The investigation confirms hypoglycaemia and hyperketonaemia as typical for ketosis. A reverse correlation has been found between the level of blood sugar and that of ketone bodies. The changes observed in both blood sugars and ketones have not been shown to manifest any definite relation to the feeding regime, which may not have been perfectly balanced but has supplied metabolizable energy and digestible protein for maintenance and production in excess. Five-day starving of freshly calved, clinically normal cows may cause a dependable drop of the blood sugar level and rise of the ketone bodies to an extent that vary from that observed in ketosis-affected cows showing no clinical picture characteristic of the disease. The glycaemic reaction observed with the glucose-tolerance and the adrenaline tests shows that there is no apparent differences in the carbohydrate metabolism and the glycosynthetic capacity of the liver between diseased and clinically normal animals. The use of the two tests as auxiliary clinical methods has not been effective in registering the deviations in the status of the adrenal cortex and the pituitary gland.