ON PHYSIOLOGICAL CHROMOMERES. 623 



must be expected to be far less common than the losses themselves, 

 but already numerous instances have been studied. They indicate 

 a latent or inactive condition of the hereditary units as the internal 

 cause of the external losses. These are, so to say, only apparent. 



The main lines of the evolution of the vegetable and animal 

 kingdoms are evidently due to successive changer of a progressive 

 nature. Each step may have produced a new species, but it seems 

 clear that many more species must have originated than were fit 

 for their surroundings. Unsuccessful or tentative forms must have 

 disappeared after a shorter or longer period of existence, and only 

 those which proved to be fit, or adapted as it is often called, to their 

 environment could be expected to survive and to spread. From 

 these considerations a curious consequence may be derived, viz. that 

 in experimental work and in the field, unsuccessful new types must 

 be expected to occur far less rarely than successful ones. It is there- 

 fore for those, that we must look in our cultures. 



Whenever new characters arise in a progressive way it is clear 

 that some internal change in the hereditary material must preceed 

 the externally visible change. It may be asked whether the two 

 processes follow one another at once, or whether a smaller or longer 

 period may separate them. 



Probably both cases occur; in the latter the new units remain 

 for some time in a latent or inactive condition. From this it may 

 turn into the active state from time to time and thereby cause the 

 phenomenon which is known as repeated mutation and which is 

 of so common occurrence among the changes of our evening primrose. 

 The first, or internal change is usually called premutation; it may 

 be due to quite different external agents than those, which provoke 

 in our gardens the visible mutations from the latent ones. The 

 stimulation which produces the praemutation must evidently be 

 supposed to be of a far more potent character than those, needed 

 for rendering active the previously inactive units. 



We now come to a discussion of the different chromosomes of 

 our Oenothera and of the classes of mutable characters for which 

 they are individually supposed to contain the hereditary factors, 

 and begin with the central group. This contains the large majority 

 of the mutants with an unchanged number of chromosomes, viz. 

 14 in the vegetative cells. These mutants behave, as a rule, as 

 recessives and do not show any definite relation to the typical 

 mutations, which are included in the other chromosomes. 



The most common among them are the dwarfs (0. mut. nanella). 



