390 



Prof. W. M. Bayliss. 



and suspending the deposit of corpuscles in a sufficient volume of 7-per-cent. 

 gum solution to make up the volume removed. Since, as Table II shows, 

 the viscosity of the blood is mainly due to the corpuscles, such a solution as 

 that mentioned would have a considerably greater viscosity than that of the 

 blood, because that of the gum solution alone is equal to that of blood. It 

 is somewhat surprising, however, that these solutions of high viscosity have 

 no better effect than those whose viscosity is no higher than that of blood. 

 It is possible that the heart may be overloaded, although the veuous pressure 

 did not rise higher than 36 mm. of sodium sulphate solution. Even after 

 50 c.c. more gum solution and 25 c.c. of saline, the venous pressure was only 

 60 to 70 mm. of sodium sulphate solution, although in asphyxia it rose to 

 100 mm. We may call to mind the experiments of Evans and Ogawa(13), 

 in which it was found that the output of the heart-lung preparation was 

 decreased by increase in viscosity of the blood above the normal value. 

 These observers hold that the effect is mainly due to increased resistance to 

 inflow through the tubes supplying the heart. The practical point is that no 

 advantage is to be gained by increasing the viscosity of the blood above its 

 normal value. 



This statement applies also where the blood pressure was low on account 

 of haemorrhage. If the viscosity of the fluid run in to replace the loss was 

 increased by the addition of corpuscles from centrifuged blood, the rise of 

 pressure was no greater than if the solution had only the normal viscosity of 

 blood. It is probable that the accommodation mechanism to be referred to 

 below comes into play under such circumstances. 



My experiments fail to supply an answer to the question why the effect 

 of intravenous injections, even of gum solutions, is so much less lasting 

 when the low pressure is due to vaso-dilatation than it is when due to loss of 

 blood. 



Since the fall of pressure in the former case is due to arterial dilatation, 

 it is natural to test whether the administration of a substance which causes 

 vaso-constriction, such as adrenaline, is what is needed. But the effect of 

 adrenaline is very transitory, so that it would be necessary to give it con- 

 tinuously or in repeated doses. Barium chloride is advocated by Langley (14), 

 and has a more prolonged action. I am able to confirm its value. In doses 

 of 2 mgrm. per kilogramme of body weight in the cat, it does not affect vaso- 

 motor reflexes, and produces a large and prolonged increase in arterial 

 pressure. It may, with advantage, be combined with a moderate amount of 

 gum solution, and given in 1 mgrm. doses, as the following experiment 

 shows : — 



Cat of 1'25 kgrm. weight. Cord cut at foramen magnum and vagi cut. 



