In the present study the actual role of propionic acid in the control of fatty acid and cholesterol synthesis was investigated in isolated liver cells from fed rats maintained in the presence of near-physiological concentrations of glucose, glutamine and acetate. Using 3H2O for lipid labelling, propionate appears as an effective inhibitor of fatty acid synthesis and to a lesser extent of cholesterol synthesis, even at the lowest concentration used (0·6 mmol/l). Butyrate is a potent activator of both synthetic pathways, and the activating effect was not counteracted by propionate. Using 1-[14C]acetate, it was observed that propionate at a moderate concentration, or 1 mmol oleate/l, are both very effective inhibitors of 14C incorporation into fatty acid and cholesterol. This incorporation was drastically inhibited when propionate and oleate were present together in the incubation medium. The net utilization of acetate by rat hepatocytes was impaired by propionate, in contrast to oleate. 1-[14C]butyrate was utilized at a high rate for fatty acid synthesis, but to a lesser extent for cholesterol synthesis; both processes were unaffected by propionate. Intracellular citrate concentration was not markedly depressed by propionate, whereas it was strongly elevated by butyrate. In conclusion, propionate may represent an effective inhibitor of lipid synthesis when acetate is a major source of acetyl-CoA, a situation which is encountered with diets rich in readily-fermentable fibres. The present findings also suggest that propionate may be effective at concentrations close to values measured in vivo in the portal vein.

Diets enriched in retrograded amylose (RS3) have been shown to lower serum cholesterol concentrations in rats. The possibility was tested that this hypocholesterolaemic effect of RS3 is caused by an increase in excretion of neutral steroids and/or bile acids. Six groups of ten rats were fed on purified diets containing either 12 or 140 g RS3/kg solid ingredients with and without added cholesterol (5 g/kg). Low-RS3 diets, with and without added cholesterol, to which the bile-acid-binding resin cholestvramine (20 g/kg) was added, were used as reference. The high-RS3 diets v. the low-RS3 diets tended to reduce the increase in the total serum cholesterol concentration during the course of the experiment (P = 0 067), decreased serum triacylglycerol concentrations, raised total neutral steroids and total bile acids in caecal contents and faecal excretion of total bile acids, but lowered faecal excretion of neutral steroids. In addition, the serum concentration of total 3α-bile acids was markedly raised by the high-RS3 diets. The high-RS3 diets raised the faecal excretion of lithocholic and muricholic acids, but lowered that of hyodeoxycholic acid, and increased the caecal amounts of lithocholic, ursodeoxycholic, β-muricholic and ω-muricholic acids. Apart from the stimulation of faecal bile acids excretion, the effects of cholestvramine on bile acid metabolism differed at various points from those of RS3. Cholesterol feeding had predictable effects on cholesterol metabolism and led to greater elevating effects of RS3 on the faecal and caecal amounts of muricholic acids. The results suggest that the serum-cholesterol-lowering effect of high-RS3 diets may be explained by an increased influx of neutral steroids and bile acids into the caecum, and increased faecal excretion of bile acids, and/or by an altered intestinal bile acid profile.

Thirty male rats were randomly assigned to one of three dietary groups in which the source of dietary fat was either a mixed oil, maize oil or fish oil. Effects of dietary fatty acid composition on in vitro rates of [U-14C]glucose incorporation into hepatic total lipids and into hepatic triacylglycerol were measured under basal, insulin (4 nM)-, gastric inhibitory polypeptide (GIP; 6 nM)- and insulin + GIP (4 nM + 6nM)-stimulated conditions. Effects of the three diets on postprandial plasma triacylglycerol, cholesterol, insulin and GIP concentrations were also measured. The fish-oil diet decreased rates of basal glucose incorporation into hepatic total lipids (P < 0·05) and hepatic triacylglycerol (P < 0·01) compared with the mixed-oil diet. The presence of insulin and GIP in the incubation medium stimulated glucose incorporation into hepatic total lipids in the maize-oil (P < 0·01) and fish-oil groups (P < 0·05), as well as into hepatic triacylglycerol in the maize-oil group (P < 0·005). In addition, the fish-oil diet decreased postprandial plasma triacylglycerol levels compared with both other dietary groups (P < 0·05 both cases), and the mixed-oil diet markedly increased postprandial plasma insulin levels compared with the other dietary groups (P < 0·001).

The effects of various sources of dietary fibre on the high glycaemic index of an Israeli ethnic food, melawach, were investigated in subjects with non-insulin-dependent diabetes mellitus (NIDDM). Locust-bean (Ceratonia siliqua) gum significantly decreased the glucose response to, and glycaemic index of, melawach in these diabetic subjects (P < 0·05). It also tended to decrease their insulinaemic response and insulinaemic index, but differences were not significant. Dietary fibre from lupin (Lupinus albus) and insoluble maize-cob fibre did not affect glucose and insulin levels in NIDDM volunteers. Subjects with a BMI < 30 kg/m2 exhibited similar glucose, but not insulin, responses to fibre. Locust-bean gum had no significant effect on glycaemic response in NIDDM subjects with a BMI > 30 kg/m2, whereas insulinaemic response decreased. The results indicate that foods containing the same nutrients in almost the same amounts, but differing in added dietary fibre, lead to different physiological responses in diabetic subjects. Furthermore, insulin response should be considered when fibre is incorporated into the diabetic's diet.

The short-term kinetics of Zn turnover were studied in Irish setters with gluten-sensitive enteropathy and control dogs following intravenous injection of 0·25 mg 96·5% enriched 70ZnCl2. The 70Zn enrichment of serum was found closely to obey two-compartment kinetics and the derived two-compartment decay equation was used to calculate the size and turnover of the two initial rapidly exchanging pools of body Zn. In normal Irish setters isotopic Zn initially equilibrates with a pool (a) of size 1·27 (SD 0·46) μmol/kg and then with a second pool (b) of size 6·83 (SD 1·72) μmol/kg. The fractional turnover of pool (b) was approximately one eighth that of pool (a). Enteropathic dogs showed no reduction in the size of either rapidly exchangeable Zn pool, reduction in serum Zn concentration or abnormality in Zn balance and hence these results do not support the possibility of an underlying Zn deficiency in this disorder.

The association of radiolabelled taurocholic acid with the solid fraction of a faecal fermentation mixture was measured. A human faecal inoculum was incubated with [24-14C]taurocholic acid and several non-starch polysaccharide sources (pectin, wheat bran, ispaghula (Plantago ovata) husk and seed), glucose or a substrate-free control. Portions of fermentation mixture were taken at 0, 3, 6, 21 and 24 h and centrifuged to acquire a supernatant fraction and a pellet containing the fermentation residue. 14C was measured in supernatant fractions and pellets at all time points. Volatile fatty acids (VFA) were measured at 0 and 24 h to confirm bacterial growth. Radioactivity in the pellet increased over time for all substrates. Glucose resulted in the greatest incorporation of taurocholic acid into the pellet, followed by pectin. At 24 h the proportion of the total radioactivity found in the pellet was 92% for glucose, 79% for pectin, 60% for wheat bran, 59% for ispaghula seed, 53% for ispaghula husk and 26% for the control (mean of duplicates). Glucose and pectin produced the greatest quantity of VFA at 24 h. VFA production was highly correlated with radioactivity in the pellet (r 0·976, P <0·005). These results suggest that the bile acid binding capacity of a faecal culture mixture may be strongly influenced by the fermentability of the available substrate and hence related to bacterial metabolic activity.

Malnutrition in infancy is a global problem which leads to retardation of childhood growth and development. There is a pressing need to improve weaning strategies for infants of the developing world. Traditional Gambian weaning foods are watery and of low energy density, but addition of energy in the form of fat and carbohydrate leads to thick, viscous gruels which are difficult to ingest. Partial digestion with amylase (EC3·2·1·1)-rich flour reduces their viscosity while retaining their energy density. The aim of the present study was to measure the digestibility of a maize-based weaning food, before and after amylase digestion, in malnourished children using a 13C breath test. Ten children (aged 7–16 months; mean weight-for-age Z score −0·8) received isovolumetric and isoenergetic quantities of a maize-based weaning food naturally abundant with 13C. Breath samples were collected at intervals of 30 min for 5 h thereafter and 13CO2 enrichment was measured by isotope-ratio mass spectrometry. Percentage dose of 13C recovered increased from a mean 13·7 (SD 3·7)% before, to 18·3 (SD 5·6)% after ingestion of amylase-treated weaning foods (P < 0·1). There was a significant inverse relation between age and weight, and percentage dose of 13C recovered in children receiving amylase-treated feeds. There were no differences in concentrations of amylase in saliva of infants or breast milk of their mothers. Partial digestion of supplementary foods may improve the nutrition of undernourished weaning children, not only by reducing their viscosity, thereby increasing ingestion, but also by improving their digestion and thereby their absorption.

The effect of long-term sucrose intake on lipid metabolism and milk composition was examined in lactating mice and correlated with the development of their pups. Female mice were weaned onto semi-purified diets containing 685 g starch or sucrose/kg diet. Food intake was similar on the two diets. At mid-lactation, milk was collected and analysed. Milk from sucrose-fed dams had a lower percentage of linoleic acid (18:2n-6) and a higher percentage of oleic acid (18:ln-9), although milk fat content was similar. Hepatic fatty acid synthesis (as measured by 3H2O incorporation into fatty acid) was three times faster in sucrose-fed than in starch-fed dams. Dietary carbohydrate had no effect on fatty acid synthesis or functional lipoprotein lipase (EC 3.1.1.34) activity in mammary gland, nor on plasma triacylglycerol concentration. Pups from first litters were killed at 11–12 d and body composition analysed. Pups born to sucrose-fed dams were significantly heavier with increased body fat and higher levels of plasma glucose and triacylglycerol. Litters of sucrose- and starch-fed dams were cross-fostered at birth and killed at 10–12 d. Analysis of variance indicated that maternal sucrose intake, both during pregnancy and lactation, increased body fat of offspring. The increases in body weight and plasma triacylglycerol of pre-weaning offspring were caused primarily by maternal sucrose intake during lactation.

Two experiments were carried out to examine the short (1–3 d) and medium (14 d) term adaptations of rat large-bowel (LB) fermentation to alterations in substrate supply brought about by including cooked haricot beans (Phaseolus vulgaris) in diets based on white bread. Changes in organic matter (OM) flow from the ileum occurred within 1 d and were stable for 14 d but the pattern of caecal short-chain fatty acids took much longer to stabilize with considerable increases in butyrate between 1 and 3 d and up to 14 d. This suggests that the metabolic activity of the LB microflora may take a considerable time to stabilize after an abrupt change in substrate supply. Despite an almost fourfold change in OM supply to the LB, the proportion of this OM apparently fermented in that organ (0·46) remained fairly constant. None of the apparent resistant starch measured in the diets could be detected in faeces. Dietary non-starch polysaccharides (NSP) were extensively fermented with similar values (0·82) for both bread and bean NSP and there was little indication of any interaction between the two diet components on NSP fermentation. An attempt was made to fractionate ileal and faecal OM to provide a basis for a quantitative model of LB stoichiometry.

Dietary Mg restriction is generally considered to contribute to the prevention of struvite urolithiasis in cats, but its effects on faecal and urinary excretion of Ca and P have not been systematically investigated. The present study seeks to fill the gap. In a 4×4-week crossover study, ovariectomized cats were fed on purified diets containing either 0·40, 0·79,1·59 or 3·17 mmol Mg/MJ (0·19, 0·38,0·76 and 1·52 g Mg/kg diet). Increasing the dietary Mg level from 0·40 to 3·17 mmol Mg/MJ by the addition of extra MgCO3 raised urinary and faecal excretion of Mg from 0·14 to 0·68 mmol/MJ and from 0·28 to 1·66 mmoI/MJ respectively. The 8-fold increase in Mg intake significantly raised urinary excretion of Ca from 0·06 to respectively when the dietary Mg level was raised from 0·40 to 3·17 mmol/MJ. A dietary level of 0·40 mmol Mg/MJ (0·19 g Mg/kg diet) was found to be sufficient to maintain Mg balance in the adult ovariectomized cats fed on the purified diet.

The effects of oat saponins (a mixture of avenacosides A and B) and dietary fibre (cellulose and guar gum) on the disaccharidase activities in the proximal small intestine of the rat were investigated. The influence of avenacosides A and B on the activity of disaccharidases and α-amylase (EC 3·2·1·1) was also studied in vitro. In vivo, oat diets with three avenacoside contents (negligible, normal and twice normal) were used. No significant differences in sucrase (EC 3·2·1·48), maltase (EC 3·2·1·20), trehalase (EC 3·2·1·28) and lactase (EC 3·2·1·21) activities were found between the oat groups after 19 d feeding. The rats that were given cellulose tended to have higher disaccharidase activities compared with the other groups. The avenacosides inhibited the lactase activity significantly in vitro while no or small effects on the other disaccharidases were found. In contrast, the in vitro hydrolysis of starch by α-amylase was increased in the presence of saponins, probably due to their detergent effect. Thus, the in vitro studies showed that the avenacosides could influence the enzyme activities. In vivo, these effects are probably minor due to the low avenacoside concentrations found in oats.

In order to induce a range of vitamin A-deficient states in young growing rats and to study the effect of vitamin A deficiency on Fe status, we designed the following two-generation experiment. Dams were fed on diets with one of five vitamin A levels from 2 weeks before and throughout pregnancy and lactation. The pups received the same diets as their mothers both before and after weaning. The five dietary levels of vitamin A were 1200, 450, 150, 75 and 0 retinol equivalents/kg feed. Vitamin A intake did not affect reproduction outcome, nor were body and liver weights of the pups affected when they were 3·5 weeks old. Male pups with normal vitamin A status had higher plasma retinol levels than female pups. Vitamin A status of the offspring was affected from 3·5 weeks onwards. Body and liver weights were decreased in the male pups given the lowest dietary vitamin A levels from week 6·5 onwards but not in the female pups. Fe status was marginally affected. Haemoglobin levels were increased and total Fe-binding capacity was decreased in the groups given no dietary vitamin A at week 9·5. Splenic Fe was increased only in the male pups given the lowest levels of dietary vitamin A. However, as a whole, Fe status was only mildly affected and subject to considerable variation. We conclude that the two-generation rat model described here is not suitable for studying effects of vitamin A deficiency on Fe metabolism.

The present study was designed to determine the quantitative effects of starchy meals containing guar gum on rates of net apparent glucose absorption and net apparent insulin and gastric inhibitory polypeptide (GIP) production in growing pigs. The effects of these meals on the viscosity of jejunal digesta were also examined and correlated to changes in glucose absorption. Four growing pigs were each given either a low-fat semi-purified diet (control) or the same diet supplemented with a high-molecular-weight guar gum at concentrations in the diet of 20 or 40 g/kg. Blood samples were removed simultaneously via indwelling catheters from the mesenteric artery and the hepatic portal vein. Samples of jejunal digesta were removed via a T-piece cannula and used immediately for viscosity measurements at 39°. The ‘zero-shear’ viscosity of each sample was then calculated. Blood-flow measurements were made using an ultrasonic flow probe fitted to the hepatic portal vein. All measurements were made at intervals of 10 or 30 min during a 4 h postprandial period. Meals containing guar gum significantly increased (P < 0·05) the viscosity of jejunal digesta, an effect that was strongly dependent on the concentration of guar gum in the original diet. No significant differences in blood-flow rates were found between the control and guar-containing diets. Both concentrations of guar gum significantly reduced (P < 0·05) glucose absorption and insulin and GIP secretion rates over the 4 h postprandial period. An inverse relationship between the rate of glucose absorption and the ‘zero-shear’ viscosity of jejunal digesta was found. This study also provides direct evidence for the important role played by the entero-insular axis in modifying the glycaemic response to a meal containing guar gum.

In balanced experiments with rats the influence of the period of adaptation on nutrient digestibility in diets containing cellulose (CEL), guar gum (GG), pectin (PEC) or retrograded high amylose maize starch (RS) was studied. Inclusion level was 80 g/kg diet DM except for the retrograded high-amylose maize starch, where the level was 316 g/kg diet DM. A diet containing normal maize starch only acted as a controi diet (FF). The apparent digestibilities of DM, NSP, starch and protein were determined after adaptation periods of 4 d and 2, 4, 6 and 8 weeks. The digestibility of nutrients was affected by diet (P < 0.001) as well as by the period of adaptation (P < 0.001). The digestibilities of DM, NSP and starch increased asymptotically during the course of the experiment. The asymptotic progress over time was most pronounced for the GG, PEC and RS diets. The estimated periods of adaptation required for stable DM digestibility were approximately 1 week for the GG, PEC and RS diets and < 4 d for the FF and CEL diets. The digestibility of NSP in the GG and PEC diets was also stable after approximately 1 week, while it was stable from < 4 d for the CEL diet. However, PEC increased the faecal content of uronic acids for at least 2 weeks. A stable starch digestibility required 1 month in the case of RS and 3–10 d for the other diets. The high faecal content of glucose for the RS diet decreased during all 8 weeks but was still high at the close of the experiment. The apparent protein digestibility changed over time in a parabolic rather than an asymptotic fashion. The GG, PEC and RS diets increased the amount of N excreted by the faecal route.

A previous study has shown that malnourished, clinically stable patients with liver cirrhosis are in protein and energy balance at their spontaneous dietary intake and that an improvement in nutritional status cannot be anticipated at this intake (Nielsen et al. 1993). In the present study we examined to what extent oral intake could be increased by nutritional support, and to what extent dietary protein would be retained with increased intake. The techniques used for balance studies were also validated since this information is not available for patients with liver cirrhosis. Fifteen malnourished patients with alcoholic liver cirrhosis were given increasing amounts of a balanced ordinary diet for 38 (SE 3) d. Intakes of protein and energy were recorded by weighing servings and leftovers on food trays. Protein intake was calculated from food tables. Total N disposal was calculated after measurement of urinary N excretion, and protein balance was calculated from the N balance. A validation study of protein balance in a subgroup of patients (analysis of N in food by the duplicate portion technique, correction for incomplete recovery of urine by measurement of urinary para-aminobenzoic acid (PABA) after administration of PABA tablets, and measurement of faecal N) did not change protein balance values. Protein intake increased from 1. 0 (SE 0.1) g/kg per d to 1·8 (SE 0·1)g/kg per d. With increasing protein intake, 84 (SE 8)% of the increase in intake was retained. The rate of protein retention was not saturated at the intakes obtained in this study. Protein intolerance was only encountered in one patient. Available evidence indicates that the requirement for achieving N balance is increased in these patients but protein retention is highly efficient with increased intake. Protein retention is dependent on energy balance. Energy intake was calculated from food tables and total energy expenditure was calculated by the factorial method. A validation study was performed in a subgroup of patients. The energy contents of food sampled by the duplicate portion technique, and of urine and faeces were measured by bomb calorimetry. Resting energy expenditure (REE) was measured by indirect calorimetry before and at the end of the study, and O2 uptake during bicycle exercise was measured before and at the end of the study. The measured intake of metabolizable energy was on average 13% lower than the value given in food tables. Calculated energy expenditure was not changed by the validation study. Mean energy intake was 163 (SE 10) kJ/kg per d and mean energy expenditure was 134 (SE 5) kJ/kg per d (P = 0·007), indicating that the protein retention described occurred at a positive energy balance. It is concluded that a substantial retention of dietary protein can be obtained by oral nutrition support over a prolonged period of time in patients with liver cirrhosis. Requirements of protein for maintenance and repletion in these patients are discussed.

The effects of oral administration of linoleic- and γ-inolenic-acid-rich oils on the clinical and histopathological manifestations of experimental autoimmune encephalomyelitis (EAE) were investigated in Lewis rats 7 d post-inoculation. γ-Linolenic-acid-rich fungal (Mucor javanicus) oil at 500 mg/kg body weight abrogated clinical and histological signs of EAE although at doses of 200 and 1000 mg/kg body weight it was only effective in delaying the onset of clinical disease. Linoleic-acid-rich safflower-seed (Carthamus tinctorius) oil at 500, 750 and 1000 mg/kg body weight decreased the severity of clinical EAE. disease in a dose-dependent manner. The effects in healthy animals of orally administered γ-linolenic-acid-rich fungal oil (500 mg/kg body weight) and linoleic-acid-rich safflower-seed oil (1000 mg/kg body weight) on splenic lymphocyte proliferative responses to the T-cell mitogen concanavalin-A (Con A), membrane fatty acid composition and lymphocyte sub-sets were also studied. Both treatments enhanced the T-cell proliferative response to Con A. There was no significant effect on the proportion of splenic CD8+ or CD4+ lymphocytes. Compositional studies on splenic phosphoglyceride fatty acids of oil-treated animals suggest the above responses were associated with increases in spleen dihomo-γ-linolenic and arachidonic acids.

Given the current interest in the cardiovascular complications of Mg deficiency, the aim of the present experiment was to investigate the effect of Mg deficiency on the time-course of lipoprotein oxidation and to assess whether short-term Mg deficiency results in vitamin E depletion that predisposes lipoproteins and tissues to subsequent oxidation. Weanling rats were pair-fed for 8 d with control and Mg-deficient diets respectively. Plasma triacylglycerol and α-toeopherol levels were significantly greater in Mg-deficient rats compared with control animals. The increase in plasma apolipoprotein B concentration indicated that a corresponding increase in plasma triacylglycerol-rich lipoproteins (TGRLP) occurred in Mg-deficient animals. Hyperlipaemia was associated with modifications in the composition of TGRLP. The proportion of triacylglycerols was elevated whereas that of cholesterol and protein was reduced, and Mg deficiency resulted in a slight significant reduction in α-tocopherol content. When the TGRLP fractions were subjected to in vitro Cu-induced oxidation the lipoprotein fractions from Mg-deficient rats were more susceptible to oxidative damage than lipoprotein fractions from control rats. Mg deficiency did not modify the α-tocopherol content of liver, heart and skeletal muscle. However, after exposure of tissue homogenates to Fe-induced lipid peroxidation, thiobarbituric acid-reactive substances were significantly higher in tissues from Mg-deficient rats compared with those from control rats. These results complement previous findings, showing that Mg deficiency increases the susceptibility of TGRLP and tissues to peroxidation and suggest that oxidative damage is not the result of a decrease in vitamin E antioxidant status.

It is known that the physico-chemical characteristics of fibre modify their fermentation characteristics in the colon. Previously we showed the varying effects of inulin and different types of fibre on the hepatic and intestinal xenobiotic-metabolizing enzymes (XME) in initially germ-free rats inoculated with a human, methanogenic, whole-faecal flora (Roland et al. 1994). The aim of the present work was to assess whether or not these effects could be related to differences in production of fermentation metabolites (gases excreted in vivo and caecal metabolites) due to the different compositions of fibre. The different types of fibres were analysed with regard to their solubility and their composition of neutral monomers and uronic acids. Inulin was totally soluble, carrot (Daucus carota), cocoa (Theobroma cacao) and wheat bran were partially soluble; pea (Pisum sativum) and oat were nearly totally insoluble. Uronic acids were found mostly in carrot and cocoa fibre. Glucose was present as the main neutral monomer in each fibre type. Xylose was found also in wheat bran, pea and oat fibres, and arabinose was found in wheat bran. Inulin consumption led to high levels of H2 production but no CH4 production, to a 4-fold greater caecal concentration of butyrate than with the other fibres and to a decrease in caecal pH. Conversely, rats fed on carrot or cocoa fibre produced a large amount of CH4 but no H2 and generated a different profile of short-chain fatty acids (SCFA). The lowest amounts of gases and SCFA were found in rats fed on wheat bran, pea and oat fibre. We observed a relationship between the caecal concentration of SCFA and the activity of hepatic glutathione-S-transferase (EC 2·5·1·18) but no direct link was shown between the other XME and the fermentation profile.

Absorption and endogenous excretion of P by male broiler chicks (14–29 d old) were quantitatively evaluated at different Ca: P ratios (1, 1:1; 2, 1·5:1; 3, 2:1; 4, 2·5:1) in four groups given experimental diets ad lib. The P content was the same in all diets. An isotope-dilution technique was used to determine endogenous faecal and renal excretion. Ca and P retentions in the whole body were estimated according to the comparative slaughter technique. P absorption was calculated from retention and endogenous excretion. Absorption and endogenous excretion of P amounted to (mg P/d per chick): 304, 270,160 and 158; and 135, 109, 31 and 30 in groups 1, 2, 3 and 4 respectively. Widening of the Ca:P ratio in the feed limited the P absorption. Availability of feed P amounted to (%): (1) 66, (2) 57, (3) 32 and (4) 30, and the amounts of absorbed P retained were (%): (1) 56, (2) 60, (3 and 4) 81. The increasing Ca concentration in the feed showed a greater effect on P absorption than on P retention. The ratios of relative retention to relative endogenous excretion of absorbed P were: (1) 1·27, (2) 1·50, (3 and 4) 4·26.

The effects of dietary P restriction to half the recommended minimum level on growth, bone and renal mineralization and urinary composition were studied in female kittens. In two separate experiments, 8-week-old weanling kittens were fed on purified diets containing either 4·6 or 9·2 mmol P/MJ (2·8 or 5·6 g P/kg diet). In the second experiment there was an additional low-P diet in which the Ca concentration was reduced from 9·5 to 4·8 mmol/MJ (7·5 v. 3.8 g Ca/kg diet). P restriction slightly but systematically reduced weight gain (to a maximum of 16%) and growth of the tibia (by 1–4%); the former effect was statistically significant (P < 0·05) between the ages of 15 and 20 weeks in Expt 1 only, and the latter did not reach statistical significance at any time point (P g 0·13). No adverse effect of P restriction was found on mineralization of femur at the age of 39 weeks. Kidney Ca concentrations were significantly lowered (Expt 1, 6 v. 20/μmol/g dry weight, P < 0-001; Expt 2, 7 v. 16/μmol/g dry weight, P < 0-01) in cats fed on the low-P diets, this effect not being affected by the dietary Ca:P ratio. Urinary P concentration was significantly depressed (by 50–96%) after feeding the low-P diets (P < 0-001). P intake did not influence P, Ca and Mg retention during the period of 15 to 39 weeks of age.