138 
PACIFIC SCIENCE, Vol. Ill, April, 1949 
implied that his subgenus Dissuraspermum 
might be wholly detached from Alstonia with 
better reason than Blab er opus had been. 
The character of syncarpous ovary and united 
follicles found in Winchia are in themselves 
not of sufficient importance to maintain a dis- 
tinct genus. There is experience in another 
genus of Apocynaceae, Wrightia, in which the 
character of connate follicles in certain species 
is of minor significance and cannot serve to 
segregate them generically from other species 
having widely divergent follicles. The close 
relationship between § Winchia and § Bala is 
very marked. 
Amblyocalyx owes its origin to an erroneous 
observation of the ovary of a specimen of A. 
angustifolia, in which the ovule masses were 
mistaken for single ovules. It was originally 
published without a specific name: "Species I 
(v. 2?) Borneensis, Beccari 1628 et 3207” 
Subsequently it was typified by 1628, Amblyo- 
calyx Beccarii. 
The genus Alstonia is amply distinct, its only 
close relative being Dyera. The § Bala is closest 
to Dyera . Besides its habit and vegetative sim- 
ilarity, the dehiscence of the anthers in some 
of the species in this section indicates affinity. 
The form of ciliation in the seeds of Alstonia 
is almost unique in the Apocynaceae. A similar 
ciliation is found in Laxoplumeria, a South 
American genus of trees having alternate leaves. 
The seeds of Alstonia are invested with cilia at 
both ends, the distinct hairs being in some 
species sparsely distributed and in others rather 
densely set, but at no time truly comose as in 
Holorrhena and Wrightia, for example, or in 
other apocynaceous genera. 
There is no easy way of distinguishing Al- 
stonia from some other genera. Rauvolpa, Taber- 
naemontana, and Ochrosia have been frequently 
confused with Alstonia. With the examination 
of the seeds no such confusion could be pos- 
sible. If the plant is a tree or shrub with ver- 
ticillate or opposite leaves, and its seeds are 
ciliate, it is an Alstonia. Given seeds, not only 
the genus but also the section can almost always 
be determined; without seeds, the assigning of 
an unfamiliar species to Alstonia would be a 
very difficult task. 
The distribution of native species of Alstonia 
in the major land groups of its range is as fol- 
lows: 
Africa — Boonei, congensis 
India — neriifolia, scholaris , Sebusi, venenata 
Burma to China — angustifolia, glaucescens, mac- 
rophylla, Mairei, scholaris, Sebusi, spatulata, 
venenata, yunnanensis 
Malayan Peninsula and Archipelago (Malaysia) 
— angustifolia, angustiloba, Curtisii, glauces- 
cens, macrophylla, neriifolia, parvifolia (Bor- 
neo ?), pneumatophora, rupestris, scholaris, 
spatulata, spectabilis 
Philippines — angustiloba, macrophylla, parvi- 
folia, scholaris, spectabilis 
New Guinea — actinophylla, Brassii, glabriflora, 
Muelleriana, scholaris, spectabilis 
Australia — actinophylla, constricta, linearis, 
Mtielleriana, ophioxyloides, scholaris, specta- 
bilis 
Melanesia (excluding New Caledonia) — scho- 
laris, spectabilis, vitiensis 
New Caledonia — Comptonii, Deplanchei, lance- 
olata, Legouixiae, Lenormandi, plumosa, qua- 
ternata, Roeperi, saligna, Vieillardi 
Fiji and eastern Pacific islands — costata, Godef- 
froyi, montana, Reineckeana, vitiensis 
Key to the Sections of Alstonia 
1. Corolla lobes with left margins over- 
lapping; seeds not acuminate or cau- 
date; stamens inserted manifestly 
above middle of corolla tube, located 
at or near throat; leaves 3- to 11- 
verticillate, lateral nerves close and 
numerous 2 
Corolla lobes with right margins over- 
lapping; seeds acuminate or acute at 
one or both ends 4 
2(1). Disc annular or not apparent, not lobed; 
seeds with thick rounded margins at 
ends; leaves rounded to acuminate 
but not finely pointed at apex; large 
trees 3 
Disc with two conspicuous deltoid to 
linear lobes; seeds with thin, often 
