as) 
Oct. 23. C. dentata from roadside stand in New Jersey; said to 
come from native trees in Va.; through Miss Maud 
Ee Purdy: 
Oct. 29. C. dentata from Asheville, N. C., through J. Stuart 
Thomson, Glen Rock, N. J. 
. mollissima from Mr. L. N. Senor, Medford, L. I., 
from U. S. D. A. trees shipped in 1926. 
From Lu-Shan Arbore- 
tum and Botanical Gar- 
en at Han-Po-Kou, 
— 
es, 
ef 
= 
© 
= 
=| 
CO 
oOo 
—— 
N 
N 
Nov. mollissima (from Hupeh) 
Henryi | from vicinity 
. Seguinit | of Garden 
oS 
Lu-Shan, Kiukiang, 
China; through Mr. R. 
C. Ching. 
SyYSTEMATIC BOTANY 
The Classification of Dicotyledons 
3y ALFRED GUNDERSEN 
The study of flower structures and flower buds has been con- 
tinued with carefully made drawings by Miss Purdy. I have 
given special attention to flowers with parietal placentation and 
groups suspected of being related to them. | have given less at- 
tention to Sympetalae. Numerous additional cases where axile 
placentation is clearly preceded in the bud by parietal placentation 
have been found. In the older systems, including Jussieu, Ben- 
tham and Hooker, and [ichler, the two chief groups of parietal 
placentation represented by Papaver and Cistus were placed to- 
gether. This is followed also in the more recent systems by 
Wettstein, by Warming, and by Rendle. In the Engler system, 
alone, Cistus and its relatives were moved to a higher place to be 
near Cactus and Myrtus. However, these genera may all belong 
together nearer the beginning; and thus at the same time flowers 
with parietal placentation come clearly before those with axile 
placentation. Further, such a position suggests that Australian 
plants, e.g. Eucalyptus, may be rather primitive forms of Angto- 
sperms. The Dicotyledons (Sympetalae excepted) include about 
two hundred families. heir changing arrangement through half 
