228 Cornell Extension Bulletin io 



The results of Jackson's crosses are here noted, without the knowl- 

 edge whether these hybrids are the results of selfed individuals. The 

 results approximate a second-generation 1:2:1 ratio in the inheritance 

 of markings. 



Inheritance of markings 

 Theoretical 

 amount that 



should have Actual result 



obtained 

 (per cent) 



25 26 per cent marked on lower segments with purpureo-auratus blotch; 



in many cases not maroon-crimson but a rich crimson. (A 

 color found at the base of the blotch in the parent species.) 

 50 53 per cent possessed a combination of the linear stripe of ganda- 



vensis and a blotch-like stripe of piirpureo-auratus. 

 25 18 per cent had gandavensis penciling. 



The occurrence of a white patch in the petalage seems to be the effect 

 of an inheritance of the pattern. The yellow splash at the margin of the 

 maroon blotch is often wanting, but usualh^ inherited, showing that 

 this yellow splash and maroon blotch are not inseparable, but are trans- 

 mitted independently. 



In the case of the other characters noted, which may be due to multiple 

 factors, the results are not of the simple 1:2:1 ratio, but are of interest 

 to note: 



Inheritance of shape of bloom 



80 per cent, form sub-open or flaring; an intermediate between the two 

 parents. Some were as widely flaring as the magnolia. This proportion 

 may be a little large since the shape is difficult to determine. 



8 per cent were of the bell-shaped bloom of the purptireo-aiiratus. 



Inheritance of foliage and habit of plant 

 90 per cent of cases intermediate. 



10 per cent, tendency toward the greater size and increased rigidity of foliage; 

 a character of the gandavensis. 



Inheritance of stoloniferous habit 



Generally inherited. Contributed by purpureo-auratus. 



Inheritance of aestivation 



This character concerns the arrangement of segments of perianth which are, 

 in most species, disposed so that there are two inner lower petals; but often 

 there is but one. 



Most of the flowers show the two-lipped type of aestivation; at least 75 

 per cent should have done so, for the species purpureo-auratus is character- 

 istically so and the species psittacinus, a parent of gandavensis, possesses 

 both forms. The one-lipped aestivation occurred only as scattering individ- 

 uals upon a spike. 



Fischer (19 14) writes: 



I see no reason why we should not benefit by the use of the Mendelian method in 

 the practical side of gladiolus breeding; that is in the creation of new types by the 

 recombination of pre-existing characters. To begin, one must have an ideal form or 



