England and Wales Court of Appeal (Civil Division) Decisions

IN THE COURT OF APPEAL (CIVIL DIVISION)
ON APPEAL FROM THE HIGH COURT OF JUSTICE
CHANCERY DIVISION (PATENTS COURT)
John Baldwin QC sitting as a Deputy High Court Judge
[2012] EWHC 1569 (Ch)

		Royal Courts of Justice Strand, London, WC2A 2LL
		30th July 2013

Lord Justice Kitchin:

Introduction

The subject matter and the parties

1. This is an appeal against the judgment and consequential order of Mr John Baldwin QC sitting as a deputy judge of the Patents Court in a patent action concerning organic pigments. These are employed commercially in the colouration of plastics, synthetic fibres and surface coatings such as inks and paints. In contrast to organic dyes, which are generally applied from solution, organic pigments consist of finely divided, particulate solids which are practically insoluble in the media to which they are applied, and so are usually incorporated by some form of dispersive or milling process that can break up and evenly distribute the pigment particles.
2. The appellant, Clariant, is a well known manufacturer of such pigments. For many years it has made a particular yellow pigment called PY 191 which it has sold under the brand name HGR. It is what is known as a calcium laked azo pigment and was protected by a series of patents which Clariant secured in Europe, the USA and Japan. There was no dispute between the parties that the teaching of all of these patents is the same in all relevant respects and so reference was made at the trial to US patent 5,047,517 ("the 517 patent"), this being the only one written in English.
3. In the period from 1995 to 1997, Clariant suffered production problems which led it to vary its production method and, in particular, to introduce into it a new solvent conditioning process.
4. Clariant formed the view that the product of this modified process was a new crystalline form of PY 191. It called this new form the beta polymorph so as to distinguish it from the old form, which it called the alpha polymorph. It also filed an application for the patent in suit, EP 1 170 338 ("the 338 patent") which claims a priority date of 4 July 2000. In its granted form, the 338 patent contains claims to the solvent conditioning process for making the beta polymorph (claims 1 to 5), the beta polymorph itself (claim 6), a PY 191 mixture containing 10%, 20%, 50%, 75% or 90% of the beta polymorph (claim 7), and the use of the beta polymorph or a mixture containing the beta polymorph for pigmenting various different kinds of industrial products including polymers (claim 8).
5. It is now known that PY 191 exists in different crystalline forms as a result of different degrees of hydration. In particular, it is known that the alpha polymorph is a monohydrate salt and the beta polymorph is a trihydrate salt. They are therefore pseudo-polymorphs rather than true polymorphs. Clariant markets both the alpha and beta polymorphs, the former as HGR and the latter as H2GR, and it claims that the beta polymorph has a "more greenish, yellow colour" and a "higher colour strength and higher transparency" than the alpha polymorph.
6. The respondent, Sudarshan, is another well known pigment manufacturer with its headquarters in Pune in India. In 2005 Sudarshan developed its own brand of PY 191 which it called Sudafast Yellow 132. It was aware of the 517 patent and adopted a manufacturing process which it believed was a routine implementation of its teaching. It also decided that while the 517 patent and its equivalents remained in force, it would only sell Sudafast Yellow 132 in India, this being a territory to which they did not extend.
7. In September 2009 the 517 patent and its equivalents expired due to effluxion of time and Sudarshan thereupon decided to launch Sudafast Yellow 132 outside India. It was aware of the 338 patent but was not concerned because it believed that it was using a routine implementation of the process described in the 517 patent, and the 338 patent itself states that the product of this process is the alpha polymorph. Further, it knew it was not using the solvent conditioning process described in the 338 patent.
8. Sudarshan was therefore very surprised to receive a letter dated 25 November 2009 from Dr Klaus Hütter of the intellectual property department of Clariant asserting that Sudafast Yellow 132 contained a significant proportion of the beta polymorph and that the importation and sale of this product was prohibited because it amounted to an infringement of the 338 patent. The letter concluded with a warning that such patent infringement might "involve the most serious penalties to the infringer and his customers".
9. Sudarshan took the threat of patent infringement extremely seriously and so immediately stopped marketing Sudafast Yellow 132 outside India, notwithstanding that the commercial consequences of so doing were likely to be very significant. It also began an investigation of PY 191 to try and determine how it could be that the product of what it thought was an old process could contain what was said to be the new beta polymorph.
10. As a first step, Sudarshan managed to secure a sample of HGR sold by Clariant to a Swiss customer in 1998 and which must therefore have been made by Clariant's old process. Tests revealed that this contained over 35% of the beta polymorph. This too was surprising to Sudarshan because it was again inconsistent with the statement made in the 338 patent that the process of the 517 patent would produce the alpha polymorph.
11. Further work by Sudarshan led it to conclude the following, all of which are now accepted by Clariant. First, the process described in the 517 patent produces as an intermediate product a filtered suspension of PY 191 which is known as wet cake, and this contains pigment which is entirely in the form of the beta polymorph. Second, drying the wet cake at elevated temperature results in a gradual conversion of the beta polymorph to the alpha polymorph. Third, and conversely, addition of the water to the alpha polymorph by, for example, wetting its surface, results in its conversion back to the beta polymorph. So also, exposure of the alpha polymorph to humidity in the atmosphere will result in its gradual conversion to the beta polymorph, the rate of conversion depending upon the temperature and degree of humidity. Finally, and as may be inferred from the foregoing, the beta polymorph is thermodynamically more stable than the alpha polymorph under ambient conditions.
12. Sudarshan also took the step of modifying its own commercial process to conform as closely as possible to that disclosed in the 517 patent. In particular, it raised the drying temperature from the 80-85°C it had been using to 120°C, this being the particular temperature disclosed in the 517 patent; and it eliminated all additives save for one called rosin which is routinely used to inhibit crystal growth and so produce transparent pigments of fine particle size. It has used this modified process, which it calls the new process, for its Indian production ever since.

The proceedings

13. On 16 June 2010 Sudarshan issued these proceedings seeking revocation of the 338 patent so far as it claimed the beta polymorph and the use of the beta polymorph for the purpose of pigmenting other products, that is to say claims 6 to 8 (as granted); a declaration that Sudafast Yellow 132 made by its old process or its new process would not infringe the 338 patent; and damages or an account of profits in respect of unjustified threats of patent infringement.
14. Sudarshan's attack on the validity of claims 6 to 8 of the 338 patent was made on two distinct fronts. The first, described by the parties as the Gillette defence, was that the old and the new processes which Sudarshan had used to make Sudafast Yellow 132 were entirely obvious implementations of the teaching of the 517 patent. Accordingly, in so far as Sudarshan's activities fell within the scope of any of the claims of the 338 patent, those claims must be invalid.
15. The second was that the teaching in the 517 patent of a process which involves the production of a wet cake which contains 100% of the beta polymorph, albeit as an intermediate, invalidates all the product claims of the 338 patent. Moreover, Sudarshan continued, although the 517 patent teaches the drying of this wet cake at 120°C, it was also obvious to use it for pigmenting polymers and the like, and so the claim to the use of the beta polymorph for these purposes was invalid too.
16. Clariant responded with a number of applications to amend the patent and a counterclaim for infringement. The claim for infringement was based upon the supply by Sudarshan to a customer called Colormatrix of a sample of Sudafast Yellow 132 in dry powder form which contained in excess of 80% of the beta polymorph. It also disputed that it had made any actionable threat.
17. Clariant's applications to amend the 338 patent were presented as a series of fall back positions. First, prior to trial, it applied to divide claims 6 and 7 into separate claims (proposed claims 6 to 10) containing respectively 10%, 25%, 50%, 75%, and 90% of the beta polymorph. The purpose of these amendments was to distinguish PY 191 made according to the 338 patent from PY 191 made according to the 517 patent.
18. As what it called a second line of amendment, Clariant also applied to add a new use claim of more limited scope than original claim 8, which reads:

"8. The use of C.I. Pigment Yellow 191 as claimed in claim 6 for pigmenting varnishes, polymers, printing inks, aqueous or solventborne pigment preparations, electrophotographic toners and developers, powder coating materials, inks, preferably ink-jet inks, color filters, and for coloring seed."

19. This claim remained as proposed amended claim 11. However, Clariant sought to add a new claim 12:

"12. The use of C.I. Pigment Yellow 191 as claimed in any of claims 6 to 10 for pigmenting polymers and powder coating materials."

20. This claim is, so Clariant contended, limited to uses where it would not be technically possible or obvious to use PY 191 in wet cake form.
21. Shortly before trial Clariant also made a conditional application to add a new claim 13 in this form:

"13. The use of C.I. Pigment Yellow 191 as claimed in any of claims 6 to 10 and which has been subjected to fine division for pigmenting polymers and powder coating materials."

22. It will be appreciated this introduced a further limitation to proposed claim 12, namely that the pigment must be subjected to fine division, presumably on the assumption that it would further distinguish the use of wet cake. When this assumption was questioned by Sudarshan in its opening submissions, not least because the 338 patent itself states that fine division can be by wet or dry grinding or kneading, Clariant abandoned its original proposal and made a conditional application to add a revised amended claim 13 in these terms:

"13. The use of C.I. Pigment Yellow 191 as claimed in any of claims 6 to 10 for pigmenting polymers and powder coating materials and which has been subjected to mechanical fine division by dry grinding."

23. Sudarshan submitted that the court should only consider the amendment on the basis that Clariant could not complain about any inadequacies in the evidence resulting from the late stage at which it had been proposed. This was accepted by Clariant, as the judge recorded at [57]:

"Mr Howe initially objected to this proposed amendment being considered at all since it was introduced after the evidence had been completed and so he had been denied a proper opportunity to lead or cross examine to all matters which were relevant. He argued that if I were to consider it at all, then I must do so bearing in mind that the limited evidential support for his argument should not be held against him; in effect I was being invited to take a broad view of the evidence. Mr Watson agreed that he could not object to this course."

24. Finally, Clariant deployed what it called a second auxiliary request which embodied this "disclaimer":

"EP 0 361 431 [the 517 patent] discloses the production of a pigment of Formula (1) as an aqueous suspension which is then filtered and washed and dried. We disclaim the said pigment as an aqueous suspension or in a form which has not been dried. "

25. This was then followed by a claim set incorporating proposed claims 1 to 11 to which I have referred. The purpose of this disclaimer was again to avoid any allegation of invalidity based upon the wet cake described in the 517 patent.

The judgment

26. The judge concluded that Sudarshan had made out both of its attacks on claims 6, 7 and 8 as granted. He was satisfied that the wet cake produced by following the teaching of the 517 patent would contain 100% of the beta polymorph and so fell within the scope of claims 6 and 7, and that it was obvious to use the wet cake for pigmenting polymers, with the result that claim 8 fell too.
27. As for the Gillette defence, the judge held that both Sudarshan's old and new processes were obvious implementations of the teaching of the 517 patent; that pigment made using the old process contained over 80% of the beta polymorph; and that pigment made using the new process contained about 40% of the beta polymorph. It followed that claims 6 and 7 were invalid for obviousness, and so also was claim 8, it being accepted (and indeed alleged) that Sudarshan's product was intended for use in pigmenting polymers.
28. Turning to the proposed amendments, the judge dealt first with the proposed new product claims 6 to 10 and he held, inevitably in the light of his earlier findings, that these were all invalidated by the teaching of the 517 patent to produce a wet cake; and so also, it was obvious to use this wet cake for purposes falling within proposed use claims 11 and 12. He also found that proposed claim 12 would result in the specification disclosing additional matter contrary to s.76(3) of the Patents Act 1977 ("the 1977 Act"), a provision which corresponds to 123(2) EPC.
29. Proposed claim 13, in its revised form, is directed to the use of PY191 in accordance with one of the earlier product claims for pigmenting polymers and powder coatings, but it must first have been dry ground. Clariant argued that subjecting a wet cake to a process of dry grinding will inevitably result in the conversion of all or most of the beta form to the alpha form. It followed, so Clariant contended, it was not obvious to perform a method within this claim starting with any aspect of the teaching of the 517 patent.
30. Sudarshan responded that this claim would also result in the specification disclosing additional matter. Moreover, it was in any event obvious to take the product of dry grinding and then subject it to water treatment before using it for pigmenting polymers, and this would result in any alpha polymorph being converted back to the beta form.
31. The judge agreed with Sudarshan. He found the claim would result in the specification disclosing additional matter. He also found that it was perfectly normal and obvious to water treat pigment which had been finely ground, and that was so whether the pigment had been ground when dry or wet. It followed that claims 13 was not allowable either.
32. The judge then addressed the disclaimer, the final fall back. He refused this for two reasons; first, it was not clear in its terms; and second, it would again result in the specification disclosing additional matter.
33. Finally, the judge turned to the question of threats. Sudarshan contended that the letter dated 25 November 2009 was a threat made both to it and its customers. Clariant responded first, that the threat was justified; second, that it was only made to Sudarshan, as manufacturer, and so was not actionable; and third, that at the time the threat was made it did not know and had no reason to suspect that the patent was invalid, and so it had a defence under s.70(2A)(b) of the 1977 Act.
34. On this issue the judge again preferred the submissions of Sudarshan. He held that the threat was not justified; that it was made to Sudarshan's customers, albeit not communicated to them; and that Clariant did have reason to suspect that the patent was invalid.

The appeal

35. On this appeal Clariant recognises that it cannot challenge the judge's conclusions in relation to original claims 6 to 8 and proposed claims 6 to 11 in the light of the wet cake. But it contends that the judge fell into error in making all his other findings. Helpfully, it explained that it relies first and primarily upon proposed claim 12 as dependent on upon any of the proposed product claims. Alternatively, and as a second string to its bow, it relies upon proposed claim 13 on the same basis. And finally, if it fails on both proposed claims 12 and 13, it relies upon the disclaimer.
36. Clariant also submits that the judge fell into error in relation to threats because the letter it wrote on 25 November 2009 cannot be interpreted as a threat to Sudarshan's customers; and further, Clariant did not at any material time know or have reason to suspect that the patent was invalid, and the judge ought to have so held.
37. It is convenient to address these submissions in the following order:

i) Did the judge fall into error in finding that it was obvious to use wet cake for pigmenting polymers?

ii) Did the judge fall into error in concluding that it was obvious to water treat dry ground wet cake?

iii) Was the judge right to hold that proposed claims 12 and 13 would result in the specification disclosing additional matter?

iv) Was the proposed disclaimer allowable?

v) Did the judge fall into error in relation to the Gillette defence?

vi) Was the letter of 25 November 2009 a threat to Sudarshan's customers? If it was, does Clariant have a defence under s.70(2A)(b) of the 1977 Act?

Was it obvious to use a wet cake for pigmenting polymers?

38. It is convenient to take this issue first because it is potentially determinative of the appeal in relation to proposed claim 12.
39. The judge heard evidence on this issue from two experts in pigment chemistry, Dr McHugh on behalf of Sudarshan and Dr Heron on behalf of Clariant. He was also referred to evidence given by Professor Schmidt, a witness of fact but also an expert in crystal chemistry and a named inventor of the 338 patent. Clariant placed particular reliance on the fact that PY 191 had never been sold by anyone as a wet cake. Conversely, Sudarshan led evidence that it had frequently sold other pigments in wet cake form, and had done so for use in pigmenting polymers.
40. The judge summarised the evidence he heard in this way at [42]:

"Furthermore Dr McHugh's evidence was to the effect that it was obvious to use pigment wet cake in the plastic, ink and coating/paint applications for the 338 patent. He described the rationale and technical advantages of using wet cake and explained that it would be obvious to the skilled addressee that he could use PY 191 wet cake for these various applications. The only real challenge to this evidence was based on the fact that it had not been done before; there was no real challenge to the rationale and advantages of doing it. Dr Heron pointed out that there are problems manipulating wet cake due to its sticky nature but he accepted that a benefit was that there were no dust problems. My conclusion on Dr Heron's testimony on the point is that pigments are most commonly sold in dry form but there are occasions where wet cake is the more suitable medium. Professor Schmidt said that to his knowledge it was not possible successfully to use wet cake pigment in plastics without drying the pigment before use, but I was not satisfied he had any basis for his remarks (and he had no relevant experience)"

41. The judge then concluded that the likely explanation for there being no previous sales of PY 191 wet cake was that Clariant had chosen never to sell it in that form. He noted that Clariant had offered no explanation why that was so and indicated that he might have inferred that had it been for a good technical reason then Clariant would have explained what it was. However, he considered that he did not need to make that inference because he found Dr McHugh's evidence persuasive and was satisfied by the reasons he gave for his opinion. It followed that it was obvious to use PY 191 containing 100% of the beta polymorph for pigmenting polymers.
42. Mr Antony Watson QC, who has appeared with Mr Mitcheson for Clariant, sought to persuade us that the judge fell into error in making this finding because it had no evidential basis. Moreover, he continued, the judge wrongly failed to attach any or at least any sufficient weight to the fact that PY 191 had never been sold in wet cake form.
43. We were then taken to the evidence given on this issue on behalf of Clariant. Professor Schmidt said in his witness statement that, to the best of his knowledge, it was not possible to use wet cake for pigmenting polymers. Perhaps more importantly, Dr Heron said this in his first report:

"Example 1 provides for the drying of the press cake in order to make PY 191 and I can see no reason why a person skilled in the art following the example would carry out all the other steps of the example, but not the very last step. As I have said in paragraph 13 it is normal practice for multi-purpose pigments such as this to be sold as dry powders and so there would be no reason not to dry the product. Having a wet product would be particularly problematic when used in non-aqueous processes such as pigmenting of polymers and powder coating materials. Indeed, as far as I am aware PY191 has never been sold in the wet cake form."

44. Mr Watson submitted that this evidence was not challenged and accordingly it was not open to Sudarshan to argue that wet cake could be used to pigment polymers or that it was obvious to do so. As for Dr McHugh, Mr Watson recognised that he suggested it would have been obvious to use PY 191 wet cake because other pigments are sometimes used in that form, but contended that this evidence had no probative value because he accepted in cross examination that he was not a polymer expert and did not know what would be suitable for incorporation into polymers and what would not.
45. I have to say I regard this as a less than complete picture of the evidence as a whole. It is convenient to begin, as did Mr Martin Howe QC, who has appeared with Mr Ward on behalf of Sudarshan, with the evidence of Mr Satyanarayan, who is known as Mr Satya. He has worked with Sudarshan for over 34 years and is now Senior Technical General Manager and responsible for its research and development activities. He explained that Sudarshan sells laked azo pigments to customers in wet cake form for the purpose of pigmenting polymers. Now it is fair to say that much of this activity has occurred after the priority date but it establishes an important fact: there is no technical reason why PY 191 could not have been supplied in wet cake form for this purpose.
46. I come next to Dr McHugh. He explained that the use of pigment wet cake for plastics applications is described in the standard work Industrial Organic Pigments by Willy Herbst and Klaus Hunger. He continued with a detailed description of the rationale and technical advantages of using the pigment in this form. In brief summary, the process of azo pigment manufacture typically involves synthesis, isolation, drying and milling stages. Drying and milling are detrimental to the overall technical performance of the product as they tend to promote agglomeration of the primary particles which must be dispersed by the customer at additional cost. So an alternative is to use a flush paste. This involves the direct transfer of pigment from an aqueous phase to a non-aqueous phase and avoids the need for drying and milling altogether. The non-aqueous phase is typically composed of materials that are common to the application medium such as resins, polymers and oils. As Dr McHugh said, flushing of calcium laked azo pigments was well known and common practice at the priority date of the 338 patent.
47. In cross examination Dr McHugh maintained that the use of wet cake was perfectly standard although he accepted that it was less common than the use of dry powders and he had no evidence it was ever done with PY 191. When asked if he had any hands on knowledge as to why that was so, he accepted that he did not, it being an area outside his expertise, and that he was not a plastics maker. By this I do not understand him to have been disclaiming any knowledge of the use of wet cake. To the contrary, he made clear that he had personal experience of such use. Nor do I understand him to have been disclaiming expertise in the field of pigment manufacture, or the forms in which and purposes for which pigment manufacturers routinely sold their products. Here again he made it quite clear that this is precisely where his expertise lay. Indeed he gave a good deal of evidence to this effect. What he was disclaiming was particular expertise as a manufacturer of plastics, and personal knowledge as to why PY 191 had not been used in wet cake form. Notable too is that Dr McHugh was not challenged in relation to his exposition of the rationale for using wet cake either generally or specifically in relation to PY 191. I am therefore entirely satisfied the judge was entitled to rely upon Dr McHugh's evidence in the way that he did.
48. As for Dr Heron, his evidence was, I think, broadly consistent with that of Dr McHugh. In his reports, he too said that pigments were usually sold as dry powders. However, he accepted that flush pastes had their advantages although he fairly said that they also have disadvantages such as making it difficult to standardise colour properties and a susceptibility to attack by fungi and bacteria when stored. He also explained that manipulation of a wet cake can cause problems arising from its sticky nature but it has the benefit that no dust is created. Similarly, in cross examination he accepted that one way of shortening the overall process of pigment preparation was to make direct use of the wet cake. Overall, therefore, he pointed out that wet cake has both advantages and disadvantages but, most importantly, he did not disagree with Dr McHugh's evidence that the use of wet cake for pigmenting polymers was perfectly obvious and routine, albeit less common that the use of powders.
49. Professor Schmidt was the last of the witnesses upon whom our attention was focussed. I have referred to the evidence he gave in his witness statement. But there are two important points which must be borne in mind in considering that evidence. The first is that Professor Schmidt was not an expert witness. Secondly, as Sudarshan accepts, he is a highly skilled and distinguished crystallographer; but he is not an expert in the uses to which calcium laked azo pigments in their various forms may be put. As he himself volunteered in the course of his cross examination, he is not and has never been an application specialist.
50. Finally, I must deal with the submission: if it was obvious, why was it not done before? The fact something could have been but was not done before may be a powerful indicator of non obviousness. But all must depend upon the particular circumstances of the case. Here, as I have explained, Clariant had a monopoly over the use of PY 191 in a number of important jurisdictions but it chose to lead no technical evidence as to why it had not made sales of the pigment as wet cake. It may be the case that its facilities or those of its customers were not configured to deal with the pigment in that form, in which case the point would have no force at all. In the absence of any elucidation of the particular circumstances surrounding the production and sale of PY 191, the judge was entitled to say that this was a matter upon which he could place no reliance one way or the other.
51. In summary, I do not accept that there was no evidence upon which the judge could properly conclude that it was obvious to use the wet cake of PY 191 for pigmenting polymers. To the contrary, it seems to me that his finding had an ample evidential basis. I would therefore dismiss the appeal against the finding that proposed claim 12 was not allowable.

Was it obvious to water treat dry ground wet cake?

52. This issue follows naturally from the first, and it is potentially determinative of the appeal in relation to proposed claim 13. I have set out the history of this amendment earlier in this judgment and this is of some importance in considering the quantity and quality of the evidence available to the judge.
53. First of all the specification itself reads (at page 7, line 26):

"Depending on the desired field of application it may be sensible to subject the resulting pigment to mechanical fine division. This fine division may be carried out by wet or dry grinding or by kneading. The grinding or kneading operation may be followed by treatment with a solvent, with water or with a solvent/water mixture in order to convert the pigment into a useful form".

54. Second, the matters here described were put to Dr Heron in cross examination and he accepted they were all routine. The following passage of the transcript is particularly illuminating:

"Q. So how would water treatment be carried out on a dry-milled product?

A. It could, essentially, be dispersed in that pigment in water and heating it and then allowing it to cool and then you have opportunities for, we are back to almost pressed cake stage where you have water on the sample and you can change the smaller crystallites into agglomerates and potentially aggregates at that point, so you are increasing the particle size, perhaps, into a more desired range or a more refined range.

THE DEPUTY JUDGE. That sort of activity will be standard stuff. Standard optimisation.

A. I think so, yes. I think the grinding, at that step, may relate to -- I think Dr Schmidt related it to products coming off a dryer as being sausage-like or spaghetti-like. So, breaking those particles down and then solvent treatment afterwards which could be water or a mixed solvent. I do not think there is anything particularly unusual about that other than you have gone to the effort of drying a pigment and then you are going to wet it again to get a new -- but if that requires to be done to produce a product spec then that is what people will do.

MR HOWE. Indeed. We have discussed – you have mentioned a number of times – particle size and particle size distribution and crystallite size and form as important factors in the production of a usable pigment at the end of the day.

A. In adjusting the shade, yes."

55. In substance Dr Heron was here accepting that it would be standard practice to take wet cake, dry grind it in order to break up the sausage or spaghetti-like structures and then water treat it to produce agglomerates or aggregates within a desired size range. In the light of this evidence, the stage at which the proposed amendment was put forward and the concession made by Clariant at trial which I have recited at [23] above, it seems to me the judge cannot possibly be criticised for finding that it was obvious to use wet PY 191 which had earlier been dry ground for the purpose of pigmenting polymers. It being accepted that water treatment will inevitably convert PY 191 into its beta form, it necessarily follows that proposed claim 13 was obvious and not allowable, and the judge was right to so hold.

Was the judge right to hold that proposed claims 12 and 13 would result in the specification disclosing additional matter?

56. As I have explained, proposed claim 12 is directed to the use of the polymer as claimed in any of claims 6 to 10 for pigmenting polymers and powder coating materials.
57. Similarly, proposed claim 13 is directed to the use of PY 191 as claimed in any of claims 6 to 10 for pigmenting polymers and powder coating materials but now with the further limitation that it must have been subjected to mechanical fine division by dry grinding.
58. The judge held that both of these proposed claims were not allowable because they would result in the specification disclosing additional matter.
59. The judge does not explain how he reached this conclusion in relation to proposed claim 12 but it seems he must have accepted Sudarshan's submission that the amendment is impermissible because it selects two applications from a class of applications, and suggests that the use of the claimed PY 191 for these particular applications has technical significance.
60. This emerges most clearly from the way the judge addressed proposed claim 13 which is directed to the use of the claimed PY 191 for the same two applications and also requires the PY 191 to have been subjected to mechanical fine division by dry grinding. Sudarshan submitted that this proposed claim discloses for the first time the use of fine division in combination with these two particular applications, and further suggests for the first time that the use of fine division in combination with these two applications has technical significance.
61. The judge accepted these submissions at [55]:

"…. there is no disclosure in the original specification either that it is inventive or that there is technical significance in the use of PY 191 in polymers or powder coating materials in particular, or in such use when the pigment has been subject to mechanical fine division by dry grinding in particular. Accordingly, and following the approach set out in the above mentioned citations, the effect of the amendment is to add subject matter relevant to the invention. Therefore, the amendment is contrary to s.76 of the Act and is not allowable."

62. The correct approach to an allegation of added matter of this kind has been considered in a number of cases, and most recently by this court in Nokia OKJ (Nokia Corporation) v IPCom GMBH [2012] EWCA Civ 567. The crucial question is whether a skilled person would, on looking at the amended specification, learn anything about the invention which he could not learn from the application as filed. The comparison is a strict one in the sense that subject matter will be added unless it is clearly and unambiguously disclosed in the application either explicitly or implicitly. Conversely, it has always been permissible to add a feature to a claim if it simply excludes protection for part of the subject matter of the claimed invention and does so in a manner which does not add to the disclosure.
63. In my judgment the application of the test in the present case produces a clear answer. So far as proposed claim 12 is concerned, this involves limiting the claim to the use of the claimed PY 191 to two uses which were specifically identified in original claim 8. This does not add to the disclosure. The uses disclosed in that broader claim are separate and distinct and to exclude some of them from the scope of protection does not teach the skilled person anything new about the invention.
64. Turning to proposed claim 13, this adds the further limitation to proposed claim 12 that the PY 191 must have been subjected to mechanical fine division by fine grinding. But this is taught in the body of the application as filed which says:

"Depending on the desired field of application it may be sensible to subject the resulting pigment to mechanical fine division. This fine division may be carried out by wet or dry grinding or by kneading. The grinding or kneading operation may be followed by treatment with a solvent, with water, or with a solvent/water mixture in order to convert the pigment into a useful form."

65. Mr Watson says, and I agree, that it is implicit if not explicit that PY 191 which has been treated in this way can be used for any of the disclosed applications. Once again, therefore, the skilled person is not taught anything new about the invention. Proposed claim 13 limits the scope of the claimed monopoly to the use of PY 191 which has been treated in a particular way, and it does so in a manner which is taught clearly and unambiguously in the application as filed.
66. Faced with the logic of this argument, Mr Howe sought to deploy a further argument at the appeal hearing, namely that the application as filed does not disclose the levels of beta polymorph in the pigment mixture after the process of dry grinding, so there is, for example, no disclosure of such a dry ground mixture containing at least 90% of the beta polymorph. This, it seems to me, is a point which Sudarshan could and should have taken at the trial if it wished to take it at all, and it is one to which Clariant may well have wished to direct evidence. In my judgment, it is too late to take it now.
67. It follows that the judge fell into error in concluding that proposed claims 12 and 13 were not allowable because they would result in the specification disclosing additional matter. This error does not, however, carry Clariant home because, for the reasons I have given, the claims are obvious over the wet cake.

Was the proposed disclaimer allowable?

68. This proposed amendment is of considerable importance because it seeks to exclude from the scope of protection any PY 191 in wet cake form, and the use of such wet cake for pigmenting polymers, powder coating materials and the like. Mr Howe characterised this as a yet further attempt by Clariant to wriggle away from the consequences of the stark fact that, contrary to the assertions made in the 338 patent, the beta polymorph was not a novel product. Graphic though this characterisation may be, any legitimate objection to the proposal must, in my judgment, find a more principled foundation.
69. I must begin, however, by saying a little more about the teaching of the 338 specification. At the outset it explains that the invention relates to a novel crystal polymorph of PY 191, to its preparation and to its use as a pigment. On page 1, line 15 it says that PY 191 is described in EP-A-0 361 431 (the 517 patent). A little later, at line 25, it continues:

"To date only one crystal polymorph of PY 191 has been disclosed. It is referred to below as the [alpha] polymorph and has the following characteristics lines in the X-ray powder diagram …"

70. This is a clear reference back to the form of PY 191 disclosed in the 517 patent. Further teaching about how the alpha polymorph is made appears on page 2, line 31:

"The [alpha] phase is formed in the synthesis of PY 191 if the coupling of the disulfo acid with the sulfonated pyrazolone is conducted in accordance with the details given in [the 517 patent]."

71. The specification then explains (from page 3, line 3) that a new crystal form is formed if the alpha polymorph is subjected to a particular solvent treatment:

"It has now been surprisingly found that by treating PY 191 with certain solvents, a new crystal polymorph is formed. The new polymorph is called the [beta] form. It features the following characteristic lines …"

72. There follows a long description of the qualities of the new beta polymorph, how it may be subjected to different processing techniques and the uses to which it may be put. Then (from page 10, line 15) a comparative example is given which is said to be in accordance with the 517 patent, and is in fact example 1 of that patent. This concludes with the statement that the "presscake" is dried at 120°C and:

"This gives Pigment Yellow 191 in the alpha polymorph."

73. The specification continues with a description of three examples of the invention, each of which involves the recrystallisation of the alpha polymorph from a solvent, in the first case N-methylpyrrolidone; in the second, dimethylsulfoxide; and in the third, dimethylacetamide.
74. The clear teaching of the application for the 338 patent as filed is, therefore, that the alpha and beta forms are true polymorphs; that the alpha form is produced by the process described in the 517 patent; and that the beta polymorph is produced by treating the alpha polymorph with particular solvents.
75. I turn then to the proposed disclaimer which, for convenience, I set out again:

"EP 0 361 431 [the 517 patent] discloses the production of a pigment of formula (1) as an aqueous suspension which is then filtered washed and dried. We disclaim the said pigment as an aqueous suspension or in a form which has not been dried."

76. There are two possible objections to it: first, that it and the proposed claims which follow are not clear and concise as required by s.14(5)(b) of the 1977 Act; and second, that it would result in the specification disclosing additional matter. I will take them in turn.
77. The requirement that the claims must be clear and concise is a fundamental one and necessary to provide legal certainty. The purpose of the claims is to define the monopoly and so they must be drawn in terms which are sufficiently clear for the skilled addressee to determine their scope. That does not mean that it is never permissible to use relative terms in a claim. Such terms may be used provided that their meaning is clear to the skilled addressee who reads them with the benefit of the common general knowledge in the field of the invention.
78. Sudarshan argued at trial that this disclaimer is not clear because the skilled person would not know what is meant by the expression "in a form which has not been dried". The judge appears to have accepted that submission, observing that "there are problems with this disclaimer" and that it created an "ambiguity".
79. Upon this appeal Mr Watson submitted the judge fell into error on this issue and that, in context and upon reference back to the 517 patent, it is quite clear that what is disclaimed is the wet cake form and that by the word "dried" the skilled person would understand that what is covered is oven drying to a low residual water content. Typically, said Mr Watson, this means drying to a constant weight. Furthermore, Mr Watson continued, this disclaimer will create no difficulty in practice because the evidence established that the skilled person will use either an oven baked product or the wet cake.
80. I am unable to accept these submissions. There was a good deal of evidence before the judge that those in this field will not always dry a pigment such as PY 191 at the same temperature or for the same length of time. Indeed, they may well choose to dry the pigment only to the extent they consider necessary to allow it to be handled and processed. Moreover, the drying conditions adopted are likely to have a significant effect on the polymorphic composition of the finished product. Thus, as Mr Watson accepted, Clariant's own HGR product has always contained 10-15% of the beta polymorph. Similarly, as Mr Watson also recognised, Sudafast Yellow 132 produced by Sudarshan using its original process and a drying temperature of 80-85°C contained around 40% of the beta polymorph. Further, the Sudafast Yellow 132 produced by Sudarshan using its new process and an elevated drying temperature of 120°C still contained 16-20% of the beta polymorph.
81. Moreover, this was entirely consistent with the evidence of the experts. As for drying conditions, Dr McHugh explained in his first report that a drying temperature of 120°C was unusually high and that the skilled person would be reluctant to use such a temperature unless there was a good reason for it. He thought no such reason was given in the 517 patent. So the skilled person would carry out investigations and try and reduce the temperature to make the process as economical as possible.
82. Turning to the impact of different drying conditions on polymorphic composition, Dr Cockcroft, an expert in crystallography who gave evidence on behalf of Sudarshan, explained in his first report that air drying wet cake under ambient conditions produced a product which was essentially entirely in the beta form, a result which was also obtained by Professor Schmidt when he vacuum dried wet cake under ambient conditions.
83. I am therefore satisfied the judge had ample evidence before him to justify the conclusion that the disclaimer would result in a monopoly of ambiguous and uncertain scope.
84. That brings me to the second objection, that the addition of this disclaimer would result in the specification disclosing additional matter. I have already explained the general approach to be adopted in considering an allegation that matter has been added contrary to Article 123(2) EPC. It might be thought that the addition of a reference to prior art in order to disclaim it must inevitably add to the disclosure. But that is not necessarily so, as the Enlarged Board of Appeal explained in decisions G 1/03 and G 2/03. In summary, and so far as relevant to this case, such a disclaimer may be allowed to restore novelty against an accidental anticipation, that is to say, an anticipation which is so unrelated to and remote from the claimed invention that the person skilled in the art would never have taken it into account when making the invention.
85. The issue came before this court in LG Philips LCD Co Ltd v Tatung (UK) Ltd and others [2006] EWCA Civ 1774. Neuberger LJ, with whom the other members of this court agreed, elaborated the approach to be adopted in these terms at [33]-[34]:

"33. The law on added matter was considered again by the Enlarged Board of Appeal, in a case where the amendment involved a disclaimer narrowing the claim, in G 1/03 PPG/Disclaimer [2004] EPOR 331. The effect of that decision is that a specific disclaimer does not add matter (contrary to Article 123(2) of the European Patent Convention – equivalent to section 76), if it is inserted into a claim to avoid an "accidental" anticipation, but it does add matter if it is inserted to avoid a "non-accidental" anticipation – see Part 2 of the decision. An accidental anticipation involves a "disclosure … belong[ing] to a remote technological field or [one whose] subject-matter suggested it would not help to solve the problem [addressed by the patent in question]". In other words, "the disclosure has to be completely irrelevant for assessing the inventive step" – see paragraph 37.

34. A little later in the same paragraph, the Enlarged Board identified an accidental anticipation in slightly different words, but to much the same effect, namely that "the disclosure in question must be so unrelated and remote that the person skilled in the art would never have taken it into consideration when working on the invention". In conclusion on this topic, in paragraph 44, the Enlarged Board said that:

"When an anticipation is taken as accidental, this means that it appears from the outset that the anticipation has nothing to do with the invention. Only if that is established, can the disclaimer be allowed."

86. More recently, the issue was again considered by this court in Napp Pharmaceutical Holdings Ltd v Ratiopharm GmbH [2009] EWCA Civ 252, [2009] RPC 18. Jacob LJ, giving the judgment of the court, emphasised that there is, at the end of the day, one crucial question, namely whether the amendment results in the specification disclosing additional subject matter relevant to the invention. If a disclaimer simply excludes subject matter from protection, then it will not offend against Article 123(2) EPC.
87. Application of these principles in this case can, in my judgment, produce only one result. The disclaimer does add to the technical disclosure because it teaches the skilled person that the wet cake described in both the 517 patent and the 338 patents comprises or consists of the supposedly new beta polymorph. It also teaches the skilled person that there is some kind of difference between the wet and dry forms of PY 191. This teaching is not to be found in the application for either patent as filed. Further, it is a teaching which is contrary to the disclosure of the 338 patent that the alpha form of PY 191 is the result of the method of synthesis described in the 517 patent and that the beta form is the result of treating the alpha form with particular solvents. Moreover, far from the disclosure of the 517 patent being in a remote technical field or something which is irrelevant for assessing inventive step, it is the most directly relevant earlier disclosure and is described in the 338 patent as the starting point from which the invention has been made.
88. Accordingly, I believe the judge was right to reject the proposed disclaimer.

Did the judge fall into error in relation to the Gillette defence?

89. It follows from all of the foregoing that the judge was right to find the 338 patent invalid and to refuse the proposed amendments. It also follows that it is not strictly necessary to deal with the other attack on the patent, namely that both the old and the new processes used by Sudarshan to make its Sudafast Yellow 132 pigment are no more than routine implementations of the prior art and so any patent covering them or any product made using them must be obvious. Nevertheless, it formed a major part of the trial and appeal and so it is right that I should address it.
90. Mr Watson submitted that, in finding in favour of Sudarshan on this issue, the judge fell into error in the following three respects. First, Sudarshan did not seek simply to carry out Example 1 of the 517 patent to produce a yellow pigment; rather it combined the teaching of the example together with certain characteristics measured from commercial PY 191 made after the priority date in order to develop and refine a process which would produce a comparable pigment. This was an impermissible mosaic of two separate pieces of art.
91. Secondly, driven by a desire to match the commercial product, Sudarshan added in extra ingredients and went through some 70 different iterations in order to arrive at the old process. To establish the new process Sudarshan went through a further 50 iterations. The results of the two development processes were not obvious ways of carrying out example 1 of the 517 patent, nor were they meant to be.
92. Thirdly, the aim of both the old and new processes was to produce a dry product, and on the evidence this would have been in the alpha form. It was not obvious either inadequately to dry the product or handle it after drying in such a way that it was able to hydrate to high levels of the beta polymorph. Further, it was never established that the sample of the old process that came to the United Kingdom and gave rise to the claim for infringement was ever made by the old process, or at least any process that was merely an obvious modification of the 517 patent.
93. Mr Watson developed his first submission by referring to Mr Satya's evidence as to how Sudarshan came to make Sudafast Yellow 132. In 2003 Reddy Polymers of Hyderabad asked a member of Sudarshan's sales team whether Sudarshan could develop PY 191. The sales team asked Mr Satya the same question and gave to him a sample of the pigment which had been purchased by them from a supplier.
94. Sudarshan then carried out a patent search on PY 191 which led it to the 517 patent and its equivalents. It also found the 338 patent and its equivalents. But there was no patent in India and so Mr Satya was satisfied that it was permissible to make and sell PY 191 in that country. Sudarshan recognised that the 517 patent described a process for making PY 191 but did not set out all the details of every possible manufacturing variable. So, where the description was not prescriptive, Mr Satya followed what he considered to be standard procedures for making laked azo pigments, and optimised the various process parameters such as pH and laking temperature, and made appropriate use of various standard additives such as co-amines and co-couplers. As for the drying temperature, Mr Satya noted that Example 1 of the 517 patent gave a temperature of 120°C but considered this to be unusually high and, based upon Sudarshan's past experience, he thought a temperature of 80-85°C more appropriate. He also tried a drying temperature of 120°C but considered this did not make any difference and so finalised the drying temperature at 80-85°C, which made the process cheaper and easier to operate. Importantly, and as an additional check on quality, Sudarshan obtained a sample of HGR and tested the qualities of the pigment it had prepared against this sample. It found them to be the same.
95. Sudarshan developed its commercial process by scaling up the process it had developed in its laboratory. Having done so, it produced what is known as a master sample which it again tested against HGR. It then sent a sample of the product to Reddy Polymers which approved it. The pigment was branded Sudafast Yellow 132 and was placed on the market.
96. So, Mr Watson submitted, it is clear that Sudarshan did not work solely from the 517 patent. It combined the teaching of that patent with a commercial product which had been sold into the market some three years after the priority date of the 338 patent and then tested the new pigment it prepared against HGR. This involved combining different pieces of art, some of which post-dated the priority date.
97. Taking the various aspects of this submission in turn, it was never suggested at trial that the product handed to Sudarshan by Reddy Polymers was a PY 191 product made by any company other than Clariant. Both parties assumed it was HGR and it is now too late to suggest otherwise. Further, there was never any evidence to suggest that the HGR produced after the priority date was any different from the HGR produced before that date. Indeed Mr Satya gave unchallenged evidence that to his knowledge HGR had been on the market since 1992. It was never suggested to him that it had changed in that time. Nor did Clariant lead any evidence of any change in its production process.
98. So I come to the real point: was it permissible to combine a commercial sample of HGR with the teaching of the 517 patent? I believe it was. Both were relied upon as prior art and the judge heard evidence from Dr Heron and Dr McHugh to the effect that anyone seeking to copy HGR would carry out a patent search and that this would uncover the 517 patent. So, on the evidence, it was indeed obvious to combine the two.
99. Mr Watson's second submission was founded upon Mr Satya's evidence that Sudarshan's old and new processes involved significant development work and very many iterations. Mr Watson accepted that there may well be obvious process variants of the 517 disclosure but, he submitted, unless the precise way chosen by Sudarshan to implement that teaching is itself shown to be obvious then this attack on the patent must fail. As to that, Mr Watson argued that the expert evidence was all one way, and he took us to this passage in the cross examination of Dr McHugh:

Q. You would not suggest that the actual process in its full details, temperatures, pH, the choice of co-couplers, the further addition of rosin, you would not suggest that that is an obvious example of carrying out example 1. I mean, if you did example 1, you would not come up with that process, would you?

A. If someone asked me to repeat example 1, I would not come up with that process. If someone asked me to produce a commercially viable pigment, that is the kind of process I would probably come up with, based on what is in example 1.

Q. It is the kind, I understand that, that you would optimise example 1. But whether anybody would optimise example 1 to arrive at the Sudarshan old process, you simply do not know. Mr. Satya spent a lot, 70 iterations?

A. Of course, I could not tell you what anyone is going to optimise that process, all that I am saying is that you would always take that patent example and, based on the skeleton of that particular patent, develop your own product with your own knowledge and your own expertise.

THE DEPUTY JUDGE: Is this the sort of thing you would come up with, or do you think this is pretty extraordinary?

A. No. This is normal process development, normal product development. These are using, effectively, a toolbox of materials, a toolbox of process variables to arrive at something that is commercially sensible.

MR. WATSON: I accept all that, but I think what my Lord is after, or certainly I am after anyhow, is this, that, yes, you are using the toolbox of putting in some co-couplers, choosing your co-coupler, your co-amine, using the tool box of putting in a rosin. You are using your tool box to adjust temperatures. You are using your tool box to adjust pHs. But the final recipe is something which is going to be almost unique. You cannot say that another worker, starting with example 1, would arrive at that recipe.

A. No, I could not say that.

100. Founding himself particularly on the last question and answer, Mr Watson submitted that Dr McHugh accepted that Sudarshan's process is not one he would have arrived at. Therefore, Mr Watson continued, it cannot have been obvious.
101. It seems to me that there are a number of matters which must be borne in mine in assessing this submission. The first is that the 517 patent does not describe a commercial process; it describes a laboratory experiment and an incomplete one at that. A skilled person seeking to put it into practice must fill in a number of process parameters.
102. Second, optimisation of pigment production was, as the judge found, a common and necessary industry practice. So, for example, it was common general knowledge to add co-couplers and co-amines to improve colouristic properties; to use diazotisation and coupling processes; to adopt finishing and after treatments; and to use a wide range of drying temperatures.
103. Third, it was never suggested in evidence that 60-70 iterations was other than routine. To the contrary, Mr Satya gave evidence that Sudarshan merely undertook standard development work and, in doing so, had to optimise many different variables. Dr Heron gave a list of process conditions that would be routinely altered. And Dr McHugh considered both the old and the new processes were no more than ordinary and obvious implementations of the 517 patent teaching.
104. Against this background, the passage of Dr McHugh's cross examination upon which Mr Watson relied can be seen to support precisely the opposite conclusion to that for which he contended. Dr McHugh made clear that Sudarshan had simply undertaken a process of entirely normal product development. It involved the use of a toolbox of materials and techniques to arrive at a commercially sensible outcome. There was nothing about it which was in any way out of the ordinary and it did not produce a surprising result. The fact that Dr McHugh accepted he would not have come up with the precise combination of variables settled upon by Sudarshan gets Mr Watson nowhere. It simply illustrates there are many obvious ways to implement the teaching of the 517 patent.
105. As a second tier point, Mr Watson also argued that the specification for Sudarshan's old process did not include any criterion for residual moisture, and this too rendered it non obvious. But the answer to this is that there was no reason why it should. All Sudarshan was seeking to do was to produce a product which was dry enough to do the job. And that is precisely what it did.
106. That brings me to Mr Watson's third submission. This related to the sample supplied by Sudarshan to Colormatrix in the United Kingdom, and it was this supply which gave rise to the infringement claim. The sample contained over 80% of the beta polymorph; indeed one experiment suggested its beta polymorph content was over 90%.
107. Mr Watson developed his argument as follows. Sudarshan had not established that this sample was made by the old process. It produced no batch records relating to it; nor did it call a witness who could deal with it. Further, it is inconsistent with the samples of the old process which were tested and which contained, at most, 40%-45% of the beta polymorph. So it must have been produced by a completely different process, or by the old process which was not carried out properly. In either event the process which was actually used has not been shown to be obvious.
108. Mr Watson supported this submission by referring us to the evidence concerning Sudarshan's reference samples produced by its old and new processes, and to Sudarshan's own experiments on the alpha polymorph which show that when it is exposed to a moist atmosphere at ambient temperatures, it takes up only a very small amount of water leading to the production of only trace quantities of the beta polymorph. In these circumstances, Mr Watson continued, the judge had no basis for holding that the process used to make the Colormatrix sample was obvious; indeed the judge appears to have overlooked the fact that this sample was quite different from the products of the old and new processes as demonstrated by the reference samples.
109. Mr Howe responded that it is not fair to criticise Sudarshan for a failure to produce batch records in the light of the procedural history of the allegation. He pointed out that it was only shortly before the trial that Clariant disclosed that the sample complained of had been received from Colormatrix in September 2009 and served a notice of experiments revealing its very high beta polymorph content. Mr Satya then made enquiries and produced in the course of his evidence in chief the documents which Sudarshan still had relating to it. These showed it originated from a blend of batches which were manufactured on or about 31 August 2008 and that it was then put into storage in Sudarshan's warehouse. Mr Satya was asked which process was used to make those batches and he said it was the old process. In the course of his cross-examination he also said that Sudarshan's normal practice was to discard its batch records after two years, and that the records relating to this sample had been discarded in accordance with that normal practice a considerable time before the trial.
110. Mr Howe also submitted that it is not right to say that Sudafast Yellow 132 only contains the alpha polymorph at the end of the drying step. In that connection he referred us to Dr Cockroft's evidence that drying the product on an industrial scale may not result in its complete conversion to the alpha polymorph. Further, as I have explained, the fact that a product may contain the beta polymorph does not mean that it has been incompletely or not properly dried.
111. Mr Howe's points concerning the absence of batch records and the composition of Sudafast Yellow at the end of the drying stage are well made. Nevertheless, the fact remains that the level of beta polymorph found in the Colormatrix sample was considerably higher than that found in any other sample made by the old process. Specifically, the old process master lot contained 30% of the beta polymorph; and the old process commercial lot contained 44% of the beta polymorph. I recognise that Mr Satya gave evidence that the batches from which the Colormatrix sample came were made during the monsoon season. But this aside, Sudarshan could not and did not attempt to put forward a specific explanation for the level of beta polymorph found in it. This, it seems to me, leaves this court in a somewhat unsatisfactory position. Mr Watson forcefully contended that the burden lies upon Sudarshan to show that the process actually used to make the Colormatrix sample was no more than an obvious modification of the prior art. Mr Howe responded that Mr Satya's evidence that the sample was made by the old process was never properly challenged. The judge did not deal with this issue in any detail in his judgment and since it is not necessary for me to express a final conclusion upon it in order to decide this appeal, I prefer not to do so.
112. Stepping back, the broad merit of the Gillette defence cannot be denied, however. I am entirely satisfied that, subject to what I might call the Colormatrix sample outlier, the old and new processes used by Sudarshan to make Sudafast Yellow 132 are routine and obvious implementations of the teaching of the 517 patent and that the judge was right to so hold. This finding is fatal to proposed claims 6 and 7 and all claims dependent on them.

Threats

113. The claim for threats is founded on the letter written by Clariant to Sudarshan on 25 November 2009 which said:

"Certainly you will not be interested in continuously effecting a patent infringement, which may involve the most serious penalties to the infringer and his customers."

114. As I have indicated, it gives rise to three issues. The first is whether the threat was justified because the activities and threatened activities of Sudarshan and its customers did amount to an infringement of a valid claim. In light of my findings, I can dispose of this issue straight away. The patent is, in my judgment, wholly invalid and so Clariant cannot prevail on this point. The second is whether the letter constituted an actionable threat to sue Sudarshan's customers even though it was only communicated to Sudarshan. The third is whether Clariant has a defence that, at the time it made the threat, it did not know and had no reason to suspect that the patent was invalid. I will address these two remaining issues in turn. But at the outset it is helpful to have the relevant statutory provision in mind.
115. Section 70 of the 1977 reads, so far as material:

"70. Remedy for groundless threats of infringement proceedings.

(1) Where a person (whether or not the proprietor of, or entitled to any right in, a patent) by circulars, advertisements or otherwise threatens another person with proceedings for any infringement of a patent, a person aggrieved by the threats (whether or not he is the person to whom the threats are made) may, subject to subsection (4) below, bring proceedings in the court against the person making the threats, claiming any relief mentioned in subsection (3) below.

(2) In any such proceedings the claimant or pursuer shall, subject to subsection (2A) below, be entitled to the relief claimed if he proves that the threats were so made and satisfies the court that he is a person aggrieved by them.

(2A) If the defendant or defender proves that the acts in respect of which proceedings were threatened constitute or, if done, would constitute an infringement of a patent –

(a) the claimant or pursuer shall be entitled to the relief claimed only if he shows that the patent alleged to be infringed is invalid in a relevant respect;

(b) even if the claimant or pursuer does show that the patent is invalid in a relevant respect, he shall not be entitled to the relief claimed if the defendant or defender proves that at the time or making the threats he did not know, and had no reason to suspect, that the patent was invalid in that respect.

(3) The said relief is:

(a) a declaration or declarator to the effect that the threats are unjustifiable;

(b) an injunction or interdict against the continuance of the threats; and

(c) damages in respect of any loss which the claimant or pursuer has sustained by the threats.

(4) Proceedings may not be brought under this section for-

(a) a threat to bring proceedings for an infringement alleged to consist of making or importing a product for disposal or of using a process, or

(b) a threat, made to a person who has made or imported a product for disposal or used a process, to bring proceedings for an infringement alleged to consist of doing anything else in relation to that product or process…."

Actionable threat to sue Sudarshan's customers?

116. The letter undoubtedly constituted a threat to sue Sudarshan. But, as both parties accepted, this is not actionable because Sudarshan is the maker and importer of the product complained of, and s.70(4) makes clear that proceedings may not be brought for a threat in respect of its activities. The question to which this aspect of the appeal gives rise is whether the letter also constituted an actionable threat to sue Sudarshan's customers. Sudarshan says it did because it conveyed to Sudarshan that Clariant intended to sue its customers. Clariant says it did not because a threat must be communicated to the person threatened. Further, the scheme of the 1977 Act is to permit threats to be made to the primary infringer, and that is just what happened here: the threat was communicated to Sudarshan.
117. The question whether an actionable threat must be communicated to the person threatened arose very many years ago in Skinner & Co v Perry (1893) 10 RPC 1. The claim in that case was made under s.32 of the Patents, Designs and Trade Marks Act of 1883 which provided broad relief against unjustified threats of the kind referred to in s.70(1) of the 1977 Act unless the person making the threats commenced and prosecuted with due diligence an action for infringement. Skinner complained about a letter written by Perry to a third party, the London Stereoscopic Company, which suggested that a camera made by Skinner was an infringement of its patent and intimated that proceedings would be taken against Skinner. After referring to the language of the section including, in particular, the words "by circulars, advertisements, or otherwise", Bowen LJ said at page 7:

"… But using language in its ordinary sense, it is difficult to see that an intimation ceases to be a threat because it is addressed to a third person in answer to an enquiry, or because it is addressed to the person himself. We are not dealing with libel or questions of publication – we are dealing with threats. If I threaten a man that I will bring an action against him, I threaten him none the less because I address that intimation to himself, and I threaten him none the less because I address the intimation to a third person."

118. Similarly, Smith LJ at observed at page 8:

"…What is said by Lord Justice Bowen, I entirely agree with; if we had not had in this section circulars or advertisements, a threat to a third person by a letter might have been argued not to have been a threat within the section…"

119. This provision was re-enacted with only minor modification in s.36 of the Patents and Designs Act 1907. That section was then itself the subject of minor amendment by the Patents and Designs Act 1919, and later more substantial amendment by the Patents and Designs Act 1932 to provide a saving if the defendant proved that the acts in respect of which proceedings were threatened constituted or, if done, would have constituted an infringement of a claim of the patent not shown by the plaintiff to be invalid.
120. It was in the context of a claim under s.36 of the 1907 Act as so amended that the issue of a threat made to a third person came to be considered again by Romer J in John Summers & Sons Ld v The Cold Metal Process Coy (1948) 65 RPC 75. He said at page 97:

"The argument is founded upon the words which appear at the beginning of the section, namely, "Where any person by circular, advertisements, or otherwise threatens any person "with an action for infringement of patent," etc. The question which would appear to arise is, do the words, "threatens any person" mean only "communicates a threat to any person," or do they also include the meaning "expresses a threat in relation to any person"? Although in the absence of a context the words would more usually, I think, bear the first of these meanings, the second is not necessarily inapt – unless one assumes that the threat is made in such circumstances that it could not possibly come to the knowledge of the person threatened.

Hitler undoubtedly threatened Britain with invasion in 1940, but it is not known that he wrote to Mr. Churchill acquainting him of his intention to do so. Be that as it may, there is in Sec. 36 a context and one which negatives any such assumption as I have mentioned. The section envisages that the threats may be expressed through the medium of "Circulars, advertisements or otherwise." The assumption to which I have referred cannot be reconciled with the conception of publicity which is involved in such words as "circulars" and "advertisements". But, further, these words and particularly the word "advertisement" point, in my judgment, with some force to the conclusion that a person can be "threatened" for the purposes of Sec. 36 by documents sent or published to persons other than himself. It may be that it was to some extent with reference to this context that Bowen, L.J., in Skinner & Coy v Perry (supra) said (at 10 R.P.C. p. 7 and he was speaking, of course, of the threats section):

"If I threaten a man that I will bring an action against him, I threaten him none the less "because I address that intimation to himself, and I threaten him none the less because I address the intimation to a third person."

It is true that the letter to which the Lord Justice was referring in making that observation was in some sense at threat to the addressee as well as being a threat to the plaintiff in the subsequent threats action. Nevertheless the observation, directed, as it was, to the threats section appears to be in perfectly general terms and I respectfully agree with it.

Fortified then by the views of Bowen L.J., and mindful (though, I hope, not too mindful) of the motives of the Legislature to which I have already referred, I hold that the words in Sec. 36 "threatens any person" do not only mean "communicates a threat to any person" but include also the expression of a threat, by circulars, advertisements or otherwise, in relation to any person."

121. Section 36 of the 1907 Act (as amended) was then re-enacted in a re-arranged but otherwise substantially unchanged form in s.65 of the Patents Act 1949. There it remained until it was repealed by the 1977 Act. Throughout this time the approach taken by the courts to the definition of an actionable threat remained consistent. Thus in Bristol Myers Co. v Manon Freres Ltd [1973] RPC 836 Whitford J said at page 868:

The object of the section is to stop patentees using a threat of proceedings in such a way that the business of third parties in the manufacture and sale of alleged infringing articles is curtailed or hampered. If patentees want to stop alleged infringers they must do it by taking proceedings for infringement and not making threats. Nothing can be more mischievous than a threat, whether directed against a customer or a customer's supplier, that has the result of the cancellation of an order. Unless such a threat were justified, it should in my view be stopped. Of course, I have held that in this case such a threat was justified.

What we have here is in fact a case in which a person, the plaintiffs, threaten another person, though not to their face, with proceedings for an infringement of a letters patent. That resulted in the cancellation of an order, and, in my view, the defendants, though not directly threatened, could claim to be aggrieved by that threat; and, had it not been a justified threat, I think it would have been actionable.

122. Similarly, in Speedcranes Ltd v Thomson [1978] RPC 221, a decision of the Court of Session, Lord Emslie referred to John Summers and observed that a threat within the meaning of the section need not be made directly to the person or persons against whom proceedings for infringement would be directed.
123. Section 70 of the 1977 Act again provided a remedy against groundless threats in which the definition of a threat remained largely unchanged. However, it included (in subsection (4)) a proviso that it was permissible to make a threat to bring proceedings in respect of primary infringements, namely making or importing a product or using a process. The scope of the remedy was further restricted by the Patents Act 2004 with effect from 1 January 2005. This Act introduced a defence of innocence (s.70(2A)(b)) and extended the proviso by permitting a patentee to threaten a manufacturer or importer in respect of his own acts of disposal and the like (s.70(4)(b)).
124. Mr Watson accepted that the words of s.70(1) of the 1977 Act are similar to those of s.36 of the 1907 Act but submitted they now have to be considered in the context of s.70(4) which allows threats to a primary infringer. In this case, he continued, the threat made by Clariant was communicated only to Sudarshan, the primary infringer, and Sudarshan's customers were not threatened at all. Further, he argued, it is difficult to discern any policy which allows threats to the primary infringer but which does not allow the person making the threat to point out to the primary infringer that his customers may be liable; indeed, since an open dialogue is now permitted, there is no reason why customers should not be referred to.
125. Attractively though these submissions were presented, I am unable to accept them. Mr Watson is right to accept that the words of s.70(1) of the 1977 Act are similar to those of s.36 of the 1907 Act and, I would add, to those of the 1883 Act. Indeed, I consider that in all material respects they have remained unchanged since a threats provision was first enacted. Moreover, they have been the subject of consistent judicial interpretation since their consideration by this court in the Skinner case in 1893. As I have also sought to show, Parliament has had many opportunities to intervene but has chosen not to do so. In these circumstances it seems to me that Parliament must be taken tacitly to have approved of that interpretation. Can it be said that Parliament has nevertheless implicitly reversed the rulings of the courts by the introduction of s.70(4)? I do not believe it can. It seems to me that, set against background of a constant line of authority over so many years, it cannot be inferred that Parliament intended to modify in such an oblique manner the meaning of the words it has consistently chosen to use.
126. Further, I do not accept that it is impossible to discern any reason for not allowing the person making the threat to point out to the primary infringer that his customers may be liable. A threat to sue a manufacturer in respect of his own activities is one thing. In a culture which encourages parties to discuss their differences before going to court, that is something with which he may be expected to cope. But a threat to sue his customers is quite another. A manufacturer may be very fearful of the damage this would cause to his goodwill and customer connection and so prefer to cease manufacture and sale altogether. A patentee may of course be prepared to sue the customers and if he does so the manufacturer must take the consequences. But Parliament may well have thought it wrong that if the patentee is not prepared to do so, perhaps because he knows it will damage his reputation too, that he should nevertheless be permitted to secure a commercial advantage by telling the manufacturer that that is what he intends to do.
127. I come then to Clariant's defence that, at the time it made the threat, it did not know and had no reason to suspect that the patent was invalid. Here Clariant relied upon the evidence of Dr Hütter who was the patent agent responsible for drafting the 338 patent and was employed in Clariant's patent management department. He said, and the judge accepted, that at the time he wrote the letter he personally did not know and had no reason to suspect that the patent was invalid.
128. That was not the full extent of the knowledge within Clariant, however. Professor Schmidt was the inventor of the 338 patent and from March 1995 until March 2002 was head of its laboratory for crystal engineering and polymorphism studies. He described in his evidence in chief how he carried out an investigation into HGR and H2GR in 1998 and found that that the standard HGR product contained from 10-15% of the beta polymorph. Perhaps more importantly for present purposes, he explained in cross examination that in early 2000 he also found that vacuum dried wet cake produced by the process of the 517 patent contained 100% of the beta polymorph. Both of these experiments were recorded in the laboratory notebooks of the company before July 2000, the priority date of the 338 patent. I should also say that Professor Schmidt took away copies of the notebooks when he left Clariant in March 2002, but the company retained the originals.
129. There can be no doubt that the Professor Schmidt's knowledge would have given Dr Hütter and, through him, Clariant reason to believe that the 338 patent was invalid. But, submitted Mr Watson, Dr Hütter did not have that knowledge and since he was the relevant officer in Clariant at the relevant time, Clariant was entitled to rely on s.70(2A)(b) and the judge should have so held.
130. The approach to be adopted in deciding whether the knowledge of an agent is to be imputed to a company principal has been considered in many cases but for present purposes I need only refer to Meridian Global Funds Management Asia Ltd v Securities Commission [1995] 2 AC 500. In that case, Lord Hoffmann, in giving the advice of the Privy Council, said at page 507:

"The company's primary rules of attribution together with the general principles of agency, vicarious liability and so forth are usually sufficient to enable one to determine its rights and obligations. In exceptional cases, however, they will not provide an answer. This will be the case when a rule of law, either expressly or by implication, excludes attribution on the basis of the general principles of agency or vicarious liability. For example, a rule may be stated in language primarily applicable to a natural person and require some act or state of mind on the part of that person "himself", as opposed to his servants or agents. This is generally true of rules of the criminal law, which ordinarily impose liability only for the actus reus and means rea of the defendant himself. How is such a rule to be applied to a company?

One possibility is that the court may come to the conclusion that the rule was not intended to apply to companies at all; for example, a law which created an offence for which the only penalty was community service. Another possibility is that the court might interpret the law as meaning that it could apply to a company only on the basis of its primary rules of attribution, i.e. if the act giving rise to liability was specifically authorised by a resolution of the board or an unanimous agreement of the shareholders. But there will be many cases in which neither of these solutions is satisfactory; in which the court considers that the law was intended to apply to companies and that, although it excludes ordinary vicarious liability, insistence on the primary rules of attribution would in practice defeat that intention. In such a case, the court must fashion a special rule of attribution for the particular substantive rule. This is always a matter of interpretation: given that it was intended to apply to a company, how was it intended to apply? Whose act (or knowledge, or state of mind) was for this purpose intended to count as the act etc. of the company? One finds the answer to this question by applying the usual canons of interpretation, taking into account the language of the rules (if it is a statute) and its content and policy. "

131. How then are these principles to be applied to s.70(2)(b) of the 1977 Act? Plainly this provision applies to companies and to individuals but, in the case of a company, whose knowledge is to be attributed to it? I did not understand Mr Watson to contend that the provision should apply only on the basis of a company's primary rules of attribution, so that, for example, what matters is the state of mind of its board of directors. On the other hand, it also seems to me unlikely that Parliament intended that the knowledge of any agent acting within the scope of his authority or any employee acting in the scope of his employment should be imputed to the company. That would be to impose a wholly unreasonable obligation upon a corporate patent department to make wide ranging enquiries throughout the whole company before it seeks to enforce a patent.
132. So I consider this is one of those cases where we must look to see whose knowledge was for the purpose of this provision intended to count as the knowledge of the company. Where, as here, a company has secured the grant of a patent and then seeks to enforce it, to limit that knowledge to that of an uninformed patent attorney in the patent department who writes a threatening letter would, in my judgment, be too restrictive. A patent attorney only learns about the invention because it is reported by the inventor. The patent attorney then has to ascertain from the inventor which features distinguish the invention from the prior art and he has to exercise great skill in drafting a detailed description of the invention in precise legal terms. It seems to me that it would be wholly wrong for a company to take advantage of a breakdown in communication between the patent attorney and the inventor which results in the company securing a patent of unjustified scope. So if a company has applied for and secured the grant of a patent and then seeks to enforce it, the controlling mind of the company must include at least the mind of the properly informed patent attorney or, put another way, the minds of the patent attorney and the inventor.
133. In my judgment it follows that Clariant did have reason to believe that the 338 patent was invalid in a relevant respect and the judge was right to dismiss this defence.

Conclusion

134. For all the reasons I have given I would dismiss the appeal.

Floyd LJ:

135. I agree.

Moore-Bick LJ:

136. I also agree.