LECTURES IN BIOLOGICAL SCIENCES 



changes were the result of natural selection and have told us 

 much about how selection has acted. In other examples, re- 

 sistance of insects to insecticides has been shown to result largely 

 from the selection of single mutations with large effects (Dob- 

 zhansky, 1951). 



In plants, genetic information recently obtained about the 

 columbine, Aquilegia (Prazmo, 1960), suggests that the distinc- 

 tive features of its flower were originally acquired through the 

 occurrence and establishment of a single gene with large effects. 

 The columbine differs from all of its relatives in that its petals 

 bear long spurs containing nectar. This serves to attract animal 

 pollinators, which may be bumblebees, hawk moths, or hum- 

 mingbirds, depending upon the species of columbine. The re- 

 lated genera all have white flowers without spurs and are 

 pollinated by various kinds of insects. A single species of 

 Aquilegia closely related to the columbine, but without spurs 

 on its petals, is found in eastern Asia. When this species was 

 crossed with the common European columbine, the spur of the 

 latter was found to be determined by a single dominant gene, 

 although many different genes affected its length. We can sup- 

 pose, therefore, that the new direction of evolution, which 

 started long ago in the columbine genus and evolved species 

 having flowers with a variety of shapes, sizes, colors, and spur 

 lengths, was originally triggered off by the establishment in a 

 population of a single mutation with a conspicuous and highly 

 adaptive effect. 



The Positive Effect of Natural Selection 



The fourth basic postulate is stated as follows: 

 The role of natural selection is much more than the purely 

 negative one of eliminating unfit types. By greatly increasing 

 the frequency of gene combinations which otherwise have a very 

 low chance of appearing, selection has an esssentially creative 

 and progressive effect. 



The experiments on the genetic structure of populations 

 which have already been reviewed can themselves be regarded 

 as strong evidence in favor of this postulate. Even more strik- 

 ing evidence has recently been obtained from experimental 



48 



