The relationship between blood pressure and the rate of flow of the blood during the cardiac cycle is one of the most fundamental problems in haemodynamics. The electromagnetic flowmeter is an instrument that not only offers a convenient means of studying the phasic blood flow, but also seems, on theoretical grounds, to be almost ideal. Reliable reports concerning the investigations on phasic changes of rate of blood flow in the cardiac cycle (i.e. the flow pulse) are scarce, especially systemic records by the electromagnetic method (Katz & Kolin, 1938;Richards & Williams, 1953). On the other hand, as has already been discussed by Shipley, Gregg & Schroeder (1943) and by McDonald (1955), there are some discrepancies between patterns of flow pulse recorded by the electromagnetic flowmeter and those recorded by other methods. We have therefore re-examined the reliability of the electromagnetic method in recording phasic flow changes both in vitro and in experimental animals. The results of the experiments in vitro have already been published (Inouye & Kuga, 1954) but records in animals were presented only in a preliminary form in that report. The present paper is concerned principally with the pattern of pulsatile flow recorded by our electromagnetic flowmeter in rabbits and dogs under nearly normal circulatory conditions.
METHODSRabbits weighing 2-3 kg were used to observe carotid flow. A carotid artery was exposed under local anaesthesia, the whole area having been previously infiltrated with cocaine hydrochloride. The femoral arteries were either too narrow to apply our cannulating method or the blood flow through them was too small to allow accurate measurement. Observations of femoral flow were, therefore, made on dogs weighing 5-14 kg, which were anaesthetized by intramuscular injection of 'Ouropan-soda' (N-methyl-cyclohexenyl-methyl barbiturate) in a dose of 0-1-0-2 g/kg. An * Present address: