At first sight it may seem strange that it should be necessary at this date to call attention to some of the earliest observations in clinical hematology. Nevertheless, it is not difficult to show that then, and even now, the principles underlying these observations have not always been kept clearly in mind. It will be remembered that Hayem1called attention to the disproportionately high colorimetric values of the blood in pernicious anemia as compared with the colorimetric power of the same number of normal red blood-corpuscles. Quincke2noted an apparent increase in the average hemoglobin content of the red blood-corpuscles in pernicious anemia. Later Laache3made a systematic investigation of the hemoglobin content and of the number of erythrocytes of the blood in health and in disease, and defined precisely what we now call the color-index of the red blood-corpuscles. Laache obtained his values for