152 PHILOSOPHICAL NOTES ON BOTANICAL SUBJECTS. 



seeds have an aril, and the embryo is curved or spiral. But none 

 of the latter characters are constant, and consequently the definition 

 of the order becomes very difficult (!)." 



18. It is no wonder that botanists are sometimes puzzled about 

 the true affinities of a plant, considering tha,t fusio7i and suppression 

 of jDarts may have frequently masked their derivation. If it were 

 not for the persistent inheritance of certain characters, which often 

 occurs, botanical classification would have been impossible, or little 

 better than a chaos. 



19. Just look into another order, that of the Liliacece. Who 

 would suppose that the onion, the asparagus, the butcher's broom, 

 the dragon-tree, and the grass-tree are all made to claim relation- 

 ship with the lily and the tuberose. It would seem that Aspidistra 

 can be included in the same category only by a sort of botanical 

 courtesy. 



20. Even in such natural groups as the Limes and the Oaks, 

 we find foliage so distinct that it might as well belong to totally 

 difi'erent plants. An inspection of the collections of these genera 

 in the Koyal Kew Gardens is very instructive. 



21. Asa Gray* gives a drawing of the summer shoot of the 

 Barberry. The lower leaf is normal, and only toothed, and 

 stipulate ; the next has the blade more contracted, and the teeth 

 longer and broader, the stipules having disappeared ; by greater 

 contraction the whole blade disappears, and the leaf is reduced to 

 something like a branched liair. 



* "Struct. Bot.," p. 117. 



