liOTANICAL SUBJECTS. 137 



conclusion that there was no essential difference between the leaf 

 and the branch. The last, of course, everyone recognizes as a 

 repetition of the stem. The leaf I considered was, like the branch, 

 only a sub-division of the stem, greatly modified to meet the 

 changed conditions to which plant life had been subjected. 



It may be objected that the leaf is deciduous by a sort of joint, 

 and the stem is not. The answer \fi— firstly, that all leaves are 

 not desiduous. Some decay and remain attached to the stem, like 

 dead l^ranches ; secondly, that in a family so well defined as that 

 of orchids, some species have deciduous leaves and others not ;* 

 thirdly, the joints or internodes of the stem of Begonia argyro- 

 stigma become deciduous by a sort of joint, like that of deciduous 

 leaves, when, through some cause, the internode decays and dies ; 

 fourthly, the branches of Xylophylla are deciduous. f Begonias 

 probably teach us more about the nature of hairs, leaves, and buds 

 than most plants. 



Asa Gray (" Structural Botany," p. 45) says : " In Begonia, a 

 leaf, used as a cutting, will root from the base of the petiole, stuck 

 in the soil, and produce buds on the blade, at the junction with the 

 petiole or elsewhere." 



The nerves of the Begonia leaf are like verticillate branches, 

 fasciated by means of mesophyl. 



It would appear that not only are leaves homologous with stems 

 but, as Prof. Caruel observes, that the very hairsj of Begonia 

 phyllomanlaca can make themselves homologous with branches 

 by developing into leaves and buds. The hairs of Drosera have 

 under certain circcmstances, done the same thing. 



After all these speculations of one's mind, it is comforting to 

 find that I have not been thinking out and elaborating heresies 

 for in a paper on the " Comparative morphology of the leaf in the 

 vascular Cryptogams and Gymnosperms," by F. O. Bower com- 

 municated to the Royal Society by W. F. Thiselton Dyer,S the 

 following is the conclusion come to, " Now the leaves in the lower 

 forms naturally lend themselves to a treatment throughout as 

 branch systems, Avhilc there is even in some complicated fern leaves 

 no structure which can warrant a distinction of the foliar base 



* Bauer's illustrations of orchid plants. 



f As also those of Quercus Robur. 



X Vide Discussion on Hairs. 



§ Vide "Proc. Roy. Soc," vol. 37, p. 61 (1884). 



