56 Cinci7inati Society of Natuial History. 



I sent a set of such as I obtained to Harvard University, 

 labeling all A. sagittifolius, except tlie doubtful A. Lindley- 

 anus. Mr. Sereno Watson said in reply : 



" The specimens differ considerably. They are certainly 

 part and probably all A. sagittifolius. -'^ * * The most 

 interesting plant of your collection is the A. Lindleyanus, 

 T. and G., which, so far as I know, has not been reported 

 south of the great lakes. -•■ * * Aster Lindleyanus is 

 unfortunately a rather indefinite thing, and seems to grade off 

 imperceptibly into A. sagittifolius. Indeed, some of your 

 specimens of A. sagittifolius, with large heads and loose 

 involucre, may be regarded as more or less transitional." 



Aster sagittifolius, as I find it in northern Ohio, is the most 

 constant in general aspect and characteristic details of any of 

 this section of Asters. The plants invariably live up to the 

 typical characters, such as erect and rigid stems, inflorescence 

 closely paniculate, and scales of the involucre linear with 

 long attenuate tips. 



The Columbus plants that we must under present arrange- 

 ments refer to, A. sagittifolius, are just the opposite of this. 

 I have thus far been able to detect but one character that 

 appears at all constant. This is the late flowering of the 

 larger forms approaching tlie description of A. Lindleyanus. 

 I found one .specimen having the erect stem and in a modified 

 way the inflorescence of the northern Ohio form, thus showing 

 affinity to what maj^ be considered the type of the species. 



Referring to the Columlnis plants, A. sagittifolius, I spoke 

 of their polymorphic habits. Taking as a type the smaller- 

 leaved, loosely-flowered form (which is the one commonly 

 met with), I find three distinct lines of variation, i.st. Into 

 A. Lindleyanus; 2d, toward the northern Ohio plants, and 

 3d, connecting with cordifolius. In the light of my present 

 knowledge of this Aster I do not think it worthy of specific 

 rank, yet it seems to have good varietal characters. It is 

 probable that the ranges of the typical plants of northern 

 and the central Ohio forms meet and overlap. 



We all know that among a lot of seedlings, the progeny 

 of one plant will show many slight diff'erences. In some 

 cases these indi\i(lual difl'erenccs will l)e well marked. When 

 such is the case, and when the characters are constant, but vet 



