Gladiolus Studies — II 



233 



and especially Safrano. Kunderd has for a number of years selected 

 such varieties, and has bred them together until he has a type that is 

 rather distinct from any of the others, not only in the matter of ruffling 

 but also in shape of bloom 



(fig- 30) • 



There are several ways of 

 explaining the origin of these 

 varieties. There is a possi- 

 bility that they are progres- 

 sive mutations; in other 

 words, that a tendency 

 toward waving arose by a 

 sport and continued to in- 

 tensify. Another explana- 

 tion is to consider ruffling as 

 due to several factors vari- 

 ously combined to cause a 

 gradual progression in the 

 degree of variation. 



CROSSING TECHNIQUE 



The normal, complete 

 flower of the gladiolus con- 

 sists of a show^ six-parted 

 perianth, of no practical 

 value in the production of 

 seed. Attached to the peri- 

 anth are three stamens, 

 with rather fleshy filaments, 

 and anthers which in many 

 cases are rather large; the 

 varieties differ widely in 

 this character. Most varie- 

 ties are abundant pollen 

 bearers, but the variety 

 Rochester "WTiite, because 

 of its complete albinism, 

 produces no pollen. The 

 pollen is of various colors ranging from white to almost blue. From the 

 center of the flower rises the long pistil, bearing aloft the three-forked 

 stigma. When the stigma is mature, which is a little time after the 

 stamens begin to shed pollen, the surface becomes rather feathery and 



Fig. 29. AMERICA X GLADIOLUS DRACOCEPH.\LUS 



The hybrid is indian red penciled and flecked with yellow 



and bronze. This is one of Ernest Braunton's hybrids 



