76 JOURNAL OF THE ROYAL HORTICULTURAL SOCIETY. 



tendency to appear first in the most highly specialised parts of a 

 flower, as on the lips of orchids, the noses of hard drinkers (the 

 so-called " grog blossom "), and the cups of the Narcissi. Thus, 

 as a rule, we find the trumpets of Daffodils, and the chalices or 

 cups of the Narcissi, the most highly coloured. 



And, if ever we do obtain what our good friend Max Leichtlin, 

 of Baden Baden, says we may hope for, viz. "a scarlet Daffodil," 

 we may be quite sure that the trumpet will become red or scarlet 

 long before the outer divisions of the flower ; and that N. poeticus, 

 having already a purple rim to its saucer, may some day mount a 

 step higher and obtain a blue one. Colour variation is mainly a 

 chemical question, and Professor Sorby tells us that it is due to 

 oxidisation as influenced by soils and chmate in a very subtle 

 and complex manner. Grant Allen has shown us that these 

 colour-changes, which are primarily chemical ones, do not take 

 place haphazard, but that the colours advance or revert in a 

 regular way, passing through all shades from yellow through red 

 to, or towards, blue, which is the highest note in floral colouring. 

 I do not ask you to believe what I say of this colour question, 

 but I do ask you to copy this diagram, and to observe in your 

 gardens for yourselves, as to whether it is not true. 



BLUE 



PURPLE / . R£EN 



RED 



i F 



WHfTl 



ORANGE 



Why so many flowers should evade passing through the 

 orange stage on their way from yellow to red, is more than I 

 can explain, but it is a fact that many flowers do pass by a short 



