Court of Justice of the European Communities (including Court of First Instance Decisions)

JUDGMENT OF THE GENERAL COURT (Eighth Chamber)

15 December 2010 (*)

(Competition - Administrative procedure - Decision finding a breach of seal - Article 23(1)(e) of Regulation (EC) No 1/2003 - Burden of proof - Presumption of innocence - Proportionality - Obligation to state the reasons on which the decision is based)

In Case T-141/08,

E.ON Energie AG, established in Munich (Germany), represented initially by A. Röhling, C. Krohs, F. Dietrich and R. Pfromm, and subsequently by A. Röhling, F. Dietrich and R. Pfromm, lawyers,

applicant,

v

European Commission, represented by A. Bouquet, V. Bottka and R. Sauer, acting as Agents,

defendant,

APPLICATION for annulment of Commission Decision C (2008) 377 final of 30 January 2008 relating to a fine pursuant to Article 23(1)(e) of Council Regulation (EC) No 1/2003 for breach of a seal (Case COMP/B-1/39.326 - E.ON Energie AG),

THE GENERAL COURT (Eighth Chamber),

composed of M.E. Martins Ribeiro (Rapporteur), President, S. Papasavvas and N. Wahl, Judges,

Registrar: K. Andòva, Administrator,

having regard to the written procedure and further to the hearing on 14 April 2010,

gives the following

Judgment

Legal context

1 Article 20(2)(d) of Council Regulation (EC) No 1/2003 of 16 December 2002 on the implementation of the rules on competition laid down in Articles 81 [EC] and 82 [EC] (OJ 2003 L 1, p. 1) provides that ‘[t]he officials and other accompanying persons authorised by the Commission to conduct an inspection are empowered … to seal any business premises and books or records for the period and to the extent necessary for the inspection’.

2 Under Article 23(1)(e) of Regulation No 1/2003, ‘[t]he Commission may by decision impose on undertakings and associations of undertakings fines not exceeding 1% of the total turnover in the preceding business year where, intentionally or negligently … seals affixed in accordance with Article 20(2)(d) by officials or other accompanying persons authorised by the Commission have been broken’.

Background to the dispute

3 By decision of 24 May 2006, in accordance with Article 20 of Regulation No 1/2003, the Commission of the European Communities ordered an inspection at the premises of E.ON AG and the undertakings controlled by it, in order to ascertain the validity of suspicions as regards their participation in anti-competitive agreements. The inspection of the applicant, E.ON Energie AG, a wholly-owned subsidiary of E.ON, commenced in the afternoon of 29 May 2006 at its business premises in Munich. On learning of the inspection decision, the applicant declared that it had no objection to it.

4 The inspection was carried out by four representatives of the Commission and six representatives of the Bundeskartellamt (the German competition authority). The documents selected during the inspection on 29 May 2006 for more detailed examination by those representatives were stored in room G.505, which was made available to the Commission by the applicant. Since the inspection could not be completed on the same day, the leader of the inspection team locked the door of that room, consisting of painted sound-insulating door leaf panels and of a door frame made of anodised aluminium, and affixed to it an official seal measuring 90 mm by 60 mm (‘the seal at issue’). The latter was affixed with up to approximately two thirds of its surface area on the door leaf and the remainder on the door frame. A record of sealing was drawn up and signed by representatives of the Commission, the Bundeskartellamt and the applicant. The inspectors then left the applicant’s premises, taking with them the key to the door of room G.505 which had been handed over to them. In reply to a request for information, the applicant indicated that, in addition to that key handed over to the Commission, 20 other ‘master keys’ giving access to room G.505 were also in circulation (recital 19 to the contested decision).

5 The seal at issue was a blue adhesive label with yellow lines on the upper and lower edges and the yellow stars of the European flag. The lower yellow area contained a notice to the effect that the Commission may impose a fine in the event of a breach of the seal. The security label used to make the seal at issue (‘the security label’) had been manufactured by 3M Europe SA (‘3M’) in December 2002. The Commission then had the abovementioned details printed on the security label by a printing firm in the first quarter of 2004.

6 When a plastic seal, such as the seal at issue, is broken, the white adhesive by means of which the seal is fixed to the substrate remains on the latter in the form of ‘VOID’ messages about 12 Didot points (approximately 5 mm) high, distributed over the whole surface of the adhesive label. The removed seal becomes transparent at those points, so that the ‘VOID’ messages are also visible on the seal.

7 On its return, at around 08.45 hrs on 30 May 2006, the inspection team found that the condition of the seal at issue, which was still adhering to the door of room G.505, had changed.

8 At around 09.15 hrs, the leader of the inspection team opened the door of room G.505. The opening of the door caused the detachment of the part of the seal at issue which was stuck to the door leaf, whereas the other part remained stuck to the door frame.

9 A breach-of-seal report was drawn up, which states inter alia the following:

‘…

- The entire seal was displaced by about 2 mm upwards and sideways so that traces of adhesive were visible at the lower and right-hand edges of the seal.

- The “VOID” message was clearly visible over the entire surface of the seal, which was nevertheless still affixed crosswise from the door frame to the surface of the door and was not torn.

- After the door was opened by the Commission [official] (Mr K.), the seal remained intact, that is, it did not tear, and on the rear of the seal (the adhesive side) white traces of the “VOID” message were evident.

- When the seal is removed, the white “VOID” message normally remains on the substrate, and that was largely the case here, too, the message being in fact on the door surface.

- Numerous white traces were, however, also visible on the adhesive surface of the seal, not on, but beside the corresponding transparent sections of the “VOID” message on the rear of the seal.’

10 The breach-of-seal report was signed by a representative of the Commission and by a representative of the Bundeskartellamt. The applicant refused to sign it.

11 During the afternoon of 30 May 2006, some digital photographs of the seal at issue were taken with a mobile telephone.

12 On 31 May 2006, the applicant made a ‘supplementary statement … to the record of sealing of 30 May 2006’, which reads as follows:

‘1. When the door was opened, no change in the documents stored in the room was ascertained.

2. On removal of the seal on the evening of 30 May prior to resealing, the “VOID” message on the door frame was completely intact.

3. Mr K. was present when the seal was affixed on the previous evening and had the impression that the process was unusually protracted.’

13 On 9 August 2006, the Commission sent a request for information to the applicant in accordance with Article 18 of Regulation No 1/2003. The applicant replied by letter of 23 August 2006. Other requests for information were sent on 29 August 2006 to 3M, on 31 August 2006 to the cleaning company working for the applicant (‘the cleaning company’) and on 1 September 2006 to the applicant’s security service.

14 The 10 members of the inspection team completed questionnaires relating to their observations on the affixing of the seal at issue and on its condition on the morning of 30 May 2006.

15 On 2 October 2006, the Commission sent a statement of objections to the applicant. On the basis of the available information, it concluded inter alia that the seal at issue had been broken and that, because of the applicant’s organisational power in the building in question, the applicant had to be held responsible for that breach of seal.

16 On 13 November 2006, the applicant submitted its comments on the statement of objections.

17 On 6 December 2006, at the applicant’s request, an oral hearing took place, held by the Hearing Officer, which 3M also attended.

18 On 21 December 2006, at the Commission’s request, 3M confirmed in writing certain statements made during the oral hearing.

19 During the administrative procedure, the applicant forwarded to the Commission three expert opinions from an institute of natural sciences and medicine (‘the institute’).

20 On 21 March 2007, the institute produced a first expert opinion (‘institute expert opinion I’), which analysed the reaction of the seal at issue to shearing and peeling stresses.

21 On 11 April 2007, the Commission asked Mr Kr., a sworn expert in adhesives technology and the material behaviour of plastics, to draw up a report on certain aspects of the performance and handling of the seal at issue. His first report (‘Kr. report I’) was drawn up on 8 May 2007.

22 On 15 May 2007, the institute produced a second expert opinion (‘institute expert opinion II’), which analysed the reaction of the seal at issue to pulling-shearing and compression stresses and to peeling stresses following the use of the cleaning product Synto (‘Synto’).

23 By letter of 6 June 2007, the Commission informed the applicant of the new facts ascertained since the statement of objections, based on 3M’s statements and Kr. report I, and gave it the opportunity to comment on them in writing.

24 On 6 July 2007, the applicant sent the Commission written comments and requested a further hearing. That request was refused.

25 On 1 October 2007, the applicant sent the Commission the institute’s third expert opinion, dated 27 September 2007 (‘institute expert opinion III’), which analysed the reaction of the seal at issue to peeling stresses following ageing, the use of Synto and air humidity.

26 The Commission then asked Mr Kr. to comment on the arguments and remarks contained in the applicant’s letter of 6 July 2007 and in institute expert opinions II and III. Mr Kr. drew up his second report on 20 November 2007 (‘Kr. report II’).

27 By letter of 23 November 2007, the Commission informed the applicant of the additional facts identified since its letter of 6 June 2007. At the same time, it granted the applicant access to the relevant documents, and in particular to Kr. report II.

28 On 10 December 2007, the applicant commented on the documents sent on 23 November 2007.

29 On 15 January 2008, the Commission received another letter from the applicant, to which were attached affidavits from 20 persons who, according to the applicant, were in possession of a key to room G.505 on the evening of 29 May 2006 (‘the keyholders’). Those persons declared in those affidavits that, during the period in question (between 19.00 hrs on 29 May 2006 and 09.30 hrs on 30 May 2006), they had either not been in building G or had not opened the door of the room in question (recital 42 to the contested decision).

30 On 30 January 2008, the Commission adopted Decision C (2008) 377 final relating to a fine pursuant to Article 23(1)(e) of Council Regulation (EC) No 1/2003 for breach of a seal (Case COMP/B-1/39.326 - E.ON Energie AG (‘the contested decision’), a summary of which is published in the Official Journal of the European Union of 19 September 2008 (OJ 2008 C 240, p. 6).

31 The operative part of the contested decision states:

‘Article 1

E.ON Energie AG broke a seal affixed by representatives of the Commission pursuant to Article 20(2)(d) of Regulation No 1/2003 and at least negligently infringed Article 23(1)(e) of Regulation No 1/2003.

Article 2

A fine of EUR 38 000 000 is hereby imposed on E.ON Energie AG for the infringement referred to in Article 1.

…’

Procedure and forms of order sought

32 By application lodged at the Registry of the Court on 15 April 2008, the applicant brought the present action.

33 The applicant claims that the Court should:

- annul the contested decision;

- in the alternative, reduce the amount of the fine imposed to an appropriate amount;

- order the Commission to pay the costs.

34 The Commission contends that the Court should:

- dismiss the action in its entirety;

- order the applicant to pay the costs.

35 Upon hearing the report of the Judge-Rapporteur, the Court (Eighth Chamber) decided to open the oral procedure. The parties presented their oral arguments and their replies to the oral questions put by the Court at the hearing on 14 April 2010.

Law

36 The applicant raises nine pleas in support of the action. The first seven pleas concern the finding of breach of seal, whereas the last two relate to the amount of the fine.

37 The first plea alleges failure to have regard to the burden of proof, the second, infringement of the ‘principle of inquisitorial procedure’, the third, an erroneous assumption that the seal was affixed properly, the fourth, an erroneous assumption of the ‘obvious condition’ of the seal at issue on the day following the inspection, the fifth, an erroneous assumption of the suitability of the security label for official sealing by the Commission, the sixth, disregard by the Commission of ‘alternative scenarios’ which may have been the cause of the condition of the seal at issue, the seventh, breach of the principle of the presumption of innocence, the eighth, infringement of Article 23(1) of Regulation No 1/2003 inasmuch as no fault on the applicant’s part was established and, finally, the ninth, infringement of Article 253 EC and of the principle of proportionality when the amount of the fine was set.

The first plea, alleging failure to have regard to the burden of proof

Arguments of the parties

38 The applicant submits that, in accordance with the principle of in dubio pro reo, the principle of the presumption of innocence, enshrined in Article 6(2) of the Convention for the Protection of Human Rights and Fundamental Freedoms, signed in Rome on 4 November 1950 (‘the ECHR’), and the first sentence of Article 2 of Regulation No 1/2003, the Commission bears the burden of proof in proceedings capable of resulting in fines imposed under antitrust law. Since the Commission is required to observe the fundamental guarantees of criminal law and to prove to the requisite legal standard that an infringement has occurred, the undertaking in question should receive the benefit of any doubts on the Commission’s part. According to the applicant, it is apparent from the case-law that, where the Commission acts on the basis of the assumption that facts found cannot be explained in any way other than as being the result of an infringement, it is sufficient to establish circumstances which cast the facts established by the Commission in a different light and which thus allow a different explanation of the facts to be substituted for that adopted by the Commission in order to conclude that an infringement has occurred.

39 As regards the Commission’s assertion that the alteration of the seal at issue is in any event ‘prima facie evidence’ of the material ingredient of breach of seal, the applicant submits that such evidence is incompatible with the principle of in dubio pro reo. In proceedings under Article 23(1) of Regulation No 1/2003, ‘prima facie evidence’ does not constitute admissible evidence and is not under any circumstances sufficient to prove an infringement penalised by a fine. Even if, by referring to ‘prima facie evidence’, the Commission actually meant proof by circumstantial factors, it has not adduced any proof since it has not established any circumstantial factors.

40 Furthermore, the burden of proof resting on the Commission is increased in this case as a result of its own conduct.

41 Firstly, when it used the seal at issue, the Commission did not take appropriate measures to limit the risk of ‘false positive reactions’ (that is, the appearance of ‘VOID’ messages on the seal at issue in the absence of its detachment), in particular due to the failure to comply with the shelf life of its seal. Thus, the Commission should have proved that, despite the fact that the maximum shelf life of the seal at issue had been exceeded, the seal was suitable and operational during the night of 29 to 30 May 2006. The manufacturer’s information is insufficient in that regard, since 3M mentions only a maximum shelf life of two years in the data sheet on the security label (‘the technical bulletin’) and was also unable to give a definitive opinion on the exact life of the product in the reply to the Commission’s questionnaire. Nor did the Commission adduce such proof by means of Mr Kr.’s tests which, moreover, were not carried out on the seal at issue itself. Finally, the institute’s expert opinion mentions that self-adhesive seals may be characterised by greater sensitivity depending on the circumstances in which they are affixed and the level of air humidity.

42 Secondly, the Commission failed to take, on the spot, the necessary measures to preserve evidence by taking photographs of the seal at issue prior to the opening of the door, in particular in the light of the observations relating to the condition of the seal at issue which were made by the applicant’s representatives to the representatives of the Commission on the morning of 30 May 2006. In that regard, the record of breaking of the seal does not in itself constitute sufficient proof of the condition of the seal at issue, since it was drawn up after the inspection.

43 Having regard to the principle of in dubio pro reo and to the increase in the burden of proof which resulted from the Commission’s conduct, it is not possible to find that a breaking of the seal occurred for which the applicant can be held responsible. The Commission has thus failed to prove beyond all reasonable doubt the material ingredient of the infringement.

44 Contrary to what the Commission contends, the present plea is not abstract, but specifically asserts the absence of proof that circumstances attributable to the applicant led to the alteration in the condition of the seal at issue. Thus, even without the expert opinions submitted by the applicant, the imposition of a penalty in the form of a fine would not be justified. In a proceeding resulting in a decision imposing a fine, it is not for the undertaking concerned to establish proof of exculpatory factors or ‘alternative scenarios’. On the contrary, the Commission must examine comprehensively all the inculpatory and exculpatory circumstances and adduce absolute proof beyond all reasonable doubt that circumstances attributable to the applicant led to the alteration in the condition of the seal at issue. A mere probability that the infringement occurred is not sufficient for the purpose of imposing a fine, a fortiori since the applicant has raised sufficient doubts as regards the adduction of evidence.

45 Even if it were to be accepted that, initially, the Commission provided seemingly convincing evidence establishing the constituent elements of a breach of seal, the applicant has nevertheless successfully adduced evidence in rebuttal. It has, in any case, succeeded in raising doubts as to the fact that the Commission’s evidence was sufficient to prove the infringement. Contrary to what the Commission suggests in recital 44 to the contested decision, the applicant did not, as regards the existence of the breach of seal, ‘simply … claim that alternative explanations are possible’ or merely ‘claim that an [atypical] alternative scenario was theoretically possible’, but relied on a number of expert opinions by the institute in order to adduce proof that certain circumstances, namely the use of an out-of-date seal, air humidity, the vibrations to which the door and its frame were subjected and the resultant shearing stresses as well as the effect of Synto, could have caused creeping of the seal at issue, creating the ‘impression of damage’ found by the inspection team. In proceedings culminating in a decision imposing a fine, any particular circumstances surrounding the choice of the seal’s intermediate product (in this case, the security label), its storage and its use by the Commission cannot be regarded as being purely and simply irrelevant.

46 In the present plea, the applicant proposes, in accordance with Article 65(c) of the Rules of Procedure of the General Court, that the oral testimony of its lawyer and an employee of E.ON should be heard with regard to the circumstances in which the seal at issue was found on the morning of 30 May 2006.

47 The Commission contends that the first plea should be rejected since it is worded abstractly and without examining the actual effects on the assessment of the evidence and on the contested decision. In the alternative, it also disputes the applicant’s arguments.

Findings of the Court

48 Under Article 2 of Regulation No 1/2003 and according to settled case-law relating to the application of Articles 81 EC and 82 EC, in the field of competition law, where there is a dispute as to the existence of an infringement, it is for the Commission to prove the infringements found by it and to adduce evidence capable of demonstrating to the requisite legal standard the existence of the circumstances constituting an infringement (Case C-185/95 P Baustahlgewebe v Commission [1998] ECR I-8417, paragraph 58; Joined Cases C-2/01 P and C-3/01 P BAI and Commission v Bayer [2004] ECR I-23, paragraph 62; and Case T-201/04 Microsoft v Commission [2007] ECR II-3601, paragraph 688). For that purpose, it must gather sufficiently precise and consistent evidence to support the firm conviction that the alleged infringement took place (see, to that effect, Joined Cases 29/83 and 30/83 CRAM and Rheinzink v Commission [1984] ECR 1679, paragraph 20; Joined Cases C-89/85, C-104/85, C-114/85, C-116/85, C-117/85 and C-125/85 to C-129/85 Ahlström Osakeyhtiö and Others v Commission [1993] ECR I-1307, paragraph 127; and Joined Cases T-185/96, T-189/96 and T-190/96 Riviera Auto Service and Others v Commission [1999] ECR II-93, paragraph 47).

49 Next, it must be recalled that, in proceedings for annulment brought under Article 230 EC, all that is required of the Courts of the European Union is to verify the legality of the contested measure (Joined Cases T-67/00, T-68/00, T-71/00 and T-78/00 JFE Engineering and Others v Commission [2004] ECR II-2501, paragraph 174).

50 Thus, the role of a Court hearing an application for annulment brought against a Commission decision finding the existence of an infringement in the field of competition law and imposing fines on the addressees consists in assessing whether the evidence and other information relied on by the Commission in its decision are sufficient to establish the existence of the alleged infringement (JFE Engineering and Others v Commission, paragraph 49 above, paragraph 175).

51 Where there is doubt, the benefit of that doubt must be given to the undertakings accused of the infringement (see, to that effect, Case 27/76 United Brands and United Brands Continentaal v Commission [1978] ECR 207, paragraph 265). The Court cannot therefore conclude that the Commission has established the existence of the infringement at issue to the requisite legal standard if it still entertains doubts on that point, in particular in proceedings for the annulment of a decision imposing a fine (JFE Engineering and Others v Commission, paragraph 49 above, paragraph 177).

52 In the latter situation, it is necessary to take account of the principle of the presumption of innocence resulting in particular from Article 6(2) of the ECHR, which is one of the fundamental rights which, according to the case-law of the Court of Justice and as reaffirmed by the preamble to the Single European Act, by Article 6(2) EU and by Article 47 of the Charter of Fundamental Rights of the European Union proclaimed on 7 December 2000 in Nice (OJ 2000 C 364, p 1), are protected in the Community legal order. Given the nature of the infringements at issue and the nature and degree of severity of the ensuing penalties, the principle of the presumption of innocence applies in particular to the procedures relating to infringements of the competition rules applicable to undertakings that may result in the imposition of fines or periodic penalty payments (see, to that effect, in particular the judgments of the European Court of Human Rights of 21 February 1984 in Öztürk v. Germany, Series A no. 73, and of 25 August 1987 in Lutz v. Germany, Series A no. 123, of the Court of Justice in Case C-199/92 P Hüls v Commission [1999] ECR I-4287, paragraphs 149 and 150, and Case C-235/92 P Montecatini v Commission [1999] ECR I-4539, paragraphs 175 and 176; and JFE Engineering and Others v Commission, paragraph 49 above, paragraph 178).

53 The applicant relies on the case-law concerning concerted practices prohibited by Article 81 EC, according to which parallel conduct of the undertakings concerned cannot be considered to prove concerted action contrary to that provision unless concerted action is the only plausible explanation for it (CRAM and Rheinzink v Commission, paragraph 48 above, paragraph 16). As regards concerted practices, it is thus for the Commission, in the light of the arguments put forward by the undertakings concerned during the administrative procedure, to examine all the possible explanations for the conduct at issue and to find that it is in the nature of an infringement only where the infringement constitutes the only plausible explanation.

54 Consequently, if the Commission finds, on the basis of the conduct of the undertakings concerned, that there has been an infringement of the competition rules, the Courts of the European Union will find it necessary to annul the decision in question where those undertakings put forward arguments which cast the facts established by the Commission in a different light and thus allow another plausible explanation of the facts to be substituted for the one adopted by the Commission in order to conclude that an infringement occurred (CRAM and Rheinzink v Commission, paragraph 48 above, paragraph 16, and Ahlström Osakeyhtiö and Others v Commission, paragraph 48 above, paragraphs 126 and 127).

55 However, in the same way as, where the Commission relies, in establishing an infringement of Articles 81 EC and 82 EC, on documentary evidence, the burden is on the undertakings concerned not only to put forward a plausible alternative to the Commission’s view but also to allege that the evidence relied on in the contested decision to establish the existence of the infringement is insufficient (see, to that effect, Joined Cases T-305/94 to T-307/94, T-313/94 to T-316/94, T-318/94, T-325/94, T-328/94, T-329/94 and T-335/94 Limburgse Vinyl Maatschappij and Others v Commission [1999] ECR II-931, paragraphs 725 to 728, and JFE Engineering and Others v Commission, paragraph 49 above, paragraph 187), it must be held that, in a case such as this, where the Commission relies on direct evidence, it is for the undertakings concerned to demonstrate that the evidence relied on by the Commission is insufficient. It has already been ruled that such a reversal of the burden of proof does not infringe the principle of the presumption of innocence (see, to that effect, Montecatini v Commission, paragraph 52 above, paragraph 181).

56 It is also important to point out that an undertaking cannot transfer the burden of proof to the Commission by relying on circumstances which it is not in a position to establish (see, to that effect, Case T-44/00 Mannesmannröhren-Werke v Commission [2004] ECR II-2223, paragraph 262, and JFE Engineering and Others v Commission, paragraph 49 above, paragraph 343). In other words, when the Commission relies on evidence which is in principle sufficient to demonstrate the existence of the infringement, it is not sufficient for the undertaking concerned to raise the possibility that a circumstance arose which might affect the probative value of that evidence in order for the Commission to bear the burden of proving that that circumstance was not capable of affecting the probative value of the evidence. On the contrary, except in cases where such proof could not be provided by the undertaking concerned on account of the conduct of the Commission itself (see, to that effect, Mannesmannröhren-Werke v Commission, paragraphs 261 and 262, and JFE Engineering and Others v Commission, paragraph 49 above, paragraphs 342 and 343), it is for the undertaking concerned to prove to the requisite legal standard, on the one hand, the existence of the circumstance relied on by it and, on the other, that that circumstance calls in question the probative value of the evidence relied on by the Commission.

57 In the present plea, the applicant submits that the Commission was required to prove, in the contested decision, beyond all reasonable doubt, that the alteration of the condition of the seal at issue, found on 30 May 2006, was attributable to it, after demonstrating that the various circumstances alleged by it could not have explained that condition. According to the applicant, it is not for it to establish proof of exculpatory factors or ‘alternative scenarios’. A mere probability that the infringement occurred is not sufficient for the purpose of imposing a fine, a fortiori since the applicant has raised sufficient doubts regarding the adduction of evidence. In its first plea, the applicant thus refers to the age of the seal at issue, to the air humidity, to the vibrations to which the door and door frame were subjected, to the resultant shearing stresses and to the effect of Synto, all of which could have caused creeping of the seal at issue, creating the ‘impression of damage’ found by the inspection team.

58 In that regard, it must be noted that, contrary to what the Commission contends, the plea raised by the applicant is not abstract, since, in essence, the applicant asserts that, in the light of the Commission’s disregard of the principles governing the burden of proof in Community competition law, the Commission has not proved to the requisite legal standard that circumstances attributable to the applicant resulted in the alteration of the condition of the seal at issue, so that the contested decision should be annulled.

59 However, it is clear from the contested decision that the Commission did not disregard the principles governing the burden of proof. Firstly, recital 44 to the contested decision expressly states that ‘it is the Commission’s duty to submit the facts to prove the breach of the seals’. Secondly, the Commission based its finding of a breach of seal on the condition of the seal at issue on the morning of 30 May 2006, which, according to the Commission, had ‘VOID’ messages over its entire surface and traces of adhesive on its back, as is clear inter alia from the statements of the Commission and the Bundeskartellamt inspectors and from the findings contained in the breach-of-seal report (recitals 75 and 76 to the contested decision).

60 Consequently, referring inter alia to the statements of six inspectors present on the spot and to the signature by the applicant of the record of sealing, the Commission first found that the seal at issue was affixed properly on the evening of 29 May 2006 (recitals 50 and 51 to the contested decision). The Commission then found, as was noted in paragraph 59 above, an alteration in the condition of that seal on the morning of 30 May 2006, which, according to it, proves the breach-of-seal infringement

61 Irrespective of the probative value of the evidence on which the Commission relied and which will have to be assessed in the context of the examination of the third, fourth and fifth pleas, the Commission was fully entitled to consider in recital 44 to the contested decision that ‘[t]he mere claim that an [atypical] alternative scenario was theoretically possible cannot be regarded as sufficient’ to refute the existence of an infringement. Under the principles set out in paragraphs 55 and 56 above, it was for the applicant to prove not only the genuineness of the various circumstances alleged by it for the purpose of explaining the condition of the seal at issue on 30 May 2006, but also that those circumstances call in question the probative value of the evidence produced by the Commission.

62 In the contested decision, the Commission examined the alternative explanations concerning the condition of the seal at issue on 30 May 2006 put forward by the applicant during the administrative procedure. However, the Commission took the view that those explanations did not demonstrate that that condition resulted from circumstances other than a breach of seal (recitals 62 to 68 and 77 to 98 to the contested decision). In those circumstances, no infringement of the principles relating to the burden of proof can be found to have occurred.

63 Finally, the applicant cannot claim that two circumstances, namely the alleged obsolescence of the seal at issue and the absence of photographs showing the condition of the seal at issue prior to the opening of the door, increased the burden of proof on the Commission. Assuming that the existence of those circumstances is sufficiently established, it will be necessary to examine whether, in the light of the arguments put forward by the applicant relating to those circumstances, the evidence put forward by the Commission substantiates to the requisite legal standard the finding of a breach of seal as referred to in Article 23(1)(e) of Regulation No 1/2003. That examination will be carried out in the context of the analysis of the third, fourth and fifth pleas.

64 It follows from all the foregoing that the first plea must be rejected.

The second plea, alleging infringement of the ‘principle of inquisitorial procedure’

Arguments of the parties

65 Referring to the judgments of the Court of Justice in Joined Cases 56/64 and 58/64 Consten and Grundig v Commission [1966] ECR 299 and Case C-269/90 Technische Universität München [1991] ECR I-5469, the applicant recalls that, under the ‘principle of inquisitorial procedure’, the Commission is required of its own motion to clarify the facts and to examine carefully and impartially all the relevant aspects of the individual case. That ‘principle’ was infringed in this case.

66 Firstly, the Commission should have remedied the ‘manifest uncertainties’ relating to the composition of Synto. The Commission should not have contented itself with asserting that it did not know which cleaning agent the institute had used in its expert analyses (recital 85 to the contested decision). In institute expert opinion II, the institute established the presence of the component 2-(2-Butoxyethoxy)ethanol in Synto. That component attacks a large number of organic substances. Conversely, Mr Kr. did not himself carry out an analysis of the composition of Synto, but was of the view that the cleaning product consisted of ‘a surfactant watery dilution containing elements of 2-Butoxyethanol and 2-Propanol (isopropyl alcohol)’. Such a substance produces only effects similar to those of an alcohol whereas the substance analysed by the institute also produces the same effects as an ether and, therefore, an additional solvent effect vis-à-vis inter alia adhesives and felt-tip pen marks. The results of the institute’s expert analyses should have prompted the Commission to carry out additional analyses of the composition of Synto. The Commission cannot refrain from doing so solely on the ground that the only variant of Synto which is more or less free of water (‘Synto Forte’) is not, according to the manufacturer’s product description, sold in the bottles of one litre which were used by the cleaning company (recital 85 to the contested decision). It is impossible to exclude the possibility that the manufacturer made incorrect statements and/or that the packaging used for Synto changed subsequently.

67 The Commission also failed to take account of the fact that the cleaning product used previously by the cleaning company (Synto Forte) was replaced by Synto shortly before the inspection, which was pointed out to the Commission in the applicant’s reply to the request for information of 19 October 2007. In that regard, it is impossible to discount the possibility that the cleaning company still had some Synto Forte left over. The Commission could easily have carried out an analysis of the cleaning product used, given that the applicant offered to send it some of the remaining contents of the bottle.

68 Secondly, the Commission infringed its obligation to investigate by not inquiring into the alleged possibility that the keyholders gave access to room G.505 to third parties and into the alleged possibility that someone entered that room in another way. The Commission thus disregarded, in recitals 98 and 100 to the contested decision, the fact that the door of the room in question had been not only sealed, but also locked to prevent any intrusion. The affidavits of the keyholders show that, during the night in question, the door of the room concerned was neither unlocked nor opened. The applicant requests that the necessary proof be adduced by the hearing of oral testimony from those persons, in accordance with Article 65(c) of the Rules of Procedure.

69 In so far as the Commission asserts that other persons could have obtained a key to room G.505 from the keyholders (recital 98 to the contested decision), the applicant maintains that, having regard to its obligation to clarify the facts completely, the Commission should have demanded that the affidavits be supplemented or itself inquired into the location of the keys.

70 Similarly, in so far as the Commission asserts that the affidavits do not exclude the possibility that ‘the door might have been opened by other means’ (recital 98 to the contested decision), it should have carried out investigations into the lock and the door of room G.505, which might have established breaking and entering or attempted tampering of some kind. An analysis of the surface of the door would have made it possible to establish that the opening of the door by other means could be ruled out.

71 The applicant adds that it cannot be assumed that it knowingly encouraged one of the keyholders to damage the seal at issue and/or to open the door. It submits that, under German law, by giving a false affidavit, that third party would be liable to prosecution and possibly to the payment of heavy damages.

72 Thirdly, by its framing of question No 6 in the questionnaire for the inspectors, which precluded or influenced the reproduction of the inspectors’ own observations, the Commission infringed the ‘principle of inquisitorial procedure’.

73 The Commission submits that the present plea must be rejected since the applicant merely makes general comments without showing how the complaints raised could affect the lawfulness of the contested decision. In the alternative, it disputes the applicant’s arguments.

Findings of the Court

74 As was recalled in paragraph 48 above, in the field of competition law, it is for the Commission to prove the infringements found by it and to adduce evidence capable of demonstrating to the requisite legal standard the existence of the circumstances constituting an infringement. For that purpose, it must gather sufficiently precise and consistent evidence to support the firm conviction that the alleged infringement took place.

75 It should also be noted that, for the purpose of complying with the principle of good administration, the Commission must play its part, using the means available to it, in ascertaining the relevant facts and circumstances (Consten and Grundig v Commission, paragraph 65 above, p. 347).

76 The guarantees conferred by the Community legal order include, in particular, the duty of the competent institution to examine carefully and impartially all the relevant aspects of the individual case (see, to that effect, Technische Universität München, paragraph 65 above, paragraph 14, and Case T-167/94 Nölle v Council and Commission [1995] ECR II-2589, paragraph 73).

77 In that regard, it must be noted as a preliminary point that the applicant’s plea seeks to show that the Commission did not examine the relevant aspects of the individual case, in particular by not remedying the uncertainties relating to the composition of Synto and by not inquiring sufficiently into the possible access to room G.505. However, such shortcomings, assuming that they may, in some circumstances, diminish the probative value of the evidence accepted by the Commission in the contested decision, could affect the lawfulness of that decision.

78 In the first place, as regards the applicant’s argument that the Commission allowed uncertainties to subsist regarding the composition of the cleaning product used on 30 May 2006, firstly, it must be pointed out that, in recital 85 to the contested decision, the Commission did not simply state that it did not know which product the institute had used for its tests. In that recital, on the one hand, the Commission stated that it was apparent from Kr. report I and Kr. report II that the effect of Synto on the surface of the seal at issue could not have had any impact on its functioning. On the other hand, it rejected the applicant’s assertion that Mr Kr. had not used the original cleaning product in his tests.

79 First of all, the Commission did indeed state that it had obtained from the cleaning company itself exactly the same detergent as that which had been used in the applicant’s rooms during the night of 29 to 30 May 2006 and that only that cleaning product had been used in the different test series. The fact also remains that the assertion, contained in recital 85 to the contested decision, that the Commission did not know which product the institute had used for its tests, is in reply to the applicant’s argument that the institute tested the product which had been sent to it by the applicant and observed that it was a water-free solvent with 2-(2-Butoxyethoxy)ethanol as its main component. However, according to the manufacturer’s product description, Synto Forte, the only variant of Synto which is more or less free of water, is not sold in the bottles of one litre which were used to clean the door of room G.505 and is not used as a cleaning product, but as a stain remover.

80 Secondly, the Commission was not obliged to analyse the composition of Synto, since, in carrying out its tests, it used the Synto used by the cleaning company on the door of room G.505, which was sent to it directly by the cleaning company, a fact which the applicant did not dispute when questioned on this point at the hearing. In addition, it is apparent from the letter of 5 September 2006 sent by the cleaning company to the Commission, and in particular from the reply given by that company to the Commission’s second question, that Synto was indeed used to clean the door of that room. Finally, the safety data sheet on Synto does not mention the presence in Synto of the component. 2-(2-Butoxyethoxy)ethanol.

81 Thirdly, the applicant does not dispute that, according to the information on the manufacturer’s internet site, Synto Forte is not sold in the bottles of one litre which were used by the cleaning company. In that regard, the applicant’s arguments to the effect that it is impossible to exclude the possibility that the manufacturer made incorrect statements or that the packaging changed subsequently are unconvincing and, in any case, unsubstantiated.

82 Fourthly, the applicant’s arguments to the effect that it is not inconceivable that the cleaning company had left over some of the more aggressive variant of Synto, which was supposedly used previously, must also be rejected. On the one hand, the applicant does not explain why that variant, which is more harmful to wood surfaces, would have been used to clean the doors of its rooms. On the other hand, it is apparent from recital 85 to the contested decision that the Commission obtained from the cleaning company itself exactly the same cleaning agent as was used in the applicant’s rooms on 30 May 2006 and used only that product in the numerous series of tests. That assertion is not disputed by the applicant.

83 Since the cleaning agent on which the expert authorised by the Commission carried out his tests was precisely the cleaning agent used by the cleaning company during the night of 29 to 30 May 2006, the Commission had no reason to analyse its composition.

84 In the second place, the applicant submits that the Commission infringed the ‘principle of inquisitorial procedure’ by not inquiring into the possibility that the keyholders gave access to room G.505 to third parties and into the possibility that someone entered that room in another way.

85 Under Article 23(1)(e) of Regulation No 1/2003, the Commission may impose fines where, intentionally or negligently, ‘seals affixed … by officials or other accompanying persons authorised by the Commission have been broken’. Thus, under that provision, the onus is on the Commission to prove the breach of seal. On the other hand, it is not its responsibility to demonstrate that the room which had been sealed was actually entered or that the documents stored there were tampered with. In this case, it is apparent from recitals 74 to 76 to the contested decision that the Commission did indeed consider that the seal at issue had been broken. For that purpose, the Commission stated inter alia (recital 74) that ‘[t]he condition of the seal on the morning of 30 May 2006 point[ed] clearly to the conclusion that the seal was removed from the office door during the night, thus allowing the office door to be opened during that period’. In the light of the foregoing considerations, the applicant’s assertions to the effect that the door of the room concerned was neither unlocked nor opened during the night in question, which are purportedly substantiated by the keyholders’ affidavits, are irrelevant.

86 In any event, as pointed out by the Commission, the keyholders’ affidavits, made between 2 September and 22 December 2007, that is, nearly a year and a half after the facts, cannot alter its conclusion in the contested decision as regards the existence of a breach of seal, since, as is clear from the applicant’s replies to its request for information, other persons potentially had access to a key allowing the door of room G.505 to be opened. The Commission was therefore not obliged to inquire into any possibility that the keyholders gave access to the room to third parties or entered room G.505 in another way.

87 In the third place, the applicant clams that the Commission infringed the ‘principle of inquisitorial procedure’ by the framing of question No 6 in the questionnaire for the inspectors, which precluded or influenced the reproduction of the inspectors’ own observations.

88 That argument must be rejected. That question sought to ask the members of the inspection team about indicia which pointed to a finding of breach of seal, in particular in the light of the findings recorded in the breach-of-seal report, namely the presence of ‘VOID’ signs over the entire surface of the seal at issue and the presence of adhesive around and on the back of the seal. The framing of the questionnaire therefore did not preclude the reproduction of the inspectors’ own observations.

89 It is indeed clear from the answers given by the inspectors that the latter did in fact indicate, under question No 6, the elements which they had individually remembered in that regard. Thus, for example, Mr Kl. stated that he had immediately had the impression ‘that the seal had been altered since its affixing [and that he had] recorded in writing the observations allowing such a conclusion and … [had] attached them as an annex to the breach-of-seal report’. Mr Ko. stated that he had ‘noticed that the seal was “displaced” and that the “VOID” sign was visible’, but that he had ‘not looked at the back of the seal’. Mr L. for his part indicated that he had ‘personally checked the condition of the seal on the following day’, noting that it ‘was displaced by approximately 2 mm’, but that he had ‘not paid particular attention to the presence of a “VOID” sign on the seal’. Mr N. also indicated that he ‘clearly recall[ed] having observed personally that the “VOID” sign was visible over the entire surface of the seal’ and that ‘the traces of adhesive on the door, in immediate proximity to the edge of the seal, also seemed to point to a breach of seal’. Finally, Mr M. noted that ‘[t]he description in the report [was] accurate’ and that he ‘w[ould] word point (b) more specifically as follows: traces of adhesive on two edges of the seal; these were 1 to 2 mm long fragments of the “VOID” sign’.

90 The second plea must therefore be rejected.

The third plea, alleging an erroneous assumption that the seal was affixed properly

Arguments of the parties

91 The applicant claims that the Commission was wrong to consider, in recital 5 to the contested decision, that it was established that the seal at issue was intact when it was affixed to room G.505 on 29 May 2006 and that it was affixed tightly to the door and door frame when the inspection team left the premises at about 19.30 hrs.

92 Firstly, it is not undisputed that the seal at issue was affixed properly to the door. In its reply to the statement of objections, the applicant already disputed the perfect adhesion of the seal at issue to the surface of the substrate in question. The content of the file supports, at the very most, the conclusion that, in accordance with the superficial impression of the inspectors who were present when the seal was affixed, the seal at issue was stuck to the surface of the substrate in question, which is not sufficient to justify the finding that it was properly affixed. The applicant points out that the technical bulletin explains that this type of seal adheres to various specified surfaces provided that they have been cleaned beforehand. However, no such cleaning was carried out. In addition, the door concerned is made of painted sound-insulating door leaf panels and the frame of anodised aluminium, materials which are not listed in the technical bulletin.

93 Nor is it established that the seal at issue was detached from its protective liner in accordance with the manufacturer’s instructions. Even 3M acknowledges the possibility of prior damage if the seal were mishandled (recital 60 to the contested decision). It is clear from institute expert opinion III that failure to affix a seal in accordance with the manufacturer’s instructions does not necessarily result in the immediate appearance of ‘VOID’ messages on the seal. Consequently, the assertion that any sign of inadequate adhesion of the seal at issue would have been noticed immediately by those present when the seal was affixed (recital 54 to the contested decision) is irrelevant. In addition, the applicant disputes the fact that the seal at issue was removed properly from its protective liner, that it was able to be affixed without any problem, that it adhered to the door and door frame of room G.505 intact and without any visible ‘VOID’ messages and that it was carefully observed by certain inspectors. The applicant has been unable to verify this and the Commission has not adduced proof of it. It adds that the inspectors’ statements regarding the process of affixing the seal at issue are contradictory.

94 As regards the assertions in recital 56 to the contested decision, the applicant submits that it is unlikely that the manufacturer would give such detailed instructions concerning the handling of the product if they were in any case superfluous. Furthermore, as the manufacturer, 3M has no interest in calling in question the reliability of its product. Consequently, the Commission could not, with sufficient certainty, have established that the seal at issue was ‘intact’ and that it was ‘affixed tightly to the door and door frame of room G 505’ (recital 5 to the contested decision).

95 Secondly, the fact that a representative of the applicant signed the record of sealing on 29 May 2006 confirms only the official affixing of a seal, but not that it was affixed properly, since the applicant was not immediately able to detect any prior damage to or defects in the application of the seal at issue.

96 Thirdly, contrary to what the Commission maintains, its claimed experience and Mr Kr.’s tests are irrelevant. The mere fact that there had allegedly been no problems of adhesion or ‘false positive reactions’ with other seals used since 2004 and originating from the same batch (recital 55 to the contested decision) does not support the conclusion that such a reaction is impossible or unlikely. Mr Kr. himself acknowledged that his analyses did not make it possible to determine to what extent his observations could be generalised. The results of the analyses carried out by Mr Kr. should have been ‘statistically confirmed’.

97 Fourthly, the Commission’s assertion that where the seals are used on ordinary office doors made of (painted) aluminium they can be expected to function properly (recital 56 to the contested decision) is irrelevant, since it has never been established that the door of room G.505 was made of aluminium. Even the frame of the door concerned is not made of painted aluminium, but of anodised aluminium, that is to say, aluminium coated with a protective oxide film for protection against corrosion.

98 The Commission contends that the plea should be rejected.

Findings of the Court

99 It is common ground that on 29 May 2006, at about 19.30 hrs, room G.505 was sealed by means of an official Commission seal. However, according to the applicant, it is not established that that seal was affixed properly. According to the applicant, the sole permissible inference from the superficial impression of the inspectors present during the sealing is that the seal at issue stuck to the surface of the substrate in question. However, it cannot be considered, with sufficient certainty, that the seal at issue, which was intact, adhered firmly to the door and door frame of room G.505 on the evening of 29 May 2006.

100 It must be pointed out that, in the contested decision, the Commission found that ‘[t]he sealing [had] … [taken] place in a proper manner’, that ‘[t]he seal adhered perfectly to the substrate consisting of the door and door frame and [that], after it had been applied …, [it] did not display any VOID markings on its blue and yellow surface’ (recital 50).

101 It relied, for that purpose, in the contested decision (recitals 5, 50 and 51) on the record of sealing and on the replies of six inspectors, who were present when the seal at issue was affixed, to question No 3 of the questionnaire addressed to the inspectors.

102 In the first place, it must therefore be examined whether the evidence mentioned in the contested decision supported the conclusion that the seal at issue was affixed properly.

103 Firstly, it must be observed that the record of sealing notes, on the one hand, that on 29 May 2006 at 19.15 hrs, room G.505 was sealed in accordance with Article 20(2)(d) of Regulation No 1/2003, and, on the other, that a representative of the applicant, Mr P., was informed of the provisions of Article 20(2)(d) and Article 23(1)(e) of Regulation No 1/2003 and of the possibility of the imposition of a fine, in accordance with Article 23(1)(e) of that regulation, where seals are broken either intentionally or negligently. The record was signed by Mr Kl., a Commission official and leader of the inspection team, Mr J., a Bundeskartellamt official, and Mr P., a representative of the applicant.

104 Contrary to what the applicant claims, the record of sealing demonstrates sufficiently that the seal at issue was affixed properly. The record in question certifies sealing ‘in accordance with Article 20(2)(d) of Regulation No 1/2003’, which was acknowledged by a representative of the applicant by affixing his signature, after having been informed of the relevant provisions. Only a proper sealing can be regarded as being in accordance with Article 20(2)(d) of Regulation No 1/2003.

105 In any event, if, on the evening of 29 May 2006, the applicant had observed an irregularity in the affixing of the seal at issue or had observed the appearance of ‘VOID’ signs on it, it may be assumed that it would immediately have made comments in that regard, given that it was perfectly aware of the significance of such signs (see also recital 51 to the contested decision). In addition, as was pointed out by the Commission, the applicant’s security service stated that, on checking the seal at issue, several hours after it had been affixed, no alteration in the seal had been observed during two patrols in building G. The applicant’s argument that the ‘VOID’ messages could have appeared only later, due to prior damage to the seal at issue caused by mishandling, must therefore be rejected.

106 Secondly, it must be observed that the replies of six Commission and Bundeskartellamt inspectors present during the sealing, to which the Commission refers in recitals 5 and 50 to the contested decision, confirm that the seal at issue was affixed properly.

107 Thus, Mr Kl., a Commission official and leader of the inspection team, stated that he was ‘absolutely certain that the seal was intact [… and that he had] personally taken special care in satisfying [him]self that it was’. He added that the seal at issue ‘adhered firmly to the door and door frame and [that] no “VOID” messages were visible’.

108 Similarly, Mr L., a Commission official, stated that he ‘[was] certain that the seal was intact when [the inspectors] left the building [and that they had] again had a good look at the seal and checked that it was well installed’.

109 Ms W., a Commission official, confirmed for her part that ‘[i]t was certain [that the seal] adhered correctly to the door[, that the] affixing ha[d] been correctly carried out [… and that] the seal seemed “normal”’. She added that the seal at issue ‘ha[d] been correctly placed on the door, [that it] had the normal dark blue and yellow colours [and that] no “VOID” message was visible’.

110 Mr N., a Bundeskartellamt official, explained that he had ‘personally observed that the seal was intact when the inspection team left the building [… and that he had] taken an even closer look at it’.

111 Mr M., a Bundeskartellamt official, stated that, ‘after Mr [Kl. had] affixed the seal, several officials, including [himself], had checked that the seal was affixed correctly’. He added that, ‘[w]hen the inspection team, including the Bundeskartellamt officials, left the corridor where the sealed room is situated, the seal was intact’. He claimed to have ‘observed [it] for [him]self’.

112 Finally, Mr B., a Bundeskartellamt official, confirmed that ‘Mr [Kl.], and other members of the team as well as [him]self, [had] checked that the seal had been correctly affixed [and that he had] personally observed that the seal was intact when the inspection team left the building’.

113 Thirdly, it must be observed that the replies to question No 3 in the Commission’s questionnaire of the other four inspectors, who had participated in the inspections at the applicant’s premises, do not call in question the probative value of the abovementioned evidence. One inspector claims not to have participated in the sealing whereas three inspectors provided further indications that the seal at issue was affixed properly and therefore confirmed the essential content of the evidence on which the Commission relied in the contested decision.

114 Thus, Mr K. stated that, ‘when [he had seen] it for the last time, the seal was intact’. Mr Me. confirmed for his part that ‘[he had] checked, at least mechanically, that the seal was intact’. Finally, Mr J. maintained that, ‘[a]s far as he remembered, the seal was intact when the inspection team left that wing of the building on 29 May 2006’.

115 It follows from the foregoing that the evidence on which the Commission relied in the contested decision supported the finding that the seal at issue had been affixed properly on 29 May 2006, that it therefore adhered to the door and door frame of room G.505 and that it was intact, inasmuch as it was not showing ‘VOID’ messages at the time when the inspection team left the applicant’s premises.

116 In the second place, it must be examined whether the circumstances alleged by the applicant are such as to call in question the probative value of the abovementioned evidence. The applicant refers for that purpose to the fact that, firstly, the door and door frame of room G.505 were not cleaned before the seal at issue was affixed, secondly, the materials of that door and door frame are not listed in the technical bulletin and, thirdly, it is not established that the seal at issue was detached from its protective liner in a manner complying with the manufacturer’s instructions.

117 Firstly, it must be observed that, in the contested decision, the Commission explained that its representatives and those of the Bundeskartellamt had ‘checked that the [substrate] was clean, so that particular cleaning of the relevant door and door frame was not necessary’ (recital 49).

118 Although it is, admittedly, recommended in the technical bulletin that the surface of the substrate be cleaned before the seal is affixed, the fact remains that that recommendation is linked to the fact that a contaminated surface could affect the adhesive force of the seal, with the result that the ‘VOID’ signs could fail to appear in the event of a breach of seal. The technical bulletin expressly states that ‘[a]ny surface contamina[nt] will adversely affect the adhesion [of the seal] and the destruct message’. The seal manufacturer, 3M, also expressly confirmed that the recommendation relating to prior cleaning of the surface of the substrates referred to primarily concerned cases in which such a surface is contaminated by oil or grease. Dust normally found in an office does not, according to the manufacturer, affect the performance of the seals.

119 However, the applicant, on which the burden of proof of the circumstances alleged by it rests, has not established that, on the evening of 29 May 2006, the door and door frame of room G.505 were covered with contaminants other than dust normally found in an office. Nor, moreover, has it established that, on the evening of 29 May 2006, the condition of the surface of the door and door frame in question was such that it could normally have affected the performance of the seal at issue. On the contrary, it may be assumed that the door of room G.505 was properly cleaned by the cleaning company. In those circumstances, the applicant’s first argument must be rejected.

120 Secondly, as regards the applicant’s arguments relating to the materials of which the door and door frame of room G.505 are made, it must be observed that the Commission noted, in recital 56 to the contested decision, that, ‘where the seals are used on ordinary office doors made of (painted) aluminium, they [could] be expected to function properly’. The Commission pointed out that ‘[t]his [had been] unequivocally confirmed by the tests carried out by [its …] appointed expert, both in situ and in the laboratory, on the sealing products and their adhesion behaviour on the substrate actually present’.

121 The applicant asserts that the frame of the door concerned is not made of painted aluminium, but of anodised aluminium. However, the applicant, on which the burden of proof of the circumstances alleged by it rests, does not put forward any evidence which would permit the inference that the fact that the door frame is made of anodised, and not painted, aluminium might have had any effect on the performance of the seal at issue.

122 In any event, 3M stated, in its reply to the Commission’s request for information, that the adhesive used in that type of seal was suitable for practically all substrates, so that the list of possible substrates in the technical bulletin (stainless steel, acrylonitrile-butadiene-styrene (ABS), polypropylene, painted metal, polyester, high-density polyethylene (HDPE), nylon, glass, polycarbonate) was not exhaustive, but was intended to provide an approximate indication of the nature and range of substrates on which the product could be used. 3M also stated that such a seal would function properly on aluminium and painted aluminium doors, while pointing out that, if a seal did not adhere sufficiently to the substrate, the ‘VOID’ messages could not appear if it was displaced, which was not the case here. The observations contained in Kr. report I and those of Mr Kr. of 9 July 2008 also confirm those statements. In those circumstances, the applicant’s second argument must be rejected.

123 Thirdly, as regards the fact that it is allegedly not established that the seal at issue was detached from its protective liner in a manner complying with the manufacturer’s instructions, it is sufficient to observe that the applicant has not adduced any evidence of the truth of that alleged fact, with the result that that argument must be rejected. In any event, according to 3M, any damage to the product when detaching its protective liner would have had the effect of causing the ‘VOID’ messages to appear even before the seal was affixed to the substrate, whereas there is no possibility of a ‘false positive reaction’ occurring subsequently. Kr. report I also rejected the relevance of the speed of separation between the seal at issue and the protective liner as regards the appearance of ‘VOID’ messages or the fact that those messages might appear only after a certain period of time. In those circumstances, the applicant’s third argument must also be rejected.

124 It is apparent from all the foregoing that the third plea cannot be upheld.

The fourth plea, alleging an erroneous assumption of the ‘obvious condition’ of the seal at issue on the day following the inspection

Arguments of the parties

125 The applicant submits that the Commission was wrong to consider, in the contested decision (recitals 9, 24, 55, 61 and 75), that, on 30 May 2006, ‘VOID’ messages were visible over the entire surface of the seal at issue.

126 Firstly, according to the perception of the applicant’s representatives, the ‘VOID’ messages were only very faintly visible and certainly not over the entire surface of the seal at issue. A Bundeskartellamt official also stated, in reply to the Commission’s questionnaire, that ‘in some places, the “VOID” sign was appearing faintly through the paper’. The members of the inspection team and the applicant’s representatives initially had doubts as to the existence of any alteration of the seal at issue. The applicant proposes for that purpose, that the oral testimony of its lawyer be heard, pursuant to Article 65(c) of the Rules of Procedure.

127 The fact that the inspectors sought to make a comparison of the condition of the seal at issue with that of the seals affixed in other parts of the building (recital 76 to the contested decision) confirms that the ‘VOID’ messages were by no means clearly visible over the entire surface of the seal at issue. In any event, in that regard, there can be no question of a perfectly convergent perception on the part of all the officials present. Furthermore, certain inspectors’ statements conflict with the breach-of-seal report.

128 Secondly, the Commission wrongly relied on the photographs of the part which remained stuck to the frame of the door of room G.505 and which were taken with a mobile telephone only during the afternoon of 30 May 2006. In that regard, the Commission’s assertion that, on the morning of 30 May 2006, the door of room G.505 was opened by the leader of the inspection team without further damaging the seal at issue (recital 10 to the contested decision) is incorrect. The act of detaching the seal at issue from the door would have been bound to damage the seal’s adhesive surface. As the inspection continued intensively from 09.30 hrs and the door of room G.505 was therefore regularly being opened and closed, the seal at issue became detached and was reaffixed on several occasions. It is inevitable that the resultant tensile and shearing stresses caused some displacement of the seal at issue which would have had the effect of making the ‘VOID’ messages appear. Moreover, after the continuation of the inspection, the seal at issue was no longer constantly monitored and it is impossible to rule out the possibility that it may have been touched or tampered with in any other way.

129 Against that background, it is contradictory for the Commission, on the one hand, to take the view that the mere appearance of ‘VOID’ messages is sufficient for it to be considered that there was a breach of seal and, on the other, to maintain that it could not have concluded definitively that a breach of seal had occurred without additional detachment of the seal at issue. It is also contradictory that, according to the statements of certain inspectors, ‘VOID’ messages were visible over the entire surface of the seal at issue, but only after its detachment.

130 As regards its reply to question No 15 in the Commission’s request for information, cited by the Commission, the applicant points out that it concerns only the visibility of the ‘VOID’ messages on the door and door frame of room G.505 after the door was opened by the inspectors and not the visibility of the ‘VOID’ messages on the seal at issue from the time of first observation on the morning of 30 May 2006. By stating that the ‘VOID’ messages are more or less contrasting, depending on the substrate, and appear to be the same colour as the seal at issue, which renders close observation necessary, the Commission confirmed that the ‘VOID’ messages were only very faintly visible, in fragments and not over the entire surface of the seal at issue.

131 The supplementary declaration to the report of 30 May 2006 has no evidential value as regards the condition of the seal at issue on the morning of 30 May 2006. It contains no statement in that regard and merely supplements some of the Commission’s own observations as regards the condition of the seal at issue after the opening of the door of room G.505.

132 Finally, the Commission does not explain how the ‘VOID’ messages appeared over the whole of the seal at issue whereas, in any event, the part of the seal at issue which was stuck to the door frame of room G.505 was not detached. According to the Commission’s account, the ‘VOID’ messages can appear only if the seal is detached (recital 75 to the contested decision). At the same time, it is practically impossible to reaffix the seal in exactly the same place, so that detachment inevitably results in traces of adhesive on the back of the seal (recital 74 to the contested decision). However, no such traces of adhesive were found on the part of the seal at issue on the frame side of the door in question. On the contrary, after the definitive detachment of the seal at issue, the ‘VOID’ messages in that location were intact and there were no residues of adhesive on that part (recital 13 to the contested decision). If, however, the part of the seal at issue which was stuck to the door frame had not been detached, it would then be impossible to explain how visible ‘VOID’ messages appeared on it. The Commission’s account, according to which ‘VOID’ messages were visible over the entire surface of the seal at issue, therefore necessarily implies that this was a ‘false positive reaction’.

133 The Commission disputes the applicant’s arguments.

Findings of the Court

134 It must be observed that, in the contested decision, the Commission considered that ‘[t]he condition of the seal on the morning of 30 May 2006 point[ed] clearly to the conclusion that the seal [had been] removed from the office door during the night, thus allowing the office door to be opened during that period’ (recital 74).

135 It relied for that purpose, in the contested decision, on the breach-of-seal report and on the replies of eight inspectors, who were present at the time of the finding of the breach of seal (recitals 8, 12, 75 and 76). It also observed, in recital 13 to the contested decision, that the applicant’s in-house and external representatives who were present on that day during the inspection had not disputed the alteration in the condition of the seal at issue, but had refused to sign the breach-of-seal report.

136 In the first place, it must therefore be examined whether the evidence relied on by the Commission in the contested decision supported the conclusion that a breach of seal had been found.

137 Firstly, as is clear from recital 7 to the contested decision, unlike a paper seal, a breach of seal in the case of a plastic seal, such as the seal at issue, is not shown by a tear. Once affixed, a plastic seal can no longer be removed from its substrate without its removal remaining visible. It is impossible to reaffix it without leaving traces. Once the product is removed, the white adhesive remains on the substrate in the form of ‘VOID’ messages about 12 Didot points (approximately 5 mm) high distributed over the entire surface of the adhesive label. The removed seal becomes transparent at these points, so that the ‘VOID’ messages are clearly visible on the seal as well. In view, in particular, of the large number and size of the ‘VOID’ messages, it is impossible in the final analysis to affix the seal in exactly the same position as that in which it was located previously. Even if that is accomplished, the ‘VOID’ signs remain clearly visible.

138 In that regard, it must be observed that the breach-of-seal report (see paragraph 9 above) states, on the one hand, that the entire seal was displaced by about 2 mm upwards and sideways so that traces of adhesive were visible at the lower and right-hand edges of the seal, and, on the other, that the ‘VOID’ message was clearly visible over the entire surface of the seal, which was nevertheless still affixed crosswise from the door to the door frame of room G.505 and was not torn. Thus, contrary to what the applicant maintains, the findings in that report, signed by Mr Kl., a Commission official and leader of the inspection team, and by Mr J., a Bundeskartellamt official, demonstrate sufficiently the existence of a breach of seal.

139 Secondly, it must be observed that the replies of the eight inspectors present during the finding of the breach of seal, to which the Commission referred in recital 75 to the contested decision, confirm that conclusion.

140 Thus, Mr Kl., a Commission official and leader of the inspection team, stated that ‘this [had] immediately [given] the impression that the seal had been altered since its affixing’. He indicated that he had ‘recorded in writing the observations authorising such a conclusion’ and that he had ‘attached them as an annex to the breach-of-seal report’.

141 Similarly, Mr Ko., a Commission official, stated ‘[that he had] noticed that the seal was “displaced” and that the “VOID” message was visible’.

142 Ms W., a Commission official, also confirmed that she had noticed that ‘the seal did not have the same appearance as on the previous evening’, that the ‘VOID’ message was legible over the entire surface of the seal at issue and that there were 2 mm traces of adhesive on the perimeter and on the back of the seal at issue next to the ‘VOID’ messages. She added that ‘the seal was not the same dark blue colour as before because the “VOID” messages were visible’.

143 Mr N., a Bundeskartellamt official, explained that he ‘recall[ed] perfectly that the “VOID” message was visible over the entire surface of the seal [and that] the traces of adhesive on the door, in the immediate proximity of the edge of the seal, also seemed to point to a breach of seal’.

144 Mr M., a Bundeskartellamt official, stated that ‘[t]he description contained in the report [was] accurate [and that he] w[ould] word point (b) more specifically as follows: traces of adhesive on two edges of the seal; these were 1 to 2 mm long fragments of the “VOID” sign’.

145 Similar observations were made by Mr Me., Mr J. and Mr B., Bundeskartellamt officials.

146 It follows from the foregoing that the evidence on which the Commission relied in the contested decision (recitals 8, 9, 74 and 75) supported the conclusion that the seal at issue had been removed from the door of room G.505 during the night of 29 to 30 May 2006 and that that door could therefore have been opened during that period, in the light of the ‘VOID’ messages over the whole surface of the seal at issue and of the presence of traces of adhesive close to and on the back of that seal on the morning of 30 May 2006. There is therefore no need to rule on the applicant’s claim that the ‘VOID’ messages on the door frame were completely ‘intact’, which means that the part of the seal at issue located on the door frame, although it had not been detached, showed ‘VOID’ messages and pointed to a ‘false positive reaction’. In any event, as the Commission pointed out, it has never adopted the views expressed by the applicant in point 2 of the supplementary declaration of 31 May 2006 (see paragraph 12 above). In addition, the mere assertion by the applicant that ‘the “VOID” message on the frame was completely intact’ cannot, in itself, demonstrate the existence of a ‘false positive reaction’ of the seal at issue.

147 In the second place, it must be examined whether the circumstances alleged by the applicant are such as to call in question the probative value of the abovementioned evidence. The applicant refers for that purpose to the fact that the ‘VOID’ messages were only very faintly visible, and certainly not over the entire surface of the seal at issue, and to the fact that the Commission wrongly relied on the photographs, taken on the afternoon of 30 May 2006, of the part of the seal at issue which remained stuck to the frame of the door of room G.505.

148 Firstly, in order to maintain that the ‘VOID’ messages were only very faintly visible, and only on one part of the seal at issue, the applicant relies on the fact that a Bundeskartellamt official stated that, in some places, the ‘VOID’ sign was appearing faintly through the paper and on the fact that the members of the inspection team and the applicant’s representatives initially had doubts as to the existence of an alteration of the seal at issue, which is purportedly attested by a visual comparison which was made between the condition of the seal at issue and the condition of the seals affixed in other parts of the building. Furthermore, certain inspectors’ statements allegedly conflict with the breach-of-seal report.

149 First of all, it must be stated that the applicant does not dispute that the ‘VOID’ messages were in fact visible on the seal at issue or, at least, on one part of it, during the morning of 30 May 2006. Nor does the applicant dispute that there were traces of adhesive below and to the right of it. As 3M confirmed in its reply to the Commission’s request for information of 5 September 2006, the appearance of the ‘VOID’ messages indicated that the adhesive label had been displaced. It follows that the Commission was fully entitled, in the breach-of-seal report, to find that the seal at issue had been broken. Furthermore, as the Commission has pointed out, although the findings relating to the appearance of the ‘VOID’ messages were set out in the breach-of-seal report, the applicant did not, in its supplementary declaration to the breach-of-seal report, make any observations in that regard (recital 13 to the contested decision).

150 Next, the condition of the seal at issue was confirmed by the statements of the eight inspectors present on the spot (see paragraph 139 above). In that regard, the applicant’s assertion that the statements of certain inspectors contradict the findings of the breach-of-seal report cannot be accepted. Thus, neither the statement of Ms P., that, ‘in some places, the “VOID” sign was appearing faintly through the paper’, nor the assertions of Mr L., that he saw square marks on the door on the left-hand side of the seal at issue and did not pay particular attention to the fact that a ‘VOID’ sign was appearing on the seal at issue, do not call in question the finding of the appearance of the ‘VOID’ messages on the seal at issue, and are not such as to invalidate either the findings contained in the breach-of-seal report or all the other inspectors’ statements to which reference is made in paragraphs 140 to 145 above.

151 Finally, as regards the comparison between the condition of the seal at issue and the condition of the seals affixed in other parts of the building, which points, according to the applicant, to the existence of doubts regarding the alteration of the seal at issue, it must be held that, as the Commission explained in recital 76 to the contested decision, since this case of a breach of seal was the first and was not a breach of seal by tearing, it seems reasonable that the inspectors checked by comparing it with the other seals. In any event, the fact that the inspection team conducted a comparison between the seal at issue and the seals affixed in other parts of the building cannot affect the findings relating to the physical condition of the seal at issue, as recorded in the breach-of-seal report, and, therefore, the applicant’s claim is irrelevant.

152 Secondly, as regards the argument that the Commission wrongly relied, in order to establish the condition of the seal at issue, on photographs taken with a mobile telephone on the afternoon of 30 May 2006, it is clearly based on an erroneous premiss.

153 It is apparent from recitals 74 and 75 to the contested decision that, in order to establish that a breach of seal had occurred, the Commission relied on the condition of the seal at issue on the morning of 30 May 2006, which showed inter alia ‘VOID’ messages over its entire surface. As regards the proof of that finding, the Commission stated, in recital 76 to the contested decision, that the leader of the inspection team and a representative of the Bundeskartellamt had drawn up the breach-of-seal report in the presence of representatives of the applicant. It also added that the condition of the seal at issue, which is described in that report, in particular the appearance of the ‘VOID’ messages over a large part of its surface, had been unanimously confirmed by the inspection team when questioned on that point. As pointed out in paragraph 146 above, that evidence is sufficient to prove the infringement.

154 Against that background, without there being any need to rule on the evidential value of the photographs of the seal at issue, the applicant’s arguments regarding the fact that those photographs were taken by the Commission after the door of room G.505 had been opened are not such as to call in question the probative value of the evidence mentioned in paragraph 153 above and must be rejected.

155 It follows from the foregoing that the applicant, on which the burden of proving the circumstances alleged by it rests, has not established the unlawfulness of the finding of the breach of seal on the morning of 30 May 2006.

156 In those circumstances, the applicant’s fourth plea must be rejected.

The fifth plea, alleging an erroneous assumption of the suitability of the security label for official sealing by the Commission

Arguments of the parties

157 The applicant submits that the Commission erred in considering that the security label was suitable for official sealing in an investigation procedure.

158 Firstly, the security label was designed to prove that a ‘secured container or product’ has under no circumstances been opened. In that regard, the user of a security label accepts that, in the event of a positive reaction of the security label, it is not possible to know, a posteriori, whether it is a case of a ‘false positive reaction’ or whether there has actually been tampering with the product in question.

159 By contrast, in proceedings conducted pursuant to Regulation No 1/2003, it is not appropriate to make the undertaking concerned bear the risk of a ‘false positive reaction’, in particular in the light of the fines incurred for breaches of seal. The Commission should use a label in regard to which the appearance of ‘false positive reactions’ is precluded from the outset.

160 The Commission cannot rely on the assertion by 3M that, to date, it has not had a single complaint about poor performance of labels of the security label type (recital 55 to the contested decision), since the users of such labels would have reason to submit a complaint only in the event of a ‘false negative reaction’.

161 Secondly, the applicant maintains that the date of expiry of the seal at issue was exceeded. However, the institute’s expert opinions show that the more the security label ages, the less reliable it is and the more sensitive to ‘external influences’ it becomes.

162 Even though, in its reply to the request for information of 8 December 2006, 3M intimated for the first time that the security label is likely to function correctly after a storage period of more than two years, 3M avoids taking a definitive view on the length of the security label’s maximum shelf life. In addition, by minimising the effects of time on the security label, 3M fails to have regard to the current state of science and technology in the field of signs of ageing on pressure-sensitive adhesives. In any event, information supplied by the manufacturer cannot provide any proof as to the effectiveness of the seal at issue at the precise material time and cannot take the place of an assessment by a ‘neutral expert’. The results of Mr Kr.’s expert assessments do not remove the doubts regarding the effectiveness of the seal at issue, since they fail to take into account not only the effect of time or Synto on the seal, but also the consequences of its being under tension for a long period in the area of the crack between door and door frame, in the event of the door being subjected to sustained vibrations and simultaneous shearing stresses.

163 As regards the Commission’s criticisms relating to the ageing simulation used by the institute, the applicant claims that it satisfies scientific requirements. Ageing accelerated by a slight rise in temperature was necessary, otherwise ageing could not be measured.

164 Thirdly, 3M’s refusal to acknowledge the possibility of the existence of ‘false positive reactions’ (recital 68 to the contested decision) is irrelevant and in any event unsubstantiated. The possible absence of customer complaints about ‘false positive reactions’ is not sufficient to justify the claim that such reactions are impossible. In addition, 3M’s statements relating to the unlimited shelf life of the seal are not credible and contradict the particulars in the technical bulletin. Moreover, it is clear from that bulletin that, where there is a risk of considerable economic damage, the security label is not, on its own, a suitable means of security. In that regard, 3M recommends the use of additional means of security in cases where serious consequences could arise and the Commission could therefore have used several seals per door. The Commission should be held responsible for the consequences of the fact that no additional measures were adopted by it and for the problems of proof arising from that.

165 The Commission disputes the applicant’s arguments.

Findings of the Court

166 By its plea, the applicant seeks to demonstrate that the security label was designed to prove that a ‘secured container or product’ has under no circumstances been opened. It claims that the seal is nevertheless not suitable for official sealing in the context of a Commission investigation in the field of competition law.

167 In the first place, it should be noted that it is apparent from the technical bulletin that this product line was designed to indicate tampering by self-destructing when an attempt is made to remove the label. However, that is precisely the use to which this type of security label is put by the Commission in its investigations. Admittedly, under the heading ‘Application ideas’, 3M suggests, for such a security label, the following applications: ‘non-transferable labels for automotive, appliance and electronics industries; tamper-indicating labels and seals for over-the-counter drugs and other packaging applications’. However, the fact that the use by the Commission of the security label in its investigations does not expressly appear in the technical bulletin cannot be construed as excluding such a use of the security label, since the list of applications suggested by the manufacturer is not exhaustive. In any event, the applicant does not adduce proof that the use of this type of security label in the context of such investigations is inappropriate.

168 Although, admittedly, as the applicant points out, the manufacturer recommends the use of additional methods of security in combination with its product in instances where ‘the consequences of tampering could be severe, such as … significant monetary loss’, it must be stated that, as is apparent from the safety data sheet, that type of measure is recommended to the extent that 3M cannot exclude the possibility of a ‘false negative reaction’.

169 In the second place, as was stated in paragraph 103 above, the record of sealing refers to the affixing of the seal at issue, in accordance with Article 20(2)(d) of Regulation No 1/2003, and to the possibility of the imposition of a fine, in accordance with Article 23(1)(e) of that regulation, where seals have been broken intentionally or negligently. In addition, a sealing in accordance with Article 20(2)(d) of Regulation No 1/2003 was recognised by a representative of the applicant by signing the record of sealing. As was noted in paragraph 104 above, only the affixing of a seal suitable for such a use can be regarded as being in accordance with Article 20(2)(d) of Regulation No 1/2003. Thus, if the applicant had had doubts as to the suitability of the security label used by the Commission for sealing in accordance with the abovementioned provision, it may be assumed that it would immediately have raised objections in that regard at the time of affixing the seal at issue, the significance of which it was well aware. However, the applicant did not make any observations of that type.

170 In the third place, so far as concerns the applicant’s arguments relating to the use beyond the date of expiry of the seal at issue, which allegedly affected its sensitivity to ‘external influences’, to sustained vibrations of the door, to simultaneous shearing stresses and to the use of Synto, it must be held that those arguments do not concern the suitability of the security label for official sealing, but rather the alleged disregard by the Commission of ‘alternative scenarios’ which may have been the cause of the condition of the seal at issue found on 30 May 2006, which is covered by the applicant’s sixth plea. Reference is therefore made to the considerations relating to the examination of that plea.

171 It follows from the foregoing that the fifth plea must be rejected.

The sixth plea, alleging disregard by the Commission of ‘alternative scenarios’ which may have been the cause of the condition of the seal at issue

Arguments of the parties

172 The applicant claims that, having regard to the principle of in dubio pro reo, the Commission has not adduced sufficient proof of a breach of seal.

173 In the first place, the applicant submits that, in the light of the institute’s expert opinions, it has demonstrated that ‘external influences’ other than detachment of the seal at issue may have caused the appearance of ‘VOID’ messages on it.

174 Firstly, the applicant claims that the maximum shelf life of the seal at issue was exceeded. The applicant’s experts demonstrated that the reliability of the security label decreases with the passage of time and that its sensitivity to ‘external influences’ increases as a result. It is established, in this case, that the seal at issue exceeded by at least a year and a half the maximum shelf life stated by the manufacturer.

175 Secondly, the applicant alleges the decisive influence of Synto. In its reply to the request for information of 9 August 2006, the applicant already explained that the cleaning company’s employee did not exclude the possibility that she may have rubbed the seal at issue with a cloth soaked with Synto. Institute expert opinion III demonstrates a reduced performance of the security label and an increased tendency for it to produce ‘false positive reactions’ when that label has previously been rubbed with Synto. In connection with the security label’s being brought into contact with Synto, institute expert opinion II also raises the possibility of a phenomenon of creeping of the security label as a result of tensile and compressive shearing stresses.

176 The applicant also points out that the microfibre cloth used was, according to the statement made by the cleaning company’s employee, extremely wet when it was wiped over the seal at issue, with the result that contact of the latter with a large quantity of Synto cannot be excluded. Contrary to what the Commission suggests, the institute’s tests were carried out with Synto and not with Synto Forte. The name of the product in question does not give any indication of its actual composition in that regard. Whilst, admittedly, institute expert opinion II does contain references to Synto Forte, the detergent used was Synto. The applicant disputes the result of Mr Kr.’s reports, according to which Synto does not alter the effectiveness of the seal. Mr Kr. did not examine how lateral penetration of Synto under the label acts over a prolonged period in the event of tangential forces, which may lead to traces of adhesive next to the seal. He did not exclude the possibility that the combined effect of Synto on the seal’s acrylic adhesive, which is sensitive to humidity, and of a small mechanical load could have led to the formation of ‘VOID’ messages on the seal at issue.

177 Thirdly, the applicant refers to the decisive influence of air humidity. It is clear from institute expert opinion III that an air humidity greater than 60% has a considerable effect on the performance of the security label and causes an increased possibility of ‘false positive reactions’. During the night of 29 to 30 May 2006, the air humidity in Munich was over 80%.

178 The air conditioning installed in the applicant’s building G is not, in principle, switched on in the event of high outside air humidity, in order to avoid the formation of condensation on the cooling ceilings. Moreover, during the period in question, the regulating technology of the air-conditioning installation often posed problems which led to malfunctioning of the automatic shutdown of the supplementary humidification system incorporated into the air-conditioning installation as from a certain level of outside air humidity, which has not been disputed by the Commission.

179 Against that background, the Commission’s assertion that the applicant neither stated nor, a fortiori, proved why the seal at issue was exposed to a high humidity is irrelevant (recital 94 to the contested decision). The Commission itself, during the proceedings, attached decisive importance to air humidity. In its request for information of 19 October 2007, the Commission requested detailed information on the air humidity during the night in question.

180 Fourthly, the applicant alleges that vibrations had a decisive effect. It is apparent from institute expert opinion I that the alleged condition of the seal at issue can also be explained, with sufficient probability, by the vibrations to which the door and walls of room G.505 were subjected as a result of the use of the adjacent rooms, and by a sufficiently large play of the door, even when locked. Those results are, furthermore, illustrated by films which the applicant showed during the hearing on 6 December 2006 before the Hearing Officer. The rooms adjacent to room G.505 were booked for a meeting on the following day. There were also incessant comings and goings during which the door of the adjacent room could have been slammed, which could have caused vibrations. It is also impossible to exclude the possibility that persons who came to the wrong room by mistake or who were not informed of the change of use of room G.505 pulled on the door of that room. The applicant cannot be accused of not having prevented that kind of incident given that the hypersensitivity of the security label and its tendency to produce ‘false positive reactions’ were not foreseeable and that the applicant was therefore entitled to be sure that the locking of the door protected the seal at issue sufficiently. The applicant proposes for that purpose, pursuant to Article 65(c) of the Rules of Procedure, that the oral testimony of an employee of E.ON Facility Management GmbH be heard.

181 Fifthly, the applicant states that a combination of the fact that the maximum shelf life of the seal at issue was exceeded, of the effect of Synto, of the air humidity and of the vibrations probably caused hypersensitivity of the security label, which was the cause of the condition of the seal at issue.

182 Sixthly, it is clear from the investigations concerning the place where the keys to the door of room G.505 were kept and from the behaviour of the keyholders that that door could not have been opened during the night of 29 to 30 May 2006. The applicant requests that proof of this be adduced by hearing the oral testimony of the keyholders, pursuant to Article 65(c) of the Rules of Procedure.

183 In the second place, the applicant submits that the Commission errs in considering that the conditions under which the institute’s experts carried out their tests did not correspond, to a large extent, to the technical circumstances observed on the spot (recital 67 to the contested decision).

184 Firstly, the applicant submits that the Commission’s argument regarding the failure to use the original seals for the institute’s tests is irrelevant (recitals 27, 29 and 35 to the contested decision). The applicant’s expert never conceded that conclusions relating to a seal of a given size can be relevant only to a seal of the same size. He merely observed that resistance values cannot be transposed at will from ‘small to large’. It stresses the fact that proof of a tendency of the seals to show ‘false positive reactions’ does not depend on the observation or measurement of certain absolute values. Given that, in the analyses carried out, the samples used by the institute were always of the same size, its conclusions are correct even though the absolute values may differ. That is also the reason why it is not necessary to use samples of an identical size.

185 The applicant adds that, should the Court hold that the properties of the original seal are decisive, the Commission prevented an adduction of evidence respecting its rights of defence, and that the Commission is therefore not entitled to plead failure on the part of the institute’s experts to use original seals. By letter of 10 October 2006, the applicant asked to be sent original seals (see also recital 21 to the contested decision) which the Commission could easily have invalidated by piercing holes in them, by permanently marking them or by some other way. The risk of falsification would have been minimal and the applicant and its experts could, moreover, have signed declarations of liability. The Commission would only agree to the applicant’s experts carrying out analyses on the original seals if Commission officials were present, which would have been impossible in practice, since the ‘statistically confirmed’ conduct of numerous tests would have necessitated a large number of individual analyses lasting several weeks and needing to be conducted in a laboratory. It is unlikely that the Commission would have been prepared to place a member of staff at the applicant’s disposal for the duration of its tests.

186 Secondly, the applicant was forced, on account of the refusal to make the original seals available, to simulate the fact that the maximum shelf life was exceeded. In that regard, due to the permeability of the protective liner to vapour, the ageing method chosen delivers perfectly reliable results. Similarly, as regards the absence of periodic ‘power peaks’ in the diagrams of institute expert opinions II and III, peaks which were, however, observed in Mr Kr.’s reports (recital 67 to the contested decision), pleaded by the Commission, it is explained by the fact that the institute used testing apparatus equipped with an air-cushioned sliding carriage which largely suppressed the ‘slip-stick effect’ responsible for the formation of such power peaks. In that regard, the applicant requests the hearing of oral testimony from an expert from the institute, pursuant to Article 65(c) of the Rules of Procedure.

187 Thirdly, the lack of reaction of the other seals affixed on 29 May 2006 is irrelevant. Those seals were used in another building which was constructed in a totally different way. Quite apart from the differences existing between the buildings in terms of the construction materials used and the sensitivity of the doors to vibration, the other seals were not necessarily exposed to the same air humidity (recital 92 to the contested decision).

188 In the third place, the applicant claims that Mr Kr.’s reports are not convincing either from a factual point of view or from a technical and scientific point of view.

189 Firstly, Mr Kr.’s reports are based on the incorrect assumption that the photographs taken during the afternoon of 30 May 2006 show the seal at issue in the condition in which it was on the morning of 30 May 2006. However, the photographs were taken only after the door of room G.505 had been opened and closed several times. In the absence of a well-founded premiss, the account has no evidential value.

190 Secondly, Mr Kr. based his assessment on an excessively small ‘play of the door’ of 0.53 mm, whereas the institute found that, between the door leaf and door frame of room G.505, there could have been a minimum play of 2 mm. Consequently, the estimate made by Mr Kr. of the possible amount of stretching of the seal at issue due to vibrations is too small. Furthermore, contrary to what the Commission claims (recital 79 to the contested decision), a 2 mm ‘play of the door’ and the other circumstances which prevailed on the day in question explain why the seal at issue was able to slide on the substrate.

191 Thirdly, it is apparent from the contested decision (recital 91) that the Commission itself acknowledged that the security label is permeable to air humidity. In that regard, first of all, the particulars given by 3M in the technical bulletin are not applicable to the present case since they are merely guidelines concerning the properties of the product and apply only to a stainless steel test surface. Also, the security label was exposed to air humidity not only after it had been detached from its liner and stuck to the door, but also when it was still on its liner, which is made of siliconised paper. Finally, the Commission’s unsubstantiated assertion that, of all the types of adhesive, the acrylates group is the most resistant to humidity is irrelevant, since the institute established that the acrylate adhesive used on the seal at issue is not sufficiently resistant to humidity. Mr Kr.’s expert opinions did not examine whether the effect of humidity could have induced ‘false positive reactions’ of the security label, that is to say, whether ‘VOID’ messages could have appeared even in the absence of ‘external influences’.

192 Fourthly, the applicant states that the tests carried out on the spot cannot be ‘statistically confirmed’, which is essential for a scientifically-based assertion. The tests carried out in Mr Kr.’s laboratory are also irrelevant since most of the tests were carried out on painted metal sheets. In the present state of science, it is recognised that acrylate adhesives behave differently on a painted surface from how they behave on anodised aluminium. Consequently, any conclusion concerning the door of room G.505 is precluded from the outset.

193 Fifthly, the applicant submits that, in Kr. report II, the expert disregarded the possibility that the seal at issue was affixed in such a way that it was under tension at the crack between the door leaf and the door frame. During the onsite inspection on 26 April 2007, Mr Kr. avoided any tension on the seal at issue in the area of the crack between the door and the door frame and also affixed several self-adhesive seals on the door before shaking it, which caused a weakening of the force of movement of the door. According to institute expert opinion I, a self-adhesive seal is sensitive to creeping if forces are exerted over a long period and on several occasions. However, it is not inconceivable that the seal at issue may have been affixed on 29 May 2006 in such a way that it was under tension at the crack between the door leaf and the door frame.

194 Sixthly, the Commission did not take account of the fact that all the security labels affixed by its expert during the onsite inspection were dog-eared after detachment. Thus, obvious signs of damage would have resulted if the seal at issue had been tampered with for the purpose of detaching it. However, the seal at issue did not have any dog-eared corners. The seal at issue cannot therefore have been removed during the night of 29 to 30 May 2006. In that regard, the applicant proposes, pursuant to Article 65(c) of the Rules of Procedure, that the oral testimony of its lawyer and of an E.ON employee be heard.

195 Seventhly, Mr Kr. likewise did not take sufficient account of the possible combined existence of certain effects (such as the fact that the shelf life of the seal at issue was exceeded, the other previous damage, the existence of vibrations, the high air humidity and the effect of a cleaning product). The seal at issue was affixed for approximately 14 hours and was therefore exposed to ‘external influences’ such as air humidity and any vibrations. As regards the application of the cleaning product, Mr Kr. also disregarded the ‘conceivable scenarios’. In particular, he based his assessment only on a period of action of the product of 30 minutes whereas, in the present case, it is not possible to exclude the possibility that the harmful action of the cleaning product on the seal at issue may have lasted longer. Nor is it possible to exclude the possibility that the seal at issue, which remained under tension for a long period, moved very slightly.

196 In the fourth place, the applicant raises the point that ‘false positive reactions’ are possible in accordance with the account of the Commission itself in the contested decision (recitals 7, 74 and 75). According to the Commission, the appearance of the ‘VOID’ messages together with the traces of adhesive on the back of the seal can only be explained by detachment and reaffixing of the seal. It must therefore be concluded by converse reasoning that intact ‘VOID’ messages prove that detachment followed by reaffixing of the seal is ruled out. However, in the supplementary declaration of 31 May 2006, the applicant indicated, without this being disputed by the Commission, that the ‘VOID’ messages on the door frame of room G.505 (and not, as the Commission claims, on the door and door frame; see recital 75 to the contested decision) were completely intact and that they were, therefore, completely intact on removal of the seal at issue prior to resealing on the evening of 30 May 2006 (recital 13 to the contested decision). Consequently, the account of the Commission itself proves that the part of the seal at issue adhering to the door frame of room G.505 was not detached from its substrate during the period in question, but that it showed ‘VOID’ signs.

197 The applicant adds that it is natural that the various spontaneous attempted explanations which it put forward with regard to the condition of the seal at issue on 30 May 2006 were not totally consistent. Neither the employees nor the service providers of the applicant know the actual reason for the alleged alteration of the condition of the seal at issue.

198 The Commission contends that the plea should be rejected.

Findings of the Court

199 As was recalled in paragraphs 55 and 56 above, concerning the burden of proof of an infringement in competition law, where the Commission acts on the basis of direct evidence which is in principle sufficient to demonstrate the existence of the infringement, it is not sufficient for the undertaking concerned to raise the possibility that a circumstance arose which may affect the probative value of that evidence in order for the Commission to bear the burden of proving that that circumstance could not have affected the probative value of that evidence. On the contrary, except in cases where such proof cannot be provided by the undertaking concerned due to the conduct of the Commission itself, it is for the undertaking concerned to prove to the requisite legal standard, on the one hand, the existence of the circumstance which it alleges and, on the other, that that circumstance calls in question the probative value of the evidence relied on by the Commission.

200 As was held in paragraph 146 above, the evidence on which the Commission relied in the contested decision (recitals 8, 9, 74 and 75) supported the conclusion that the seal at issue had been removed from the door of room G.505 during the night of 29 to 30 May 2006 and that that door could therefore have been opened during that period, in the light of the ‘VOID’ messages over the whole surface of the seal at issue and of the presence of traces of adhesive close to and on the back of that seal on the morning of 30 May 2006. As regards the circumstances which, according to the applicant, affect the probative value of that evidence, it must therefore be established whether the applicant has proved to the requisite legal standard, on the one hand, the existence of those circumstances and, on the other, that they call in question the probative value of the evidence relied on by the Commission.

201 In the first place, the applicant maintains that it has adduced proof of the existence of ‘external influences’ which were the cause of the ‘VOID’ messages on the seal at issue, which include the fact that its maximum shelf life was exceeded, the influence of the cleaning product Synto, the air humidity, the vibrations or else a combination of those various factors.

202 Firstly, as regards the applicant’s argument that the ‘VOID’ messages appeared on the seal at issue due to the fact that its maximum shelf life had been exceeded, and without there being any need to rule on the applicant’s argument that the seal at issue exceeded by at least a year and a half the maximum shelf life recommended by the manufacturer, it must be stated that the applicant does not adduce proof of a causal link between any such obsolescence and the appearance of the ‘VOID’ messages on the surface of the seal at issue.

203 In that regard, it must be pointed out that, on the one hand, the seal at issue and the other seals used on the other doors the opening of which was prohibited, were from the same batch (recital 69 to the contested decision). However, only the seal at issue bore the ‘VOID’ messages, which tends to preclude the possibility that the alleged fact that the maximum shelf life of the seal at issue had been exceeded was the cause of those messages. On the other hand, institute expert opinion III produced by the applicant does not in any case prove the existence of a ‘false positive reaction’ where an artificially aged security label is used, but makes reference to ‘significant reduction of the adhesive power of the sealing film aged in an accelerator and [to] an increased sensitivity to the appearance of “VOID” markings’. It follows that the first argument must be rejected.

204 Secondly, as regards the alleged decisive influence of Synto, it must be observed that the applicant does not adduce proof that the use of Synto leads to a risk of ‘false positive reaction’ of the seal.

205 First of all, it must be observed that the applicant’s assertion that it is not inconceivable that the employee of the cleaning company may have wiped the seal at issue with a cloth impregnated with a large quantity of Synto seems to be qualified by the observations of Mr Kr., which are not disputed by the applicant, that the application of Synto over the seal by means of a cloth produces an abrasive effect, with the result that the dark blue colour of the seal comes off onto on the cloth. However, the employee of the cleaning company never mentioned such an alteration of the seal at issue after cleaning the door of room G.505. On the contrary, the cleaning company stated, on 6 September 2006, in reply to a Commission request for information, that the cleaner had not observed any alteration of the seal at issue after cleaning that door. Nor is it apparent from the breach-of-seal report or from the applicant’s supplementary declaration that the inspectors observed any abrasion of the dark blue colour of the seal at issue at the time of the finding of the breach of seal.

206 Next, the expert opinions produced by the applicant do not prove that the use of a cleaning product gives rise to a risk of ‘false positive reaction’ of the seal, since those expert opinions demonstrate only a ‘significantly increased sensitivity’ of the seal. In addition, even assuming that the expert opinions produced by the applicant demonstrate the existence of such a risk, it must be observed that it is not established that the institute’s tests were conducted using Synto, since the results of institute expert opinion II state, on at least one occasion, that Synto Forte was used for those tests. In that regard, it must be remembered that, as was observed in paragraph 80 above, it is apparent from the letter of 5 September 2006 sent by the cleaning company to the Commission, and in particular from the reply given by that company to the Commission’s second question, that Synto (and not Synto Forte) was in fact used to clean the door of room G.505. It is not inconceivable that the use of Synto Forte may have altered the conclusions of the institute’s expert opinions.

207 Finally, it is apparent from 3M’s reply to a Commission request for information that cleaning agents do not, in principle, have any effect on seals. Thus, 3M stated that ‘cleaning products in principle [had] no impact on the label’, that ‘[t]he substrate of the product [was] a polyester which is solvent resistant’ and that ‘[t]he product should withstand exposure to cleaning products in common use’. Although 3M admitted to ‘not [having] carried out specific tests relating to [Synto]’, it stated that, in its opinion, ‘the main risk in using cleaning products would be to affect the surface imprint of the converted product ­- in this instance, the blue and yellow colours on the seal used by the Commission’ and that ‘[c]leaning products should not affect the adhesion of the adhesive layer on the back of the product’. It must be observed that it was precisely only such an abrasive effect that the Commission’s expert found during his tests, as was noted in paragraph 205 above. Furthermore, the Commission’s expert did not find any ‘false positive reaction’ of the seal when Synto was applied to it.

208 In any event, as was stated by the Commission in recital 84 to the contested decision, it was the applicant’s responsibility to inform the cleaning company about the significance and handling of the seal at issue and to make sure that the seal at issue was not, where relevant, broken by its employee, a fortiori since, as is clear from the record of the interview with the cleaning company’s employee, that employee receives a plan specifying the occupancy of the meeting rooms before she starts cleaning them.

209 Thirdly, as regards the alleged air humidity, the applicant produces a document purporting to prove that, during the night of 29 to 30 May 2006, the prevailing humidity in Munich was 80%. However, according to institute expert opinion III, an air humidity greater than 60% would lead to a significant increase in the sensitivity of the security label.

210 In that regard, it is sufficient to observe that, on the one hand, the applicant does not adduce proof that an air humidity of over 60% prevailed in building G during the night of 29 to 30 May 2006. The document produced by the applicant as evidence of the air humidity during the night of 29 to 30 May 2006 concerns only the air humidity outside that building and is therefore not directly relevant for the purpose of demonstrating the humidity which was present in the building in question. As regards the latter, the applicant informed the Commission that it no longer had the data relating to 29 and 30 May 2006. In addition, the document of 14 July 2006, drawn up by the applicant for the attention of its staff, refers to an air humidity reaching 55% outside building G and 50% inside it during the days preceding that date. The applicant’s arguments relating to the regulating technology of the air-conditioning installation or to the malfunctioning of the automatic shutdown of the supplementary humidification system incorporated into the air-conditioning installation are no more convincing, since the applicant does not adduce proof that those events led to a humidity of over 60% in building G during the night of 29 to 30 May 2006. On the other hand, nor does the applicant adduce proof that an increased air humidity leads to the occurrence of ‘false positive reactions’, since institute expert opinion III mentions only an ‘increased sensitivity’ of the security label to air humidity.

211 In any event, the applicant’s claims are contradicted by the particulars contained in the technical bulletin, according to which that product withstands exposure to 90% humidity for a period of 168 hours at a temperature of 32°C, which are confirmed by the findings of the Commission’s expert. It follows that the applicant’s third argument must be rejected.

212 Fourthly, as regards the alleged decisive influence of vibrations, which could explain the condition of the seal at issue, it is sufficient to observe that the applicant does not adduce proof that the door and walls of room G.505 were subjected to vibrations. As the Commission points out, it is impossible to ascertain the conditions under which the video recordings produced by the applicant, intended to demonstrate that vibrations are capable of causing the appearance of ‘VOID’ messages on certain parts of a seal affixed to a closed door, were made, or whether those recordings do in fact show the door of room G.505. In addition, as the Commission pointed out, those recordings do not, in any event, demonstrate the appearance of ‘false positive reactions’, but only the appearance, where vibrations were caused by the alleged violent closing of the door of the adjacent room, of ‘VOID’ messages in the crack between the door and its frame, which does not correspond to the findings contained in the breach-of-seal report.

213 Moreover, the circumstance, raised as a hypothesis by the applicant, that the rooms adjacent to room G.505 were booked for a meeting on the following day and the door of the adjacent room may thus have been slammed, which could have caused vibrations, or else that certain persons coming to the wrong room could have pulled at the door of room G.505, does not explain the condition of the seal at issue on the morning of 30 May 2006.

214 Since the seal at issue was still intact at around 19.30 hrs on the evening of 29 May 2006 (recital 5 to the contested decision) and the inspection team found the alteration of the seal at issue at around 08.45 hrs on the morning of the following day (recital 8 to the contested decision), the hypothesis put forward by the applicant implies that there were ‘incessant comings and goings’ in room G.506, the only meeting room adjacent to room G.505, during the night of 29 to 30 May 2006, which is not demonstrated. In that regard, it is apparent from the office occupancy schedule that room G.506 was occupied on 30 May 2006 only between 10.00 and 16.00 hrs, that is, after the breach of seal had been found. While, admittedly, the applicant stated in its reply to the Commission’s request for information of 9 August 2008 that the door of that room had been opened at around 05.00 hrs. by the security service, the applicant does not allege that the security service closed it violently, which would have caused the alleged vibrations. Moreover, as regards the applicant’s arguments, reaffirmed at the hearing, that a meeting in room G.506 required the installation of heavy equipment from 05.00 hrs, it is sufficient to observe that, in the light of the fact that the room in question was cleaned only at 07.00 hrs. and that that meeting started at 10.00 hrs, such an assertion is not credible.

215 Finally, the applicant’s argument that some people may have come to the wrong room or may not have been informed of the change of use of room G.505 and pulled at its door, thus damaging the seal at issue, cannot succeed, since the door of that room was locked and the applicant does not adduce proof of the causal link between any shaking of that door and the condition of the seal at issue on the morning of 30 May 2006.

216 In any event, even if the circumstances described in paragraphs 212 to 215 above were proved to be true, it was the applicant’s responsibility to take the necessary precautions for the purpose of informing the members of its staff and any visitors present in building G during the night of 29 to 30 May 2006 of the existence of the seal at issue and its handling, in order to avoid any breach of seal. It follows that the applicant’s fourth argument must be rejected.

217 Fifthly, as regards the fact that a combination of the storage beyond the maximum shelf life of the seal at issue, the effect of Synto, air humidity and vibrations may have caused a hypersensitivity of the seal at issue, it must be stated that the very existence of those circumstances, even taken individually, or their impact on the seal at issue has not been demonstrated to the requisite legal standard. Moreover, the applicant does not maintain that the combination of those factors would give rise to a risk of ‘false positive reaction’, but only that ‘prior damage to the security label due to ageing, the product Synto or a combination of those two factors leads, in the event of high air humidity, to a significant increase in the sensitivity of that label’. That argument must therefore also be rejected.

218 Sixthly, as regards the argument that the door of room G.505 was not opened during the night of 29 to 30 May 2006, it must be recalled that, as was pointed out in paragraph 85 above, pursuant to Article 23(1)(e) of Regulation No 1/2003, the burden of proving the breach of seal rests on the Commission. However, it is not for the Commission to demonstrate that the room which was sealed was actually entered. That argument cannot therefore succeed.

219 It follows from all the foregoing considerations that the applicant has not demonstrated the existence of circumstances capable of calling in question the probative value of the evidence on which the Commission relied in order to conclude that a breach of seal occurred.

220 In the second place, since, as is apparent from the foregoing considerations, the expert opinions produced by the applicant do not prove that the circumstances referred to above were capable of giving rise to the condition of the seal at issue as observed on the morning of 30 May 2006, the possible shortcomings of those opinions, alleged by the Commission, are irrelevant in this case.

221 In any event, as pointed out by the Commission, those expert opinions do indeed suffer from a number of shortcomings. Thus, firstly, the tests conducted by the institute were not conducted on original Commission seals, but were conducted on samples of a very small size (4 sq. cm instead of 54 sq. cm). The applicant’s expert himself stated in that context that the resistance values of the seals could not be transposed at will from ‘small to large’. In addition, as was noted by the Commission’s expert, the size of a seal is capable of influencing the results of expert assessments as regards the possible effect of Synto, vibrations or air humidity on the performance of the seal. In that regard, the applicant’s argument that the Commission prevented an adduction of evidence respecting its rights of defence by refusing to send it original seals cannot be accepted. The applicant does not dispute that the Commission offered to provide it with original seals, on condition that Commission staff were allowed to be present at the tests. However, the applicant declined that offer (recital 65 to the contested decision). Against that background, the applicant’s assertion that the Commission was unlikely to be willing to make a member of staff available on the spot for observation purposes for the duration of the tests cannot be accepted either, since any such responsibility lay with the Commission and not with the applicant, so that it cannot be pleaded by the applicant in order to justify the failure to conduct tests on original seals.

222 Secondly, since the applicant refused to carry out its tests on original seals in the presence of a Commission official, it used seals which were artificially aged. In that regard, the seals used by the applicant were stored in a climate chamber in which the samples were exposed for more than 40 days to 60% relative humidity, which exceeds the storage recommendations in the technical bulletin, namely 50% relative humidity.

223 Thirdly, it is apparent from institute expert opinions II and III that the applicant’s expert carried out its tests, or at least some of them, by soaking the strips of plastic film with 100 mg of cleaning agent, which would correspond to a much larger quantity than the quantity of cleaning agent which would, where appropriate, have been used by the cleaner, since she stated that the cloth ‘was extremely wet - but [that] it did not have a large quantity of cleaning product on it’. Moreover, it is not inconceivable that the cleaning agent used by the institute for its tests was not Synto, which was used by the cleaner and was sent to the Commission for its tests by the cleaning company itself, but its variant, Synto Forte. Thus, institute expert opinion II expressly states that certain tests were carried out using the Synto Forte variant. Furthermore, it is apparent from institute expert opinions II and III that the product used for the experiments contains the component 2-(2-Butoxyethoxy)ethanol, which is not present in Synto, but only in its variant, Synto Forte. It is common ground that that component makes Synto Forte a much more aggressive cleaning agent.

224 In the third place, in the light of the fact, on the one hand, that the evidence on which the Commission relied in the contested decision (recitals 8, 9, 74 and 75) supported the conclusion that the seal at issue had been removed from the door of room G.505 during the night of 29 to 30 May 2006 and that that door could therefore have been opened during that period (see paragraph 146 above) and, on the other, that the applicant, on which the burden of proving the facts alleged by it rested, has not adduced proof of the circumstances which it alleges were the cause of the condition of the seal at issue thus established, the alleged weaknesses of Mr Kr.’s reports, pleaded by the applicant, cannot call in question the conclusion set out in paragraph 219 above.

225 In any event, firstly, contrary to what the applicant maintains, the expert opinions of Mr Kr. are not based on the assumption that the photographs taken during the afternoon of 30 May 2006 show the seal at issue in the condition in which it was during the morning of the same day, since the reference to those photographs, which appears inter alia in Kr. report II, constitutes only a supplementary indication of the condition of the seal at issue, as was apparent from the photographs taken on 30 May 2006.

226 Secondly, as regards the alleged ‘play of the door’, which was purportedly too small during Mr Kr.’s tests, it must be pointed out that, at the time of Mr Kr.’s on-the-spot inspection in the presence of the applicant, he found, with the aid of a digital ruler, a ‘play of the door’ of 0.53 mm, without that finding being called in question by the applicant in its observations on the account of the facts or in the reports of institute expert opinions II and III. In any event, on the one hand, as is clear from recital 79 to the contested decision, even assuming that the ‘play of the door’ was 2 mm, the amount of stretching of the seal at issue would have remained very small. On the other hand, the applicant does not put forward the slightest evidence to support the conclusion that a wider ‘play of the door’ would have been capable of giving rise to a risk of ‘false positive reaction’ of the seal at issue.

227 Thirdly, as regards the air humidity, the applicant does not prove that the particulars in the technical bulletin, on which the Commission based a number of its observations, are not applicable in this case, in particular in the light of the material of which the door and door frame of room G.505 are made. Nor does it establish that the security label was sensitised as a result of the presence of humidity in the air at the time of the detachment of the security label. Finally, contrary to what the applicant maintains, it is clear from Kr. report II that even a high humidity could not have caused alteration of the seal at issue.

228 Fourthly, as regards the lack of ‘statistical confirmation’ of Mr Kr.’s tests, it is for the applicant to demonstrate the circumstances which it alleges, with the result that the lack of ‘statistical confirmation essential to a scientifically proven assertion’ is irrelevant. In any event, the Commission indicated that the experimental process using powder-coated metal sheets was particularly prudent as compared with experiments on anodised aluminium.

229 Fifthly, the applicant’s assertion that Mr Kr. disregarded the possibility that the seal at issue may have been affixed in such a way that it was under tension at the crack between the door leaf and door frame of room G.505, which would have resulted in greater stresses on the seal at issue than those assumed in Kr. report II, must be rejected, since that assertion is completely unsubstantiated. In any event, Mr Kr. observed that stretching of the seal at issue would have required considerable force (recital 89 to the contested decision), which is out of the question when a seal is affixed by hand.

230 Sixthly, the alleged absence of dog-ears on the seal at issue cannot demonstrate that it was not removed during the night of 29 to 30 May 2006. Apart from the fact that ‘dog-ears’ do not constitute a characteristic of a breach of seal, it must be pointed out that detachment by hand does not necessarily cause the appearance of dog-ears, since a seal can be removed with care or can also be detached starting from the crack between the door and door frame on which it is placed. This complaint must therefore also be rejected.

231 Seventhly, contrary to what is maintained by the applicant, it is apparent from Kr. report II that the combined influence of the circumstances alleged by the applicant was indeed examined in that report. Consequently, this complaint cannot succeed either.

232 In the fourth place, in the light of the conclusion drawn in paragraph 219 above, there is no need to rule on the alleged ‘possibility of false positive reactions’ which is purportedly apparent from the account of the Commission itself in recitals 7, 74 and 75 to the contested decision.

233 In any event, as was stated in paragraph 146 above, this line of argument must be rejected, since the Commission has never adopted the views expressed by the applicant in point 2 of its supplementary declaration of 31 May 2006 (see paragraph 12 above) and since, moreover, the applicant’s mere assertion that ‘the “VOID” message on the door frame was completely intact’ cannot, in itself, demonstrate the existence of a ‘false positive reaction’ of the seal at issue.

234 It follows that the sixth plea must be rejected in its entirety.

The seventh plea, alleging breach of the principle of the presumption of innocence

Arguments of the parties

235 The applicant claims that, during the administrative procedure, the Commission did not ‘sufficiently observe’ the principle of the presumption of innocence. It points out that, on 16 October 2007, the Commission asked Mr Kr. to submit his observations on institute expert opinions II and III. The Commission’s questions addressed to Mr Kr. for the purpose of preparing Kr. report II were, for the most part, of a ‘leading’ nature, the Commission having thereby infringed its obligation of impartiality, the neutrality of the investigation, the principle of the presumption of innocence and the ‘right to a fair hearing’.

236 Since the Commission did not respect the neutrality of the investigation as required by the principle of the presumption of innocence, there is no point in answering the question whether there are serious doubts as to the credibility and neutrality of the expert (recital 37 to the contested decision), since the complaint of failure to observe the principle of the presumption of innocence is not directed against the Commission’s expert, but against the Commission itself. The mere fact that there are serious doubts as to the Commission’s neutrality is sufficient to prove an infringement of the obligation of impartiality resulting from the principle of the presumption of innocence. Moreover, the expert’s conduct in reformulating the questions cannot compensate for the Commission’s failure to observe the principle of the presumption of innocence.

237 The Commission contends that this plea should be rejected.

Findings of the Court

238 It must be recalled that the principle of the presumption of innocence, as it results in particular from Article 6(2) of the ECHR, is one of the fundamental rights which, according to the case-law of the Court of Justice, reaffirmed by the preamble to the Single European Act, by Article 6(2) EU and by Article 47 of the Charter of Fundamental Rights of the European Union, are recognised in the Community legal order. Given the nature of the infringements in question and the nature and degree of severity of the ensuing penalties, the principle of the presumption of innocence applies inter alia to the procedures relating to infringements of the competition rules applicable to undertakings that may result in the imposition of fines or periodic penalty payments (see Case T-279/02 Degussa v Commission [2006] ECR II-897, paragraph 115, and the case-law cited).

239 The principle of the presumption of innocence implies that every person accused is presumed to be innocent until his guilt has been established according to law (Joined Cases T-22/02 and T-23/02 Sumitomo Chemical and Sumika Fine Chemicals v Commission [2005] ECR II-4065, paragraph 106).

240 In this case, the applicant submits that the Commission breached the principle of the presumption of innocence by sending to Mr Kr., by letter of 16 October 2007, a questionnaire in which the questions were of a ‘leading’ nature.

241 In that regard, it must be observed that the questions addressed to Mr Kr. in that letter were worded as follows (see also recital 36 to the contested decision):

‘1. Please comment on the methods, analysis and conclusions of the [institute] expert opinions sent to the Commission by letters of 6 June 2007 and 1 October 2007. In your statement, explain in particular the reasons why those [institute] expert opinions do not challenge your own expert opinion of 8 May 2007 on the functionality of Commission seals with regard to methods, analyses and conclusions. If you consider further tests necessary to confirm/back up your previous statements, describe them briefly.

2. Please also answer the above question in the light of the arguments/factors which [the applicant] - beyond the [institute] expert opinions - put forward in its letter of 6 July 2007 (for example, lack of statistical relevance of your tests).

3. Please confirm that the combination of the factors/arguments put forward by [the applicant] (or [by the institute]) (amongst others, lack of preliminary cleaning of the substrate, the use of the cleaning agent Synto on the seal, shaking/vibrations of the door, humidity, alleged excessive period of storage of the seals) cannot have led to a false positive [reaction] without the seal’s having been removed from the substrate. Please also confirm that the combination of the factors/arguments put forward is not able to explain the other circumstances witnessed by the Commission on the morning of 30 May 2006 (traces of adhesive around the edge of the seal and on the underside of the seal). If you consider further tests necessary to confirm/back up your statements, describe them briefly.’

242 As is evident from the examination of the fourth plea, the evidence relied on by the Commission in the contested decision (recitals 8, 9, 74 and 75) supported the conclusion that the seal at issue had been removed during the night of 29 to 30 May 2006 and that the door of room G.505 could therefore have been opened during that period, in the light of the ‘VOID’ messages over the entire surface of the seal at issue and of the presence of traces of adhesive close to and on the back of it on the morning of 30 May 2006. Against that background, as is evident from the examination of the first plea, it was for the undertaking concerned to prove to the requisite legal standard, on the one hand, the existence of the circumstances alleged by it and, on the other, that those circumstances call in question the probative value of the evidence relied on by the Commission.

243 In that regard, it must be stated that, as the Commission points out, Kr. report II was drawn up with the aim of establishing whether the conclusions of Kr. report I were called in question by institute expert opinions II and III. In Kr. report I, the Commission’s expert did in fact state that ‘creeping’ of the original seal, which had been raised as a possibility in institute expert opinion I, could not be the cause of a ‘false positive reaction’, regardless of the age, method of affixing and speed of detachment of the seal, even in the event that it had been previously cleaned with Synto and had then been subjected to the effects of shearing and peeling forces.

244 Thus, the Commission’s questionnaire was designed to determine, in the light of the conclusion contained in Kr. report I, whether such a conclusion was called in question by the reports of institute expert opinions II and III. The Commission further pointed out, without being contradicted in this regard by the applicant, that the wording of those questions also resulted from the context in which they arose, namely that Mr Kr. had already made some comments orally on the findings contained in the reports of institute expert opinions II and III and had indicated that he had no reason to doubt the conclusions contained in Kr. report I.

245 Finally, as the Commission pointed out in recital 37 to the contested decision, it is apparent from the wording of the questions by Mr Kr. himself in Kr. report II that the Commission’s expert regarded those questions as open questions, since they had been reformulated as follows: ‘Do the institute’s evaluation reports call in question … the conclusions of the last report?’; ‘Do the arguments put forward by [the applicant’s lawyer] call in question the conclusions of the last report?’; ‘May a combination of the factors/arguments put forward lead to a false positive reaction and explain the circumstances in which the seal was found?’.

246 It follows from the foregoing considerations that the Commission did not breach the principle of the presumption of innocence by the wording of the questions contained in its letter of 16 October 2007 to Mr Kr.

247 The seventh plea must therefore be rejected.

The eighth plea, alleging infringement of Article 23(1) of Regulation No 1/2003

Arguments of the parties

248 The applicant points out that, in the contested decision, the Commission accused it of having negligently committed a breach of seal, but without clarifying what it considered to be negligent conduct.

249 Firstly, Article 23(1) of Regulation No 1/2003 is addressed exclusively to undertakings or associations of undertakings which act through their employees or their representatives whose actions could, therefore, be attributed to them. Referring to the judgment in Joined Cases 100/80 to 103/80 Musique Diffusion française and Others v Commission [1983] ECR 1825, paragraph 97, the applicant submits that the Commission wrongly considered, in recital 101 to the contested decision, that actions of third parties were attributable to it. No member of staff or authorised representative of the applicant opened the door of room G.505, which is clear from the affidavits produced by the applicant. Against that background, the alleged opening of the door or any other act which caused the detachment of the seal at issue constitutes an ‘act exceeding the authority of the persons acting’, which cannot be attributed to the applicant. Whether the cleaning company was warned of the existence of the seal at issue is irrelevant, since it did not have a key to room G.505. In that regard, the purpose of the seal at issue, namely the preservation of the documents in the file which were still to be examined, could have been jeopardised only by the keyholders, so that there was no point in informing the cleaning company of the existence of the seal at issue.

250 Secondly, the complaint of negligence put forward by the Commission is not relevant. There is negligence only if the person concerned could and should have known that he or she was carrying out the act constituting the infringement. The Commission relied solely on alleged alterations of an out-of-date seal used incorrectly.

251 Contrary to what the Commission alleges (recital 102 to the contested decision), the employee of the cleaning company could not have known that the mere act of wiping the seal at issue with a cloth impregnated with an ordinary cleaning product could have led to the appearance of characteristics similar to those of a breach of seal. Nor can the applicant be accused of not having prevented such an act (footnote 176 to the contested decision). The seal at issue did not carry any mention of its possible sensitivity to superficial cleaning and the applicant was not informed of any sensitivity of the seal at issue either at the time of its affixing or in the record of sealing. The applicant can be held responsible only for the unauthorised opening of the sealed door. Even members of the inspection team obviously had no knowledge of how the seal at issue worked.

252 Furthermore, the fact that the Commission did not inform the applicant that the maximum shelf life of the seal at issue had been exceeded cannot be to its disadvantage. In the event that the maximum shelf life of a seal is exceeded, malfunctioning of that seal cannot be ruled out. In German law, the mere fact that the final date for storing a product is exceeded constitutes a defect of the product. Having regard to the fact that the maximum shelf life of the seal at issue had been exceeded and in the light of the fine which might be imposed in the event of a breach of seal, the Commission should in any event have mentioned that circumstance to the applicant.

253 The Commission disputes the applicant’s arguments.

Findings of the Court

254 In the contested decision, the Commission considered that ‘[l]eaving aside cases where forces of nature are involved, it must be assumed in principle that a seal can be removed only by a deliberate action. This is also borne out by the fact that, after it had been removed, the seal was obviously reaffixed in such a way as to conceal the fact that it had been broken’. It added that ‘[a]ccount must also be taken of the fact that only persons authorised by [the applicant] (including the staff of [the cleaning company], a wholly-owned … subsidiary [of the applicant]) were in the building’ (recital 101 to the contested decision). It also considered that it could be ‘concluded that the breach of the seal happened through negligence’, that ‘[i]t must be borne in mind that, when the seal was being affixed, representatives of [the applicant] were informed by the inspection team leader of the significance of the seal and the consequences of any breach of it’ and that ‘[t]his [was] also indicated on the seal itself’ (recital 102 to the contested decision). Thus, the nature of the alteration of the seal at issue led it, in the light of the factors set out inter alia in recital 9 to the contested decision and in the breach-of-seal report, to consider that a breach of seal had been committed intentionally or, at least, negligently.

255 In the present plea, the applicant claims, in essence, that the Commission wrongly considered, in recital 101 to the contested decision, that actions of third parties were attributable to it and that there can be no question of negligence in this case, since the employee of the cleaning company was not in a position to know that she was carrying out the material act constituting the breach of seal.

256 Firstly, it has already been recalled that, under Article 23(1)(e) of Regulation No 1/2003, the Commission is entitled to impose fines where, ‘intentionally or negligently’, seals affixed by officials or other accompanying persons authorised by the Commission have been broken. As was pointed out in the examination of the second plea, it is not for the Commission, pursuant to Article 23(1)(e) of Regulation No 1/2003, to demonstrate that the room which was sealed was actually entered. The applicant’s arguments to the effect that the door of room G.505 was not opened by the keyholders, which is apparent from the affidavits of those keyholders, or by the employee of the cleaning company, since she did not have a key to that room, are therefore irrelevant.

257 Secondly, as regards the applicant’s argument that none of its staff or authorised representatives opened the door of room G.505, which it claims is apparent from the affidavits of the keyholders and from the fact that, against that background, any detachment of the seal at issue would in any event exceed the authority of those persons, it must be held that, in recital 101 to the contested decision, the Commission stated that only persons authorised by the applicant (including the staff of the cleaning company, a wholly-owned subsidiary of the applicant) were in building G. In addition, in recital 103 to the contested decision, the Commission held that the possibility that unauthorised persons may have gained access to the building during the night of 29 to 30 May 2006 could be ruled out and that the applicant had never claimed that an unauthorised person had entered that building.

258 First of all, it must be observed that those statements are not disputed by the applicant in the present action. In addition, the applicant’s staff or authorised representatives must be regarded as performing tasks for and under the direction of the applicant (see, to that effect, Joined Cases 40/73 to 48/73, 50/73, 54/73 to 56/73, 111/73, 113/73 and 114/73 Suiker Unie and Others v Commission [1975] ECR 1663, paragraph 539). In that regard, the applicant’s argument based on the judgment in Musique Diffusion française v Commission, paragraph 249 above, that only keyholders are authorised staff or representatives, must be rejected. As the Commission observes, the Commission’s power to impose a sanction on an undertaking where it has committed an infringement presumes only the unlawful action of a person who is generally authorised to act on behalf of the undertaking (see Joined Cases T-236/01, T-239/01, T-244/01 to T-246/01, T-251/01 and T-252/01 Tokai Carbon and Others v Commission [2004] ECR II-1181, paragraph 277, and the case-law cited).

259 Thirdly, as regards the argument that the employee of the cleaning company could not have known that the act of wiping the seal at issue with a cloth impregnated with an ordinary cleaning product could lead to the appearance of characteristics similar to those of a breach of seal, that argument is based on an erroneous premiss that it has been demonstrated that the condition of the seal at issue on the morning of 30 May 2006 was attributable to the cleaning agent allegedly used by that employee.

260 In any event, even assuming that the condition of the seal at issue may have been altered by that employee using a cloth and a cleaning agent (recital 102 to the contested decision), it must be stated that it was for the applicant to take all the necessary measures to prevent any tampering with the seal at issue, a fortiori since the applicant had been clearly informed of the significance of the seal at issue and of the consequences of a breach of seal (recital 5 to the contested decision).

261 Fourthly, so far as concerns the applicant’s arguments alleging that the Commission relies only on alterations of an out-of-date seal used incorrectly or on the alleged maximum shelf life of the seal at issue, it must be pointed out that those arguments have already been rejected in the context of the examination of the sixth plea.

262 In the light of the foregoing, it must be held that the Commission was fully entitled to take the view that this case concerned at least a breach of seal through negligence.

263 The plea must therefore be rejected.

The ninth plea, alleging infringement of Article 253 EC and of the principle of proportionality when the amount of the fine was set

Arguments of the parties

264 The applicant claims, in the alternative, that the fine imposed on it in the contested decision should be annulled and, in the further alternative, that it should be reduced.

265 In the first place, the fine imposed on the applicant infringes both the ‘prohibition of arbitrary measures’ and Article 253 EC. In the contested decision, the Commission failed to specify the criteria on which it based its determination of the amount of the fine imposed, so that the imposition of a fine of EUR 38 million is incomprehensible, in so far as the Guidelines on the method of setting fines imposed pursuant to Article 15(2) of Regulation No 17 and Article 65(5) [CS] (OJ 1998 C 9, p. 3) provide, even in the case of serious infringements, for a basic amount of only EUR 20 million. In the light of the inadequacy of the statement of reasons for the contested decision, the applicant was unable to exercise effectively its rights of defence.

266 Firstly, in the contested decision (recital 104 et seq.), the Commission merely listed a series of circumstances which either apply to any breach of seal or, in general, constitute aggravating or attenuating circumstances, but without expressing in figures the basic amount or the aggravating or attenuating circumstances, either in absolute terms or as a percentage. Consequently, the Commission failed, with regard to an essential aspect of the contested decision, in contravention of Article 253 EC, to set out the decisive aspects of its assessment.

267 Secondly, the contested decision creates the impression that the Commission started from the assumption of an intentional infringement whereas, in the relevant passages of the contested decision, it held that it was an infringement committed ‘at the very least through negligence’ (recital 102).

268 In the second place, the aggravating circumstances alleged are erroneous from a factual point of view, apply to all cases of breach of seal and amount to nothing more than abstract and general remarks completely unconnected with the specific case (see recitals 105 to 108 to the contested decision). A distinction must be made on the basis of the extent and effects of the breach of seal. The Commission did not adopt a position in this case on the specific circumstances of the breach of seal. The other reasons cited in the contested decision, namely, firstly, the reinforcement of the penalties for breach of seal in Regulation No 1/2003 as compared with the previous provisions, secondly, the circumstance that this was an inspection in the field of competition law, thirdly, the fact that the documents stored in room G.505 had not been photocopied or catalogued, fourthly, the size of the applicant and, finally, the failure to take measures to secure the seal at issue, are irrelevant as regards the determination of the gravity of the infringement.

269 In the third place, the Commission did not take sufficient account of several attenuating circumstances in favour of the applicant, which would have justified a considerable reduction of the amount of the fine.

270 Firstly, a breach of seal through negligence should be penalised by a lower fine than that imposed for an intentional breach of seal. In addition, in this case, the breach of seal is the result of a combination of more or less fortuitous circumstances.

271 Secondly, the members of the inspection team did not inform the applicant of the particular sensitivity of the security label, which contributed to the possible negligence constituted by the absence of preventive measures on the part of the applicant.

272 Thirdly, it was not possible to establish that documents were removed from room G.505.

273 Fourthly, contrary to what the Commission asserted in recital 112 to the contested decision, the applicant cooperated with it to an extent which exceeded that required of it, in particular by providing it with expensive expert opinions.

274 In the fourth place, the mere reference to the percentage of the fine set in relation to the applicant’s total turnover is not sufficient to exclude a breach of the law in the setting of the fine. The amount of the fine is disproportionate to the infringement, in the light of the doubts as to the actual existence, in this case, of a breach of seal attributable to the applicant. Nor is there any requirement for the fine to have a deterrent effect. Furthermore, in a similar application of the principle of proportionality, the Nederlandse Mededingingsautoriteit (the Netherlands competition authority; ‘the NMa’) recently imposed a fine for breach of seal of EUR 269 000 or 0.0028% of the total turnover of the undertaking in question, even though the Wet houdende nieuwe regels omtrent de economische mededinging (Mededingingswet) (Netherlands Law on Competition) (Staatsblad 1997, No 242), as amended, allowed a range of fines to be set, with an upper limit 1% of the global turnover (Article 70b(1) of the Mededingingswet).

275 The Commission contends that the applicant’s arguments should be rejected.

Findings of the Court

276 In the first place, the applicant argues that the contested decision is insufficiently reasoned, since the Commission failed to specify in it the criteria on which it based its setting of the amount of the fine imposed. That inadequacy of the statement of reasons infringed the applicant’s rights of defence.

277 According to settled case-law, the statement of reasons required by Article 253 EC must be appropriate to the act at issue and must disclose in a clear and unequivocal fashion the reasoning followed by the institution which adopted the measure in question in such a way as to enable the persons concerned to ascertain the reasons for the measure and to enable the competent Court of the European Union to exercise its power of review. The requirements to be satisfied by the statement of reasons depend on the circumstances of each case, in particular the content of the measure in question, the nature of the reasons given and the interest which the addressees of the measure, or other parties to whom it is of direct and individual concern, may have in obtaining explanations. It is not necessary for the reasoning to go into all the relevant facts and points of law, since the question whether the statement of reasons meets the requirements of Article 253 EC must be assessed with regard not only to its wording but also to its context and to all the legal rules governing the matter in question (see Case C-367/95 P Commission v Sytraval and Brink’s France [1998] ECR I-1719, paragraph 63, and the case-law cited).

278 In this case, as regards the criteria on the basis of which the Commission acted in setting the amount of the fine imposed on the applicant, the Commission stated that the level of the fine was determined in particular by the seriousness of the infringement and the particular circumstances of the case (recitals 104 and 113 to the contested decision).

279 The Commission thus argued, firstly, that, quite apart from this specific case, a breach of seal was a serious infringement and a fine imposed for a breach of seal had to have a deterrent effect (recital 105 to the contested decision), so that it should not pay for an undertaking involved in an inspection to break a seal.

280 Secondly, it pointed out that inspections were generally ordered only where there was evidence pointing to infringements of the competition rules and that that was the case here. In that regard, the inspection carried out at the applicant’s premises was intended to enable the Commission to investigate evidence pointing to infringements of the competition rules; in addition, uncatalogued documents, in particular those found on the first day of the inspection, were stored in the sealed room (recitals 107 and 108 to the contested decision).

281 Thirdly, it stated that it had taken into account, in calculating the amount of the fine imposed on the applicant, the fact that this was the first case to which Article 23(1)(e) of Regulation No 1/2003 had applied, while at the same time taking the view that that circumstance could not result in the fine set being incapable of ensuring the deterrent effect of that provision (recital 109 to the contested decision).

282 Fourthly, it pointed out that, apart from the fact that Regulation No 1/2003 had reinforced the provisions on fines for procedural infringements three years prior to the inspection and that sealings had already taken place in the buildings of the same group a few weeks previously, the applicant was one of the largest European undertakings in the energy sector, which had at its disposal extensive expertise with regard to antitrust law, and its attention had been drawn, at the time of the sealing, to the significant level of fines provided for in the case of a breach of seal (recital 110 to the contested decision).

283 The Commission then rejected the applicant’s arguments, put forward by way of attenuating circumstances, to the effect that the Commission did not provide any proof of the fact that the door of room G.505 had been opened or that documents had been removed, or to the effect that the applicant cooperated with the Commission beyond the extent required (recitals 111 and 112 to the contested decision).

284 Since, as regards Article 23(1)(e) of Regulation No 1/2003, the Commission has not adopted any guidelines setting out the method of calculation which it is required to follow when setting fines under that provision and since the Commission’s reasoning is disclosed in a clear and unequivocal fashion in the contested decision, the Commission was not, contrary to what the applicant maintains, required to express in figures, in absolute terms or as a percentage, the basic amount of the fine and any aggravating or attenuating circumstances. It follows that the complaint alleging infringement of Article 253 EC must be rejected. Since the applicant’s complaint alleging infringement of its rights of defence arises from the alleged inadequacy of the statement of reasons for the contested decision, it must also be rejected.

285 In the second place, the applicant claims that the fine imposed on it is disproportionate.

286 In that regard, it must be recalled that the principle of proportionality requires that measures adopted by the institutions do not exceed the limits of what is appropriate and necessary in order to attain the objectives legitimately pursued by the legislation in question; when there is a choice between several appropriate measures recourse must be had to the least onerous, and the disadvantages caused must not be disproportionate to the aims pursued (Case C-180/96 United Kingdom v Commission [1998] ECR I-2265, paragraph 96, and judgment of 12 September 2007 in Case T-30/05 Prym and Prym Consumer v Commission, not published in the ECR, paragraph 223).

287 It follows that fines must not be disproportionate to the aims pursued, that is to say, to the observance of the competition rules, and that the amount of the fine imposed on an undertaking for a competition infringement must be proportionate to the infringement, assessed as a whole, taking account, inter alia, of its gravity (Prym and Prym Consumer v Commission, paragraph 286 above, paragraph 224). In that regard, in accordance with a consistent line of cases, the gravity of an infringement is assessed in the light of numerous factors, in respect of which the Commission has a margin of discretion (Case C-328/05 P SGL Carbon v Commission [2007] ECR I-3921, paragraph 43; see also, to that effect, Joined Cases C-189/02 P, C-202/02 P, C-205/02 P to C-208/02 P and C-213/02 P Dansk Rørindustri and Others v Commission [2005] ECR I-5425, paragraphs 240 to 242).

288 Firstly, in recitals 105 to 108 to the contested decision, the Commission rightly set out the reasons why the infringement of breaking a seal was, as such, a particularly serious infringement, referring primarily to the purpose of seals, which is to prevent evidence from being lost during the inspection, and the need to ensure that the fine imposed has a sufficiently deterrent effect. In that regard, it is also important to point out that, on the one hand, as regards the infringement of breaking a seal, the legislature has, in Regulation No 1/2003, laid down more severe penalties than those which were provided for in the previous set of rules, in order to take into account the particularly serious nature of that infringement. On the other hand, it is clear from the case-law that, in setting the amount of fines, the Commission is justified in taking into consideration the need to ensure that they have a sufficiently deterrent effect (see, to that effect, Musique Diffusion française and Others v Commission, paragraph 249 above, paragraph 108), to which even greater importance attaches in the context of an infringement of breaking a seal, in order that undertakings should not be able to consider that it would pay them to break a seal in the course of an inspection (recital 105 to the contested decision). In the light of the foregoing, and contrary to what the applicant maintains, the Commission did not find that there were aggravating circumstances in its case, but mentioned the circumstances which justified the imposition of a sufficiently deterrent fine for any infringement constituting a breach of seal.

289 Secondly, as regards the attenuating circumstances of which the Commission allegedly did not take sufficient account, first, the applicant’s argument that the breaking of a seal through negligence constitutes an attenuating circumstance for the undertaking concerned must be rejected. In that regard, it must be observed, first of all, that, contrary to what the applicant maintains, the Commission did not start from the assumption of an intentional infringement, since it took the view that it had to be conceded in this case that, ‘at the very least’, this was a breach of seal through negligence (recital 102 to the contested decision). In addition, under Article 23(1)(e) of Regulation No 1/2003, the infringement constituted by the breach of seal may be committed intentionally or negligently. As the Commission points out, the mere fact that the seal is broken nullifies its safeguarding effect and is therefore sufficient to constitute the infringement.

290 Second, as regards the applicant’s argument that the Commission should have informed it of the particular sensitivity of the security label, it must be held that, as is apparent from the analysis of the fifth and sixth pleas, that alleged sensitivity has not been demonstrated in the case of the Commission’s official seals and, in any event, the applicant has not established that it was capable of leading to the occurrence of ‘false positive reactions’ of the seal. In addition, as is apparent from the examination of the eighth plea, it was for the applicant to take the necessary measures in order to prevent any tampering with the seal at issue.

291 Third, the fact that it was not possible to establish that documents had been removed from room G.505 is irrelevant, since the objective of sealing is specifically to prevent any handling of the documents placed in the sealed room in the absence of the Commission’s inspection teams. In this case, as the Commission observed in recitals 11 and 111 to the contested decision, the documents stored in room G.505 had not been catalogued, mainly because of their quantity. The inspection team was therefore unable to ascertain whether documents stored in that room were missing.

292 Fourth, the alleged costly efforts which the applicant agreed to make for the purpose of producing the institute’s expert opinions or questioning of staff and keyholders cannot be regarded as a clarification of the facts, going beyond what was required of it and justifying a reduction of the amount of the fine, since those efforts were made in the context of the exercise of the applicant’s rights of defence and did not facilitate the Commission’s investigation.

293 Thirdly, it must be noted that, in any event, the Commission took into consideration the fact that the breach of seal in question was the first case to which Article 23(1)(e) of Regulation No 1/2003 had applied (recital 109 of the contested decision), while pointing out that, quite apart from that circumstance, first, the applicant had at its disposal extensive legal expertise in antitrust law, second, the amendment of Regulation No 1/2003 dated from more than three years prior to the inspections in which the applicant had been involved, third, the applicant had been informed of the consequences of a breach of seal and, fourth, other seals had already been affixed in the buildings of other companies of the applicant’s group a few weeks previously.

294 Fourthly, contrary to what the applicant maintains, a fine of EUR 38 million cannot be regarded as disproportionate to the infringement, in the light of the particularly serious nature of a breach of seal, the size of the applicant and the need to ensure that the fine has a sufficiently deterrent effect, in order that it cannot pay for an undertaking to break a seal affixed by the Commission in the context of inspections.

295 In that regard, the argument concerning the decision-making practice of the NMa in the Netherlands cannot be accepted. Apart from the fact that the Commission cannot, in any event, be bound by the decision-making practices of national competition authorities, it must be stated that the comparison made by the applicant between, on the one hand, the percentage of the fine imposed by the NMa in relation to the total turnover of the undertaking concerned and, on the other, the percentage of the fine imposed on the applicant in this case in relation to its turnover is not relevant, since that comparison was made, as regards the Netherlands infringement, in relation to the turnover of the group of companies concerned and, as regards the present case, in relation to the turnover of the applicant alone, and not in relation to that of the E.ON group as a whole.

296 It is clear from all the foregoing considerations that the fine imposed on the applicant by the Commission, which corresponds to approximately 0.14% of its turnover, is not disproportionate.

297 It follows that the plea must be rejected, as must the action in its entirety, without there being any need to grant the applications for measures of inquiry made by the applicant (see, to that effect, Case C-260/05 P Sniace v Commission [2007] ECR I-10005, paragraphs 77 to 79, and the case-law cited).

Costs

298 Under Article 87(2) of the Rules of Procedure, the unsuccessful party is to be ordered to pay the costs if they have been applied for in the successful party’s pleadings.

299 Since the applicant has been unsuccessful, it must be ordered to pay the costs, as applied for by the Commission.

On those grounds,

THE GENERAL COURT (Eighth Chamber)

hereby:

1. Dismisses the action;

2. Orders E.ON Energie AG to pay the costs.

[Signatures]

Table of contents

Legal context

Background to the dispute

Procedure and forms of order sought

Law

The first plea, alleging failure to have regard to the burden of proof

Arguments of the parties

Findings of the Court

The second plea, alleging infringement of the ‘principle of inquisitorial procedure’

Arguments of the parties

Findings of the Court

The third plea, alleging an erroneous assumption that the seal was affixed properly

Arguments of the parties

Findings of the Court

The fourth plea, alleging an erroneous assumption of the ‘obvious condition’ of the seal at issue on the day following the inspection

Arguments of the parties

Findings of the Court

The fifth plea, alleging an erroneous assumption of the suitability of the security label for official sealing by the Commission

Arguments of the parties

Findings of the Court

The sixth plea, alleging disregard by the Commission of ‘alternative scenarios’ which may have been the cause of the condition of the seal at issue

Arguments of the parties

Findings of the Court

The seventh plea, alleging breach of the principle of the presumption of innocence

Arguments of the parties

Findings of the Court

The eighth plea, alleging infringement of Article 23(1) of Regulation No 1/2003

Arguments of the parties

Findings of the Court

The ninth plea, alleging infringement of Article 253 EC and of the principle of proportionality when the amount of the fine was set

Arguments of the parties

Findings of the Court

Costs

* Language of the case: German.