THE CIRCULATION OF THE BLOOD 323 



upon the blood inside them which keeps up the flow through the 

 capillaries, the heart serving to keep the big arteries tightly filled 

 and so to call the elastic reaction of their walls into play. The 

 whole circulation depends primarily, of course, upon the beat of 

 the heart, but this only indirectly governs the capillary flow, and 

 since the latter is the aim of the whole vascular apparatus, it 

 is of great importance to know as much as possible about arterial 

 pressure; not only how great it is on the average, but how it is 

 altered in different vessels in various circumstances so as to make 

 the flow through the capillaries of a given part greater or less 

 according to circumstances ; for, as blushing and pallor of the face 

 (which frequently occur without any change in the skin elsewhere) 

 prove, the quantity of blood flowing through a given part is not 

 always the same, nor is it always increased or diminished in all 

 parts of the Body at the same time. Most of what we know about 

 arterial pressure has been ascertained by experiments made upon 

 the lower animals, from which deductions are then made concern- 

 ing what happens in man, since Anatomy shows that the circula- 

 tory organs are arranged upon the same plan in all the mammalia. 

 A great deal can, however, be learnt by studying the flow of liq- 

 uids through ordinary elastic tubes. Suppose we have a set of 

 such (Fig. Ill) supplied at one point with a pump, c, possessing 

 valves of entry and exit which open only in the direction indi- 

 cated by the arrows, and that the whole system is slightly over- 

 filled with liquid so that its elastic walls are slightly stretched. 

 These will in consequence press upon the liquid inside them and 

 the amount of this pressure will be indicated by the gauges; so 

 long as the pump is at rest it will be the same everywhere (and 

 therefore equal in the gauges on B and A), since liquid in a set of 

 horizontal tubes communicating freely, as these do at D, always 

 distributes itself so that the pressure upon it is everywhere the 

 same. Let the pump c now contract once, and then dilate: dur- 

 ing the contraction it will empty itself into B and during the dila- 

 tation fill itself from A. Consequently the pressure in B, indi- 

 cated by the gauge x, will rise and that in A will fall. But very 

 rapidly the liquid will redistribute itself from B to A through D, 

 until it again exists everywhere under the same pressure. Every 

 time the pump works there will occur a similar series of phenom- 

 ena, and there will be a disturbance of equilibrium causing a 



