Evidence has been presented during the past few years that intravenous administration of hypertonic solutions of urea has the effect of reducing intracranial pressure and inducing shrinkage of the brain in patients undergoing neurosurgical treatment (Javid & Settlage, 1956;Stubbs & Pennybacker, 1960). In considering mechanisms by which such effects might arise one possibility which comes to mind is that the urea may act in a similar manner to other solutes, such as sodium chloride, which have similar effects when injected in hypertonic solution (Weed & McKibben, 1919) and which are believed to bring about withdrawal of water from the cerebral tissue by virtue of the increased osmotic pressure of the blood plasma which results from their presence in it. However, an obvious difficulty about accepting such an explanation in the case of urea is the fact that this substance is generally held to distribute itself freely throughout the whole of the intracellular as well as the extracellular water of the human body (McCance & Widdowson, 1951) and, in particular, it appears not to behave as an osmotically active solute in relation to the membrane of the receptor organs responsible for controlling the output of antidiuretic hormone which are believed to lie in the hypothalamic area of the brain (Verney, 1954).The experiments reported in this paper were undertaken in order to determine directly the extent to which intravenously injected urea would penetrate the water of central nervous structures, and hence to study its ability to promote the transfer of water from the brain cells under a simple osmotic gradient. The opportunity was also taken to collect data on changes in water and electrolyte content of brains from animals exposed to constantly elevated concentrations of urea in their blood for periods of several hours. The majority of the experiments were done on cats but isolated observations were also made on a rabbit and a dog.27-2 424 M. W. B. BRADBURY AND R. V. COXON METHODS Preparation of animal. Seventeen adult cats, fourteen males and three females, weighing between 2-0 and 4-8 kg, were used for the main investigation. All animals were deprived of food for 16 hr and were initially anaesthetized with intraperitoneal sodium pentobarbitone 49 mg/kg body wt., supplemented by further small doses of the same drug given intravenously as required. Plastic tubing inserted through the left femoral vein into the inferior vena cava was used for infusions, and the right femoral artery was cannulated to provide for the withdrawal of blood samples. The urethra was then catheterized and the abdomen opened in order that the bladder might be compressed to facilitate complete emptying. For the priming dose 'Urevert' (Baxter Laboratories Ltd.), which is a sterile solution containing 30 g/100 g urea and 6-5 % invert sugar, was injected over 5-10 min into the inferior vena cava, 4-0 ml. being given/kg body wt. of cat. Further urea was administered as a 3 or 6 % solution with 5 % glucose by means of an electrically driven syringe; con...