PRESENT TENDENCIES OF MORPHOLOGY 271 



ical sciences and the results which may follow for the later develop- 

 ment of these sciences. 



An experiment always necessitates the preliminary analysis of the 

 phenomena by which the fact that one wishes to observe and if pos- 

 sible to measure is conditioned. It assumes a hypothetical solution 

 of which it will show the reality or the non-existence. Every experi- 

 ment is then preceded by an induction and followed by one or more 

 observations. The experimental method is always, as Chevreul called 

 it, a method a posteriori. 



Experiment creates nothing; it has precisely the same value and 

 the same logical significance as the proof of a mathematical opera- 

 tion. 



For an experiment it is not necessary to demand, as some seem to 

 believe, a complicated plan, a richly equipped laboratory and costly 

 apparatus. 



It is necessary indeed not to confound the precise measure of a 

 phenomenon which often is obtained by the aid of very delicate in- 

 struments w r ith the pure and simple establishment of a causal relation 

 between one fact and the other facts which determine it, an establish- 

 ment which is the basis of the experiment itself. Even if the fact were 

 accidental, as the fall of the apple before the eyes of Newton, its de- 

 termination may nevertheless become an experiment. And it is only 

 the mind of the observer which will give it this character. 



Das 1st ja was den Menschen zieret 

 Und dazu ward ihm der Verstand 

 Dasz im innern Herzen spiiret 

 Was er erschafft mit seiner Hand. 



Where the unscientific person sees without interpreting and 

 takes a purely contemplative attitude, the naturalist worthy of the 

 name supplies the supposition of voluntary acts the action of whose 

 factors he wishes to study. 



An animal receives in the hunt or by some other accident a ball 

 in the left side of the neck; the right side is paralyzed. If the fact 

 is well determined and freed from all cause of error its voluntary 

 reproduction in the laboratory would be only the verification of 

 an experiment already realized. 



Not only does nature at present offer us, as we have said-, nu- 

 merous experiments, many of which are very difficult to repeat, 

 but we may also say that paleontology furnishes us experimental 

 data of incalculable value. The arguments which it furnishes to 

 transformational morphology are not, as is sometimes pretended, 

 purely conjectural; the degree of certainty that they possess is not 

 inferior to that in astronomy or in the other divisions of the phys- 

 ical sciences whose objects are partly inaccessible to us. 



