153 
(4) Semiaquilegia :—species 4; one in Japan, Chusan and 
South-east China; two in Centra] China; one in North-Eastern 
America—the American form differs from Aquilegia in the reduc- 
tion of the petal-spur solely. 
(5) Enemion :—species 5; one in N.E. Asia and Japan; the 
rest in N.W. America 
(6) Souliea:—a single species, recorded from Kansu, Szechuan 
and Yunnan. 
Pending further study of the living plants, and particularly of 
ripe seeds which seem often not to be perfected, we have not 
attempted to indicate the exact distribution of the grandiflora 
group of species of Paraquilegia; the exsiccata have been placed 
under one or other of the types recognised by Maximowicz to the 
best of our ability, but with great diffidence. Except where other- 
wise stated the specimens quoted have been examined by us. 
ur views on the probable phylogeny of this group may be 
shown as follows :— 
| to capsular 
Berberidaceae 
VTE oe 
To port of Anemone | 
and Thalittrum . 
(Semaguilegia | 
| LNeTMLON 
e of t 
of development proceeded any further than Aquilegia. Then 
and Actaea, which in their turn may have been the progenitors 
certain herbaceous Berberidaceae (for instance Epimedium). 
