It may be difficult to differentiate the possible causes of hyperglycemia in the cat. The following investigation was on transient hyperglycemia in 320 cats. The frequency and degree of stress-hyperglycemia and its relation to different types of primary diseases was proven. Animals with overt diabetes mellitus or pancreatic diseases were not included in this study. Plasma glucose was analyzed with the o-Toluidine- or with the glucose oxidase-method. Plasma immune-reactive insulin (IRI) was estimated by RIA. A glucose concentration of greater than or equal to 140 mg/dl (7.77 mmol/l) in the fasted animal was defined as hyperglycemia. The results of the retrospective investigation show in cats transient hyperglycemia occurs more often (3.2%) than permanent hyperglycemia (0.57%). Contrary to overt diabetes mellitus there is no sex- or age-related predisposition. Glucose values above 300 mg/dl (16.65 mmol/l) with maximal values up to 620 mg/dl (34.42 mmol/l) were estimated in 12% of the animals. In 12 from 19 hyperglycemic animals the basal IRI values were between 4 and 14 microU/ml, 11 animals had a lowered I/G ratio. In the following order of frequency the primary diseases found in combination with transient hyperglycemia were: dys- and stranguria, viral and bacterial infections, gastrointestinal diseases, neoplasia, renal insufficiency, cardiopathies etc. With improvement of the underlying disturbance the stress-related hyperglycemia normalized without insulin therapy within a few days. In conclusion, it is necessary to initially identify basic diseases that may trigger the onset of stress-hyperglycemia in the cat. Neither the extent of the glucose level nor a single plasma-insulin-value are valid indicators of confirming the diagnosis of diabetes mellitus.(ABSTRACT TRUNCATED AT 250 WORDS)