BOTANICAL GAZETTE 



[FEBRUARY 



subulata. It is of course impossible to establish absolute identities 



among seedling plants of types which have not yet been seen in 



flower. Consequently the Fi progeny of mut. nummularia are 



classified either as true to t)^e or as secondary mutations in 



table I, which shows the composition of the cultures now under 



observation. 



TABLE I 



Composition of Fi generation of mxft. nummularia (seedling stage) 



* Indicates that the seeds were from one capsule. 



Table I shows that only 34 per cent of the seeds of mut. num- 

 mularia germinated. In order to obtain as many plants as possible, 

 a large number of seeds were counted into the seed pans which 

 seemed too immature to germinate; 65 such seeds, planted by 

 themselves, produced 9 plants. Part of the seeds planted were 

 obtained from self-pollinated capsules, others from capsules which 

 had been cross-poUinated. Table I shows that the progeny from 

 the self-poUinated seeds includes secondary mutations and typical 

 nummularia plants in the ratio 1:6. The same ratio for the progeny 

 from cross-poUinated seeds is about 1:3.2. Although the differ-" 

 ence in the ratio seems very marked, it may be due to the fact 

 that the germination was poor and the cultures small. 



