A high level of non-protein nitrogen in the blood is characteristic of the toxemia incident to upper gastrointestinal tract obstruction (1, 2). The explanation for this change is not entirely clear. Whipple, Cooke and Stearns found an increase in nitrogen excretion in the dog after intestinal obstruction (3), and no marked evidence of renal insufficiency (4). They attributed the heaping up of non-protein nitrogen in the blood to accelerated protein destruction caused by some toxic body arising in the wall of the obstructed gut. More recently there has been a tendency to doubt the existence of a toxic body and to consider the increase in the non-protein nitrogen of the blood as a true retention due entirely to renal insufficiency, possibly caused in turn by dehydration. Convincing proof for this view is lacking. The high blood urea concentration observed in children suffering from water loss produced by gastrointestinal tract disturbances, has also been interpreted as due to renal insufficiency without any adequate proof (5). Gamble and Ross (6) in studying experimentally the dehydration of rabbits, after pyloric obstruction, make a similar assumption.The question can be settled only by determining accurately the nitrogen excretion. In experimental pyloric and intestinal obstruction this is most difficult, since collections of urine can seldom be accurately made on account of vomiting. We have shown that the toxemia incident to obstruction of the cardiac end of the stomach is similar to but more marked than that following high intestinal or pyloric obstruction (7). Since here no vomiting is possible, the urine can be quantitatively collected. We have also found that if the pylorus is obstructed for 48 hours and then released, although the vomiting