60 APPLICATION OF PRINCIPLES OF IIEKEDITY TO BREEDING. 



to the shade of coloring, or a variety may change in size, and so on. 

 Again, a character may become latent, possibly lost entirely. Thus, 

 a purple-flowered species by the loss or latency of a factor for purple, 

 may become red. Again, a red flower might become purple by the 

 revival of the latent factor for purple. 



It is probably safe to say that most evolutionary changes are of 

 the classes mentioned in the preceding paragraph. Take, for instance, 

 the color of wfld mammals. Nearly all mammals, so far as they have 

 been studied, have the same factors for color. The differences in the 

 colors of the different species have come about simply by modiflca- 

 tions in these same factors. Yet, some evolutionary changes result 

 in the development of new characters. Beards on grasses must 

 at one time have been new organs. But changes of this kind are 

 comparatively rare, and occur so seldom that we can take little 

 cognizance of them in practical breeding work. 



As stated before, we do not know the cause of these changes. 

 One school of biologists maintains that evolutionary changes are 

 slow and gradual, another that they take place by instantaneous 

 steps, which may be large or small — that is, that they are 'discon- 

 tinuous." We are more interested here in the amount of change 

 that may occur in a given time than in the manner in which such 

 changes take place. The important point is that when evolutionary 

 changes do occur they are usually permanent changes, and the new 

 forms resulting are subject to the laws of selection and hybridization 

 which have already been outlined. That these permanent changes 

 do occur can not be questioned. That in general they are merely 

 changes in hereditary characters already present is equaUy certain. 

 Doctor Nilsson in his work with the cereals at Svalof has many times 

 taken an unselected lot of seed from a standard variety of field 

 grain and found it in the main to consist of a large number of fijced 

 types differing from each other in various ways. When the same 

 character is studied throughout the numerous strains that occur in 

 a field it is found to present nearly every possible gradation in 

 different strains, but generally speaking in each of the strains the 

 gradation found is fixed. 



Jennings, in his study of Paramecium, found in wild cultures almost 

 an. indefinite number of strains, each differing permanently in size, 

 and these differences undoubtedly are due to permanent changes 

 of hereditary characters, as in the case of Nilsson's cereals. Jennings's 

 investigations indicate that there might be found in Paramecium 

 almost every gradation in size, but that the size of each particular 

 strain is fijced. 



The principal relation of these changes to the work of the plant 

 breeder lies in the fact that a crop as grown under field conditions will 



165 



