26 ESTABLISHMENT OF VARIETIES IN COLEUS 



There were few cases of sudden, clear-cut bud variation in this 

 group. One was a very decided and ahnost complete development of 

 yellow to yellow-red blotched; 3 were sectorial variations to green-yellow- 

 red blotched; 2 cases gave loss of yellow to green-red blotched; 2 involved 

 gain of epidermal red to solid red, and in two cases there was loss of 

 epidermal red. All these cases were sectorial for a plant or in some 

 cases for a single branch. The ratio giving the frequency of bud varia- 

 tions involving color in this group is 1 to 3,960. 



Plants with laciniate leaves. — Until the winter of 1912-13, all the 

 plants in the cultures had been constant and very uniform for leaf- 

 shape, showing no greater variation in this respect that is seen in fig- 

 ures 2, 6, and 9. During that Mdnter it was noted in 11 cases that as 

 new leaves developed they were more and more deeply cut and lobed 

 until in Januar}^ and February the uppermost leaves were in extreme 

 cases much divided and deeply laciniate, as shown in figure 7. The 

 plants appeared like the middle plant in plate 4; 3 of these plants 

 were from cuttings of branches that were pure-green bud variations; 

 the others were plants which had a somewhat fluctuating green-yellow 

 spotted pattern. Nine of these plants were grown during the following 

 summer, when it was noted that without exception the new leaves 

 gradually became more entire until by late summer all the leaves then 

 hanging to the plants were entire. At the same time 5 of the plants 

 became decidedly more yellow, even becoming quite uniform for type 

 green-yellow-red blotched. Cuttings were made in April from each of 

 these plants and from these 16 plants were grown during the summer 

 of 1913. All of these fluctuated to entire leaves, and in regard to color 

 gave plants some of which were uniform for type green-yellow-red 

 blotched, others for tj^pe green-yellow spotted-red blotched, while the 

 foliage of others showed mixtures of these patterns with also pattern 

 green-red blotched. 



The further generations in subclone 12 exhibited, as shown in table 

 4, the same periodicity in change of leaf-form, except that in late sum- 

 mer of 1914 a rather large number of the plants showed new leaves 

 that were laciniate. One plant of this subclone was grown in the winter 

 of 1913-14 from a pure-green bud-sport that appeared during the 

 previous summer. This plant remained constant for entire leaves of 

 pattern green-red blotched during the winter, but as it died early in the 

 summer no further progeny were grown. Further generations of the 

 plants with laciniate leaves in subclone 14 were not grown. 



The laciniate leaf-form appeared anew in the winter of 1913-14 in 

 clone 3 in two instances. One was a plant whose line of descent 

 showed bud variation from type green-yellow-red blotched to type green- 

 yellow spotted-red blotched; that of the other showed bud variation from 

 green-yelloio-red blotched to green-yellow solid red, and then from this to 

 green-yellow-solid red upper center. The latter was the only one of 



