268 



BOTANICAL GAZETTE 



[OCTOBER 



equal numbers of the mutant t3^e and of the type of O. Lamarck- 

 iana. Moreover, a mutant dwarf has been produced. 



O. Lamarckiana mut. liquida (fig. 5, A). — In 1912 and 1913 

 a new type of mutant was discovered, which came next to 0. scin- 

 tillans, had flat and smooth leaves Uke that form, but the foliage 

 was much broader and lighter green. The individual mutants 

 resembled one another in all respects, save the size of the flowers; 

 they were very slender and had leaves about two-thirds the size 

 of those of 0. Lamarckiana. The pollen was sometimes abundant, 

 but in other flowers rather scanty. 



All in all, I had 6 mutants of this type. One came from seed 

 of 0. Lamarckiana, but unfortunately it failed to produce good 

 seeds, although the fniits were well developed. Four others arose 

 from 0. lataXLamarckiand, two in 1912, which also did not jdeld 

 fertile seeds, and two in 1913. The last one was found in the 

 culture of 0. pallescens mut. no. i, and mentioned previously. The 

 seeds of the 3 fertile mutants were sown in 19 14 and yielded small 

 cultures, which split up into two types, one repeating the mutant 

 parents in all respects, and the other differing in no visible way 

 from ordinary 0. Lamarckiana. Besides these there were some 

 mutants which happened to belong to allied types. Table VII 

 gives the size and constitution of these cultures. 



TABLE VII 

 Second generation of O. liquida 



In each of these cultures 15 individuals have been allowed to 

 flower, about one-half of these being the liquida type and the other 

 half the Lamarckiana type.. The 3 mutants were scintillans, 

 pallescens, and cana. All 3 have grown vigorously and flowered 

 in August and September; they differed in no respect from the 

 races of the same names. 



