90 The Source of Variability 



basis of the establishment of different chromosome numbers in dif- 

 ferent species of plants (Stebbins, 1950; Clausen, 1951). 



We therefore have every reason to believe that mutants similar 

 to laboratory types exist and are spontaneously and continuously 

 arising in all natural populations. We may also assert that these are 

 the true building blocks of evolution. 



MAGNITUDE OF CHARACTER CHANGE 



A lively controversy has progressed for many years concerning the 

 magnitude of character change involved in evolution. Darwin 

 (1859) believed that evolution proceeded by the accumulation of 

 small differences. Goldschmidt (1940, 1948) argued that such small 

 changes cannot lead to the large differences found between major 

 groups and that the latter must arise by the occurrence of large 

 mutant differences (saltation or macroevolution ) . Because of dis- 

 continuities in the geologic record, Clark (1930) also argued that 

 major groups must arise by saltation. Simpson (1944) took the 

 position that saltation is not necessary to explain the discontinuities 

 in the fossil record and agreed with Dobzhansky ( 1937 ) that 

 evolution proceeds on the basis of genetic changes of small magni- 

 tude. Later both Goldschmidt (1955) and Dobzhanky (1955) 

 pointed out that large differences in characters might arise from 

 relatively small differences in genetic makeup. 



In discussing character change at the inter-generic level, experi- 

 mental genetic analysis is usually impossible because either the 

 organisms are too different to produce fertile offspring, or they 

 represent extinct groups, or they cannot be reared under suitable 

 experimental conditions. Hence, any extrapolation from characters 

 to genetic makeup must be by inference. In spite of these difficulties 

 the phylogenetic analysis of certain groups does offer interesting 

 speculation regarding magnitude of character change. 



Small Changes 



The evolution of the male claspers in one branch of the caddisfly 

 genus Helicopsijche can be traced through a series of small steps 

 from the ancestral type H. mexicanus to the specialized type H. 

 vergelana (Fig. 37). In mosquitoes of the genus Ciilex many 

 evolutionary lines can be followed in each of which only a small 

 morphological step separates one species from another, as is shown 

 by characteristics of the male genitalia in one of these lines 



