THE INFLUENCE OF CHEMICAL FACTORS 77 



apparatus. The authors assume that these two effects will counter- 

 balance each other, but this assumption is not warranted, and the 

 nitrogen washed out must be in excess of the oxygen taken up from 

 physical causes, because the oxygen pressure in the organism as a 

 whole is low and remains low even while pure oxygen is breathed, 

 whereas the tissues and fluids are normally saturated with nitrogen at 

 a pressure amounting to J of an atmosphere, the whole or almost the 

 whole of which is washed out by breathing oxygen. It seems, there- 

 fore, that breathing of oxygen does increase the metabolism to some 

 slight extent, and it must be borne in mind that a large effect is not 

 in any case to be expected, because the oxygen supply to the tissues 

 will only be increased by the extra amount of oxygen physically 

 dissolved in the arterial blood which cannot be more than 10 per 

 cent, of the total quantity carried. 



The results obtained on warm-blooded animals point, when con- 

 sidered in their entirety, in the opinion of the writer, towards the con- 

 clusion that the oxygen pressure is practically the limiting factor for 

 the oxidations, but that it is so regulated as to be just sufficient. A 

 diminution of the oxygen supply to the tissues, which will take place 

 when the oxygen pressure in the inspired air falls below something 

 like 85 mm., causes a decrease in the rate of oxidation, while an in- 

 crease in the oxygen pressure appears to produce a slight increase. 



That the oxygen tension in the tissues must be the limiting factor 

 for the oxidations would be a self-evident proposition if, as is very 

 generally believed, this tension were practically nil. It should be 

 pointed out, however, that the direct evidence in favour of such a view 

 is extremely defective and uncertain. It is now generally admitted 

 that the method invented by Ehrlich [1885] to determine the presence 

 or absence of free oxygen is untrustworthy, as the stains employed 

 may very well become reduced even if free molecular oxygen is pre- 

 sent. The determinations of oxygen quantities (PflUger [1868, 1870], 

 Hammarsten [1871]) or tensions (Strassburg [1872], Fredericq 

 [1911]) in secretions, which have in most cases (though not always) 

 shown that oxygen was practically absent, are untrustworthy, because 

 oxygen is used up in the fluids in question. This was shown by 

 Pfluger [1870] for milk and bile and has been observed by the writer 

 for urine (unpublished experiments). The oxygen tension of freshly 

 voided urine amounted in one experiment to about 20 mm., but the 

 oxygen disappeared completely in less than half an hour. In very 

 dilute urine a tension of about 35 mm. was determined, and as it was 



