Sampling shall be performed by an authorised person as specified by the Member States.

Each lot which is to be examined must be sampled separately.

In the course of sampling and preparation of the samples precautions must be taken to avoid any changes, which would affect the benzo(a)pyrene content, adversely affect the analytical determination or make the aggregate samples unrepresentative.

As far as possible incremental samples should be taken at various places distributed throughout the lot or sublot. Departure from this procedure must be recorded in the record.

The aggregate sample is made up by uniting all incremental samples. This aggregate sample is homogenised in the laboratory unless this is incompatible with implementation of point 3.6.

Replicate laboratory samples for enforcement, trade (defence) and referee purposes shall be taken from the homogenised aggregate sample unless this conflicts with Member States’ rules on sampling.

Each sample shall be placed in a clean, inert container offering adequate protection from contamination and against damage in transit. All necessary precautions shall be taken to avoid any change in composition of the sample, which might arise during transportation or storage.

Each sample taken for official use shall be sealed at the place of sampling and identified following the Member State’s rules.

A record must be kept of each sampling, permitting each lot to be identified unambiguously and giving the date and place of sampling together with any additional information likely to be of assistance to the analyst.

In the case of oils, for which a homogeneous distribution of benzo(a)pyrene can be assumed within a given lot, it is sufficient to take three incremental samples per lot to form the aggregate sample. Reference to the lot number shall be given. For olive oil and olive pomace oil further information on sampling is given in Commission Regulation (EC) No 1989/20034.

For other products, the minimum number of incremental samples to be taken from the lot shall be as given in Table 1. The incremental samples shall be of similar weight, no less than 100g each, resulting in an aggregate sample of no less than 300g (see point 3.5).

If the lot consists of individual packages, then the number of packages which shall be taken to form the aggregate sample is given in Table 2.

Sampling of foodstuffs at the retail stage should be done where possible in accordance with the above sampling provisions. Where this is not possible, other effective sampling procedures at retail stage can be used provided that they ensure sufficient representativeness for the sampled lot.

A number of the most commonly used definitions that the laboratory will be required to use are given below:

r

Repeatability, the value below which the absolute difference between two single test results obtained under repeatability conditions (i.e., same sample, same operator, same apparatus, same laboratory, and short interval of time) may be expected to lie within a specific probability (typically 95 %) and hence ${\mathrm{r\; =\; 2.8\; \times \; s}}_r$.

s_r

Standard deviation, calculated from results generated under repeatability conditions.

RSD_r

Relative standard deviation, calculated from results generated under repeatability conditions $\left[\left(s_r/\stackrel{-}{x}\right)\mathrm{\times \; 100}\right]$.

R

Reproducibility, the value below which the absolute difference between single test results obtained under reproducibility conditions (i.e., on identical material obtained by operators in different laboratories, using the standardised test method), may be expected to lie within a certain probability (typically 95 %); ${\mathrm{R\; =\; 2.8\; \times \; s}}_R$.

s_R

Standard deviation, calculated from results under reproducibility conditions.

RSD_R

Relative standard deviation calculated from results generated under reproducibility conditions $\left[\left(s_R/\stackrel{-}{x}\right)\mathrm{\times \; 100}\right]$, where $\stackrel{-}{x}$ is the average of results over all laboratories and samples.

HORRAT_r

the observed RSD_r divided by the RSD_r value estimated from the Horwitz equation (1) using the assumption r = 0.66R.

HORRAT_R

the observed RSD_R value divided by the RSD_R value calculated from the Horwitz equation.

U

the expanded uncertainty, using a coverage factor of 2 which gives a level of confidence of approximately 95 %.

Methods of analysis used for food control purposes must comply with points 1 and 2 of the Annex to Council Directive 85/591/EEC.

The analytical result is to be reported corrected or uncorrected for recovery. The manner of reporting and the level of recovery must be reported. The analytical result corrected for recovery is used for checking compliance (see Annex I, point 5).

The analyst should note the ‘European Commission Report on the relationship between analytical results, the measurement of uncertainty, recovery factors and the provisions in EU food legislation’ (2).

The analytical result has to be reported as x +/– U whereby x is the analytical result and U is the measurement uncertainty.

Laboratories must comply with Directive 93/99/EEC.

W. Horwitz, ‘Evaluation of Analytical Methods for Regulation of Foods and Drugs’, Anal. Chem., 1982, 54, 67A-76A.

(http://europa.eu.int/comm/food/food/chemicalsafety/contaminants/index_en.htm).

ISO/AOAC/IUPAC International Harmonised Protocol for Proficiency Testing of (Chemical) Analytical Laboratories, Edited by M. Thompson and R. Wood, Pure Appl. Chem., 1993, 65, 2123-2144 (Also published in J. AOAC International, 1993, 76, 926).

ISO/AOAC/IUPAC International Harmonised Guidelines for Internal Quality Control in Analytical Chemistry Laboratories, Edited by M. Thompson and R. Wood, Pure Appl. Chem., 1995, 67, 649-666.