32 • MUSCULAR WORK 
connection with the mouthpiece designed for the muscular-work experiments, 
and previously described. The general routine for the no-load experiments 
was the same as that for the hard-work experiments. 
Work experiments. — In these experiments it was necessary to make sure 
that a continual supply of fresh air passed directly by the mouthpiece, so that 
the subject would not rebreathe any of the expired air. A slightly different 
routine was therefore required from that employed in the resting, sitting, and 
coasting experiments. After throwing the valve m (see fig. 3) connecting the 
subject with the main air-supply, the 3-way valve r was immediately turned 
so as to have the air pass by the mouth; if done quickly this valve could be 
turned without producing a disturbance in the circulation of the air. Toward 
the end of the working period, which usually did not last more than 11 
minutes, the valve r was thrown a moment or two before the valve m, and thus 
when this latter valve was finally thrown the conditions were exactly the same 
as when the experiment began. Simultaneously with the throwing of the 
valve m at the beginning and end of the experiment, both counters on the 
ergometer were read and the observations recorded, and the oxygen-meter like- 
wise read and the data recorded. As the oxygen-meter used during these ex- 
periments was not of a self-registering type, an assistant was detailed to 
watch the meter and record each revolution of the drum. Throughout the 
entire work experiment a liberal supply of soapsuds was applied to the mouth- 
piece to insure absence of a leak. 
Experiments with controlled diet. — On a few days it was possible to control 
the diet of the subject by having him take his meals at the laboratory. Since 
other experiments were being carried out in the laboratory at the same time 
with essentially the same diet, it was deemed unnecessary to analyze specially 
the food materials used in the muscular-work experiments. The composition 
of the foods ingested was obtained, therefore, for the most part, either by 
using average values computed from previous analyses, or by the direct 
analysis of simultaneously sampled food products. As there was an uncer- 
tainty about the proper collection of the urine, and the collection of feces 
was absolutely impossible, it was argued that any control of the diet would 
be at best somewhat tentative, and while an attempt was made at one time to 
maintain a carbohydrate-rich diet, and at another time a carbohydrate-low 
diet, the actual data obtained are somewhat questionable. It is believed, 
however, that the subject honestly adhered to the diet given him in the 
laboratory. 
COMPUTATION OF THE ENERGY OUTPUT FROM THE OXYGEN CONSUMPTION. 
In our experiments, while no direct calorimetric measurements were made, 
the oxygen consumption was determined with great care, and the data thus 
secured were used for the computation of the energy output — a method which 
has been frequently employed by a number of investigators. No one has more 
painstakingly worked out the theory and principle of such computations than 
has Zuntz. This method, first elaborated and published by Zuntz ° in 1897, 
has been utilized in several more recent publications. 6 - 
° Zuntz, Pflttger's Archiv f. d. ges. Physiol., 1897, 68, p. 202. 
* Zuntz and Schumburg, Physiologie des Marsches, Berlin, 1901, p. 260; Zuntz, Loewy, Mttller, and Caspari, 
Hohenklima u. Bergwanderungen, Berlin, 1906, p. 96; and Loewy, Oppenheimer's Handbuch der 
Biochemie, Jena, 1911, 4 (1), p. 280. 
