BOTANICAL SUBJECTS. 



171 



veins as roots, growing in a medium of mesophyl, where the 

 changes influenced by light and air occur, and the result sucked 

 up by the leaf rootlets (the veins), and distributed in the same 

 way that the soil rootlets suck up mineral water and distribute it. 



In Ouvirandra fenestralisj we have anastromosing leaf -roots 

 (veins) with a minimum of mesophyl. In fennel, and others, the 

 veins do not anastomose, and they also have a minimum of meso- 

 phyl. In several water species of Ranunculus, the submerged 

 leaves are only veins, without the mesophyl, such as in R. hetero- 

 phyllus^ and others, while the floating leaves exposed to the air 

 have their veins fasciated by mesophyl, although from the nodes 

 roots proper are also given out. It would appear very probable 

 that the submerged leaves of this Ranunculus do not function as 

 leaves, but as roots. 



If this be so, why do these leaf -roots retain the character of 

 true leaves in their phyllotaxis, and petiole ? Because they 

 descend from the leaf-lorm. of root, and heredity keeps them in 

 that groove. 



No doubt this form of root-like division in these plants has 

 been inherited from leaves, but we should always bear in mind 

 that originally there were no roots, but only numbers of dichoto- 

 inous veins spread out in the fashion of a fan, and all attached to 

 an anchoring disk, and functioning both as leaves and roots, as in 

 Jania fastigiatUf Fig. 48, Gymnocongrus fastigiatus. Fig. 18, 

 and others. 



Fig. 48. Part of Jania fastigiata, Harv. ("Harvey, Phyc. Austr.," pi. 251). 



It is in Callithamnion Griffithsioides and others, Fig. 49, that 

 we begin to find the disk fringed, and spht up into root fibrils. 



