In the past ten years there has been an increasing interest in the electrolyte structure of the blood. This interest grew from a realization of the impossibility of interpreting isolated changes in single constituents. Methods for partitioning the anions developed with much greater rapidity than did a similar approach to the cations. In 1923 the authors published with Loeb and Palmer a series of studies on the serum acids including a few base partitions. Although the acid determinations were quite consistent, base results were unsatisfactory due to the unreliability of the sodium and potassium methods. With the development of an accurate method for estimating total cations, a more satisfactory understanding of the total electrolyte structure is possible. This method has evolved slowly and in our hands has not been considered reliable prior to the past three years, when our data became sufficiently consistent to report. We have discarded most of our work before that time.Even before an acceptable method for total base appeared, Van Slyke and co-workers filled the only essential remaining gap in anion analysis by determining a formula for base bound to protein. It is now possible, therefore, to arrive at a fairly accurate picture of the total electrolyte distribution in serum. It is proposed in this paper to discuss the results obtained by studies carried out on normals and patients with pneumonia, nephritic acidosis, nephritic edema, and diabetic acidosis. A few observations during diuresis are included.
265