4.10 Glyphosate (158)

RESIDUE AND ANALYTICAL ASPECTS

Glyphosate is a herbicide with uses on many crops. Glyphosate has been evaluated several times with the initial evaluation in 1986 and the latest in 1997. It was listed under the Periodic Re-evaluation Programme of the 34^th Session of the CCPR for residue review by 2005 JMPR (ALINORM 03/24). The Meeting received information on glyphosate metabolism and environmental fate, methods of residue analysis, freezer storage stability, national registered use patterns, supervised residue trials on conventional and glyphosate tolerant crops and national MRLs.

The 2004 JMPR concluded that the metabolite AMPA is of no greater toxicological concern than the parent glyphosate and established an ADI for the sum of glyphosate and AMPA of 0-1 mg/kg bw. The same meeting considered an ARfD unnecessary.

Some information on GAP and national MRLs were submitted by Australia and The Netherlands.

The following abbreviations are used for the metabolites discussed below:

Glyphosate is available in a variety of different salt forms including as the sodium, potassium, ammonium and isopropylamine salts. To assist uniform interpretation of GAP application rates have been expressed in terms of glyphosate acid equivalents (ae). Applications of glyphosate can be made at different stages of crop growth. The following abbreviations are used for the main stages of application:

PRE = pre-emergent or pre crop emergence

EPO = early post-emergence

LPO = late post-emergence

PH = pre-harvest, specifically within a few weeks prior to harvest

Animal metabolism

The Meeting received metabolism studies for glyphosate in rats, lactating goats and laying hens.

The biotransformation and degradation pathways in the goat and hen are similar to those established in rat metabolism. In animals, ¹⁴C-labelled glyphosate is excreted unchanged. The only residue identified in tissues of goats and laying hens was glyphosate although there were indications that small amounts of AMPA may be formed. A significant proportion of the ¹⁴C in the goat and hen metabolism experiments was retained on columns used for separation and characterization. The possible identity of the retained ¹⁴C activity was not explored. Metabolism studies on rats samples were analysed both with and without clean-up on cation and anion columns detecting only unchanged glyphosate and small amounts of AMPA. Experiments on rats determined that no AMPA was formed following intravenous administration suggesting that microbial degradation on oral administration may be responsible for the minor amounts of AMPA detected.

Plant metabolism

The Meeting received plant metabolism studies for glyphosate on coffee, corn, cotton, soya beans, wheat, pasture grasses and alfalfa as well as on the glyphosate tolerant crops cotton, soya beans and sugar beet. Pre-emergent application of [¹⁴C]glyphosate at application rates equivalent to 4.5 kg ae/ha resulted in low levels of ¹⁴C in plants collected 4-8 weeks after application. Control plants also contained ¹⁴C probably from incorporation of ¹⁴CO₂ liberated during soil microbial degradation of [¹⁴C]glyphosate.

For the same crops grown hydroponically and exposed to sublethal doses of [¹⁴C]glyphosate in the growth solution, glyphosate was the major component of the total radioactive residue (TRR) in the aerial parts of the plants (21-69%). Other compounds identified were AMPA (4.2-28%), N-methyl AMPA (0-2.0%) as well as small amounts of natural products. The proportion of ¹⁴C extracted with water was higher for aerial parts (70-90%) compared with roots (36-87%). In the case of roots, glyphosate was the major compound detected (7.6-57%) together with smaller amounts of AMPA (2.8-7.4%), N-methyl AMPA (0-0.4%) and natural products (1-11%). In summary, several minor metabolites were present at < 2% TRR (N-methyl AMPA, methylphosphonic acid and N-methyl glyphosate) though their origin was unclear; in vivo metabolism, microbial degradation or impurities in the material administered.

The metabolism of [¹⁴C]glyphosate on both immature and mature coffee plants following uptake from soil, from hydroponic solution, stem injection and foliar application was studied. ¹⁴C was translocated from the sites of application. In all cases glyphosate was the major component of the ¹⁴C residue. For example 5 weeks after foliar application to mature coffee plants glyphosate comprised 72-99% of the residue in leaves, 91% in stems, 96% in roots and 94% in beans. Residues of AMPA were 5% or less of the TRR and were generally present in ratios with glyphosate comparable to that present in the administered formulation.

Only low levels of radioactivity were recovered in grasses, alfalfa and clover grown on [¹⁴C]glyphosate treated soil or in soil into which [¹⁴C]glyphosate treated quackgrass was incorporated. Glyphosate was the only component of the ¹⁴C residue of foliar treated grass and alfalfa extracted with water, a process that recovered > 95% of the TRR. Small amounts of AMPA were detected in grass samples following foliar application of [¹⁴C]glyphosate. Drying grass and alfalfa to form hay did not alter the ¹⁴C residues.

Metabolism studies have been completed in glyphosate tolerant soya beans, sugar beet, and cotton crops that contain the CP4-EPSPS gene. In tolerant soya beans, glyphosate is metabolized substantially to AMPA, the latter can be conjugated with natural plant constituents to give trace level metabolites, or degraded to one carbon fragments that are incorporated into natural products. None of the trace level metabolites account for greater than 2% of the TRR in any soya bean raw agricultural commodity. Glyphosate plus AMPA account for at least 66% of the total radioactive residues in forage, hay, and grain. Glyphosate residues differ among the plant components accounting for about 90% of the TRR in forage but only about 25% of the TRR in grain. AMPA accounted for only 6.8% of the TRR in forage, but was the major ¹⁴C compound in grain accounting for up to 49% of the TRR. About 9% of the TRR in grain was shown to be due to incorporation of ¹⁴C into natural products; in the oil as fatty acids, in the aqueous extract as soluble components, and in the acid hydrolysate of the extracted grain as amino acids and natural organic acids.

In glyphosate tolerant cotton, glyphosate and AMPA account respectively for 91-95% and 0.7-1.6% of the TRR in forage. In cottonseed, glyphosate is the major extractable radiolabeled compound (12-24 % of the TRR) and only trace levels of AMPA are present (< 2% of the TRR). A significant fraction of the residues in the seed are attributed to incorporation into natural products; 10-12% of the TRR was characterized as saponifiable fatty acids in oil and 54-75% of the TRR was present as natural products.

The metabolism of [¹⁴C]glyphosate in tolerant sugar beet was very similar to soya beans and cotton. Glyphosate is partially metabolized to AMPA and low levels of AMPA conjugates. Glyphosate and AMPA together account for at least 99 and 81% of the TRR in roots and tops, respectively. AMPA is further converted, to a limited degree, to low levels of simple conjugates. In addition to conjugation, ¹⁴C is broadly incorporated into a wide variety of natural products and plant constituents.

The results of all the studies demonstrate that the metabolic fate of glyphosate in tolerant plants is the same as in non-tolerant plants.

Environmental fate

The Meeting received information on the behaviour and fate of glyphosate during solution photolysis and aerobic soil metabolism. Consistent with the policy outlined by the 2003 JMPR only the environmental fate data relevant to the residues of glyphosate in crops were evaluated.

Crop rotation studies were not provided. However, aerobic soil metabolism of glyphosate was rapid with inferred degradation half-lives of 3.6-25 days depending on the soil system studied. The major metabolite formed was AMPA which was further degraded to CO₂. In aqueous solution glyphosate is stable to hydrolysis. The rate of degradation in field and in aquatic environments is such that glyphosate is not expected to persist in the environment.

Methods of analysis

Glyphosate and AMPA residues are measured as derivatives following clean-up of aqueous extracts by cation and anion exchange, the derivatization reaction varying with the chromatographic method used for separation (GC, HPLC) and detection system employed (FPD in phosphorous mode, fluorescence detector, UV, MS and MS/MS). Satisfactory recoveries at the LOQs of 0.05 mg/kg for both glyphosate and AMPA were reported for numerous commodities.

Stability of pesticide residues in stored analytical samples

The Meeting received information on the stability of glyphosate residue samples during storage of analytical samples at freezer temperatures. The available storage stability data indicate that residues of glyphosate and AMPA are stable under frozen storage conditions (-20°C) in/on the following commodities (storage interval in parentheses): beans, rape and linseed (18 months), wheat grain and straw, rye grain and straw (1349 days), pasture grass (362 days), soya bean seed (183 days) soya bean straw (398 days), corn grain (944 days), sorghum forage (958 days), sorghum straw (958 days), clover (944 days) and tomatoes (938 days). Residues were stable in animal commodities (pig, cow and chicken tissues and milk) for at least 700 days. For eggs residues were stable for 431 days.

Residues of AMPA are stable under frozen storage conditions (-20 °C) in/on the following commodities (storage interval in parentheses): pasture grass (362 days), soya bean seed (183 days) soya bean straw (398 days), corn grain (944 days), sorghum forage (958 days), sorghum straw (958 days), clover (944 days) and tomatoes (938 days). Residues were stable in animal commodities (pig, cow and chicken tissues and milk) for at least 700 days. For eggs residues were stable for 431 days.

Definition of the residue

The metabolism studies in coffee, corn, cotton, soya beans, wheat, pasture grasses and alfalfa as well as on the glyphosate tolerant crops cotton, soya beans and sugar beet patterns of metabolites were similar in different species of plants. The main metabolite found in plant metabolism studies was AMPA. The Meeting agreed that glyphosate together with AMPA should be regarded as the residues of toxicological concern.

For the purposes of estimation of dietary intake and to enable comparison of the calculated intakes with the ADI it is preferable to express the residues in terms of glyphosate (glyphosate = 1.5 × AMPA).

Currently, the residue definition for glyphosate is "glyphosate". In national systems the residue definition for glyphosate is generally also the parent compound.

The Meeting agreed that the residue definition applicable to glyphosate would continue to be the parent compound. As for estimation of dietary intake and the risk assessment component relating to exposure, the 2004 JMPR concluded that AMPA was of no greater toxicological concern than its parent compound and set a group ADI of 0-1 mg/kg bw for the sum of glyphosate and AMPA.

For glyphosate STMR estimation, residue = glyphosate + 1.5 × AMPA

Definition of glyphosate residue (for compliance with MRLs): glyphosate

Definition of glyphosate residue (for estimation of dietary intake): sum of glyphosate and AMPA, expressed as glyphosate.

These definitions apply to plant and animal commodities.

Results of supervised trials on crops

The Meeting received data from supervised trials on the following crops: olives, bananas, kiwifruit, beans (dry), peas (dry), lentils, soya beans (conventional and tolerant), sugar beet (glyphosate tolerant), barley, maize (conventional and tolerant), oats, rye, sorghum, wheat, sugarcane, almonds, pecan, macadamia, walnuts, cotton (conventional and tolerant), linseed, mustard, rape, sunflower, coffee, tea, alfalfa and grasses.

Glyphosate may be applied prior to crop emergence (pre-emergence = PRE), shortly after crop emergence (early post-emergence = EPO), between EPO and a few weeks before harvest (late post-emergence = LPO) and prior to harvest (pre-harvest = PH). In addition glyphosate may be applied to weeds in the crop as a spot treatment or by wiper application to weeds in which case the area treated is generally less than 10% of the area planted and a directed sprays in which the crop is not exposed. The Meeting considered these later methods of application as unlikely to result in significant residues in crops. As such spot, wiper and directed sprays (e.g., hooded sprayers) were not included in consideration of GAP.

When applied as pre-harvest residues in the raw agricultural commodity (RAC) are mainly determined by applications made when the plant is growing and transport from the application site to the RAC occurs, rather than applications made to senescing crops where the RAC is protected from the spray, such as beans in pods. For commodities that are exposed and glyphosate is applied as a pre-harvest application to senescent crops, it is the pre-harvest spray that has the greatest influence on residues. If a range of application protocols involving different numbers of sprays, timing and application rates were used for a crop grown at a single location, the highest residue from any trial at the location and carried out with numbers of applications and rates within the range permitted by GAP was selected.

The limits of detection of glyphosate and AMPA are typically 0.05 mg/kg. When glyphosate and AMPA were summed, AMPA was converted to glyphosate equivalents (AMPA mg/kg × 1.5). If AMPA residues are < 0.05, they are not summed with glyphosate, because they are typically much less than glyphosate residues. If both glyphosate and AMPA are < LOQ, then sum is < LOQ of glyphosate. The exception is where there is evidence that AMPA residues are comparable to glyphosate residues such as for soya beans in which case the residues are summed and if both glyphosate and AMPA residues are < LOQ, the sum is less than the combined LOQs for glyphosate and AMPA.

Olives

Trials on olives were conducted in Greece (no GAP provided), Italy (GAP 4.3 kg ae/ha; no PHI specified) and Spain (directed sprays/spot sprays GAP 2.5 kg ae/ha, PHI not specified, assumed 0 days or 4.3 kg ae/ha, PHI 7 days for fruit on the ground, the lower rate, shorter PHI was taken to be the critical use for Spain). Two trials from Italy and two from Spain matched Italian GAP. Glyphosate residues found in fruit, harvested from the ground one or more days after application were 0.39 and 0.66 mg/kg, for the Italian trials, and 0.17 and 0.12 mg/kg for the Spanish trials. Residues of AMPA were not measured in these trials. Four trials from Spain matched that countries GAP (2.5 kg ae/ha). Residues in fruit collected from the ground were 6.7, 12, 12 and 12 mg/kg for glyphosate. Residues of AMPA were all < 0.05 mg/kg; the sum of glyphosate and AMPA residues were 6.7, 12, 12 and 12 mg/kg.

The Meeting decided that the residues from trials complying with the GAP of Italy and Spain were from different populations and that they could not be combined for estimating a maximum residue level. The Meeting agreed that the number of residue trials was insufficient to estimate a maximum residue level for olives.

Bananas

Trials on bananas were conducted in Brazil (GAP of 4.3 kg ae/ha, with a PHI of 30 days), Honduras, Panama, Colombia and Ecuador (no GAP supplied). For three trials from Brazil that matched Brazilian GAP residues found of glyphosate and AMPA were < 0.05 mg/kg. In four trials conducted in Honduras, Panama, Colombia and Ecuador, at applications rates higher than Brazilian GAP, residues found in pulp and peel, when expressed as whole bananas, were < 0.05 mg/kg for both glyphosate and AMPA. The Meeting decided to utilize the trials at higher application rates in support of the Brazil trials to recommend a maximum residue level of 0.05 (*) for glyphosate in bananas. The HR and STMR for total residues are both 0.05 mg/kg.

Kiwifruit

Trials on kiwifruit were conducted in Italy (GAP of 4.3 kg ae/ha ground directed spray, with no PHI specified). None of the trials matched GAP. The Meeting agreed to withdraw its previous recommendation of 0.1 (*) mg/kg for kiwifruit.

Beans, dry

Trials on beans, dry were conducted in Belgium (no GAP provided), Denmark (no GAP provided), the UK (GAP of 1.4 kg ae/ha when grain moisture is <30% generally applied 7-14 days before harvest), the USA (GAP 0.43-4.2 kg ae/ha pre-emergent). The Netherlands GAP is 0.72-2.2 kg ae/ha, PHI 7 days. In Canada GAP is 0.9 kg ae/ha when grain moisture is < 30% generally 7-14 days before harvest. Trials from Belgium and Denmark were evaluated against UK GAP.

Five trials from the UK matching GAP had residues of 0.11, 0.12, 0.16, 0.20 and 1.8 mg/kg. One trial each from Belgium and Denmark matched UK GAP with residues of < 0.05 and 0.17 mg/kg respectively. Residues of AMPA were < 0.05 mg/kg.

None of the USA trials matched GAP for that country and were evaluated against the GAP of Canada. Thirteen trials conducted in the USA approximated Canadian GAP. Residues found in beans (dry) were < 0.05, 0.07, 0.09, 0.10, 0.11, 0.13, 0.19, 0.30, 0.32, 0.37, 0.38, 0.68 and 1.6 mg/kg. Residues of AMPA were all < 0.05 mg/kg.

The Meeting considered that the field trials conducted according to the GAP of the UK and the USA were from similar residue populations and could be combined for the purposes of estimating a maximum residue level. Glyphosate residues, in ranked order were (n = 19): < 0.05, < 0.05, 0.07, 0.09, 0.10, 0.11, 0.12, 0.13, 0.16, 0.17, 0.19, 0.20, 0.30, 0.32, 0.37, 0.38, 0.68, 1.6 and 1.8 mg/kg. The Meeting confirmed its previous recommendation of a maximum residue level for glyphosate in beans (dry) of 2 mg/kg.

As residues of AMPA were < 0.05 mg/kg, total residues for the purposes of estimating an STMR and highest residues are the same as the glyphosate values. The highest residue and STMR are estimated to be1.8 and 0.17 mg/kg respectively.

Peas, dry

Trials on peas, dry were conducted in Belgium (no GAP provided), Canada (GAP of 0.9 kg ae/ha, when grain moisture is < 30% generally applied, 7-14 days before harvest), Denmark (no GAP provided), the UK (GAP 1.4 kg ae/ha, when grain moisture is < 30% generally 7-14 days before harvest) and the USA (GAP 0.43-4.2 kg ae/ha pre-emergent). The Netherlands GAP is 0.72-2.2 kg ae/ha, PHI 7 days. Trials from Belgium and Denmark were evaluated against UK GAP.

Residues in six UK trials that approximated GAP of that country were 0.13, 0.16, 0.17, 1.7, 1.8 and 2.1 mg/kg. Residues in a single trial from Belgium and Denmark that approximated the UK GAP were 0.17 and 0.5 mg/kg respectively. When measured, residues of AMPA were < 0.05 (4) mg/kg.

In four trials from Canada that approximated GAP of that country, residues of glyphosate were 0.5, 0.82, 1.4 and 8.9 mg/kg. AMPA residues were < 0.05 mg/kg.

The Meeting considered that the field trials conducted according to the GAP of the UK and Canada were from similar residue populations and could be combined for the purposes of estimating a maximum residue level. Glyphosate residues, in ranked order were (n = 11): 0.13, 0.16, 0.17, 0.17, 0.5, 0.5, 0.82, 1.4, 1.7, 1.8 and 2.1 mg/kg. The Meeting estimated a maximum residue level for glyphosate in peas (dry) of 5 mg/kg confirming its previous recommendation.

As residues of AMPA were < 0.05 mg/kg, total residues for the purposes of estimating an STMR and highest residue are the same as the glyphosate values. The STMR is estimated to be 0.5 mg/kg and highest residue 2.1 mg/kg.

Lentils

Trials on lentils were conducted in Canada (GAP of 0.9 kg ae/ha, when crop has < 30% grain moisture content and lowermost pods (bottom 15%) are brown and seeds rattle, with a 7-14 days PHI). Two trials matched GAP of Canada with residues of glyphosate of < 0.05 and 3.0 mg/kg and AMPA of < 0.05 mg/kg. The total residues were < 0.05 and 3.0 mg/kg.

The Meeting considered there were insufficient trials to recommend a maximum residue level for lentils.

Soya beans

Trials on conventional soya beans were conducted in the USA (GAP of 4.2 kg ae/ha PRE, 4.2 kg ae/ha PH, with a PHI of 7 days). Four trials approximated GAP for the USA had glyphosate residues of 0.45, 5.4, 13 and 17 mg/kg. Corresponding AMPA residues were < 0.05, 1.2, 1.9 and 1.8 mg/kg respectively.

Additionally, trials were conducted on glyphosate tolerant soya beans (GAP of 0.43-4.2 kg ae/ha PE, 1.7 kg ae/ha LPO, 0.83 kg ae/ha PH, combined LPO+PH < 2.5 kg ae/ha, PHI 14 days). GAP allows pre-harvest applications together with post-emergent directed sprays as well as pre-harvest over the top sprays. The Meeting considered that a single pre-harvest application made close to harvest would not give rise to residues in beans (dry) representative of GAP as when the last application is made the crop has entered into senescence, limiting transport of residues to the seed. Only trials that included pre-emergent and in-crop applications were considered as compliant with USA GAP. The Meeting also noted that in trials conducted in the USA the pre-emergent application was typically at a higher rate than permitted by GAP, 6.4 versus 4.2 kg ae/ha, but considered the difference in application rates to account for less than 10% difference in the residue at harvest and that the later post-emergent sprays determined the residue. In a metabolism study on soya beans residues in seed after a single pre-emergent application at 5.4 kg ae/ha were < 0.01 mg/kg. While residues found after one or two post-emergent applications at 0.84 or 1.7 kg ae/ha, with the last application occurring 61 days prior to harvest, were 0.04 and 4.4 mg/kg respectively. Thirty-two trials from the USA approximated GAP of that country. Residues of glyphosate were 0.27, 0.28, 0.34, 0.37, 0.42, 0.44, 0.51, 0.56, 0.60, 0.70, 1.0, 1.1, 1.4, 1.4, 1.5, 1.7, 1.8, 1.9, 1.9, 1.9, 2.0, 2.6, 2.7, 2.7, 3.0, 3.3, 3.5, 3.6, 3.7, 4.4, 5.3 and 5.6 mg/kg. Total residues were 0.59, 0.78, 0.89, 1.0, 1.1, 1.1, 1.2, 1.2, 1.5, 1.6, 2.4, 3.2, 4.0, 4.0, 4.3, 4.7, 4.9, 5.1, 5.4, 5.7, 6.2, 6.6, 7.1, 7.6, 7.6, 7.9, 8.2, 8.5, 11, 11, 11 and 17 mg/kg.

The Meeting considered that the field trials conducted on conventional and glyphosate tolerant soya beans according to the GAP of the USA to be from similar residue populations and could be combined for the purposes of estimating a maximum residue level. Glyphosate residues, in ranked order were (n = 36): 0.27, 0.28, 0.34, 0.37, 0.42, 0.44, 0.45, 0.51, 0.56, 0.60, 0.70, 1.0, 1.1, 1.4, 1.4, 1.5, 1.7, 1.8, 1.9, 1.9, 1.9, 2.0, 2.6, 2.7, 2.7, 3.0, 3.3, 3.5, 3.6, 3.7, 4.4, 5.3, 5.4, 5.6, 13 and 17 mg/kg. The Meeting confirmed its previous recommendation of a maximum residue level for glyphosate in soya beans (dry) of 20 mg/kg.

Total residues were (n = 36): 0.45, 0.59, 0.78, 0.89, 1.0, 1.1, 1.1, 1.2, 1.2, 1.5, 1.6, 2.4, 3.2, 4.0, 4.0, 4.3, 4.7, 4.9, 5.1, 5.4, 5.7, 6.2, 6.6, 7.1, 7.2, 7.6, 7.6, 7.9, 8.2, 8.5, 11, 11, 11, 16, 17 and 20 mg/kg. The highest residue and STMR for total residues are 20 and 5.0 mg/kg respectively.

No residue data was available for immature seed and the Meeting agreed to recommend withdrawal of its previous recommendation for soya bean (immature seed) of 0.2 mg/kg.

Sugar beet (glyphosate tolerant)

Trials on sugar beet (glyphosate tolerant) were conducted in the USA (GAP of 0.43-4.2 kg ae/ha PRE, 1.3 kg ae/ha EPO from emergence to 8-leaf stage, 0.87 kg ae/ha from 8-leaf stage to canopy closure, 3.8 kg ae/ha combined maximum rate for all applications from emergence to harvest, PHI 30 days). The metabolism study indicated that pre-emergent applications of glyphosate do not make a significant contribution to the residue in sugar beet roots at harvest. An examination of the trial data indicated that it was probable that the last application contributed most to the residue at harvest. The Meeting considered that none of the trials matched GAP.

Cereal grains

The Meeting decided to evaluate the residue trial data for barley, maize, oats, rye, sorghum and wheat for a possible cereal grains recommendation. Estimates of values for total residues (HR and STMR) that are required for dietary intake and animal dietary burden calculations are discussed under each commodity while maximum residue level estimation is discussed at the end after wheat.

Barley

Trials on barley were conducted in Belgium (no GAP provided), France (no GAP provided) and the UK (GAP of 0.54-1.4 kg ae/ha, when grain moisture is < 30% generally 7-14 days before harvest). Trials conducted in Belgium and France were evaluated against the GAP of the Netherlands (GAP of 0.72-2.2 kg ae/ha, with a PHI of 7 days).

Residues of glyphosate in four UK trials approximating UK GAP were 1.4, 3.3, 4.4 and 11 mg/kg. Total residues were 1.4, 3.3, 4.4 and 11 mg/kg. Residues in two trials from Belgium (10 and 20 mg/kg), two from the UK (6.3 and 8.4 mg/kg) and nineteen from France (1.5, 2.2, 2.8, 2.9, 3.3, 5.5, 5.9, 6.3, 6.7, 7.2, 7.9, 8.5, 9.6, 13, 14, 15, 19, 19 and 19 mg/kg) approximated the GAP of the Netherlands.

The Meeting considered the trials to all be from similar populations and decided to combine the results for the purpose of maximum residue level recommendation. Glyphosate residues in rank order were (n = 27): 1.4, 1.5, 2.2, 2.8, 2.9, 3.3, 3.3, 4.4, 5.5, 5.9, 6.3, 6.3, 6.7, 7.2, 7.9, 8.4, 8.5, 9.6, 10, 11, 13, 14, 15, 19, 19, 19 and 20 mg/kg. Total residues (where glyphosate and AMPA were reported) were (n = 22): 1.1, 2.2, 2.8, 3.0, 3.3, 3.3, 4.4, 5.6, 6.0, 6.8, 7.3, 8.0, 8.6, 9.7, 10, 11, 14, 15, 19, 19, 19 and 20 mg/kg. The Meeting estimated a high residue and STMR for total residues in barley of 20 and 7.65 mg/kg respectively.

Maize

Trials on conventional maize were conducted in the USA (GAP of 0.43-4.2 kg ae/ha PRE, 0.87 kg ae/ha directed spray when the crop is > 30 cm tall and 2.5 kg ae/ha PH when grain moisture is < 35%, with a PHI of 7 days).

Trials on conventional maize were conducted in the USA (GAP of 0.43-4.2 kg ai/ha PRE, 0.87 kg ai/ha directed spray when crop > 30 cm tall and 2.5 kg ai/ha PH grain moisture < 35%, with a PHI of 7 days). From 21 trials that approximated USA GAP, which involved a single pre-harvest application to conventional maize residues of < 0.05 (12), 0.05 (2), 0.06 (2), 0.07, 0.09, 0.19, 0.54 and 3.0 mg/kg were found. Corresponding total residues were < 0.12 (11), < 0.14 (2), 0.14, < 0.16, 0.19, < 0.23, < 0.25, < 0.26, < 0.62 and 3.0 mg/kg respectively.

Additionally, trials were conducted on glyphosate tolerant maize (GAP 0.43-4.2 kg ae/ha PRE, 0.83 kg ae/ha EPO, 1.7 kg ae/ha LPO, total EPO and LPO applications < 2.5 kg ae/ha, 0.87 kg ae/ha PH < 35% grain moisture, PHI 7 days). None of the trials on glyphosate tolerant maize incorporated a pre-harvest application. Trials on conventional maize showed that pre-harvest applications made a significant contribution to the final residues found. The Meeting considered that none of the trials on glyphosate tolerant maize matched GAP of the USA.

Oats

Trials on oats were conducted in Canada (GAP of 0.18-4.3 kg ae/ha PRE, 0.9 kg ae/ha PH, application at < 30% grain moisture typically 7-14 days before harvest), Denmark (no GAP provided) and the UK (GAP 0.54-1.4 kg ae/ha, < 30% grain moisture typically 7-14 days before harvest). Three trials from Canada matched GAP of that country with glyphosate residues of 0.70, 3.1 and 4.6 mg/kg (total residues 0.70, 3.2 and 4.8 mg/kg). Eight trials from the UK and three from Denmark approximated UK GAP with glyphosate residues of 0.9, 3.4, 3.4, 4.1, 4.9, 4.9, 5.2, 6.0, 8.1, 8.6 and 14 mg/kg. Total residues in three trials that also measured AMPA levels were 3.5, 6.1 and 8.4 mg/kg.

The Meeting considered the trials approximating GAP in Canada and the UK to be from the same population and decided to combine the results for the purpose of estimating the maximum residue level and STMR. Residues of Glyphosate in ranked order were (n = 14): 0.7, 0.9, 3.1, 3.4, 3.4, 4.1, 4.6, 4.9, 4.9, 5.2, 6.0, 8.1, 8.6 and 14 mg/kg. AMPA residues were only measured in six of the fourteen trials considered giving total residues of 0.7, 3.2, 3.5, 4.8, 6.1 and 8.4 mg/kg. The Meeting estimated a high residue of 14 mg/kg and an STMR of 4.15 mg/kg.

Rye

Trials on rye were conducted in Denmark (no GAP provided) and were evaluated against the GAP of the UK (GAP 0.54-1.4 kg ae/ha, < 30% grain moisture PHI 7-14 days). Three trials approximated UK GAP with glyphosate residues of 1.6, 1.6 and 2.2 mg/kg (AMPA not measured).

The Meeting considered three trials insufficient to estimate a maximum residue level for rye.

Sorghum

Trials on sorghum were conducted in the USA (GAP of 0.43-4.2 kg ae/ha PRE, 0.87 kg ae/ha directed spray when crop > 30 cm tall and 1.7 kg ae/ha PH grain moisture < 35%, PHI 7 days). Thirteen trials matched GAP for the USA with glyphosate residues of 1.1, 1.3, 1.4, 1.7, 1.8, 4.4, 4.6, 5.3, 6.0, 6.3, 6.4, 12 and 13 mg/kg (total residues 1.1, 1.4, 1.6, 1.8, 1.8, 4.5, 4.8, 5.4, 6.2, 6.6, 6.6, 12 and 13 mg/kg). The Meeting estimated a highest residue of 13 mg/kg and an STMR of 4.8 mg/kg for total residues in sorghum grain.

Wheat

Trials on wheat were conducted in Belgium (no GAP provided), France (no GAP provided) and the UK (GAP of 0.54-1.4 kg ae/ha, < 30% grain moisture with a PHI of 7 days). Trials conducted in Belgium and France were evaluated against the GAP of the Netherlands (GAP is 0.72-2.2 kg ae/ha, PHI 7-14 days). Seven trials approximated GAP of the UK with glyphosate residues of 0.1, 0.1, 0.3, 0.3, 0.5, 0.7 and 1.0 mg/kg (residues of AMPA were all < 0.05 mg/kg). Two trials from Belgium (1.7 and 3.6 mg/kg), two from the UK (1.1 and 1.2 mg/kg) and nineteen from France (0.16, 0.53, 0.66, 0.80, 0.90, 0.99, 1.1, 1.2, 1.3, 1.5, 1.7, 2.1, 2.4, 3.8, 3.9, 4.0, 4.9, 6.3 and 9.5) approximated GAP of the Netherlands.

The Meeting decided that the residues conducted according to GAP of the Netherlands and the UK could be combined for the purposes of STMR and maximum residue level recommendation. Residues of glyphosate in rank order were (n = 30): 0.1, 0.1, 0.16, 0.3, 0.3, 0.5, 0.53, 0.66, 0.7, 0.80, 0.90, 0.99, 1.0, 1.1, 1.1, 1.2, 1.2, 1.3, 1.5, 1.7, 1.7, 2.1, 2.4, 3.6, 3.8, 3.9, 4.0, 4.9, 6.3 and 9.5 mg/kg. Total residues were (n = 24): 0.1, 0.1, 0.24, 0.3, 0.3, 0.5, 0.53, 0.66, 0.7, 0.80, 0.90, 1.0, 1.1, 1.1, 1.2, 1.5, 1.9, 2.2, 3.7, 3.8, 3.9, 4.0, 4.9 and 6.5 mg/kg where measured. The Meeting recommended a high residue of 9.5 mg/kg and an STMR of 1.05 mg/kg for total residues in wheat grain.

Data are available for a large range of cereal grains and the Meeting considered it appropriate to estimate a group maximum residue level for cereal grains. As the residues of glyphosate in maize are much lower than in the other cereal grains due to the protection afforded by the husk and also the absence of data for rice, the Meeting decided to recommend a group maximum residue level for cereal grain except maize and rice of 30 mg/kg. The estimated maximum residue level replaces the previous recommendations for barley, oats and sorghum of 20 mg/kg and wheat of 5 mg/kg.

Total residues for cereal grains except maize and rice were (n = 65) 0.1, 0.1, 0.24, 0.3, 0.3, 0.5, 0.53, 0.66, 0.7, 0.7, 0.80, 0.90, 1.0, 1.1, 1.1, 1.1, 1.1, 1.2, 1.4, 1.5, 1.6, 1.8, 1.8, 1.9, 2.2, 2.2, 2.8, 3.0, 3.2, 3.3, 3.3, 3.5, 3.7, 3.8, 3.9, 4.0, 4.4, 4.5, 4.8, 4.8, 4.9, 5.4, 5.6, 6.0, 6.1, 6.2, 6.5, 6.6, 6.6, 6.8, 7.3, 8.0, 8.4, 8.6, 9.7, 10, 11, 12, 13, 14, 15, 19, 19, 19 and 20 mg/kg. The Meeting estimated an STMR of 3.7 mg/kg for total residues in cereal grains except maize and rice. This STMR value will be used in the dietary intake calculations for cereal grain commodities other than barley, maize, oats, sorghum and wheat.

Using the results for conventional maize of (n = 21) < 0.05 (12), 0.05 (2), 0.06 (2), 0.07, 0.09, 0.19, 0.54 and 3.0 mg/kg and corresponding total residues of < 0.12 (11), < 0.14 (2), 0.14, < 0.16, 0.19, < 0.23, < 0.25, < 0.26, < 0.62 and 3.0 mg/kg, the Meeting recommended a maximum residue level of 5 mg/kg for maize. The Meeting estimated highest residue and STMR levels for total residues in maize of 3.0 and < 0.12 mg/kg respectively.

Sugarcane

Trials on sugarcane were conducted in the USA (GAP of 0.49 kg ae/ha, PHI 21-35 days, 0.84 kg ae/ha; PHI 28-70 days). Seven trials from the USA matched GAP with residues 0.07, 0.13, 0.21, 0.27, 0.28, 0.69 and 0.97 mg/kg (total residues 0.07, 0.13, 0.21, 0.27, 0.28, 0.69 and 0.97 mg/kg).

The Meeting recommended a maximum residue level of 2 mg/kg for residues of glyphosate in sugarcane. The high residue and STMR levels for total residues are 0.97 and 0.27 mg/kg respectively.

Tree nuts

Trials on tree nuts (almonds, pecans, macadamias and walnuts) were conducted in the USA (GAP 0.43-4.3 kg ae/ha, directed applications, PHI 3 days). None of the trials matched GAP.

Cottonseed

Trials on conventional cotton were conducted in the USA (GAP of 0.43-4.2 kg ae/ha PRE, 0.43-4.2 kg ae/ha directed spray and 0.43-1.7 kg ae/ha PH, PHI 7 days). No trials matched GAP for conventional cotton.

Additionally, trials were conducted on glyphosate tolerant cotton (USA GAP of 0.43-4.2 kg ae/ha PRE do not exceed 4.2 kg ae/ha/season for pre-emergent application, 0.83 kg ae/ha for in-crop directed applications which must not exceed 3.3 kg ae/ha/season, 1.7 kg ae/ha PH do not exceed 1.7 kg ae/ha/season for pre-harvest application; combined applications must not exceed 6.6 kg ae/ha/season, PHI 7 days). USA GAP allows pre-harvest applications together with post-emergent directed sprays as well as pre-harvest over the top sprays. The Meeting considered that a single pre-harvest application would not give rise to residues in cottonseed representative of GAP as the timing last application coincides with crop senescence, limiting potential transport of residues to the seed. Only trials that included pre-emergent and in-crop applications were considered as compliant with USA GAP.

Twenty-three trials from the USA approximated USA GAP with glyphosate residues of 0.46, 0.50, 0.69, 1.2, 1.3, 2.5, 2.8, 3.6, 4.2, 4.6, 4.9, 5.0, 7.2, 7.5, 9.7, 13, 16, 18, 18, 18, 21, 22 and 28 mg/kg (total residues 0.46, 0.58, 0.69, 1.2, 1.4, 2.5, 2.9, 3.7, 4.4, 4.6, 5.1, 5.2, 7.5, 7.9, 9.8, 14, 17, 18, 19, 19, 22, 23 and 28 mg/kg). The Meeting estimated a maximum residue level of 40 mg/kg for glyphosate to replace its previous recommendation of 10 mg/kg and estimated a highest residue of 28 mg/kg and an STMR of 5.2 mg/kg for total residues in cottonseed.

Linseed (flax)

Trials on linseed were conducted in the UK (GAP 1.1-1.4 kg ae/ha, < 30% grain moisture PHI 7-28 days). Two trials from the UK matched GAP from that country with glyphosate residues of 2.0 and 4.6 mg/kg. The Meeting agreed the number of trials was insufficient for the purposes of estimating a maximum residue level.

Mustard seed

Trials on mustard were conducted in the UK (GAP 1.1-1.4 kg ae/ha, < 30% grain moisture, with a PHI of 8-10 days). Two trials matched GAP in the UK with glyphosate residues of 0.25 and 2.6 mg/kg (total residues 0.25 and 2.6 mg/kg). The Meeting considered two trials insufficient to estimate a maximum residue level.

Rape (Canola)

Trials on rape were conducted in Belgium (no GAP provided), Canada (0.18-4.3 kg ae/ha PRE; 0.9 kg ae/ha PH, < 30% grain moisture, with a PHI of 7-14 days), Denmark (no GAP provided), Finland (no GAP provided) France (no GAP provided), Sweden (no GAP provided) and the UK (GAP 1.1-1.4 kg ae/ha, < 30% grain moisture with a PHI of 14-21 days). Trials from Belgium, France, Denmark, Finland and Sweden were evaluated against the GAP of the UK.

Four trials from Canada matched GAP of that country with residues found of 0.61, 1.8, 2.1 and 3.6 mg/kg (total residues 0.61, 1.8, 2.1 and 3.7 mg/kg). Two trials from Belgium (0.23 and 4.6 mg/kg), ten trials from France (0.21, 0.23, 0.35, 0.50, 0.87, 0.93, 0.96, 1.4, 1.9 and 5.6 mg/kg), ten from the UK (0.16, 0.4, 0.35, 0.60, 0.7, 0.7, 0.80, 0.9, 1.5 and 2.7 mg/kg), five from Denmark (4.1, 6.7, 8.6, 10 and 12 mg/kg), one from Finland (1.5 mg/kg) and three from Sweden (0.40, 2.0 and 2.8 mg/kg) approximated GAP from the UK. The Meeting agreed that the residue populations for trials approximating the GAP of Canada and the UK could be combined for the purpose of recommending maximum residue levels and STMRs.

Residues of glyphosate in ranked order were (n = 35): 0.16, 0.21, 0.23, 0.23, 0.4, 0.35, 0.35, 0.40, 0.50, 0.60, 0.61, 0.7, 0.7, 0.80, 0.87, 0.9, 0.93, 0.96, 1.4, 1.5, 1.5, 1.8, 1.9, 2.0, 2.1, 2.7, 2.8, 3.6, 4.1, 4.6, 5.6, 6.7, 8.6, 10 and 12 mg/kg. The Meeting recommended a maximum residue level of 20 mg/kg for glyphosate residues in rape seed. The new recommendation replaces the previous recommendation of 10 mg/kg.

Total residues in rank order were (n = 31): 0.16, 0.21, 0.23, 0.23, 0.35, 0.35, 0.4, 0.40, 0.50, 0.60, 0.7, 0.7, 0.80, 0.87, 0.9, 0.93, 0.96, 1.4, 1.5, 1.6, 1.9, 2.0, 2.7, 3.0, 4.1, 4.6, 5.7, 6.7, 8.6, 10 and 12 mg/kg. The Meeting estimated a highest residue of 12 mg/kg and an STMR of 0.93 mg/kg for total residues in rape seed.

Sunflower

Trials on sunflowers were conducted in Hungary (GAP 0.54-4.3 kg ae/ha PRE, 1.8 kg ae/ha PH 20-30% grain moisture, with a PHI of 6 days if rate is 0.54 kg ae/ha PH, otherwise a PHI of 21 days). Eight trials matched GAP of Hungary with glyphosate residues of < 0.05, < 0.05, 0.16, 0.39, 0.40, 3.7, 4.9 and 5.6 mg/kg. Total residues in the three trials that also measured AMPA were < 0.05, < 0.05 and 0.39 mg/kg. The Meeting estimated a maximum residue level of 7 mg/kg for residues in sunflower seed. Available evidence suggests residues of AMPA in sunflower seed are unlikely to exceed 10% of the glyphosate residue. The Meeting decided to utilize the glyphosate residues to estimate a highest residue and an STMR for total residues in sunflower seed of 5.6 and 0.395 mg/kg respectively.

Coffee beans

Trials on coffee were conducted in Brazil (no GAP provided), Columbia (no GAP provided), Costa Rica (no GAP provided) and the USA (GAP 0.43-4.3 kg ae/ha PRE and as directed sprays, with a PHI of 28 days). All trials were evaluated against the GAP of the USA yielding four trials that approximated GAP with glyphosate residues of < 0.05, < 0.05, 0.30 and 0.58 mg/kg (total residues < 0.05, < 0.05, 0.30 and 0.58 mg/kg). The Meeting considered four trials insufficient to estimate a maximum residue level for coffee beans.

Tea

Trials on tea were conducted in China (no GAP provided), India (no GAP provided), Sri Lanka (no GAP provided) and Japan (GAP single application at 0.9-2.3 kg ae/ha directed spray, with a PHI of 7 days). All trials were evaluated against the GAP of Japan. Four trials from Sri Lanka matched GAP for Japan with glyphosate residues of 0.12, 0.21, 0.27 and 0.42 mg/kg (total residues 0.12, 0.21, 0.27 and 0.42 mg/kg). The Meeting considered four trials insufficient to estimate a maximum residue level for tea.

Alfalfa

Trials on alfalfa were conducted in Canada (GAP of 0.9-1.8 kg ae/ha, PHI not specified but typically 3-7 days before last cut before rotation of renovation) and the USA (GAP 0.43-4.3 kg ae/ha PRE, 1.7 kg ae/ha PH for renovation, PHI 1.5 days).

Twenty trials approximated the GAP of the USA with glyphosate residues (as received) in forage of 54, 54, 55, 57, 58, 61, 64, 66, 70, 74, 76, 77, 85, 94, 98, 99, 107, 114, 122 and 153 mg/kg. Total residues were 54, 54, 55, 57, 58, 61, 64, 66, 71, 74, 76, 78, 86, 95, 99, 100, 108, 115, 123 and 154 mg/kg. The Meeting estimated a high residue of 154 mg/kg together with a median residue of 75 mg/kg for total residues in alfalfa forage all on an as received basis. The Meeting agreed to withdraw its previous recommendation for alfalfa forage.

Three trials from Canada matched GAP of that country with glyphosate residues in fodder (hay) of 57, 77 and 78 mg/kg (total residues 58, 78 and 79 mg/kg). Residues in hay from USA trials that matched GAP were 0.45, 83, 97, 97, 117, 131, 148, 187, 189, 195, 196, 204, 208, 214, 219, 256, 257, 280, 335 and 341 mg/kg. Total residues were: 0.79, 84, 98, 98, 119, 132, 149, 188, 190, 197, 197, 206, 210, 215, 221, 260, 259, 282, 338 and 344 mg/kg.

Residues of glyphosate in alfalfa fodder (hay) in ranked order were (n = 23): 0.45, 57, 77, 78, 83, 97, 97, 117, 131, 148, 187, 189, 195, 196, 204, 208, 214, 219, 256, 257, 280, 335 and 341 mg/kg (as received). Total residues in rank order were; 0.79, 58, 78, 79, 84, 98, 98, 119, 132, 149, 188, 190, 197, 197, 206, 210, 215, 221, 260, 259, 282, 338 and 344 mg/kg.

The Meeting recommended maximum residue level of 500 mg/kg (dry weight basis) for glyphosate in alfalfa fodder based on a high residue of 383 mg/kg (341 mg/kg ÷ 0.89 default dry matter content) together with highest residue and median residue levels of 344 and 190 mg/kg (as received) respectively for total residues.

Grass pasture

Trials on grass pasture were conducted in the USA (GAP of 0.31-4.3 kg ae/ha PRE, 4.3 kg ae/ha PH for pasture renovation, if application is less than 2.3 kg ae/ha no grazing or harvest interval is required, if > 2.3 kg ae/ha a waiting period 8 weeks applies before grazing or harvesting). Thirteen trials matching USA GAP were provided. Residues of glyphosate found in forage (from grass 15-20 cm high to boot stage) were 431, 456, 527, 616, 657, 664, 689, 713, 773, 869, 881, 884 and 1093 mg/kg (dry weight basis). Total residues were 435, 460, 530, 623, 660, 668, 691, 718, 777, 875, 881, 891 and 1099 mg/kg. The Meeting decided to utilize the averages of residues over 7 days (263, 273, 311, 333, 348, 430, 431, 449, 449, 478, 511, 612 and 615 mg/kg) to estimate a highest residue level of 615 mg/kg and a median residue of 431 mg/kg for total residues in grass forage, both on a dry weight basis.

Residues found in hay, cut when the grass was at boot to early head growth stage, were (n = 13): 7.3, 38, 66, 75, 100, 122, 187, 203, 212, 215, 233, 244 and 259 mg/kg (as received). Total residues were: 9.9, 39, 67, 77, 101, 124, 190, 210, 214, 218, 240, 248 and 262 mg/kg. The Meeting recommended a maximum residue level of 500 mg/kg (dry weight basis) for glyphosate based on a high residue of 294 mg/kg (259 mg/kg ÷ 0.88 default dry matter content) and highest and median residue levels of 262 and 190 mg/kg (as received) respectively for total residues in hay or fodder (dry) of grasses. The recommended maximum residue level replaces the previous recommendation of 50 mg/kg.

Bean fodder

Trials on beans haulm/straw were conducted in Belgium (no GAP provided) and the UK (GAP of 1.4 kg ae/ha, with a PHI of 7 days). Trials from Belgium were evaluated against UK GAP. Residues of glyphosate in haulm/straw at harvest were (n = 10): 3.4, 4.4, 7.8, 16, 17, 28, 46, 50, 51 and 93 mg/kg. AMPA was measured in four of the trials, with AMPA residues found to be less than 10% of the glyphosate residues. The therefore, Meeting agreed to use glyphosate residues to estimate the high and median residue levels. The Meeting estimated a maximum residue level of 200 mg/kg (dry weight basis) based on a highest residue of 103 mg/kg (93 mg/kg ÷ 0.90 default dry matter content) and high and median residue levels of 93 and 22.5 mg/kg (as received) for bean fodder.

Pea fodder

Trials on pea haulm/straw were conducted in Belgium (no GAP provided), Canada (GAP of 0.9 kg ae/ha, when the crop has < 30% grain moisture content and lower most pods (bottom 15%) are brown and seeds rattle, with a 7-14 days PHI) and the UK (GAP of 1.4 kg ae/ha, with a PHI of 7 days). Some trials from the UK conducted at higher rates were evaluated against the GAP of the Netherlands (0.72-2.2 kg ae/ha, PHI 7 days). Trials from Belgium were evaluated against the GAP of the UK. Residues in pea straw from Canada (16, 19, 20 and 28 mg/kg) appeared to be from a different population to that data approximating the GAP of the UK and the Netherlands; the latter were used for the purposes of maximum residue level estimation. Residues in ranked order were (n = 10): 27, 27, 31, 78, 79, 125, 154, 179, 200 and 320 mg/kg (as received). Total residues were 79, 80, 127, 155 and 181 mg/kg. AMPA was measured in five of the trials, with AMPA residues found to be less than 10% of the glyphosate residues. The Meeting therefore agreed to use glyphosate residues to estimate the high and median residue levels. The Meeting estimated a maximum residue level of 500 mg/kg (dry weight basis) based on a highest residue of 364 mg/kg (320 mg/kg ÷ 0.88 default dry matter content) and high and median residue levels of 320 and 102 mg/kg (as received) respectively for pea hay or fodder (dry).

Lentil fodder

Trials on lentil straw were conducted in Canada (GAP of 0.9 kg ae/ha, when crop has < 30% grain moisture content and lowermost pods (bottom 15%) are brown and seeds rattle, with a 7-14 days PHI). Residues of glyphosate were < 0.05 and 11 mg/kg. The Meeting considered two trials inadequate for the purposes of estimating an MRL for lentil straw.

Soya bean forage and fodder

Trials on conventional soya beans were conducted in the USA (GAP 3.7 kg ae/ha for PH use, do not graze or harvest treated hay or fodder for livestock feed within 25 days of last application. If application rate for PH is less than 0.74 kg ae/ha the livestock feed interval is reduced to 14 days after last application). The Meeting considered that pre-harvest application would give rise to residues in bean forage and fodder representative of GAP. Four trials approximating GAP for the USA had glyphosate residues in hay of 2.5, 3.4, 8.5 and 9.9 mg/kg (as received). Corresponding AMPA residues were 0.16, 0.18, 0.30 and 0.1 mg/kg respectively. The Meeting considered four trials insufficient to estimate a maximum residue level for soya bean hay. The Meeting agreed to withdraw its previous recommendation of 200 mg/kg for soya bean fodder.

Additionally, trials were conducted on glyphosate tolerant soya beans (GAP 1.7 kg ae/ha LPO; 0.83 kg ae/ha PH, do not graze or harvest treated hay or fodder for livestock feed within 14 days of last application). Residues of glyphosate in forage in trials that approximated USA GAP were 4.1, 4.5, 9.1, and 12 mg/kg (as received). No trials matched GAP for hay. The Meeting considered four trials inadequate for the purpose of estimating a maximum residue level and agreed to withdraw the previous recommendation of 5 mg/kg for soya bean forage.

Sugar beet tops

Trials were provided on sugar beet (glyphosate tolerant) from the USA (GAP 0.43-4.2 kg ae/ha PRE, 0.43-1.3 kg ae/ha LPO, 0.43-0.87 kg ae/ha PH, PHI 30 days). No trials matched GAP.

Barley straw

Trials on barley straw were conducted in Belgium (no GAP provided), France (no GAP provided) and the UK (GAP of 0.54 - 1.4 kg ae/ha, PHI 7 days). Trials conducted in Belgium, France as well as some UK trials were evaluated against the GAP of the Netherlands (GAP is 0.72-2.2 kg ae/ha, PHI 7 days). Residues in straw in four UK trials that matched the GAP of that country were 13, 41, 47 and 62 mg/kg. In twenty-three trials that matched GAP of the Netherlands residues of glyphosate in straw were 29 and 86 mg/kg for Belgium trials, 6.0, 6.9, 12, 15, 17, 33, 39, 40, 43, 59, 71, 80, 96, 102, 110, 126, 140, 147 and 160 mg/kg for trials conducted in France and 22 and 56 mg/kg for two UK trials.

The Meeting considered the residues to be from the same population and decided to pool the results. Residues of Glyphosate in barley straw in ranked order were (n = 27): 6.0, 6.9, 12, 13, 15, 17, 22, 29, 33, 39, 40, 41, 43, 47, 56, 59, 62, 71, 80, 86, 96, 102, 110, 126, 140, 147 and 160 mg/kg (as received). Total residues, where measured, were: 6.1, 6.9, 13, 14, 15, 17, 22, 30, 34, 39, 41, 42, 44, 48, 56, 61, 64, 73, 82, 89, 100, 105, 115, 126, 142, 151 and 162 mg/kg (as received). The Meeting recommended a maximum residue level for glyphosate of 400 mg/kg (dry weight basis) based on a highest residue of 180 mg/kg (160 mg/kg ÷ 0.88 default dry matter content) and highest and median residue levels for total residues of 162 and 48 mg/kg (as received) for residues in barley straw.

Maize forage and fodder

Trials on conventional maize were conducted in the USA (GAP 4.2 kg ae/ha PRE; 0.87 kg ae/ha hooded sprayers, Do not graze or feed maize forage or fodder following hooded sprayer applications; 2.5 kg ae/ha PH grain moisture < 35%, PHI 7 days). Glyphosate residues in stover/fodder 7 days after a pre-harvest application according to USA GAP were 2.1, 2.6, 3.4, 3.7, 4.8, 6.7, 8.4, 8.8, 11, 18, 23, 28, 35, 43, 43, 44, 53, 54, 55, 82, and 92 mg/kg. Total residues were: 2.1, 2.6, 3.5, 3.8, 4.8, 6.8, 8.8, 9.0, 11, 18, 24, 29, 36, 44, 45, 45, 54, 55, 56, 83 and 93 mg/kg.

Additionally, trials were conducted on glyphosate tolerant maize (GAP 4.2 kg ae/ha PRE; 1.7 kg ae/ha LPO, allowing a minimum of 50 days between application and harvest of corn forage; 0.87 kg ae/ha PH < 30% grain moisture, combined LPO+PH < 2.5 kg ae/ha, PHI 7 days). Seventeen trials on forage but no trials on tolerant maize fodder matched GAP. Glyphosate residues were (n = 17): 0.30, 0.50, 0.54, 0.66, 0.73, 0.79, 0.87, 0.92, 1.1, 1.1, 1.2, 1.3, 1.3, 1.8, 1.8, 2.2 and 4.6 mg/kg. Total residues were: 0.35, 0.50, 0.54, 0.75, 0.78, 0.84, 0.92, 0.98, 1.2, 1.2, 1.3, 1.4, 1.4, 1.9, 1.9, 2.4 and 4.7 mg/kg.

Using the residue trials for conventional maize crops, the Meeting recommended a maximum residue level of 150 mg/kg (dry weight basis) for maize fodder based on a highest residue of 111 mg/kg (92 mg/kg ÷ 0.83 default dry matter content). The Meeting also estimated a highest residue of 93 mg/kg and a median residue of 24 mg/kg for total residues in maize fodder, both on an as received basis.

The highest and median residues for total residues in maize forage were 4.7 and 1.2 mg/kg respectively, both on an as received basis. The Meeting considered that maize forage is not traded and agreed to withdraw its previous recommendation of 1 mg/kg.

Oat straw

Trials on oats straw were conducted in Canada (GAP 0.9 kg ae/ha PH, PHI growth stage dependent < 30% grain moisture 7-14 days) and the UK (GAP 0.54-1.4 kg ae/ha, < 30% grain moisture PHI 7-14 days). Residues in the trials from Canada conducted according to GAP were 3.5, 27 and 33 mg/kg. Residues in the UK trials were 12, 16, 21, 25, 33, 35, 49 and 64 mg/kg. Total residues where AMAP residues were also measured were: 26, 28, 34, 37 and 50 mg/kg. In all cases AMPA residues were much less than 10% of the glyphosate residue. The Meeting decided to pool the data from the Canada and UK trials to estimate a maximum residue level for glyphosate in oat straw of 100 mg/kg (dry weight basis weight) based on a highest residue of 71 mg/kg (64 mg/kg ÷ 0.90 default dry matter content) and to utilize the glyphosate residues to estimate highest and median residue levels of 64 and 27 mg/kg (as received).

Rye straw

Trials on rye straw were conducted in Denmark (no GAP provided) and were evaluated against the GAP of the UK (GAP 0.54-1.4 kg ae/ha, < 30% grain moisture PHI 7-14 days). No trials matched GAP.

Sorghum fodder and hay

Trials on sorghum were conducted in the USA (GAP of 1.7 kg ae/ha PH grain moisture < 35%, PHI 7 days). Residues of glyphosate in fodder (stover) in trials approximating USA GAP were (n = 10): 2.9, 7.0, 8.2, 16, 16, 21, 28, 29, 30 and 33 mg/kg (total residues 2.9, 7.1, 8.4, 16, 16, 22, 28, 29, 30 and 33 mg/kg). The Meeting recommended a maximum residue level for residues of glyphosate in sorghum fodder of 50 mg/kg (dry weight basis) based on a highest residue of 37.5 mg/kg (33 mg/kg ÷ 0.89 default dry matter content) and total residues highest and median levels of 33 and 19 mg/kg (as received) respectively.

Wheat straw

Trials on wheat were conducted in Belgium (no GAP provided), France (no GAP provided) and the UK (GAP of 0.54-1.4 kg ae/ha, < 30% grain moisture PHI 7 days). Trials conducted in Belgium, France and some UK trials were evaluated against the GAP of the Netherlands (GAP is 0.72-2.2 kg ae/ha, PHI 7-14 days).

Five trials in the UK matched GAP of that country with residues in straw of 6.3, 7.3, 18, 47 and 47 mg/kg (total residues 6.5, 7.8, 18, 48 and 48 mg/kg).

In four trials conducted in the UK according to GAP of the Netherlands residues in straw were 23, 27, 68 and 109 mg/kg. Two trials from Belgium matched GAP of the Netherlands with residues of 103 and 198 mg/kg (total residues 105 and 202 mg/kg). In eighteen trials conducted in France matching the GAP of the Netherlands glyphosate residues in straw were 7.8, 16, 20, 23, 24, 25, 25, 32, 46, 58, 70, 77, 90, 96, 98, 107, 120 and 130 mg/kg (total residues 7.9, 17, 23, 24, 25, 26, 26, 33, 60, 71, 78, 90, 101, 111, 123 and 132 mg/kg).

The Meeting considered the residues from trials conducted according to the GAP of the UK and the Netherlands to be from the same population and to combine the residues for the purposes of estimation of a maximum residue level and STMR. Residues in wheat straw, in rank order were (n = 29): 6.3, 7.3, 7.8, 16, 18, 20, 23, 23, 24, 25, 25, 27, 32, 46, 47, 47, 58, 68, 70, 77, 90, 96, 98, 103, 107, 109, 120, 130 and 198 mg/kg (where measured total residues were 6.5, 7.8, 7.9, 17, 18, 23, 24, 25, 26, 26, 33, 48, 48, 60, 71, 78, 90, 101, 105, 111, 123, 132 and 202 mg/kg). The Meeting recommended a maximum residue level for glyphosate in wheat straw of 300 mg/kg (dry weight basis) based on a highest residue of 225 mg/kg (198 mg/kg ÷ 0.88 default dry matter content) as well as highest and median values for total residues of 202 and 48 mg/kg respectively, both on an as received basis.

The Meeting agreed to withdraw its previous recommendation for straw and fodder (dry) of cereal grains of 100 mg/kg.

Almond hulls

Trials on tree nuts (almonds, pecans, macadamias and walnuts) were conducted in the USA (GAP of 0.43-4.3 kg ae/ha, directed applications, with a PHI of 3 days). No trials matched GAP.

Cotton gin by-products

Trials on conventional cotton were conducted in the USA (GAP of 1.7 kg ae/ha, with a PHI of 7 days). No trials matched GAP for conventional cotton.

Trial data on glyphosate tolerant cotton (GAP 1.7 kg ae/ha, PHI 7 days) was also submitted. Residues of glyphosate in cotton gin by-products were found to be (n = 16): 5.8, 8.6, 16, 19, 21, 31, 34, 36, 37, 41, 42, 67, 70, 84, 91 and 126 mg/kg (total residues 5.9, 8.7, 16, 19, 21, 31, 35, 37, 37, 41, 42, 68, 71, 85, 92 and 128 mg/kg).

The Meeting estimated a highest residue level for cotton by-products of 128 mg/kg and a median level of 37 mg/kg, both on an as received basis.

Rape straw

Trials on conventional rape were conducted in the Sweden (No GAP) and were evaluated against the GAP of the UK (GAP 1.4 kg ae/ha, PHI 14 days). One trial matched GAP of the UK with residues in straw of 30 mg/kg The Meeting considered a single trial insufficient for estimation of a maximum residue level.

Fate of residues duiring processing

The Meeting received processing studies for glyphosate in olives, soya beans, sugar beet, barley, maize, oats, sorghum, wheat, cotton seed, linseed, rape seed, sugarcane, coffee beans and tea leaves, investigating the effects of washing and further processing on incurred residues of glyphosate and AMPA in a range of processing fractions. Only the processing studies relevant to commodities for which maximum residue levels were estimated are reported below.

In trials from the USA conventional and glyphosate tolerant soya beans were processed according to simulated commercial practices into hulls, meal and oil (crude and refined). Median processing factors for hulls, meal and oil prepared according to commercial procedures were 4.5 (n = 3, range 3.8-5.2) for hulls, 1.0 (n = 3, range 0.8-1.0) for meal and < 0.01 (n = 4, range < 0.01-0.02) for crude and refined oil. Median processing factors for total residues were 4.1 (n = 3, range 2.1-5.0) for hulls, 0.89 (n = 3, range 0.83-0.95) for meal and < 0.02 (n = 4, range 0.01-< 0.04) for crude and refined oil.

The Meeting considered that using the median processing factors from the various studies would be appropriate, to reflect the different commercial practices, and estimated soya bean processing factors for glyphosate of 4.5 in hulls, 1.0 in meal and < 0.01 in oil. For total residues, processing factors of 4.1 in hulls, 0.89 in meal and < 0.02 in oil are established. As residues did not concentrate in oil the Meeting did not consider it necessary to recommend a maximum residue level.

Processing studies for barley to beer and distilled spirit were reported however the reported processing factors would exceed the theoretical maximum transfer and the results were not considered further.

In a study on processing (wet and dry milling) of glyphosate tolerant maize, processing factors for aspirated grain dust were 1.6 for both glyphosate and total residues. For bran, the processing factor for dry milling was 1.2 and for wet milling 0.45. Flour and meal had processing factors of 1.1 for glyphosate and total residues while the processing factors for gluten, starch and refined oil were all < 0.05 for glyphosate and < 0.33 for total residues.

Median processing factors for glyphosate in oat processed commodities for hulls, kernels and rolled oats were 1.8 (n = 4, range 1.5-2.3), 0.2 (n = 4, range 0.2-0.2) and 0.2 (n = 4, range 0.1-0.3) respectively.

Sorghum was processed in a study that approximated commercial practices to yield bran, flour, germ, grain dust, grits (medium) and starch. Mean processing factors (n = 2) for glyphosate residues were 5.0, 0.34, 0.02, 4.9, 0.47 and 0.01 respectively for bran, flour, germ, grain dust, grits (medium) and starch.

Four wheat processing studies were made available to the Meeting. Median glyphosate or best estimates of processing factors for bran, whole meal, flour and whole meal bread were 1.7, 0.46, 0.105 and 0.36. As residues concentrate in bran, the Meeting decided to estimate highest anticipated residues in bran based on the highest residue found from trials used to estimate the maximum residue level and mean processing factor. The Meeting confirmed its previous recommendation of a maximum residue level for wheat bran (unprocessed) of 20 mg/kg based on a high glyphosate residue of 9.5×1.7 = 16 mg/kg. The Meeting agreed to withdraw its recommendations for other processed commodities for which residues did not concentrate, i.e., wheat flour and wheat wholemeal.

Data on processing of sugar cane approximating commercial practices were made available to the Meeting. Although the application rates used on the cane processed to bagasse, molasses, raw and refined sugar were higher than the current GAP in the USA the Meeting decided to use the processing data for cane harvested 28 to 35 days after the last application. Median or best estimates of glyphosate processing factors for bagasse, molasses, raw and refined sugar were: 0.275, 8.25, 0.80 and 0.24 respectively. Using the STMR of 0.27 and high residue of 0.97 for sugar cane and the relevant processing factors, the Meeting estimated a maximum residue level for glyphosate of 10 mg/kg for sugar cane molasses together with a median residue of 2.3 mg/kg for total residues.

In a cotton processing study approximating commercial practices glyphosate processing factors for kernels, hulls, meal, crude oil, refined oil and bleached oil were 0.07, 0.33, 0.11, < 0.1, < 0.1 and < 0.1 respectively. As residues of AMPA were all below the limit of quantification, the processing factors for total residues are the same as for glyphosate. Residues did not concentrate in cotton seed oil. The Meeting agreed to withdraw its previous recommendations for the commodities cotton seed oil (crude) and cotton seed oil (edible).

Processing factors and high residue values relevant to maximum residue estimation (glyphosate) and STMR and median residue based on total residues are summarized below.

¹ Processing factors for total residues were not available, however residue data suggests AMPA is either present at less than 10% of the glyphosate residue level or not detected at levels above the limit of quantitation. In these cases the glyphosate processing factors and total residue processing factors would not be significantly different

² Where residues of AMPA were measured they were <LOQ. The Meeting decided to use the larger database of glyphosate processing factors to estimate the median processing factor for total residues

Farm animal dietary burden

The Meeting estimated the farm animal dietary burden of glyphosate residues using the diets in Appendix IX of the FAO Manual (FAO 2002).

The calculation from the MRLs provides the feed levels suitable for animal commodity MRL estimation, while the calculation from feed STMRs is suitable for estimation of animal commodity STMRs. DM is dry matter. The percent dry matter (DM) is taken as 100% where MRLs and STMRs are already expressed on a dry weight.

Calculation of the dietary burden for maximum residue estimation

Calculation of the dietary burden for STMR estimation

The glyphosate dietary burdens for animal commodity MRL and STMR estimation (residue levels in animal feeds expressed on dry weight, figures in brackets are for STMRs) are: beef and dairy cattle 381 (266) ppm, swine 23 (11.5) ppm and poultry 23 (11.4) ppm.

Farm animal feeding studies

The Meeting received information on the residue levels arising in animal tissues and milk when dairy cows were fed rations containing a 9:1 mixture of glyphosate and AMPA at total combined daily dietary levels of 40, 100 and 400 ppm. No residues were detected in milk from animals receiving the highest feed level. Tissue residues in single animals slaughtered after 28 days of feeding treated rations were < 0.05 mg/kg at all feed levels in fat and muscle for both glyphosate and AMPA. Glyphosate residues in liver were 0.06, 0.07 and 0.21 mg/kg for the 40, 100 and 400 ppm feed levels respectively (total residues 0.06, 0.07 and 0.47 mg/kg). In kidney, glyphosate residues were 0.32, 0.82 and 3.3 mg/kg for the three feed levels (total residues 0.42, 1.2 and 4.5 mg/kg). By 28 days after feeding treated rations ceased, residues in tissues were < 0.05 mg/kg in all tissues and milk.

The Meeting also received information on the residue levels arising in tissues when pigs were fed a ration with glyphosate and AMPA in a 9:1 ratio for 28 days at 40, 120 and 400 ppm in the diet. No residues above LOQ were detected in fat at any feed level. Maximum residues in tissue samples were for animals fed at 400 ppm and were liver 0.72 (total residue 1.4) mg/kg, kidney 9.1 (11) mg/kg, muscle 0.06 (0.06) mg/kg and fat < 0.05 (< 0.05) mg/kg. At the 40 ppm feed level, maximum residues were liver 0.06 (total residue 0.06) mg/kg, kidney 0.32 (0.42) mg/kg, muscle < 0.05 (< 0.05) mg/kg and fat < 0.05 (< 0.05) mg/kg. Residues in tissues were < 0.05 mg/kg for all feed levels at 28 days after access to treated feed was stopped.

A residue study on laying hens fed a diet incorporating glyphosate and AMPA in a 9:1 ratio at 40, 120 and 400 ppm for periods of up to 28 days was made available to the Meeting. At the highest feeding levels, maximum glyphosate residues in eggs were 0.12 (total residues 0.16) mg/kg, no residues were detected in eggs at the lowest feeding level. Maximum residues in tissues at the highest feeding level were < 0.05 mg/kg for fat and muscle and 0.61 (total residue 1.1) mg/kg for liver and 4.3 (4.8) mg/kg for kidney. At the lowest feed level of 40 ppm maximum residues in tissues were < 0.05 mg/kg for fat and muscle and 0.06 (total residue 0.06) mg/kg for liver and 0.35 (0.35) mg/kg for kidney

Animal commodity maximum residue levels

The dietary burdens used for maximum residue and STMR estimation for beef and dairy cattle at 381 and 266 ppm are close to the maximum feed level of 400 ppm and this feed level was used to estimate residue levels in milk and cattle tissues. Maximum residues of glyphosate expected in tissues are: fat < 0.05 mg/kg, muscle < 0.05 mg/kg, liver 0.20 mg/kg, kidney 3.1 mg/kg and the mean residue for milk < 0.05 mg/kg. The STMR dietary burden for beef and dairy cattle is 266 ppm. The Meeting estimated STMR values for total residues from the mean total residues obtained at the 400 ppm feeding level. The estimated STMRs were: meat < 0.05 mg/kg, fat < 0.05 mg/kg, kidney 2.9 mg/kg, liver 0.29 mg/kg and milks < 0.05 mg/kg.

¹ Values in parentheses are the estimated residues at the dietary burdens

² Values in square brackets are the actual feeding levels in the transfer study

³ Residue values in parentheses in italics are obtained from the dietary burden, feeding levels in the transfer study and the residues found in the transfer study. Mean is mean animal tissue (or milk) residue in the relevant feeding group. The residues for HR calculations are glyphosate residues while those for STMR calculations are total residues (glyphosate + 1.5×AMPA).

The maximum dietary burden for pigs is 23 ppm. The levels of residues in tissues other than kidney all expected to be < 0.05 mg/kg for both glyphosate and AMPA when fed at this level. Residues in kidney at the 40 ppm feed level were a maximum of 0.6 mg/kg (total residues 0.72 mg/kg, mean total residues 0.43 mg/kg). Residues of glyphosate in kidney of animals fed at 23 ppm in the diet are estimated to be 0.345 mg/kg. Mean total residues in kidney of animals fed at 11.5 ppm in the diet are estimated to be 0.12 mg/kg. The STMR values for pig meat and pig edible offal as estimated to be 0 and 0.12 mg/kg.

The Meeting estimated a maximum residue level of 0.5 mg/kg for pig edible offal. The Meeting also estimated maximum residue levels for meat (from mammals other than marine mammals) of 0.05 (*) mg/kg; edible offal mammalian [except pigs] of 5 mg/kg and milks 0.05 (*) mg/kg. The recommendations replace the previous recommendations of 0.1 (*) mg/kg for cattle meat, 2 mg/kg for cattle edible offal, 0.1 (*) mg/kg for cattle milk, 0.1 (*) mg/kg for pig meat and 1 mg/kg for pig, edible offal.

The maximum dietary burden for poultry is 23 ppm. The levels of glyphosate and AMPA residues in fat, muscle and eggs are all expected to be < 0.05 mg/kg when fed at this level. Residues of glyphosate in liver and kidney at the 40 ppm feed level were 0.06 mg/kg for liver and 0.35 mg/kg for kidney. Mean total residues in liver and kidney at the 40 ppm feed level were 0.055 and 0.31 mg/kg respectively. Residues of glyphosate in liver and kidney of birds fed at 23 ppm in the diet are estimated to be < 0.05 and 0.20 mg/kg. Mean total residues in liver and kidney of birds fed at 11.4 ppm are estimated to be < 0.05 and 0.088 mg/kg.

The Meeting estimated maximum residue levels for poultry meat 0.05 (*) mg/kg; poultry edible offal 0.5 and eggs 0.05 (*) mg/kg. The recommendation for poultry meat and eggs replace the previous recommendation, both 0.1 (*) mg/kg. As no residues are expected at the dietary burden for STMR estimation, the poultry meat and eggs STMRs are zero. The STMR for liver and kidney are estimated to be 0.05 and 0.088 mg/kg respectively.

DIETARY RISK ASSESSMENT

Short-term intake

The 2004 JMPR concluded that it was unnecessary to establish an ARfD for glyphosate. The Meeting therefore concluded that short-term dietary intake of glyphosate residues is unlikely to present a risk to consumers.

Long-term intake

The evaluation of glyphosate has resulted in recommendations for MRLs and STMRs for raw and processed commodities. Consumption data was available for 32 food commodities and were used in the dietary intake calculation. The results are shown in Annex 3.

The International Estimated Daily Intakes for the 5 GEMS/Food regional diets, based on estimated STMRs were in the range 0-1% of the maximum ADI of 1 mg/kg bw for the sum of glyphosate and AMPA (Annex 3). The Meeting concluded that the long-term intake of residues of glyphosate and AMPA from uses that have been considered by the JMPR is unlikely to present a public health concern.