1078 The American Natnralist. [December, 
provisionally admitted species which closely surround the typical 
Quercus robur. Finally De Candolle admits that out of the 
three hundred species which will be enumerated in his Prodro- 
mus as belonging to the oak family, at least two-thirds are pro- 
visional species that are not known strictly to fulfill the definition 
above given of a true species." 
Of our own botanists, the late Dr. Asa Gray was one of the 
most conservative. But he could not but recognize the wonderful 
variableness of certain genera, and he has left upon record his 
opinion of two of them (Proc. of the Amer. Academy, Vol. XVII., 
p. 163). He says : "Aster and Solidago in North America, like 
Hieracium in Europe, are among the larger and are doubtless 
the most intractible genera of the great order to which they belong- 
In these two genera, along with much uncertainty in the limitation 
of species as they occur in Nature, there is an added difficulty 
growing out of the fact that many of the earlier ones were founded 
upon cultivated plants, some of which had already been long in 
the gardens, where they have undergone such changes that it has 
not been easy, and in several cases not yet possible, to identify 
them with wild originals. Late flowering Composita^, and Asters 
especially, are apt to alter their appearance under cultivation in 
European gardens. For some the season of growth is not long 
enough to assure normal and complete development, and upon 
many the difference in climate and exposure seems to tell in un- 
usual measure upon the ramification, inflorescence and involucral 
braets, which afford principal and comparatively stable characters 
to the species as we find them in their native haunts. I am not 
very confident of the success of my prolonged endeavors to put 
these genera into proper order, and to fix the nomenclature of the 
older species ; and in certain groups absolute and practical defini- 
tion of the species by written characters or descriptions is beyond 
my powers. But no one has ever seen so many of the type speci- 
mens of the species as I have, nor given more time to the systematic 
study of these genera." 
I have myself noticed the variation presented by two reputed 
species of rock cress (Cardamine), or, as it is usually called, 
Dentaria. Some years ago I collected at Lookout Mountain Ten- 
