6 



JOURNAL OF THE ROYAL HORTICULTURAL SOCIETY. 



which he, in common with all American lovers of flowers, owes 

 to the Eoyal Horticultural Society of Great Britain. 



The following statistics may be of interest before we enter on 

 the task of analysing Dr. Gray's work. 



The genus Aster comprises 200 species. These are found 

 chiefly in the northern hemisphere, particularly in North 

 America, where about 120 of them occur. Europe and Asia 

 share nearly all the rest, while Australasia is said to possess none. 



The order to which the Asters belong is the largest among 

 flowering plants, comprising one-tenth of all flowering plants, 

 and about one-eighth of those which occur in North America. 

 The Asteroid tribe includes the Golden-reds and true Daisies, 

 and many genera which have only a few species each ; in some 

 genera there is only a single species. That which it most con- 

 cerns us to know with regard to the relatives of the Asters is the 

 capability of the most closely allied genera for improvement. On 

 the one side we find Sericocarpus, a genus of rather unattractive 

 low herbs with small heads ; on the other side stand the species 

 of Erigeron, or Fleabane, a few of which are rather showy and 

 Aster-like, with broad ray-flowers. Botanically speaking, the 

 differences between the Fleabanes and the Asters are very slight, 

 no natural lines of demarcation existing. 



Dr. Gray divides our American species into thirteen sub- 

 genera, arranged under two heads, based on their lease of life. 

 The first series comprises the perennials, and has ten subgenera ; 

 the second includes the biennials and annuals, and has three 

 subgenera. But even at the outset there is an anomaly, showing, 

 as every part of the genus abundantly demonstrates, that one 

 cannot draw hard and fast lines : the species A. coloradoensis is 

 apparently perennial, but in all other particulars it is closely 

 related to the section which contains annuals and biennials. 

 The distinction is biological rather than morphological. 



The sections or subgenera grade into each other in a manner 

 which forbids the establishing of any precise limits, and the 

 same is true of the species themselves. The latter are connected 

 in many cases by intermediate forms which conceal from view 

 all lines of demarcation. We may go one step farther, and say 

 that almost every species has varieties which at one time or the- 

 other have laid claim to being considered of the rank of true 

 species. Of course these statements may be applied with little 



