Cardiac levels of ATP, glycogen, glucose, glucose 6-phosphate, fructose 6-phosphate, fructose 1,6-diphosphate, triose phosphates, glycerol 3-phosphate, pyruvate, lactate, citrate, S-oxoglutarate, malate, acetoacetate, 3-hydroxybutyrate and free fatty acids, as well as the liver glycogen content were determined simultaneously by instant deep-freezing of the tissues i n situ in urethanechloralose anaesthetized, normal and alloxan-diabetic rats. I n comparison to the mean values obtained in fed, normal rats the following changes in the pattern of substrate levels were found to be typical of: a) Starvation in normal rats : significant increases in citric acid cycle intermediates, in ketone bodies and in cardiac glycogen; an increase in the levels of hexose monophosphates and a decrease in fructose diphosphate, indicating inhibition of phosphofructokinase activity, and also a decrease in glycerol %phosphate and triose phosphates ; increase in malate jpyruvate ratio ; no change in free fatty acids, ATP and pyruvate; and a most marked decrease in liver gIycogen content.b) Acute alloxan diabetes: in general, changes similar to, but greater in magnitude than found in starved, normal rats, except for an unchanged 2-oxoglutarate level ; moreover, significant increases in lactatelpyruvate and 3-hydroxybutyrate/acetoacetate ratios and a concomitant decrease in ATP were recorded. On starvation of acute diabetic rats the changes were attenuated, most significantly in regard to ketone bodies and ATP; the effect of starvation was exactly the opposite of that seen on starvation of normal rats. c) Chronic alloxan diabetes : no significant alterations except for a somewhat raised citrate level, a lowered FDP level with an associated decrease in phosphofructokinase activity, also decreased glycerol 3-phosphate and ATP levels ; liver glycogen content reduced. Starvation produced a picture of general substrate exhaustion, in which citrate, malate and the intermediates of glycolysis reached very low levels although phosphofructokinase activity was high, in association with low acetoacetate values and a decrease in even cardiac glycogen and FFA levels; liver glycogen stores were depleted.Regression analyses were carried out on the above data and it was established that the FDP/F6P ratio showed a highly significant negative correlation to cardiac citrate, but not to ATP levels, under all conditions. Furthermore, ketone-body levels and also the lactate/pyruvate ratio showed a steep positive correlation to citrate above a critical level of 60-70 pmolesj100 g citrate, whilst the 3-hydroxybutyrate/acetoacetate ratio was positively correlated to citrate over the entire range of concentrations. A further aspect was the demonstration of a highly significant negative correlation between cardiac citrate and liver glycogen in normal and in acute diabetic rats, and between cardiac ketone bodies and liver glycogen under all conditions ; cardiac ketone bodies did not accumulate until the liver glycogen stores were depleted below a critical level of a...