Arginase synthesis appears to be under the control of at least two distinct processes. In addition to substrate induction which operates through a Jacob and Monod mechanism, one may define a process of nitrogen catabolite repression which is a part of the ammonium effect. The two processes show a large degree of independence most specifically shown by mutations which independently cancel each of the two mechanisms without affecting the gross physiology. In this line we show that controversial conclusions can be reached when using different approaches to provoke catabolite derepression. The nitrogen catabolite repression is abolished by the gdhA− mutation which affects the structural gene for the NADP-specific glutamate dehydrogenase. The effect of the gdhA− mutation is retained when the metabolic defect is compensated by addition of glutamate as well as the best nitrogen nutrients, glutamine and asparagine. This control of arginase and of some, but not all, other nitrogen catabolic enzymes may operate at constant levels of glutamate dehydrogenase (NADP). It occurs when the glutamate dehydrogenase (NADP) is in its catalytically active state, which requires the simultaneous presence of two substrates, NH+4 and 2-oxoglutarate. In fungi the glutamate dehydrogenase (NADP) is the first enzyme of the assimilatory pathway of ammonium, which, when functioning, makes arginase unnecessary. However the regulatory function seems to operate by an intrinsic modification of the enzyme itself rather than by the result of its catalytic action. Carbon catabolite repression appears largely independent from nitrogen catabolite repression and vice versa. However, arginase synthesis is enhanced in mutants lacking aconitase as well as by very poor carbon nutrition. Both conditions decrease the level of 2-oxoglutarate which is required for the expression of nitrogen catabolite repression. Once specific induction and nitrogen catabolite repression are clearly defined, it becomes possible to show that additional regulatory processes are involved in arginase sythesis (Wiame, 1973).