1903-4.] Prof. Schafer on Artificial Respiration in Man. 49 
and 6760 c.c. per minute. It is the only method which, in this 
series of experiments, gave an amount equal to the normal tidal 
air of the individual — which was, in fact, somewhat exceeded. 
Not that it is impossible by other methods (especially those of 
Howard and Marshall Hall) to obtain larger figures for the ex- 
change air than those given in the tables here shown — figures 
equal to or even larger than the tidal air — but merely because it 
is more difficult to do so at the rate of artificial respiration at 
which these experiments were carried on. The most important fact 
which the tables show is that at this rate (which is the normal 
rate of this particular individual, and not by any means a fast 
rate), it is easily possible to pump far more air into and out of the 
chest by the prone-pressure method than by any of the methods 
generally employed. The actual pressure exerted upon the prone 
subject was not greater, probably rather less, than upon the supine 
subject, in which the full weight of the fore part of the operator’s 
body was certainly thrown upon the lower ribs, whereas in the 
similar experiments upon the prone subject the outflow of air on 
making pressure on these ribs was so abundant and easy that there 
was a tendency for the operator not to throw the whole weight on 
the hands; even more air, therefore, could have been exchanged 
if desired. 
Table YII. — The following Table gives the main results of all 
the foregoing Tables in a summarised form. 
Mode of Respiration. 
Number 
per Minute. 
Amount of Air 
exchanged per 
Respiration. 
Amount of Air 
exchanged 
per Minute. 
Natural (supine), 
13 
489 c.c. 
6,460 c.c. 
Natural (prone), 
12*5 
422 „ 
5,240 ,, 
Prone pressure, . 
13 
520 ,, 
6,760 ,, 
Supine pressure, 
13 6 
295 ,, 
4,020 ,, 
Rolling (with pressure), 
13 
254 ,, 
3,300 ,, 
Rolling (without pressure), 
12 
192 ,, 
2,300 ,, 
Traction (with pressure), . 
12*8 
178 ,, 
2,280 ,, 
Results similar in character to the above have been yielded by 
many experiments, both upon the same and upon different individuals. 
These experiments all show that by far the most efficient method 
PROC. ROY. SOC. EDIN. — VOL. XXV. 4 
