May, 1920.] YASUI .'—OEyETICAL STUDIES. 59 



to the doubleness and d its absence, then each individual in Nos. 38 

 and 39 is to be represented with Dd. The experiment 2 gives us 

 another proof of the recessiveness of the singleness to the doubleness ; 

 because if it be not so there must have some double individuals 

 appeared in this experiment. Moreover the experiment 4 explains 

 plainly that those doubles were heterozygous, as they threw singles, 

 when mated with a fixed single. 



In short, in Portulaca gran.diB.ora, at least in the races used in 

 these experiments, the difference of ids related to doubleness and 

 singleness is one, but if it be more than one, all those differing ids 

 should be in one and the same chromosome. 



There are several other plant species, which include the races 

 with double and single flowers ; but the relation of their dominancy 

 and recessiveness in those various cases is not the same even in the 

 same genus. The cases in which the doubleness is recessive to the 

 singleness are reported in Delphinium (Bateson, 1909), Primula 

 (Bateson and Gregory, 1909), Petunia, (Saunders, 1910), Matthiola 

 (Saunders, 1911), Sweet William (Saunders, 1911), Japanese Morning 

 Glory " Tenaga-botan " (Takezaki, 1918), " Kujyaku-zaki " (Miyake 

 and Imai, 1920), and others. I have also experienced, in a race of 

 Japanese Morning Glory, other than " Tenaga-botan " and " Kujvaku- 

 zaki ", that the doubleness is recessive to singleness ; while the cases 

 of the dominant double were reported hy Correns (1905) in Mimulus 

 and Campanula, and by de Vries (1902) in Dianthus and Campanula ; 

 Bateson also suggested a similar case in Carnation. But these records 

 were not accompanied by the numerical data. In 1917 Saunders re- 

 ported the similar cases on Meconopsis, Althaea, and Carnation, with 

 numerical data; and especially the cases of the last two plants worth 

 attention. 



Although the number of individuals in my experiments on double- 

 ness and singleness in Portulaca grandiBora was not so large as it was 

 desirable, it is I believe one example of the dominancy of the 

 doubleness of petals, which bears the proof of numerical data. 



B. Colours of the Petals 



Experiment 1. Pale yellow single X White single. The colour of 

 the F 1 plants was magenta (Plate I, fig. 3), the reciprocals always 

 giving the same results. The segregation in F 2 generation is shown 

 in the table 4. 



