146 Nature of the Genetic Material 



it concurs with the material presented up to now in proving the 

 identity of position effect and point mutation, the topic of the 

 present discussion. 



Mutable or unstable genes have been known for a long time, and 

 there is an extensive literature on the subject for plants (Emerson, 

 Stadler, Rhoades, Ernst, Demerec, Imai), but rarely have cases been 

 reported in animals. The one example in Drosophila, studied by 

 Demerec (1941^), cannot be interpreted with certainty, I have shown 

 ( Goldschmidt, 1943 ) that facts very closely paralleling those described 

 by Demerec may be based upon special genetic situations combined 

 with special morphological conditions. For the time being, the exist- 

 ence of the phenomenon in animals cannot be regarded as demon- 

 strated beyond a doubt, and it will be very difficult to prove it finally. 

 The prototype of the phenomenon is found in pericarp color in maize, 

 which frequently shows variegation. Emerson showed (1914ff.) that 

 this variegation is based upon "mutation" from a colorless allele to a 

 dominant colored one, taking place in somatic cells at definite places 

 and times of development, and leading to colored spots. Their number 

 and size give information about frequency, place, and time of these 

 "mutations." If a colored spot engulfs sex cells (or their precursors) it 

 can be proved that mutation is involved, or seems to be involved. 

 (The latter fact puts Drosophila at a disadvantage. The assertion 

 cannot be made that a mosaic spot on a wing engulfs the sex cells, 

 for they are completely removed from the soma from the beginning of 

 development. This is one reason why Demerec's interpretation had to 

 be regarded with caution. ) It is strange that "color mutations" always 

 occur in the direction from recessive to dominant, also with inter- 

 mediate alleles. Very few occurrences in the opposite direction have 

 been described. This unexpected behavior, that is, not expected on the 

 basis of our knowledge of mutation in maize, indicates some special 

 feature. 



Older authors like Correns assumed that these ever-sporting or 

 mutable loci were "sick genes" and could not throw light upon 

 genuine mutation, a notion which I accepted for some time but which 

 can no longer be held, since a much better solution is available now. 

 The link with our present discussion was forged by Rhoades' (1938) 

 discovery that in a special line of Mexican maize a dominant locus 

 Dt in the ninth chromosome exists in the presence of which the color 

 locus aa in the third chromosome — no anthocyanin pigment — 

 "mutates" after the fashion of unstable loci to its higher alleles; and 

 only this one locus does it. Again it could be shown that the effect is 



