THE EVOLUTION OF ANIMAL FUNCTION 325 



defined case as the cardiac muscle-cell of the amphibia, we can 

 make but a poor guess at the successive steps of past functional 

 evolution. 



Our present knowledge of the history of animal evolution is, 

 in fact, a singularly lifeless affair. It is as lifeless as was our 

 knowledge of embryological development before the method of 

 experimental morphology came into being. Our only acquaint- 

 ance with the effect of evolution upon the actual working of 

 the organisms concerned is that limited inference, which mor- 

 phological observation will support, as to the functional oppor- 

 tunities enjoyed by various groups of cells. We have hitherto 

 taken little account of the qualification of each cell to use its 

 opportunities. 



Here, then, as it seems to me, is the primary problem of 

 comparative physiology — a problem whose investigation is 

 wholly necessary for the understanding of the evolutionary 

 process. It may be stated in general terms as the question to 

 what extent and along what lines the functional capabilities of 

 animal cells have been changed in the course of evolution. It 

 would not be profitable at this early stage to enter upon any 

 schematic and formal discussion of the many subsidiary ques- 

 tions which are included in this general statement. The better 

 way, perhaps, to define the problem, and to make it more real, 

 will be to attempt some preliminary consideration of the 

 methods of investigation. 



After such constant insistence upon the difference between 

 functional behaviour and functional capability, it is scarcely 

 necessary to repeat the warning that, if the investigation is 

 to keep clear of confusion between questions of function and 

 questions of structure or of the arrangement of parts, functional 

 capability must be tested by experimental methods under chosen 

 standard conditions and stimuli, and must never be inferred 

 from normal behaviour under the special conditions presented 

 by each organism. This inquiry means, in fact, the investigation 

 of the reactions of cells under conditions which will necessarily 

 be in a certain sense artificial, because if they are to be as nearly 

 as possible similar for all cells investigated, they cannot be for 

 all cells the normal conditions of life in the organism. 



A simple example may make this point clear. Suppose that 

 we have observed the excitatory process to proceed with greater 

 rapidity in the motor nerves of the frog than in those of, the 



