which he would be in a position to oppose”.

As for the Court of Justice’s reference in Budvar at [49] to administrative action, the judge appears to have thought that this must mean a cancellation action or the equivalent before a national office. I accept that it is not clear from Budvar what the Court had in mind, but it is not necessarily the case that it meant a cancellation action or equivalent. Counsel for Combe accepted that it may be possible under some systems of law and in some circumstances to restrain infringement by administrative action. Counsel for the Defendants gave a simple example of this, namely where the use of the later trade mark is through importation of the goods, and action is taken by the proprietor of the earlier trade mark under the applicable customs regulations to stop importation. Another potential example might be to request trading standards officers (or equivalent bodies) to seize infringing goods.

Turning to Heitec, the Court’s reasoning when answering the first and second questions, and when answering the fourth question, supports the Defendants’ case. I acknowledge that what the Court said when answering the third question at [58]-[59] appears to support Combe’s case, but for the reasons given above I do not consider that, upon analysis, it really does. In any event, I find the former persuasive and not the latter.

Finally, I would add that the judge also thought that Combe’s argument received some support from what I had said sitting at first instance in W3 Ltd v easyGroup Ltd [2018] EWHC 7 (Ch), [2018] FSR 16 at [349]-[350], but the passages in question merely reflect the arguments of counsel in that case.

Cancellation in context. The judge considered that the surrounding context made it clear that, by bringing the EUIPO cancellation action, Combe were opposing the use of the Wolff EUTM. In my judgment, however, context is incapable as a matter of law of turning a cancellation action into something that amounts to opposing use of the later trade mark. As explained above, use and registration are different things. A cancellation action attacks the registration of the later trade mark, but use can only be opposed by administrative or court action to restrain infringement (or action with equally legally binding effect). Mr Santini gave evidence that, as a commercial person, he thought that there was no distinction between objecting to registration and objecting to use. The judge said he agreed with this. As a matter of law, however, there is a sharp distinction between the two.

In this context both sides sought to rely upon W3 v easyGroup. In that case easyGroup’s solicitors sent W3 a letter before action followed by draft Particulars of Claim within the relevant period, but easyGroup did not follow through by commencing proceedings. I found that the steps taken by easyGroup amounted to serious, detailed and credible threats of infringement proceedings, and held that that was sufficient to stop time running.

In the light of Heitec, it is clear that even serious, detailed and credible threats of infringement proceedings are not enough to preclude acquiescence unless they are followed by administrative or court action (or action with equally legally binding effect). In the present case, Combe made it clear to Dr Wolff in 2014-2015 that Combe did not agree to coexistence of the trade marks in the UK, but on the contrary objected to Dr Wolff’s use of VAGISAN. This message was reiterated in January 2018. At no stage during the relevant period, however, did Combe threaten to bring infringement proceedings against Dr Wolff in the UK. Still less was there a letter before action from solicitors, let alone draft Particulars of Claim. So far as restraining infringement of the Combe Trade Marks was concerned, Combe were entirely passive.

Continuous use on a meaningful or commercial scale. There are two questions here: first, does the use need to be continuous; and secondly, does it need to be on a meaningful or commercial scale? I shall address them separately, but as will appear they are related.

Although the proposition that the use of the later trade mark must be continuous throughout the five year period receives support from some distinguished commentators, I do not accept it. Counsel for the Defendants gave a simple example of a case where it could not be right to require continuous use, namely seasonal goods. Some goods are only sold at certain times of year such as Easter or Christmas. As counsel for Combe was driven to accept, use of the later trade mark during the relevant season for five successive years must be sufficient even if there is no use of it outside the relevant season. More generally, a requirement of continuous use would be unworkable. As counsel for Combe accepted, it could not mean that there had to be use of the later trade mark every second of every hour of every day of every week of every month of every year for five years. How then would one judge whether the use was sufficiently continuous? Would it be enough to use it once an hour? Once a day? Once a week? Once a month? Once a year? And how active would the use have to be for this purpose? Would it be sufficient that goods bearing the trade mark were sitting on the shelf of a shop even if no one bought them for months at a time?

On the other hand, as counsel for the Defendants accepted, it cannot be sufficient for there to be use for a short period of time, say one week, at the beginning of the five year period in question. As counsel for the Defendants submitted, the crux of the single condition in section 48(1) is acquiescence by the proprietor of the earlier trade mark for five successive years in the use of the later trade mark. If there is no use, there cannot be acquiescence in the use. The same message is conveyed by the English language version of recital (12) to Directive 2008/95, which refers to the proprietor of the earlier trade mark having “knowingly tolerated the use [of the later trade mark] for a substantial length of time”.

In this connection, it is pertinent to note that, as Advocate General Tstenjak pointed out in her opinion in Budvar at [69], all the Romance language versions, and the German and Dutch versions, of Article 9(1) use the equivalents of the words “toleration” and “tolerated” rather than the equivalents of the words “acquiescence” and “acquiesced”. Furthermore, taking the German version by way of example, this expresses the condition in the following way: “Hat ... der Inhaber einer älteren Marke … die Benutzung einer jüngeren eingetragenen Marke … während eines Zeitraums von funf aufeinander folgenden Jahren in Kenntnis dieser Benutzung geduldet (Where the proprietor of an older trade mark has tolerated the use of a younger registered trademark for a period of five consecutive years with knowledge of this use)”.

In my judgment the answer to this conundrum is that there must be continued, but not necessarily continuous, use of the later trade mark throughout the five year period. What amounts to continued use for this purpose will be a fact-sensitive question which depends on the nature of the goods or services and on the characteristics of the average consumer of those goods or services. Thus fewer instances of use may suffice for expensive or specialised goods that are only bought occasionally than for low-cost everyday items which are purchased regularly. Even in the case of inexpensive commonplace goods, periodic use over five years may well suffice.

As for the level of use required, the Defendants contend that any use which is an infringing use is sufficient, while Combe contend that the use must amount to genuine use such as would defeat a claim for revocation of the later trade mark on the ground of non-use. In support of the latter contention, counsel for Combe argued that it would not make sense to protect use of the later trade mark if it was liable to revocation for non-use.

Although it again has the support of some distinguished commentators, in my judgment genuine use cannot be the right test. Revocation for non-use takes effect from the end of the five year period in question; it does not operate ab initio (unlike a declaration of invalidity). Furthermore, a claim for revocation on the ground of non-use can be defeated by demonstrating genuine use of the trade mark at any time within that period. It may be enough to use the trade mark only in the last month. Thus a requirement of genuine use would be inconsistent with the requirement of continuous use which Combe themselves advocate. Moreover, the purposes of the two sets of provisions are different. The purposes of the provisions on acquiescence are those set out above. The purpose of the non-use provisions are to sanction a failure to put the trade mark to use within the five years allowed (absent proper reasons for non-use).

Accordingly, I agree with the Defendants that all that is required is use which infringes the earlier trade mark. This is not to say that the scale of use is irrelevant. On the contrary, it may be relevant in two ways. First, as indicated above, it may be relevant to the question of whether there has been continued use of the later trade mark for five years. The acid test, it seems to me, is that there must be use for five years which a vigilant trade mark proprietor could be expected to oppose. Secondly, as discussed below, it may be relevant to the question of whether the proprietor of the earlier trade mark is aware of the use.

Turning to the present case, the judge held that Dr Wolff’s use did not qualify for reasons he expressed as follows at [251]:

“... there was little if any active marketing of VAGISAN from June 2014 onwards, and that led to  sales levels which can very fairly be described as de minimis, at least during the 11 month period between [September] 2015 and August 2016.  [Counsel for the Defendants] submitted that ‘use’ does not have to be on a large or competitive scale, but it seems to me, as [counsel for Combe] submitted, that it needs to be on a meaningful or commercial scale.  By that I mean on a scale which consistent with there being a continuing intention to sell the relevant products commercially.  Here, that was not the position for a period of about two years.  The low levels of activity were equally well consistent with the idea that Defendants’ foray into the market had failed, or at least was suspended for a prolonged period pending further review.  In either case, I see little to persuade me that there was use in any active, commercial sense, and that is surely what the statute requires.”

In my judgment there are a number of flaws in this reasoning. First, the judge’s focus upon sales ignores the fact that, on Combe’s own case, Dr Wolff was continuously infringing the Combe Trade Marks throughout the relevant period by using the Wolff EUTM on the Wolff Website. Although the earliest instance of this use pleaded by Combe was 9 July 2014, Combe’s witness Ms Want gave evidence that Combe UK had discovered “a Vagisan UK targeted website” (i.e. the Wolff Website) in February 2014 and had noted in April 2014 that the website “appeared to have been further developed”.

Secondly, even if consideration is restricted to the sales during the period between 7 September 2015 and 8 August 2016, and use through Dr Wolff’s distributors stocking and offering the goods under the trade mark is ignored, there appears to have been fairly continual use of the Wolff EUTM, albeit that the sales were at a relatively low level.

Thirdly, I cannot agree with the proposition that sales of 498 units of VAGISAN during the period from 7 September 2015 to 8 August 2016 was de minimis at least when viewed from the perspective of infringement. There is very little authority as to the application of the de minimis doctrine to trade mark infringement, but as a matter of principle use of a sign which is truly de minimis is not actionable: see Merck KgaA v Merck Sharp & Dohme Corp [2017] EWCA Civ 1834, [2018] ETMR 10 and cf. Napp Pharmaceutical Holdings td v Dr Reddy’s Laboratories (UK) Ltd [2016] EWHC 1517, [2017] RPC 4. In the present case there has never been any suggestion by Combe that sales of 498 units would not be actionable. On the contrary, the findings of infringement made by the judge include those sales, and by virtue of the judge’s order Combe are entitled to financial relief in respect of them. This is certainly use on a scale which a vigilant trade mark proprietor could be expected to oppose.

Fourthly, although it is Combe’s case in this Court that what is required is not merely continuous use but also genuine use, the judge did not consider whether the use was genuine. Still less did he cite or apply any of the case law of the CJEU as to what amounts to genuine use (or any of the domestic case law applying the principles laid down by the CJEU), no doubt for the excellent reason that he was not referred to any of that case law. If he had been referred to that case law, he would have discovered that, in that context, there is no de minimis rule: see Case C-259/02 La Mer Technology Inc v Laboratoires Goemar SA [2004] ECR I-1159 at [25].

In summary, Dr Wolff did in fact make continuous infringing use of the Wolff EUTM in the UK throughout the five years commencing in late January 2014; but in any event Dr Wolff certainly made continued use of the Wolff EUTM at a level which was more than sufficient for the purpose of section 48(1).

Awareness. Although I disagree with the judge’s first three reasons for rejecting the Defendants’ section 48(1) defence, I agree with the fourth. Section 48(1) is explicit that the proprietor of the earlier trade mark must be aware of the use of the later trade mark for five years. This is perfectly consistent with Article 9(1). If the proprietor is not aware of the use for a significant part of that period, it cannot be said to have acquiesced in that use or knowingly tolerated it.

The Defendants contend that, once the proprietor of the earlier trade mark becomes aware of the use of the later trade mark, then it will continue to be aware of such use if the use continues. I do not agree that this is the correct way of looking at the issue. Acquiescence or toleration requires continued awareness of the use. That does not mean that the proprietor can say that it has ceased to be aware of the use merely because of the passage of time, or that it can turn a blind eye to the question of whether the use is continuing. But if the proprietor genuinely and reasonably believes that the use has ceased for a significant period of time, then I do not consider that the proprietor can be said to be acquiescing in or tolerating the use during that period.

In the present case the judge found as follows:

“253.

On this point, I accept Ms Want’s evidence that VAGISAN was kept on a watching brief by Combe, but had no commercial profile for a long period.  In her witness statement she said that she thought VAGISAN had either dropped off the market entirely or were so small that she may not have noticed them.  In her oral evidence, she explained that during 2015, VAGISAN was not ‘coming up in our IRI.’  As she explained, IRI obtains and reports records of all sales through tills at major multiples and at a representative sample of independent pharmacies.  The lack of any presence in the IRI returns is significant, because it suggests there were no sales at all in major multiples, supermarkets or monitored pharmacies.  Hence, Ms Want said when pressed in cross-examination …:

‘Q.      You did not think they had withdrawn from the market though?

A.        They might well have done, but I do not know what - obviously I did not know what their actual strategy was and where they were selling. Could they still be in some tiny little pharmacy that is not being picked up somewhere, I do not know. But from what I can see there is nothing there.

Q.        That is not the question. I said did you think at the time that they had withdrawn from the market, Ms Want? It is a very simple question.

A.       Then, I guess ---

Q.         Did you think it or not?

A.

Yes, I did. I must have done, because that is why they were not there.’

254.

In light of this evidence, I find that the Claimants were simply not aware for a continuous period of five years of the Defendants’ use of their later mark.  Ms Want and her team monitored the market with a degree of care for competitor products, and it seems to me Ms Want was entitled reasonably to assume, as she did, that for whatever reason Wolff were simply not active in the market, in the sense of actively promoting the sale of their products, for a long period during 2015 and 2016.  They simply ‘were not there.’  Such limited use as there may have been was at such a low level that it did not register with Ms Want, when it was her job to know about it.”

Thus Combe were not aware of the use of VAGISAN for a long period during 2015 and 2016 despite keeping a watch for such use. In those circumstances, it was reasonable for Combe to assume that Dr Wolff had withdrawn from the UK market. That being so, I agree with the judge that Combe did not acquiesce in the use that did in fact occur during that period.

Counsel for the Defendants submitted that the judge’s analysis was flawed because he had failed to take into account the undisputed fact that Combe had become aware of the Wolff Website in February 2014. It was not put to Ms Want in cross-examination, however, that she was aware of that use of VAGISAN during the period from early 2015 to late 2016. She may have assumed, without checking, that the website had ceased to be targeted at the UK.

I would add that this period followed the abortive negotiations between the parties which came to an end in February 2015. As discussed above, during those negotiations Combe had objected to Dr Wolff’s continuing use of VAGISAN in the UK. Although Dr Wolff did not back down in the negotiations, Dr Wolff’s apparent absence from the market afterwards goes a long way to explain why Combe evidently did not feel the need to threaten, let alone bring, infringement proceedings during the period from then until late 2016. (Combe’s dilatoriness in bringing proceedings after actual confusion became evident in early 2017 is inexplicable, but immaterial.)

Conclusion. I would therefore dismiss the Defendants’ appeal against the judge’s rejection of their section 48(1) defence.

As originally enacted, section 11(1) of the 1994 Act provided as follows:

“A registered trade mark is not infringed by the use of another registered trade mark in relation to goods or services for which the latter is registered (but see section 47(6) (effect of declaration of invalidity of registration)).”

This provision was entirely “home-grown” and had no basis in Directive 89/104. Nor was there any counterpart in Regulation 40/94.

With effect from 14 January 2019, section 11(1) was amended by the Trade Marks Regulations 2018 (SI 2018/825) regulation 12(2), made pursuant to section 2(2) of the European Communities Act 1972, to read as follows:

“A registered trade mark is not infringed by the use of a later registered trade mark where that later registered trade mark would not be declared invalid pursuant to section 47(2A) or (2G) or section 48(1).”

This version implemented Article 18(1),(2) of Directive 2015/2436. Parallel provisions are contained in Article 16(1),(2) of Regulation 2017/1001.

As noted above, in Fédération Cynologique the Court of Justice ruled thatArticle 9(1) of Regulation 207/2009 must be interpreted as meaning that the exclusive right of the proprietor of a Community (now EU) trade mark to prohibit all third parties from using, in the course of trade, signs identical with or similar to its trade mark extends to a third-party proprietor of a later registered Community (now EU) trade mark, without the need for that latter mark to have been declared invalid beforehand. It is common ground that the same must apply to Article 5(1) of Directives 89/104 and 2008/95, and therefore that the original version of section 11(1) was incompatible with those Directives.

In his judgment the judge only considered the amended form of section 11(1), because that is all he was asked to consider. The judge rejected the Defendants’ defence under the amended version on the ground that there had been no acquiescence by Combe in the use of the Wolff EUTM in the UK, and therefore the Wolff EUTM was not a trade mark which would not be declared invalid pursuant to section 48(1).

On the appeal the Defendants sought permission to raise a new argument, namely that Dr Wolff had a defence under the original version of section 11(1) in respect of the period prior to 14 January 2019 because the Wolff UKTM was deemed to have been registered as of 18 November 2011 (see paragraphs 7 and 12 above) unless and until the Wolff UKTM was declared invalid. Combe did not object to the new argument being raised, but argued that the Court should adopt a conforming interpretation of the original version of section 11(1) which accorded with the amended version. The Defendants’ response to that argument was to contend that it was not possible to interpret the original version of section 11(1) in that manner.

After counsel for the Defendants had opened the appeal, counsel for Combe informed the Court that Combe accepted that the Defendants were entitled to rely upon the original version of section 11(1) as providing them with a defence in respect of acts committed prior to 14 January 2019.

Combe sought to turn this concession to their advantage by arguing that it followed that Combe were not in a position to oppose use of the Wolff UKTM prior to 14 January 2019, and therefore their claim for infringement of the Combe Trade Marks could not be barred by acquiescence applying the reasoning in Budvar. The Defendants disputed this, pointing out that in fact Combe had issued infringement proceedings at a time when the relevant mark was the Wolff EUTM, thus demonstrating that Combe were in a position to oppose use of the Wolff UKTM even though it had not yet come into existence. Fortunately, it is not necessary to resolve this issue given my conclusion that the Defendants’ defence under section 48(1) fails anyway.

So far as the amended version of section 11(1) is concerned, the Defendants contend that the judge was wrong to reject their defence for three reasons. First, because he was wrong on acquiescence. I have concluded that he was right about that, however.

Secondly, the Defendants say that the judge focussed on the wrong trade mark, namely the Wolff EUTM rather than the Wolff UKTM. Combe counter that, as the judge noted at [31], it was common ground before him that the Wolff UKTM was a clone of the Wolff EUTM that came into existence on 31 December 2020 and whose validity depended on the validity of the Wolff EUTM. It seems to me that, on any view, the judge was correct to consider the Wolff EUTM so far as the period prior to 31 December 2020 is concerned. Furthermore, subject to the point discussed below, the replacement of the Wolff EUTM by the Wolff UKTM on 31 December 2020 cannot make any difference to the Defendants’ defence under the amended version of section 11(1).

Thirdly, the Defendants contend that the judge wrongly pre-judged the validity of the Wolff EUTM, and hence of the Wolff UKTM. On the face of it, the judge did no such thing: he simply noted at [269] that, as matters stood, the Wolff EUTM had been declared invalid. It is true that the judge did not consider what the consequences would be if Dr Wolff’s appeal to the General Court were to succeed, but it appears that the reason for this is that the question was not explored in argument before the judge.

There are two different ways in which Dr Wolff’s appeal could succeed. First, it could succeed on the merits if Dr Wolff can persuade the General Court to take a different view on likelihood of confusion between the Wolff EUTM and the Combe Trade Marks to that taken by the judge. Secondly, it could succeed on the basis that a decision as to whether there is a conflict between the Wolff EUTM and the Combe Trade Marks is devoid of purpose, in which case the Cancellation Division would have to take a new decision based on Combe’s other registrations. Either way, the upshot would be that the validity of the Wolff UKTM would be unaffected by the outcome of the cancellation action (in the second case, by virtue of Schedule 2 paragraph 21A(4) of the 1994 Act).

But would that mean that the Defendants could rely upon the amended version of section 11(1)? Although the Wolff UKTM would not be declared invalid on this hypothesis, this would not be “pursuant to … section 48(1)” (it is common ground that section 47(2A) and (2G) are not relevant). It does not matter for this purpose whether one is considering the outcome of the EUIPO cancellation action (in which Dr Wolff did not appeal the Cancellation Division’s decision that Combe had not acquiesced in the use of the Wolff EUTM for five years as at the date that the cancellation action was commenced) or the outcome of these proceedings (in which the judge held, and I have agreed, that Combe had not acquiesced in the use of the Wolff EUTM for five years as at the date they were commenced).

The Defendants accept that this is the result if section 11(1) is given a literal interpretation, but argue that it should be purposively interpreted. The Defendants say that it should be interpreted as meaning that the defence is available if the later trade mark would not be declared invalid including pursuant to section 47(2A) or (2G) or section 48(1). I am unable to accept this argument. That is simply not what the subsection says, and the Defendants’ argument amounts to rewriting it under the guise of purposive interpretation. This is particularly inappropriate given that it implements Article 18 of Directive 2015/2436.

I therefore conclude that it does not matter whether or not Dr Wolff’s appeal to the General Court is successful. If the appeal is dismissed, the existing declaration of invalidity of the Wolff EUTM will take effect and the validity of the Wolff UKTM will fall with it. Even if the appeal is allowed, however, that will not enable the Defendants to rely upon the amended version of section 11(1). I would therefore dismiss the Defendants’ appeal against the judge’s rejection of that defence.

For completeness, I would add that neither side suggests that Combe’s recent application to the UKIPO for a declaration of invalidity of the Wolff UKTM is relevant to this issue.

There is no challenge by the Defendants to the judge’s conclusion that their use of the sign VAGISAN gave rise to a likelihood of confusion so as to infringe the Combe Trade Marks pursuant to section 10(2)(b) of the 1994 Act. In that context the judge set out the principles for assessing whether there is a likelihood of confusion at [43]-[47], and there is no suggestion that he misdirected himself. Nor is there any challenge to the judge’s findings that (i) the average consumer’s attentiveness would be towards the lower end of the spectrum ([79]), (ii) the Defendants’ goods were identical to goods for which the Combe Trade Marks were registered ([85]) and (iii) VAGILSIL had acquired enhanced distinctiveness through use ([94]).

When it came to determining whether the Defendants’ use of DR WOLFF’S VAGISAN gave rise to a likelihood of confusion, the judge cited what I had said in Whyte & Mackay Ltd v Origin Wine UK Ltd [2015] EWHC 1271 (Ch), [2015] FSR 33:

“18.

The judgment in [Case C-591/12 P Bimbo SA v Office for Harmonisation in the Internal Market (Trade Marks and Designs) [EU:C:2014:305]] confirms that the principle established in [Case C-120/04 Medion AG v Thomson Sales Germany & Austria GmbH [2005] ECR I-8551]is not confined to the situation where the composite trade mark for which registration is sought contains an element which is identical to an earlier trade mark, but extends to the situation where the composite mark contains an element which is similar to the earlier mark. More importantly for present purposes, it also confirms three other points.

19.

The first is that the assessment of likelihood of confusion must be made by considering and comparing the respective marks - visually, aurally and conceptually - as a whole. In Medion v Thomson and subsequent case law, the Court of Justice has recognised that there are situations in which the average consumer, while perceiving a composite mark as a whole, will also perceive that it consists of two (or more) signs one (or more) of which has a distinctive significance which is independent of the significance of the whole, and thus may be confused as a result of the identity or similarity of that sign to the earlier mark.

20.

The second point is that this principle can only apply in circumstances where the average consumer would perceive the relevant part of the composite mark to have distinctive significance independently of the whole. It does not apply where the average consumer would perceive the composite mark as a unit having a different meaning to the meanings of the separate components. That includes the situation where the meaning of one of the components is qualified by another component, as with a surname and a first name (e.g. BECKER and BARBARA BECKER).

21.

The third point is that, even where an element of the composite mark which is identical or similar to the earlier trade mark has an independent distinctive role, it does not automatically follow that there is a likelihood of confusion. It remains necessary for the competent authority to carry out a global assessment taking into account all relevant factors.”

The judge held the Defendants’ use of DR WOLFF’S VAGISAN gave rise to a likelihood of confusion for the reasons he expressed at [289]:

“i)

I of course accept that the phrase ‘DR WOLFF’s’ provides a degree of phonetic distinction to the composite mark, and moreover has surnominal and possessive significance - i.e., as [counsel for the Defendants] indicated, it suggests that ‘whoever Dr Wolff is, Vagisan belongs to him or her.’

ii)

That much is true, but to my mind it still does not prevent the word VAGISAN having its own independent, distinctive role.  VAGISAN is the product being sold.  That is what the average consumer is interested in, given her condition.  In the composite phrase, therefore, it has a natural emphasis and distinction.

iii)

As Ms Want pointed out in her witness statement for trial, it is common for pharmaceutical and cosmetic brands to use ‘Dr’ followed by a name as the ‘house’ brand for a range of sub-products (she gave a number of examples.)  In such cases, it seems to me, the use of ‘Dr’ followed by a name is intended to seek to establish the credibility of the product - since it gives the impression it has the imprimatur of someone who is medically qualified.  It is therefore not, it seems to me, likely to be understood as a signifier of trade origin, but rather an indicator of the authenticity and likely effectiveness of the product which is then described.   That conclusion, it seems to me, is reinforced by the evidence of Mr Bowman (see above at [277]), to the effect that he still refers to the Defendants’ Moist Cream product as VAGISAN despite the rebranding.  It seems quite natural to me that he should do so.

iv)

The next point of course is the one addressed at some length above, namely that VAGISAN is likely to be confused with VAGISIL.  They are obviously not the same word, but they are sufficiently similar as to give rise to confusion and it seems to me that on the face of it, that remains so even if the comparison is between the composite phrase DR WOLFF’s VAGISAN and VAGISIL.  VAGISAN retains its own independent, distinctive role even in the composite phrase and is very similar to the word VAGISIL.  The strong similarity suggests they spring from the same commercial source, or at any rate, from sources which are linked commercially. 

v)

That conclusion is reinforced by the fact that DR WOLFF is not and never has been a well-known brand name in the UK.  Wolff’s products, such as Alpecin Caffeine Shampoo, may be well known, but the brand name DR WOLFF is not.  Thus, the average consumer, seeing the phrase ‘DR WOLFF’s’ used in conjunction with ‘VAGISAN’, would not be inclined to think - yes of course, that is Dr Wolff the manufacturer of Alpecin Shampoo which is different to the company which manufactures VAGISIL.  Moreover, as Ms Want pointed out in her evidence, neither are consumers familiar with Combe as the manufacturers of VAGISIL.  In other words, there is nothing in the overall context which is likely to help the average consumer make a clear distinction between the separate trade origins of the two brands, even if one is described as DR WOLFF’s VAGISAN.

vi)

The upshot, in my view, is that the average consumer faced with DR WOLFF’s VAGISAN would nonetheless still consider it to be associated with VAGISIL.  She would naturally place emphasis on the word ‘VAGISAN’ as having an independent distinctive role, and would naturally associate it with VAGISIL.”

As is common ground, this is a multi-factorial evaluative assessment, and therefore this Court is not entitled to interfere with it in the absence of an error of law or principle. The Defendants contend that the judge erred in principle in no less than six respects.

First, the Defendants assert that the judge wrongly equated DR WOLFF’S VAGISAN with VAGISAN, and thus removed the DR WOLFF’s element of the sign from the comparison with VAGISIL. In my judgment, however, the judge did no such thing. On the contrary, he compared the sign DR WOLFF’S VAGISAN with VAGISIL, but in making that comparison he correctly recognised that DR WOLFF’S VAGISAN was a composite sign consisting of two elements, DR WOLFF’S and VAGISAN.

Secondly, the Defendants argue that the judge failed to compare the visual, aural and conceptual similarity of DR WOLFF’S VAGISAN as a whole with VAGISIL. Again, I do not accept this. The judge had already concluded that VAGISAN was very similar to VAGISIL, and he recognised that the DR WOLFF’S element of the composite sign was quite different.

Thirdly, the Defendants complain that the judge failed to take into account the rule of thumb that consumers pay most attention to the beginning of a trade mark: see e.g. Case T-183/02 and T-184/02 El Corte Inglés SA v Office for Harmonisation of the Internal Market (Trade Marks and Designs) [2004] ECR II-965 at [81]. It appears that the reason for this is that the judge was not referred to the relevant case law, but it does not matter. The rule of thumb has no relevance to the present case because the composite sign consists of two elements, one of which has no counterpart in the trade mark and the other of which is very similar to the trade mark.

Fourthly, the Defendants assert that the judge wrongly failed to have proper regard to the DR WOLFF’S element of the composite sign because he considered that the average consumer would ignore it. The judge did not find that the average consumer would ignore the DR WOLFF’S element, however. What he found was that the VAGISAN element retained an independent distinctive role in the composite sign. It is fair to say that, in reaching that conclusion, the judge contradicted himself at [104(iii)] when he said that it was common for DR NAME signs to be used as house brands in the pharmaceutical and cosmetic field and therefore such a sign was not likely to be understood as a signifier of trade origin. Nevertheless, the point that DR NAME signs are commonly used as house marks supports his conclusion that VAGISAN played an independent distinctive role in the composite sign. In my judgment that is a conclusion which was at least open to him. Indeed, I agree with it.

Fifthly, the Defendants argue that the judge wrongly took into account the distinctive character of the composite sign when assessing the likelihood of confusion. It is correct that, in general, the distinctive character of the sign is irrelevant to the assessment of likelihood of confusion: see e.g. Case T-533/18 Wanda Films SL v European Union Intellectual Property Office [EU:T:2019:727] at [54]-[55]. The judge’s reasoning, however, was that the fact that DR WOLFF’S was not a house mark which was well known to the average consumer in the UK meant that the presence of that element in the composite sign did not eliminate the likelihood of confusion arising from the independent distinctive role of the VAGISAN element. I see no error in that analysis. I would add that, even if DR WOLFF’s had been well known to the average consumer, that would not necessarily have avoided the risk of confusion.

Lastly, the Defendants assert that the judge failed to take into account either the context of the use of the sign or the descriptiveness of the VAGI- element of VAGISIL and VAGISAN. The first part of this assertion makes no sense. The Defendants have not identified what “context” the judge should have taken into account but did not. As for the second part, the judge did consider this when assessing the likelihood of confusion due to the use of VAGISAN He concluded that, although the VAGI- element was allusive, this did not detract from the likelihood of confusion ([105]). As I have already noted, there is no challenge to that assessment. This conclusion is equally applicable to the composite sign.

I would therefore dismiss the Defendants’ appeal against the judge’s conclusion that their use of DR WOLFF’S VAGISAN gave rise to a likelihood of confusion.

As noted above, it is now conceded by Combe that the Defendants have a defence in respect of the period prior to 14 January 2019. Subject to that, I would dismiss this appeal.