I9is] BARTLETT— MUTATION IN OENOTHERA 117 



none of the seedlings had been discarded. It must be remembered 

 that each pan is a fair sample of a whole culture, for the seedlings 

 were pricked off when very small and were taken from the seed pan 

 as they came, with no attempt at sorting. 





Fig. 13. — F2 progeny of Lexington C, Oenothera pratincola (pan 17 of the progeny 

 of C-52); one example of mut. nummularia, C-52-28, is shown; the remaining plants 

 are typical. 



Figs. 5, 6, 12, 13, 14, 15, and 17 show 7 of the 50 occurrences of 

 mut. nummularia in cultures aggregating 13,035 plants. Three 

 more of the original plants of this mutation are shown in figs. 3, 4, 

 and 16. The figures showing entire pans should give a fairly clear 

 idea of what the writer interpreted as fluctuating variation. It is 

 believed that very few if any mutations escaped detection in the 



