JuxE, 1930 



E^•OLUTION 



P/ 



i\GE FIFTEEN 



yet very important, 

 had to do with the 



Most of modern genetics has been occupied with 

 tracing down the above "facts." Thej' relate essen- 

 tially to the method of transmission, to later gener- 

 ations, of gene-differences that are already found 

 to exist between individuals. They show the uni- 

 versality of these differences, their comparative cap- 

 abilities. But they leave untouched what now be- 

 comes the major question — how do such dift'erences 

 originate in the first place.'' What is the origin 

 of variations.' 



A hitlierto rather incidental 

 part of modern genetics has 

 problem just raised. It has been discovered definite- 

 ly that differences between genes do arise, de novo, 

 as it were. That is, not all gene-differences now 

 existing in a population have existed in it from the 

 beginning. New differences are continually arising, 

 somehow, and the differences now existing have un- 

 doubtedly arisen in the past in a manner similar to 

 these. 



Where Variations Begin 



Each gene-difference arises suddenly and full- 

 fledged, though we may not be aware of it at once. 

 Thus, in a population of gray-colored mice, sud- 

 denly, in a certain cell of one individual, one of the 

 genes whose cooperation is necessary for the pro- 

 duction of the gray color undergoes a change into 

 a gene of different composition that tends, in its in- 

 teraction with the other (unchanged) genes for 

 color, to produce a yellow tinge instead of a gray. 

 In this single cell, however, the change will not be 

 observed by us. But if this cell, or one of the cells 

 derived from it, happens to be a germ-cell, an off- 

 spring-individual may be formed in the next genera- 

 tion all of whose cells carry this new gene. Then 

 if the new gene is dominant (as it happens to be in 

 the case of yellow versus gray in mice) to the old 

 gene for gray which the oft'spring has received 

 from its other parent, the coat of the new animal 

 will be yellow, and we will see that a mutation has 

 occurred. But if the new gene had been recessive, 

 the gray dominant, the offspring would have ap- 

 peared gray like its parents and we should not yet 

 have been aware of the mutation. The new gene 

 might persist none-the-less, and be inherited by 

 generation after generation in invisible fasjiion, being 

 meanwhile "dominated over" by the gray from the 

 other parent. If in a later generation two descend- 

 ants both of which carried the mutated gene hap- 

 pened to mate together, an egg with the yellow 

 might become fertilized by a sperm also carrying 

 yellow, neither, therefore, carrying the dominant 

 gray, and from such a union a visibly yellow off- 

 spring would emerge for the first time. A mutation, 

 when recessive, may accordingly fail to manifest 

 itself for many generations, or may never have a 

 chance to sliow itself at all, before the line of in- 

 dividuals carrying it becomes extinguished. (It has 

 been shown by Fisher that most mutations must 

 meet this mute inglorious fate.) 



The new gene, once it has arisen, is ordinarily 



as stable as the old. The change is definite and 

 fixed, evidently of a chemical nature. Once it has 

 occurred, we have a new mutant gene which will 

 eventually either spread throughout the population 

 or be killed off, according to whether the individuals 

 which carry it reproduce more offspring or fewer. 

 The effects of mutations are of course as varied 

 as the gene-differences which are found to occur 

 within populations, since these gene-differences orig- 

 inated by mutation. Some gene-differences, some 

 mutations, produce large and startling effects, like 

 growing a leg on a fly's forehead. Some affect the 

 whole body in practically all its parts, others ap- 

 parently but one. But the less conspicuous changes, 

 tlie insignificant effects that are easily overlooked, 

 or that even, in many individuals, quite overlap the 

 normal type, seem at least as apt to occur as do 

 the pyrotechnical varieties. Evidence is not lacking 

 that physiological changes, and changes that can 

 only be detected physiochemically, are probably as 

 frequent as changes in visible structures, but gene- 

 ticists have till now had to have a predominantly 

 morphological training, and anyhow the morpho- 

 logical is easier to see and deal with. It would be 

 absurd and scholastic to try to classify mutations 

 according to the nature of their effects. A muta- 

 tion can do practically anything that life can do — 

 or at least a little of it, for life is built out of mu- 

 tations. 



ir. THE RANDOMNESS OF MUTATIONS 



The statement just made does not necessarily 

 mean, however, that the average mutation does very 

 much in the furthering of life. You will recall that 

 perhaps the biggest question among the older schools 

 of evolutionists was this : Do variations have a 

 tendency to be adaptive, to further life.' Is there 

 any evidence in them of an internal or external 

 adapting or perfecting principle, call it what you 

 will.' Any kind of inheritance of acquired adapta- 

 tion of the germ-plasm to environment, or single- 

 stepping origination of species requires this. What 

 do the data on the actual occurrence of mutations 

 show .' 



They show just the opposite, and in so doing 

 they support Darwin . The vast majority of ob- 

 served mutations are positively detrimental, and 

 handicap the individual less or more in the strug- 

 gle for its survival and reproduction. In fact, as 

 Altenburg and I showed in some studies on the fruit- 

 fly, Drosophila, in 1919, by far the greater number 

 of detectable mutations in it are actually lethal: 

 their effect is to kill the animal before it becomes 

 adult (though of course their effect may be pre- 

 vented if they are recessive and if the dominant nor- 

 mal gene has been received by the individual from 

 its other parent). Evidence is accumulating that 

 the same situation probably holds in other forms of 

 life. Now this is just what we should expect, and did 

 expect, on the basis of tlie theory that a mutation 

 is a chemical change in a gene, occurring at ran- 



