1.

Under the REACH Regulation, ( 2 ) the European Union is able to restrict the manufacturing, use or placing on the market of substances of very high concern. The Commission can thus make the use of such substances subject to authorisation. In preparation for this measure, the European Chemicals Agency (ECHA) must first place the substances in question on the ‘candidate list’, as a result of which certain requirements are imposed.

2.

Substances which present a very high concern include those which are persistent, bioaccumulative and toxic (PBT) or very persistent and very bioaccumulative (vPvB). ( 3 ) These substances are degraded only very slowly or not at all and therefore have the potential to accumulate in parts of the environment. The effects of this accumulation are unpredictable in the long term. Such accumulation is also difficult to reverse in practice. Furthermore, PBT or vPvB substances may contaminate remote areas that should be protected from further contamination by hazardous substances resulting from human activity. ( 4 )

3.

The present appeal was lodged because ECHA regards the substances at issue, octamethylcyclotetrasiloxane (D4), decamethylcyclopentasiloxane (D5) and dodecamethylcyclohexasiloxane (D6), as PBT/vPvB substances. By the contested decision, ( 5 ) ECHA therefore included those three substances in the candidate list for the imposition of an authorisation requirement.

4.

The appeal concerns, first and foremost, the rules governing the method for assessing bioaccumulation and certain scientific/factual objections to the application of those rules.

5.

Article 13 of the REACH Regulation lays down general requirements for generation of information on intrinsic properties of substances. Test methods are addressed in paragraph 3:

‘Where tests on substances are required to generate information on intrinsic properties of substances, they shall be conducted in accordance with the test methods laid down in a Commission Regulation or in accordance with other international test methods recognised by the Commission or the Agency as being appropriate.’

6.

The REACH Regulation empowers the Commission to make the use of certain substances subject to an authorisation requirement (Title VII).

7.

PBT and vPvB substances in particular may be considered for an authorisation requirement under Article 57(d) and (e) of the REACH Regulation:

‘The following substances may be included in Annex XIV in accordance with the procedure laid down in Article 58:

…

8.

Article 59 of the REACH Regulation sets out the procedure for the identification of substances referred to in Article 57 and provides that the ECHA must place the substances thus identified on a candidate list for eventual inclusion in Annex XIV.

9.

Annex XIII to the REACH Regulation lays down the criteria for the identification of PBT and vPvB substances.

10.

The second paragraph in the preamble to Annex XIII to the REACH Regulation describes the method for applying the criteria:

‘For the identification of PBT substances and vPvB substances a weight-of-evidence determination using expert judgement shall be applied, by comparing all relevant and available information listed in Section 3.2 with the criteria set out in Section 1. This shall be applied in particular where the criteria set out in Section 1 cannot be applied directly to the available information.’

11.

The fourth paragraph in the preamble to Annex XIII to the REACH Regulation states that the ‘information used for the purposes of assessment of the PBT/vPvB properties shall be based on data obtained under relevant conditions’.

12.

Under the fifth paragraph in the preamble to Annex XIII, ‘the identification shall also take account of the PBT/vPvB properties of relevant constituents of a substance and relevant transformation and/or degradation products’.

13.

The sixth paragraph in the preamble to Annex XIII to the REACH Regulation provides that the Annex applies ‘to all organic substances, including organo-metals’.

14.

Bioaccumulation is defined in Sections 1.1.2 and 1.2.2 of Annex XIII to the REACH Regulation by reference to the bioconcentration factor:

‘1.1.2. Bioaccumulation

A substance fulfils the bioaccumulation criterion (B) when the bioconcentration factor in aquatic species is higher than 2000.

…

1.2.2. Bioaccumulation

A substance fulfils the “very bioaccumulative” criterion (vB) when the bioconcentration factor in aquatic species is higher than 5000.’

15.

In addition, Section 3.2 of Annex XIII to the REACH Regulation concerns the information to which regard is to be had:

‘3.2. Assessment Information

The following information shall be considered for the assessment of P, vP, B, vB and T properties, using a weight-of-evidence approach:

3.2.1. …

3.2.2. Assessment of B or vB properties:

16.

On the basis of Article 13(3) of the REACH Regulation, the Commission adopted Regulation (EC) No 440/2008, ( 6 ) which, in Section C.13 of its Annex, regulates bioaccumulation in fish: aqueous and dietary exposure.

17.

The first paragraph of the introduction to Section C.13 explains the background to the test method:

‘This test method (TM) is equivalent to OECD test guideline (TG) 305 (2012). The major goal of this revision of test method is two-fold. Firstly, it is intended to incorporate a dietary bioaccumulation … test suitable for determining the bioaccumulation potential of substances with very low water solubility.’

18.

The second paragraph of the introduction to Section C.13 gives the reasons for introducing a dietary test method:

‘… In addition, it has been recognised that testing very poorly water soluble substances may not be technically feasible. In addition, for substances with very low water solubility in the aquatic environment, exposure via water may be of limited importance in comparison to the dietary route. This has led to the development of a test method in which fish are exposed via their diet.’

19.

The sixth paragraph of the introduction to Section C.13 explains the requirement for the aqueous exposure test:

‘… The aqueous exposure test is most appropriately applied to stable organic substances with log K OW values between 1,5 and 6,0 (13) but may still be applied to strongly hydrophobic substances (having log K OW > 6,0), if a stable and fully dissolved concentration of the test substance in water can be demonstrated. If a stable concentration of the test substance in water cannot be demonstrated, an aqueous study would not be appropriate thus the dietary approach for testing the substance in fish would be required (although interpretation and use of the results of the dietary test may depend on the regulatory framework).’

20.

The eighth paragraph of the introduction to Section C.13 addresses the choice between different test methods:

‘The decision on whether to conduct an aqueous or dietary exposure test, and in what set-up, should be based on the factors in paragraph 3 considered together with the relevant regulatory framework. For example, for substances which have a high log K OW but still show appreciable water solubility with respect to the sensitivity of available analytical techniques, an aqueous exposure test should be considered in the first instance. However it is possible that information on water solubility is not definitive for these hydrophobic types of substances, so the possibility of preparing stable, measurable dissolved aqueous concentrations (stable emulsions are not allowed) applicable for an aqueous exposure study should be investigated before a decision is made on which test method to use … It is not possible to give exact prescriptive guidance on the method to be used based on water solubility and octanol-water partition coefficient “cut off” criteria, as other factors (analytical techniques, degradation, adsorption, etc.) can have a marked influence on method applicability for the reasons given above. However, a log K OW above 5 and a water solubility below ~ 0,01 – 0,1 mg/l mark the range of substances where testing via aqueous exposure may become increasingly difficult.’

21.

The fifth paragraph of the chapter entitled ‘Principle of the test’ in Section C.13, Part I, concerns growth corrections in carrying out aqueous exposure tests:

‘The increase in fish mass during the test will result in a decrease of test substance concentration in growing fish (so-called growth dilution), and thus the kinetic [bioconcentration factor] will be underestimated if not corrected for growth …’

22.

On 14 October 2014, the Executive Director of ECHA requested the ECHA Member State Committee to deliver an opinion on the persistence and bioaccumulation of D4 and D5 in the light of the criteria laid down in Annex XIII to the REACH Regulation. On 22 April 2015, the Committee adopted an opinion according to which D4 and D5 meet the criteria for vPvB substances and D4 also meets the criteria for PBT substances. ( 7 )

23.

On 1 March 2017, the competent authority of the Federal Republic of Germany submitted dossiers under Article 59(3) of the REACH Regulation in accordance with Annex XV, proposing that D4 and D5 be identified as substances of very high concern on account of their PBT and vPvB properties. On 21 December 2017, the Commission also requested ECHA pursuant to Article 59(2) to prepare a dossier in accordance with Annex XV for the identification of D6 as a substance of very high concern. In that dossier, ECHA concluded that D6 meets the PBT and vPvB criteria.

24.

On 13 June 2018, the Member State Committee submitted support documents in respect of all three substances at issue, in which it stated that those substances meet the conditions for PBT and vBvP substances.

25.

On 27 June 2018, pursuant to Article 59(8) of the REACH Regulation, ECHA adopted the contested decision ED/61/2018 regarding the inclusion of D4, D5 and D6 in the candidate list for eventual inclusion in Annex XIV on the ground that they had been identified as PBT and vPvB substances within the meaning of Article 57(d) and (e). At the same time, ECHA updated the candidate list to include entries for D4, D5 and D6. The remarks on the respective candidate list entries state that D5 and D6 fulfil the criteria laid down in Article 57(d) for a PBT substance when they contain a quantity of D4 greater than or equal to 0.1%.

26.

The first appellant, Global Silicones Council, represents companies which manufacture and sell silicone products throughout the world. The other appellants are companies which manufacture, sell and supply silicone products, in particular the substances at issue. They brought an action against the contested decision, which the General Court dismissed by the judgment under appeal of 30 June 2021, Global Silicones Council and Others v ECHA (T‑519/18, not published, EU:T:2021:404).

27.

On 8 September 2021, the appellants lodged the present appeal. They claim that the Court should:

28.

American Chemistry Council (ACC) intervened in support of the appellants in the proceedings before the General Court and supports the forms of order sought by them.

29.

ECHA contends that the Court should:

30.

Germany and the Commission intervened in support of ECHA in the proceedings before the General Court. They support ECHA and contend that the Court should order the appellants to pay the costs.

31.

The parties submitted written observations. The Court did not consider it necessary to hold a hearing as it did not require any further information on the basis of the written procedure.

32.

The contested decision is based inter alia on the finding that the substances at issue are toxic and bioaccumulative or very bioaccumulative.

33.

Only the first part of the fourth ground of appeal concerns the finding that D5 and D6 are toxic because D4 is present in them as an impurity (see B). The appeal focuses instead on the finding of the bioaccumulation of the substances at issue (see A).

34.

Bioaccumulation is generally referred to as a process in which the chemical concentration of a substance in an organism achieves a level that exceeds the concentration in the surrounding medium (such as water for a fish or air for a mammal), the diet, or both. ( 8 ) Bioaccumulative substances are problematic because even a low environmental impact from those substances can lead to accumulation in the affected organisms and thus have a high impact on them.

35.

The contested decision is based, with regard to bioaccumulation, primarily on studies on the bioconcentration factor of the substances at issue. ( 9 ) This is the concentration of the test substance in or on a fish or specified tissues thereof divided by the concentration of the substance in the surrounding medium. ( 10 ) Germany describes the test method used, ‘aqueous exposure’, as a process where a constant concentration of the substance in question is maintained in an aquarium without sediment, such that the fish in the aquarium absorb the substance directly from the water, in particular through the gills.

36.

The appellants consider this method and the related studies to be inappropriate for assessing the bioaccumulation of the substances at issue that is actually occurring. They asserted before the General Court that, due to their low water solubility and their partitioning properties in particular, those substances, if released into water, would either bind to particulates that deposit to sediment or partition readily to air. Consequently, the bioconcentration factor determined in laboratory tests – the absorption of the substances directly from the water – is not based on data obtained under realistic conditions. Instead, regard should have been had primarily to studies of dietary accumulation – the scientific terms biomagnification ( 11 ) and trophic magnification ( 12 ) are used. Some of those studies conclude that there is no bioaccumulation.

37.

The first three grounds of appeal and parts of the fourth ground of appeal combine objections regarding the interpretation of the significant rules, which are laid down mainly in Annex XIII to the REACH Regulation (see 1), and factual arguments concerning the scientific information on the substances at issue (see 2).

38.

The disagreement over Annex XIII to the REACH Regulation concerns the legal test to be applied in assessing bioaccumulation. Although Sections 1.1.2 and 1.2.2 of Annex XIII contain clear thresholds based on bioconcentration, the appellants and ACC instead wish for the probative value of all available studies to be comprehensively assessed and, in particular, for greater weight to be attached to certain studies on dietary bioaccumulation (first ground of appeal in particular, see a). To that end, by the third ground of appeal they attempt to demonstrate that the laboratory tests to determine the bioconcentration factor do not produce data obtained under ‘relevant conditions’ (see b). In the first limb of the third part of the fourth ground of appeal, they claim, with the same aim, that a provision of the CLP Regulation ( 13 ) should be applied mutatis mutandis (see c). By the second ground of appeal, they object that the General Court distorted some of their pleas (see d). All these objections cannot, however, call into question the judgment under appeal.

39.

By the first ground of appeal, the appellants and ACC object in particular to paragraphs 68 to 71 of the judgment under appeal, where the General Court finds that bioaccumulation is established primarily by determining the bioconcentration factor. With regard to D6 specifically, the appellants reiterate this view in the third limb of the third part of the fourth ground of appeal in order to challenge paragraph 226 of the judgment under appeal, where the General Court also made this finding against them.

40.

In paragraph 68 of the judgment under appeal, the General Court explains that in Sections 1.1.2 and 1.2.2 of Annex XIII to the REACH Regulation, the legislature specified that bioaccumulation is to be assessed in relation to the bioconcentration factor in aquatic species.

41.

This finding follows from the wording of those provisions, under which the ‘bioaccumulative’ criterion is fulfilled – without any limitation – when the bioconcentration factor is 2000 and the ‘very bioaccumulative’ criterion is fulfilled when the bioconcentration factor is 5000.

42.

The appellants and ACC do rightly stress that, according to the second paragraph in the preamble to Annex XIII to the REACH Regulation, all relevant and available information must be taken into consideration in the form of a ‘weight-of-evidence determination’ and compared with the criteria set out in Section 1, in this case Sections 1.1.2 and 1.2.2. They conclude that, in addition to the bioconcentration factor, ECHA should have attached greater importance to dietary bioaccumulation, in particular the biomagnification factor and the trophic magnification factor.

43.

Both factors are in fact to be taken into consideration in the assessment of bioaccumulation under Section 3.2.2(c) of Annex XIII to the REACH Regulation.

44.

However, that information is merely an aid to be used if the bioconcentration factor cannot be reliably determined, as the General Court correctly explained in paragraphs 69 and 70 of the judgment under appeal. This is clear from recital 6 of Regulation No 253/2011 ( 14 ) and from the second sentence of the second paragraph in the preamble to Annex XIII to the REACH Regulation, according to which a weight-of-evidence determination and a comparison with the information listed in Section 3.2 of Annex XIII is necessary in particular where the criteria set out in Section 1 of Annex XIII, in this case the abovementioned thresholds laid down in Sections 1.1.2 and 1.2.2, cannot be applied directly to the available information. It follows, conversely, that the weight-of-evidence determination must first clarify whether the available studies have reliably determined the bioconcentration factor.

45.

In so far as the appellants refer to statements made by ECHA in guidance, ( 15 ) according to which regard should be had not only to the bioconcentration factor, they cite that text very selectively. It is also clear from its overall context, rather, that the determination of the bioconcentration factor is the preferred method and other information is of interest primarily if the information on the bioconcentration factor is incoherent or doubtful for other reasons. ( 16 )

46.

The first sentence of Article 13(3) of the REACH Regulation suggests the same conclusion. Under that provision, tests to generate information on intrinsic properties of substances are to be conducted in accordance with the test methods laid down in a Commission regulation or in accordance with other international test methods recognised by the Commission or the Agency as being appropriate.

47.

When the Member State Committee made the findings regarding the bioaccumulation of D4 and D5 in 2015, the Commission had fully recognised the aqueous exposure test applied in this instance to determine bioconcentration in Section C.13 of the Annex to Regulation No 440/2008.

48.

However, this entry recognising the test method no longer applied in 2018 when the Member State Committee submitted the support documents for the contested decision. The Commission had amended it on the basis of revised OECD Guidelines. ( 17 )

49.

The Commission newly incorporated a dietary bioaccumulation test (through biomagnification or trophic magnification) for substances with very low water solubility. ( 18 ) Accordingly, the decision on the test set-up should be based on water solubility and the octanol/water partition coefficient. ( 19 ) Regard must be had, on the one hand, to the regulatory framework and, on the other, to practical method applicability as testing via aqueous exposure may become increasingly difficult in the case of certain substance properties. ( 20 )

50.

The aqueous exposure test for the bioconcentration factor nevertheless continues to be the preferred method for determining bioaccumulation even after these amendments. Recourse must be had to test methods based on food intake only if that method causes difficulties.

51.

In paragraph 71 of the judgment under appeal, the General Court therefore correctly holds that the EU legislature, in this case the Commission, chose (on the basis of the OECD Guidelines) to give a certain priority to the results of reliable studies on the bioconcentration factor of a substance in aquatic species. In the light of the consideration of the third ground of appeal, it is possible to examine whether that choice was justified.

52.

The General Court therefore also correctly ruled in paragraphs 73 and 74 that information other than the results of studies on bioconcentration would have to invalidate or refute those results in order to call into question the findings regarding bioaccumulation. The objections raised by the appellants and ACC in this regard must therefore also be rejected.

53.

Accordingly, the first ground of appeal and the third limb of the third part of the fourth ground of appeal are unfounded in so far as they contest the findings regarding the legal test to be applied in assessing bioaccumulation.

54.

By the third ground of appeal, the appellants also challenge the assessment of bioaccumulation by reference to the bioconcentration factor. To that end, they rely on the fourth paragraph in the preamble to Annex XIII to the REACH Regulation, under which the information used for the purposes of assessment of the PBT/vPvB properties is to be based on data obtained under relevant conditions. They dispute that the laboratory tests used are ‘relevant conditions’ for obtaining the significant data because, by virtue of their properties, the substances at issue are almost not present in the environment in that form.

55.

The Commission contends that the appellants did not make this submission in good time and that it is therefore inadmissible. The appellants did mention the notion of ‘relevant conditions’ a number of times in their action before the General Court but merely inferred from it that the different criteria should be assessed in the same environmental medium (compartment). ( 21 ) As the General Court states in paragraph 105 of the judgment under appeal, however, the appellants raised the complaint of unrealistic laboratory tests only in the reply in connection with the ‘relevant conditions’. In paragraph 132, it nevertheless does not consider it necessary to determine whether that submission is out of time.

56.

Like the General Court, I do not think it appropriate to reject this submission in isolation as inadmissible. The legal characterisation of the properties of the substances at issue, which it is admissible for the appellants to question, ( 22 ) requires clarification of the legal test to be applied. Consideration should be given in this connection to the notion of ‘relevant conditions’.

57.

As far as that legal test is concerned, under this ground of appeal it is claimed, in essence, that the clear definition of bioaccumulation based on the bioconcentration factor under Sections 1.1.2 and 1.2.2 of Annex XIII to the REACH Regulation is not intended to be significant where the test conditions for determining that factor are not actually present in nature. In that case, the test results would not be obtained under ‘relevant conditions’.

58.

If this argument were accepted, this would constitute an additional restriction of the application of that criterion and the associated test method in relation to the previous considerations. It would be relevant not only if there are difficulties in applying the method, but also if the results are realistic in view of the actual environmental properties of the substances concerned.

59.

The General Court refutes this argument, in essence, in paragraphs 88 and 89, 116 to 119 and 133 of the judgment under appeal, asserting that the contested decision is based on an assessment of the hazards arising from the properties of substances, but not on the assessment of the risks associated with the use of the substance.

60.

This distinction between hazards and risks appears to be very artificial at first sight, a point also criticised by the appellants. It has, however, rightly been recognised by the Court on many occasions, ( 23 ) as it has in view different assessment procedures. The risk assessment is governed by Section 6 of Annex I to the REACH Regulation, ( 24 ) while the hazard assessment is regulated in Annex XIII. The former is based on exposure scenarios, a real-life analysis of the envisaged use of a substance and the specifically associated risks. The hazard assessment, on the other hand, is intended to identify the intrinsic properties of a substance irrespective of its use.

61.

Accordingly, the General Court’s argument is, in essence, that the ‘relevant conditions’ under the fourth paragraph in the preamble to Annex XIII to the REACH Regulation do not relate to how the substance in question behaves when used under actually existing environmental conditions but only to whether the test methods are suitable for identifying the hazardous intrinsic properties of the substances concerned.

62.

If such a test method is not just to provide theoretically interesting results but is also to be of practical benefit, it must fulfil two conditions in particular. First, the results of the test of intrinsic properties of different substances must be comparable with one another. Second, it must be structured in such a way that it allows conclusions to be drawn regarding the risks associated with different uses. The tests must therefore be based on standardised experimental conditions which provide comparable results and allow conclusions to be drawn regarding the risks of different uses. ( 25 )

63.

ECHA and the Commission thus argue convincingly that relevant conditions merely means the generally recognised experimental procedures for hazard assessment. ( 26 ) This is consistent with the abovementioned first sentence of Article 13(3) of the REACH Regulation, according to which tests to generate information on intrinsic properties of substances are to be conducted in accordance with the test methods laid down in a Commission regulation or in accordance with other international test methods recognised by the Commission or the Agency as being appropriate.

64.

The Commission regularly updates these test methods in order to take account of technical progress. For that reason, in 2011 it amended Annex XIII to the REACH Regulation ( 27 ) and in 2017 the relevant Section C.13 of the Annex to Regulation No 440/2008. ( 28 ) In these adaptations, the Commission has placed greater emphasis on the assessment of bioaccumulation based on food intake. It has not stated, however, that this method is preferable where the determination of bioconcentration is based on conditions which are not normally present in nature. As has already been stated, ( 29 ) it rules out that method only where it does not provide reliable results because of difficulties in conducting the test.

65.

Consequently, the legislature, in this case the Commission, has already defined on the basis of the significant scientific findings the relevant methods for identifying intrinsic properties which are to be applied in assessing the bioaccumulation of the substances at issue. In doing so, it must be considered to have already taken sufficient account of the actual environmental fate of such substances.

66.

This appraisal is not precluded by the fact that different views are held within the scientific community on how bioaccumulation should be assessed. Since the definition of these methods is characterised by a high degree of scientific complexity, the Commission enjoys a broad discretion in this regard, ( 30 ) allowing it to decide between different scientific views and to choose one of them.

67.

Moreover, the appellants have not called these rules into question as such either before the General Court or in the appeal proceedings.

68.

The judgment under appeal is not therefore vitiated by an error in law because the General Court declined to understand the notion of ‘relevant conditions’ to the effect that the test methods for assessing bioaccumulation must be consistent with the actual presence of the substances in question in the environment.

69.

To this extent, the third ground of appeal is also unfounded.

70.

As a further argument against the primary consideration of bioconcentration, by the first limb of the third part of the fourth ground of appeal in connection with the determination of the bioaccumulation of D6, the appellants rely on the CLP Regulation. ( 31 ) They object that in paragraphs 96 and 225 of the judgment under appeal, the General Court declined to have regard to Section 1.1.1.3 of Annex I to that regulation.

71.

It is not clear, however, what additional benefit the appellants expect to gain from the consideration of that provision as it contains only an abstract explanation of the establishment of probative value which is largely the same as the second and third paragraphs in the preamble to Annex XIII to the REACH Regulation. If these (undoubtedly applicable) provisions cannot call into question the clear thresholds for the determination of bioaccumulation based on the bioconcentration factor, a mutatis mutandis application of a similar provision of the CLP Regulation cannot produce a different outcome.

72.

Consequently, this argument is not relevant to the decision (‘ineffective’) and must therefore be rejected.

73.

The second ground of appeal concerns the rules governing the assessment of bioaccumulation only on a superficial examination; in fact, the appellants and ACC are addressing the legal characterisation of the properties of the substances at issue. The General Court, however, erroneously considered the corresponding plea as an argument against the applicability of those rules.

74.

In this regard the appellants challenge paragraphs 141 and 150 of the judgment under appeal, where the General Court refers to the sixth paragraph in the preamble to Annex XIII to the REACH Regulation, according to which that annex applies to all organic substances, including organo-metals.

75.

The appellants argued before the General Court that the ‘hybrid nature’ of the substances at issue should be taken into account in the assessment of bioaccumulation. They refer in this respect to the organic-inorganic or ‘mainly inorganic’ structure of the substances at issue. ( 32 )

76.

In paragraph 141 of the judgment under appeal, the General Court answers that a substance with a ‘hybrid’ structure would not necessarily be excluded from the scope of Annex XIII to the REACH Regulation. In paragraph 150, the General Court further states that Annex XIII is applicable to the substances at issue because they are organic substances.

77.

The appellants rightly argue, however, that in these findings the General Court failed to respond to its arguments on this point. They did not submit that Annex XIII to the REACH Regulation was not applicable to the substances at issue. As is shown by reading the application before the General Court, they criticised the manner in which that annex was applied. ( 33 ) Specifically, they raised doubts as to the suitability of the bioconcentration factor for determining bioaccumulation based on the fact that the substances at issue have a hybrid organic-inorganic structure. ( 34 )

78.

The findings in paragraphs 141 and 150 of the judgment under appeal do not therefore respond to the appellants’ pleas and are based on a distortion of those pleas. ( 35 )

79.

If the General Court had also ignored the objections which the appellants based on the hybrid nature of the substances at issue, the judgment under appeal would be vitiated by an insufficient statement of reasons and would have to be set aside.

80.

The appellants themselves nevertheless rightly acknowledge that the General Court did address those objections. I will deal with this point in connection with the legal characterisation of the scientific information on the substances at issue.

81.

The distortion of the appellants’ pleas underlying paragraphs 141 and 150 of the judgment under appeal therefore has no consequences to the detriment of the appellants. Accordingly, the second ground of appeal is not relevant to the decision (‘ineffective’) and must also be rejected.

82.

In summary, it must be stated that under Annex XIII to the REACH Regulation the aqueous exposure test for the bioconcentration factor is the preferred method for determining bioaccumulation. However, recourse may also be had to test methods based on food intake if that method causes difficulties. The appellants’ arguments against this interpretation of Annex XIII are unfounded.

83.

The appellants also present factual arguments in connection with all the grounds of appeal asserting that the specific properties of the substances at issue preclude the use of the bioconcentration factor to determine bioaccumulation. It is necessary in this regard to examine, first, the admissibility of the submission and to clarify the standard of review in the appeal (see a), second, the objections regarding the investigation of the facts (see b) and, lastly, the assessment of the relevant information (see c).

84.

The Commission, and to some extent also ECHA and Germany, take the view that the assessment of the relevant information is not subject to review by the Court on appeal because it requires a fresh assessment of the facts. The appeal is thus inadmissible.

85.

They are correct in their view that the evaluation of individual studies cannot be requested on appeal, as the assessment of facts and evidence does not, save where the clear sense of the facts and evidence has been distorted, constitute a question of law which is subject, as such, to review by the Court of Justice in the context of an appeal. ( 36 )

86.

Although the appellants attempt to counter this argument by alleging that the General Court distorted facts, in practice they are objecting, first and foremost, to the conclusions which the General Court draws from those facts.

87.

This does not mean, however, that the appellants’ submissions regarding the facts are inadmissible. Rather, the Court of Justice has jurisdiction under Article 256 TFEU to review their legal characterisation and the legal conclusions which were drawn therefrom, where the General Court has determined or assessed the facts. ( 37 ) Accordingly, the Court has even already ruled that whether the low solubility of a substance must be taken into consideration for the purposes of classifying the aquatic environmental hazards posed by that substance is a question of the legal characterisation of facts which is subject to review on appeal. ( 38 )

88.

In the present case, the subsumption of the (essentially undisputed) scientific information on the substances at issue must therefore be reviewed on the basis of the legal test developed above for the determination of bioaccumulation.

89.

However, it is primarily the responsibility of ECHA to assess the properties of the substances at issue with a view to whether their bioaccumulation should be determined by reference to the bioconcentration factor or other test methods. Since this is a scientifically and technically highly complex question, ECHA enjoys a broad discretion. The EU judicature may review such measures only to verify whether manifest errors of appraisal have been committed or the limits of discretion have been manifestly exceeded. In such a context, the EU judicature cannot substitute its assessment of scientific and technical facts for that of the authorities on which alone the legislature has placed that task. ( 39 )

90.

In particular, where a party claims that the authority competent in the matter has committed a manifest error, the EU judicature must verify whether that authority has examined, carefully and impartially, all the relevant facts of the individual case on which that assessment was based. ( 40 ) Although a broad discretion is also available in the finding of the basic facts, the competent authority must show before the EU judicature that it has taken into consideration all the relevant factors and circumstances. ( 41 )

91.

In judicial proceedings, this means, in principle, that the party which is claiming an inadequate investigation of the relevant factors or manifest errors of assessment must first adduce factors justifying the doubts, which are relevant to the decision, regarding the plausibility of the assessment made by the EU authority in question. ( 42 ) Only then must that authority rebut those doubts, if necessary. This is not a reversal of the burden of proof, contrary to the appellants’ view, but – on the contrary – a fundamental principle of EU procedural law. Since the EU judicature does not normally investigate of its own motion, the burden of proof nearly always rests with the party which invokes a certain fact.

92.

In the present case, the hurdle which the appellants must overcome in order to be successful is even higher as the determination of bioaccumulation is not based on a free, purely scientific assessment of the available information on the properties of the substances in question. Rather, according to the above explanation it is clear from Annex XIII to the REACH Regulation that the determination of bioaccumulation is to be based, as a rule, on an aqueous exposure test for the bioconcentration factor. Recourse may be had exceptionally to other test methods, in particular connected with food intake, only if that method raises particular difficulties.

93.

Consequently, the complaint raised by the appellants in the first and second grounds of appeal that, in paragraphs 74 and 156 of the judgment under appeal, the General Court reversed the burden of proof is not convincing. In making the statements in question, the General Court merely made clear that the appellants’ submissions did not give sufficient grounds for doubts as to the plausibility of the assessment by ECHA.

94.

With regard to the investigation of the facts, by the second limb of the fourth ground of appeal, the appellants object in essence that the General Court stated insufficient reasons for rejecting a complaint in this regard. They had submitted that ECHA had failed to take adequate account of studies published between the two opinions of the Member State Committee in 2015 and 2018.

95.

The General Court rejected this complaint in paragraphs 170 and 171 of the judgment under appeal, listing those studies and stating that they had been taken into consideration.

96.

Viewed in isolation, this statement of reasons does actually seem very brief.

97.

As ECHA explains, however, in paragraph 169 of the judgment under appeal the General Court also refers to the fact that it is apparent from the support documents of the Member State Committee submitted in 2018 that the studies in question were taken into consideration. It would be unreasonable to expect the General Court, as it were, to ‘recapitulate’ the assessment of the studies in those documents in order to demonstrate that the studies were not only mentioned but also substantively taken into consideration.

98.

Furthermore, at various points, particularly in paragraphs 185 to 188 and 196 to 200 of the judgment under appeal, the General Court discusses the way in which ECHA and the Member State Committee took some of those studies into consideration. In addition, in paragraphs 73 to 77 of the judgment under appeal, the General Court discusses the examination by the Member State Committee of studies on dietary bioaccumulation.

99.

Lastly, the appellants’ argument that the inadequate consideration of those studies can be seen from the fact that ECHA determined the bioaccumulation of the substances at issue is not relevant to this question. It does not concern the examination of the relevant factors, but the conclusions drawn from that information, that is to say, the legal characterisation of the facts, which will be addressed below.

100.

Consequently, the General Court rightly rejected the complaint of an inadequate investigation of the facts, and the appellants’ plea in this regard is unfounded.

101.

As regards the legal characterisation of the available information, the appellants submit, in essence, that because of their hybrid nature (second ground of appeal), the substances at issue have certain properties as a result of which the aqueous exposure test for bioconcentration produces unrealistic results (third ground of appeal). This should have been identified by way of a weight-of-evidence determination of the available studies and led to the use of studies based on food intake (first ground of appeal).

102.

In particular, the reliance on the need for a weight-of-evidence determination is based on an already rejected interpretation of Annex XIII to the REACH Regulation, as bioaccumulation is to be assessed principally by reference to bioconcentration. ( 43 ) This does not mean, however, that the appellants’ factual submissions cannot demonstrate difficulties precluding the application of that test method.

103.

The ‘hybrid’ nature of the substances at issue refers to the fact that the essential characteristic of their structure is an inorganic ring of silicon and oxygen to which organic groups, so-called alkyl groups consisting of carbon and hydrogen, are bound. ECHA, ( 44 ) for example, presents the structural formula of D4 as follows:

104.

The specific properties of the substances at issue are based on this structure. In particular, their solubility in water is low and the octanol-water partition coefficient is high. This partition coefficient serves as a measure of the relationship between lipophilicity (fat solubility) and hydrophilicity (water solubility) of a substance. The value is greater than one if a substance is more soluble in fat-like solvents such as n-octanol and less than one if it is more soluble in water. ( 45 ) For D4, ECHA gives a value of 56 μg/L for water solubility and a value of 6.98 for the partition coefficient. ( 46 ) For D5, those values are 17 μg/L and 8.07 ( 47 ) and for D6 5.1 μg/L and 8.87. ( 48 )

105.

These properties form the basis for the appellants’ submission, which is not disputed by ECHA and the Commission, that the substances at issue are not normally present under natural conditions in water but would either deposit into the water sediment or evaporate into the air. They therefore take the view that ECHA should not have been permitted to base the assessment of bioaccumulation on the bioconcentration factor in the light of aqueous exposure, but should have relied on studies on the intake of the substances at issue through diet.

106.

The General Court responds to this submission in paragraph 117 of the judgment under appeal, stating that the appellants had not explained how the substances at issue could be present in certain organisms if they were not present in water. It relies in this regard on the support documents of the Member State Committee which mention, in addition to laboratory studies, studies according to which the substances at issue are present in certain organisms in nature. The General Court’s argument is not convincing, however, as those organisms could have absorbed the substances directly or indirectly from the sediment through diet, as is also argued by the appellants.

107.

The appellants’ view is further corroborated by the second paragraph of the introduction to Section C.13 of the Annex to Regulation No 440/2008, according to which the determination of the bioconcentration factor by aqueous exposure for very poorly water soluble substances may not be technically feasible. In addition, for substances with very low water solubility in the aquatic environment, exposure via water may be of limited importance in comparison to the dietary route. According to the eighth paragraph, a partition coefficient above 5 and a water solubility below 10 to 100 μg/l mark the range of substances where testing via aqueous exposure may become increasingly difficult.

108.

Nevertheless, the rejection of the appellants’ submission by the General Court proves to be legitimate on other grounds.

109.

In particular, it is stated in the sixth paragraph of Section C.13 of the Annex to Regulation No 440/2008 that the aqueous exposure test may also be applied under certain conditions to substances with a partition coefficient above six. The eighth paragraph makes clear that it is not possible to give exact prescriptive guidance on the method to be used based on water solubility and the partition coefficient alone as other factors (analytical techniques, degradation, adsorption, etc.) can have a marked influence on method applicability.

110.

Consequently, in the light of the possible difficulties mentioned in paragraph 107, the legislature nevertheless decided that in principle the aqueous exposure test method for determining the bioconcentration factor is preferable. It is only if this method cannot be implemented in practice that the assessment of bioaccumulation through diet takes on crucial importance.

111.

According to the opinions of the Member State Committee from 2015 and 2018, reliable studies on the bioconcentration of the substances at issue are available.

112.

It is true that by the fourth limb of the third part of the fourth ground of appeal, the appellants complain about the treatment of objections to the growth corrections applied to a CERI study from 2010 in paragraphs 231, 234 and 237 of the judgment under appeal. ( 49 ) By those corrections, account is taken of the effects of the growth of the test animals on the results. ( 50 )

113.

However, in paragraph 237, the General Court rightly rejected the objection to the CERI study as out of time because it was not raised in the application, as the General Court also explained in paragraphs 200 and 201 of the judgment under appeal. Under Article 84(1) of its Rules of Procedure, no new plea in law may be introduced in the course of proceedings unless it is based on matters of law or of fact which come to light in the course of the procedure.

114.

Furthermore, this complaint actually no longer concerns the legal characterisation of facts, but the assessment of the abovementioned study and of the objections raised. It is not therefore an admissible subject matter for the appeal.

115.

Furthermore, ECHA rightly explains that this (doubly inadmissible) objection relating to growth corrections does not demonstrate a manifest error of assessment, but merely a difference of scientific opinion on the importance of those corrections for the reliability of the results.

116.

Lastly, by the fifth limb of the third part of the fourth ground of appeal, the appellants assert that, in paragraph 241 of the judgment under appeal, the General Court wrongly found that there is nothing in the file to indicate the existence of studies refuting the determination of the bioconcentration factor. However, the General Court did not thereby state that there were no studies at all arguing against the use of the bioconcentration factor. The appellants have indisputably cited studies raising methodological objections to the studies relied on by ECHA. However, the contested finding of the General Court merely referred to the fact that the appellants did not cite any studies which concerned the determination of the bioconcentration factor for the same species.

117.

In the light of the rules governing the relevant test method for assessing bioaccumulation under Sections 1.1.2 and 1.2.2 of Annex XIII to the REACH Regulation and Section C.13 of the Annex to Regulation No 440/2008 in particular, the determination made by ECHA on the basis of the bioconcentration factor that the substances at issue are bioaccumulative and very bioaccumulative is thus not only free from manifest errors of assessment, but is even more consistent with the legal criteria to be applied than the view taken by the appellants.

118.

It was not therefore necessary for ECHA additionally to make comments on dietary bioaccumulation (biomagnification and trophic magnification) in stating reasons for the contested decision. Consequently, the objections raised by the appellants concerning these supplementary reasons are no longer material.

119.

The conclusion reached by the General Court is therefore correct, even though its statement of reasons is not entirely convincing in every respect.

120.

The objections to the legal characterisation of the properties of the substances at issue in respect of bioaccumulation are also therefore unfounded.

121.

By the first part of the fourth ground of appeal, the appellants object, lastly, to paragraph 256 and paragraphs 270 to 272 of the judgment under appeal, where the General Court accepted that the identification of toxic properties of D5 and D6 is based solely on the presence of D4, whose toxic properties are established, as an impurity in those substances.

122.

This finding is based on the fifth paragraph in the preamble to Annex XIII to the REACH Regulation, according to which the PBT/vPvB properties of relevant constituents of a substance are also to be taken into consideration.

123.

While the appellants concede that account must be taken of the presence of D4 as an impurity in the assessment of PBT properties of D5 and D6 as a constituent, they criticise the General Court for accepting, in paragraph 256, the failure to carry out an assessment of the toxic properties of both substances as such.

124.

The General Court also gives convincing reasons for its position in paragraph 256, however, stating that an assessment of the toxicity of D5 and D6 as such is not necessary if that property is already apparent from the presence of D4 as an impurity and the data are not complete in other respects. Furthermore, in paragraph 257 of the judgment under appeal, which is not contested, the General Court rightly explains that it would not be compatible with the objective of a high level of protection to forgo the assessment of toxicity in such a situation.

125.

In addition, the appellants object to the fact that, in paragraph 271, the General Court rejected as unconvincing their argument that D4 would be released as an impurity in D5 and D6 only in such small quantities that it would not reach a toxic level. On the basis of the statements regarding bioaccumulation, however, the General Court argues convincingly that even D4 released in small quantities could accumulate (in organisms) to such an extent that it has toxic effects.

126.

The first part of the fourth ground of appeal is therefore also unfounded and the appeal should be dismissed in its entirety.

127.

Under Article 184(2) of the Rules of Procedure, where the appeal is unfounded, the Court is to make a decision as to the costs.

128.

Under Article 138(1) of the Rules of Procedure, which is applicable to appeal proceedings by virtue of Article 184(1) thereof, the unsuccessful party is to be ordered to pay the costs if they have been applied for in the successful party’s pleadings.

129.

Since the appellants have been unsuccessful and ECHA has applied for costs, the appellants must be ordered to pay those costs.

130.

Germany and the Commission, interveners at first instance, should be ordered to bear their own costs pursuant to Article 140(1) of the Rules of Procedure. ( 51 )

131.

Furthermore, under Article 184(4) of the Rules of Procedure, where the appeal has not been brought by an intervener at first instance, but it participated in the written or oral part of the proceedings before the Court of Justice, the Court may decide that it is to bear its own costs. Since ACC took part in the proceedings and has been unsuccessful, it should be ordered to bear its own costs.

132.

I therefore propose that the Court should: