Xo. 496] ASPECTS OF THE SPECIES QUESTION 259 



along three parallel lines, which have been designated 

 as (1) natural experiments, where advantage is taken 

 of natural movement into new habitats, (2) habitat cul- 

 tures, where reciprocal plantings are made in original 

 and new habitats, and (3) control cultures, where the 

 experiments are carried on in plant-houses, where they 

 can be controlled and directed at will. In addition to 

 this, definite experiments for the determination and pro- 

 duction of constancy are of fundamental value. In 

 regard to the evaluation of the new form when once 

 produced, it will suffice here to state that this should be 

 concerned both with the individual and with the group 

 of individuals. The proper determination of the latter 

 is the task of biometry, when it has attained the neces- 

 sary development. 



For the preservation of the results obtained by the 

 ecologic methods briefly sketched above, an evolution 

 herbarium is proposed.' It is felt that the usual taxo- 

 nomic herbarium will have its usefulness restricted more 

 and more to the preservation of types, and to the pur- 

 poses of instruction. The evolution herbarium will be 



A species or a form will be represented in it by all the 

 variations that can be discovered, and each variation by 

 a large number of individuals. The new forms produced 

 hy nature, and by experiment in nature or under control, 

 will likewise be adequately represented. This repre- 

 sentation will take the form not only of the usual dried 

 specimen, but of photographs, drawings, slides, pre- 

 served material, etc. With all this, however, the evolu- 

 tion herbarium is still to be regarded as a record merely. 

 It is not to replace the taxonomic herbarium as a mass 



