362 VICTOR E. SHELFORD. 



true to fact rather than to schemes. Figuratively speaking, an 

 ecological classification cuts taxonomy vertically, showing many 

 structural adaptations as matters of stratum. It also cuts it again 

 horizontally, showing ecological similarity in organisms, struc- 

 turally and phylogenetically diverse. It therefore provides a 

 new and different means of organization of data. 



What is the significance in the fact (Shelf ord, 'I2 1 ) that Cicindela 

 lepida belongs to the ecological group, the cottonwood association, 

 which we may say corresponds to an order, and to the subterranean 

 ground stratum (corresponding to family) and to the Cicindela 

 lepida mores? Furthermore that Cicindela lecontei and Cicindela 

 sexguttata belong to respective different and older situations or 

 associations? We note that the habitats in which the species 

 occur are characterized by distinctly different soils, moisture, 

 amounts of shade and light. We note furthermore that these 

 animals are possessed of unusual powers of flight and are able 

 to select conditions suited to their physiological constitution. 

 Their mores characters are definite characters, which can be 

 measured in terms of reactions to measured complexes of physical 

 and other environmental factors. They are as clearly defined 

 as any morphological taxonomic characters and can be meas- 

 ured with the accuracy of physical phenomena. 



Doubtless to the student of genetics, the question of the origin 

 of such characters and their fixation in heredity is a leading 

 question. At this point we know little or nothing. Since nearly 

 all species have definite habitat preferences and since many 

 varieties differ slightly from the related species form in the matter 

 of habitat preference, it is probable that origin of a slight change 

 in habitat preference, meaning a slight change in reaction to 

 physical factors, a change in ecological optimum, is usually an 

 early correlative of the origin of new races. Still the so-called 

 taxonomic characters may remain apparently unchanged while 

 marked changes in habitat preference and in reaction to physical 

 factors are being brought about in plastic animals (Allee, '12, 

 p. 341). On the other hand, the segregation in the so-called pure 

 lines and races, accomplished in experimental breeding, often 

 appears to take place without any regard to environment (Cock- 

 erell, '08, p. 547). These two facts accepted as they stand are in 



