72 ESTABLISHMENT OF VARIETIES IN COLEUS 



If we turn our attention to the various phenomena associated with 

 the laciniate character of leaves and petals, we find the same degree of 

 irregularity and diversity of origin, expression, and inheritance that is 

 exhibited by variegations. Cramer (1907, chap. 21) devotes a chapter 

 to an excellent survey of the facts regarding the behavior of the char- 

 acter. It is interesting to note that Cramer observes that varieties 

 quite constant in seed progeny are likewise very constant in vegetative 

 propagation. There are numerous cases known of the spontaneous 

 development of a laciniate type from one with entire leaves, both as 

 seed mutations and as vegetative mutations; likewise of return to the 

 types with entire leaves. While some cases are quite constant, others 

 are widely fluctuating, even exhibiting a marked periodicity. 



The behavior of this character in Coleus is most striking in its varia- 

 bility of expression and of its inheritance as a periodic variation through 

 vegetative propagation. 



Fundamentally, the processes involved in the development of leaf- 

 shape are quite different from those involved in the production of 

 pigments such as green, yellow, and red. The shape of the leaf in 

 general depends on the rate, number, and regularity of cell divisions 

 in the different planes of growth. It would seem that a general and 

 quite uniform series of cell-divisions would give a leaf of more regular 

 outline, and that if the cell-divisions in the growing leaf occur irregu- 

 larly, giving, so to speak, points or lines of more rapid growth somewhat 

 analogous to apical growth, with a more or less multiple dichotomy, 

 then cut, lobed, or laciniate leaves would result. The shape of the leaf, 

 it would seem, is determined by intercellular relations concerned with 

 the manner of cell-divisions. 



A comparison of the variability that develops in vegetative propa- 

 gation with that occurring in the seed progeny reveals some essential 

 differences between the inheritance of the characters involved. While 

 the range of variation is quite the same, there is a marked difference 

 in what we may call the intensity of variation. In vegetative propaga- 

 tion the degree of the intensity was low, with reference to the frequency 

 of the appearance of new color patterns or to the development of the 

 laciniate character. A large number of plants were grown by vege- 

 tative propagation. The bud variations were comparatively infrequent, 

 occurring something like once for every 10 plants grown, giving in the 

 course of 6 generations the different types of pattern described. In a 

 seed progeny, however, practically the entire range of variation which 

 appeared in vegetative propagation was seen in a single seed genera- 

 tion comprising 50 plants. The processes concerned with reduction and 

 fertilization increased the intensity of variation and brought out in a 

 single progeny of no great number the full extent of variations. 



Color patterns, which are intercellular characters and in a sense 

 vegetative types, are inherited through vegetative propagation in 

 marked degree, while in seed propagation there is no evidence that they 



