246 ARBORETUM AND FRUTICETUM. PART III. 
London Horticultural Society, where one plant, ona wall exposed to the east, 
grows vigorously without any protection, and flowers freely in warm summers. 
On the whole, it is a very desirable species. It grows best in “ a loamy soil, 
and is readily multiplied by layers.” (D. Don.) 
App. i. Doubtful Species of Clématis. 
In Sweet’s Hort. Brit. are enumerated C. terniflora Dec., said to be in- 
troduced (from what country is uncertain) in 1826; C. diterndta Dec., from 
Japan, in 1825; C. dahirica Dec., from Dahuria, in 1822; C. diversifolia Dec., of 
uncertain origin; C. japénica Dec., from Japan, in 1826; C. semitriloba Dec., 
from Spain, in what year is uncertain; C. viorndides Jac. (which we have con- 
sidered as a synonyme of C. campaniflora), in 1826. These species, or 
names, will be found followed by specific characters in De Candolle’s Pro- 
dromus, and in Don’s Miller ; but, as they are not now to be met with in the 
gardens about London, we have deemed it not advisable to occupy our pages 
by describing them. 
App. il. Anticipated Introductions. 
On recurring to the lists in the historical part of this work, and comparing 
the names there given with those of the species described in the foregoing 
pages, it will be found that we are already in possession of all the species of 
Clématis which are considered to be natives of Europe. Of those of Asia, 
C. nepalénsis, pubéscens, vitifolia, and Buchananiana, according to the list 
in p. 173., may be expected from the Himalaya. From China, C. intricdta (men- 
tioned p. 176.) may be expected; and, from North America, there are the names 
C. holosericea, Walteri, and Catesbyana, which are not in our Catalogues as 
already introduced. There are probably other species in the mountainous 
regions of Asia, and in China, which will endure the open air in England ; and, 
as the seeds of the genus are light, not bulky, and very tenacious of life, the 
probability of their growth will amply repay travellers for the trouble of col- 
lecting them. 
Genus II. 
Vila 
ATRA’GENE L. Tue Arracene. Lin. Syst. Polyandria Polyggnia. 
Identification. Lin. Gen., p.615.; Willd. Baum., p. 45.; Dumont, 4. p. 426.; Hayne Dend., 118. ; 
Don’s Mill., 1. p. 10. 
Synonymes. Clématis Lam, and Dec.; Atragene, Fr. and Ger. 
Derivation. The name of Atragene appears to be taken from two Greeks words ; athros, pressed, and 
genos, birth; alluding, as it is supposed, to the manner in which the branches press against or clasp 
the — raed eee them. It was first used by Theophrastus, and was by him applied to Cle- 
matis Vitalba L. 
Gen. Char. Involucre none. Sepals 4, somewhat induplicate in the bud. 
Petals numerous, shorter than the sepals. Cariopsides (carpels) terminated 
by a bearded tail. Cotyledons approximate in the seed.— Climbing deci- 
duous shrubs, with variously cut opposite leaves. (Don’s Miil., i. p. 10.) 
Perhaps no genus was ever distinguished from another on slighter differences 
than fhose extant between Clématis and Atragene. These are, the presence 
of petals in the flowers of the latter genus; though this is scarcely the case 
in A. ochoténsis; and, in the double-flowered variety of C. Viticélla, the meta- 
morphosed stamens, which give the flower its fulness, are considered to be 
petals. Hence De Candolle regards the different species of Atragene only 
as a section of the genus Clématis ({§ iv. Atragenc, Prod.,i. p. 9.); but, as 
