1909] LEAVITT—HOMOEOSIS IN PLANTS 31 



The present paper is sequel to a casual observation made several 

 years ago, when as I was passing near a horsechestnut tree in the autumn 

 I noticed, upon a depending branch, a leaf which had lost several of 

 its leaflets. I found that the palmately compounded leaves of this 

 species fall to pieces as they are cast off by the tree, or even before 

 this event. The leaflets are removed as if by a clean cut; and some- 

 times the petiole, quite devoid of blades, is left standing alone upon 

 the stem. 



This complete dismemberment seemed to me a curious thing. 

 Abscission at the base of the petiole is not a simple decay, but is, as 

 everyone knows, a somewhat complicated process, wherein there is 

 formed across the leaf-stalk, through suitable cell divisions, a plate 

 or layer of cells, by the disintegration of which the leaf is ultimately 

 allowed to fall away. The cell walls of the scar-surface become 

 suberized, so that the wound is sealed against the loss of water. 

 Leaf-abjection is thus a complex adaptive process. 



The question arose, Why should the abscission proper to the 

 petiole-foot be repeated in all of its details at the bases of the petio- 

 lules? — for I ascertained that the process is carried out in full at 

 these points in this species. Of what use to the tree is it that the 

 frame of the leaf, now emptied of its valuable contents which have 

 been withdrawn into the stem, should be carefully disarticulated ? I 

 nave never been able to imagine any utility nor have those to 

 whom I have propounded the riddle suggested any. I believe that 

 in fact the habit is neutral in the tree's economy. 



If this is so, the evolution of absciss-layers in petiolules cannot 

 be referred to natural selection. For in the plan of evolution by 

 the accumulation of variations under the guidance of natural selec- 

 tion, developing organs must pay their way as they go. 



An explanation of the presence of the useless structures occurred 

 to me. It seemed probable that the absciss-layer perfected in the 

 natural course of evolution at the foot of the leaf-stalk had, sub- 

 sequently to the compounding of the blade in this species, been 

 transferred, so to speak, to the bases of the petiolules. The disartic- 

 u ations of the leaflets seemed to be a series of ecologically meaning- 

 ess echoes of the primal, useful disarticulation at the junction of stem 

 and petiole. 



