PHYSIOLOGICAL. EEGULATIONS 447 



whole functioning organism and its ancestors would be described 

 in all possible aspects. But approximate description is facilitated 

 by selected circumscription. In this way the breathing of man was 

 studied as an isolated activity, so long as control rates of pulmo- 

 nary ventilation showed variabilities less than the rates that were 

 found after disturbances. In this way the respiratory changes in 

 blood were considered by themselves. The isolation is ordinarily 

 accomplished by defining distinctive conditions so that the correla- 

 tives that are significant are either recognized or controlled. Acu- 

 men in biology is often in the recognition of highly correlated 

 variables and the neglect of others. 



When a biologist studies a ''closed" system he may profitably 

 remember that the act of limiting the independent variables also 

 limits the manifestations of regulations. Thus, the study of the 

 respiratory cycle in blood limits the regulations to those compo- 

 nents that change during this cycle. Limitation is desirable for 

 many reasons; but only on comparing additional states are the 

 regulations exhibited in wider variety. 



In general, each variable and condition brings to light certain 

 aspects of regulations. Identity and similarity of components is 

 in the end judged by arbitrary criteria of resemblance in the quan- 

 titative manifestations observed {e.g., rates of exchange) and in 

 the procedures used in measuring them. 



Continuance of life itself is not always a correlative of recovery. 

 Frogs that were desiccated by 32% of Bo or more, often died (fig. 

 QQ) ; but whether or not they died, water was regained equally fast. 

 Hence death did not depend on inability rapidly to regain water 

 content of the body as a whole. A deficit of water that exceeded 

 the tolerated water load (as judged by survival) did not indicate 

 inability to gain water faster than it was lost. Conversely, recov- 

 ery toward water balance did not mean recovery of all functions 

 required for survival ; and there was no fast relation between the 

 kind of deprivation to which the animal was subjected and the kind 

 of recovery that was required for survival. Other functions than 

 ability to gain net water content were concurrently upset, so that 

 what was described as regulation of water content was insufiScient 

 to insure survival. 



Someone can always assert: the organism is not "primarily" 

 regulating water content, but is adjusting the concentration or the 

 permeability. How does he tell ? There appears to be no criterion 



