The biological sciences employ a concept of normality that must be dis
tinguished from statistical or value concepts. The concept of normalit
y is presupposed in the standard explications of biological functions,
and it is crucial to the strategy of explanation by approximations in
, for example, physiology. Nevertheless, this concept of normality doe
s not seem to be captured in the language of physics. Thus attempts at
explaining the methodological relationship between the biological sci
ences and the physical sciences by concentrating only on the concept o
f biological function cannot go very far. An analysis of the concept o
f normality is also necessary.