Evaluation of acrylamide and selected parameters in some Turkish coffee brands from the Turkish market

The purpose of this study was to measure acrylamide (AA) levels and selected parameters of different traditional Turkish coffees. AA, 5-hydroxymethylfurfural (HMF), total reducing sugar, protein, pH, moisture, dry matter (DM) as well as ash, caffeine and soluble solids content (SSC) in DM, L*, a*, b* colour parameters of coffee samples were determined and the correlation between AA level and these parameters were investigated. A total of 36 coffee samples (20 Turkish, 8 Dibek and 8 Terebinth coffee) from the Turkish market were examined. A liquid chromatography-tandem mass spectrometry (LC-MS/MS) method was validated for the detection and quantitation of AA in coffee samples. The calibration curve was linear (R² ≥ 0.999) over the range of 30–1000 μg kg^?1. The limit of detection (LOD) and limit of quantification (LOQ) were found as 4.6 μg kg^?1 and 15.5 μg kg^?1, respectively. The amounts of AA in analysed coffee samples were in the range 31.1 ± 0.6 to 323.4 ± 5.4 µg kg^?1. The highest mean AA levels were found in Terebinth coffees (240.3 μg kg^?1) followed by Turkish coffees (204.3 µg kg^?1) and then Dibek coffees (78.6 µg kg^?1). No tested Turkish coffee samples had an AA concentration above the indicative value (450 µg kg^?1) for roast coffee recommended by the European Commission (EC) in 2011. In addition, a strong positive correlation was found between HMF values and AA amounts of selected coffee types.     

Dietary intake assessment of caramel colours and their processing by-products 4-methylimidazole and 2-acetyl-4-tetrahydroxy-butylimidazole for the Chinese population

ABSTRACT

The aim of the study was to assess the dietary intake of caramel colours and their by-products 4-methylimidazole (4-MEI) and 2-acetyl-4-tetrahydroxybutylimidazole (THI) for the Chinese population. Based on the typical and maximum reported use levels of caramel colours in 15 food categories, the dietary intakes of combined and single-class caramel colours of Classes I, III and IV were estimated with the food consumption data from the China National Nutrient and Health Survey. Using the mean values of 4-MEI and THI contents in Class III and Class IV Caramel colour samples, the exposures to 4-MEI and THI from dietary caramel colours were derived. The results showed that the combined and individual average dietary caramel colour intakes for the Chinese population of different age groups were estimated to be 232–60.3 mg kg^?1 bw day^?1 for combined caramels, 5.9–29.2 mg kg^?1 bw day^?1 for Class I, 7.7–29.6 mg kg^?1 bw day^?1 for Class III, 21.2–54.3 mg kg^?1 bw day^?1 for Class IV, which were far below the group acceptable daily intake (ADI) and respective ADIs. The combined intake of 4-MEI from Class III and IV caramel colours was estimated to be 3.8–5.2 μg kg^?1 bw day^?1 on average, and 12.9–27.1 μg kg^?1 bw day^?1 at 95th-97.5th percentile for the general population. The anticipated exposure to THI from Class III caramel colours was estimated to be 0.1–0.3 μg kg^?1 bw day^?1 on average and 0.5–1.7 μg kg^?1 bw day^?1 at 95th–97.5th percentile for the general population. The dietary caramel colours intakes and the exposures to 4-MEI and THI from dietary caramel colour for the Chinese population were considered to be of low health concern based on the present toxicological data. Soy sauce, vinegar and compound seasonings were found to be the main contributors to the dietary intake of caramel colours.     

T-2 and HT-2 toxins in cereals and cereal-based products in South Korea

A total of 214 samples, consisting of brown rice, barley, mixed grains, corn, wheat and wheat flour were analysed for T-2 and HT-2 toxins using high-performance liquid chromatography with fluorescence detection. Recovery and repeatability were 79.9%–107.5% and 4.9%–14.5% for T-2, and 74.0%–106.1% and 5.0%–17.9% for HT-2, respectively. T-2 toxin was detected in 11 (5.1%) of all samples. The highest incidence was found in corn (21.7%) followed by mixed grains and brown rice. Mean of all samples was 1.5–4.1?µg?kg^?1, the maximum level being 41.5?µg?kg^?1 in corn. HT-2 toxin was detected in 126 (58.9%) of all samples, and the mean values were 26.4–59.2?µg?kg^?1. The estimated daily intakes for the sum of T-2 and HT-2 toxins were 2.56, 3.22, 2.53, 0.03, 0.01 and 2.45?ng (kg?bw)^?1?day^?1 in brown rice, barley, mixed grains, corn, wheat and wheat flour, respectively.     

Total mercury in milled rice and brown rice from China and health risk evaluation

Total mercury (THg) levels in 440 pairs of milled rice samples and brown rice samples from 15 major rice grain-producing provinces of China were measured and the associated health risk via rice consumption for different age categories of Chinese population was also assessed. THg contents were measured by a direct mercury analyser and the limit of detection (LOD) was 1.5 μg kg^?1. The THg levels for milled rice samples and brown rice samples varied from non-detected to 17.8 μg kg^?1 and 1.5 to 25.4 μg kg^?1, respectively, with a mean level of 3.4 μg kg^?1 and 4.9 μg kg^?1, respectively. The THg levels in all milled and brown rice samples were generally low, except three brown rice samples having concentrations above the legally set value for cereals (20 μg kg^?1 Hg). THg intakes for different age categories were estimated according to THg content and corresponding rice consumption and the associated health risk was evaluated by the corresponding provisional tolerable weekly intake (PTWI) for THg (5.0 μg kg^?1 bw week^?1), which was established by the Joint FAO/WHO Expert Committee on Food Additives (JECFA). The 50th percentile of the THg intakes via milled rice and brown rice consumption for different age categories was in the range 0.09–0.19 μg kg^?1 bw week^?1 and 0.14–0.27 μg kg^?1 bw week^?1, respectively, well below the PTWI, suggesting that the associated health risk is relatively low. However, the 99.9th percentile of the THg intakes for 2–4-year-old children amounted up to 20.6% of the PTWI (milled rice) and 29.5% of the PTWI (brown rice), which deserves attention.     

Furanic compounds and furfural in different coffee products by headspace liquid-phase micro-extraction followed by gas chromatography–mass spectrometry: survey and effect of brewing procedures

In this study, the levels of furan, 2-methylfuran, 2,5-dimethylfuran, vinyl furan, 2-methoxymethyl-furan and furfural in different coffee products were evaluated. Simultaneous determination of these six furanic compounds was performed by a head space liquid-phase micro-extraction (HS-LPME) method. A total of 67 coffee powder samples were analysed. The effects of boiling and espresso-making procedures on the levels of furanic compounds were investigated. The results showed that different types of coffee samples contained different concentrations of furanic compounds, due to the various processing conditions such as temperature, degree of roasting and fineness of grind. Among the different coffee samples, the highest level of furan (6320 µg kg^?1) was detected in ground coffee, while coffee-mix samples showed the lowest furan concentration (10 µg kg^?1). Levels in brewed coffees indicated that, except for furfural, brewing by an espresso machine caused significant loss of furanic compounds.     

Packaged food intake by British children aged 0 to 6 years

The European Union approach to assessing exposure to chemical migrants from plastic food-contact materials has been to assume an intake of 1 kg of food in contact with a particular material, per 60 kg person per day, which equates to 16.7 g kg^?1 body weight. A food packaging surface area–food mass ratio of 6 dm²/1 kg is assumed, equivalent to 0.1 dm² kg^?1 of body weight. Children might be at increased risk to exposure from migrants as they have higher intakes of food per kg body weight compared with adults. In addition, much of the food marketed for/to children is in small portions and therefore the food-contact material area–food mass ratio is relatively high. To determine if, and how, the European Union model might be modified to ensure specific protection against chemical migration into food marketed for children, data on 4-day food intakes of 297 children aged 0–6 years were collected including information on pack size, pack type and food-contact material area–food mass ratio. The 297 children consumed a total of 1646 kg of food and drink (including tap water), of which 978 kg (59%) was packaged with 67% of this packaged in plastics. Mean intakes of food packaged in plastic ranged from 27 g kg^?1 body weight (for the infants under 1 year) to 51 g kg^?1 body weight (for the 1–4-year-olds). This was higher than the 16.7 g kg^?1 body weight derived from the European Union convention. The mean area of packaging in contact with the food consumed daily per kg body weight were 0.65 dm² kg^?1 for the infants under 1 year, 0.81 dm² kg^?1 for the 1–4-year-olds, and 0.66 dm² kg^?1 for the 4–6-year-olds. All 297 children had intakes that exceeded 0.1 dm² of packaging per kg of body weight assumption.     

Surveys of aflatoxin B1 contamination of retail Turkish foods and of products intended for export between 2007 and 2009

Surveys were carried out between 2007 and 2009 to determine the aflatoxin B₁ content of 3345 commercial Turkish foodstuffs supplied by producers for testing for their own purposes or for export certification. To simplify the reporting of data, foods were categorized as: 1, high sugar products with nuts; 2, nuts and seeds; 3, spices; 4, grain; 5, cocoa products; 6, dried fruit and vegetables; 7, processed cereal products; 8, tea; and 9, baby food and infant formula. Aflatoxin analysis was carried out by high-performance liquid chromatography with fluorescence detection after immunoaffinity column clean-up, with a recoveries ranging from 91% to 99%, depending on the matrix. Of the 3345 samples analysed, 94% contained aflatoxin B₁ below the European Union limit of 2 µg kg^?1, which applies to nuts, dried fruit, and cereals products. The 6% of the 206 contaminated samples were mainly nuts and spices. For pistachios, 24%, 38%, and 42% of the totals of 207, 182, and 24 samples tested for 2007, 2008 and 2009, respectively, were above 2 µg kg^?1, with 50 samples containing aflatoxin B₁ at levels ranging from 10 to 477 µg kg^?1.     

Quantification of ethyl carbamate in soy sauce consumed in Korea and estimated daily intakes by age

To estimate the daily intake of ethyl carbamate (EC, a possible human carcinogen) from soy sauce consumed in Korea, 136 soy sauce samples were collected from various regions and analysed by gas chromatography/selected ion mode mass spectrometry (GC/SIM‐MS). The distribution of EC varied significantly among samples, ranging from not detected to 128.9 µg kg^?1, with the highest level in Japanese‐style soy sauce. Based on individual intakes in the Complementary Report on 1998 National Health and Nutrition Survey and EC contents analysed in the present study, the contribution of soy sauce to EC intake was estimated by multiplying the mean concentration of EC by individual soy sauce intake data. Daily intakes of EC kg^?1 body weight by high consumers were particularly high in age groups 1–2 years (78.59 ng), 3–6 years (86.37 ng), 50–64 years (86.24 ng) and ≥ 65 years (76.86 ng). Based on a benchmark dose confidence limit (BMDL) of 0.3 mg EC kg^?1 body weight day^?1, the margin of exposure (MOE) calculated from the daily intake of EC in soy sauce for high consumers ranged from 3488 to 7317, which is of concern. Daily EC intakes are likely to be exceeded in groups who consume other fermented foods as well as alcoholic beverages. Copyright © 2006 Society of Chemical Industry     

Acrylamide, 5-hydroxymethylfurfural and Nε-carboxymethyl-lysine in coffee substitutes and instant coffees

Sensitive analytical methods were developed and validated for the quantification of acrylamide, N^ε-carboxymethyl-lysine (CML) and 5-hydroxymethylfurfural (HMF) in 24 commercial coffee substitutes (CSs) and 12 instant coffees (ICs). Acrylamide levels varied widely from 200 to 4940 µg kg^–1 with higher levels in CSs. Only two out of 24 CSs had a level of acrylamide above the indicative value set for this food category by the European Commission (4000 µg kg^–1). None of the ICs tested in this study exceeded the indicative value set for this foodstuff (900 µg kg^–1). CML ranged from 0.17 to 47 mg kg^–1 and it increased in proportion to the protein content of the samples. The highest concentrations were found in IC partly due to the relatively high protein content of this food group. HMF was the most abundant neoformed compound (NFC) found in the tested commercial samples. It was found between 0.59 and 13 g kg^–1. Among other food categories IC and CS could appear to be major contributors to the exposure to NFCs if consumed on a daily basis. Further investigations are needed to elucidate the acrylamide formation during processing and to determine the daily intake level of frequent consumers of these products.     

Electrochemical behavior of chlorogenic acid at a boron-doped diamond electrode and estimation of the antioxidant capacity in the coffee samples based on its oxidation peak

Abstract: In this study, an electroanalytical methodology for the determination of chlorogenic acid (CGA) was achieved at a boron‐doped diamond electrode under adsorptive transfer stripping voltammetric conditions. The values obtained for CGA were used to estimate the antioxidant properties of the coffee sample based on CGA oxidation. By using square‐wave stripping mode, the compound yielded a well‐defined voltammetric response at +0.49 V with respect to Ag/AgCl in Britton–Robinson buffer at pH 3.0 (after 120 s accumulations at a fixed potential of 0.40 V). At the optimum experimental conditions, linear calibration curve is obtained within the concentration range of 0.25 to 4.0 μg mL^?1 with the limit of detection 0.049 μg mL^?1. The developed protocol was successfully applied for the analysis of antioxidant capacity in the coffee products such as Turkish coffee and instant coffee.     

QUALITY OF FLAVORED YOGURT CONTAINING ADDED COFFEE AND SUGAR

Yogurts flavored with instant coffee (0.5, 0.7 and 0.9%) and sweetened with sugar (4 or 5%) and a control yogurt (no coffee or sugar) were evaluated for chemical, physical, microbiological and sensory properties. The added ingredients generally had no effect on the chemical, physical and microbiological quality of yogurts at day 1 after preparation as compared to the control. During 15 days at 5–7C, pH and lactic acid bacteria counts decreased (P < 0.05) and titratable acidity increased (P < 0.05) in all samples. Yogurts with 0.5% coffee flavoring and 4 and 5% sugar met Turkish Institute Standards for yogurt sensory quality when evaluated by a trained 10‐member panel. Yogurt flavored with 0.5% coffee and containing 5% sugar had most attributes rated in the “like” category by 50% or more of 51 consumer panelists.     

Natural occurrence of type-B trichothecene mycotoxins in Korean cereal-based products

Type-B trichothecenes (deoxynivalenol (DON), nivalenol (NIV), fusarenone-X (FUS-X), 15-acetyldeoxynivalenol (15ADON), and 3-acetyldeoxynivalenol (3ADON)) were determined in 338 cereal-based products. Detection limit, quantification limit and mean recovery for five toxins were in the ranges 0.7–2.6?µg?kg^?1, 2.1–7.8?µg?kg^?1 and 73–110%, respectively. The range of occurrence and average level in samples were, respectively, 21–88% and 5.2–121.8?µg?kg^?1 for NIV, 10–96% and 1.7–109.5?µg?kg^?1 for DON, 2–39% and 0.4–3.6?µg?kg^?1 for FUS-X, 0–80% and 0–17.3?µg?kg^?1 for 15ADON, and 0–29% and 0–1.5?µg?kg^?1 for 3ADON. Regarding co-occurrence, 64% of samples had more than two type-B trichothecenes. The estimated daily intakes of NIV, DON, FUS-X, 15ADON, and 3ADON were 0.077, 0.048, 0.004, 0.006 and 0.002?µg?kg^?1?bw?day^?1, respectively. These results suggest that current exposure levels do not indicate the possibility of adverse effects, but consideration of the combined exposure of type-B trichothecenes may be required due to the high frequency of co-occurrence.     

Determination of furan levels in commercial samples of baby food from Brazil and preliminary risk assessment

Commercial baby food samples available on the Brazilian market (n = 31) were analysed for furan content using a gas chromatography-mass spectrometry method preceded by solid-phase microextraction. A limit of detection of 0.7 µg kg^?1, a limit of quantitation of 2.4 µg kg^?1, mean recoveries varying from 80% to 107%, and coefficients of variation ranging from 5.6% to 9.4% for repeatability and from 7.4% to 12.4% for within-laboratory reproducibility were obtained during an in-house validation. The levels of furan found in the samples were from not detected to 95.5 µg kg^?1. Samples containing vegetables and meat showed higher furan levels as compared with those containing only fruits. An exposure assessment showed furan intakes up to 2.4 µg kg^?1 body weight day^?1 (99th percentile) for babies fed exclusively with commercial baby foods. Margins of exposure obtained from intakes estimated in this work indicated a potential public health concern.     

Post-harvest practices linked with ochratoxin A contamination of coffee in three provinces of Cordillera Administrative Region,Philippines

One of the emerging concerns in the Cordillera Administrative Region, Philippines is ochratoxin A (OTA) contamination in coffee. During 2015 to 2016, a total of 51 Arabica (Coffea arabica) coffee samples from Benguet province and 71 Robusta (Coffea canephora var. Robusta) coffee samples from the provinces of Ifugao and Kalinga were analysed for OTA contamination. The OTA-producing fungal contaminants during drying and storage of Arabica and Robusta coffee were Aspergillus niger and Aspergillus ochraceus. Ochratoxin A was more commonly detected in Robusta coffee (36.6%) than in Arabica coffee (21.6%). Among the contaminated samples, Robusta coffee cherries in the drying yard had the highest mean OTA level (120.2 μg kg^?1, n = 10) while roasted Robusta coffee beans had the lowest mean level (4.8 μg kg^?1, n = 9). The onset of contamination of Arabica coffee occurred during storage, with a mean OTA level of 46.7 μg kg^?1 (n = 9). Roasted coffee had lower OTA content although five samples had levels >5.0 μg kg^?1. Pearson Chi-square analysis (χ²) and Fisher’s exact test revealed that several post-harvest practices involving non-removal of the husk or hull and mixing of defective coffee were significantly associated with the occurrence of OTA during drying and storage (p < 0.05). No significant associations, however, were identified during roasting. This study suggests that the post-harvest practices in Cordillera Administrative Region should focus on the removal of defective coffee in all stages of post-harvest and rapid reduction of moisture content particularly during drying.     

Survey of fumonisins B1, B2 and B3 in conventional and organic retail corn products in Spain and Italy and estimated dietary exposure

A survey was conducted to determine the occurrence of fumonisin B₁, B₂ and B₃ during 2007 in 186 samples of organic and conventional locally available corn products. Samples included baby food (n = 62), corn flour (11), cornflakes (23), pasta (14), cookies (17) and other corn products (59) were obtained from popular markets of Valencia (Spain) and Perugia (Italy). The analytical method used pressurized liquid extraction and liquid chromatography/electrospray ionization tandem mass spectrometry with a triple quadrupole (QqQ) analyser. Of the 104 Spanish samples, 22% contained levels in the range of 2–449 µg kg^?1, 2–229 µg kg^?1 and 6–105 µg kg^?1 for FB₁, FB₂ and FB₃, respectively, while 19 (23%) of the 82 Italian samples were positive with quantifiable levels between 2–235 µg kg^?1, 3–187 µg kg^?1, and 4–40 µg kg^?1 for fumonisins B₁, B₂ and B₃, respectively. Overall, none of the Italian samples and only one organic baby food sample from a Spanish market was above the maximum permitted levels established by European legislation. Fumonisins were found mostly in corn flour followed by cookies and cornflakes. Eleven samples from Spain and nine samples from Italy were organic products, being contaminated the 72% and 77% of the samples, respectively. Analysis of the results showed that levels of fumonisins in corn products were similar in Italy and Spain. The safety of fumonisin intake through corn products was demonstrated by the calculation of the estimated daily intake of both populations considering organic and conventional products separately, which ranged from 1.7 × 10^?3 to 0.72 µg kg^?1 bw day^?1 and comparing them with the provisional maximum total daily intake (PMTDI) of 2 µg kg^?1 bw day^?1 established by the European Union.     

Simultaneous determination of type A,B and D trichothecenes and their occurrence in cereals and cereal products

A sensitive LC–MS/MS method for the simultaneous determination of type A, B and D trichothecenes in cereals is presented. The limits of detection ranged between 0.1 and 0.7 µg kg^?1 for all analytes. The method was applied to 289 representatively drawn samples of wheat, rye and oat products. Ninety-four percent of the wheat samples (n = 130), 95% of the rye samples (n = 61) and 100% of the oat samples (n = 98) were contaminated with the type A trichothecenes T-2 and HT-2 toxin. Median levels of T-2/HT-2 (sum of the toxins) were 0.91, 0.53 and 8.2 µg kg^?1, respectively. Highest levels were found in wheat bran (24 µg kg^?1), rye kernels (3.1 µg kg^?1) and oat flakes (85 µg kg^?1). All wheat and rye samples and 75% of the oat samples were contaminated with the type B trichothecene deoxynivalenol. Median levels of this toxin were 23, 15 and 0.53 µg kg^?1, respectively. Highest levels were found in wheat bran (1160 µg kg^?1), rye kernels (288 µg kg^?1) and oat flakes (55 µg kg^?1). The type B trichothecene nivalenol was detected in 67% of the wheat samples, in 3% of the rye samples and in 24% of the oat samples with highest levels in wheat bran (96 µg kg^?1), rye kernels (1.8 µg kg^?1) and in oat flakes (17 µg kg^?1), respectively. Levels of other type A and B trichothecenes played a minor role, although the rates of contamination were often high. Neither macrocyclic type D trichothecenes (satratoxin G and H, verrucarin A, roridin A) nor diacetylverrucarol and verrucarol (type A trichothecenes), were detected in any of the samples.     

Assessment of caffeine intake in the Korean population

An improved method for the analysis of caffeine in foods by HPLC was validated by measuring several analytical parameters. The caffeine contents of 1202 products available from Korean markets were analysed. A consumption study was conducted by using data from the Korea National Health and Nutrition Examination Survey (KNHANES), 2010–12, to estimate the caffeine intakes of the Korean population. The mean intakes of caffeine from all sources in the general population and consumers were 67.8 and 102.6 mg day^?1 for all age groups, respectively. The 95th percentile intakes of the general population and consumers were 250.7 and 313.7 mg day^?1, respectively. In those aged 30–49 years, the caffeine intakes of the general population and consumers were highest at 25.5% (101.8 mg kg^?1 day^?1) and 36.6% (0.9 mg kg^?1 day^?1), respectively, compared with the maximum recommended daily intake (400 mg day^?1) for adults. In the general population, the main contributors to the total caffeine intake were carbonated beverage for the younger age groups and coffee for the adults. These data provide a current perspective on caffeine intake in the Korean population.     

Determination of lead and cadmium content in sausages from Iran

The contents of lead and cadmium in five major brands of six types of cooked beef sausages consumed in Iran were determined by a graphite furnace atomic absorption spectrometer (GFAAS) after hydrogen peroxide/nitric acid digestion. The metal content in the samples, expressed in µg?kg^?1 wet weight, varied from 24.0 to 158.7 with an average of 53.5 for lead and from 2.2 to 13.5 with an average of 5.7 for cadmium. The highest lead and cadmium concentrations were obtained from a German sausage (158.7?µg?kg^?1; brand B) and hot dog (13.5?µg?kg^?1; brand D), respectively. The results indicate that the sausages from Iran have concentrations below the permitted levels for these heavy metals. The daily dietary intakes and the percentage contribution of the two considered metals to the provisional tolerable weekly intake (PTWI) were calculated for sausages.     

Influence of roasting and different brewing processes on the ochratoxin A content in coffee determined by high-performance liquid chromatography-fluorescence detection (HPLC-FLD)

A rapid and reliable procedure has been developed for the determination of ochratoxin A (OTA) in green and roasted coffee. The method consists of extraction of the sample with methanol–5% aqueous sodium hydrogen carbonate/1% PEG8000 (20:80), followed by immunoaffinity column (IAC) clean-up and, finally, high-performance liquid chromatography (HPLC) determination with fluorimetric detection. Mean recoveries for green and roasted coffee spiked at different levels ranging from 94 and 105% were obtained. The limit of determination (S/N = 3) was 0.032 ng g^?1 and the precision (within-laboratory relative standard deviation) was 6%. The method described has been used to assess the influence of roasting and different brewing processes on OTA content in commercial lots of green and roasted coffee. The results provided evidence that roasting led to a significant drop on OTA levels (65–100%). Also, the way coffee is prepared affects the OTA content: brewing using a Moka Express (Italian coffee) led to a significant reduction of OTA concentration (50–75%) since hot water stays in contact with coffee for a short time. On the contrary, Turkish coffee-making (infusion for about 10 min) cause poor reduction in OTA.     

A study of the need for fibre by the growing New Zealand white rabbit

New Zealand White rabbits, aged between 5 and 8 weeks, were offered diets based on oatmeal together with up to 500 g kg^?1 of ground oat husk, or 500 g kg^?1 of ground barley straw or 400 g of a purified cellulose. The rabbits gained about 40 g liveweight per day when fed a well-balanced control diet (in which oatmeal, grassmeal, corn oil and fishmeal were the main constituents) and, apart from one occasion, there was no significant reduction in this rate of gain even when the diets offered contained up to 500 g kg^?1 of these fibre sources. Food consumption increased from 80 g day^?1 to 115 g day^?1 as the acid detergent fibre concentration in the diet increased from 39 to 270 g kg^?1. Digestible and metabolisable energy contents of the diets fell as the fibre concentration rose but the rabbits were able to adjust their intakes and maintain their daily metabolisable energy intakes constant at about 1100 kJ. The proportion of fat in the body dry matter fell as the fibre concentration in the diet was raised. A minimum dietary fibre content of 100 g kg^?1 as measured by the acid detergent fibre or crude fibre techniques is suggested for optimum healthy growth.