58 ESTABLISHMENT OF VARIETIES IN COLEUS 



DISCUSSION. 



The appearance and subsequent behavior of bud variations in Coleus 

 present numerous analogies to various phenomena of variation exhib- 

 ited by members of a seed progeny of hybrid origin. 



In respect to the definiteness of the characters contrasted, green and 

 yellow, red and non-red, the color patterns arising by bud variations 

 are as different, at least in their extremes, as one could expect in the 

 members of a seed progeny even of hybrid origin. This is especially 

 noticeable in comparison with the seed progenies of Coleus itself. The 

 solid-red epidermis and the no-red epidermis represent two extremes 

 fully commensurate with the so-called presence and absence of a 

 character and the bud variations giving these were fully as different as 

 the types arising in seed progeny. The intermediates red blotched and 

 solid red upper center are pattern characters equally distinct both in 

 manner of appearance and in vegetative constancy. The same is true 

 of the extremes of development of both green and yellow. 



Plants of the same pattern in a single line of descent, both in the 

 same generation and in successive generations, frequently produced 

 the same type of variation independently, a behavior quite analogous 

 to the segregations that reappear in each generation of a hybrid line 

 or in the successive generations of the progeny of a mutating plant. 

 Many of these variations show a return to a parental pattern, just as 

 a recessive parental quality reappears as a result of segregation in 

 hybrid progeny. There is much in such reappearance of patterns that 

 is quite identical with the phenomena of alternative inheritance. 



Two plants identical in appearance and derived from adjacent 

 branches on the same plant may give quite different progenies in suc- 

 sessive generations. One line may be very constant and uniform, the 

 other may give numerous bud variations of wide range. This is a 

 familiar phenomenon in hybrid seed progenies where certain of the 

 plants of any generation, although apparently identical, give quite 

 different progenies. In Mendel's experiments with Pisumjor example, 

 although the F2 yellow peas of the cross green X yellow were similar, 

 they gave different progenies. Some produced only yellow peas, while 

 the progeny of others gave both yellow and green. 



In the strains of Coleus studied b}^ the writer, certain types of 

 bud variation occur more frequently than others. Loss of yellow was 

 more frequent than loss of green, and loss or decrease of red was 

 more frequent than increase of red. In the entire series of plants 

 derived by vegetative propagation there was decided predominance of 

 green over yellow, of red blotched and no red over solid red, of the blumei 

 character of leaf-base, and of the shallow or crenate lobing of the 

 leaves. These same characters show marked tendencies for domi- 

 nace among the members of the seed progeny. 



