Draft standard considered by the Codex Alimentarius Commission
in first reading and now referred to Governments for detailed comments.
SAMPLING
Text prepared by the former “European Council of the Codex Alimentarius”, now the “Advisory Group for Europe of the Joint FAO/WHO Codex Alimentarius Commission”.
Introduction
§ 1
When taking samples from foodstuffs on the market or intended for the market to ensure that they are of satisfactory quality it must always be borne in mind that the analyst can only issue a reliable and useful report if the material supplied for testing is suitable to produce the required results with the testing methods to be used. In the case of food samples care must be taken not only to secure the quantity required to test the food concerned by the normal methods but also to ensure that the sample is packed, labelled and delivered to the analyst in such a way that it cannot be confused and that its condition does not change prior to delivery. Samples of easily perishable goods must reach the analyst as quickly as possible and should be tested by him as soon as possible, even if he has means available to prevent them from perishing further. The rules given below also apply where appropriate to samples of food additives and of utensils and packaging materials for foodstuffs.
Control sample
§ 2
A control sample is a sample which, at the request of the person with authority over the goods, is left behind after a sample has been removed, and which is taken in the same way, in the same quantity and with the same precautions as the sample which is removed, so that it has the same test possibilities.
Packaging of samples
§ 3
The sample removed and where applicable the control sample must be appropriately packaged immediately. A packing is appropriate if it is suitable to prevent any change in the sample until it is delivered to the testing centre. However the selection of the type of packing or the type of container for the sample depends not only on the goods from which the sample was taken but also-insofar as the purpose of the test is known to the sampler-on the type of test which is to be carried out. For example, it is normally essential to pack a sample for bacteriological tests in a sterile container or in sterile packing material. (See also paragraphs 19 and 20.)
§ 4
It is generally sufficient to pack materials in grain or powder form in paper bags, provided they are not hygroscopic. It is recommended to use receptacles of glass, glazed earthenware, etc. for fats, fatty or damp materials and hygroscopic products (meat etc.). Samples of liquid foodstuffs are generally put into bottles. Samples of pre-packed (pre-bottled) foodstuffs do not usually require any special wrapping but may be left in their original packets (containers). (See also paragraphs 7 and 18).
Sealing samples
§ 5
The sealing of sample and where applicable control sample wrappings is intended to prevent any deliberate alteration of the contents of the sealed packets. ‘Sealing’ covers not only the use of sealing wax, but also the use of other types of fastenings which cannot be opened without this fact being obvious. Thus lead deals with stamps, press buttons which can only be closed once and similar devices are also suitable for sealing samples and control samples.
§ 6
Sufficiently strong paper bags can be used for samples of all types of food as an outer wrapping to be fastened (sealed) as above. The use of sealed paper bags is particularly recommended where the fastening could only be attached with difficulty to or would not hold securely on the receptable containing the sample. When several samples are taken at the same time sealing may be effected by enclosing the individual containers of such samples in a collective container which is then sealed as described above.
Sample labelling
§ 7
Each sample removed and control sample left behind is to be labelled so that it cannot be confused. The simplest method is to write the description directly on to paper wrappings. Labels or tags on which the description may be written should be stuck or securely tied on containers on which it is impossible to write. In the case of pre-packed (pre-bottled) samples care should be taken to ensure that the description of the goods and other important details on the original package (container) are not obscured by the sample label.
Sending the samples to the test centre
§ 8
A sample must be sent to the respective test centre without delay and care must be taken to ensure that the purpose of the test is not endangered by the choice of an unsuitable method of transport, e.g. one which takes too long or in which the sample container might be broken.
Covering letter for the sample
§ 9
Each sample must be sent to the test centre with a covering letter which should as far as possible contain all the sampler's comments and finding for the attention of the analyst, e.g. time and reason for taking the sample; established description of goods, information on any steps taken to preserve the sample, obvious defects, results of any preliminary examination, origin of the goods and date of their purchase or production, type and length of storage and quantity of the stocks of the goods to be tested still remaining when the sample was taken. It is also recommended that the covering letter should specify whether a control sample was left behind.
Procedure for taking samples from large stocks of non-homogeneous substances
§ 10
When an opinion is to be given on a relatively large stock of goods which is probably not homogeneous (e.g. because it tends to separate) samples must be taken from various different storage places (or containers). If there is reason to suspect that certain parts of the stock are in such a condition that they should not be mixed with other quantities of the goods in stock, these parts must be disregarded in obtaining an average sample from the total stock. Separate samples must therefore be taken from them.
§ 11
To obtain a sample which enables a conclusion to be reached on the average quality of the total stock-or of a part of it suitable for mixingcare must be taken to ensure that the sample volume, which is usually negligible in comparison with the volume of the goods to be inspected, is of the same quality as would be obtained if the total quantity of goods were thoroughly mixed before the sample was taken. The greater the stock of goods to be tested for its average quality, the larger the number of places within the individual containers or the greater the number of individual containers from which samples must be taken; these must then be thoroughly mixed and the correct volume for the sample to be sent to the test centre must be taken from the resulting mixture. Obvious impurities in the goods may only be removed before sampling if it is normal trade practice to remove them before passing the goods on to the purchaser.
§ 12
However taking an average sample from large quantities of goods is only justifiable if there is no doubt from the way and the minimum quantities in which the goods concerned are sold to the customers of the business being inspected that each purchaser either receives properly mixed goods of average quality or is supplied with sufficient quantities of varying quality to enable him to mix them himself in accordance with normal trade practice and thus obtain this average quality before passing them on to his customers.
§ 13
When taking samples of foodstuffs stored in sacks, barrels, crates or other large containers (including goods in large pieces such as potatoes, fruit, etc.) it will sometimes be necessary to empty the containers and consider all the layers to obtain a true average sample. In the case of large stocks of foodstuffs stored in this way but normally presumed to be homogeneous it is sufficient to take the average sample from 1 container if the total number of container does not exceed 5, from 10% of containers with a minimum of 5 if they do not exceed 100, from 5% with a minimum of 10 if they do not exceed 500, from 3% with a minimum of 25 if they do not exceed 2000 and from 1% with a minimum of 60 if they exceed 2000. A minimum of 5 samples must be taken from different places from truck loads and large heaps of goods, and care must be taken that pieces of different sizes are taken proportionately from the various layers of the stocks.
§ 14
The constituents are never equally distributed in granular substances (e.g. grain). The smaller or heavier particles such as earth, sand, small seeds from weeds, etc. usually fall to the bottom of the container or sack. Therefore the different layers must be taken into account to obtain an average sample. The contents of small drawers or chests should for practical purposes be emptied on to a pile and the samples taken from this after it has been thoroughly mixed.
§ 15
The contents of sacks are sampled by taking a part sample from the top, middle and bottom of the sack. These part samples are well mixed together to give an average sample. For larger quantities of goods in sacks the number of sacks to be sampled is determined according to the pattern given in § 13. A sample which comes from one place only is called a random sample and does not enable any conclusion to be drawn regarding the quality of a large volume of goods.
§ 16
Liquids which separate must be carefully mixed before the sample is taken. Liquids in small barrels are best mixed by rolling the cask. The contents of tins must be mixed by shaking up and down before the sample is taken. Small quantities of liquid are best mixed by being poured several times from one container to another or by stirring and shaking. A sample of liquids which do not separate should where possible be taken from the centre of the container with a siphon or pipe. Partly or completely frozen, crystallised or congealed liquids must be fully liquefied and thoroughly mixed as described above before the sample is taken. The procedure described for sampling liquids should also be adopted when the merchandise to be sampled is of an oily, viscous or unctuous consistency. However if such merchandise cannot be mixed by rolling or shaking the container and if it is also impossible to stir it like liquids, then it must be throughly mixed with a spade-like implement before samples are taken.
§ 17
Obviously the covering letter to be sent to the test centre with various samples from different parts of a large stock of merchandise or with an average sample from a stock of merchandise presumed not to be homogeneous should, in addition to the usual details, also contain special details of any particular observations on the differences in quality between various parts of the stock or any other details likely to be useful to the analyst.
Procedure for taking samples from merchandise which is pre-packed or filled into containers
§ 18
Merchandise which is pre-packed or filled into containers must be sampled so that in the case of smaller packets or containers at least one full packet or container, and where necessary more, is taken as a sample. When samples are taken from large packets or large containers the contents of these large packets or containers must where necessary be thoroughly mixed beforehand, if the merchandise in question has a tendency to separate and is not homogeneous. Naturally anyone who puts, or instructs his staff to put, merchandise into packets or containers (barrels, bottles, jars, etc.) for sale to his customers is responsible for seeing that each customer receives satisfactory merchandise and that therefore even in the case of merchandise which is not homogeneous and which tends to separate the amount in each packet or special container complies with the respective minimum standards when well mixed. If the purpose of sampling pre-packed foodstuffs is to ascertain whether an existing large stock of pre-packed merchandise is totally satisfactory the number of packets or containers to be taken as a sample must conform with the pattern given in § 13.
Samples for microbiological tests
§ 19
Sampling for microbiological tests must be conducted under conditions which prevent the sample from being infected with micro-organisms by the sampling process and during the period prior to the tests, and which if possible also check the growth of any micro-organisms which may have been on or in the sample when it was procured until the test is carried out. Thus it is only merchandise which is protected by its packing (e. g. sealed jars, bottles, tins, plastic bags, etc.) against infection by micro-organisms which present no special difficulties in sampling for microbiological tests. It is also possible to take samples for microbiological tests without special precautions from solid foods (e.g. sausages, bread, certain types of uncut cheese, etc.) the nature of which prevents micro-organisms from penetrating into their interior in the period between sampling and testing. However in this case it is necessary for the microbiological test to be carried out as quickly as possible after sampling.
§ 20
Special care must be taken in procuring samples for microbiological tests in all cases where samples of liquid or semi-solid merchandise (e.g. water, milk, wine, jam, ice cream, etc.) are to be taken from pipes or large containers and it is often difficult to remove the sample under sterile conditions. Such samples should be put into sterile bottles or jars of 100 to 150 ccm capacity and tested as soon as possible after being procured. If such samples have to undergo a long journey to the test laboratory they must be suitably refrigerated (with ice, dry ice, etc.) to check the further growth of any micro-organisms which might be present in the sample so that a reliable conclusion regarding the condition of the merchandise at the time of sampling may be drawn from the results of the test.
