(153) 
subspecies or forms. It may best be illustrated by two examples. 
The two forms described as V. Baldwini Torr. and V. ¢xtertor 
Small were regarded by Mackenzie and Bush as two forms of one 
species ; the former was reduced to a variety of the latter, and a new 
name, V. interior Baldwini, was published. The same course is 
followed in this paper. Nomenclatorially the name should be V. 
Baldwint interior, since Baldwint, the oldest name, must apply 
to the species according to current rules of nomenclature. How- 
ever, V, Baldwinz is not the species, but only a local eastern form 
of the more widely distributed V. zzterzor. This is shown by the 
migration routes of species from the Ozarkian center. One of them 
extends northeast through southern Missouri into Illinois, and it is 
along this extension that the variety occurs. If species migrated in 
the other direction, from Illinois toward Arkansas, which is not the 
case, then there would be some reason for regarding V. zuterdor as 
the variety. A second and similar case is that of V. oval¢folia, 
which, as its morphological structure shows, belongs to the group 
Altissimae of the Appalachian center. The Floridian form of the 
plant with different foliage is therefore a variation from the normal, 
although unfortunately the type specimens of Torrey and Gray be- 
long to it. Further details of the significance of geographical dis- 
tribution will be given under the appropriate genera and species. 
In the treatment of the species and genera keys have been given 
to all the species whose validity has been admitted. Some descrip- 
tions have been inaccessible, and a few others are so fragmentary 
that the names must be regarded as zomina subnuda. Specific 
descriptions have been included in most cases, but omitted for the 
commoner well-known forms. Specimens have been cited in many 
cases where the species are obscure or little known. The nomen- 
clature naturally follows the principles set forth in the Philadelphia 
Code of 1904.* Especial care has been taken to ascertain the 
specific type of each genus, since it is believed that final stability of 
nomenclature can only be attained by grouping the species of a 
genus around one definite type. In one case only is a name intro- 
duced in direct contravention of the rules cited; that of Vernonza 
interior Baldwind already mentioned. In that case it would be op- 
posing the known facts of the relationship of the two forms if the 
code were followed in detail, an event certainly neither expected 
nor desired by its founders. 
* Bull. Torrey Club 31: 249-290. 1904. 
