Six male children, aged 8–28 months, were studied for three consecutive periods of 1 week each. They were given diets that provided 1.7 g protein/kg per d and supplements of minerals and vitamins, with a metabolizable energy intake during the 1st, 2nd and 3rd week of 419, 293 and 335 kJ (100, 70 and 80 kcal)/kg per d, diets 1, 2 and 3 respectively. All the food offered was consumed. Each child was weighed at the same time each day on an electronic balance. On the 6th and 7th day of each study period urine and stool were collected for 24 h to assess nitrogen balance. Using linear regression analysis it was shown that all children gained weight on diet 1, 2.3 (sd 1.3) g/kg per d, and five of six children gained weight on diet 3, the mean for the whole group being 2.7 (sd 2.3) g/kg per d, not significantly different. On diet 2 all children lost weight, -5.4 (sd 1.3) g/kg per d, highly significantly different from each of the other dietary periods. Using asymptotic regression analysis it could be shown that the rate of weight loss during the first part of the week on diet 2, - 11 g/kg per d, was greater than at the end of the week, - 2 g/kg per d, with a tendency towards a steady weight by day 7. Apparent N retention (mg/kg per d) was positive at the end of each of the three dietary periods: diet 1, 112 (sd 25); diet 2, 54 (sd 34); diet 3, 82 (sd 20). In five of the six children there was a marked reduction in stool frequency on diet 2 compared with diet 1, that was maintained to the period on diet 3. The findings suggest that during the period on diet 2 there was a saving of energy of the order of 105 kJ (25 kcal)/kg per d, which lasted through into the period on diet 3.

The relationship between diet and the development of chronic disease still remains a controversial area. One major difficulty is to obtain a valid estimate of habitual pattern and level of food consumption for each individual. There is, in fact, a voluminous and largely negative literature on the validity of dietary assessment methods. In the present paper the utility of the most frequently used dietary assessment method in epidemiological studies is discussed in terms of precision and accuracy.

Sixteen adult castrated male goats of the Granadina breed, with initial live weights ranging from 26.0 to 33.3 kg were used in two experiments to determine their protein and energy requirements for maintenance. Digestibility, nitrogen and energy balance measurements were made during the experiments. Two diets, which were based on pelleted lucerne (Medicago sativa) hay alone or on this forage and barley, were individually given at about maintenance level once daily. Gas exchange was measured using open-circuit respiration chambers. Fasting heat production was also determined. By regression analysis endogenous urinary N and maintenance requirements for N were estimated to be 119 mg/kg body-weight (W)0.75 per d and 409 mg total N/kg W0.75 per d respectively. Fasting heat production was 324 kJ/kg W0.75. The energy requirement for maintenance was calculated by regression of energy balance on metabolizable energy (ME) intake and a value of 443 kJ/kg W0.75 per d was found. The overall efficiency of utilization of ME for maintenance was 0.73.

The metabolism of maltitol (4-α-D-glucosylsorbitol) was assessed in fasting conventional (C) rats, C mice and germ-free (GF) mice, using [U-14C]maltitol. The radiorespirometric patterns of 14CO2 collected for 48 h after the administration of labelled maltitol were characterized by a constant rate of 14CO2 production lasting 4 h for both C rats and mice. The pattern for the GF mice showed a peak at the second hour followed immediately by a slow decrease. The percentage recovery of 14CO2 was significantly lower for the GF mice (59%) compared with C animals (72–74%). Urine, faeces and intestinal contents after 48 h totalled 19% of the administered radioactivity in the C rats and mice and 39% in the GF mice. The digestibility of maltitol and the absorption of sorbitol in GF mice was also assessed. The caecum and small intestine of GF mice, 3 h after administration of equimolar quantities of maltitol (140 mg/kg body-weight) or sorbitol (70 mg/kg body-weight), contained 39 and 51 % of the ingested dose respectively, present mostly in the caecum as sorbitol. The α-glucosidase (maltase) (EC 3.2.1.20) activity of the small intestine was appreciably higher (1·5–1·7 times) in the GF mice than in the C mice. These results suggest that the enzymic activities in the small intestine of mice and rats are sufficient to hydrolyse maltitol extensively. Consequently, the slow absorption of sorbitol seems to be an important factor limiting the overall assimilation of maltitol in the small intestine.

Twelve goats of the Granadina breed in mid- and late lactation were used in two consecutive years to determine their protein and energy requirements for lactation. The animals were individually fed on diets based on pelleted lucerne (Medicago sativa) hay and barley. A total of six balance experiments were carried out. Gas exchange was measured using open-circuit respiration chambers. Milk yield ranged widely from 0.649 to 1.742 kg/d in the first year and from 0.222 to 1.989 kg/d in the second year, a steady decline in milk output being observed as lactation progressed. Milk composition remained rather constant during the midstage of lactation, with an average content (/kg milk) for total solids, total nitrogen, fat and gross energy of 149.7 g, 5.39 g, 58.8 g and 3.59 MJ respectively. Total endogenous N, endogenous urinary N and maintenance requirement for N in lactating goats were estimated to be 244, 218 mg N/kg body-weight (W)0.75 per d and 478 mg total N/kg W0.75 per d respectively from regression equations. A constant efficiency of use of dietary N for milk N plus retained N of 51.0 % was found. By regressing milk energy plus apparent body energy retention or loss on metabolizable energy (ME) intake, the maintenance energy requirement was estimated to be 401 kJ ME/kg W0.79 per d. When estimating the corrected milk yield as milk energy +(0.84 x negative energy retention) +(1.05 x positive energy retention), regression analysis indicated that the overall efficiency of use of ME for lactation was 66.7%. Also, from a plot of apparent body energy retention v milk energy yield, both expressed as a percentage of ME intake above maintenance, the efficiency with which ME was used to promote energy retention in the body during lactation was calculated to be 0.907 times that for milk secretion.

The effects of the inclusion of raw and autoclaved whole faba beans (Vicia faba; RFB and AFB respectively) or faba bean fractions (cotyledons and hulls) in diets for growing broiler chickens (0–4 weeks of age) on performance, intestinal physiology and jejunal histological structure have been studied in three experiments. Significant decreases in body-weight as well as lower food consumption and higher food intake:weight gain ratio were observed in those animals fed on diets containing 250, 350 and 500 RFB'kg in the diet. Birds fed on AFB diets (500 g/kg) had significantly greater body-weights than chicks fed on RFB or raw faba bean cotyledons (RC). Significant increases in the relative lengths of duodenum, jejunum, ileum and caeca, pancreas relative weight, and intestinal transit time of birds fed on diets containing 250,350 and 500 g RFB/kg compared with control birds were observed. Including AFB (500 g/kg) in the diet significantly increased body-weight and significantly decreased pancreas weight compared with RFB (500 g/kg)-fed birds. The inclusion of RFB hulls had no effect on these variables. Dehulling or autoclaving of faba beans, or both, proved to have no significant effect on relative lengths of duodenum, jejunum, ileum and caeca, nor on caecal volatile fatty acid concentration in birds fed on 500 g faba beans/kg diet. Electron microscopy of the jejunal mucosa revealed discrete hyperplasia of polysomes and mitochondria1 swelling in those animals fed on AFB (500g/kg) or AC (4264g/kg). Pronounced strangulations were also observed along the microvilli, whose length was similar to that of control birds. The inclusion of RFB hulls, either autoclaved or raw, led to no ultrastructural changes in the enterocytes, as detected by electron microscopy. Birds fed on diets containing the cotyledons of RFB (RC, 426 4 g/kg) rather than whole RFB showed the same ultrastructural disorders as RFB (500 g'kg)- fed birds. The present study shows that factors other than those usually claimed, i.e. protease inhibitors, phytates, tannins and lectins, may be contributing to the low nutritional value of V. faba seeds for growing chickens.

An in vitro faecal incubation system was used to study the metabolism of complex carbohydrates by intestinal bacteria. Homogenates of human faeces were incubated anaerobically with added lactulose, pectin, the hemicellulose arabinogalactan, and cellulose, both before and after subjects had been pre-fed each carbohydrate. Fermentation of added substrate was assessed by the production of short-chain fatty acids (SCFA) and suppression of net ammonia generation over 48 h of incubation. Control faecal homogenates to which carbohydrate was not added yielded an average increment of SCFA of 43 mmol/l, equivalent to 172 mmol/kg in the original stool. The addition of lactulose, pectin and arabinogalactan each increased the yield of SCFA by a similar amount, averaging 6·5 mmol/g carbohydrate or 1·05 mol/mol hexose equivalent; organic acid yield was not increased by pre-feeding these substances for up to 2 weeks. Acetate was the major SCFA in all samples at all times and, after pre-feeding with extra carbohydrate, butyrate concentrations exceeded propionate in all samples. Faecal homogenates incubated with cellulose showed no greater SCFA production than controls over the first 48 h, but there was a slight increase when samples from two subjects pre-fed cellulose were incubated for 14 d. Net ammonia generation was markedly suppressed by addition of lactulose to faecal incubates with an initial period of net bacterial uptake of ammonia. Pectin and arabinogalactan also decreased ammonia generation, but the reductions were not significant unless subjects were pre-fed these materials; cellulose had no effect on ammonia generation.

Four groups of seven dairy cows were given hay plus high-fibre concentrates based on sugar-beet feed (hay-concentrate, 40:60 w/w) or high-starch concentrates based on flaked maize (hay-concentrate, 20:80 w/w), with a crude protein (nitrogen x 6.25) content of either 160 or 220 g/kg dry matter, over weeks 4–18 of lactation. Performance during week 3 of lactation, when all cows were fed on a standard ration, was used as a covariate. For diets with a high-fibre content, higher protein concentrations led to increases in yields of milk and milk fat, with no effect on live-weight loss. For diets with a high-starch content, higher protein concentrations did not affect milk yield or composition but resulted in an increase in live weight rather than a decrease. Diets with a high-starch content led to increased proportions of propionic acid in the rumen and increased concentrations of insulin in the blood. It is concluded that the source of carbohydrate needs to be taken into account when predicting the response to protein supply by dairy cows.

The effect of long-term (10 weeks) treatment with growth hormone (GH) was investigated in twin lambs, one sibling being a control and the other treated with GH (0.1 mg/kg live weight per d). The lambs were fed on a concentrate-grass cube (9: 1 w/w) diet at a daily rate of 40 g fresh weightlkg Live weight. The average daily live-weight gain of the GH-treated lambs was 36% greater than that of the controls (307 v. 225 g/d, P < 0.01). The carcass composition of the GH-treated lambs changed: fat content was decreased (P < 001) and protein content was increased (P< 0.05) when expressed relative to carcass dry matter. The absolute weights and the weights when expressed relative to fleece-free empty body of some muscles were significantly increased in GH-treated lambs. The mean retention times of both particulate- and liquid-phase components of the digesta were unchanged by GH treatment, when calculated for the rumen or for the entire gastrointestinal tract. The feed conversion ratio was significantly greater (P < 0.01) in GH-treated lambs compared with controls. Nitrogen retained per g N intake was also significantly increased (P < 0.05) by GH treatment. Plasma urea concentrations were decreased (P < 005) and glucose concentrations were increased (P< 001) in GH-treated lambs, whereas non-esterified fatty acid concentrations were unchanged. Plasma insulin and total insulin-like growth factor-1 concentrations progressively increased in GH-treated lambs as treatment time continued. They were significantly correlated after week 4 of treatment. Two types of hepatic GH- binding site were detected, with high and low affinities for GH. The capacities of both binding sites were significantly increased (P < 0.05) in GH-treated lambs when expressed per unit microsomal protein but, when expressed per liver, only the capacity of the high-affinity site was increased.

Recent studies suggest that dexfenfluramine (D-fenfluramine), because of its pure serotonergic effect, may be a more potent anti-obesity agent, associated with fewer side-effects than the racemate DL-fenfluramine. The effect of dexfenfluramine on gastric emptying of a mixed solid and liquid meal was assessed with a double-isotope scintigraphic technique in eleven obese patients. Each subject took a placebo capsule on the morning and evening of the day before, and on the morning of the first gastric emptying measurement. Dexfenfluramine was then taken at a dose of 15 mg twice daily and gastric emptying measurements were performed at 5 and at 29 d after the initiation of active treatment. Dexfenfluramine significantly slowed gastric emptying of the solid meal at both 5 and 29d when compared with the placebo (P < 0.05) and also delayed emptying of solid food from the proximal stomach (P < 0.01), but no significant effect on liquid emptying was observed. No significant side-effects were reported and there was a marginal weight loss (P< 0.005) during treatment. We conclude that inhibition of gastric emptying may contribute to the efficacy of dexfenfluramine in the treatment of obesity.

Vitamin B6 deficiency was induced in mice by maintenance for 4 weeks on a vitamin B6-free diet. Tryptophan metabolism was assessed by determining the urinary excretion of tryptophan metabolites, the metabolism of [14C]tryptophan in vivo and the formation of tryptophan and niacin metabolites by isolated hepatocytes. The vitamin B6-deficient animals excreted more xanthurenic acid and 3-hydroxykynurenine, and less of the niacin metabolites N1-methyl nicotinamide and methyl-2-pyridone-4-carboxamide, than did control animals maintained on the same diet supplemented with 5 mg vitamin B6/kg. After intraperitoneal injection of [14C]tryptophan, vitamin B6-deficient mice showed lower liberation of14CO2 from [methylene-14C]tryptophan and [U-14C]tryptophan than did controls, indicating impairment of kynureninase (EC 3.7.1.3) activity. There was no difference between the two groups of animals in the metabolism of [ring-2-14C]tryptophan. Hepatocytes isolated from the vitamin B6-deficient animals formed more 3-hydroxykynurenine and xanthurenic acid than did cells from control animals, but also formed more NADP and free niacin.

Six dairy cows of two breeds were fed during three alternating periods with products from C3- and C4-plants to yield different natural 13C enrichments of the diet (δ13C range: -28.0 to -13.7‰). The resulting changes in the 13C enrichment of breath carbon dioxide, serum and milk of the animals followed the 13C: 12C of the food, in agreement with the individual biological half-lives of those products, and established isotope discriminations. Breath CO2 was more enriched in 13C than expected. This could be related to isotope discriminations during rumen fermentation. From these results an isotopic balance model for the breath CO2 could be established.

Healthy male volunteers consumed hot lunches consisting of cooked white rice, fried chicken fillet, raisins, bigarreaus (sweet cherries) and curry sauce with carbohydrate: fat ratios of either 0·77 or 2·04, and polysaccharide: (mono+di)saccharide ratios (P: MD) varying between 5·3 and 0·95. Before and at various time intervals after the start of the meal, blood samples were analysed for glucose, insulin, triacylglycerol (TG), free glycerol (FG), free fatty acids (FFA) and cholesterol. Elevated postprandial glucose and insulin peaks were induced by the meals containing a larger amount of carbohydrate. The type of carbohydrate in the meal appeared to have little or no effect on the peak maximum. A small second glucose peak was seen at 2 h after the meals with P: MD > 3·1. The TG concentration in the blood showed a similar and rapid rise. After the meal containing the largest amount of simple sugars, the TG curve levelled off 1 h after the start of the meal and then remained at a nearly constant level (1·5 mM). In contrast, a larger amount of complex carbohydrates induced a TG concentration which rapidly rose to a maximum of 1·75 mM, and subsequently decreased slowly to somewhat below that of the simple sugar-rich meal at 4 h. The postprandial curves after the fat-rich meals showed a continuous rise of TG up to a maximum of about 2 mM at 2 h, and a subsequent slow decrease. FG and FFA all showed a rapid drop immediately after the start of the consumption of the meals, followed by a rebound at 1 h. The postprandial curves of total cholesterol, as well as the areas below the curve of both total cholesterol and free, unesterified cholesterol were lower after the carbohydrate-rich meals than after the fat-rich meals. This is attributed to the larger amount of cholesterol in the fat-rich meals.

The saliva of sheep was shown to contain significant concentrations of uric acid (16 (sd) 4.5) μmol/l) and allantoin (120 (sd 16.4) μmol/l), sufficient to recycle purine derivatives equivalent to about 0.10 of the normal urinary excretion. When allantoin was incubated in vitro in rumen fluid, it was degraded at a rate sufficient to ensure complete destruction of recycled allantoin. In a series of experiments in which allantoin was infused into the rumen of sheep fed normally, or into the rumen or abomasum of sheep and the rumen of cattle completely nourished by intragastric infusion of volatile fatty acids and casein, no additional allantoin was recovered in the urine. These losses were probably due to the degradation of allantoin by micro-organisms associated with the digestive tract. It is concluded that all allantoin and uric acid recycled to the rumen via saliva will be similarly degraded. Therefore, the use of urinary excretion of purine derivatives as an estimator of the rumen microbial biomass available to ruminants will need to be corrected for such losses.