-1 
PS 
MR. К. Hy COMPTON: AN INVESTIGATION OF THE 
SOPHORELE. 
This tribe, regarded as one of the most primitive of the Leguminosee and 
as affording a transition between the Papilionatee and the Ceesalpinioidese, 
has been imperfectly studied so far in respect of seedling anatomy. 
The three species examined have hypogeal germination *, but while in 
Castanospermum australe the collet abuts on the cotyledonary node, in the 
species of Sophora a few millimetres of hypocotyl intervene. Castano- 
spermum australe has the largest seed and seedling of any of the Leguminosee 
so far examined; but those of Sophora spp. are comparatively small, as in 
many Vicieæ. The cotyledons are inserted opposite one another. 
The type of symmetry varies widely ; diarchy occurs in Sophora spp., 
tetrarchy in Virgilia capensis according to Van Tieghem and Douliot (1888, 
p- 174), while the complex Castanospermum australe may be hexarch or octarch. 
All the protoxylems of the root are cotyledonary in the cases studied ; but 
in Sophora spp. a very definite direct connection occurs between plumular 
xylem and the metaxylem of the hypocotyledonary axis; a similar con- 
nection was observed in one case in Castanospermum. 
PoDALYRIE®, 
This tribe, with the Sophorew considered as the most primitive of the 
Leguminose, exhibits certain important features in seedling structure, 
though this has not been adequately studied. 
Germination in all the cases on record is epigeal; the seedlings are 
comparatively small, the hypocotyl uniform in diameter throughout, the 
cotyledons oblong or ovate, slightly thickened, 
The mature plants are small or medium-sized shrubs, or herbaceous 
perennials; in the examples studied the seedlings of the former show a 
tetrarch, of the latter а diarch root. Уйлана denudata exhibits an 
ordinary type of transition, in which the polar xylems are converted into 
double bundles and are joined by the halves of the lateral xylems ; all four 
root-poles being cotyledonary. In Pultenva daphnoides and Chorizema 
cordatum the transition begins in the same way from a tetrarch root, but the 
lateral xylems never enter the cotyledon, dwindling away in the upper part 
of the hypocotyl ; the cotyledon trace being simply a double bundle derived 
from a polar root xylem. Physiologically the lateral root-poles have a 
relation, not with the cotyledons as in Viminaria denudata, but with the first 
two plumular traces. In Baptisia leucantha and Thermopsis montana no 
lateral root-poles are developed, the transition being on the diarch plan 
* Lubbock (1892, p. 452) also describes as hypogeal Sophora secundiflora, Edwardsia 
chilensis ( = Sophora macrocarpa), and Myroxylon peruiferum. (The epigeal Нетаѓіотуіоп 
campechianum is included in the Sophorez by error.) According to De Candolle (1825, pl. 4), 
however, Sophora japonica and Virgilia aurea (= Calpurnea lasiogyne) are epigeal. 
