Unquestionably one of the most important tests in the examination of the spinal fluid is that for protein. While normally present in small amounts, excess is probably always pathologic. To be sure, a diagnosis seldom rests on protein determination alone, but, taken together with other tests, a knowledge of the total protein content of the cerebrospinal fluid is of great value in differential diagnosis. It has been the experience of one of us that the tests for protein are often performed inaccurately, and wrongly or insufficiently interpreted. The chief reason for this is that the tests ordinarily used, such as trichloracetic acid, phenol, nitric acid, butyric acid, etc., give a precipitate in normal fluids, and that considerable experience is required to recognize normal from abnormal quantities of precipitated protein. More difficult, even for those well trained in such examinations, is the recording of those tests for intelligent comparison of fluids