E. DONG, G. C. M. WIEDERHOLD 
1097 
C..S. 13 MONTHS PDST-OP 
200 
175 - 
150 
0 
0 
I 
100 
75 
50 
X 
o 
c 
1 p'-n r 
0 - RECIPIENT HERRT RRTE 
+ = DONOR HERRT RHTE 
^ CD 
CD 
CD 
(T) CD 
71 -i<i 
CD 
J I \ I I 1 1 I -Lr- y l I 
XX 
* % X 
XX _ X 
o 
a 
in 
5 
2 3 4. 5 6 7 8 9 tO^-^ t^O 
Minut-cs oP exercise Minufcs 
aPtcr exercise 
Figure 9. — Heart frequency response in exercise. Heart response to cessation of stimulation. 
struck by the observation that the rate con- 
stants after the first two beats were relatively 
consistent, but the change of heart period im- 
mediately following the inhibited beat were con- 
sistently greater. These plots shown in figure 
10 calculated the rate constant, plotted the ex- 
perimental data, the best fit exponential and 
the residual. 
A single exponential will increase the overall 
residual error, fitting neither the early nor late 
phases. 
In pursuit of examining the heart rate con- 
trol by the vagus, we have demonstrated the 
following features: (1) experimental control, 
(2) data storage and measurement, (3) graphic 
display, (4) curve fitting, and (5) use of sta- 
tistical libraries. In order to formulate our un- 
derstanding of the process involved we now 
turn to model building. 
A well-known model of nerve potential in the 
axon has been proposed by Hodkin and Huxley." 
This model relates alterations in cell membrane 
permeability to membrane potential. Noble, 
after experiments on Purkinje tissue, modified 
the axon model to fix the properties of the car- 
diac pacemaker — the spontaneous depolariza- 
tion leading to rhythmicity.^*^ " We implemented 
this model on ACME. 
Trautwein has demonstrated that acetylcho- 
line causes a marked change in potassium per- 
meability in cardiac tissue which is the basis 
of the inhibitory effect of ACH.^^ We therefore 
perturbed potassium permeability in the Noble 
model with a pulse decayed with a single ex- 
ponential. You must remember that the Noble 
model was built to stimulate the membrane 
potential and not the beat-to-beat heart rate. 
Two problems arose. First, when given a time 
constant of 2.7 seconds, the model went into ar- 
rest with multiple inhibitions. Second, when the 
model was driven as if excited by an electrical 
simulus, the model returned to the control heart 
rate within one beat after discontinuation of 
the stimulus. These data were modeled by add- 
ing an RC circuit across the model.^^ This modi- 
fication suggests that during pacing there is a 
net loss of positive charge. Following cessation 
