M. N. LEVY 
657 
arterial blood pressure was also recorded from 
a brachial artery by means of a Statham 
P23AA strain gauge. 
One series of experiments was conducted on 
11 dogs in which the circulatory system was al- 
lowed to remain intact, except for third-degree 
heart block. The ventricles were paced with 
square wave electrical pulses delivered from a 
Grass S4 stimulator. The time of each pacing 
stimulus was recorded on the Brush oscillo- 
graph. Also, the atrial electrogram and the pac- 
ing stimulus served as inputs to a parallel logic 
analog computer (EAI 580). The computer cal- 
culated the P-P interval (time between atrial 
activations) and P-St interval (time from the 
beginning of the P wave to the pacing stimu- 
lus) on a beat-by-beat basis, and the results 
were recorded on the oscillograph. A record 
taken at fast paper speed is shown in Figure 3. 
The tall spikes in the upper tracing (labelled P) 
are the spontaneous atrial activations. Between 
the P waves are small deflections which repre- 
sent the current spread of the ventricular pac- 
ing stimulus, followed by the QRS complex (re- 
corded from the atrial cavity, and hence much 
smaller than the P wave). The ventricular pac- 
ing stimulus appears in the middle channel, la- 
p 
St -I — I — I — I — I — I — I — I — I — \ — I — i — I — I — I — [ 
1000 — 
P-St 
MSEC 
Figure 3. — Record from a typical experiment showing 
the atrial electrogram (labelled P), the ventricular 
pacing stimulus event marker (St), and the P-St in- 
tervals, in msec. The small deflections between the 
atrial activations (tall spikes) represent the pacing 
stimulus artifact and the subsequent QRS complex, 
which are much smaller than the atrial activation de- 
flections since the recording electrodes were located 
in the right atrial cavity. The P-St interval is re- 
corded at the time of each atrial activation and de- 
notes the P-St interval for the preceding cardiac 
cycle. The time marker at the top of the record de- 
notes 1 sec intervals. 
belled St. The bottom tracing shows the com- 
puted P-St intervals, which are recorded at the 
time of each P wave for the preceding cycle. 
In each experiment in this first series, the 
ventricular pacing frequency was varied in a 
stepwise fashion over a range of values near the 
spontaneous atrial contraction frequency. In 
some experiments, observed variations in blood 
pressure were temporarily attenuated by un- 
clamping a tube which connected a wide-bore 
cannula in the left subclavian artery to a blood 
reservoir which was elevated to a hydrostatic 
level which was approximately equal to the pre- 
vailing mean arterial pressure of the animal. In 
many of these experiments, these procedures 
were repeated after bilateral vagotomy, bilat- 
eral stellatectomy, or both. 
In a second series of experiments conducted 
on 7 animals, the circulatory system was not in- 
tact, but the descending thoracic aorta was li- 
gated at the level of the fifth rib. A wide-bore 
cannula was inserted retrograde just proximal 
to the ligature, and the blood pressure of the 
proximal portion of the arterial system was 
regulated by means of a servo-controlled pump. 
The regulation of the blood pressure was gov- 
erned by the analog computer such that the 
maximum blood pressure was produced at an 
optimal P-R interval. The shape of the function 
describing the relationship between the arterial 
blood pressure and the P-R interval was pro- 
grammed to resemble that described by 
Brockman*'^ for dogs with complete heart 
block. The analog computer program for accom- 
plishing this has been described previously.^ 
The program permits establishment of the op- 
timal P-R interval and the rate of change of 
blood pressure as a function of the deviation of 
the P-R interval from this optimal value. 
The same computer-generated function could 
also be used to control the frequency of cardiac 
neural stimulation. After the procedures involv- 
ing the changes in blood pressure described 
above were completed, the vagi were sectioned 
in the neck, and shielded bipolar platinum elec- 
trodes were applied to the distal segments of 
these nerves. On a beat-by-beat basis, the fre- 
quency of vagal stimulation was varied as a 
function of the deviation of the P-R interval 
from the optimum value. 
