102 ORGANISM AND ENVIRONMENT 
dioxide in the alveolar air, we first test it under vary- 
ing conditions so as to make sure of its relative 
stability, and then proceed to investigate its connection 
with and subordination to other normals. Thus we 
find that the normal concentration of carbon dioxide 
in the alveolar air is connected with or subordinate to 
the normal composition of the blood, the normal 
activity of the respiratory centre, heart, kidneys, and 
other organs, the normal composition and amount of 
the food and the normal concentration of oxygen in 
the air. Our general working hypothesis would have 
told in a general way that connections of this kind 
must exist; but special investigation could alone tell 
us how they exist and how one is directly subordinate 
to another. It is this kind of investigation that is 
experimental physiology. The normals of anatomy 
are not mere physical structure, nor are the normals of 
physiology mere averages: they are manifestations of 
the life of an organism regarded as a whole. We 
have seen, for instance, in the case of the alveolar 
carbon dioxide pressure, in the percentage of haemo- 
globin in the blood, in the structure of bone-marrow, 
how a subordinate normal alters as the organism 
adapts itself so as to preserve its more fundamental 
normals under new conditions. In pathological condi- 
tions we find remarkable alterations in subordinate 
normals, and these alterations are undoubtedly the 
expression, to a large extent, of adaptations to the 
altered conditions. Pathological phenomena are not 
mere chance effects of the environment on the organ- 
