When glucose is added to cells of the yeast Succharomyces cerevisiae grown on non-fermentable carbon sources, a cAMP signal is induced which triggers a protein phosphorylation cascade. Addition of glucose or fructose to cells of a phosphoglucose isomerase mutant also induced the cAMP signal indicating that metabolization of the sugar beyond the sugar phosphate step is not necessary. Glucose 6-phosphate might stimulate the triggering reaction since induction with fructose shows a significant delay. Experiments with double and triple mutants in hexokinase 1, hexokinase 2 or glucokinase indicated that the presence of one of the three kinases was both necessary and enough for induction of the cAMP signal by glucose and the presence of one of the two hexokinases necessary and enough for induction by fructose. The product of the kinase reaction itself however does not appear to be the trigger of the reaction: when the increase in the level of glucose 6-phosphate and fructose 6-phosphate was measured as a function of time after addition of different glucose concentrations, no correlation was observed with the increase in the cAMP level. From the dependence of the cAMP increase on the external concentration of glucose, a rough estimate was obtained of the K , of the triggering reaction: about 25 mM. This value clearly fits with the K , of the low-affinity glucose carrier (about 20 mM) and differs by at least an order of magnitude from the K , values of the high-affinity glucose carrier and the three kinases. The present results situate the primary triggering reaction at the level of transport-associated phosphorylation. The main (= low-affinity) glucose carrier appears to be the receptor while association of the corresponding kinase is needed for induction of the signal. Since it is known that the presence of the kinases influences the characteristics of sugar transport, no definite conclusion can be given on whether the necessity of the kinases reflects the need for a certain type of transport or the need for phosphorylation of the sugar. The increase in the level of fructose 1,6-bisphosphate, on the other hand, correlated very well with the cAMP increase. However, it clearly lagged behind the cAMP increase, confirming the previously suggested importance of the cAMP signal for the stimulation of glycolytic flux at the level of phosphofructokinase 1. The importance of the cAMP signal for the stimulation of phosphofructokinase 1 also provides an explanation for the transient overshoot in the levels of glucose 6-phosphate and fructose 6-phosphate which are observed after addition of glucose to derepressed yeast cells. Addition of glucose to glucoserepressed wild-type cells triggers no or just a very weak cAMP signal, indicating that one of the intermediates in the induction sequence must be glucose repressible. This conclusion was confirmed by experiments with wild-type cells grown on either galactose or maltose (which have less glucose repression) and with cells of the hxk2 mutant which is deficient in glucose repression...