THE H.EMORRHAGIC DIATHESIS 133 



We have yet one more provision to refer to. The 

 cells lining the blood-vessels, and the leucocytes 

 themselves, are not immortal. When they die, 

 thrombokinase is shed out, and so thrombin would 

 be formed and induce local clotting. This does 

 actually occur in phlebitis and other forms of 

 venous or arterial thrombosis. In the physiological 

 state, however, the Uver secretes into the blood an 

 antithromhin sufficient in amount to deal with small 

 formations of thrombin, but not sufficient to interfere 

 wath the natural process of arrest of haemorrhage. 



Considerable variations take place in the readiness 

 with which the blood coagulates, and it is often easier 

 to understand why than how this is brought about. 

 For instance, at the end of pregnancy clotting is 

 rapid ; in the diseases mentioned above it is deficient 

 or slow. After a haemorrhage, the fibrinoplastic 

 (clot-forming) power rises quickly. Information may 

 be obtained by means of the coagulimeter, a standard 

 capillary tube into which the blood is sucked up so 

 that the time which it takes soUdif3dng may be 

 measured. It requires some care in practice to avoid 

 variations in the calibre, variations in temperature, 

 the inclusion of Ijnnph or clots, etc. 



Associated \\ith deficient coagulability there is 

 often a tendency to effusions of plasma through the 

 capillary walls on account of the low viscosity of the 

 blood. The symptoms of such a tendency to effusion 

 are liabiUty to chilblains, headaches, nettlerash or 

 patchy oedema, and transient or functional albu- 

 minuria. 



The conversion of fibrinogen into fibrin is only the 



