SOURCES OF CHANGE 73 



ical countries should be evenly distributed through 

 the year, we might expect the wet- and dry-season 

 forms of butterflies to give place to successive 

 generations of similar individuals. 



The difference would still exist among specific 

 characters that some were due to response of herit- 

 age to factors of the external environment while 

 for the others the internal environment would 

 normally provide the proper stimulus. The same 

 thing is true of characters which we now consider 

 in the establishment of species and their sub- 

 divisions. Sumner's mice demonstrate this fact. 

 The characters which they usually present in size 

 of feet, ears, and tail, are as definitely a response 

 to external conditions as are the changes produced 

 experimentally. 



It will be noted that all of these changes, be 

 they of the "inherited" or of the "acquired" 

 type, deal with degrees of development. Degrees 

 are important. Most species differ in degree rather 

 than in kind from the nearest related forms and 

 such transitions are to be expected first if we ever 

 succeed in producing new species in the laboratory. 

 Nevertheless, change to the extent of complete 

 loss of an existing structure has been noted in the 

 classical example of the mutant, polled Hereford 

 cattle, and almost complete loss in the wingless 

 mutant of Drosophila. Such losses may well be 

 due to accidents of meiosis, and so products of the 



