2m 



Account of the Genus Dahlia. 



species ; the branches of some of the varieties of the super- 

 flua are smooth, and covered with a glaucous bloom, and I 

 have found many instances, in which the florets of the ray 

 are entirely destitute of stigmata. The leaves in all these 

 doubtful cases preserve the true character of the superflua, 

 whilst the flower approaches more to the frustranea. The 

 knowledge of the failure of the specific characters of Mons. 

 De Candolle in these varieties, may have caused the opi- 

 nion, which I understand is prevalent in France, that the two 

 species should be reduced into one. I have considerable 

 doubts of the propriety of such change, for I do not think it 

 probable, that so distant a variety as the scarlet one must be, 

 if there be but one species, should have been obtained at a 

 period, when scarce any other existed ; whilst all subsequent 

 cultivation has not produced another so distant, from the 

 original pinnata and rosea as this. 



Subjoined is a table of all the published figures, of the dif- 

 ferent varieties of those plants, which have come under my 

 observation. 



Figures of varieties of D. superflua. 

 Single Flowers. 

 Cavanilles, Icones, vol. 3 tab. 365. D. rosea; not coloured. 

 Annales du Museum, vol. 3. page 420. tab. 37. fig. 3. D 



rosea ; pale lilac. 

 Botanist's Repositonj, tab. 408. D. pinnata ; bright purple. 

 Paradisus Londinensis, tab. 16. D. sambucifolia ; pale rose 



with white at the base. 

 Macdonald's Gardeners Diet., tab. 19* fig. 1. D. pinnata ; 

 dull purple. 



