1. The effects of unhydrogenated and partially-hydrogenated marine and rapeseed oils on the relative content of n-6 fatty acids in rat liver phospholipids were studied by gas–liquid chromatography.

2. The type of dietary oil strongly affected the pattern of n-6 fatty acids in the liver phospholipids even after a relatively short feeding period (3 weeks).

3. With diets deficient in linoleic acid, accumulation of the 20:3 n-9 fatty acid was observed in the animals receiving partially hydrogenated oils, but not in the animals receiving the unhydrogenated marine oil which contained relatively high amounts of polyunsaturated fatty acids of the n-3 family.

4. The results presented are in line with reports in the literature that dietary trans fatty acids may inhibit the desaturation–elongation enzyme systems of rat liver. The observed effects did not appear to be directly correlated to the dietary content of very-long-chain monounsaturated fatty acids.

1. The effects of differing degrees of experimental protein-energy malnutrition on the response of myofibrillar protein turnover rates to administration of corticosteroid has been studied in two experiments on rats. The basal control diet, offered ad lib. in each case, contained 40 g protein/kg, and other groups received diets containing 62.5, 95 or 220 g protein/kg at 0.67, 1 or 1.5 times the level of the control energy intake.

2. Daily administration of 25 or 30 mg corticosterone/kg body-weight after 18 d pre-feeding caused an increase in plasma protein, glucose and insulin concentrations, but a decrease in the corticosterone: insulin values. Liver size and protein content increased, as did the fractional excretion of dietary nitrogen as urea-N in all treated groups. However, whereas a fall in food intake and body-weight occurred in one experiment the reverse occurred in the other.

3. NT-Methylhistidine excretion was 12% lower for rats receiving 40 v. 220 g protein/kg diet and excretion was increased by only 57 v. 90% respectively, when the two groups of rats were given 30 mg corticosterone/kg per d. Rats which received 25 mg corticosterone/kg per d and up to 95 g protein/kg diet increased excretion of NT-methylhistidine by an average 35%.

4. The fractional degradation rate of myofibrillar protein (kd) was reduced by about 10% by the low-protein diet from 3.1 to 2.8%/d. During corticosterone treatment the increment in kd for rats on this diet was only 60% of that for rats receiving the 220 g protein/kg diet, i.e. an increase of 1.8 v. 3.0%/d. Energy restriction further reduced kd during low-protein intake but did not affect the response to the corticosterone. Variations in dietary protein from 40 to 95 g/kg had little effect on the increase in kd during steroid treatment. The effect of corticosterone on calculated synthesis rates (kg) differed markedly between experiments. While kg fell by 50–65% in rats which lost weight on treatment, it rose by up to 60% in rats where carcass non-collagen-protein accretion remained unchanged or increased, despite an increase in kd

5. Protein deficiency decreases the catabolic response to glucocorticoid, but the net metabolic response appears crucially dependent on changes in food intake or the stage of growth of the rat or both. A net anabolic response with increased fractional rates of myofibrillar protein breakdown, synthesis and accretion was observed in growing rats fed on relatively-low-protein diets and given 25 mg corticosterone/kg per d. This novel finding indicates that a particular role for cortisol in the adaptation to protein-energy malnutrition by humans should be ascribed only with caution.

1. Male, Wistar rats (50–74 g) were given a semi-purified diet containing cholesterol (10 g/kg diet) for 4 weeks, groups of control and experimental animals killed, and the remainder of the cholesterol-fed animals given either a semi-purified cholesterol-free diet without a fibre source or a similar diet with pectin (50 g/kg diet) for 8 weeks. Animals were killed at 2-week intervals and serum and liver cholesterol and triglycerides, faecal neutral steroids and acidic steroids measured.

2. Animals given pectin had significantly lower serum cholesterol levels than animals given the basal diet after 4 and 6 weeks.

3. Both experimental groups efficiently decreased their liver cholesterol levels during regression.

4. Bile acid excretion was elevated in both groups, especially in rats receiving pectin, but neutral steroid excretion was unchanged.

5. The increase in bile acid excretion was primarily in the form of chenodeoxycholic acid and its metabolites.

6. This increased bile acid excretion is postulated to result from stimulation of the normal mechanism for increased steroid excretion in the rat, i.e. a change in the value of cholic: chenodeoxycholic acid or from a mechanism not yet elucidated.

1. Between September 1979 and July 1981, 405 northern-English children, initially aged 11–12 years, each recorded all food and drink consumed on five occasions on each of three consecutive days. Sugars and nutrient intakes were calculated using food tables.

2. The average total sugars intake was 118 g/d or 21% of the energy in take and 43% of the carbohydrate intake; these percentages were similar for both sexes, all social class groups and surveys.

3. Sugars were found to be derived from a variety of food sources with confectionery being the single largest source.

4. Sugars were fairly evenly consumed over the average day with a high amount in foods eaten between meals. Snacks accounted for 65% of the sugars intake but only 46% of the energy intake.

5. Sugars intake and snacking therefore seem to have been major components of the eating habits of these children. If dietary changes towards lower sugars intakes are to be achieved a vigorous, informed health education campaign is necessary together with effective labelling of manufactured foods, the development of new products and the modification of some existing ones.

1. Effects of a feeding programme on maternal weight, triceps and subscapular skinfolds during pregnancy and lactation were assessed in a marginally undernourished population of rural Taiwan. Mothers participated during two pregnancy and lactation periods. At 3 weeks after the delivery of the first infant, one group of 114 mothers began consuming a nutrient-dense supplement (A), while another group of 111 received a placebo (B).

2. There was no significant A–B difference in pregnancy weight gain (A 7.52 kg, B 7.75 kg) or in mean maternal weight, triceps or subscapular skinfolds at any time during pregnancy or lactation. Despite a general trend toward moderate weight loss during lactation, one-third of mothers in both groups gained weight. Highly significant increases in weight and skinfold thicknesses from one lactation period to the next characterized both groups.

3. The absence of demonstrable supplement-effects on maternal anthropometry concomitant with increments in infant birth weight supports the notion that infants can benefit from maternal supplementation without changes in maternal nutritional status.

1. The effects of incorporating Fybogel (3·5 and 7 g doses), Metamucil (7 g) or guar gum (2·5 and 14·5 g doses) in a drink containing 50 g glucose on plasma glucose, plasma insulin and gastric emptying were studied in thirty-eight normal volunteers. In addition, the effects of Fybogel (7 g) on glucose tolerance, plasma insulin and gastric emptying were measured in fourteen non-insulin-dependent diabetics.

2. Both doses of guar gum significantly lowered plasma glucose and plasma insulin responses to the oral glucose load in normal subjects, although 14·5 g guar gum did not delay the half-time for gastric emptying.

3. Neither Fybogel nor Metamucil had significant effects on plasma glucose responses in normal subjects. In addition, Fybogel (at either dose) had no significant effects on plasma insulin levels, or on gastric emptying in normal subjects or on plasma glucose and insulin responses in diabetic patients.

4. The viscosity of ispaghula solutions (Fybogel) was lower than that of guar gum solutions.

1. Male, Sprague-Dawley (Charles-River) rats, of initial weight 272 g, were given a powdered stock diet (T1) ad lib. force-fed a synthetic diet (T2) or offered a range of palatable foods in conjunction with the powdered stock diet (T3) or a similar diet supplemented with certain minerals and vitamins (T4).

2. Metabolizable energy (ME) intake (kJ/d) averaged 303, 453, 402 and 383 for T1, T2, T3 and T4 respectively and corresponding weight gains were 5.5, 6.9, 8.2 and 7.9 g/d and were significantly different (P < 0.001).

3. The intakes of T3 and T4 rats ranged from 10 to 60% above the mean value for T1.

4. Crude protein (CP; nitrogen × 6.25) retentions were similar for T1, T3 and T4 rats and significantly lower (P < 0.01) for T2 rats. Fat retentions were 1.1, 4.1, 2.9 and 2.4 g/d for T1 to T4 respectively (P < 0.001).

5. The energy contents of the gain (MJ/kg) were 12.7, 26.0, 16.7 and 14.9 for T1 to T4 respectively (P < 0.001) and energy retentions (kJ/d) were 70, 179, 139 and 117 respectively (P < 0.001).

6. A linear regression of energy retention (ER) on ME yielded a slope of 0.78 and a mean energy requirement for zero balance of 510 kJ/kg body-weight0.75.

7. These results are in conflict with reports of ‘diet-induced thermogenesis’ in ‘cafeteria’-fed rats.

1. Newborn rabbits were over-fed by encouraging them to suck from two lactating does. These double-fed animals were compared with single-fed litter-mate controls.

2. Single- and double-fed rabbits were compared with respect to body-weight, bone length and weights of heart, kidney, lung, liver, spleen, stomach, brain and adipose lobes at ages 7, 14 and 21 d and at 25 weeks. Body composition with respect to fat, protein and water was also analysed in animals killed during the first month of life.

3. Double feeding had a profound effect on body-weight, such that by 21 d of age these animals weighed 65% more than controls. Longitudinal growth was also increased, but to a lesser extent. Organ weights were largely determined by body-weight with the exception of the brain which was unaffected by double feeding and the heart and adipose lobes which were represented to excess. Body composition studies demonstrated a marked increase in body fat and a decrease in body water as a percentage of body-weight in the double-fed.

4. When mature, double-fed rabbits weighed a little more than controls and had accumulated significantly more fat. No other difference between them was demonstrated.

5. Other experiments on rats, mice and pigs, along with these experiments on rabbits, suggest that young mammals respond to over-feeding with a limited increase in real growth associated with excess deposition of adipose tissue. Studies on infants of diabetic mothers have demonstrated that the human fetus responds to over-nutrition in a very similar way. Therefore we should be cautious about over-feeding human infants, especially premature ones, as it seems unlikely that excessive deposition of fat will be beneficial.

1. To investigate the role of dietary components in the hyperinsulinaemia of the obese hyperglycaemic (ob/ob) syndrome, plasma insulin responses to fasting and refeeding were examined in Aston ob/ob mice supplied with standard diet, non-digestible-carbohydrate test food, and isoenergetic test foods from which either carbohydrate, protein or fat was omitted.

2. During fasting, plasma insulin concentrations fell more rapidly and to a greater extent than plasma glucose. Refeeding the standard diet raised insulin concentrations above normal, associated with a 25% compensatory increase in food intake over 24 h.

3. Test foods supplied to previously fed or fasted mice produced glucose responses consistent with the available carbohydrate content. Carbohydrate-free food (protein and fat) provided a small insulinotropic stimulus; the effect of protein-free food (carbohydrate and fat) was greater; and the combination of carbohydrate with protein (fat-free food) evoked a marked insulin response. In contrast, insulin concentrations declined in mice given the non-digestible-carbohydrate food.

4. Consumption of the standard diet was increased after 24 h feeding non-digestible-carbohydrate food, but was unaffected by a 30 h fast initiated 54 h previously.

5. These results demonstrate that hyperinsulinaemia in ob/ob mice is not merely triggered by the ingestion of bulk, but depends on the type of nutrient ingested. Dietary carbohydrate appears to be the major stimulus to the hyperinsulinaemia, with an important augmentation in the presence of protein. Since direct glucose stimulation of insulin release is defective in ob/ob mice, the hyperinsulinaemia must be mediated by increased activity of the enteroinsular axis.

1. The volume and number of adipocytes were measured in fourteen anatomical sites of adult guinea-pigs kept in small cages and fed ad lib. kept in small cages and on restricted diet or fed ad lib. and exercised.

2. In the sedentary ad lib.-fed animals, there was no significant correlation between percentage body-weight as adipose tissue, as determined by direct dissection, and mean adipocyte volume based on samples from many different sites. The correlation was significant, though not high, for sedentary, restricted-diet animals and for exercised specimens.

3. The correlations between the volume of adipocytes from left–right pairs of sites, and from sites around the same limb, were highly significant under all conditions studied. The correlation between the volume of adipocytes from sites other than left–right pairs of sites was weaker and in some cases statistically insignificant in sedentary ad lib.-fed guinea-pigs. The volume of adipocytes from sites in the groin region, the mesenteries and medial to the trapezius muscle failed to correlate in many cases with the volume of adipocytes from other sites sampled.

4. The number of adipocytes at each site was similar in the exercised and sedentary ad lib.-fed animals. The restricted-diet, sedentary group had fewer adipocytes at all sites studied except the omental and mesenteric fat mass and the groin sites.

5. It is suggested that moderate regular exercise or fasting gives rise to closer coordination between adipocytes at different sites because central factors regulating adipocyte volume become more prominent than local factors.

1. Two intermuscular adipose depots, the popliteal mass (POP) and the fat under the gluteus superficialis muscle (GLS) were studied in guinea-pigs maintained under three different regimes of diet and exercise.

2. Eleven different sites in POP and four sites in GLS were defined using associated muscle and fascia as ‘landmarks’. Samples of twenty to twenty-five adipocytes from each site were removed and their mean volume measured; a consistent pattern of distribution of larger and smaller adipocytes within POP and GLS was found in all animals.

3. The correlation between the volume of adipocytes from sites between the same pairs of muscles was substantially greater than the correlation between the volume of adipocytes from comparable sites between different muscles, whether the sites were within the same fat mass (POP) or in different fat masses (POP and GLS).

4. The volume of adipocytes from the edge of the fat mass correlated more closely with those from the edge of the same or a synergistic muscle than with the volume of cells from the centre sites or edge sites from an interface with another muscle. These effects were most pronounced in exercised animals and weak or insignificant in sedentary ad lib.-fed animals.

5. The volume of adipocytes from edge sites correlated with the gross weight of the adjacent muscles in exercised animals only.

6. It is concluded that muscle activity has a local effect on the volume of adjacent adipocytes. It is suggested that the mechanism involved is mechanical rather than biochemical.

1. Two groups of children with phenylketonuria (PKU) received protein at two different levels. The protein scource was a protein hydrolysate, devoid of phenylalanine, and intact protein from milk, vegetables and fruit. One group (RDA group) was given protein at a level based on the recommendations of the (US) Food and Nutrition Board (1974, 1980). The other group (FAO group) was given protein at the level of intake corresponding to the Joint FAO/WHO ad hoc Expert Committee (1973) safe levels of intake of egg or milk protein. The children were monitored very closely for several years. From an earlier study evaluating the protein intake of the two groups it was suspected that the Joint FAO/WHO ad hoc Expert Committee (1973) recommendations were marginal.

2. In the present study the phenylalanine intake of the two groups required to maintain the plasma phenylalanine concentration at the required level was established. The results showed that the RDA group required more phenylalanine than the FAO group. This difference was statistically significant from the age of 5–15 months.

3. We have interpreted the greater requirement for phenylalanine in the RDA group as a result of a greater nitrogen intake and thus a more rapid chemical maturation of N (increase in protein concentration of the body with age). It is known that up to the age of 6 months the chemical maturation of N is related to the N intake. In the present study we have found that this difference in chemical maturation lasted up to the age of 15 months. The conclusion drawn from the study was that a protein intake slightly higher than the Joint FAO/WHO ad hoc Expert Committee (1973) recommendations might be desirable.

1. Groups of rainbow trout (Salmo gairdneri) of approximate mean initial weight 8 g were grown in outdoor tanks over a 14-week period at water temperatures between 12° (start) and 6° (end). Four diets were used. Two contained non-oxidized fish oil (120 g/kg) with or without supplementary DL-α tocopheryl acetate and two contained moderately oxidized fish oil again with or without DL-α-tocopheryl acetate. The measured selenium content of the diets was 0.10 mg/kg.

2. No significant differences occurred as a consequence of the use of moderately oxidized oil compared with the corresponding treatments using non-oxidized oil. Significant differences did occur between dietary treatments that contained supplementary DL-α-tocopheryl acetate and those that did not. These differences applied to weight gain, haematocrit, erythrocyte fragility, mortalities, liver and muscle tocopherol concentrations and lipid peroxidation of liver mitochondria in vitro. Liver glutathione peroxidase (EC 1.11.1.9) activity was unaffected by the dietary treatments used and the proportions of fatty acids in polar lipids of liver and muscle were little changed by the diets used. Severe muscle damage occurred in trout given diets lacking supplementary DL-α-tocopheryl acetate.

3. Previous experiments carried out on rainbow trout at a constant water temperature of 15° (Hung et al. 1981; Cowey et al. 1981, 1983), using diets lacking supplementary vitamin E, did not lead to differences in weight gain, pathological changes or mortalities.

4. Vitamin E requirement may increase as water temperature decreases; minimum dietary requirements for vitamin E measured at a constant water temperature of 15° may not be valid under practical conditions where water temperatures vary over the year.

1. The effects of dietary phosphorus and sulphaguanidine levels, and sex differences on: (a) phytate digestibility, (b) calcium and P utilization, (c) the activities of alkaline phosphatase (EC 3.1.3.1), alkaline phytase (EC 3.1.3.8) and acid phosphatase (EC 3.1.3.2) in the intestinal mucosa of male and female rats were investigated.

2. There was a linear increase in femur ash, Ca and P contents and the maximum force withstood by the fresh femurs as dietary P level was increased from 1.5 to 3.0 to 4.5 g/kg diet.

3. The apparent digestibilities of Ca, P and phytate-P decreased as the level of P in the diet increased. Rats given the diets with 1.5 or 3.0 g P/kg were hypercalciuric and hypophosphaturic compared with rats receiving 4.5 g P/kg diet.

4. The level of Ca retained was similar for all treatments. The level of P retained increased as the dietary P level increased. This suggests that P deprivation was a result of inadequate amounts of P retained and not due to the absorption of inositol phosphates formed during the enzymic hydrolysis of phytate.

5. The addition of sulphaguanidine increased phytate digestibility without changing the activities of acid and alkaline phosphatase or alkaline phytase of the intestinal mucosa. This suggests that these enzymes did not play a role in the increase in phytate digestibility. However, dietary sulphaguanidine enhanced phytate digestibility, suggesting that alterations in the diet which modify either the composition or metabolism of the gastrointestinal microflora may be beneficial in enhancing the in vivo hydrolysis of phytate.

6. Differences between males and females are reported and discussed.

1. Chicks and turkey poults were fed for 3 weeks on low-zinc diets, prepared from purified ingredients, supplemented with zinc oxide at graded levels.

2. Birds of both species given the unsupplemented basal diets grew poorly, with high mortality rates. All had severe hyperkeratosis but bone development was normal. Only when birds received diets with low concentrations of added Zn were leg abnormalities observed.

3. Zn requirements were assessed visually from dose-response graphs. The chick required 18 mg Zn/kg diet for maximal live weight and 24 mg Zn/kg for maximal Zn concentration in blood serum. The responses of tibial Zn and net retention of Zn did not reach plateaux within the range of dietary Zn concentrations studied. The turkey poult's Zn requirement for maximal live weight was 25 and 28–29 mg/kg for net retention of Zn and for maximal concentration of Zn in blood plasma and in the tibia; 41 mg Zn/kg diet was required for maximal Zn in blood serum.

4. Liver Zn was not correlated with dietary Zn in either species.

1. The dietary intake of male mice from a line selected for high body-weight was restricted to 60% by weight of the ad lib. intake of the control litter-mates between days 21 and 42 post partum.

2. One group was killed and compared with controls at 42 d of age. A further group then resumed ad lib. feeding. Mice which had regained the control body-weight by 61 d of age were killed and also compared with controls.

3. Muscle weight, fibre length, sarcomere length, sarcomere number per fibre, actin length and myosin length were measured in the biceps brachii and tibialis anterior muscles.

4. Muscle weight, fibre length and the number of sarcomeres per fibre were significantly reduced in both muscles following dietary restriction, but regained their control values following a resumption of normal feeding. The other indices remained unchanged throughout the study.

5. The pattern of fibre length changes, due entirely to alterations in sarcomere number, is similar to that reported previously during growth, senescence and immobilization and following selection for high and low body-weights.

1. Voluntary intake was determined with vegetative Lotus pedunculatus cut and fed fresh to growing sheep of 42–46 kg live weight. Effects attributable to condensed tannins were assessed by growing the plant under high and low levels of soil fertility, inducing low and high concentrations of tannin (Expt 1), or by binding the tannins through spraying the herbage with polyethylene glycol (molecular weight 3350, PEG; Expt 2). Primary-growth lotus was used in Expt 1 and secondary-growth lotus in Expt 2. Concentrations of total and free condensed tannin were determined in fresh lotus, free tanning being defined as that not bound by mascerates of the plant.

2. In Expt 1 the herbages fed contained respectively 46 and 106 g total condensed tannin/kg dry matter (DM) and 3 and 14 g free condensed tannin/kg DM. Mean metabolizable energy (ME) intakes were 0.89 and 0.77 MJ/kg live weight0.75 per d (P < 0.05) respectively.

3. The lotus used in Expt 2 contained 63 and 5 g total reactive condensed tannin and free condensed tannin/kg DM respectively. After spraying with PEG at 2.4 g/g total condensed tannin, these values were reduced to 7 and 0.5 g/kg DM respectively. PEG addition increased apparent digestibility (proportion of each nutrient ingested) of cellulose, hemicellulose and nitrogen by 0.05, 0.08 and 0.26, and increased ME intake from 0.48 to 0.69 MJ/kg live weight0.75 per d.

4. It was concluded that high concentrations of condensed tannins depressed ME intake, due to depressions in both the voluntary intake and digestion of organic matter. The stimulation of hemicellulose and cellulose digestion by PEG addition suggests that the condensed tannin was depressing rumen digestion.

5. It is considered that free tannins are most likely to have been responsible for the depressive effects in the present study. This could be due to their reaction with microbial enzymes in the rumen, the tannin content of the plant exceeding the capacity of the plant protein to bind it, or to reaction with enzymes secreted into the small intestine following the absorption of amino acids. In either instance free tannins could also react with proteins of the gut wall.

1. Primary growth vegetative Lotus pedunculatus containing 46 and 106 g/kg dry matter (DM) of total condensed tannin and 3 and 14 g/kg DM of free condensed tannin, was cut and fed fresh at hourly intervals (750 g DM/d) to sheep fitted with permanent cannulas into the rumen and duodenum. Low- and high-tannin lotus contained respectively 41.3 and 31.6 g total nitrogen/kg DM and 132 and 152 g lignin/kg DM. The two forms of lotus were similar in carbohydrate composition.

2. Nutrient intake was recorded and faecal output measured by direct collection. Digesta flow to the duodenum was estimated by measuring dilution at the duodenum of inert ruthenium phenanthroline (Ru-P) and chromium-EDTA markers continuously infused into the rumen.

3. Effects attributable to condensed tannins were assessed by comparing the digestion of the two diets, and by comparing the digestion of each with predicted values for non-tannin-containing fresh forages fed at similar intakes.

4. Apparent digestibility of all nutrients measured was less for high- than for low-tannin lotus (P < 0.01). The levels of cellulose digested ruminally and post-ruminally in both forms of lotus were similar to predicted values. However, less hemicellulose and readily fermentable carbohydrate (RFC; soluble carbohydrate + pectin) was digested in the rumen in sheep given both forms of lotus than would be predicted for non-tannin-containing fresh forage diets, but this was compensated for by greater post-ruminal absorption of both nutrients.

5. Total N gains across the rumen (duodenal N flow – total N intake) were 1.8 and 10.5 g/d for low- and high-tannin lotus v. predicted losses of 3.7 and 2.1 g/d for non-tannin-containing fresh forages given at the same total N intakes. Post-ruminal digestion of non-amonia-N (NAN; proportion NAN flowing at duodenum) was 0.71 and 0.67 for low- and high-tannin lotus respectively v. 0.76 for comparable non-tannin-containing fresh forages. Energy absorbed as amino acids from the small intestine was calculated to be 0.29 of metabolizable energy for both forms of lotus, compared with 0.17 and 0.21 for perennial ryegrass and white clover.

6. It was concluded that the presence of condensed tannins in lotus markedly increased post-ruminal NAN absorption compared with non-tannin-containing fresh forage diets, but depressed ruminal digestion of RFC and hemicellulose.

1. Concentrations of reducing sugars, glucose, fructose and lactic acid in blood obtained from arterial and portal catheters were measured for periods of 8–24 h in twenty-three unanaesthetized pigs (mean body-weight 50 kg). From 6 to 8 d after implantation of catheters, the animals received experimental meals containing different levels (400, 800, 1200, 1600 g respectively) of different sugars (glucose ten meals, sucrose eighteen meals, lactose nine meals, maize starch sixteen meals) as well as a protein–mineral–vitamin premix.

2. After each meal the reducing sugars appeared in the portal blood in successive waves. The porto-arterial differences in the concentration of reducing sugars, representing the real appearance of sugar-hydrolysis products in the animal, varied greatly according to the sugar ingested and its level of intake. For each level of intake, these differences were larger, but of shorter duration, for glucose and sucrose than for maize starch. For these three carbohydrates, the higher the level of ingestion, the larger and the more persistent the porto-arterial differences. Lactose represented a special case, as the porto-arterial differences of reducing sugars were always much lower than those obtained with the other sugars and they did not vary with the level of intake.

3. Our findings show that the products formed by feeding glucose and sucrose appear more rapidly in the portal blood than those formed by feeding lactose. Accordingly, the length of time of digestion of glucose and sucrose is shorter than that of maize starch and lactose.

1. Concentrations of reducing sugars, glucose, fructose and lactic acid in blood obtained from arterial and portal catheters were measured together with the portal hepatic blood flow-rate for periods of 8–24 h in twenty-six unanaesthetized pigs (mean body-weight 51 kg). The animals received experimental meals containing different amounts (100–1600 g) of different sugars (glucose fifteen meals, sucrose twenty-four meals, lactose fourteen meals, maize starch nineteen meals) together with a protein–mineral–vitamin mixture (150 g) 6–8 d after implantation of the catheters and an electromagnetic flow probe.

2. Because the portal blood flow pattern did not differ between test meals, net absorption followed the same trends as for porto-arterial concentration differences (Rérat et al. 1984). Apart from lactose, the amounts of reducing sugars appearing in the portal vein correlated with the intake of the test meal, but the absorption pattern was different for each sugar. The appearance of reducing sugars was faster and earlier after intake of glucose than after sucrose and the same was the case for sucrose relative to maize starch. The differences between the three carbohydrates tended to increase with the level of intake.

3. With a test meal containing 1 kg carbohydrate, i.e. a normal meal in a 50 kg pig, digestion of sucrose and maize starch was not finished 8 h after the meal since only 60 and 52% respectively of their hydrolysis products were recovered in the portal blood. In the case of lactose, the amounts of reducing sugars appearing in the portal blood were always very small and constant (113–118 g within 8 h) whatever the level of intake, i.e. 30 and 15% of their hydrolysis products for intakes of 400 and 800 g respectively.

4. Depending on the carbohydrate ingested, the uptake of glucose by the gut cell wall ranged from 14 to 21 g/h and the production of lactic acid from 2.5 to 3.5 g/h.