124 PHILOSOPHICAL NOTES ON 



this epidermis cannot morphologically mean anything different 

 from the subjacent cells, which in an air-life it protects by 

 becoming hardened. Le Maout and Decaisne (" Syst. of Bot./' 

 p. 131) state as follows : — 



" Submerged leaves have no epidermis, stomata, fibres, or 

 vessels ; their parenchyma is reduced to elongated cells, arranged 

 in few series, and it is consequently very permeable by water." 



From this it would appear that an undue stress has been put 

 by botanists on the distinction between cellular and vascular 

 plants. We might with equal cogency divide plants into 

 epidermic and non-epidermic. 



We will now turn to another form of leaf, that of ferns. 

 First, however, I should like to say that although it is generally 

 true that the venation of the leaves of dicotyledons is reticulate, 

 that of monocotyledons parallel, and that of ferns dichotomous, 

 there is no hard-and-fast law that does not permit of deviation 

 from these general rules. We have plants classed among mono- 

 cotvledons, such as Smilax, which have a reticulate venation. 

 The leaves of the Elm, the petals of the Rose, the Gladiolus, the 

 Tulip, the Begonia, and many others, as well as the leaves of 

 Strangeria (a cycad) and Salishuria (a conifer) have a dichoto- 

 mous venation, similar to that of ferns. The venation of the 

 frond of Ophioglossum, as given in " Hooker's Brit. Ferns," pi. 46, 

 is certainly reticulate. Then the venation of the India-rubber 

 plant is pinnate and parallel like that of the Canna. Near the 

 margin of the leaf the parallel veins anastomose into a continuous 

 vein, which festoons itself along the margin. There the venation 

 is slightly reticulate. 



How did the dichotomous venation of ferns and of the petals 

 of phaenogams originate ? If we turn again to seaweeds we 

 shall probably find a solution of this question. Some seaweeds 

 consist of nothing but dichotomous threads of cells, strung 

 together end to end, without any parenchyma whatever. 

 Gymnocongrus fastigiatus. Fig. 18, Jania fastigiata, Chal- 

 lithamnion Griffithsioides, and others consist of nothing but 

 fronds which are like fern veins. If, for instance, we took the 

 frond of the filmy fern Trichomanes reniformis^ Fig. 17, and 

 divested it of the intravenous parenchyma, we would have some- 

 thing very like Gymnocongrus fastigiatus and others. Fig. 18 

 represents the latter seaweed having the fronds or cladophyls 



