136 
PACIFIC SCIENCE, Vol. Ill, April, 1949 
of the above three herbaria, then complete 
citation is made; also complete references to 
herbaria are given in special cases. 
Abbreviation of herbaria: The depositories 
of the specimens examined are abbreviated 
as proposed in Chronica Botanica (Lanjouw, 
1939), where complete names and addresses 
of the herbaria are given, arranged in alphabetic 
sequence according to city in which they are 
located. The following abbreviations appear in 
this paper: A (Jamaica Plain), Bish (Hono- 
lulu), BM (London), Br (Bruxelles), Bri 
(Brisbane), Cal (Calcutta), F (Chicago), Fi 
( Firenze ) , G ( Geneve ) , GH ( Cambridge ) , 
K (Kew), L (Leiden), Min (Minneapolis), 
Mo (St. Louis), NY (New York), P (Paris), 
Sing ( Singapore ), UC ( Berkeley ), US (Wash- 
ington). 
Acknowledgments : Acknowledgment is made 
to the directors and curators of institutions 
from which herbarium material was borrowed; 
my particular gratitude is expressed for the 
aid rendered to me by Mr. B. A. Krukoff, Dr. 
H. A. Gleason, Dr. Bassett Maguire, and Dr. 
H. N. Moldenke. 
DISCUSSION 
Two new sections of the genus Alstonia are 
proposed in the present paper. One is based 
on the genus Winchia and consists, at present, 
of but a single species, including in its syn- 
onymy A. p achy carp a and A. ro strata. The 
section Winchia has syncarpous ovaries and 
connate follicles, features not previously recog- 
nized in Alstonia. The other new section em- 
braces species which have hitherto been incor- 
porated in § Dissuraspermum. Seed mor- 
phology is stressed as being of sectional im- 
portance. 
It is not surprising that only one new 
species ( Alstonia Brassii) is described, for in 
a genus consisting of conspicuous, greatly 
polymorphic species superfluous names are 
usually freely published until a comprehensive 
treatment is rendered. 
Study has forced a disposition of the well- 
known A. villosa in synonymy under the neg- 
lected but prior A. spectabilis. A. congensis is 
identified with A. Gilletii; consequently the 
greater part of the material hitherto distributed 
as A. congensis is now referred to A. Boonei. 
Blaberopus Sebusi, long since buried and for- 
gotten as merely a mixture of A. venenata and 
A. neriifolia, is resurrected as a legitimate 
species of Alstonia, with the description of a 
variety adduced as supporting evidence of its 
integrity. A . Muelleriana is reinstated. Several 
other novel interpretations, sometimes radically 
divergent from those currently accepted, have 
been necessary. 
Eight new varieties 2 are introduced in the 
present work. This is partly a reflection of the 
polymorphic nature of Alstonia; these varieties 
often serve as convenient intermediates between 
species that are clearly distinct only in their 
typical forms. 
1 am not completely satisfied with my treat- 
ment of several problems in Alstonia. No 
authenticated specimens of A. spectabilis, which 
is one of the three basic species founded at the 
same time as the genus, were available. It is 
here designated as type of a new section and 
interpreted as widespread in the Malay Archi- 
pelago. Neither types nor authenticated ma- 
terial of A. glabri flora and A. Kurzii were seen. 
In § Blaberopus, I have a somewhat uneasy 
feeling concerning the great distributional gap 
shown by A. neriifolia, which is known from 
sub-Himalayan India and apparently appears 
also in Java. 
Most disturbing are several cardinal problems 
still unsolved concerning the species of Alstonia 
in the Pacific. A. plumosa was not identified; 
this, being the second Alstonia described from 
the area, has involved species both in New 
Caledonia and in the more eastern Pacific 
islands. No specimen of A. Godeffroyi was 
identified, and one may wonder whether the 
2 No use is made of trinomials with the formal 
designation "var. Pypica,” etc., for the nomenclaturally 
typical element. 
