OPINION OF ADVOCATE GENERAL

delivered on 23 September 2004 (1)

CEVA Santé Animale SA and Pfizer Enterprises SARL

1.In the present case the Commission has appealed against a judgment of the Court of First Instance ruling that the Community is liable in damages pursuant to Article 288 EC for, essentially, the Commission’s delay in classifying progesterone (a naturally occurring steroid hormone) in the list of substances for which a maximum residue limit need not be set, to be drawn up pursuant to legislation on veterinary medicinal products. (2)

Relevant legislation in force at the commencement of proceedings

The scientific committees and other relevant bodies

2.The legislation – and the parties’ pleadings – refer to the following scientific committees and other bodies.

3.The Committee for Veterinary Medicinal Products (‘CVMP’) was set up under Article 16 of Council Directive 81/851/EEC (3) in order to facilitate the adoption of a common position by the Member States with regard to marketing authorisations. It is composed of representatives of the Member States and of the Commission. Since 1995 the CVMP has been part of the European Agency for the Evaluation of Medicinal Products.

4.The European Agency for the Evaluation of Medicinal Products (‘EMEA’) was established by Article 49 of Council Regulation (EEC) No 2309/93 (4) with responsibility for coordinating the existing scientific resources put at its disposal by the competent authorities of the Member States for the evaluation and supervision of medicinal products. It was subsequently given the role of dealing with applications to establish maximum residue limits. (5)

5.The Scientific Committee on Veterinary Measures relating to Public Health (‘SCVPH’) was set up by Commission Decision 97/579/EC, (6) which provides that its members are to be scientific experts in one or more of its fields of competence and are collectively to cover the widest range of disciplines. At the relevant time the SCVPH was part of the Commission (within the Health and Consumer Protection Directorate-General). (7)

6.The Codex Alimentarius, (8) or food code, was established in 1961 by the Food and Agriculture Organisation of the United Nations (‘FAO’) and the World Health Organisation (‘WHO’). Two years later those bodies approved the establishment of the Joint FAO/WHO Food Standards Programme and adopted the statutes of the Codex Alimentarius Commission to develop food standards, guidelines and related texts, which constitute the Codex. The Joint FAO/WHO Expert Committee on Food Additives (‘JECFA’) is an international expert scientific committee that is administered jointly by the FAO and the WHO. It has been meeting since 1956, initially to evaluate the safety of food additives. Its work now also includes the evaluation of contaminants, naturally occurring toxicants and residues of veterinary drugs in food. While not officially part of the Codex Alimentarius Commission structure, the JECFA provides independent scientific expert advice to the Commission and its specialist Committees.

Regulation No 2377/90

7.Council Regulation (EEC) No 2377/90 (9) (or ‘the Regulation’) lays down a Community procedure for the establishment of maximum residue limits of veterinary medicinal products in foodstuffs of animal origin.

8.The Regulation includes the following recitals in its preamble:

‘[1] the use of veterinary medicinal products in food-producing animals may result in the presence of residues [in] foodstuffs obtained from treated animals;

…

[3] in order to protect public health, maximum residue limits must be established in accordance with generally recognised principles of safety assessment, taking into account any other scientific assessment of the safety of the substances concerned which may have been undertaken by international organisations, in particular the Codex Alimentarius or, where such substances are used for other purposes, by other scientific committees established within the Community;

…

[6] it is therefore necessary to lay down a procedure for the establishment of maximum residue levels of veterinary medicinal products by the Community, following a single scientific assessment of the highest possible quality;

…

[9] arrangements must also be made for the establishment of maximum residue levels for substances which are currently used in veterinary medicines administered to food-producing animals; … however, in view of the complexity of this matter and the large number of substances involved, long transitional arrangements are required;

[10] after scientific assessment by the [CVMP], maximum residue levels must be adopted by a rapid procedure which ensures close cooperation between the Commission and the Member States ...’.

9.Under the Regulation the Commission is to establish the maximum residue limit in accordance with the procedure there laid down. Article 1(1)(b) of the Regulation defines ‘maximum residue limit’ (‘MRL’) as the maximum concentration of pharmacologically active substances resulting from the use of a veterinary medicinal product and remaining in foodstuffs obtained from animals which may be accepted by the Community to be legally permitted or recognised as acceptable in or on a food.

10.The Regulation provides for the inclusion in one of four annexes, to be adopted in accordance with a prescribed procedure, of pharmacologically active substances used in veterinary medicines for food-producing animals. Such substances are to be listed in Annex I where MRLs have been established, (10) in Annex II where ‘following an evaluation … it appears that it is not necessary for the protection of public health to establish a[n MRL]’ (11) and in Annex IV where it appears that an MRL cannot be established because residues at whatever limit constitute a hazard to the health of the consumer. (12) A provisional MRL may be established for a substance in use when the Regulation entered into force ‘provided that there are no grounds for supposing that residues … at the level proposed present a hazard for the health of the consumer’. Substances in respect of which a provisional MRL has been established are to be listed in Annex III. (13)

11.Articles 6 and 7 of the Regulation originally laid down the procedure for obtaining the inclusion in Annex I, II or III of, respectively, a ‘new’ pharmacologically active substance and a pharmacologically active substance which was authorised for veterinary use on the date of entry into force of the Regulation. Those articles were replaced in 1999, (14) primarily in order to involve the EMEA, established since the Regulation was adopted. (15) The amended articles establish a single procedure for both types of application.

12.Article 7(4) originally provided, in so far as is relevant:

‘Having regard to the observations formulated by the members of the [CVMP], the Commission shall prepare, within a maximum period of 30 days, a draft of the measures to be taken. ...’

13.Article 7 as amended provides, in so far as is relevant:

‘5. The [EMEA] shall forward the definitive opinion of the [CVMP] within 30 days of its adoption both to the Commission and to the applicant. The opinion shall be accompanied by a report describing the safety evaluation of the substance by the [CVMP], which shall give the grounds for its conclusions.

6. The Commission shall prepare draft measures taking account of Community legislation and shall start the procedure provided for in Article 8. …’

14.Article 8(2) of the Regulation requires the Committee for Adaptation to Technical Progress of the Directives on Veterinary Medicinal Products (‘the Standing Committee’) to deliver its opinion on the draft measures. Article 8(3) provides that the Commission is to adopt the proposed measures where they are in accordance with the Standing Committee’s opinion; where they are not in accordance with that opinion, the Commission is to propose to the Council the measures to be adopted. The Council is to act by qualified majority; if it has not acted within three months of the proposal being referred to it, the proposed measures are to be adopted by the Commission unless the Council has voted against them by a simple majority.

15.Article 14 originally provided:

‘With effect from 1 January 1997, the administration to food-producing animals of veterinary medicinal products containing pharmacologically active substances which are not mentioned in Annexes I, II or III shall be prohibited within the Community ...’

16.By the end of 1996 there were still, according to the Commission, some 188 substances awaiting classification. The relevant committee of the European Parliament explained that the delay was due to ‘difficulties for industry in collecting all the data necessary …, shortages in financial/human resources at Member State level … and the fact that the fledgling [EMEA] … only became operational in 1995’. (16) The original date mentioned in Article 14 was accordingly changed to 1 January 2000 for most substances (including progesterone) the use of which was authorised on the date of entry into force of the Regulation and in respect of which applications for an MRL had been lodged before 1 January 1996. (17)

17.The preamble to Regulation No 434/97, (18) which so amended Article 14, states:

‘… Regulation (EEC) No 2377/90 … provides for the gradual evaluation of substances the use of which was authorised on the date of entry into force of that Regulation ...

… in order to allow this Community procedure to continue on a sound scientific basis and not to deprive veterinary surgeons and users of substances needed to protect animal health, this time limit [of 1 January 1997] should be extended ...’

Directive 81/851

18.Directive 81/851 (19) concerns the issue by Member States of national marketing authorisations for veterinary medicinal products.

19.Article 4(2) provides, in so far as is relevant:

‘A Member State shall not authorise the placing on the market of a veterinary medicinal product intended for administration to food-producing animals whose flesh or products are intended for human consumption, unless:

(a) the active substance or substances capable of pharmacological action contained in the veterinary medicinal product were authorised for use in other veterinary medicinal products in the Member State concerned on the date of entry into force of Council Regulation (EEC) No 2377/90 ...;

(b) the active substance or substances capable of pharmacological action is or are mentioned in Annex I, II or III to the aforementioned Regulation.

…’

Regulation No 2309/93

20.Regulation No 2309/93 (20) governs the issue of Community authorisations for the marketing of veterinary medicines. Article 31(3)(b) provides that, in the case of a veterinary medicinal product intended for administration to food-producing animals, a statement of the MRL which may be accepted by the Community in accordance with Regulation No 2377/90 is to be produced in the course of the procedure to obtain a marketing authorisation.

Directive 96/22

21.Council Directive 96/22/EC (21) requires Member States to prohibit inter alia the administering to farm animals of hormonal substances having gestagenic action. (22) Such substances include progesterone. By way of derogation, Member States may authorise the administering to farm animals of progesterone for therapeutic purposes in limited circumstances. (23) Those circumstances do not include administration for growth promotion purposes, which remains prohibited.

Case-law on the Regulation

22.It is helpful to refer at the outset to the case-law on Regulation No 2377/90.

Lilly

23.In Lilly (24) the Court of First Instance stated, first, that where the CVMP, having all the necessary information at its disposal, has given a favourable opinion on a request for the inclusion of a substance in Annex II, the Commission is under an obligation to draw up a draft regulation including that substance in Annex II and, second, that the procedure for the establishment of an MRL under the Regulation is independent of and distinct from the procedures for the issue of marketing authorisations laid down in Directive 81/851 (25) and Regulation No 2309/93. (26)

24.However, the latter statement was explicitly rejected by the Court of Justice in Monsanto (27) and the former, although not explicitly overridden, can no longer be regarded as good law for the reasons given in paragraph 75 below.

Pharos

25.In Pharos (28) the CVMP had recommended inclusion of the substance concerned in Annex II but the Standing Committee had not agreed, being concerned that the substance could be used to boost growth. The Commission had accordingly asked the CVMP for a further opinion as to whether abuses of the product were possible.

26.In its appeal against the judgment of the Court of First Instance dismissing its claim for damages, the manufacturer of the product submitted inter alia that the Regulation did not confer on the Commission the right to seek a further opinion from the CVMP. The Court of Justice stated that the Court of First Instance had been right to hold that where it is confronted with a matter which is highly complex and sensitive the Commission must be accorded the right to seek a further opinion from the CVMP even though the Regulation is silent on the point. (29) The Court also stated that the amount of time the Commission had to consider the various courses of action open to it must be appraised in the light of the complexity of the matter concerned. In the circumstances, a period of 11 months, during which the Commission initially reconsidered the file for six months and then sought a second scientific opinion, could not be considered an excessively long period. (30)

Monsanto

27.In Monsanto, (31) the Court of Justice ruled that the procedures for establishing MRLs and issuing marketing authorisations were inherently linked, inasmuch as a marketing authorisation will not be issued in respect of a veterinary medicinal product for administration to food-producing animals unless an MRL has been established. The Court of First Instance in Lilly therefore misinterpreted those provisions when it concluded that the procedure for establishing an MRL under Regulation No 2377/90 was independent of and distinct from the procedures for the issue of marketing authorisations laid down by Directive 81/851 (32) and Regulation No 2309/93, (33) and when it inferred, in particular from the fact that Regulation No 2377/90 contained no provision authorising the Commission to take account of a marketing ban in refusing to establish an MRL, that the Commission was not legally entitled to take account of it. The Court of Justice considered it to be self-evident that, when applying a regulation, the Commission may take other provisions of Community law into account. (34)

The facts

28.CEVA Santé animale SA (‘CEVA’) and Pfizer Enterprises SARL (formerly Pharmacia Enterprises SA) (‘Pfizer’) (jointly, ‘the respondents’) are pharmaceutical companies each of which has, since before the Regulation entered into force, marketed a veterinary medicinal product containing the active ingredient progesterone. In veterinary medicine progesterone is used in cows and mares to treat disorders of the reproductive system (therapeutic purposes) and to synchronise oestrus and prepare donor and receptor animals in the case of embryo transfer (zootechnical purposes).

29.In 1993 CEVA submitted an application to the Commission for the establishment of an MRL for progesterone in cattle and horses.

30.In November 1996 the EMEA informed CEVA that, at its meeting in October 1996, the CVMP had recommended the inclusion of progesterone in Annex II to the Regulation and that the opinion of the CVMP would be forwarded to the Commission for adoption by the Standing Committee.

31.In April 1997 the Commission sent new scientific information to the EMEA and requested a re-assessment of the risks relating to the hormones oestradiol-17ß and progesterone.

32.In October 1997 the EMEA wrote to CEVA saying ‘the Commission has decided to stop the adoption procedure for progesterone as new scientific data have recently become apparent concerning oestradiol, which are considered relevant also for progesterone. The CVMP has therefore been requested to undertake a reconsideration of the assessment in light of these additional data. You will be kept informed on further developments concerning the establishment of MRLs for progesterone.’

33.In April 1998 the Commission again wrote to EMEA requesting that the CVMP should be given the opportunity to consider scientific information expected to be available in the course of 1998 from a number of sources, such as the International Agency for Research on Cancer, an advisory body to the WHO, and the United States National Institute of Health, and the results of a number of specific studies commissioned by the European Commission.

34.In May 1998 the Commission learned that the JECFA was also planning to re-evaluate the three natural hormones oestradiol-17ß, progesterone and testosterone in February 1999.

35.In February 1999 the Commission published in the Official Journal a ‘call for scientific documentation required for risk assessment of the hormonal substances oestradiol-17ß, progesterone, testosterone, zeranol, trenbelone acetate and melengesterol acetate used for animal growth promotion purposes’.

36.In April 1999 or thereabouts the JECFA’s summary assessment of the three natural hormones became available. (35) On the basis of the available data, JECFA concluded that there would be no need to specify numerical MRLs for the three hormones.

37.In April 1999 the Commission asked the EMEA to initiate ‘the update of the evaluation’, which the Commission had requested in 1997 but which it had in 1998 sought to delay pending the results of further studies, of the hormones oestradiol-17ß and progesterone ‘at your earliest convenience, in order to allow the adoption and publication of the results of this evaluation before 1 January 2000’.

38.In May 1999 the Commission forwarded to the EMEA the opinion of the SCVPH dated 30 April 1999. (36) The report’s summary conclusion was as follows:

‘Taking into account both the hormonal and non-hormonal toxicological effects as mentioned above … it has to be concluded that the issues of concern include neurobiological, developmental, reproductive and immunological effects, as well as immunotoxicity, genotoxicity and carcinogenicity. In consideration of the recent concerns relating to the lack of understanding of critical developmental periods in human life as well as the uncertainties in the estimates of endogenous [naturally occurring] hormone production rates and metabolic clearance capacity, particularly in prepubertal children, no threshold level and therefore no ADI [acceptable daily intake] can be established for any of the six hormones.’

39.By letter of 20 December 1999, the EMEA informed CEVA that, at its meeting earlier that month, the CVMP had confirmed its earlier opinion on the inclusion of progesterone in Annex II to the Regulation. The opinion of the CVMP (37) and its summary report (38) were appended to that letter.

40.The CVMP stated in its opinion:

‘The Committee, having evaluated the applications, recommended in October 1996 to include progesterone in Annex II of Council Regulation (EEC) No 2377/90. That opinion was however not adopted by the European Commission. In 1997 and 1999 the European Commission brought new data on steroidal sex hormones to the attention of the Committee and requested a re-evaluation of the substance in the light of the new data. The Committee, having considered the applications and the new data as stated in the appended summary report, confirmed the previous opinion and recommended that the above-mentioned substance shall be inserted in Annex II of Council Regulation (EEC) No 2377/90 ...’

41.On 3 May 2000 the SCVPH adopted a re-evaluation of its opinion of April 1999. It had been asked either to confirm that there was no recent scientific information that would lead it to revise its previous opinion or to revise the relevant parts thereof as necessary. It concluded that recent scientific information did not provide convincing data or arguments requiring revision of its previous conclusions, and mentioned that it had discussed again the obvious gaps in the present knowledge on target animal metabolism and residue disposition of the hormones under consideration and that it expected that the ongoing EU research programmes would provide additional data on both topics.

42.On 24 May 2000 the Commission adopted a proposal for a directive amending Directive 96/22. (39) The proposal required Member States inter alia provisionally to prohibit the administration of progesterone to farm animals while maintaining the derogation for therapeutic and zootechnical purposes. The preamble stated that the provisional prohibition ‘should apply while the Community seeks more complete scientific information from any source, which could shed light and clarify the gaps in the present state of knowledge on these substances’. (40)

43.In July 2000 the respondents put the Commission on formal notice to take the necessary measures for including progesterone in Annex II to the Regulation as soon as possible.

44.In November 2000 the respondents brought proceedings before the Court of First Instance seeking principally (i) a declaration pursuant to Article 232 EC that the Commission had failed to comply with its obligations under Community law by failing to take the necessary measures for the inclusion of progesterone in Annex II to the Regulation following the issuing of the positive opinion of the CVMP and (ii) an order pursuant to Article 288 EC that the Community, as represented by the Commission, repair the damage suffered by the respondents as a result of the Commission’s unlawful failure to act. The respondents explained that that damage was due to the fact that from 1 January 2000 they had been unable to market their products for administration to food-producing animals and several competent national authorities had withdrawn or not extended marketing authorisations for their products. Fédération européenne de la santé animale (Fedesa) intervened in support of the respondents.

Developments since the commencement of proceedings

45.On 25 July 2001 the Commission adopted a proposal for a regulation (‘the draft regulation’) (41) classifying progesterone and norgestomet (a similar hormone) in Annex I to the Regulation (which lists pharmacologically active substances for which MRLs have been fixed). The preamble to the draft regulation stated:

‘Following its evaluation, the CVMP considered that it was not necessary, for the protection of public health, to establish maximum residue limits for Progesterone and Norgestomet when used in authorised veterinary medicinal products in accordance with Community legislation in force, notably Directive 96/22/EC. The substances were therefore proposed to be included in the list in Annex II of Regulation (EEC) No 2377/90. However, the overall evaluation of the available risk assessments of these substances and of the entire body of available scientific information and data indicate that, as concerns the excess intake of hormone residues and their metabolites, and in view of the intrinsic properties of hormones and epidemiological findings, a risk to the consumer has been identified.’ (42)

46.On 1 August 2001 the draft regulation was sent to the Standing Committee in accordance with the procedure laid down by Article 8 of the Regulation. (43) That Committee did not give a favourable report and, on 26 October 2001, again in accordance with Article 8, the Commission submitted the draft regulation to the Council. It was, however, rejected at the meeting of the Council of Ministers for Agriculture of 21 and 22 January 2002. According to the Commission, that rejection was essentially due to (i) the difficulties in calculating the appropriate values of the MRLs for all kinds of food-producing animals because of endogenous production and (ii) lack of validated detection methods. Since there was no qualified majority in favour of the draft but a simple majority against its adoption by the Commission, the proposal went no further. (44)

47.In December 2002 the Commission submitted a second proposal to the Standing Committee to place progesterone in Annex III (provisional MRL). That proposal did not receive a favourable reading by that committee in February 2003, again for a variety of reasons.

48.In September 2003 Directive 2003/74/EC of the European Parliament and of the Council (45)

amending Directive 96/22 was adopted. Directive 96/22 as amended provisionally prohibits inter alia the administration of progesterone to farm animals, subject to the continued derogation for administration for therapeutic and zootechnical purposes.

49. On 24 October 2003 the Commission adopted Regulation (EC) No 1873/2003. (46) That regulation amended Annex II to Regulation No 2377/90 so as to include progesterone for female animals of bovine, ovine, caprine and equine species for intravaginal therapeutic or zootechnical use only. The preamble to Regulation No 1873/2003 states:

‘The Commission considers that safeguards as to the possibility of misuse of veterinary medicinal products are necessary. Restricting the terms of the use of progesterone to administration only via the intravaginal route in female animals of bovine, ovine, caprine and equine species provides this additional safeguard needed to avoid misuse as the relevant veterinary medicinal products cannot, due to their specific presentation, be realistically used for a prohibited purpose. It is therefore considered appropriate to include progesterone in Annex II to Regulation (EEC) No 2377/90 in accordance with the Annex to the present … Commission regulation, which limits the use of progesterone to this specific purpose and product formulation.’ (47)

The judgment of the Court of First Instance and the appeal

50. The Court of First Instance gave judgment on 26 February 2003. It ruled first that, by adopting the draft regulation on 25 July 2001 and submitting it first to the Standing Committee then to the Council, the Commission had defined its position on the matter with regard to which the respondents called upon it to act; there was accordingly no need to give judgment on the actions for failure to act.

51. With regard to the claims for damages, the Court of First Instance ruled as follows:

‘First of all, it should be observed that, in view of the 6th and 10th recitals in its preamble and Articles 7 and 8, the 1990 Regulation, before as well as after amendment by Regulation No 1308/99, lays down a reasonably speedy procedure for establishing MRLs in which the opinion of the CVMP occupies a central place. In Pharos v Commission [Case C-151/98 P [1999] ECR I-8157] the Court nevertheless acknowledged, in the particular circumstances of that case, that, where it is confronted with a matter which is scientifically and politically complex and sensitive, the Commission is entitled to seek a further opinion from the CVMP, even though the 1990 Regulation is silent on the point.

Secondly, it must be acknowledged that, quite clearly, the progesterone file is a scientifically and politically complex file. Amongst other things, progesterone is an endogenous substance and there are at present no reliable analytical methods by which to check abuse of the substance. The complexity of the file is further confirmed by what happened to the draft regulation which the Commission adopted and then submitted to the Standing Committee and the Council.

However, that complexity does not excuse the Commission’s inaction after 1 January 2000. Given that the CVMP entirely confirmed its first opinion, even after taking into consideration the new scientific data presented to it by the Commission, and the fact that the Commission itself has always maintained the view that progesterone should continue to be authorised for therapeutic and zootechnical treatment, the Commission disregarded the legitimate interests of the respondents, of which it was perfectly well aware, in a clear and serious way by failing to adopt the measures needed for its continued use, for therapeutic and zootechnical purposes, after 1 January 2000, the date from which, under Article 14 of the 1990 Regulation, the administration to food-producing animals of veterinary medicinal products containing pharmacologically active substances which are not mentioned in Annexes I, II or III to the 1990 Regulation was prohibited within the Community. It is important to note in this context that the application for an MRL to be established for progesterone was made as early as September 1993.

Even if the scientific and political complexities of the file were such as to prevent the Commission from adopting, shortly after the CVMP issued its second opinion, a draft regulation conforming to that opinion, the Commission ought to have concerned itself with the interests of the [respondents], for example by adopting draft measures establishing a provisional MRL on the basis of Article 4 of the 1990 Regulation or by arranging for a (second) deferral of the time-limit laid down in Article 14 thereof.

That being so, the inaction of the Commission between 1 January 2000 and 25 July 2001 constitutes a clear and serious breach of the principle of sound administration giving rise, in principle, to liability on the Community’s part. There is therefore no need in the present case to establish whether the Commission’s inaction was administrative or legislative in nature, or to determine the exact scope of its discretion in setting MRLs.

…

The Commission’s argument that there is no causal link between its inaction and the damage sustained, on the ground that it is for the competent national authorities to adopt marketing authorisation decisions, cannot be accepted. Indeed, if national authorities have withdrawn or suspended marketing authorisations or suspended procedures for issuing such authorisations because no MRL has been fixed for progesterone, they did so simply in order to comply with the prohibition under Article 14 of the 1990 Regulation and Article 4(2) of Council Directive 81/851/EEC …. That being so, the damage is attributable to the Commission’s inaction. …’ (48)

52. In view of the fact that it was not then possible to fix the amount of damages, the Court of First Instance deferred to a subsequent stage of the proceedings the question of assessing the damage attributable to the Commission’s inaction between 1 January 2000 and 25 July 2001. (49)

53. In May 2003 the Commission lodged an appeal against the judgment of the Court of First Instance in which it requested the Court to set aside that judgment as regards the actions for damages and rule on the substance of the applications for damages by dismissing them entirely as unfounded. The respondents, supported by the International Federation for Animal Health (IFAH), successor of Fedesa, lodged cross-appeals in which they requested the Court to rule that (i) the Court of First Instance had erred in law in finding that there was no longer a need to rule on the action for failure to act and (ii) it was in violation of its obligations that the Commission failed to take the necessary measures to ensure that progesterone be included in time in Annex II to Regulation No 2377/90.

54. On 29 October 2003 the respondents and IFAH wrote to the Court discontinuing their cross-appeals on the basis that by adopting Regulation No 1873/2003 the Commission had brought to an end the failure to act which was the subject-matter of the cross-appeals.

The grounds of appeal

55. The Commission puts forward five grounds of appeal. First, the Court of First Instance committed an error of law in interpreting Article 14 of the Regulation as imposing on the Commission an absolute obligation to take a final decision on all the applications concerning ‘old’ substances before 1 January 2000 or, failing that, to arrange for a deferral of that deadline. Second, the Court of First Instance committed an error of law in the interpretation and application of the principle of sound administration in stating that the Commission’s inaction between 1 January 2000 and 25 July 2001 constituted a clear and serious breach of that principle. Third, the Court of First Instance fundamentally misconstrued the factual and scientific record by ignoring the relevance of scientific evidence other than the CVMP opinion and instead basing its judgment on the premiss that the Commission was obliged to follow that opinion. Fourth, the Court of First Instance committed an error of law in the interpretation and application of Article 288 EC concerning the non-contractual liability of the Community: it failed to analyse whether the Commission’s alleged breach was administrative or legislative, the degree of discretion enjoyed or whether the Commission ‘manifestly and gravely disregarded’ the limits of that discretion and it incorrectly held that the damage sustained was attributable to the Commission’s inaction. Fifth, the Court of First Instance committed an error of law in not distinguishing the position of Pfizer when it considered the principle of sound administration in Pfizer’s action for damages.

56. The first and third grounds concern the interpretation and proper application of the Regulation. The second and fourth grounds concern the more general questions whether the Commission’s inaction was unlawful and whether, if so, that unlawfulness was both sufficiently serious to engage the Community’s liability in damages and a direct cause of the damage sustained. The fifth ground concerns the specific position of Pfizer as compared to CEVA.

57. I propose to deal with the grounds of appeal in that order.

The first ground of appeal: Article 14 of the Regulation

58. The Commission submits that the Court of First Instance committed a fundamental error of law in interpreting Article 14 of the Regulation as imposing on the Commission an absolute obligation to act by completing the scientific evaluation and by taking a final decision on all the applications concerning ‘old’ substances before 1 January 2000 or, failing that, to arrange for a deferral of that deadline. According to the Commission, that date was not intended as a deadline by which a final decision had to be taken in every case: in the present case, specific reasons – explained in the context of the second and third grounds of appeal – justified the continued handling of CEVA’s application for progesterone beyond 1 January 2000.

Admissibility

59. The respondents submit that the first ground of appeal is inadmissible since the Commission had not argued before the Court of First Instance that the date specified in Article 14 of the Regulation should not be viewed as a deadline.

60. In my view that submission must be rejected. Even if it were the case that the Commission had not raised such an argument before the Court of First Instance, it seems to me that if – as the Commission submits – the judgment of the Court of First Instance is based on a false interpretation of Article 14 of the Regulation, then it must be open to the Commission to challenge that interpretation. (50) If the judgment is not based on that interpretation, the ground of appeal is unfounded, not inadmissible.

61. In any event it appears from the pleadings before the Court of First Instance that the status of the date referred to in Article 14 of the Regulation was raised before that court by both the respondents and the Commission; (51) it was moreover mentioned in the replies to the written questions put by the Court of First Instance.

62. The first ground of appeal is accordingly in my view admissible.

Substance

63. The respondents submit in the alternative that the Commission’s first ground of appeal is unfounded. The date of 1 January 2000 was intended as the deadline by which the Commission had to act in order to ensure that veterinary medicinal products containing pharmacologically active substances could continue to be marketed. That is borne out by the statement in the preamble to Regulation No 434/97, (52) which substituted the date referred to in Article 14, to the effect that the time-limit should be extended ‘in order to allow this Community procedure to continue on a sound scientific basis and not to deprive veterinary surgeons and users of substances needed to protect animal health’. (53) The respondents also claim that in various correspondence the Commission treated the date of 1 January 2000 as a date by which it was obliged to adopt an MRL for substances in respect of which applications had been submitted in due time.

64. IFAH, which has submitted a very brief response, appears to endorse that argument.

65. The Commission essentially submits that that date is not an absolute deadline which automatically applies in all circumstances.

66. I consider that the Commission’s interpretation is correct. It seems implausible that the legislature intended to impose an absolute obligation on the Commission to come to a definitive decision by a given date if the complexity of the matter and/or evolving scientific opinion meant that to do so might prejudice human health, the protection of which is the principal objective of the Regulation and must of course take precedence over other considerations.

67. Moreover it may be noted that the date is not expressed in Article 14 as a deadline for classification by the Commission; on the contrary, that provision simply states that as from that date the administration to food-producing animals of veterinary medicinal products containing pharmacologically active substances which are not mentioned in Annex I, II or III is to be prohibited within the Community. Had the date been intended to operate as an absolute deadline imposed on the Commission, the legislature might have been expected to say so.

68. The Commission’s own statements in correspondence as to its interpretation of Article 14, whether or not that interpretation is correct, cannot in any event be relevant to the question whether the Court of First Instance correctly interpreted that provision.

69. The statement in the preamble to Regulation No 434/97 (54) that the time-limit should be extended ‘in order … not to deprive veterinary surgeons and users of substances needed to protect animal health’, (55) invoked by the respondents, cannot in my view have been intended to require a decision by a given date in all circumstances: human health must manifestly take precedence over animal health where there is a conflict.

70. The above interpretation does not of course mean that the Commission was at liberty indefinitely to postpone classifying in one of the annexes to the Regulation a veterinary medicinal product containing pharmacologically active substances for which application to set an MRL was duly made. The fact that a time-limit is mentioned, even though it is not absolute, means that any delay beyond that date requires justification. (56)

71. Indeed it is not entirely clear whether the Court of First Instance proceeded on the basis that Article 14 imposed an absolute deadline. In any event, the central issue in the present case relates to whether there were sufficient reasons to justify the Commission’s delay. It is to those issues which I now turn.

The third ground of appeal: the scientific evidence

72. I will deal below with the second and fourth grounds, which should be taken together. By its third ground, the Commission submits that the Court of First Instance fundamentally misconstrued the evidence in that it failed to take account of the entire factual record in completely ignoring the relevance of scientific evidence other than the CVMP opinion, in particular the risk assessment of progesterone carried out by the competent SCVPH committee. (57)

73. I accept the Commission’s submission that the judgment of the Court of First Instance, and in particular paragraph 101, does appear to be based on the premiss that, once the CVMP had issued its re-evaluation, the Commission was bound to adopt measures conforming to that opinion.

74. The respondents support that premiss, contending that, once the CVMP has made its scientific assessment and submitted its opinion to the Commission, the latter is under an obligation to prepare a draft following that opinion; they cite the judgments in Lilly (58) and Monsanto (59) and the Opinions of Advocate General Mischo in Pharos (60) and Advocate General Alber in Monsanto. The respondents accept that – as indicated in the latter Opinion – it was permissible for the Commission, when confronted with new scientific evidence, to submit it to the CVMP for review. However, after the CVMP had in December 1999 confirmed its earlier findings that the use of progesterone in veterinary medicine was safe, the respondents consider that there was nothing justifying the Commission’s further inaction.

75. I am not persuaded by that argument. Admittedly the Court of Justice in Monsanto did not specifically deal with the Commission’s submission that the Court of First Instance had misinterpreted Article 6(3) of the Regulation by taking the view that the Commission was legally bound to follow the opinion of the CVMP. (61) However, in ruling on the substance of the case – having set aside the judgment of the Court of First Instance – the Court of Justice does not appear to have endorsed the view that the Commission is bound by the CVMP’s opinion. It rejected as unfounded Monsanto’s plea that the Commission had committed a breach of the Regulation by omitting, following the delivery by the CVMP of its opinion, to submit draft measures. (62) That interpretation moreover appears to be consistent with the wording of the Regulation, which originally required the Commission’s draft measures to ‘[have] regard’ to the CVMP opinion and now simply requires those measures to ‘[take] account of Community legislation’. (63)

76. I accordingly do not consider that the case-law requires the Commission not to depart from the opinion of the CVMP. In my view the Commission was justified in taking into account scientific evidence other than the CVMP opinion. That conclusion follows inescapably from the preamble to the Regulation, which states that MRLs are to be established ‘taking into account any other scientific assessment of the safety of the substances concerned which may have been undertaken by international organisations … or, where such substances are used for other purposes, by other scientific committees established within the Community’. (64) Thus the Commission was clearly entitled or indeed required to take account of the scientific evidence which was available from other relevant sources such as the SCVPH, the JECFA and the International Agency for Research on Cancer.

The second and fourth grounds of appeal: interpretation of Article 288 EC

78. The second and fourth grounds of appeal essentially concern the Court of First Instance’s interpretation and application of the second paragraph of Article 288 EC. That court stated in paragraph 103 of the judgment that the inaction of the Commission between 1 January 2000 and 25 July 2001 (when the Commission adopted the draft regulation) constituted a clear and serious breach of the principle of sound administration giving rise, in principle, to liability on the Community’s part and that there was therefore no need to establish whether the inaction was administrative or legislative in nature or to determine the exact scope of the Commission’s discretion in setting MRLs.

79. The second paragraph of Article 288 EC states:

‘In the case of non-contractual liability, the Community shall, in accordance with the general principles common to the laws of the Member States, make good any damage caused by its institutions or by its servants in the performance of their duties.’

80. The Court of Justice in Brasserie du Pêcheur and Factortame (65) reconsidered the requirements to be met for a finding of non-contractual liability under Article 288 EC. It stated first that the system of rules which the Court has worked out in relation to such liability takes into account, inter alia, the complexity of the situations to be regulated, difficulties in the application or interpretation of the texts, and, more particularly, the margin of discretion available to the author of the act in question. It then set out the three conditions which must be met before a right to reparation arises: the rule of law infringed must be intended to confer rights on individuals, the breach must be sufficiently serious and there must be a direct causal link between the breach and the damage sustained by the injured parties.

81. In relation to the second condition, the Court added that, in a legislative context characterised by the exercise of a wide discretion, the Community could not incur liability unless the institution concerned had manifestly and gravely disregarded the limits on the exercise of its powers, explaining that, even where the legality of measures is subject to judicial review, exercise of the legislative function must not be hindered by the prospect of actions for damages whenever the general interest of the Community requires legislative measures to be adopted which may adversely affect individual interests. (66)

82. Although the Court in its earlier case-law on Article 288 EC had distinguished between liability for administrative and legislative acts, that distinction can no longer be regarded as relevant since the judgment of the Court in Bergaderm. (67)

83. In that case the Court ruled that the categorisation of the measure was irrelevant: the decisive test for finding that a breach of Community law is sufficiently serious to give rise to a right to damages is whether the institution ‘manifestly and gravely disregarded the limits on its discretion’. (68) Bergaderm has been confirmed on that point by Camar and Tico (69) and Fresh Marine. (70)

84. The Commission’s second ground of appeal concerns the question whether there was an infringement of Community law; the fourth ground concerns the question whether, if so, the infringement was sufficiently serious to give rise to liability in damages and, if so, whether there was a direct causal link between the infringement and the damage sustained by the respondents.

The second ground of appeal

85. The Commission submits that the Court of First Instance erred in law in its interpretation and application of the principle of sound administration in finding that the Commission’s inaction between 1 January 2000 and 25 July 2001 constituted an infringement of that principle. (71)

86. The Commission accepts that the principle of sound administration requires it to deal with matters within a reasonable time. The essential question therefore is whether the length of time between 1 January 2000 and 25 July 2001 was unreasonable.

87. The respondents submit that, since the Commission was under an absolute legal obligation under the Regulation to take the necessary measures before 1 January 2000, there is no issue of reasonableness. I do not accept that submission. As indicated in the context of the first ground of appeal, I do not consider that the date specified in Article 14 of the Regulation can be regarded as a deadline which automatically applies whatever the circumstances. Accordingly there is an issue of reasonableness.

88. Turning to the judgment under appeal, it is clear that the Court of First Instance accepted, first, that the progesterone file was scientifically and politically complex, (72) second, that, where the Commission was confronted with a matter which was scientifically and politically complex and sensitive, it was entitled to seek a further opinion from the CVMP, (73) and, third, that the scientific and political complexities of the file might have been such as to prevent the Commission from adopting a draft regulation conforming to the CVMP’s second opinion shortly after its issue. (74)

89. In my view those statements are all manifestly correct.

91. I also consider that it was reasonable for the Commission to postpone its proposal for the classification of progesterone under the Regulation until those decisions had been taken. The complexity of the progesterone dossier was reflected in the different scientific positions taken by the CVMP on the one hand and the SCVPH and other relevant international scientific bodies on the other, and the preamble to the Regulation explicitly recognises that such assessments should be taken into account. (76) It appears that the CVMP had not examined in its risk assessment the risks from the potential misuse or abuse of progesterone as a growth promoter; the SCVPH had however examined those risks and concluded that there was a risk of potential abuse against which specific measures had to be taken. It must also be borne in mind that progesterone, being an endogenous hormone, presents particular problems of both potential abuse and the detection of residues.

92. In particular it seems reasonable for the Commission to have waited until the SCVPH had delivered its re-evaluation in May 2000 before taking the position that progesterone should continue to be authorised for therapeutic and zootechnical treatment by adopting a proposal to amend Directive 96/22. (77) As the Commission submits, it was appropriate to wait until that position had been taken before classifying progesterone in one of the annexes to the Regulation because it is that directive, and not the Regulation, which determines whether progesterone may be used for therapeutic or zootechnical purposes; it is clear from the judgment of the Court of Justice in Monsanto (78) that no MRLs may be set for substances the marketing of which in veterinary medicinal products is prohibited by other provisions of Community law. Moreover Article 7(6) of the Regulation (79) explicitly requires the Commission to take account of Community legislation.

93. The respondents and IFAH submit that the only scientific doubts about the safety of progesterone concerned its use as a growth promoter, a completely different use which is prohibited and which is impossible to achieve with the respondents’ products. I am not convinced that those points are relevant. First, it is stated explicitly in the preamble to the Regulation that account is to be taken of scientific assessments by other scientific committees established in the Community of the safety of the substances concerned ‘where such substances are used for other purposes’. (80) Second, the fact that the respondents’ products cannot be used for growth promotion purposes cannot be relevant to the classification of progesterone for the purposes of the Regulation, since the Regulation concerns the classification of pharmacologically active substances in veterinary medicinal products and not the veterinary medicinal products themselves.

94. The Commission also submits – as a separate point – that the Court of First Instance did not give proper weight to the other interests which were at issue and which the Commission was obliged to take into account, including in particular the objective of the Regulation to protect the health of consumers, which according to settled case-law takes precedence over economic considerations. That argument however seems to me to be simply a further aspect of the submission that the Court of First Instance erred in law in finding that there was unlawful inaction after 1 January 2000 in circumstances where the Commission was still awaiting the outcome of the scientific stage of the process of establishing MRLs and deciding whether to allow progesterone to be used for therapeutic or zootechnical purposes, the MRL to be set and the annex in which to place it, with which I have dealt above.

95.The Commission next submits that the Court of First Instance erred in law in finding that, even if the complexities of the progesterone file were such as to prevent the Commission from adopting a draft regulation shortly after the CVMP’s second opinion, ‘the Commission ought to have concerned itself with the interests of the [respondents], for example by adopting draft measures establishing a provisional MRL on the basis of Article 4 of the 1990 Regulation or by arranging for a (second) deferral of the time-limit laid down in Article 14 thereof’. (81)

96.I agree that the statement that the Commission should have established a provisional MRL on the basis of Article 4 of the Regulation seems dubious: that article applies only ‘provided that there are no grounds for supposing that residues of the substance concerned at the level proposed present a hazard for the health of the consumer’ and thus is presumably not intended to be used in circumstances where the normal timetable is disrupted precisely because of concerns about public health. (82) And, as the Commission points out, the difficulties posed by the fact that progesterone is an endogenous substance and by the lack of reliable analytical detection and control methods are equally relevant to provisional and definitive MRLs.

97.With regard to the Court of First Instance’s alternative suggestion, namely that the Commission should have arranged for a second deferral of the time-limit laid down in Article 14 of the Regulation, it may be doubted whether – even assuming that such a deferral would have assisted the respondents – such a proposal by the Commission would have been effective. When the Commission proposed the previous extension of the time-limit to 1 January 2000, the Economic and Social Committee, in its Opinion on that proposal, stated that ‘the deadline [should] be set definitively for 1 January 2000, with no possibility of extension’. (83) The Commission adds that the Treaty of Amsterdam amended the EC Treaty with effect from 1 May 1999 so as to require co-decision on measures in veterinary fields which have as their direct objective the protection of public health. (84) But the issue of the deferral of the time-limit is secondary; again, the central issue remains whether there were sufficient reasons for the Commission’s delay in taking action.

98.I accordingly conclude that the Court of First Instance erred in law in stating that the Commission’s inaction between 1 January 2000 and 25 July 2001 constituted a breach of the principle of sound administration.

<i>The fourth ground of appeal </i>

99.The Commission submits that the Court of First Instance committed an error of law in finding, first, that the Commission’s inaction between 1 January 2000 and 25 July 2001 constituted a clear and serious breach of the principle of sound administration giving rise, in principle, to liability on the Community’s part (85) and, second, that the damage sustained by the respondents was attributable to the Commission’s inaction. (86)

100.First the Commission submits that the Court of First Instance failed to take into account the complexity of the situation or the degree of discretion enjoyed by the Commission in cases such as the present: on the contrary, the Court of First Instance stated that there was no need to determine the exact scope of the Commission’s discretion in setting MRLs. According to the Commission, that reasoning is circular and flawed because the gravity of the alleged breach depends on whether the Commission’s inaction was administrative or legislative in nature, on the degree of discretion enjoyed and on whether the Commission ‘manifestly and gravely disregarded’ the limits of that discretion.

101.In my view, it follows from the case-law of the Court discussed above (87) that the Court of First Instance, in determining whether the Commission’s inaction gave rise to liability in damages pursuant to the second paragraph of Article 288 EC, should have taken into account the complexity of the situation and, more particularly, the margin of discretion available to the author of the act in question. I accordingly agree with the Commission that that court erred in law in stating that there was no need to determine the exact scope of the Commission’s discretion.

102.It is also clear from the case-law that the question whether the alleged unlawfulness consisted in legislative or administrative action is no longer relevant to the Community’s non-contractual liability. (88) The Court of First Instance was accordingly right to state that there was no need to establish whether the Commission’s action was administrative or legislative in nature.

103.Second the Commission submits that the Court of First Instance misinterpreted the Regulation and its relationship with other provisions of Community law by accepting causality between the alleged damage and the Commission’s ‘inaction’, since national marketing authorisations should have been suspended only in so far as concerned the <i>administration to food-producing animals</i> of veterinary medicinal products containing progesterone but remained valid as regards the <i>production, distribution and exportation </i> of such products and their <i>administration to non food-producing animals</i>. There is accordingly no direct causal link between the Commission’s inaction and the alleged damage.

104.Although that submission may be strictly correct, the significant economic impact none the less presumably arises precisely from the prohibition on administering veterinary medicinal products containing progesterone to food-producing animals. As the respondents submit, by virtue of Article 4(2) of Directive 81/851 (89) veterinary medicinal products containing active substances for which no MRL has been established cannot be marketed. I accordingly do not accept the second plea in the Commission’s fourth ground of appeal.

105.My conclusion on the second and fourth grounds of appeal is therefore that the Court of First Instance erred in law in, first, stating that the Commission’s inaction between 1 January 2000 and 25 July 2001 constituted a breach of the principle of sound administration and, second, failing to determine the scope of the Commission’s discretion before finding the Community liable in damages. The judgment of the Court of First Instance should consequently be set aside.

<b>The substance of the case </b>

106.Article 61 of the Statute of the Court of Justice provides that where an appeal is well founded the Court of Justice may, having quashed the decision of the Court of First Instance, itself give final judgment in the matter where – as in the present case – the state of the proceedings so permits.

107.I have already set out the reasons for my view that the Commission’s delay between 1 January 2000 and 25 July 2001 did not constitute a breach of the principle of sound administration. (90)

108.That does not necessarily mean however that the Commission was justified in taking no action until it adopted the proposal for a Council Regulation on 25 July 2001.

109.I accept that, as the Commission states, (91) it was necessary to wait until it had taken a position on amending Directive 96/22 (92) so as to continue to permit the administration for therapeutic and zootechnical purposes of progesterone to farm animals in derogation from the general prohibition on the administration of hormones to such animals. It first adopted a proposal so to amend Directive 96/22 on 24 May 2000. (93) There was therefore no unjustified delay in the period ending on that date.

110.The next question is whether there was unjustified delay after 24 May 2000. The period of 14 months which elapsed between that date and the Commission’s adoption of the draft regulation seems at first sight unduly long. The Commission however explains why it was not in a position immediately to adopt a proposal for a regulation classifying progesterone. (94) The CVMP opinion had recommended that progesterone be placed in Annex II and had accordingly included no MRLs. In the light of the SCVPH risk assessments of progesterone, however, the Commission decided that that course was not an acceptable risk management measure and hence proposed to include progesterone in Annex I, which meant that MRLs needed to be set before the draft regulation could be adopted. I accept that, in the absence of existing recommended MRLs, the Commission was not in a position to adopt the draft regulation immediately after it had taken a position on amending Directive 96/22.

111.I do not however consider it necessary for the purpose of the present proceedings to take a definitive view on the question whether, and if so when, the Commission came under a duty to adopt a regulation classifying progesterone. The judgment of the Court of First Instance against which the present appeal has been brought ruled that the Community was liable in damages as a result of the Commission’s inaction. The Community will be liable in damages as a result of such inaction only if (i) the inaction is ‘sufficiently serious’ and (ii) there is a direct causal link with the alleged damage. (95) More particularly where the legislative context is characterised by the exercise of a wide discretion there will be no liability unless the institution has ‘manifestly and gravely disregarded the limits on the exercise of its powers’. (96)

112.It is clear that in the field of public health the Commission must have a wide discretion. In particular in sensitive and controversial cases it must have a sufficiently broad discretion and enough time to enable it to arrange for re-examination of the scientific issues which will determine its decision. (97)

113.In all the circumstances, I do not consider that it can reasonably be concluded that the Commission manifestly and gravely disregarded the limits on the exercise of its powers in failing to adopt the draft regulation until July 2001.

114.The developments since that date demonstrate that the question how to classify progesterone for the purpose of the Regulation is complex and divisive. (98) Although no direct comparison can be made with other cases, since each depends on the facts and the specific legislative context, the judgment in <i>Denkavit</i> v <i>Commission</i> may be mentioned. (99) In that case the Court, in dismissing an application for damages allegedly suffered as a result of the Commission’s delay in adopting a measure in the field of maximum permitted levels of substances in animal feed, ruled that the Commission ‘cannot be blamed for having waited [21 months] until it was fully informed before adopting a decision on a matter as complex as the presence in feeding-stuffs of substances which might prove to be undesirable from the point of view of human or animal health’.

115.I am accordingly of the view that the Court should dismiss the claim for damages as unfounded in so far as it relates to the period ending on 25 July 2001. In the context of the present proceedings, there is no need to express a view on whether any liability for damages might arise in respect of the period after that date.

<b>Fifth ground of appeal </b>

116.Given the conclusions I have reached on the first four grounds of appeal and on the substance, it is not necessary to consider the fifth ground of appeal.

<b>Conclusion </b>

117.I am accordingly of the view that the Court should:

(1) set aside the judgment of the Court of First Instance of 26 February 2003 in Joined Cases T-344/00 and T-345/00 <i>CEVA and Pharmacia Enterprises </i>v <i>Commission</i>;

(2) dismiss the actions for damages brought in those cases;

(3) order the respondents to pay the costs of both the proceedings at first instance and the appeal, other than those costs attributable to the cross-appeals;

(4) order the parties and IFAH to bear their own costs in respect of the cross-appeals.

1 – Original language: English.

2 – Joined Cases T-344/00 and T-345/00 <i>CEVA and Pharmacia Enterprises </i>v <i>Commission</i> [2003] ECR II-229.

3 – Directive of 28 September 1981 on the approximation of the laws of the Member States relating to veterinary medicinal products, OJ 1981 L 317, p. 1, as amended in particular by Council Directive 90/676/EEC of 13 December 1990, OJ 1990 L 373, p. 15. Directive 81/851 has been repealed and replaced by Directive 2001/82/EC of the European Parliament and of the Council of 6 November 2001 on the Community code relating to veterinary medicinal products, OJ 2001 L 311, p. 1.

4 – Regulation of 22 July 1993 laying down Community procedures for the authorisation and supervision of medicinal products for human and veterinary use and establishing a European Agency for the Evaluation of Medicinal Products, OJ 1993 L 214, p. 1.

5 – See paragraph 9 below.

6 – Decision of 23 July 1997 setting up Scientific Committees in the field of consumer health and food safety, OJ 1997 L 237, p. 18.

7 – Pursuant to Article 62 of Regulation (EC) No 178/2002 of the European Parliament and of the Council of 28 January 2002 laying down the general principles and requirements of food law, establishing the European Food Safety Authority and laying down procedures in matters of food safety, OJ 2002 L 31, p. 1, every reference in Community legislation to the five Committees set up by Decision 97/579 has been replaced by a reference to the European Food Safety Authority (EFSA). The same legislation created an EFSA Scientific Committee and eight Scientific Panels.

8 – The Codex Alimentarius draws its name from the Codex Alimentarius Austriacus, a collection of standards and product descriptions for a wide variety of foods developed in the Austro-Hungarian Empire between 1897 and 1911.

9 – Regulation of 26 June 1990 (OJ 1990 L 224, p. 1) as amended in particular by Council Regulation (EC) No 434/97 of 3 March 1997, OJ 1997 L 67, p. 1 and Council Regulation (EC) No 1308/1999 of 15 June 1999, OJ 1999 L 156, p. 1.

10 – Article 2.

11 – Article 3.

12 – Article 5.

13 – Article 4.

14 – By Regulation No 1308/1999, cited in note 9.

15 – See paragraph 4 above.

16 – Cited by the Commission in footnote 20 to the appeal.

17 – Regulation No 434/97, cited in note 9.

18 – Cited in note 9.

19 – Cited in note 3.

20 – Cited in note 4.

21 – Directive of 29 April 1996 concerning the prohibition on the use in stockfarming of certain substances having a hormonal or thyrostatic action and of beta-antagonists, and repealing Directives 81/602/EEC, 88/146/EEC and 88/299/EEC, OJ 1996 L 125, p. 3.

22 – Article 3(a). A gestagenic hormone is a female hormone which aids the development and retention of a pregnancy.

23 – Article 4(1).

24 – Case T-120/96 <i>Lilly Industries </i>v <i>Commission</i> [1998] ECR II-2571, paragraphs 83 and 88 of the judgment.

25 – Cited in note 19.

26 – Cited in note 4.

27 – Case C-248/99 P <i>France </i>v <i>Monsanto and Commission</i> [2002] ECR I-1, paragraphs 80 to 82 of the judgment.

28 – Case T-105/96 <i>Pharos </i>v <i>Commission </i>[1998] ECR II-285; on appeal, Case C-151/98 P [1999] ECR I-8157.

29 – Paragraph 26 of the judgment.

30 – Paragraphs 30 to 32 of the judgment.

31 – Cited in note 27; on appeal from Case T-112/97 <i>Monsanto </i>v <i>Commission </i>[1999] ECR II-1277.

32 – Cited in note 3

33 – Cited in note 4.

34 – Paragraphs 80 to 82 of the judgment.

35 – According to the Commission (not challenged by the respondents on this point), the full report became publicly available only in late November 2000.

36 – ‘Assessment of potential risks to human health from hormone residues in bovine meat and meat products’, XXIV/B3/SC4.

37 – EMEA/CVMP/890/99.

38 – EMEA/MRL/146/96-Rev.3.

39 – COM(2000) 320 final; OJ 2000 C 337 E, p. 163.

40 – 10th recital.

41 – Proposal for a Council Regulation amending Annex I of Regulation (EEC) No 2377/90, COM(2001) 627 final.

42 – Recitals 8 and 9.

43 – See paragraph 14 above.

44 – See paragraph 14 above.

45 – Directive of 22 September 2003, OJ 2003 L 262, p. 17. The proposal for Directive 2003/74 had been adopted on 24 May 2000 (see paragraph 42 above).

46 – Regulation amending Annex II to Council Regulation (EEC) No 2377/90, OJ 2003 L 275, p. 9.

47 – 10th recital.

48 – Paragraphs 99 to 103 and 107 of the judgment.

49 – Paragraph 108.

50 – See Case C-252/96 P <i>Parliament</i> v <i>Gutiérrez de Quijano y Lloréns </i>[1998] ECR I-7421, paragraphs 29 to 34.

51 – See for example paragraphs 53 to 55 of the Commission’s defence in Case T-344/00, responding to paragraphs 51 to 57 of the application, which clearly invoke the deadline in Article 14 and which underpin CEVA’s claim for damages (see paragraphs 83 and 103 of its application). The equivalent paragraph numbers in Case T-345/00 are paragraphs 51 to 55 of the defence and paragraphs 44 to 49, 75 and 83 of the application.

52 – Cited in note 9.

53 – Thirsd recital, set out in paragraph 17 above.

54 – Cited in note 9.

55 – Third recital, set out in paragraph 17 above.

56 – By way of comparison, the Commission stated, in reply to written questions from the Court of First Instance, that as at 1 January 2000 about 40 substances (of the 700-odd ‘old substances’) remained unclassified; almost all those dossiers were regularised in the course of 2000. By September 2002, only progesterone and norgestomet, another hormonal substance, were still unclassified.

57 – Paragraphs 99, 101 and 102 of the judgment.

58 – Cited in note 24.

59 – Cited in note 31.

60 – Cited in note 28.

61 – See paragraphs 71 and 86 of the judgment.

62 – See paragraphs 88 and 89 of the judgment.

63 – Article 7(4), original version, and Article 7(6), as amended: see paragraphs 12 and 13 above. See also by analogy Case C-120/97 <i>Upjohn </i>[1999] ECR I-223, paragraph 47 of the judgment, in which the Court stated that the opinion of the Committee for Proprietary Medicinal Products, which was set up with a similar mandate to the CVMP, is not binding.

64 – Third recital.

65 – Joined Cases C-46/93 and C-48/93 [1996] ECR I-1029, paragraphs 42, 43 and 51 of the judgment.

66 – Paragraph 45 of the judgment.

67 – Case C-352/98 P <i>Bergaderm and Goupil</i> v <i>Commission</i> [2000] ECR I-5291.

68 –

– See paragraphs 39 to 47.

—

69Case C-312/00 P [2002] ECR I-11355.

—

70Case C-472/00 P [2003] ECR I-7541.

—

71Paragraph 103 of the judgment.

—

72Paragraph 100 of the judgment.

—

73Paragraph 99 of the judgment.

—

74Paragraph 102 of the judgment.

—

75Paragraph 101 of the judgment.

—

76Third recital, set out in paragraph 8 above.

—

77Cited in note 2145.

—

78Cited in note 31.

—

79Set out in paragraph 13 above.

—

80Third recital, set out in paragraph 8 above.

—

81Paragraph 102 of the judgment

—

82It may be noted that the Commission submitted a proposal to the Standing Committee in December 2002 to place progesterone in Annex III; that proposal did not receive a favourable reading.

—

83OJ 1997 C 133, p. 27, paragraph 3.4.

—

84Article 26 of the Treaty of Amsterdam amended Article 129(4)(b) of the EC Treaty, now Article 152(4)(b) EC.

—

85Paragraph 103 of the judgment.

—

86Paragraph 107 of the judgment.

—

87See paragraphs 80 to 83.

—

88See paragraph 83 above.

—

89Cited in note 19.

—

90See paragraph 91 above.

—

91See paragraph 75 of the appeal.

—

92Cited in note 21.

—

93See paragraph 42.

—

94See paragraphs 85 and 86 of the appeal.

—

95See paragraph 80 above.

—

96See paragraph 83 above.

—

97Case T-199/96 Bergaderm and Goupil v Commission [1998] ECR II-2805, paragraph 55 of the judgment, endorsed on appeal in Bergaderm, cited in note 67, paragraph 66.

—

98See paragraphs 46 and 49 above.

—

99Case 14/78 [1978] ECR 2497, paragraph 20 of the judgment.

—