British American Tobacco Co Ltd v Philip Morris Limited

[1994] APO 64

official notice

decision of a delegate of the commissioner of patents

Application        :   No. 573088 in the name of BRITISH-AMERICAN TOBACCO COMPANY LIMITED

Title:   Improvements Relating to Smoking Articles

Action: Opposition under section 59 (Patents Act 1952) by PHILIP MORRIS LIMITED; hearing.

Decision:   Issued            .

Abstract:   Novelty and obviousness considered.  All the claims found to be novel, but obvious.  Opportunity to amend.

patents act 1990

decision of a delegate of the commissioner of patents

Re:Patent Application No. 573088 by BRITISH-AMERICAN TOBACCO COMPANY LIMITED and opposition thereto under section 59 of the Patents Act 1952 by PHILIP MORRIS LIMITED.

background

Patent application 573088 (the application) was filed as 57651/86 on 21 May 1986 by BRITISH-AMERICAN TOBACCO COMPANY LIMITED (the applicant), claiming a Convention priority of 24 May 1985.  Acceptance was advertised on 26 May 1988.

On 25 August 1988 a notice of opposition was lodged by ROTHMANS OF PALL MALL (AUSTRALIA) LIMITED (Rothmans); this was subsequently withdrawn after the service of evidence-in support and evidence-in answer.  On 23 November 1988 a notice of opposition was lodged by PHILIP MORRIS LIMITED (the opponent), together with an application for an extension of time in which to do so; this application was subsequently allowed.  Following a number of extensions, two of which were opposed, service of the evidence-in-support was completed on 26 October 1990.  Also after extensions, service of the evidence-in-answer was completed on 23 December 1992, and of the evidence-in-reply on 27 October 1993.  A hearing was set down in Canberra on 10 August 1994.

The applicant was represented at the hearing by D Catterns QC, assisted by Dr A Bennett of Counsel and Mr P Heathcote, patent attorney, of Shelston Waters.  The opponent was represented by Dr J McL Emmerson QC.

As acceptance of the application was advertised before commencement of the Patents Act 1990, the provisions of section 234(3) and regulation 23.3 of this Act apply.

The grounds of opposition are stated to be those specified in paragraphs (a) through (i) of section 59 of the Patents Act 1952, though the only grounds argued were those of novelty, obviousness, and section 40.

THE SPECIFICATION

The invention relates to the design characteristics of cigarettes, in which in this case an important consideration is the circumference of the cigarette.  In this decision I will adopt what appears to be generally accepted nomenclature in the industry.  Namely, a "conventional" cigarette is one which has a circumference (in millimetres) in the mid-twenties (say around 24-26mm), a "slim" cigarette has a circumference in the low twenties (say around 21-23mm), and an "ultra-slim" cigarette has a circumference lower than about 20mm.

I also note that another important aspect of the invention is what is known in the industry as the "mass static burn rate" of a cigarette.  This is measured in milligrams of tobacco consumed per minute (mg/min) as the cigarette smoulders under controlled conditions, and is herein referred to as "SBR".  The issue in part appears to be that if the SBR is too high too few "puffs" are likely to be realised to be commercially acceptable, and if too low the cigarette may extinguish during smouldering, again seen generally not to be commercially desirable.

Also, though not apparent from the specification, it emerges from the evidence that another factor in cigarette design is that, in Australia, excise is levied on the mass of tobacco.  There is then an economic interest in devising a cigarette which uses less tobacco, while being satisfying to the consumer.

This invention is concerned with ultra-slim cigarettes, and can be understood from the following extracts selected from the specification.

"It is a requisite of cigarettes of commercially acceptable quality that when lit but not being smoked, the coal continues to burn, at a slow rate, a condition usually referred to as smouldering.  A number of cigarette design parameters affect the smoulder rate, otherwise known as the free or static burn rate.  One of these is the air permeability of the cigarette paper; the lower the air permeability the lower is the smoulder rate.  Another factor relating to smoulder rate is the amount of tobacco which is consumed in unit time.  It has previously been thought that for an acceptable linear smoulder rate, a minimum amount of about 60mg of tobacco had to be consumed per minute to assure the maintenance in the coal of enough heat to sustain the coal in its smouldering condition.  This factor has heretofore been seen to make it necessary for a cigarette rod of the usual circular cross section to have a circumference of a least about 22mm if the cigarette was to smoulder in a reliable manner.

The present invention provides a cigarette comprising a tobacco rod which rod comprises a tobacco filler and a paper wrapper, the circumference of said rod being within a range of 10mm to 19mm, preferably 12.5mm to 19mm, the free burn rate of said rod being within a range of 25 to 50 mg/min, preferably being within a range of 25 to 45 mg/min, and more preferably within a range of 30 to 45 mg/min and the packing density of the rod being within a range of 150 mg/cc to 350 mg/cc, and preferably within a range of 200 mg/cc to 300 mg/cc.

The present invention is based on our finding that in point of fact cigarettes having circumferences of 19mm or less and free burn rates of 45-50 mg/min or less smoulder in a fully acceptable manner and that in addition such cigarettes exhibit a number of significant practical advantages.

There is no necessity in order to provide cigarettes according to the present invention which smoulder reliably between puffs to make compensatory adjustments to the cigarette paper air permeability or to provide the cigarette paper with unorthodox levels of burn rate promoting additives, or in fact to make compensatory adjustments in regard to any other parameters which affect smoulder rate.

Cigarettes according to the present invention utilise tobacco with increased efficiency.  Thus a cigarette may be provided according to the present invention which, compared with a cigarette of orthodox dimension, having, for example, a circumference of 25mm, comprises considerably less tobacco, 25% less for example, and yet which provides an equal at least number of puffs."

"For unit length of tobacco rod, the reduction in the cigarette paper requirement is, leaving the longitudinal lap seam out of account, directly proportional to rod diameter reduction, whereas the reduction in tobacco requirements is, assuming constant packing density, in proportion to the ratio of the squares of the diameters.  Thus in unit length of a cigarette according to the present invention the ratio of mass of cigarette paper to mass of tobacco is higher than that for a cigarette of conventional circumference at equal packing densities.  Thus in cigarettes according to the present invention, exceptionally firm ash formations are obtainable.

In that cigarettes may be provided according to the present invention which contains less tobacco than orthodox cigarettes at equal puff numbers, cigarettes according to the present invention exhibit lower smoke component mainstream and sidestream deliveries."

"By use of the present invention significant reductions in packaging material requirements per unit cigarette may be realised."

The specification concludes with eight claims, as follows.

"1. A cigarette which comprises a tobacco rod which rod comprises a tobacco filler and a paper wrapper, the circumference of said rod being within a range of 10mm to 19mm, the packing density of said tobacco filler being within the range of 150 mg/cc to 350 mg/cc and the free burn rate of said rod being within a range of 25 to 50 mg/min.

2. A cigarette as claimed in Claim 1, wherein said circumference is within a range of 12.5mm to 19mm.

3. A cigarette as claimed in Claim 1 or 2, wherein said free burn rate is within a range of 30 to 45 mg/min.

4. A cigarette as claimed in Claim 1,2 or 3, wherein the packing density of said tobacco filler is within a range of 200 mg/cc to 300 mg/cc.

5. A cigarette as claimed in any one of the preceding claims and further comprising a filter of the same transverse dimensions as said rod.

6. A cigarette as claimed in any one of the preceding claims giving 5 to 15 puffs at standard machine smoking conditions.

7. A cigarette as claimed in Claim 6 giving 5-10 puffs.

8. A cigarette substantially as hereinbefore described with reference to the drawing hereof.

EVIDENCE

All the declarations mentioned are statutory declarations.

The evidence lodged in support of the opposition consists of:

1. A first declaration by Robert Charles Kelson, Barrister and Solicitor in the employ of Callinan Lawrie, patent attorneys for the opponent, with exhibit RCK-1.  This exhibit is a certified copy of the file records of the United States Patent and Trade Mark Office relating to Reissue application 32401, based on United States Patent 4637410, which derives priority from the same source as the present application. 

2. A first declaration by Warren Edward Claflin, a Principal Engineer in the Research and Development Department of Philip Morris Inc, and exhibit WEC-1, being the specification of the application in suit, which Claflin discusses.

3. A second declaration by Kelson, with exhibits RCK-2 to RCK-8, relating primarily to the public availability of a book, Technique du Tabac (Tabac), and some foreign patent specifications.

4. A declaration by Ken Cotterill, Librarian, with exhibits KC-1 to KC-2, also concerned with the availability of Tabac.

5. A third declaration by Kelson, with exhibits RCK-9 to RCK-21.  The exhibits include the papers entitled Factors Affecting Static Burning Rate by Resnik et al (the Resnik paper), Effect of Cigarette Circumference on Weight Loss During Puffs by Yamamoto et al (the Yamamoto paper), Role of Cigarette Physical Characteristics on Smoke Composition by DeBardeleben et al (the DeBardeleben paper), and some patent specifications, particularly Australian 571664, (the Lephardt patent).

6. A second declaration by Claflin correcting an error in his first declaration.

7. A first declaration by Rose Cooper Arthur, Product Development Technician for Philip Morris Inc, citing some test results on Capri and Ten Cent  cigarettes.  The former is apparently a cigarette produced by the applicant according to the invention, the latter is an old cigarette in the ultra-slim category.  It was manufactured and sold in East Africa by a subsidiary of the applicant for a period extending up until at least the late 1960's.

8. A first declaration by Richard Jeremy Stone, presently an Assistant Director with Philip Morris Products Inc, citing specifications and further test results for the Ten Cent cigarette.

9. A declaration by Rosy Fischer, Librarian, concerning the publication of some journals.

10. A first declaration by Henry Goldberg, Operations Director for the opponent, with exhibits HG-1 to HG-3.  The exhibits are the present specification, a booklet describing some cigarette manufacturing machines, and some sales figures for cigarettes in Australia.

11. A first declaration by Allan Henriksen, Product Development Manager for the opponent, with exhibits AH-1 to AH-6.  The exhibits include the specification, the papers by Resnik, Yamamoto, and DeBardeleben, and the declaration made by Kenneth Charles Roberts as part of the evidence-in-support of the opposition by Rothmans, now withdrawn, together with its exhibits.

12. A second declaration by Henriksen referring to Tabac.

The evidence-in-answer consists of:

13. A declaration by Tasman I Wilson, presently employed as Manager, Wills Technical Centre, with exhibits TIW-1 to TIW-15.  The exhibits include a number of patent specifications, an affidavit concerning the properties of the Ten Cent cigarette, another affidavit concerning the "firmness" of a cigarette according to claim 1 compared with a conventional cigarette, and the paper, Studies on the Transport Phenomena in Naturally Smouldering Cigarettes by Muramatsu (the Muramatsu paper).

14. A declaration by Peter Heathcote, the applicant's responsible Patent Attorney in this matter, with exhibits PH-1 to PH-3, concerned with the publication dates of the Resnik paper and some patent specifications.

The evidence-in-reply consists of:

15. A declaration by Wallace G Lloyd, tobacco industry consultant and previously employed by Philip Morris Inc, concerning some tests performed on cigarettes about 17 years old compared with records of the tests performed on the same brand of cigarette when manufactured.  The relevance of this is the effect of ageing on cigarette characteristics, specially SBR.

16. A third declaration by Henriksen, commenting on the declaration by Wilson.

17. A second declaration by Stone, also commenting on the declaration by Wilson.

18. A third declaration by Claflin, concerning the SBR of some commercially available cigarettes and some test cigarettes.

19. A second declaration by Arthur, commenting on the Wilson declaration.

20. A fourth declaration by Claflin, commenting on the Wilson declaration.

21. A declaration by Daniel M Ennis, Associate Senior Scientist for Philip Morris Inc, commenting on the statistical validity of the "firmness" tests presented in the Wilson declaration.

22. A declaration by Robin Shiffman, Technical Manager for the opponent, commenting on the declaration by Wilson, and also on the declaration by Taj Hirji, the latter being part of the evidence-in-support of the opposition by Rothmans, now withdrawn.

23. A declaration by Charles David, Process Development Manager for the opponent, commenting on the Wilson and Hirji declarations regarding some statistical matters.

24. A declaration by Raj Vasudeva, Mathematician, commenting on the statistical validity of statements made by Roberts and Hirji.

25. A fourth declaration by Kelson, with exhibits RCK-22 to RCK-27.  The exhibits include the declarations by Hirji with its exhibits, and by Herbert Bentley, a paper entitled The Mechanism of Smouldering in Cigarettes by Egerton et al, and information regarding the public availability of this paper and the Muramatsu paper.

26. A fifth declaration by Claflin, commenting on the Wilson declaration.

27. A second declaration by Goldberg, with exhibit HG-4, commenting on Wilson.

SUBMISSIONS

Not all the extensive documentary evidence filed was referred to at the hearing.  In the following I attempt to summarise the main issues raised at the hearing and in the evidence.

From the Opponent

Dr Emmerson submitted that the conventional wisdom alleged by the specification at page 1a, namely that the maintenance of smoulder required the consumption of about 60 mg/min of tobacco as a minimum, and that this required a minimum circumference of about 22mm, simply did not exist.  This was supported by reference to the declarations by Claflin (December 1989), Goldberg (September 1990), and Henriksen (October 1990).  The tactic employed in the specification, it was argued, was to set up a false conventional wisdom, and then "discover" that it was wrong.  In fact all the applicant needed to do to make ultra-slim cigarettes was to make them, using known techniques, without making any discovery, overcoming any difficulty, or needing to make any compensatory adjustments.  This being the case, it was submitted then that the invention claimed does not involve an inventive step and is obvious.

The characteristic parameters of the claimed cigarette were considered, with the result that, it was said, virtually the whole range of ultra-slim cigarettes was claimed.  The tobacco density covered the whole useful range for cigarette manufacture generally, plus a margin.  If the density fell below about 150 mg/cc it would be too loose and not contained by the wrapper.  If above about 350 mg/cc there would be an excessive pressure drop when puffed, ie difficult to draw.  Reference was made to declarations by Goldberg, Henriksen, and Roberts.  The circumference claimed of 10-19mm was the whole range for ultra-slim cigarettes, and the claimed SBR of 25-50 mg/min was merely a reflection of the claimed circumference  range, on which it is dependent.

Regarding this, it was submitted that it was common knowledge in the industry that SBR decreased with decreasing circumference, other things being constant, and that this is shown in the Resnik paper.  This paper was accepted by both parties as being in the common general knowledge in the industry.  There was much discussion about what this paper teaches, particularly with regard to Fig 9 and the associated text.  This figure is a graph of circumference vs SBR, and for 9 test cigarettes of circumference ranging from about 23mm to about 26mm, shows SBR's increasing from about 62 mg/min to about 76 mg/min.  The text refers to the importance of this relationship for slim cigarettes. The graph has broken axes, about which there was also argument.  The opponent extends Resnik's findings from the slim range to the ultra-slim, with the intention of showing that the resulting SBR is within that range claimed.  The validity of this extrapolation was also disputed.

In response to strong argument from the applicant that the invention is for a combination, characterised by the three integers of packing density, circumference, and SBR, it was submitted that the claim in fact is not for a true combination, since the integers are not all real, are not independent, and do not interact.  The packing density claimed, it was said, was not a real integer since it covered the whole range, circumference and SBR are not independent, and there is no interaction between some integers.  Therefore, it was put, there was no need to find all these integers in a piece of prior art to find want of novelty.

A number of items of prior art were identified, which it was said anticipated claim 1.

Reference was made to the Lephardt patent, which discloses a cigarette in which the tobacco rod is surrounded by a corrugated layer sandwiched between an external wrapper and the rod.  Examples of the size of the tobacco rod are given which fall within the circumference claimed, thus anticipating (it was argued) claim 1.

The book Tabac was also considered.  This book is in French but a translation shows cigarettes of diameter range 4-12mm (about 13-38mm circumference), and in addition to disclosing claim 1 shows that it was not the conventional belief that ultra-slim cigarettes could not be made.

A number of cigarette making machines were mentioned.  The evidence shows the Hauni-Teknik machine can be equipped to produce cigarettes of 18.85mm circumference, ie just within the claimed range.  The Rizla (hand rolling) machine could be set by the user to make extremely small cigarettes.

An excerpt was produced from the Guinness Book of Records, showing that a cigarette had been manufactured with a diameter of only 3mm (9.4mm circumference).

Reference was also made to the Yamamoto paper.  This paper studies the effect of varying circumference on puff count, and on the distribution of tar and nicotine among the tobacco column, filter, and mainstream smoke.  The cigarettes were outside the circumference range claimed.

Submissions were also made concerning the dimensions and measured properties of the Ten Cent cigarette, as made in about 1969, and as samples of the same brand retested in 1990.  In both the reports (Cooper and Stone) the cigarettes have packing density within the claimed ranges; SBR was not originally tested but on retest that also was also within the claimed range.  In response to argument that the cigarettes were too old to be reliably tested for SBR, the opponent submitted evidence regarding the original and present SBR for a similarly obscure cigarette, about 17 years old at the time of retesting, and which was found to be unchanged (Lloyd).

It was submitted that the law applicable in this case was principally that set out in Werner RD & Co Inc v Bailey Aluminium Products Pty Ltd (1989) 85 ALR 679. Reference was also made to Asahi v Grace 22 IPR 491, and NV Philips Gloeilampenfabrieken v Mirabella International Pty Ltd 26 IPR 513.

Finally, brief submissions were made pointing out that there was no evidence of commercial success, or of a long felt need, that the alleged invention was more concerned with fashion and marketing than invention, and that the claims do not comply with section 40 since they do not define an invention at all.

From the Applicant

There were two preliminary matters raised by Mr Catterns.  The first was that the annexure to the Stone declaration relating to the original test results  for the Ten Cent cigarette appears to have been incomplete as served on the applicant, and consequently whether it could be considered as being before the Office.  It was agreed that it could be so considered.

In the second, a rather long list was tendered containing alleged defects in the opponent's evidence, going generally to admissibility, and specifically to questions such as relevance, speculation, qualifications, and so on.  It was pointed out that these were administrative proceedings, and the matter was not pressed.  Mr Catterns requested that it be considered just as part of the applicant's argument.

It was submitted that the invention claimed is a simple combination claim, characterised by the three integers of circumference, packing density, and SBR.  Accordingly, for a finding of want of novelty, all of these integers must be disclosed in a single piece of the prior art, and this was not done.  The correct approach, it was said, was not to regard the invention as solving a problem, but of proposing a new combination, which offered a number of real, though individually perhaps not large, benefits.

These included increased tobacco use efficiency, reduced filter material, lower SBR, reduced sidestream smoke, and improved subjective firmness.

Dr Emmerson's arguments that the claim was not for a true combination were disputed.  First, statements regarding what the law required of a patentable combination were not necessarily accepted, and second, the claims satisfied them anyway.  It was conceded that the packing density range was conventional, but it is common for integers in a combination claim to be old, and in any case it might have been that the reduced circumference required a non-conventional density.  Contrary to the opponent's argument, there is interaction among the integers, in particular SBR is influenced by the density.  Also, while the circumference and SBR are related, SBR is not determined by the circumference, as it is affected by many other factors as well.  It was pointed out that the evidence associated with Hirji showed that cigarettes could be manufactured within the claimed ranges of circumference and density, but which would fall outside the claimed range of SBR.

The existence of the conventional wisdom that satisfactory ultra-slim cigarettes could not be made was supported by reference to the Hirji declaration, and the fact that no compensatory adjustments were needed was part of the applicant's discovery.  Reference was made to the Bentley statement.  In this Bentley reports his "surprise" at the demonstrated qualities of the Capri cigarette, which is a cigarette according to claim 1, in a number of respects and generally according to the benefits claimed by the applicant.

So far as the Ten Cent cigarette is concerned, it was put that the cigarettes were too old for tests to be reliable, the cigarette was certainly not part of the common general knowledge, and in fact there was no evidence of publication of the cigarettes or the old data in Australia.

It was submitted that much of the opponent's arguments were put as though the prior art publications were part of the common general knowledge in the industry, but this was not established, and, with the exceptions of the Resnik paper and some cigarette making machines, the applicant disputed it.  This was the applicant's position in respect of the book Tabac, for which only one copy and one loan could be found, some other machines, the papers by Yamamoto and others, and the Lephardt and other patent specifications.  So far as the latter are concerned, it was argued that since the constructions disclosed stood to save tobacco, they could be seen as showing the prior existence of such a need.

Therefore, it was said, since none of the prior art items disclose all the features of the combination claimed, and are not shown to be part of the common general knowledge, they do not assist the opponent's case in any way.

There was a critical examination of the Resnik paper in particular, and of the other papers by Yamamoto, DeBardeleben, and Muramatsu (though the evidence shows that this was not published in Australia), and what they taught.  It was argued that they all taught away from the invention, and the significance of the broken axes in Resnik's Fig 9 was a recognition by Resnik that the relationship between circumference and SBR shown only applied over the limited range tested.  The opponent's extrapolation of Resnik was said to be statistically unsound.  A redrawn Fig 9 was submitted by Hirji.  It was also suggested that Resnik's test cigarettes were experimental and not commercial.

In considering the applicable law it was argued that Nicaro Holdings and Others v Martin Engineering Pty Ltd and Another 16 IPR 545 was preferable to Werner v Bailey (supra) in its approach to deciding novelty.  It was also said that the claims were not like the claims considered in Philips v Mirabella (supra) as they are not claims merely specifying criteria to be met.  Alternatively, if they are, the decision in CCOM Ltd & Anor v JiejingPty Ltd & Ors 27 IPR 577 would support their validity.

Finally, it was said that the fact the opponent took no steps to market cigarettes according to Lephardt (or others, which use less tobacco, and which they own), but did market cigarettes according to the invention after its publication, is further evidence that there was a belief in the industry that acceptable ultra-slim cigarettes could not be made.

Hardly any mention was made at any stage about the Reissue application, or the independent claims.

DECISION

I will present my decision under a number of sub-headings.

Combination

In British Celanese Ltd v Courtaulds Ltd (52 RPC 171 at 193) Lord Tomlin said:

"It is accepted as sound law that a mere placing side by side of old integers so that each performs its own proper function independently of any of the others is not a patentable combination, but that where the old integers when placed together have some working interrelation producing a new or improved result then there is patentable subject matter in the idea of the working interrelation brought by the collocation of the integers."

Or in Minnesota Mining & Manufacturing Co v Beirersdorf (Aust) Ltd (1980) 144 CLR 253 at 266 Aickin J said:

"The...(invention claimed)...is a combination patent in the proper sense of that term, ie it combines a number of elements which interact with each other to produce a new result or product.  Such a combination may be one constituted by integers each of which is old, or by integers some of which are new, the interaction being the essential requirement."

More recently, in the case of Nicaro v Martin (supra) at page 553 Lockhart J said:

"Each of the integers of a combination patent may not be novel.  They may be part of the prior art, but it is the interaction of the combined integers to produce a new result which constitutes the invention the subject of a combination patent."

The cigarette claimed is characterised by circumference, tobacco packing density, and SBR.  In contrast with the Lephardt cigarette for example, it is of otherwise conventional structure. In most of the reported cases in which "mere collocation" is an issue, the inventions have been mechanical apparatus with a number of components or kits of some kind or another.  I have some difficulty regarding the claimed features as being "integers" (which is the word commonly used in the cases) in the relevant sense; in Webster's Third New International Dictionary (1976) it is in the most appropriate sense defined to be "a complete entity".  I am also doubtful whether the relationships between circumference and SBR and density and SBR constitutes an "interaction"; according to the same dictionary an interaction is said to involve a "mutual or reciprocal action or influence", and there does not seem to be any reciprocal influence between the features claimed.  Rather circumference and density are independent variables and SBR is a dependent variable.  Nevertheless, since the argument was strongly put, and the applicant is entitled to the benefit of any doubt in the matter, I will proceed as if the features are "integers" with a degree of "interaction".

Then, there is a "new or improved result" from this interaction, since, according to the applicant, tobacco is used more efficiently, quantity of filter material used is reduced, sidestream smoke is reduced, and some other benefits previously mentioned are realised.

Therefore I conclude that claim 1 is for a patentable combination.  Given that conclusion, the features of the combination claimed are all essential features of the invention.

I should also add that even if the claim was correctly construed not to define a combination at all, but rather to be for an integrated article with characterising features, then I would still consider the features claimed in claim 1 to be essential.  This follows, not just from the fact that the features are in the broadest claim, but from a consideration of the whole specification, the applicant's stated views of their invention, and the effect of the features on the performance of the invention.  See the comments by Buckley LJ in Catnic Components Ltd v Hill and Smith Ltd [1982] RPC 183.

Novelty

The test for want of novelty is the "reverse infringement test" as set out in Meyers Taylor Pty Ltd v Vicarr Industries (1977) CLR 228 at 235 where Aickin J said:

"The basic test for anticipation or want of novelty is the same as that for infringement and generally one can properly ask oneself whether the alleged anticipation would, if the patent were valid, constitute an infringement..."

The general rule as to what constitutes infringement has been expressed by Lord Upjohn in Rodi & Weinenberger AG v Henry Showell Ltd (1969) RPC 367 at 391 as follows:

"The essential integers having been ascertained, the infringing article must be considered.  To constitute infringement the article must take each and every one of the essential integers of the claim."

Also, per Lockhart J in Nicaro v Martin (supra) at 549:

"The invention must appear in a single disclosure, so it is not permissible to make a pattern or mosaic of or to read together various pieces of the prior art..."

Therefore the present question is, does a single piece of the prior art disclose, in Australia before the priority date, a cigarette of conventional construction, with a circumference in the range 10-19mm, packing density in the range 150-350 mg/cc, and SBR in the range 25-50 mg/min?  In my opinion the answer to this question must be in the negative, for the following reasons.

The most relevant items of prior art under this heading are the Lephardt patent, the Ten Cent cigarette, Tabac, and various cigarette manufacturing and rolling machines.

The Lephardt patent discloses a cigarette which has a tobacco rod with a circumference within the claimed range, a packing density (by calculation) within the claimed range, but with no mention of SBR.  As mentioned, while the SBR is dependent on circumference and density, it is not determined by them.  Also, the Lephardt cigarette is not of a conventional structure; the tobacco rod is surrounded by a corrugated wrapper.

The Ten Cent cigarette also has a circumference and density within the claimed range, but the evidence regarding the SBR appears to be inconclusive.  As this cigarette was originally specified the SBR was not given.  The opponent conducted tests on very old samples and found the SBR fell within the claimed range.  However the applicant has disputed the validity of these tests because of the age of the cigarettes tested.  The opponent has responded by conducting tests on another cigarette, of similar age, for which the SBR was originally specified, and has found the SBR has remained substantially unchanged.  I do not think I can reach any conclusion in these circumstances.  It does not matter in any case, as there is no evidence of publication in Australia of the cigarette or its specifications.

The translated extract from Tabac discloses cigarettes within the claimed circumference range, but is silent as to density and SBR.  The same can be said for the manufacturing and hand-rolling machines cited.

Therefore I conclude that the claims do not fail for want of novelty.

Obviousness

The test for obviousness under the Patents Act 1952 is whether the invention claimed would have been obvious to a non-inventive worker in the industry, equipped with the common general knowledge (cgk) in the industry, as at the priority date of the claim; see The Wellcome Foundation Ltd v VR Laboratories (Aust) Pty Ltd 148 CLR 262. The test must be made against what is proven to be cgk in the industry, being a "general body of knowledge" held and used by workers of ordinary skill in that industry - 3M v Beiersdorf (supra).

(a) The Common General Knowledge

It was accepted at the hearing that the Resnik paper and some of the cigarette machines were part of the cgk.  I consider that the evidence also shows that at least the concept of ultra-slim cigarettes was part of the cgk, as was the tobacco packing density claimed.  In my opinion, the evidence fails to establish anything else about cgk.

(b) The Conventional Wisdom

Declarants whom I would accept on the evidence as being persons skilled in the art disagreed  as to what was the "conventional wisdom" in the trade regarding the viability of ultra-slim cigarettes.  For the applicant it was said that it was believed such cigarettes would be (if I have understood it correctly) commercially unacceptable.  For the opponent it was disputed that any such belief existed; it was said that the only barrier in existence was a marketing one.

In these circumstances I am unable to decide the issue finally.  However it does seem to me that if such a belief existed it would be recognised by both parties, and/or there would be some objective basis for the belief.  In fact, neither was the case, and I am rather inclined to doubt its existence.

(c) The Resnik Paper

This paper is entitled Factors Affecting Static Burning Rate, and was published in Australia in the industry journal Tobacco Science during 1977.  The paper discloses the test results for the effect of a number of variables in cigarette construction on measured SBR.  Each of the results are said to be the average for 24 conventional cigarettes, with packing density of 260 mg/cc.

The most relevant part of the paper for present purposes is Fig 9 and the associated text.  This figure is a Cartesian graph; the x-axis represents cigarette circumference from 0-27mm, and the y-axis represents the measured SBR from 0-82 mg/min, under controlled (and apparently standardised) conditions.  Both axes are broken near the origin, with the result that the graph effectively displays circumference from 23-27mm, and SBR from 62-82 mg/min.

There are 9 data points on the graph, for a circumference range of about 23 to about 26mm, and SBR of about 62 to about 77 mg/min.  The distribution of the points suggest a positive correlation between circumference and SBR.

Resnik appears to have applied a least squares linear regression curve to the data points, the result being a "best fit" straight line of the form y=a+mx, where 'y' is SBR, 'x' is circumference, 'm' is the positive gradient, and 'a' is a constant.  That is, SBR is more or less directly proportional to circumference.  On page 3 of his paper Resnik says:

"...we find that SBR increases with increasing circumference (Fig 9).  This phenomenon is important in "slim" (22-23mm circumference) cigarette design where the percentage decrease in puff count caused by decreasing the circumference is not as great as the percentage weight reduction"(emphasis added).

The applicant questions what Resnik actually teaches.  It was submitted that the broken axes represent an honest admission by Resnik that his findings could not be extended outside the range actually tested.  However it is common in graphical representations to make axes discontinuous near the origin so that a relationship between variables can be more clearly seen.  Of course this should not be done without some caution, else a relationship be suggested where none exists.  The applicant has redrawn Fig 9 on continuous axes, and the data points do then appear as a small and distant cluster, difficult to interpret.  I accept that the axes have been broken for the conventional reason, and that Resnik's data justifies what is shown in Fig 9 and the conclusions drawn from it.

However the opponent seeks to extrapolate the Resnik curve into the ultra-slim range, with a view to showing that the claimed SBR is more or less a consequence of the reduced circumference.  The validity of this extrapolation is questioned by the applicant.

Regression analysis may of course be used for interpolation or extrapolation.  The former is when (eg in the present case) the SBR is estimated for a circumference not actually tested but within the range tested.  The analysis provides a percent confidence interval for the estimated value of SBR, and a (larger) confidence interval for the actual value of SBR.  These confidence intervals are at a minimum at the mean circumference, and increase away from the mean.

Extrapolation occurs when (in this case) the SBR is estimated for a value of circumference outside the range actually tested, that is by extending the regression curve.  This is a common practice though risky, and in fact not strictly statistically valid, irrespective of rapidly widening confidence intervals.  See the discussion in the text by Hays (et al), Statistics - Probability, Inference and Decision (New York 1971) at pages 643-644.  Basically it would appear to be a matter for the judgement of the user, whether and how far to extrapolate.

Therefore I am unable to fully accept the extrapolation performed by the opponent, and certainly not into the lower regions of circumference claimed, where anything could be happening.  However, I am prepared to accept that, in all probability, SBR would continue to trend downward as circumference is reduced, at least into the upper region of the claimed range.  I also accept, since it is stated by Resnik, and Resnik is in the common general knowledge, that it was well-known in the trade that the percentage decrease in puff count (which relates to consumer satisfaction) which occurs on reducing circumference is less than the percentage saving in tobacco (which relates to excise levied).  This is in my opinion what Resnik teaches; it does not teach the invention claimed, but it opens a window onto what has been invented.

(d) Other Papers

The evidence does not show that the other papers were part of common general knowledge in Australia.  Nevertheless I refer to two of them, not as independent disclosures of cigarettes, but as supporting Resnik, and as providing a description, if not an explanation, of some of the inherent relationships among the variables in a smouldering cigarette.  Yamamoto finds that the puff count is approximately proportional to circumference, and also that while the total delivery of tar and nicotine decreases with decreasing circumference, the proportion in the mainstream smoke increases as a result of decreased retention by the tobacco rod and filter.  Muramatsu finds that the puff count is proportional to the product of circumference and density, and that the SBR decreases with decreasing circumference and also decreases with increasing density (Table 6.2).

The cigarettes tested in these papers were in the slim-to-conventional circumference range.  Table 6.2 of Muramatsu shows an example which has a circumference of 20mm (nearest millimetre and just outside the range claimed), density of 177 mg/cc (within the range claimed), and SBR of 43.1 mg/min (within the range claimed).  I am not suggesting that Muramatsu is a near disclosure of the invention, as the paper does not appear to have been published in Australia.  However I am suggesting that this paper (and the others) are descriptive of some fundamental relationships among the variables present and at work in a smouldering cigarette, which, it seems to me, the applicant is not entitled to monopolise.

(e) Is there an Inventive Step?

The opponent says that the applicant chose, for commercial reasons, to design, market, and attempt to monopolise, ultra-slim cigarettes, and proceeded to do so, without overcoming any difficulty or making any discovery.

In Werner v Bailey (supra) at page 523 Lockhart J said:

"Novelty alone will not sustain a patent.  There must in addition be some difficulty overcome, some barrier crossed."

Nevertheless the quantum of invention required to validate a patent is small.  Thus in Killick v Pye [1958] RPC 366 at 377 Omerod J said:

"It is well settled that the validity of a patent, challenged on the ground of inventiveness, may be established though the inventive step represent a very small advance."

However in this case I must say that I am unable to discern any element of invention at all.

The problem faced by the applicant was that of devising an ultra-slim cigarette which would be commercially acceptable while reducing the mass of tobacco required.

In my opinion Resnik shows a clear guide to the solution of this problem, and this was in the common general knowledge in the industry.  Resnik shows that reducing the circumference tends to reduce the SBR, and that SBR is affected by many other factors as well, including tobacco cut, tobacco additives, tobacco moisture content, wrapper additives, and wrapper porosity.  The solution to the applicant's problem is heralded by Resnik when he notes the significance of the differing proportionate reductions in puff count and tobacco weight when applied to slim cigarettes.

It seems to me that the applicant has just extended the teaching of Resnik to the ultra-slim range of cigarettes, and has engaged in a design exercise rather than an exercise of the inventive faculty.  This exercise would have involved the juggling of known variables with generally known interdependencies, no doubt with testing, to arrive at design parameters which satisfied their requirements.  This may have required skill, but in my opinion it was not invention.

As Tomlin J said in Safreans AB v Ford Motor Co (England) 44 RPC 49 at 61:

"...(the claim)...as I have construed it does not go beyond ordinary skilled designing work, and cannot be considered as having required the exercise of any invention."

Finally I observe that in my opinion some of the actions of the parties in this matter were driven by commercial judgement and marketing decisions, and should not be taken to imply anything much about matters such as long felt need, difficulties overcome, conventional wisdom, or the like.

Therefore I conclude that claim 1 fails for obviousness.  No significant submissions were made about the dependent claims.  These claims define some preferred ranges within the broad ranges of claim 1 or add some other minor matters.  By the same reasoning these claims are also obvious.

Section 40

I do not think that any issues are raised under this heading separate from what has been considered already.

Conclusion

I have decided that all the claims of the application are obvious.  While I have doubts that the specification discloses patentable subject matter, I nevertheless allow the applicant 60 days from the date of this decision in which to propose amendments directed toward overcoming my findings, else I will refuse the application.  Subject of course to any appeal.

Costs

As the opposition has been successful, and there are no special circumstances, I award costs against the applicant.

RG Tolhurst
Delegate of the Commissioner of Patents

Patent attorneys for the applicant  :  Shelston Waters, Sydney

Patent attorneys for the opponent   :  Callinan Lawrie, Melbourne