No. 638] 



TYPES OF MUTATIONS 



set apparently assort at random in the reduction division. 

 If, instead of acting individually without predilection one 

 for another, the four should come to assort in pairs, we 

 should have a different ratio in the F2 generation (15:1 

 instead of 35:1). There would still be duplication of 

 genes and a 4x number in reference to the parental form, 

 but independent assortment of the chromosomes would 

 have been lost. It will not be possible to go into the 

 details of the argument. It is at least suggestive that 

 Shull (12) has found 3 cases of duplicate genes in the 

 shepherd's purse, which has 32 chromosomes (tliat is 4 

 times 8)"; and Nilson-Ehle (10) has found a ca>»' of tripli- 

 cate genes in a wheat having 42 chromosonu's, which is 3 

 times the number in another variety of whi'nt (11). 



If tetraploid plants have l)oen of infhiciicc in cvohitioii, 

 it is probable that the other types of (hiplieation liave 

 also been of influence. A mutant of the (Ihil)e t>i)e with 

 a single duplication in one of the 12 sets oi'diiinrily fails 

 to hand on the duplication through the poll. 11. < »eca- 

 sionally it might do so, and we should then e.\]iret a con- 

 stant race with 4 homologous chromosomi's in oiw of the 

 12 sets. If these 4 should cease to assort at random and 

 pair, we should again have a possible du])]ieation of uvncs 

 and an added pair of chromosomes characteri>tie of the 



Thei-e is not time at my disposal to discuss mutations 

 of genes. In a recent paper (2) on a somatic mutation 

 in portulaca, I Jiave indicated my belief that mutations 

 of genes may occur at any stage in the development of 

 the plant. We have found color mutations which atfected 

 only the epidermis, and therefore could not be trans- 

 mitted through seeds. AVe have also found similar color 

 mutations which affected only the sub-e] .id.Tinal ti>sne, 

 and tlKM-efore could not show in the petal: l»nt which lu- 

 came evident from the set^ls i)ro.luce.l from this mutated 

 tissue. Then, seems to he n<. preferre<l location for the 

 ori-in ..f factorial mutations in tlowerin- plant-, althou-li 

 thev are nu.re readily tran>mltted if l\u-y occur in the 



