A study was made of all 587 new patients attending an accident and emergency department in the East End of London during one week. Two hundred and twenty-six (39%) cases were not accidents or emergencies; of these, 67% were self-referrals who had not previously seen their general practitioner (GP) and 21% were self-referrals who had previously seen their GP. The four main reasons that these self-referred patients had for attending were that they thought their condition needed immediate attention; they were insufficiently organized to see their GP; they were not registered with a GP; or they wanted a second opinion. Twenty-eight (12%) of the cases which were not accidents or emergencies were referred by a GP. Sixty-nine (12%) of all cases were not registered with a GP. The frequency of cases who were not accidents or emergencies was significantly higher in those not registered than in those registered (0.01 greater than P greater than 0.001). Nineteen (3%) patients were living rough or in hostels. Little abuse of the ambulance service was found.

1.To determine the part played by changes in interstitial tissue compliance in the mechanism of experimental renal hypertension, pressure and volume measurements were made in rats 60 days after partial constriction of one renal artery with a clip and removal of the opposite kidney. Tissue pressure and venous pressure were found to be significantly higher in rats which developed systolic arterial blood pressure of 140 mmHg or above than in loosely clipped controls with normal blood pressure. Plasma volume (PV) was also increased in the hypertensive animals, but there was no difference in interstitial fluid volume (IFV); the PV/IFV ratio was significantly higher in the hypertensive rats.2. Removal of the clip in the hypertensive rats restored blood pressure to normal in 1 h; this was associated with a fall in venous pressure and plasma volume. Tissue pressure fell despite a rise in interstitial fluid volume; the PV/IFV ratio fell.3. It is suggested that renal artery constriction, possibly by a humoral mechanism, causes a fall in interstitial space compliance and that this, by causing changes in body fluid distribution, plays a part in the mechanism of experimental renal hypertension. This mechanism may have a physiological role in the maintenance of adequate plasma volume during water shortage.The mechanism by which hypertension is sustained in the Goldblatt rat, in which one kidney is removed and the artery to the other is partially constricted, is still uncertain. Since blood levels of renin are raised soon after renal artery constriction, but fall to normal as the blood pressure

1.A method for measuring interstitial 'fluid' pressure by using a wick consisting of long-stranded cotton wool at the interface of the tissue is described.2. A correct measurement of hydrostatic fluid pressure was obtained when the wick, connected to a suitable transducer, was applied to filter paper in which the channels contained fluid at a known subatmospheric pressure.3. A mean subatmospheric pressure of -1,6 cmHzO was recorded in the subcutaneous tissue of the normally hydrated frog; pressure fell with dehydration and rose with overhydration. 4. A mean subatmospheric pressure of -2·8 cmHzO was recorded in the subcutaneous tissues of the abdominal wall and scalp of the normally hydrated rat. Simultaneous measurements made at symmetrical sites showed a high degree of correlation. 5. A comparison of interstitial 'fluid' pressure in the subcutaneous tissues of the scalp (measured by a wick) and the abdominal wall (measured by a Guyton capsule), in both anaesthetized and conscious rats showed some degree of correlation. There was, however, a wide scatter of values.6. The interstitial 'fluid' pressure in the rat, measured by both wick and capsule, became more negative when the animals underwent frusemide diuresis; the capsule pressures fell more rapidly for a given degree of fluid loss. 7. The wick method was applied to the subcutaneous tissues of the arm in normal man; a mean atmospheric pressure of -3,4 cmHzO was recorded in five subjects. There were no untoward sequelae.8. The forces responsible for the measured pressure have been analysed. The recorded subatmospheric interstitial 'fluid' pressure in the rat rose towards atmospheric pressure and sometimes became positive when the wick was removed, soaked in

S U M M A R Y1. Measurements of plasma volume, haematocrit, extracellular fluid volume, blood pressure, venous pressure and interstitial tissue pressure were made in rats 4 days after unilateral nephrectomy. Measurements were repeated either 4 days after subsequent removal of the remaining kidney, or 4 days after anastomosis of the remaining ureter with the inferior vena cava. Extracellular fluid volume was expanded by giving 0.5% saline by mouth: in one series blood volume was expanded by injections of blood.2. Plasma volume (PV) rose more after bilateral nephrectomy (BN), extracellular fluid volume (ECFV) more after unilateral nephrectomy and ureterocaval anastomosis (UNUCA). The PV/ECFV ratio was significantly higher after BN.3. Blood pressure and venous pressure rose after BN but not after UNUCA. Interstitial tissue pressure (TP) rose more after BN in spite of greater expansion of interstitial fluid volume (IFV = ECFV-PV) after UNUCA. The ratio AIFV/ATP was several times less after BN than after UNUCA. 4. Interstitial space compliance was estimated by measuring changes in IFV and TP 10 min after a saline infusion (compliance = AIFV/ATP). After UNUCA there was little change from values obtained after unilateral nephrectomy alone ; following BN compliance fell severalfold.5. It is suggested that changes in compliance of the interstitial space may be brought about by a substance secreted by the kidney and represent a hitherto undetected mechanism by which plasma volume is maintained constant under different conditions of hydration. Changes in interstitial space compliance may also play a part in the aetiology of renoprival hypertension by raising tissue pressure, and thus venous pressure and cardiac output.