114 Flowers and their Pedigrees. 



distinguish the original true leaves of each whorl from 

 the leaf-like stipules. Only two leaves out of the six 

 ever have buds or branches proceeding from their 

 .axils ; and this last token infallibly marks out for us 

 which are the real primitive opposite pair, and which 

 the spurious imitation. 



What may be the use of the square stem it would 

 be more difficult to decide. Perhaps it may serve to 

 protect the plant from being trodden down and broken ; 

 perhaps by its angularity and stringiness it may 

 render it unpalatable to herbivorous animals. This 

 much at least is certain, that very few cows or donkeys 

 will eat goose-grass. There is another large family 

 of plants — the dead-nettle tribe — all of which have 

 also square stems ; and they are similarly rejected as 

 fodder by cattle. Indeed, the very fact that the 

 stellate tribe have become thus quadrangular, while 

 the other and earlier members of the madder kind, 

 like coffee &d gardenia, have round stems, in itself 

 suggests the idea that there must be some sufficient 

 reason for the change, or else it would never have 

 taken place ; but, as in many other cases, what that 

 reason rtiay be I really cannot with any confidence 

 inform you from my simple professional chair on the 

 gate here. If I were only at Kew Gardens, now — : 

 well, that might be a different matter. 



And now let us come down to the individual 



