_Cyanosis_ of dark hue and marked degree was prevalent in the earlier severe cases, and in some cases appeared entirely out of proportion to the state of circulation and respiration and to the post-mortem findings as reported by Dr. Klotz. _Blood_ Hemorrhage being not uncommon, the blood was tested for coagulability, but in this respect no marked departure from the normal range was noted. _Coagulation._—Coagulation time was observed by stirring blood in a test tube with a wire and noting the time of the appearance of fibrin and by means of a Biffi-Brooks coagulimeter. The extreme ranges observed were from 2½ to 5½ minutes. The average by defibrination was 3 minutes and 36 seconds, and by the Biffi-Brooks method 4 minutes and 38 seconds. _Red Corpuscles._—Osmotic resistance. A number of bloods were examined by observing their resistance to osmotic laking by exposure to a series of hypotonic sodium chloride solutions. Though some differences were observed, from the evidence obtained, it is not permissible to conclude that such variations were constant or of a significant magnitude. _Color_ on exposure to air. It was early observed that venous blood from cyanotic patients was very slow to take on arterial hue on exposure to air. _Plasma Bicarbonate._—The plasma bicarbonate was determined in seven cases by Miss Waddell by the method of Van Slyke and Cullen. In all except one of these the results were within the normal range as given by Van Slyke. Three were in the lower normal range, being 54.1, 55.1 and 60.5 respectively, expressed in terms of cubic centimeters of CO_{2} reduced to 0°, 760 mm. Hg. pressure, bound as bicarbonate by 100 c.cm. of plasma. Three were in the median range, being 64, 65.5 and 71 c.cm. In one case the bicarbonate CO_{2} was reduced to 46.6 c.cm. There seemed to be no constant relation between the apparent severity of the clinical condition of the patient and the bicarbonate reading. In the one case in which this was found to be reduced below Van Slyke’s lower normal limit the blood was taken only a few hours before death. _Hemoglobin Per Cent._—As determined by the Sahli hemoglobinometer (by Miss Lee) and as estimated by the total oxygen capacity (Van Slyke method) (by Dr. Rohde and Mrs. Macklin), the hemoglobin content ranged within normal levels. _Relative Volume of Corpuscles._—A limited number of hematokrit tests on severe cases gave results in normal levels.