Determination of carbendazim residues in fruit juices by liquid chromatography-tandem mass spectrometry

In this paper a rapid optimized method for routine analysis of carbendazim residues in fruit juices is reported. The procedure is based on matrix solid-phase dispersion (MSPD) with diatomaceous earth and analysis of the extract by liquid chromatography-tandem mass spectrometry, with electrospray ionization (LC-ESI-MS/MS). In the method optimization, finding of the optimal pH for the extraction of carbendazim from juice was particularly critical. Significant matrix effects were observed, but could be eliminated using matrix-matched standards. High recoveries (82-102%), good repeatability (RSD ≤ 12%) and low limits of detection (0.03 ng ml^-1) and quantification (0.1 ng ml^-1) were achieved with this method.     

Investigation of patulin contamination in apple juice sold in retail outlets in Italy and South Africa

A study of apple juice products sold in Italy and South Africa was initially carried out on 20 samples bought in Cesena, Italy, and Tygerberg in Cape Town, South Africa. The samples were bought at random and analysed for patulin contamination. All 12 of the Italian samples had no detectable levels of patulin, except one, which was just slightly above the lowest regulatory limit of 10 ng ml^-1. On the other hand, five of the eight South African samples were all contaminated with patulin levels above 10 ng ml^-1, with one showing a concentration of 75 ng ml^-1, well above the highest regulated limit of 50 ng ml^-1. This latter result led to a more targeted investigation with 14 samples being purchased in the low-income areas of Tygerberg where the initial samples were sourced. These samples confirmed that there might be a problem of mycotoxin contamination in apple juices products sold to low-income consumers because half of the samples showed patulin contamination of which four had levels well above the acceptable limits. This is the first study in South Africa to look at apple juice products in low-income areas and it points to a need to intervene and introduce quality systems in the supply chain of the manufacture and packaging of apple juice products by independent small business.     

Occurrence of ivermectin in bovine milk from the Brazilian retail market

High-performance liquid chromatography (HPLC) with fluorescence detection was used for the quantification of ivermectin residues in bovine milk intended for human consumption. After liquid-liquid extraction of ivermectin and purification of the extract, the compound was derivatized with 1-methylimidazol in N,N-dimethyl formamide to form a fluorescent derivative, which was separated by HPLC, using reversed-phase C₁₈, with methanol : water (96 : 4 v/v) mobile phase at a flow rate 0.7 ml min^-1. The excitation and emission wavelengths of the fluorescence detector were adjusted at 360 and 470 nm, respectively. The linearity of the method was in the range 10-100 ng ivermectin ml^-1. Based on a sample of 5.0 ml, the limit of detection and the limit of quantification for ivermectin in milk were 0.6 and 2 ng ml^-1, respectively. The recovery rate varied from 76.4 to 87.2%, with an average of 77.9 ± 3.2%, at four fortification levels. The inter-day precision of the method was 13% (n = 5). Of 168 samples analysed, 17.8% contained ivermectin above the limit of quantification. Nevertheless, none of the samples contained ivermectin above the maximum residue limit (10 ng ml^-1) established by the Brazilian Ministry of Agriculture.     

Screening survey of deoxynivalenol in beer from the European market by an enzyme-linked immunosorbent assay

Deoxynivalenol (DON) was analysed in 313 beer samples collected from the European retail market using a commercially available immunoassay kit (enzyme-linked immunosorbent assay, ELISA). The incidence rate was about 87%, while most samples (73%) had contamination levels lower than 20 ng ml^-1. The contamination ranged between 4.0 and 56.7 ng ml^-1, with an average of 13.5 ng ml^-1. A statistically significant correlation between alcohol levels and DON contamination was found, as well as a significant difference between bottom, top and spontaneous fermenting beers. Twenty-seven beer samples were compared using a second ELISA kit and a good correlation was obtained between the two kits (r = 0.93). Although when compared with gas chromatography-mass spectrometry the ELISA tended to overestimate the results, a good correlation (r = 0.94) between the two methods was observed. Monitoring of DON in beer is important considering that DON production is dependent on the weather and that it can contribute significantly to the tolerable daily intake of DON, especially for frequent beer consumers.     

Development of a lateral flow immunoassay strip for screening of sulfamonomethoxine residues

A rapid lateral flow immunoassay screening method has been developed for the determination of sulfamonomethoxine (SMM) residues in swine urine. For this purpose, a specific monoclonal antibody (mAb), SMM4B9 for SMM, was generated and characterized. The mAb showed low cross-reactivity (not larger than 0.3%) to other sulfonamides and other potentially occurring analytes. Based on the competitive immunoassay principle, the strip was developed with the mAb SMM4B9 and applied to the screening of SMM residues. The test strip is made up of a sample pad, a gold-conjugate SMM4B9 reagent pad, a blotted test membrane containing a test line, a control line (a nitrocellulose membrane spotted with SMM-BSA and goat anti-mouse antibody, respectively), and an absorbent pad. The test could be accomplished within 8-10 min. It was shown that the sensitivity of the test strip was as low as 5 ng ml^-1 of SMM and the half of maximal inhibition concentration (IC₅₀) was calculated to be 10.78 ± 0.22 ng ml^-1 by relative optical density. In unaided visual assessment the detection limit of the strip was 15 ng ml^-1. For samples spiked at 20 and 30 ng ml^-1 the coefficient of variation (CV (%)) was between 2.3 and 7.1%. When the test strip was compared with high-performance liquid chromatography (HPLC) analysis for naturally contaminated swine urine samples, the difference in results was less than 6.1%. The data suggest that the method has advantages of high sensitivity, specificity, simplicity and speed of performance, as well as the characteristics of repeatability, reproducibility or accuracy and assurance. Therefore, the test strip is suitable to determine SMM residues in swine urine rapidly and reliably by quantitative, semi-quantitative or qualitative detection.     

Occurrence of aflatoxin M1 in randomly selected North African milk and cheese samples

Forty-nine samples of raw cow's milk and 20 samples of fresh white soft cheese were collected directly from 20 local dairy factories in the north-west of Libya and analysed for the presence of aflatoxin M₁ (AFM₁). The samples were analysed using a high-performance liquid chromatography technique for toxin detection and quantification. Thirty-five of the 49 milk samples (71.4%) showed AFM₁ levels between 0.03 and 3.13 ng ml^-1 milk. Multiple analyses of five milk samples free of AFM₁ artificially contaminated with concentrations of AFM₁ at 0.01, 0.05, 0.1, 1.0 and 3.0 ng ml^-1 showed average recoveries of 66.85, 72.41, 83.29, 97.94 and 98.25%, with coefficients of variations of 3.77, 4.11, 1.57, 1.29 and 0.54%, respectively. Fifteen of 20 white soft cheese samples (75.0%) showed the presence of AFM₁ in concentrations between 0.11 and 0.52 ng g^-1 of cheese. Multiple assays of five cheese samples free of AFM₁ spiked with different concentration of AFM₁ (0.1, 0.5, 1.0 and 3.0 ng g^-1) showed average recoveries of 63.23, 78.14, 83.29 and 88.68%, with coefficients of variation of 1.53, 9.90, 4.87 and 3.79%, respectively. The concentrations of AFM₁ were lower in the cheese products than in the raw milk samples.     

Determination of ochratoxin A in grapes of Greek origin by immunoaffinity and high-performance liquid chromatography

A simple analytical method for ochratoxin A (OTA) determination in grapes is described, using aqueous methanolic extraction, an immunoaffinity column clean-up step and high-performance liquid chromatography with fluorescence detection. Mean recovery was 94% (RSD = 4.0%) with a detection limit of 0.4 ng g^-1 and quantification limit of 1.20 ng g^-1. Repeatability (r) and reproducibility (R) were 1.17 and 1.34, respectively. OTA determinations were applied to 50 grape samples (23 different varieties) originating from representative regions of Greece. Results showed the presence of OTA in 86% of samples tested (n = 50). Traces were found in 56% of samples but OTA was not detectable in 14% of samples. Traces were also found in 4% of red, organically grown samples. The most contaminated were three samples of red grapes, two from Central Greece (2.69 and 1.41 ng g^-1), both table and wine-making grapes. The third sample (1.46 ng g^-1), originating from the island of Samos, was used only in wine-making. Mean (1.06 ng g^-1) and median (0.76 ng g^-1) OTA concentrations in red grapes were slightly higher compared to the mean (0.82 ng g^-1) and median (0.65 ng g^-1) concentrations in white grape samples. The study shows that the potential risk for a person of 60 kg ranged from 0.9 to 9 ng kg^-1 bw day^-1 and is dependent on the quantity of grapes consumed daily.     

N-methyl carbamate pesticide residues in conventional and organic infant foods available on the Canadian retail market, 2001-03

Seven parent N-methyl carbamate insecticides, in addition to two transformation products of aldicarb (aldicarb sulfoxide and aldicarb sulfone), and a single transformation product of carbofuran (3-hydroxycarbofuran) were measured in infant and junior foods available on the Canadian retail market between 2001 and 2003. Carbaryl and methomyl were the only analytes present at levels above the limits of detection in juice, cereals, fruit, vegetables or meat samples analysed. Carbaryl was the most frequently (7.6%) detected compound and concentrations ranged from 0.6 to 18 ng g^-1. Detectable levels of carbaryl were most frequently found in foods prepared with fruit. Methomyl was detected (0.8 ng g^-1) in one chicken with broth sample analysed in the present study. In all cases, the concentrations observed were orders of magnitude below the maximum residue limits established for these compounds in the corresponding raw food commodities in Canada (100-10 000 ng g^-1). Dietary intakes of carbaryl and methomyl based on the consumption of infant foods tested ranged between 0.2-343 and 0.4-2.0 ng kg^-1 body weight day^-1, respectively.     

Determination of ochratoxin A in virgin olive oils of Greek origin by immunoaffinity column clean-up and high-performance liquid chromatography

An improved specific analytical method for ochratoxin A (OTA) determination in olive oil is described, using a methanolic-aqueous extraction, an immunoaffinity column clean up step and high-pressure liquid chromatography with fluorescence detection. The mean recovery was found at 108% (relative standard deviation, RSD = 4.7%) and the detection limit (DL) was estimated at 4.6 ng kg^-1. Along with OTA, aflatoxin B₁ (AFB₁) was determined using the same extract. The recovery factor was 84.8% (RSD = 17.8%) and the DL was 56 ng kg^-1 olive oil. Both determinations were applied in 50 samples of olive oil originated from representative regions of Greece. Results revealed the presence of OTA in 88% of samples tested (n = 44, mean 267 ng kg^-1). Among them, 10 were contaminated with more than 500 ng kg^-1 (median 568 ng kg^-1), 10 with 200-500 ng kg^-1 (median 260 ng kg^-1), 15 with 100-200 ng kg^-1 (median 140 ng kg^-1), nine with DL-100 (median 60 ng kg^-1) and in six samples, OTA was not detectable. Interestingly, most contaminated samples were from Southern Greece. Results of AFB₁ determination showed the presence of aflatoxin B₁ (60 ng kg^-1) in only one olive oil sample also from Southern Greece. The levels of OTA found in Greek olive oil were relatively low as compared with other commodities such as cereals or wine reported in the literature.     

Multi-residue method for the determination of 450 pesticide residues in honey, fruit juice and wine by double-cartridge solid-phase extraction/gas chromatography-mass spectrometry and liquid chromatography-tandem mass spectrometry

A multi-residue method was developed for the determination of 450 pesticide residues in honey, fruit juice and wine using double-cartridge solid-phase extraction (SPE), gas chromatography-mass spectrometry (GC-MS) and liquid chromatography-tandem mass spectrometry (LC-MS-MS). The method development was based on an appraisal of the characteristics of GC-MS and LC-MS-MS for 654 pesticides as well as the efficiency of extraction and purification from honey, fruit juice and wine. Samples were first diluted with water plus acetone, then extracted with portions of dichloromethane. The extracts were concentrated and cleaned up with graphitized carbon black and aminopropyl cartridges stacked in tandem. Pesticides were eluted with acetonitrile + toluene, and the eluates were concentrated. For 383 pesticides, the eluate was extracted with hexane twice and internal standard solution was added prior to GC-MS determination. For 67 pesticides, extraction was with methanol prior to LC-MS-MS determination. The limit of detection for the method was between 1.0 and 300 ng g^-1 depending on each pesticide analyte. At the three fortification levels of 2.0-3000 ng g^-1, the average recovery rates were between 59 and 123%, among which 413 pesticides (92% of the 450) had recovery rates of 70-120% and 35 pesticides (8% of the 450) had recovery rates of 59-70%. There were 437 pesticides (97% of the 450) with a relative standard deviation below 25%; there were 13 varieties (3% of the 450) between 25.0 and 30.4%.     

Use of polarography as a quality-control method for determining diacetyl in citrus and vegetable juices, yoghurt and butter

The determination of diacetyl permits the detection of microbial growth in the processing of citrus fruit before the appearance of other organoleptic, chemical or microbiological changes. It also makes it possible to detect a break in the cold chain during distribution and sale. The study proposed a polarographic method for the determination of diacetyl that allowed routine analysis with the aim of detecting possible contamination in the citrus juice manufacturing chain (orange and orange-carrot). The analytical performance of the method in terms of a linearity from 0 to 960 μg ml^-1, a recovery of 97 to 98%, a precision of 3.2 to 4.8%, and a sensitivity of 0.2 ng ml^-1 for juices indicated the robustness of the method. The same method was also applied to butter and yoghurt with detection limits of and 0.4 ng g^-1.     

Analysis of heat-processed corn foods for fumonisins and bound fumonisins

Thirty retail samples of heat-processed corn foods, i.e. corn flakes, corn-based breakfast cereals, tortilla chips and corn chips, were analysed for fumonisins — fumonisin B₁ (FB₁), fumonisin B₂ (FB₂) and hydrolysed FB₁ (HFB₁) — as well as for protein- and total-bound FB₁. Bound (hidden) fumonisins cannot be detected by conventional analysis. Improved methods for the determination of bound FB₁ were developed. The protein-bound FB₁ was extracted with 1% sodium dodecylsulfate (SDS) solution. The SDS, which interfered with high-performance liquid chromatography (HPLC) analysis, was then separated from protein-bound FB₁ by complexing with methylene blue followed by solvent extraction and hydrolysis with 2 N KOH. To measure total-bound FB₁, the sample itself was hydrolysed with KOH. In both cases, clean-up was accomplished on an OASIS polymeric solid-phase extraction column and the bound fumonisins were determined by HPLC measurement of HFB₁. Fourteen of 15 samples of corn flakes and other corn-based breakfast cereals analysed contained detectable levels of FB₁ with a mean in positive samples of 67 ng g^-1 (13-237 ng g^-1). Two samples also had detectable levels of FB₂ (21-23 ng g^-1). Bound FB₁ was found in all samples; the mean protein-bound FB₁ measured was 58 ng g^-1 (22-176 ng g^-1) and the mean total-bound FB₁ measured was 106 ng g^-1 (28-418 ng g^-1), reported as FB₁ equivalents after correction for recoveries of HFB₁. There was an average of about 1.3 times more FB₁ in the bound form compared with extractable FB₁, and this was about twice as much as protein-bound FB_1. Seven of the 15 samples of alkali-processed corn-based foods, such as tortilla chips and corn chips, contained FB₁ and three contained HFB₁ with means in measurable positive samples of 78 (48-134) and 29 (13-47) ng g^-1, respectively. Five of these alkali-processed corn foods contained bound FB₁; the mean measurable protein-bound FB₁ was 42 ng g^-1 (39-46 ng g^-1) and the mean measurable total-bound FB₁ was 100 ng g^-1(54-209 ng g^-1). HFB₁ derived from bound FB₁ in selected samples was confirmed by HPLC with mass spectrometry (MS).     

Analysis of veterinary drug residues in fish and shrimp composites collected during the Canadian Total Diet Study, 1993-2004

Thirty shrimp, marine fish, freshwater fish, and canned fish composite samples collected and prepared as part of the Canadian Total Diet Study were analysed for 39 different veterinary drug residues. The analyses were undertaken to obtain baseline data that could be used to estimate the dietary exposure of Canadians to these residues. The most frequently observed residue was AOZ (four out of 30 samples), the metabolite of furazolidone, at a range of 0.50 to 2.0 ng g^-1 wet weight. Other residues detected included enrofloxacin (three samples; 0.3-0.73 ng g^-1), leucomalachite green (three samples; 0.73-1.2 ng g^-1), oxolinic acid (two samples; 0.3-4.3 ng g^-1), AMOZ (the metabolite of furaltadone; one sample; 0.40 ng g^-1), chloramphenicol (one sample; 0.40 ng g^-1), and SEM (the metabolite of nitrofurazone; one sample; 0.8 ng g^-1). The results of this survey indicate that Canadians are exposed to low ng g^-1 concentrations of some banned and unapproved veterinary drug residues via the consumption of certain fish and shrimp.     

Occurrence of fumonisins B1 and B2 in asparagus from Shandong province, P.R. China

Thirty samples of asparagus spears were collected from the fields in Shandong province, China, in July 2004, and were analysed for the occurrence of fumonisins B₁ and B₂ (FB₁ and FB₂) by HPLC coupled with electrospray ionization tandem mass spectrometry. Twenty-four samples (80%) contained fumonisins, ranging from 24 to 670 ng g^-1 (average 123 ng g^-1) and 17 to 138 ng g^-1(average 35 ng g^-1) for FB₁ and FB₂, respectively. The total amount of fumonisins (FB₁ and FB₂) in all samples ranged from 47 to 714 ng g^-1 (average 158 ng g^-1) (based on dry weight). This is the first report on the natural occurrence of FB₁ and FB₂ in asparagus spears in China.     

Survey of aflatoxins in beer sold in Canada

Between March 1998 and March 2002, 304 samples of domestic (Canadian) and imported beers from 36 countries were picked up for the determination of aflatoxins B₁, B₂, G₁ and G₂. Twelve samples were positive with aflatoxins greater than the limit of quantitation (LOQ) (aflatoxin B₁, 4.4 ng l^-1; aflatoxin B₂, 3.4 ng l^-1; aflatoxin G₁, 11.2 ng l^-1; and aflatoxin G₂, 6.2 ng l^-1). Five samples from Mexico, two samples from Spain and one from Portugal contained aflatoxin B₁. Four samples from India contained aflatoxins B₁ and B₂. The remaining samples contained less than the LOQ for aflatoxins B₁, B₂, G₁ and G₂. The analytical method for this survey was based on that of Scott and Lawrence (Scott PM, Lawrence GA. 1997. Determination of aflatoxins in beer. Journal of AOAC International 80:1229-1234.). Aflatoxins B₁, B₂, G₁ and G₂ were determined at parts per trillion (ng l^-1) levels in beer by immunoaffinity column cleanup followed by derivatization with trifluoroacetic acid and reversed-phase liquid chromatography with fluorescence detection.     

Effect of cooking on enrofloxacin residues in chicken tissue

The aim of this study was to determine the effect of different cooking processes (microwaving, roasting, boiling, grilling and frying) on naturally incurred enrofloxacin residues in chicken muscle. Enrofloxacin and its metabolite, ciprofloxacin, were analysed using a validated LC-MS method with limits of detection (LOD) and quantification (LOQ), respectively, of 2 and 5 ng g^-1 quinolones in muscle samples. The method was shown to be linear over the range 5-500 ng g^-1. Mean intra-day relative standard deviation (RSD) at a concentration of 50 ng g^-1 (n = 6) was 6%; inter-day RSD was 12%. A recovery study demonstrated that 65-101%, of the drug and metabolite could be recovered from the tissue. The RSD with naturally incurred roasted chicken breast was 9.18% at a concentration of 11 ± 1.01 ng g^-1 (n = 6). In water, enrofloxacin remained stable for 3 h when heated at 100°C. It was concluded that residue data from raw tissue are valid for estimation of consumer exposure to this drug, as well as the ADI calculations because cooking procedures did not affect enrofloxacin residues, which remained stable during heating. However, there was an apparent decrease in quinolone concentration in tissue because some was lost by exudation into the liquid used for cooking. Conversely, for a cooking procedure with water loss, there was an apparent increase in residue concentration.     

Dioxins and polychlorinated biphenyls (PCBs) in fish oil dietary supplements: Occurrence and human exposure in the UK

Commercially available fish oil supplements sourced from retail outlets in the UK, as well as by mail order, were surveyed in 2000-02 for dioxin (PCDD/Fs) and polychlorinated biphenyl (PCB) content. Sampled products were representative of market share. The WHO-TEQ values for these products ranged from 0.18 to 8.4 ng kg^-1 for ΣPCDD/F and from 1.1 to 41 ng kg^-1 for Σ dioxin-like PCBs. The results suggest a downward trend in the levels of dioxins in fish oil supplements over the last decade, since levels for similar products ranged from 0.3 to 10 ng kg^-1 for ΣPCDD/F WHO-TEQ in 1996. Levels of ICES (International Council for the Exploration of the Seas) 7 PCBs in the current study ranged from 8.3 to 267 µg kg^-1. Subsequent to this survey, European Union legislation has been introduced that includes a maximum limit of 2 ng kg^-1 WHO-TEQ for dioxins in fish oil products for human consumption. Twelve of the 33 products reported here would have exceeded this limit. Negotiations are in progress to incorporate dioxin-like PCBs into the European Union regulations. When manufacturer-recommended doses were applied to the observed levels, the estimated upper bound human exposure to dioxins and dioxin-like PCBs from dietary intake of these products ranged from 0.02 to 7.1 pg WHO-TEQ kg^-1 body weight day^-1 for adults and from 0.02 to 10 pg WHO-TEQ kg^-1 body weight day^-1 for schoolchildren. This level rises to 1.8-8.9 pg WHO-TEQ kg^-1 body weight day^-1 for adults and 1.4-14 pg WHO-TEQ kg^-1 body weight day^-1 for schoolchildren when combined with the average exposure from the whole diet in 1997. Again, subsequent to this survey, the Committee on Toxicity of Chemicals in Food, Consumer Products and the Environment (COT) revised the UK tolerable daily intake (TDI) for mixtures of dioxins and dioxin-like PCBs from 10 to 2 pg WHO-TEQ kg^-1 body weight day^-1. This is in line with the tolerable weekly intake (TWI) of 14 pg WHO-TEQ kg^-1 body weight set by the Scientific Committee on Food (SCF).     

Variability in captan residues in apples from a Canadian orchard

Apple trees in an orchard in Quebec, Canada were treated, following label directions, with the fungicide captan (1,2,3,6-tetrahydro-N-(trichloromethylthio)phthalimide) during the 2003 agricultural season. A total of 142 apples from three rows of trees were selected for determination of captan by GC/MS. Individual apples were found to contain captan levels ranging from 16.9 to 6350 ng g^-1. Only two individual apple samples exceeded the Canadian maximum residue limit (5000 ng g^-1) for captan in apples. Six composite samples, comprising half portions of eight individual apples, were analysed from each of the three experimental rows. Composite samples ranged in concentration from 166 to 2620 ng g^-1. The greatest uncertainty associated with the measured concentrations was due to variability among apples rather than the measurement of residue levels.     

Determination of triazamate in apples, peas and Brussels sprouts using high-performance liquid chromatography with tandem mass spectrometry

A rapid method was developed to allow the analysis of triazamate and the hydrolysis product triazamate acid in apples, peas and Brussels sprouts. The procedure was based on acidified methanol extraction with quantification by liquid chromatography coupled with tandem mass spectrometry detection. The detection limit, based on the lowest calibration level, was 0.005 mg kg^-1. The method was validated at 0.05 and 0.01 mg kg^-1, and was used for screening of triazamate acid in the surveillance of apples as part of the UK pesticide monitoring programme. Calibrations were linear over the range 0.004-0.080 µg ml^-1 (equivalent to 0.005-0.11 mg kg^-1), with a minimum correlation coefficient of 0.95. The detection response showed some matrix-dependent variation. Mean recoveries were in the range 70-108% with associated per cent coefficient of variations of less than 20%.     

Distribution and elimination of fumonisin analogues in weaned piglets after oral administration of Fusarium verticillioides fungal culture

The distribution and elimination of fumonisins after oral administration of 50 mg FB₁, 20 mg FB₂ and 5 mg FB₃ per animal day^-1 for 22 days was studied in weaned barrows. At the end of the trial, the lung, heart, liver, kidney, spleen, brain, serum, bile, muscle, fat, urine and faeces samples were collected and their content of fumonisins (FB₁, FB₂) determined by LC-MS. The highest FB₁ concentrations were found in the liver (99.4 ± 37.5 ng g^-1) and kidneys (30.6 ± 10.1 ng g^-1), whilst the highest average amount of FB₂ was in the liver (1.4 ± 2.3 ng g^-1) and fat (2.6 ng g^-1 ± 4.8) samples. Comparing the FB₁/FB₂ ratio in different organs (19/1), it was found that the ratio in the abdominal and subcutaneous fat samples (4/1) was markedly different from those in all other tissues, namely the relative proportion of FB₂ was higher in latter cases. Of the total quantity of FB₁, the 13% taken up during 5 days was excreted unchanged with the faeces and urine. On average, in the urine and faeces, FB₁ was detected in nine- and 14-fold quantities, as compared with FB₂.