1. The folate and vitamin B12 in milk are strongly bound to minor whey proteins. These binder proteins are present in excess, and so milk has capacity – which varies between milks of different mammalian species – to sequester added cyanocobalamin and folic acid.

2. A study was made of the influence of different milks on uptake of [3H]cyanocobalamin and [3H]folic acid in selected bacteria, mostly of types that are commonly found in the intestine.

3. None of the test cultures required exogenous vitamin B12. Nevertheless, when free cyanocobalamin was added, in seven of nine cultures it was rapidly taken up into the cells, though there were large differences between the different bacterial species in their absorptive capacity. In presence of sow's milk, little or none of the added cyanocobalamin was taken up, even after incubation of the test mixture for 1 h at 37°. The avidity of sow's milk for cyanocobalamin, as judged from its retention of the vitamin against competition by bacterial cells, was greater than that of a preparation of porcine Intrinsic Factor of similar binding capacity.

4. Of ten cultures of bacteria representing seven species, only the five cultures that required exogenous folate took up added folic acid into the cells. Uptake was severely depressed by the simultaneous addition of goat's colostrum, even after incubation of the test mixture for 1 h at 37°.

5. The physiological role of the binders is discussed. It is postulated that they may act in the mammary gland as trapping mechanisms to accumulate the vitamins from blood plasma into milk and in the gut to facilitate their absorption, both directly, and indirectly by preventing their uptake by intestinal micro-organisms. It is concluded that the binders might well influence the vitamin economy in the neonatal period, and the ecology of the gut microflora.

1. The catabolism of valine was estimated in vivo by measurement of the production of labelled CO2 for 2 h after the oral administration of either [U-14C]valine or [I-14C] valine. It was also estimated in vitro in homogenates of liver and muscle incubated with labelled valine. Experiments were performed in rats given diets providing either 215 g (HP) or 25 g (LP) protein per kg diet.

2. The proportion of [U-14C]-valine excreted as 14CO2 was not reduced in rats given the LP diet for 16 d but the excretion of 14CO2 from [I-14C]valine was reduced by 40% in these animals. When rats were transferred from the HP diet to the LP diet there was a reduction in the excretion of 14CO2 from [I-14C]valine; when the diet was changed from LP to HP output of 14CO2 increased to control values.

3. Homogenates of muscle and liver catabolized valine to CO2. Both liver and muscle from rats fed on the LP diet catabolized less [I-14C]valine than tissues from control animals.

4. Valine aminotransferase activity was higher in muscle than in liver, and did not change in tissues from rats fed on the LP diet. In these animals 2-ketoisovaleric acid dehydrogenase activity was reduced in both liver and muscle.

5. The production of 14CO2 was lower with [U-14C]valine as the substrate than with [I-14C]valine and there was no difference between tissues from rats fed on the HP and LP diets.

6. The results with [I-14C]valine suggest that both liver and muscle from protein-depleted rats catabolize valine at a reduced rate. The reason for the discrepancy between these results and those with [U-14C]valine is not clear. It is concluded that the results with [U-14C]valine in vitro are affected by dilution of the label before the formation of 14CO2, but that this does not hold in vivo.

1. A machine is described which is small enough to fit into a respiration calorimeter and which enables sheep to be fed automatically up to twelve times daily.