The Annex to Regulation (EC) No 333/2007 is amended as follows:

1. (1)

   in point B.1.7 ‘Packaging and transmission of samples’, the following second paragraph is added:

   ‘In case of sampling for PAH analysis plastic containers shall be avoided if possible as they could alter the PAH content of the sample. Inert, PAH-free glass containers, adequately protecting the sample from light, shall be used wherever possible. Where this is practically impossible, at least direct contact of the sample with plastics shall be avoided, e.g. in case of solid samples by wrapping the sample in aluminium foil before placing it in the sampling container.’;

2. (2)

   points B.2 and B.3 are replaced by the following:

   ‘B.2.SAMPLING PLANS

   B.2.1.Division of lots into sublots

   Large lots shall be divided into sublots on condition that the sublot may be separated physically. For products traded in bulk consignments (e.g. cereals) Table 1 shall apply. For other products Table 2 shall apply. Taking into account that the weight of the lot is not always an exact multiple of the weight of the sublots, the weight of the sublot may exceed the mentioned weight by a maximum of 20 %.

   B.2.2.Number of incremental samples

   The aggregate sample shall be at least 1 kg or 1 litre except where it is not possible, e.g. when the sample consists of 1 package or unit.

   The minimum number of incremental samples to be taken from the lot or sublot shall be as given in Table 3.

   In the case of bulk liquid products the lot or sublot shall be thoroughly mixed in so far as possible and in so far it does not affect the quality of the product, by either manual or mechanical means immediately prior to sampling. In this case, a homogeneous distribution of contaminants is assumed within a given lot or sublot. It is therefore sufficient to take three incremental samples from a lot or sublot to form the aggregate sample.

   The incremental samples shall be of similar weight/volume. The weight/volume of an incremental sample shall be at least 100 grams or 100 millilitres, resulting in an aggregate sample of at least about 1 kg or 1 litre. Departure from this method shall be recorded in the record provided for under point B.1.8 of this Annex.

   Table 1Subdivision of lots into sublots for products traded in bulk consignments

   Lot weight (ton)	Weight or number of sublots
   ≥ 1 500	500 tonnes
   > 300 and < 1 500	3 sublots
   ≥ 100 and ≤ 300	100 tonnes
   < 100	—

   Table 2Subdivision of lots into sublots for other products

   Lot weight (ton)	Weight or number of sublots
   ≥ 15	15-30 tonnes
   < 15	—

   Table 3Minimum number of incremental samples to be taken from the lot or sublot

   Weight or volume of lot/sublot (in kg or litre)	Minimum number of incremental samples to be taken
   < 50	3
   ≥ 50 and ≤ 500	5
   > 500	10

   If the lot or sublot consists of individual packages or units, then the number of packages or units which shall be taken to form the aggregate sample is given in Table 4.

   Table 4Number of packages or units (incremental samples) which shall be taken to form the aggregate sample if the lot or sublot consists of individual packages or units

   Number of packages or units in the lot/sublot	Number of packages or units to be taken
   ≤ 25	at least 1 package or unit
   26-100	about 5 %, at least 2 packages or units
   > 100	about 5 %, at maximum 10 packages or units

   The maximum levels for inorganic tin apply to the contents of each can, but for practical reasons it is necessary to use an aggregate sampling approach. If the result of the test for an aggregate sample of cans is less than but close to the maximum level of inorganic tin and if it is suspected that individual cans might exceed the maximum level, then it might be necessary to conduct further investigations.

   Where it is not possible to carry out the method of sampling set out in this chapter because of the unacceptable commercial consequences (e.g. because of packaging forms, damage to the lot, etc.) or where it is practically impossible to apply the abovementioned method of sampling, an alternative method of sampling may be applied provided that it is sufficiently representative for the sampled lot or sublot and is fully documented.

   B.2.3.Specific provisions for the sampling of large fish arriving in large lots

   In case the lot or sublot to be sampled contains large fishes (individual fishes weighing more than about 1 kg) and the lot or sublot weighs more than 500 kg, the incremental sample shall consist of the middle part of the fish. Each incremental sample shall weigh at least 100 g.

   B.3.SAMPLING AT RETAIL STAGE

   Sampling of foodstuffs at retail stage shall be done where possible in accordance with the sampling provisions set out in point B.2.2 of this Annex.

   Where it is not possible to carry out the method of sampling set out in point B.2.2 because of the unacceptable commercial consequences (e.g. because of packaging forms, damage to the lot, etc.) or where it is practically impossible to apply the abovementioned method of sampling, an alternative method of sampling may be applied provided that it is sufficiently representative for the sampled lot or sublot and is fully documented.’;

3. (3)

   in the first paragraph of point C.1 ‘Laboratory Quality Standards’, footnote 1 is deleted;

4. (4)

   in point C.2.2.1 ‘Specific procedures for lead, cadmium, mercury and inorganic tin’, the second paragraph is replaced by the following:

   ‘There are many satisfactory specific sample preparation procedures which may be used for the products under consideration. For those aspects not specifically covered by this Regulation, the CEN Standard “Foodstuffs - Determination of trace elements – Performance criteria, general considerations and sample preparation”(¹) has been found to be satisfactory but other sample preparation methods may be equally valid.’;

5. (5)

   point C.2.2.2 is replaced by the following:

   ‘C.2.2.2.Specific procedures for polycyclic aromatic hydrocarbons

   The analyst shall ensure that samples do not become contaminated during sample preparation. Containers shall be rinsed with high purity acetone or hexane before use to minimise the risk of contamination. Wherever possible, apparatus and equipment coming into contact with the sample shall be made of inert materials such as aluminium, glass or polished stainless steel. Plastics such as polypropylene or PTFE shall be avoided because the analytes can adsorb onto these materials.’;

6. (6)

   point C.3.1 ‘Definitions’ is amended as follows:

   1. (a)

      the definition for ‘HORRAT_r’ is replaced by the following:

      • ‘HORRAT7_r
        The observed RSD_r divided by the RSD_r value estimated from the (modified) Horwitz equation8 (cf. point C.3.3.1 (“Notes to the performance criteria”)) using the assumption r = 0,66 R.’

   2. (b)

      the definition for ‘HORRAT_R’ is replaced by the following:

      • ‘HORRAT9_R
        The observed RSD_R divided by the RSD_R value estimated from the (modified) Horwitz equation10 (cf. point C.3.3.1 (“Notes to the performance criteria”)).’

   3. (c)

      the definition for ‘u’ is replaced by the following:

      • ‘u
        Combined standard measurement uncertainty obtained using the individual standard measurement uncertainties associated with the input quantities in a measurement model11

7. (7)

   point C.3.2 is replaced by the following:

   ‘C.3.2General requirements

   Methods of analysis used for food control purposes shall comply with the provisions of Annex III to Regulation (EC) No 882/2004.

   Methods for analysis for total tin are appropriate for official control on inorganic tin levels.

   For the analysis of lead in wine, the methods and rules established by the OIV12 apply in accordance with Article 31 of Council Regulation (EC) No 479/200813.’;

8. (8)

   point C.3.3.1 is replaced by the following:

   ‘C.3.3.1.Performance criteria

   Where no specific methods for the determination of contaminants in foodstuffs are prescribed at European Union level, laboratories may select any validated method of analysis for the respective matrix provided that the selected method meets the specific performance criteria set out in Tables 5, 6 and 7.

   It is recommended that fully validated methods (i.e. methods validated by collaborative trial for the respective matrix) are used where appropriate and available. Other suitable validated methods (e.g. in-house validated methods for the respective matrix) may also be used provided that they fulfil the performance criteria set out in Tables 5, 6 and 7.

   Where possible, the validation of in-house validated methods shall include a certified reference material.

   1. (a)

      Performance criteria for methods of analysis for lead, cadmium, mercury and inorganic tin:

      Table 5

      Parameter	Criterion
      Applicability	Foods specified in Regulation (EC) No 1881/2006
      Specificity	Free from matrix or spectral interferences
      Repeatability (RSDr)	HORRATr less than 2
      Reproducibility (RSDR)	HORRATR less than 2
      Recovery	The provisions of point D.1.2 apply
      	Inorganic tin	Lead, cadmium, mercury
      		ML is < 0,100 mg/kg	ML is ≥ 0,100 mg/kg
      LOD	≤ 5 mg/kg	≤ one fifth of the ML	≤ one tenth of the ML
      LOQ	≤ 10 mg/kg	≤ two fifths of the ML	≤ one fifth of the ML

   2. (b)

      Performance criteria for methods of analysis for 3-MCPD:

      Table 6

      Parameter	Criterion
      Applicability	Foods specified in Regulation (EC) No 1881/2006
      Specificity	Free from matrix or spectral interferences
      Field blanks	Less than LOD
      Repeatability (RSDr)	0,66 times RSDR as derived from (modified) Horwitz equation
      Reproducibility (RSDR)	as derived from (modified) Horwitz equation
      Recovery	75-110 %
      LOD	≤ 5 μg/kg (on dry matter basis)
      LOQ	≤ 10 μg/kg (on dry matter basis)

   3. (c)

      Performance criteria for methods of analysis for polycyclic aromatic hydrocarbons:

      The four polycyclic aromatic hydrocarbons to which these criteria apply are benzo(a)pyrene, benz(a)anthracene, benzo(b)fluoranthene and chrysene.

      Table 7

      Parameter	Criterion
      Applicability	Foods specified in Regulation (EC) No 1881/2006
      Specificity	Free from matrix or spectral interferences, verification of positive detection
      Repeatability (RSDr)	HORRATr less than 2
      Reproducibility (RSDR)	HORRATR less than 2
      Recovery	50-120 %
      LOD	≤ 0,30 μg/kg for each of the four substances
      LOQ	≤ 0,90 μg/kg for each of the four substances

   4. (d)

      Notes to the performance criteria:

      The Horwitz equation14 (for concentrations 1,2 × 10^–7 ≤ C ≤ 0,138) and the modified Horwitz equation15 (for concentrations C < 1,2 × 10^–7) are generalised precision equations which are independent of analyte and matrix but solely dependent on concentration for most routine methods of analysis.

      • Modified Horwitz equation for concentrations C < 1,2 × 10^–7:

        RSD_R = 22 %

        where:

        • RSD_R is the relative standard deviation calculated from results generated under reproducibility conditions [(s_R /) × 100]

        • C is the concentration ratio (i.e. 1 = 100 g/100 g, 0,001 = 1 000 mg/kg). The modified Horwitz equation applies to concentrations C < 1,2 × 10^–7.

      • Horwitz equation for concentrations 1,2 × 10^–7 ≤ C ≤ 0,138:

        RSD_R = 2C^(–0,15)

        where:

        • RSD_R is the relative standard deviation calculated from results generated under reproducibility conditions [(s_R /) × 100]

        • C is the concentration ratio (i.e. 1 = 100 g/100 g, 0,001 = 1 000 mg/kg). The Horwitz equation applies to concentrations 1,2 × 10^–7 ≤ C ≤ 0,138.’;

9. (9)

   point C.3.3.2 is replaced by the following:

   ‘C.3.3.2.“Fitness-for-purpose” approach

   For in-house validated methods, as an alternative a “fitness-for-purpose” approach16 may be used to assess their suitability for official control. Methods suitable for official control must produce results with a combined standard measurement uncertainty (u) less than the maximum standard measurement uncertainty calculated using the formula below:

   $\mathrm{Uf\; =}\sqrt{{\left(\mathrm{LOD/2}\right)}^2+{\left(\mathrm{\alpha C}\right)}^2}$

   where:

   • Uf is the maximum standard measurement uncertainty (μg/kg).

   • LOD is the limit of detection of the method (μg/kg). The LOD must meet the performance criteria set in point C.3.3.1 for the concentration of interest.

   • C is the concentration of interest (μg/kg);

   • α is a numeric factor to be used depending on the value of C. The values to be used are given in Table 8.

   Table 8Numeric values to be used for α as constant in formula set out in this point, depending on the concentration of interest

   C (μg/kg)	α
   ≤ 50	0,2
   51-500	0,18
   501-1 000	0,15
   1 001-10 000	0,12
   > 10 000	0,1

   The analyst shall note the “Report on the relationship between analytical results, measurement uncertainty, recovery factors and the provisions of EU food and feed legislation”17.’;

10. (10)

    in point D.1.2 ‘Recovery calculations’, the second paragraph is replaced by the following:

    ‘In case no extraction step is applied in the analytical method (e.g. in case of metals), the result may be reported uncorrected for recovery if evidence is provided by ideally making use of suitable certified reference material that the certified concentration allowing for the measurement uncertainty is achieved (i.e. high accuracy of the measurement), and thus that the method is not biased. In case the result is reported uncorrected for recovery this shall be mentioned.’;

11. (11)

    in point D.1.3 ‘Measurement uncertainty’, the second paragraph is replaced by the following:

    ‘The analyst shall note the “Report on the relationship between analytical results, measurement uncertainty, recovery factors and the provisions of EU food and feed legislation”18.’.