1896.] li< [Ortmann. 



controlled by natural selection." If thus mutation is influenced by nat- 

 ural selection, it implies, that any particular mutation must advance in a 

 direction advantageous for the respective species, and, indeed, many 

 examples of mutation known among fossil animals are apparently due to 

 the advantage produced by the change.* I must add here, however 

 that probably not all mutations (in a palaeontological meaning) are due 

 to natural selection, but that many do not imply an actual improvement. 

 In this respect Elmer's investigations of the Papilionidoe are important. 

 The variations in the colors of the wings, on which Eimer exclusively 

 relies, are apparently neither useful nor injurious, yet they are caused 

 most likely by external conditions, for example, by warmth or cold dur- 

 ing the development of the imago from the larva. Eimer points out, 

 that in his butterflies a distinct direction of variation is evident, which he 

 calls " Orthogenesis." We shall see below that this is a process of inher- 

 itance. By the constant action of certain external causes upon subse- 

 quent generations, and the repeated inheritance of the characters thus 

 acquired, a certain tendency of variation in a distinct direction may 

 develop. If this tendency does not bear on utility, the degree of varia- 

 tion in the single individuals difl'ers considerably, and even individuals 

 varying in other directions are preserved. Thus a gradual transition 

 results from the less to the more changed individuals. But altogether, 

 from generation to generation, the variation in that direction increases, 

 and the changed individuals may become the most numerous, thus effect- 

 ing a slow change of the average characters of the species, which looks 

 exactly like a mutation. We may call this latter mutation, produced by 

 accumulative inheritance, by Eimer's term "orthogenesis," in contrast 

 to the "mutation" produced by natural selection. " Orthogenetic 

 mutations " are also known among fossil animals, and I referf especially 

 to the group of Ammonites whose mutations have been first studied. 

 Here most of the characters advancing in certain lines, ornaments and 

 form of the shell, etc., are apparently not subject to natural selection. Of 

 course, we do not know, in most of the cases, whether a particular trans- 

 formation is useful or not, and in many cases, where we cannot recognize 

 any advantage, the latter is pre&ent nevertheless. But since Eimer's 

 investigations have amply proved that such changes, indifferent as 

 regards utility, are certainly present in living animals, they must also 

 have been present in fossil animals X 



* I mention only the example of the transformation of the structure of the extremities 

 in the horse-phylum, as discussed by Scott (I. c, p. 368). With the change of one char- 

 acter in a useful direction the change of others may be connected, which are in correla- 

 tion with tlie first. This would be an indirect action of natural selection." 



t A very illustrative example of " Orthogenesis " is the transformation of the Miocene 

 and Pliocene Fulgur contrarius into the Pliocene and Recent Fulgurperversuf. See Leidy, 

 " Remarks on the Nature of Organic Species," Trans. Wagner Free Inst. Sci., ii, 18S9, 

 p. 51fr., Pis. 9 and 10. 



J Weismann indeed denies, even in respect to Eimer's butterflies, that there are any 

 u.seless variations, but tliis is one of his many assertions, which he does not even try to 

 establish properly (comp. "Germinal Selection," The Mmiist, Vol. 6, No. 2, Jan., 1896, 



