APPENDIX I 



TRENDS IN EVOLUTION 



THE possibility of the existence of definite trends in certain species leading 

 them to evolve in certain directions rather than in others is indicated by at 

 least two sets of phenomena. 



I have referred (page 40) to the fact that we have some quite complete 

 series of fossils in which are seen gradual modification of structure, the several 

 steps of the modification being so slight as to be of doubtful "selection value." 

 Plate 46 shows such a series in the fossil horses, and Fig. 26, page 108, shows 

 an even more instructive series of fossil Paludina shells. It is difficult to be- 

 lieve that the gradual transformation of the latter was due to some advantage 

 from the possession of a rugose shell, an advantage sufficient to cause the 

 "selection" of each slightly more corrugated variety. This series of shells 

 seems to suggest an inherent trend toward greater rugosity. 



Recent studies of variation have shown that inherent tendencies toward 

 modification in particular directions do exist in at least one species. De Vries 

 in Amsterdam, and MacDougal, at the New York Botanical Gardens, in 

 their careful and extensive experiments in rearing an evening primrose (CEno- 

 thera lamarckiana) , found that the mutants which arose were of certain definite 

 types and that these same types appeared generation after generation in con- 

 siderable numbers (compare page 18). De Vries found seven mutants; 

 MacDougal, fourteen. 



In the Amsterdam garden the mutant albida appeared in four generations 

 from lamarckiana parents, previous to 1902, 15 albida appearing in one gen- 

 eration, 25 in another, n in another, and 5 in another. The mutant nanella 



appeared 5 times in one generation, and in other generations, respectively, 



189 



