IRREVERSIBLE EVOLUTION PETRONIEVICS. 437 



difficulty of producing new tendencies in the germ by the influence 

 of external causes, and on the other the degree of correlation that 

 would be needed among these tendencies. When it is a question of 

 the reduction of an organ or part, two alternatives must be distin- 

 guished. If the reduction has gone so far that the corresponding 

 tendency in the germ is verging on complete disappearance, the re- 

 duced organ or part will find itself practically in the same condition 

 as if it were already lost. But if their reduction has not reached 

 to such an extreme their evolution in an inverse direction will not 

 be impossible. 



For the second law we must distinguish between the case of a single 

 part and that of a complex organ. The regressive evolution of a sin- 

 gle part, if during this evolution and the preceding progressive evo- 

 lution no change of form has taken place, could clearly lead back to 

 the point where the progressive evolution started. And the regres- 

 sive evolution of a single part, if the corresponding arrangement in 

 the germ is not too much enfeebled, could evidently also be followed 

 by a new progressive evolution. But if a change of form has taken 

 place during the first progressive evolution, and if this change of 

 form has been so great that a change in the corresponding arrange- 

 ment of the germ has been necessary, then neither the regressive evo- 

 lution following the original progressive evolution will be able to 

 lead back to the point of departure of the latter, nor will a new pro- 

 gressive evolution have the power to accomplish it, because to do so 

 would necessitate the return to a disappeared condition. If, for 

 instance, a tooth has first increased in size and afterwards dimin- 

 ished without change of form this tooth will be able by diminution 

 to assume the dimensions which it had at the beginning of its in- 

 crease, and a new increase of the same kind will not be impossible 

 (if the reduction has not gone too far). But if the increase in size 

 has been accompanied by a radical change of form, if, for instance, 

 a conical tooth has become laterally compressed, then a return to the 

 conical form will not be possible either during its diminution 1 or 

 during a new increase in size. 



In the case of a complex organ Dollo asserts that its return to the 

 previous condition through the action of regressive evolution is im- 

 possible on account of the " indestructibility of the past." But if 

 a single reduced part of an organ is regarded as the supposed reason 

 why reversibility is impossible, we are able to affirm almost with cer- 

 tainty that in such a case the indestructibility of the past does not 

 exist, because it would find itself in contradiction with the well es- 



1 This impossibility is exactly what W. D. Matthew supposes to have happened during 

 the evolution of the Fclidac when he supposes that the felines come from Dinictis, a 

 primitive saber-toothed cat (see W. D. Matthew, " The Thylogeny of the Felidae," Bull. 

 Amer. Mus. Nat. Hist., vol. 28, 1910, p. 290 s). Scott has clearly had a glimpse of the 

 fact that this phylogeny contradicts the law of irreversible evolution (see W. B. Scott, 

 28, p. 540 s). 



136650°— 20 29 



