12 THE COEFFICIENT OF MUTATION IN OENOTHERA BIENNIS L. 



explained as a result of such a condition, since there is not the 

 least doubt concerning the gametic purity of 0. biennis L. 1 ) 



From these discoveries it was pretty safe to deduce that the 

 pure 0. biennis must also be in a state of mutability, and the first 

 thing to do was obviously to make extensive cultures in order to 

 find the pure line mutants. Stomps cultivated over 900 individuals 

 of the third and fourth generations of a pure line, derived from a 

 rosette^collected by him in the sand dunes near Beverwyk, Holland, 

 in 1905. 2 ) Among these he found one 0. biennis mut. nanella, one 

 0. biennis mut. semigigas, and also four instances of the pale- 

 yellow variety 0. biennis sulfurea. The first two he calls parallel 

 mutations, since they are analogous to the dwarfs and semigigas 

 mutations of 0. Lamarckiana and arise in the same way and with 

 the same differentiating characters. The experimental origin of 

 0. biennis sulfurea by mutation clearly shows that this variety, 

 which is anything but rare in some parts of our sand dunes, may 

 arise in the same way in the wild condition and afterward propa- 

 gate itself by seeds. 



The production of dwarfs from 0. biennis by mutation has since 

 been repeated more than once in my cultures of hybrids between 

 this species and some of its allies, 3 ) and a lata mutant from 0. biennis 

 has been reported by Gates and described under the name of 0. 

 biennis mut. lata. Besides 0. biennis, some allied species also are 

 now known to show the phenomenon of mutation. Among these, 

 an American form of 0. biennis, which I cultivate under the 

 preliminary name of 0. biennis Chicago, has been studied more 

 extensively than any other form. I had already found in the 

 neighborhood of Courtney, Miss., in 1904, in a locality called "the 

 bottom," along the shores of the Missouri River, a single specimen 

 with narrow, almost linear leaves. Evidently it constituted a wild 

 mutation from the surrounding type. 4 ) 



Seeds taken from the normal specimens of this locality have 

 since produced in my garden two mutations, which proved, in 



i) Davis, B. M., Mutations in Oenothera biennis L. Amer. Nat. 47M16, 

 1Q13; also Parallel mutations in Oenothera biennis L. Amer. Nat. 

 48:498—501, 1914. 



2) Stomps, Th. J., Parallele Mutationen bei Oenothera biennis L. Ber. 

 Deutsch. Bot. Gesells. 32:179 — 188, 1914; also Parallel mutations in 

 Oenothera biennis L. Amer. Nat. 48:494 — 497, 19 14. 



3) Gruppenweise Artbildung, pp. 300 — 301, Berlin, 19 13. 



4) Op. cit. p. 304. 



