International food safety leaders and experts on contaminants and toxins made significant progress in adopting and advancing most of the provisions that were discussed at the last meeting of the Codex Alimentarius Committee on Contaminants
International food safety leaders and experts on contaminants and toxins made significant progress in adopting and advancing most of the provisions that were discussed at the last meeting of the Codex Alimentarius Committee on Contaminants (and Toxins) in Foods (CCCF12), held mid-March in Utrecht (The Netherlands). This article reviews those decisions and describes the pending and new work issues which will be addressed in the next 12 months through various electronic working groups or directly at the next CCCF session. CCCF12 has opened the door for Codex Alimentarius to develop guidelines which may provide – subject to confirmation – a common future backbone to help global harmonization on the ways countries may assess migration of chemicals from food contact materials to foods.
The 12th session of the Codex Alimentarius Committee on Contaminants (and Toxins) in foods (CCCF) was held in Utrecht (The Netherlands) from 12 to 16 March 2018. CCCF is the Codex Alimentarius horizontal committee in charge of the update of the General Standard on Contaminants and Toxins in Foods. Therefore, it is one of the most important Codex committees – which sets tolerable levels for environmentally and industrially-sourced contaminants in foods as low as reasonably (i.e. technically and economically) achievable to ensure an acceptable protection of consumer health, while ensuring fair practices in international trade of such foods.
CCCF12 will take important decisions on regulatory limits for foods in international trade of lead, cadmium, methyl mercury, aflatoxins, ochratoxin A
CCCF12 agreed to suggest adoption of lowered maximum tolerable levels for lead in several food categories in the Codex Alimentarius General Standard in Contaminants and Toxins in Food (GSCTF) and, where appropriate, in the corresponding Codex Commodity Standard.
CCCF12 discussed the reasons why quinoa, a pseudo-cereal, was specifically excluded from the applicable MLs for lead and cadmium to cereals. CCCF12 concluded that a further background search would be needed and that there may be room to address the issue in the future (i) to include quinoa as a pseudo cereal currently covered under “other cereal grain” in the context of the on-going revisions of the Classification of Food and Feed (CAC/MISC (CXM) 4) and/or (ii) to develop specific MLs for lead and cadmium for quinoa.
As announced (see WFFR February 2018 issue), CCCF12 agreed to cut the applicable ML for lead in grape juices by 20%, down to 0.04 mg/kg.
CCCF12 took a very surprising decision in not following the recommendation of the intersession EWG to adopt a ML for lead of 0.08 mg/kg and even concluded to stop suggesting such a level (‘revoke the existing level and discontinue any further work’). It seems that the proposed ML would have had some significant impact on some specific types of processed tomatoes, especially those with a high natural soluble solids content (e.g. TSS = 28 – 38%). Given the fact that a ML of 0.05 mg/kg for lead is already established for fruiting vegetables (which includes fresh tomatoes), it could also be used to derive a ML for processed products, using the technique of concentration factors (leading to a higher endpoint ML for such processed products).
It is public knowledge that some large tomato producing areas intended for producing specific tomato concentrates – which are of high brand value for worldwide consumers and of strategic importance for the local economy and communities – may have been affected by decades of environmental contamination due to fraudulent management of domestic and industrial waste via illegal landfilling disposal. Although this fact was not discussed as such during the session, CCCF took a wise approach in leaving countries a degree of liberty in establishing their own limits for tomato concentrates, while indicating a path forward by pointing out the Codex ML for fresh tomatoes.
In summary, this (non)decision is a perfect illustrative example of pragmatism about “agreeing to not disagree” on an international standard. But the issue might be revisited in the future based on a better geographical representation when new occurrence data become available on more achievable levels. Future may tell.
The Committee did “cut the pear in two halves” and finally reached consensus on a half-way 35% reduction of the existing ML in suggesting to adopt a level of 0.4 mg/kg.
Despite an already established ML for lead in wines at 0.15 mg/kg established by the intergovernmental organization setting up standards for wines (i.e., the OIV) and noting that wider geographical occurrence data in some grape wines and liquor wines elaborated from fruits may be needed at Codex level, CCCF12 deferred the future ML discussion to the intersession electronic working group tasked to review lead limits in the GSCTF. CCCF12 further noted that the future ML would apply to wines produced only after the date of adoption (by the CAC) of the future Codex ML and therefore such a ML would not be applicable to any wine in current aging stages, which will be marketed after that future adoption date.
Food grade salt
CCCF12 reached a consensus to suggest cutting the current ML by a half, down to 1 mg/kg for final adoption. This level would also apply to standardized products conforming with the corresponding Codex commodity standard for food grade salt. However, it was noted that salt is (i) consumed in small daily quantities, (ii) a level of 1.5 mg/kg could have been protective and (iii) an exemption could have been granted for salt harvested from marshes. The Committee did not retain those comments, except for the exemption of salts from marches, and adopted the 1 mg/kg 50%-reduced ML for lead in all salts.
Fat spreads and blended spreads and edible fats and oils
As announced, the Committee agreed to a 60% reduction of the existing ML for lead in fat and blended spreads down to 0.04 mg/kg. For edible fats and oils, the Committee reached consensus on only a 20% reduction of the existing ML down to 0.08 mg/kg, although a more ambition reduction would have allowed 97% of current global products to be compliant with the new ML. However, it should be noted that the mitigation measures are difficult to put in place for some secular oils and the Committee took that aspect into account as well. Those levels are applicable to products conforming with the relevant Codex commodity standards covering fat spreads, blended spreads and edible fats and oils.
Fresh farmed mushrooms, i.e. common mushrooms (Agaricus bisporous), shiitake mushrooms (Lentinula edodes), and oyster mushrooms (Pleurotus)
CCCF12 agreed to move forward with suggesting the adoption of a new ML for lead in fresh farmed mushrooms at 0.3 mg/kg. This is less ambitious than the EWG recommended level of 0.2 mg/kg, but still allows 96% of global products to comply with the new ML, relevant to those mushrooms conforming with the corresponding Codex Commodity Standard. The ML applies to the whole commodity and not only the edible part. In this regard, this is a good achievement full of pragmatism.
Canned brassica vegetables
While agreeing that canned brassica vegetables were part of the vanned vegetables food category, CCCF12 agreed to establish a specific new ML for lead at 0.1 mg/kg, as planned and announced.
Fruit juices obtained exclusively from berries and other small fruits
CCCF12 also discussed and agreed to provide additional clarification to the scope of applicable existing ML for lead in this food category and whether it should be lowered. While confirming that the current ML of 0.05 mg/kg applies to the whole commodity (not the concentrated ones) or the commodity reconstituted to the original juice concentration, ready to drink (applicable also to nectars, ready to drink), CCCF12 agreed to suggest an adoption of a 20% reduced ML for lead to grape juices at 0.04 mg/kg, noting however that such a reduced level may cause a 15% negative impact on the global trade of grape juices. The current ML for the other types of fruit juices remained unchanged.
The Committee also agreed to continue working on further MLs in edible offal and some wines.
Cadmium in chocolates and other types of cocoa-based foods
CCCF12 had some interesting discussion about the various proposals for establishing maximum tolerable levels (MLs) for cadmium in chocolate products and other cocoa products. The Committee finally agreed to support the adoption of two MLs for cadmium in chocolates, taking into consideration the total cocoa solids content on a dry matter basis as a criterion to apply the two different ML.
For chocolates with such cocoa solids above or equal to 70%, CCCF12 suggests the adoption of a 0.90 mg/kg ML, as a compromise between the EWG suggested 1.00 and the request for a lower value (0.80) by some large importers.
For chocolates with cocoa solids content below 70% and above or equal to 50%, a ML of 0.80 mg/kg was set for adoption, although a lower 0.60 value would have led to only a potential 4.3% global trade impact. A note has been added to both MLs making clear that they would also apply to sweet chocolate, Gianduja chocolate, semi-bitter table chocolate, vermicelli chocolate/chocolate flakes, bitter table chocolate, all subject to Codex commodity standards.
For chocolates with a solid matter level lower than 30%, CCCF12 was not able to achieve a compromise between the EWG suggested ML of 0.40 mg/kg and the suggestions for much lower levels (0.10 and 0.20), i.e. between the interests of producing countries where cocoa fields are located on volcanic soils (rich in cadmium) and the objective to protect further vulnerable groups of population, particularly keen to eat these types of chocolates.
For chocolate and chocolate products containing or declaring an amount of total cocoa solids below 50%, but above or equal to 30% (on a dry matter basis), (too) many different ML were proposed, at values around the already suggested 0.50 mg/kg. Unable to find a middle-ground agreeable value, CCCF12 tasked a new EWG to continue working on this chocolate products type and assess whether it was feasible to merge two categories and derive only one ML for chocolate and chocolate products containing or declaring a total cocoa solids content below 50% (on a dry matter basis), taking into account the small dataset for this category versus datasets available for the chocolates below 30%.
For dry mixtures of cocoa and sugars that are sold for final consumption, the Committee was torn apart about whether the Committee should continue the work or not, as some of these products are regularly consumed by children. As a compromise, CCCF12 agreed to discontinue the work on dry mixtures of cocoa and sugars sold for final consumption (therefore leaving an international vacuum which may be filled by existing or upcoming – possibly even more trade disruptive – national limits), but to continue to work on cocoa powders with 100% total cocoa solids on a dry matter basis. They noted that consistency among all these kinds of chocolate and cocoa products is important and that such a level could allow countries to derive pertinent limit(s) for the dry mixtures of cocoa and sugars sold for final consumption.
Methyl mercury in fish including associated sampling plans
CCCF12 made some progress in adopting MLs for methyl mercury in all types of tunas (1.2 mg/kg); alfonsinos (1.5 mg/kg); marlins (1.7 mg/kg) and sharks (1.6 mg/kg) based on less than 5% rejection (rather than the P95 approach) and equally applicable to whole fresh fish or frozen fish (including those intended for further processing). However, the EU28, Norway and Switzerland were unhappy with these decisions and their 5% rejection basis and officially expressed their reservations about all of them, while indicating that they won’t change the applicable MLs in the EU and EEA region. A further note provides the precisions on screening (relation with total mercury content) and national legislation for vulnerable groups, as follows: “Countries or importers may decide to use their own screening when applying the ML for methylmercury in fish by analysing total mercury in fish. If the total mercury concentration is below or equal to the ML for methylmercury, no further testing is required and the sample is determined to be compliant with the ML. If the total mercury concentration is above the ML for methylmercury, follow-up testing shall be conducted to determine if the methylmercury concentration is above the ML. The ML also applies to fresh or frozen fish intended for further processing. Countries should consider developing nationally relevant consumer advice for women of childbearing age and young children to supplement the ML.”
For Amberjack, the Committee recalled its policy about the target ML of 0.3 mg/kg and given the fact that occurrence data showed that both the average and median concentrations of total mercury and methylmercury were below that level, the Committee decided to discontinue the work on the ML for Amberjack. In doing so, the Committee also confirmed its policy that contaminant regulations and standards shall focus on those food categories which contribute the most to the exposure.
The Committee did not reach any consensus on how to make progress on a ML for swordfish and simply decided to discontinue and stop working on any ML. This decision was surprising, especially because the Committee noted that the methylmercury concentrations in swordfish were high, which was a health concern when consuming this fish. But at the same time, many delegations were unhappy with the ML itself and considered it as too high. Other delegations pointed out the role of selenium as possible remedy to the presence of methyl mercury, especially present in swordfish flesh. The reality is probably far more complex than this, with (i) developing countries highly dependent of catches of swordfish and (ii) the reluctance of the Committee to restart past discussion on Guideline Levels for special species of swordfish versus MLs, as the application of Codex GLs by official control authorities has been an issue in the past. There too, in the absence of any international standard (i.e. ML), the regulation of methyl mercury for swordfish is therefore entirely left to the appreciation of national and/or regional competent authorities, which may develop their own sanitary measures, in due line with their rights (and obligations) to do so under the WTO/SPS Agreement, and necessarily with some level of impact to trade in the absence of Codex MLs.
Beyond the MLs discussion, the Committee agreed on an amended sampling plan for methylmercury contamination in fish, which is now to be reviewed by the next CCMAS meeting (May 2018) for a possible endorsement.
Finally, the Committee agreed to establish a new EWG, chaired by New Zealand and co-chaired by Canada, to prepare a discussion paper on the establishment of other MLs in additional fish species, yet to be identified.
Total aflatoxin limit in ready-to-eat peanuts and associated sampling plan
CCCF12 reviewed the recommendation for an ML of 10 micrograms per kilogram (μg/kg or ppb) total aflatoxins in ready-to-eat peanuts for the final consumer. As announced, the debate raged for hours based on arguments that have been used and raised for many years already. The Committee oscillated between those countries in favour of a lower level than the one suggested by the EWG at 10 μg/kg (based on exposure data) and a higher level than 10 μg/kg (to protect some supplying countries’ interests). Instead of completely phasing out the objective to develop a Codex ML, the Committee resolved all these inextricable expressed positions by suggesting that more occurrence data were needed specifically on (i) ready-to-eat peanuts (not intended for further oil processing) which are subject to international trade (i.e. not to domestic consumption from local producers).
Such a call for new data will therefore only be issued by JECFA as of three-years from now to grant an additional “grace” period for the main producing and exporting countries to implement the 2004 Codex Code of Practice for the prevention and reduction of aflatoxins in ready-to-eat peanuts. Indeed, it was pointed out that some confusion exists on some current occurrence data about peanuts intended for further processing (edible oil production). Thus, the Committee decided to keep, but put on hold, the draft ML for total aflatoxins at 10 μg/kg in ready-to-eat peanuts, so that discussion will start again when JECFA reports on the new occurrence data collected in 4 or 5 years from now.
In other terms, opponents to such an international standard have bought extra time to avoid such an international ML, which then leaves the door fully open to national or regional competent authorities to set their own national limit due to this patent lack of global harmonization due to multiple diverging interests.
Total Aflatoxins and Ochratoxin A (OTA) limits in culinary herbs and spices
Same symptoms, same remedy. In view of the diverging positions between countries and business operators on the various MLs, and despite some recommendations to make progress on OTA from the dedicated past intersession EWG, CCCF12 decided to hold at their current step the draft proposed MLs (i.e. total aflatoxins at 20 or 30 micrograms per kilogram (μg/kg or ppb) in Nutmeg, Chili and Paprika, Ginger, Pepper and Turmeric and Ochratoxin A at 20 μg/kg). Therefore, JECFA is now tasked to issue a call for new occurrence data in three years from now and then the Committee may start to discuss the MLs again. Buying time is of great value to avoid adopting an international standard (i.e. ML) that nobody can agree upon.
As some Codexians would say: 'Better off with no norm rather than with a bad norm'. Maybe... Because, meanwhile, a lack of global harmonization prevails (i.e. national diverging limits coexist).
It should also be noted that this is to allow time for the Codex Committee on Spices and Culinary Herbs (CCSCH) to make progress on commodity standards on some of the above-mentioned categories, and maybe in that context, CCSCH may come up with its own views of possible MLs, which in any case must be subject to endorsement by the CCCF.
Codes of practice for the reduction of the presence of industrial or environmental contaminants
Dioxins, PCB-like dioxins (DL-PCB), non-PCB dioxin like substances (NDL-PCB)
CCCF12 went through a revised version of the draft Code of Practice text which was presented as the outcome of the EWG which worked on it in the past 12 months. Given that most of the recent amendments to the text were editorial and did not substantively change the core aspects of the Code, the Committee agreed to suggest its final adoption by the forthcoming Codex Alimentarius Commission (CAC41), which will be held this July. The final agreed text is to be found in Appendix V of the CCCF12 report.
It will be subject to a circular letter that will be issued soon in the context of the preparation of CAC41 and seeking comments on the adoption of the final text. The deadline is generally being set on May 15.
3-MCPDE and Glycidyl esters in refined vegetable oils and their products thereof, with special focus on infant formulas
CCCF12 had a fair, plain and consensual discussion on this Code of Practice. Unlike the Code on PCBs, the Committee considered it required more work, especially because parts of the Code still include texts in square brackets (i.e. on specific practices on matters such as low lipase activity, irrigation water, polar solvents, degumming, bleaching clay and two-stage deodorization, or the inclusion of specific references to fish oils). Therefore, the Committee decided to move the proposed draft Code of Practice for adoption at Step 5 (out of 8) (thus becoming such a draft Code and with a WTO/TBT value) by the next CAC41.
The Title was changed to make sure the Code would not only cover vegetable oils, and a new paragraph to the introduction to further clarify that revised title. Thus, CCCF12 established a new EWG to revise the draft Code of Practice based on the comments and information to be submitted by Codex members and observers in response to a specific circular letter, after CAC41. The EWG is tasked to resolve all outstanding issues for possible final review at the next CCCF session in 2019 (CCCF13). Before CAC41, a specific other circular letter is expected to seek comments to support the adoption at step 5 and advancing to step 6. The text of the draft Code of Practice is already available in Appendix VI of the CCCF12 report4.
Proposed draft guidelines to define the risk analysis principles about the inadvertent low level presence of chemicals in foods
CCCF12 had to review a new version of the proposed draft guidelines, instead of the one which was published as the result of the EWG. Indeed, it seems that a handful number of delegations have had an informal meeting prior to the CCCF12 session and came up with revised version delivered to the CCCF12 plenary, as a conference room document. Many improvements were brought to the text (e.g. revised title to avoid different interpretations of “emerging contaminants” or “inadvertently present”; shorter introduction; clearer scope (i.e. contaminants outside the normal regulatory framework); changes to definitions; new section on and to clarify how cut-off values should be derived with one illustrative example).
The discussion highlighted the need to clarify the scope even further to address anything else than “classical” residues or permitted addition of chemical substances to food (so food additives, pesticide residues, residues of veterinary drugs will be specifically excluded). The whole approach on how guidance levels (alert levels) are derived from values obtained by using the threshold of toxicological concern (TTC) will need to be further worked out and its applicability reassessed for the various cases considered on a case-by-case basis.
Despite the remaining work on the guidelines, the Committee considered that the text was ready for adoption at Step 5 (out of 8) by this July CAC41. The text as amended is available as appendix IX to the CCCF12 report4.
CCCF12 therefore established a new EWG to further develop the guidelines with a focus on the texts in square brackets for consideration at the next session. Although not formally established by CCCF12, CCCF12 leaves the door open to the chairs from New Zealand and the Netherlands to organize a physical working group (PWG) to meet immediately prior to CCCF13 session to make further recommendations to the Committee, based on the text to be included in the EWG report.
It is worth noting that among the case study examples listed in Annex 3 of the draft guidelines, a clear reference and link is made to the “Occurrence and Risk Characterization of Migration of Packaging Chemicals in New Zealand Foods”, published by New Zealand Ministry for Primary Industries. Therefore, it seems very clear that the intent is to provide to governments Codex guidelines to implement national policies in a harmonized fashion how they may regulate migration of those substances viewed as contaminant (i.e. non-intentionally added substances) only, from the complex chemical-based food contact material matrices into foods. This move of Codex CCCF in the food contact material area is worth noting as quite significant and somewhat historical.
Discussion papers on possible future work
Other Lead limits
The Committee agreed with the principle of three different priority groups for certain food categories (high, intermediate, low). CCCF12 established a new EWG to prepare a revised discussion paper and project document, also taking into consideration exposure data (in addition to other criteria for prioritization of commodities) in establishing the prioritization categories for MLs, and to propose, if feasible, MLs for the categories indicated with a focus on commodities identified as high in the priority list. In this context, it should be carefully noted that some of the comments were raised during the session to move the confectionery product category to that high priority group.
Certainly, an issue to be monitored during the EWG work, because not all confectionery products are intended to be marketed to children, nor are primarily consumed by children, and thus it would make no sense to move them to the high priority group at all, rather the opposite. Indeed, because Lead may not be present (even detectable) in many confectionery products, it would rather be more adequate to move that food category to the lowest priority list. But as always, future may tell.
The Committee also noted that JECFA would issue a call for occurrence data for lead in several categories listed in Table 4 of the working document. It encouraged Codex member countries to submit data on national food consumption levels and on lead occurrence data to WHO GEMS/Foods to assist in the development of the discussion paper and a possible decision on any new work (new MLs) for lead in the categories identified in the high priority group.
Aflatoxins and sterigmatocystin in cereals
CCCF12 agreed that this issue was important and therefore assigned to a new EWG the task to update the discussion paper with proposals for new work and project documents.
The aim is to suggest specific MLs for total aflatoxins in (a) wheat, maize, sorghum and rice (specifying the categories) for grains for human consumption and (b) in flour and cereal-based foods for infants and young children. Regarding the issue of the method of analysis to identify and quantify the presence of sterigmatocystin in cereals, CCCF12 encouraged standard development organizations (such as AOAC, ISO, AOACC, etc.) to develop a duly validated method and submit it for consideration by CCMAS.
Regarding the specific Annex on sterigmatocystin, CCCF12 agreed that no action was required, at this point in time.
Prevention and reduction of cadmium in cocoa (beans)
CCCF12 also considered that a specific and new Code of Practice for the prevention and reduction of cadmium in cocoa would be useful. It tasked a new EWG to further elaborate the discussion paper to (a) determine whether mitigation measures available at present would support the development of a Code of Practice, noting that such measures should be proven to be cost-effective, applicable worldwide by both large and small scale business operators, including farmers, and (b) identify the scope of such a code of practice (e.g. whether it would cover the whole production chain or only the primary production step), based on a survey.
Should those pre-conditions be met in time, then the EWG is to include a project document to define the scope and timeline of the new work and possibly append a first proposed draft of such a Code of Practice.
Hydrocyanic acid and mycotoxins in cassava and products thereof
The discussion was postponed to 2019 CCCF13 for an administrative reason.
CCCF12 Priorities for future work of JECFA on contaminants
Based on the recommendations made by an in-session working group, CCCF12 reaffirmed the need for JECFA scientific evaluations/review of (i) dioxins, (ii) arsenic (inorganic), (iii) scopoletin (in relation to Noni Juice consumption), (iv) ergot alkaloids, (v) ciguatoxins, and (vi) an updated risk assessment, including exposure assessment, of mycotoxins trichothecenes (T2 and HT2). The full details are included in Appendix X of the CCCF12 report4.
Forward Work Plan For The Committee On Contaminants In Foods
CCCF12 tasked the Codex Alimentarius Commission’s, JECFA’s and the Host Country’s secretariats, with the assistance of the EU, to continue the work on the discussion paper to focus on whether the Committee has now covered all main staple foods subject to international trade and of public health concern.
Proposal for the development of general guidance on data analysis for ML development
Based on the request of JECFA to develop a general guidance on data analysis for ML development to avoid different approaches applied by the various EWGs, e.g. on the handling of occurrence data without information on limit of quantification of method used and therefore established a new EWG to prepare such a discussion paper for further consideration by CCCF13 in 2019.
Revision of the Code of Practice for the Prevention and Reduction of Lead Contamination in Foods
Upon a proposal from the USA, CCCF12 agreed to establish a new EWG to prepare a discussion paper including a project document for a proposal for new work on the revision of the existing Code of Practice for consideration by CCCF13 in 2019.
The aim of such revision would be to add new in- formation on lead reduction strategies in the areas of (i) agricultural production (e.g. techniques to address lead contamination in soil and water) and (ii) food processing (e.g. filtration aids for juice manufacture, measures to reduce lead in foods during cooking, and minimizing introduction of lead from food processing equipment).
More details on CCCF12 working documents, CRDs are available on the webpage of the session at http://www.fao.org/fao-who-codexalimentarius/meetings/detail/en/?meeting=CCCF&session=12
CCCF12 full official report is to be available at http://www.fao.org/fao-who-codexalimentarius/meetings/en/
 CCCF does not regulate any specific or overall migration limit(s) for any chemical used as technical ingredient in food contact materials (themselves largely unregulated at the Codex alimentarius level), but this may change at one point in time in the future.
 See latest version of the GSCTF (i.e. CODEX STAN 193) at http://www.fao.org/fao-who-codexalimentarius/sh-proxy/en/?lnk=1&url=https%253A%252F%252Fworkspace.fao.org%252Fsites%252Fcodex%252FStandards%252FCODEX%2BSTAN%2B193-1995%252FCXS_193e.pdf
 CCMAS is the Codex alimentarius Committee on Methods of Analysis and Sampling. Next CCMAS webpage is available at http://www.fao.org/fao-who-codexalimentarius/meetings/detail/en/?meeting=CCMAS&session=39
 See CCCF12 CRD26 at http://www.fao.org/fao-who-codexalimentarius/sh-proxy/en/?lnk=1&url=https%253A%252F%252Fworkspace.fao.org%252Fsites%252Fcodex%252FMeetings%252FCX-735-12%252FCRDs%252Fcf12_CRD26x.pdf
 See CX/CF 18/12/14 at http://www.fao.org/fao-who-codexalimentarius/sh-proxy/en/?lnk=1&url=https%253A%252F%252Fworkspace.fao.org%252Fsites%252Fcodex%252FMeetings%252FCX-735-12%252FWD%252Fcf12_14e.pdf