T 0225/03 22-06-2007

1. The appeal complies with the requirements of Articles 106 to 108 in conjunction with Rule 64 EPC; it is therefore admissible.

2. The alleged public prior use concerns a bearing which is said to be shown in detail in the drawing E4/E4' or E5/E5'. Should a bearing according to these drawings be found to have been public before the priority date, then it seems accepted by the parties that the only difference between the claimed bearing and the alleged prior use bearing lies in the roller number coefficient defined by k = (dm/Da).sin(180º/z). For the alleged prior use bearing with 35 conical rollers (z) having a large diameter Da of 10,772 mm and an alleged diameter dm of a pitch circle in which the conical rollers are arranged of 137 mm this roller number coefficient is 1.14, thus not within the claimed range of 1.16 to 1.32. Since a reduction of the number of conical rollers from 35 to 34 would be enough for the resultant roller number coefficient (1.17) to fall within the claimed range the alleged prior use bearing is prima facie relevant for inventive step.

It is therefore unavoidable to examine whether the alleged public prior use effectively can be considered to be established.

3. According to the respondent the drawing E4 shows a series 37000 bearing which is a standard series for SKF as shown by the extract of a catalogue E2 of 1994.

The respondent's case is that Mr Stubenrauch declares in E1 that E3, E4 and E4' are evidence that series 3700 bearings were sold between 1985 and 1994 to ZF Passau, whereby the signature "Maurer" on E5 and E5' indicate the agreement of ZF for delivery of such bearings.

4. According to the established case law of the boards of appeal, the burden of proof lies with the party claiming that the information in question was made available to the public. Certain principles on the standard of proof necessary to establish the facts on which a decision is to be based have been developed.

In some decisions the boards of appeal have applied the standard of "the balance of probabilities", which means that in relation to, for example, the question of when a document was first made available to the public, the board must decide what is more likely than not to have happened (see for example decisions T 381/87, OJ EPO 1990, 213; T 296/93, OJ EPO 1995, 627; and T 729/91 of 21 November 1994).

In other decisions the boards took the view that a fact had to be proved "beyond reasonable doubt" or "up to the hilt" (see for example decisions T 782/92 of 22 June 1994; T 97/94, OJ EPO 1998, 467; T 848/94 of 3 June 1997; T 472/92, OJ EPO 1998, 161 and, in particular, T 750/94, OJ EPO 1998, 32).

The present board considers the latter approach to be the appropriate one in this case where the revocation of a granted European patent is at issue and the evidence in support of the alleged prior use is within the power and knowledge of the opponent. To base a revocation decision on the mere balancing of probabilities of what might have occurred would be difficult to reconcile with the need for reliability in the decision-making procedures of the EPO, which is of utmost importance for users of the patent system as well as the general public.

In the board's opinion as expressed in both its communications mentioned above the documentary evidence filed by the respondent is not sufficient for proving that bearings have effectively been made available to the public, under which circumstances they have been made available to the public and, if bearings have been made available to the public, what exactly the technical features of the bearing were.

4.1 In particular the board is of the opinion that the following points relating to the date on which the prior use occurred, exactly what was in prior use and the circumstances surrounding the prior use need to be confirmed or further investigated:

- It is not clear from E4' where the dimension of the pitch circle diameter dm = 137 has been derived from. The two diameters 132,6 and 141,5 said to have been used for calculating the pitch circle diameter have no direct link with the end surface of the rollers.

- It is not clear why this drawing E4/E4' dated from 1981 should be evidence for what has been sold from 1985 to 1994, particularly as the dimension 141.5 has been changed from 143.7 at a date which cannot be determined from the drawing (see amendment (6)).

- What the Bohrungs-Karteikarte E3 has been used for is not indicated on the document itself. It indicates some features of a bearing and has on it the reference to the client ZF Passau and to the number of the drawing E4/E4'. Even if it were accepted from the declaration of Mr Stubenrauch E1 that this card is evidence that something has been manufactured, it is not clear what exactly has been manufactured in 1980. It does not seem to be a bearing having the dimension dm=137 because the dimension 141.5 from which this apparently was derived was a higher value in and prior to 1981 (see amendment (2) "maß 143.7 nachgetragen" in E4/E4'). It is also not clear why E3 dated 1980 should be relevant for bearings sold from 1985 to 1994.

- Even if the number 37000 and the name ZF Passau appear on E3 this does not mean that the bearings have been made available to the public around that time. At that time the bearings may only have been confidential for test purposes.

- If 37000 is a series it is not clear what differences exist within the series and if the bearing appearing on the drawing and the ones sold had effectively the same features.

- Concerning the catalogue E2, a number on one page including the year 1994 and a copyright sign with the year 1994 alongside can be found. But there is no evidence that the catalogue has effectively been distributed or at least made available to a member of the public during that year.

- Even if it were proven that the catalogue was available in 1994 it is not clear what its relevance is to products which have been sold between 1985 and 1994, particularly as the catalogue seems not to contain all information relevant to Claim 1. There is no information in the declaration E1 or elsewhere in the file that the product shown in the catalogue is exactly the same as or corresponds to the products allegedly sold between 1985 and 1994.

- Additionally, when looking at the features of the series 37000 on the pages of the catalogue E2 it is to be noted that the inner diameter d indicated for this series (107,950) is not the same as the one on the drawing E4/E4' (109,538) and of the Bohrungskartei-Karte E3 (109,538).

- It is not clear why, if more than 10,000 bearings have been sold, apart from the declaration of Mr Stubenrauch it was not possible to bring any evidence for the sale of at least one bearing of this series to a member of the public at a specific date.

- It is not clear why Mr Stubenrauch remembers so well that over 10,000 bearings of this particular series of bearings have been sold in the particular interval of time.

4.2 On the basis of the foregoing the board shares the appellant's view that the alleged public prior use is not sufficiently proven by the documentary evidence. There remains the respondent's request to hear the witness Mr Stubenrauch in accordance with Article 117(1)(d) EPC.

5. The opposition division did not take evidence by hearing of the witness before revoking the patent because it considered the allegation of public prior use to be adequately proven by the documentary evidence. However, as set out above, the board finds that level of proof insufficient and the respondent's request to complement the documentary evidence by the witness being heard must be granted, only the witness being able to confirm or not the links between the pieces of evidence as well as the circumstances of use as explained by him in the "eidestattliche Versicherung". Under these circumstances and in order to avoid the loss of an instance the board considers it appropriate to exercise its discretion and to remit the case to the first instance for further prosecution pursuant to Article 111(1) EPC.

6. The appellant requested the board to disregard the two drawings E5 and E5' filed with letter of 9 August 2005 and the request to hear the witness as late filed and as constituting a tactical abuse of proceedings.

The respondent filed the two additional drawings E5 and E5' in response to the first communication of the board in which a number of issues concerning the alleged prior use were questioned. The board leaves open the matter of whether theses additional pieces of evidence are "prima facie" relevant enough to be introduced into the proceedings.

As regards the request to hear a witness the opponent in his notice of opposition used the wording in respect to Mr Stubenrauch "to be summoned via the opponent" ("zu laden über den Einsprechenden"). Even if this is not an explicit offer of Mr Stubenrauch as a witness it is difficult to see that it could have any other intended meaning. Moreover, the appellant himself first seriously questioned whether the alleged public prior use had been sufficiently proven not during the opposition proceedings but in the statement setting out the grounds of appeal.

In the opinion of the board the request to hear the witness therefore can not be considered to be late filed.

7. The subsidiary requests for oral proceedings to be held do not have to be considered since no decision on the substance adversely affecting a party has been taken. The case is remitted to the first instance for further prosecution and the parties still have the possibility of appealing against the final decision on the substance, possibly with oral proceedings being held (see T 315/92 of 27 April 1993, point 5 of the reasons).