Mechanism of action of pyrogen.

1. In unanaesthetized rabbits the cerebral ventricles were perfused for 30-75 min from left lateral ventricle to cisterna magna with solutions of different composition, whilst rectal temperature was continuously recorded.2. Temperature did not rise during the perfusion when the perfusing fluid consisted of artificial c.s.f.; it did rise, however, when the perfusing fluid consisted merely of a 0.9% sodium chloride solution.3. Temperature fell, though not in all rabbits, during the perfusion when the calcium in the perfusing fluid was increased from 1.25 mM, the concentration in c.s.f. to 5 mM.4. Magnesium chloride had only a weak action, in comparison to calcium, in preventing the rise produced during perfusion with 0.9% sodium chloride solution. In a concentration of 1.25 mM it had no effect, but in a concentration of 5 mM it delayed and greatly reduced the rise.5. Temperature did not rise during perfusion with an isotonic sucrose solution.6. The rise in temperature produced by an intravenous injection of leucocyte pyrogen was not prevented when the injection was made during a perfusion with artificial c.s.f., but it was prevented when the calcium concentration in the perfusing fluid was raised to 5 mM or when the perfusing fluid consisted of isotonic sucrose solution. Again, magnesium had only a weak action in comparison to calcium.7. These results support the theory put forward recently (Feldberg, Myers & Veale, 1970) that the constancy of temperature depends upon the physiological balance of sodium and calcium ions in the anterior hypothalamus, that the calcium ions act as a kind of ;brake' preventing the sodium ions from exerting their temperature raising effect, and that pyrogen acts by removing the ;calcium brake', the pyrogen fever thus being a sodium fever.