Friday, 24 October 2014

T 0718/08 - Defining 'optimal chewability'

No need for brushing your dogs teeth anymore? The present case provides a chewable product for the dental care of pets. In the claims, the chewable product is not defined directly in terms of composition, but rather in reference to a particular property, namely to the minimum force required to penetrate the product by chewing.

According to GL F-IV, 4.11, where the invention relates to a product, it may be defined in a claim exceptionally by its parameters, namely only in those cases where the invention cannot be adequately defined in any other way. A necessary condition is that those parameters can be clearly and reliably determined either by indications in the description or by objective procedures usual in the art (T 94/82). 

According to GL F-IV, 4.18, where a claim contains an ill-defined ("unclear", "ambiguous") parameter, and the skilled person is not able, on the basis of the disclosure as a whole and using his common general knowledge, to identify (without undue burden) the technical measures necessary to solve the problem underlying the application at issue, an objection under Art. 83 should be raised.

The present case illustrates the above, in that the Board concludes that the skilled person is unable to reliably measure penetration force and thus reproduce the claimed invention.

Reasons for the Decision
1. The appeal is admissible.

2. Sufficiency of Disclosure

2.1 The invention is concerned with a chewable product for the dental care of pets comprising continuous and discontinuous phases, with claims to the product itself, its method of manufacture and its use in a method for reducing tartar. The main idea of the invention is to take into account the biting force of the pet in the design of the product (specification paragraph [0016]). To this end the independent claims require the phases to be in a proportion such that a force of at least 100 N is required to penetrate the product's surface. This is greater than the anticipated bite force of the pet (paragraph [0019]) ensuring optimum chewability and improved cleaning action in particular of molars and premolars, (paragraphs [0009], [0054]).

2.2 The product (and its method of manufacture and use) is not defined directly in terms of composition, but is characterized rather in reference to a particular property, namely the minimum force required to penetrate the product. Where a product is so defined in terms of a parameter, the disclosure will normally also need to provide sufficient information as to how to reliably and objectively measure the value of the parameter in question (unless, for example, this is known to the skilled person from his common general knowledge). This requirement ensures not only that the claimed subject-matter is clearly and unambiguously defined, but also that the skilled person, using that information to supplement his common general knowledge, is able to reproduce the invention without undue burden. Without such information he or she would not be able to successfully carry out the invention, and the invention would be insufficiently disclosed. Cf. Case Law of the Boards of Appeal, 5th Edition, 2006 (CLBA), II.A.6.1, first paragraph, and the decisions cited therein.

2.3 Information regarding the method for measuring penetration force can be found in paragraphs [0059] and [0088]. These refer to a "specially constructed "model tooth"" (paragraph [0059], line 59) and an analysis system "designed to simulate the biting action of a dog's teeth"(paragraph [0088], lines 12 to 13). The teeth in question are the premolars and molars (paragraph [0009]). To this end the system, identified as a TA XT2I Texture analyser from Rheo Ltd, uses "a specially designed cone-shaped penetrometry probe of length 12mm" pushed into the product "at a rate of 2mm/s" (paragraph [0088], lines 13 to 14). The skilled person learns from these passages read in context that he is to use a cone-shaped probe of 12mm length as a model of a dog's tooth to simulate biting action under given conditions. The description, figures and claims however do not specify the particular cone angle of this specially designed probe. As stands to reason the penetration force depends significantly on this angle : a sharp cone (small angle) will penetrate the product with greater ease than a blunt one (large angle). This is borne out clearly by the results of the tests summarized in figures 3, 6 and 8 of D2 and the tables on page 14 of D12, which show a variation in the order of 1000 over the measurement range (10° to 140°). The fact that cone angle is critical to measurement of the penetration force is undisputed, as is the fact that the disclosure fails to expressly mention any value for the cone angle.

2.4 The Appellant argues that the missing cone angle can be inferred from the probe's stated function as model dog tooth, and the fact that the product is aimed mainly at better cleaning of molars and premolars. This would instruct the skilled person, using his background knowledge of premolar and molar dimensions as reflected in D19 and condensed in D20, to choose that tooth having the same height as the probe, and, equating its width to the diameter of the conical probe's base, to so arrive at the value of its cone angle.

2.5 This line of reasoning is unconvincing. It assumes firstly a particular correlation between conical shape of the model and actual teeth size and shape, in this case buccal (cheek-side) height and width, for which the Board is unable to find any basis in the patent. It is also decidedly not part of the skilled person's common general knowledge to simply equate height and base diameter of a conical model tooth to buccal height and width of a given tooth. Premolars and molars have complex non-conical shapes that vary from tooth to tooth, as the photographs 105 to 110 and 405 to 411 of D11 (boxer premolars/molars taken from different angles), or also figure 1 (top) or figure 3 of D19 (side views of beagle teeth) clearly illustrate. Such a variety of complex shapes does not lend itself to simple modelling. Thus, even if a simple cone model is adopted, it is neither immediately apparent nor obvious how to determine the cone shape and size from the wide variety of actual teeth shapes and sizes, let alone that its dimensions should be based on buccal width and height of a single tooth.

The Board can also not subscribe to the further underlying assumption that the information provided in D19 belongs to common general knowledge. D19 is a scientific paper, published in a specialist journal, the Journal of Veterinary Dentistry, in 2002, two years after priority, which presents the results of a study of buccal (cheek side) surface dimensions of beagle teeth in comparison to those of cats and humans. The narrow scope of this study, its select readership (veterinary dentists), not to mention the fact that it was made public after priority, can but lead to the conclusion that D19 and the information therein does not belong to the skilled person's common knowledge. That person is a pet food engineer specializing in dental care products, whose background knowledge of animal teeth will have been drawn from dictionaries and encyclopaedias, and textbooks and handbooks on the subject (cf. T 890/02 (OJ EPO 2005, 97) cited in CLBA, I.C.1.5, first paragraph).

2.6 In the Board's view the skilled person is much more likely to try and find the missing cone angle amongst actual teeth angles. If his background knowledge as defined above offers a particular value (or very limited range of values) that he would immediately consider both as suitable and representative than the invention can be regarded as sufficiently disclosed. The photographs of D11 and declaration D19, however, show that no such particular value, or even a very narrow range of values, exists. As noted, the subject teeth, premolars and molars have various highly complex shapes. The different angle views in photographs 105 to 110 and 405 to 411 of D11, for example, show the premolar/molar surfaces of a boxer to have a varying number of rounded projections with different angles depending on the point of view (front or side). Thus even for premolars/molars the angles are spread widely, between say 30° to 140°, and there is no single value that is prevalent. This observation is confirmed by expert declaration D21, see section 5.3, which also mentions angle variation between breeds (section 4.2, 4.3).

At best a range can be identified where observed angles occur more frequently. D21 in section 5.3 gives some examples. The Appellant has previously suggested 30° to 90°, or an even narrower range, 30° to 50° (see the table in D20). In these progressively narrower ranges variation is still by a factor of 3 to 4 and 1.6 to 1.8 respectively (cf. D2, figures 6,8; D12, page 14, Rancho). This is still to an extent so as to preclude reliable measurement of the penetration force.

2.7 As for reverse-engineering the cone angle from table 3 and the specific examples described in the preceding paragraphs, the Board is of the firm conviction that this would place an undue burden on the skilled person.

Firstly, various factors and parameters of the manufacturing process that influence the material properties are left open in the patent. Besides duration of the various stages, this includes the nature and quality of the raw materials, the particular extruder used as well as the specific mechanical energy (SME) applied during extrusion. Table 1 on page 8 of the Respondent's submission of 14 November 2008, for example, demonstrates the significance of SME for penetration force.

Secondly, the extent of testing required to unambiguously determine which cone angle was used to produce the table values would far exceed routine experimental work. For each composition it would require producing a multitude of samples for different process parameters and subjecting each to flexion tests and repeated measurements of penetration force with different cone angles until the table values are returned.

2.8 In the light of the above the Board concludes that the skilled person is unable to determine the missing cone angle on the basis of the patent and his common general knowledge. Failing a specific value of the cone angle he will be unable to reliably measure penetration force and thus reproduce the claimed invention. The invention according to the claims of the main request is thus insufficiently disclosed (Articles 83, 100(b) EPC).

2.9 The specific value of the penetration force is central to the invention as it attempts to give expression to the underlying qualitative idea (see above) in objectively verifiable terms. As it fails herein, the invention is inherently deficient and any attempt to formulate the invention more precisely must fail. The auxiliary requests are thus also not allowable for the above reason.

For these reasons it is decided that:

The appeal is dismissed.
This decision has European Case Law Identifier: ECLI:EP:BA:2009:T071808.20090911. The whole decision can be found here. The file wrapper can be found here. Photo from

1 comment :

  1. Note that this decision is dated 11 September 2009. Recent decisions typically distinguish somewhat more clearly between A84 and A83 in case of parameters used to define a product, e.g. T 608/07.