338 C. E. MCCLUNG. 



heterogeneous subtypes. We do not know certainly that this is 

 an occurrence of exactly the same nature as the present day 

 observed mutation, but it is of the same character and is an evi- 

 dence that stability of the germ plasm is disturbed by irreversible 

 changes which, as I have pointed out, may find expression in 

 different arrangements of the chromosomes. 



Opposed to the sudden and pronounced changes in character 

 combinations that mark instances of discontinuous variation stand 

 the common examples of continuous variation, where there is 

 present a graded series of inconstant and fluctuating variations. 

 To my mind the two types of variation suggest different 

 mechanisms of transmission. Discontinuous variation speaks of 

 pronounced and irreversible alterations of the chromosome com- 

 plex ; continuous variation, on the contrary, suggests the idea of 

 small changes in individual chromosomes, due to the impossi- 

 bility of mechanically producing two exactly equivalent daughter 

 cells. These latter variations are minute and of every conceiv- 

 able character, but are inconstant and indeterminate. As a 

 basis for the occurrence of these fluctuating changes \ve have the 

 observed fact of the lack of exact correspondence in the size of 

 chromosomes and other cell parts in the various cells of an 

 organism, and the probability of a difference in the interaction of 

 the chromosomes in synapsis. I can not refrain from again 

 calling attention to the strong evidence that identical twins offer 

 in support of the theory that the characters are fixed by the 

 composition of the germ cells. Since these twins are the product 

 of a single egg it is evident that they must have exactly the same 

 chromosome complex, which, in development, will bring about 

 a very close parallel in the characters produced under their con- 

 trol. No other pair of germ cells from the same two parents 

 will ever produce two individuals that bear the resemblance to 

 each other that identical twins do, because, by the laws of 

 chance, it is improbable that any two pass through synapsis and 

 the maturation mitoses and emerge from them with their chromo- 

 somes constituted in exactly the same way. Ordinary twins, 

 coming from different eggs, are no more alike than children of 

 successive births although they are developed under practically 

 identical conditions, thus removing the possibility of environ- 



