ON FOUR DOORYARD FLOWERS 



leaves. Among these crossbreds it appeared that 

 the green colored foliage was prepotent or dom- 

 inant over the white and yellow variations. The 

 horseshoe variations were more readily transmit- 

 ted, but there was a varying proportion of marked 

 and plain leaves among the hybrids. 



I also worked at one time in selecting the gera- 

 niums for the production of large flowers of daz- 

 zling brilliant scarlet color, and with a good meas- 

 ure of success. One of the varieties thus produced 

 has been greatly admired by all who have seen it, 

 and will probably be thought worthy of introduc- 

 tion. 



It will thus appear that there is abundant op- 

 portunity for improving the geraniums even by 

 working with the species ordinarily under culti- 

 vation. I repeat, however, that the best oppor- 

 tunity for work in this line will involve hybridizing 

 experiments in which the exceedingly hardy wild 

 species are utilized. It should be possible thus to 

 produce new races of geraniums that have alto- 

 gether exceptional quality. 



The wild species include some that are white 

 in color as well as those that are pink or white 

 striped with pink or with reddish veins. So there 

 is opportunity to have a wide choice as to color 

 variation. The cross might likely result also in 

 giving the geraniums enhanced vigor so that new 



[133] 



