GREEN LEAF IN CHERRY BLOSSOM 



David Fairchild. 



IT HAS long been known that the 

 reproductive organs of a flower could, 

 under certain circumstances, be re- 

 placed by leaves — either the ordinary 

 stem-leaf or the flower leaf (petal). 

 Double flowers are in most cases due 

 to a replacement of stamens or pistils 

 by petals, and are for this reason 

 sterile. Examination of a carnation 

 or almost any highly developed double 

 flower will show that it must necessarily 

 be sterile, since it altogether lacks the 

 reproductive organs. 



The appearance of an ordinary leaf 

 in place of some part of the seed- 

 producing apparatus is rarer, but an 

 excellent example is shown in Fig. 7, 

 a blossom from a Japanese flowering 

 cherry {Pninus pseiido-cerasus) at my 

 home near Chevy Chase, Maryland. 

 The pistil is lacking, and its place 

 is occupied by two well-formed green 

 leaves (a) and (b), the teeth or serra- 

 tions along the margins of the leaves 

 being clearly visible. Such a phenom- 

 enon is technically known as phyllody. 

 The older botanists looked on such 

 a change as a metamorphosis, believing 

 that the pistil had actually been changed 

 into a leaf. At present, it is more usual 

 to regard it merely as a replacement, 

 since this does not involve the assump- 

 tion, formerly made without hesitation, 

 that the rej^roductive organs are mor- 

 phologically nothing but modified leaves, 

 which might easily reappear through 

 "reversion." The sporophylls or re- 

 productive organs of the flower may be 

 modified leaves, but it is equally 

 possible that leaves are modified sporo- 

 ]jhylls; and until evidence is available 

 from which the case can be proved one 

 way or the other, it is safer not to 

 assume that the sudden appearance of 

 a little leaf in the center of the flower, 

 as shown in the illustration, is a rever- 

 sion of some organ to its ancestral tyjjc. 

 Even as to the evolutionary origin 

 of the petals, we are not vet certain. 

 A. P. de CandoUe (1817) seems to have 

 been the father of the idea which 

 prevailed for nearly a century, that all 

 the floral leaves arc derived from 

 262 



sporo]jh\'lls, the ])lant having found it 

 presumabh- advantageous to modify 

 some of its sporophylls in a conspicuous 

 way, in order that it might attract the 

 attention of insects and secure the 

 pollination of the remaining sporo- 

 phylls. Recently, however, Goebel and 

 others have contended that the j^etals 

 may in some cases be modified bracts or 

 true leaves. 



If we go back one step farther, we 

 can disregard these slight difficulties 

 and say broadly that a flower is merely 

 a highly modified stem of the bud type. 

 This will be recognized by anyone who 

 examines a flower bud before it opens. 

 The showy petals, which to the layman 

 are the distinctive part of a flower, are 

 in fact merely ornaments of compara- 

 tively late addition, from an evolu- 

 tionary point of view, and not at all 

 necessary to make a flower. From the 

 morphological viewpoint the presence 

 of the sporophylls, the organs which 

 bear ovule and pollen grains, is the 

 criterion recognized by most botanists. 

 The origin of these sporophylls, the 

 stamens and carpels, is obscure. They 

 are very ancient structures, and al- 

 though they are represented in lower 

 plants by leaves bearing sporangia, or 

 organs for the distribution of spores, it 

 is possible, as was indicated above, 

 that at a still earlier period leaves were 

 the luxury and sporophylls of some 

 kind the primary necessity of the plant. 



The causes leading to such a re])lace- 

 ment as that illustrated in the ])hoto- 

 graph are still almost wholh' obscure. 

 They are rather \'agucly ascribed, in 

 many cases, to an excess in nutrition, 

 but there is reason to believe that in a 

 large number of instances such a change 

 is to be considered as originating in the 

 germ-plasm. Evidently, a phenomenon 

 of this .sort is likely to com])licate our 

 idea of hcredit>' in a xoxy embarrassing 

 way, and to cause us to hesitate before 

 accejjting too confidently any theory 

 which will make heredity a simple 

 matter of the .shullling of unalterable unit 

 characters in a jmrely mechanical 



