350 Proceedings of the Royal Society of Edinburgh. 
has led me to modify my first method. By taking the mean instead of the 
lowest point of the curve, the inertia is as far as possible compensated for. 
Fig. 6 is a curve obtained by the method of intermittent escape, and 
illustrates this point quite distinctly. 
In order to ascertain whether the local condition of the circulation 
below the point of constriction might not be demonstrated better by means 
of the plethysmograph than by the transmission sphygmograph, a good 
many tracings have been taken by means of it. One of these is shown in 
fig. 7. It will be seen that on increasing the pressure in the armlet, the 
plethysmograph tracing at once rose to a point above the abscissa, and, 
instead of showing a curve of the pulsations, it only revealed a respiratory 
curve. As the pressure was allowed to fall, the plethysmograph curve fell 
below the abscissa, and at a certain point began to show definite pulsations. 
At this point, however, the curve began to rise to such a degree as to 
render the exact record somewhat difficult, and as, in every case, the same 
difficulty presented itself, the use of the plethysmograph was discontinued. 
There are, undoubtedly, considerable difficulties in the estimation of 
diastolic pressure, since the amplitude of the pulsations, as revealed by the 
kymograph, is subject to so many influences. It is probable that the state 
of the arterioles, as regards contraction and relaxation, is the most powerful 
factor. But there can be no doubt that fluctuations in the energy of the 
heart itself play an important part. Some of the curves which have been 
obtained afford excellent illustrations of these difficulties. Of these, that 
which is shown in fig. 8 is very striking. It will be seen that there are, on 
the descending curve, at least four considerable fluctuations, as well as a 
few smaller ones. These are often very much like Traube-Hering curves, 
and certainly owe their origin to the same influences. 
For practical purposes the observations which are made, whether with 
such a recording apparatus as has been described or any of the ordinary 
sphygmomanometers, should be noted on a graphic chart. It is difficult, 
without such a record, to watch the course of any investigations. Two charts 
of the arterial pressure and pulse rate are to be found in fig. 9 and fig. 10. 
In the former of these, with a normal pulse rate there is an extremely 
high arterial pressure. The facts were obtained from a patient with 
arterial sclerosis, cardiac hypertrophy, and interstitial nephritis. The 
latter records the observations in a case of Addison’s disease, and it will be 
seen that, with a rather high pulse rate, there is an extremely low curve of 
pressure. They are taken from an address recently delivered by me (32). 
To Professor Schafer it is an agreeable duty to express my obligations for 
kind help and valuable advice in regard to many of the matters with which 
