THE method is, in principle, essentially that of Hawkins [1923], which in turn was a modification of that of Cullen [1922], using blood instead of plasma. Hawkins collected 0-25 cc. of blood from a vein in a graduated pipette, delivered it into neutralised saline containing indicator under paraffin, centrifuged down the red cells and matched the supernatant fluid against standard phosphates.Hollo and Weiss [1924] adopted a procedure similar to that of Hawkins but differing in some details.Our method differs from that of Hawkins in that only a single drop, about 20 mm.3, of blood is required. Further, the tube containing the diluted blood is sealed up, the use of liquid paraffin is avoided and loss of C02 cannot occur during the centrifugalisation.The plasma is diluted 30 times, which, as Cullen [1922] has shown, is sufficient to reduce the influence of the proteins on the indicator used to nearly negligible proportions. This dilution also renders the disturbing effect of the colour of the serum inappreciable. The diluting fluid contains sufficient oxygen to complete the oxygenation of the haemoglobin. Accordingly, the PH measured is that of the sample when fully oxygenated, without loss of C02 and cooled to room temperature. We employ capillary blood which, as Lundsgaard and Moller [1922] and Verzar and Keller [1923] have shown, is 90-95 % saturated with oygen. The method may, therefore, be taken as measuring the PH of arterial blood cooled down to 18°The correction to be applied to this measurement to arrive at the PH of the same blood at the temperature of the body will be discussed later.DETALS OF THE METHOD.Requirements.(1) A freshly prepared saline solution containing 0 7 % NaCl and 0-2 % potassium oxalate or 0-2 % sodium fluoride. Shortly before use, one volume of 0-02 % phenol red is added to 10 volumes of saline and the PH adjusted to about 7-5 with N/50 NaOH containing the same amount of indicator. It is kept in a hard glass bottle protected from C02.