314 WILSON EXPEDITION TO CHINA 
cuneate at the base with straight sides, and nearly evenly toothed. In our No. 3546, 
leaves similar to the above and others nearly orbicular or obovate-orbicular, and 
irregularly toothed, occur on the same branch. 
Euptelea Franchetii Van Tieghem in Jour. de Bot. XIV. 272 (1900). 
— Vilmorin & Bois, Frut. Vilmorin. 9, fig. (1904). — Finet & Gagne- 
pain in Bull. Soc. Bot. France, LII. Mém. IV. 25 (1905), quoad speci- 
mina Fargesii et Henryi; Contrib. Fl. As. Or. II. 25 (1907). 
Euptelea polyandra Diels in Bot. Jahrb. XXIX. 346 (non Siebold & Zuccarini) 
(1900). — Pampanini in Nuov. Giorn. Bot. Ital: n. ser. XVII. 267 (1910); 
XVIII. 115 (non Siebold & Zuccarini) (1911). 
Euptelea Davidiana Hemsley in Hooker's Icon. XXVIII. t. 2787 (non Baillon) 
(1905), quoad tabulam et specimina Fargesiana, Wilsoniana, Henryana 
No. 10746 excepto. 
Western Hupeh: Hsing-shan Hsien, woodlands, alt. 1000- 
1600 m., May, June and August 1907 (No. 139; tree 5-13 m. tall); 
Changyang Hsien, woodlands, alt. 1000-1600 m., July and September 
1907 (No. 139*; tree 6-10 m. tall); Patung Hsien, woods, alt. 1000- 
1600 m., October 1907 (No. 219; small, bushy tree 3-10 m. tall); 
Hsing-shan Hsien, woods, alt. 1300-2000 m., November 1907 (No. 
588; tree 13 m. tall, 1 m. girth); without locality, April, August and 
October 1900 (Veitch Exped. No. 1048); without locality, A. Henry 
(No. 6455); “ Monte Si-ho, Ou-tan-scian,” July 1909, C. Silvestri 
(No. 2960). Eastern Szech’uan: without precise locality, 
A. Henry (No. 7337). South-east Szech’uan: Nanch’uan, A. 
von Rosthorn. Shensi: Tai-pei-shan, 1910, W. Purdom; “ Kin-ton- 
san,” July 1897, J. Giraldi; various localities, June, July and August 
1899, Hugh Scallan. 
As pointed out by Hemsley (1. c.) the particular characters on which Van Tieg- 
hem founded this species are not constant and are of little value. The ovérlooked 
character of the green under-surface of the leaves, with perfectly smooth non- 
papillose epidermis is, however, constant and affords a means by which this species 
may be distinguished at a glance. Specimens before us prove that the relative 
length of the filament and anther is merely a question of age and growth, as enun- 
ciated first by Hemsley, whose figure (l. c.) admirably depicts this species but 
whose description in part includes the western species E. pleiosperma, Hooker & 
omson. 
E. Franchetii is one of the most common of small trees throughout western 
Hupeh and eastern Szech’uan, occurring by the sides of streams, on the margins 
of moist woods and thickets. 
The species of this genus have proved very difficult to deal with systematically 
and various authorities have arrived at different conclusions in regard to them. 
With the mass of material before us we are convinced that in China two species, 
each occupying well-defined geographical areas, occur. The shape and size of the 
leaves and degree of their dentation, the relative length of the filaments and 
