332 WILSON EXPEDITION TO CHINA 



2600 m., July 1908 (No. 3595; bush 2-4 m. tall, flowers white); 

 Sungpan Ting, alt. 2600-3300 m., August 27, 1910 (No. 4012; bush 

 1-1.5 m, tall, fruit scarlet with orange-red fleshy peduncle) ; Mt. Omei, 

 June 1904 (Veitch Exped. No. 4879); same locality, alt. 3500 m., E. 

 Faher (No. 528); without locality, alt. 1300-4000 m., July and Sep- 

 tember 1903 (Veitch Exped. No. 3546). Yunnan: Sung-kwei, Ho- 

 ching-cho and Lichiang Fu valleys, alt. 2100-3000 m., April 1906, G. 

 Forrest (No. 2022); Lichiang range, alt. 2800-3300 m., June 1906, (?, 

 Forrest (No. 2256) ; side valleys on the eastern flank of the Tali valley, 

 alt. 2400-2700 m., June-July 1906, G. Forrest (No. 4448) ; eastern flank 

 of the Tali range, alt. 2100-2400 m., G. Forrest (No, 4451). Kansu : 

 Min-chou, alt. 2600-3000 m., 1911, W. Purdom, Shensi: '* Kian- 

 san," August 4, 1897, G. GiraUi; " Kin-tou-san," July 11, 1897, G. 

 Giraldi. 



This Rose is abundant in upland thickets, on the margins of woods and in forest 

 glades everjnvhere on the higher mountain ranges of western Hupeh and Szech uan. 

 In certain characters it is extremely variable. The shoots are smooth or densely 

 setose; the prickles are fairly numerous or wanting, and are only slightly dilated 

 at the base or are strongly decurrent and extend the full length of the internode. 

 The leaves sometimes have up to eight pairs of leaflets and thus exceed in num- 

 ber of leaflets those of any other Rose; the leaflets are nearly glabrous or silkily 

 pubescent on the under surface. In a general way it may be said that the more 

 exposed the situation in which the plant is growing the smaller the leaves, the larger 

 the prickles and the more abundant the setae. The flowers are always white and 

 solitary with four or very rarely with five petals, and the foot-stalk of the fruit 

 is always thickened, fleshy and brightly colored. It is this last character together 

 with the greater number of leaflets that at once distinguishes this species from 

 Rosa sericea Lindley, which has not yet been found in China. The specimens from 

 Yunnan have fewer and broader generally only 3 to 5 pairs of leaflets and resemble 

 in this character the Himalayan R. sericea, but the stalks of the fruit are apparently 

 fleshy. In Lindley's plant the foot-stalk of the fruit is slender and not thickened or 

 fleshy, and the leaves never have more than five pairs of leaflets. It is a native of 

 the western Himalaya, and the eastern limits of its range are not yet clearly de- 

 fined. We should not be surprised to learn of Rosa omeiensis Rolf e being found afl 

 far west as Sikkim. 



Rosa omeiensis, f. pteracantha Rehder & Wilson, n. comb. 



Rosa sericea Crdpin in Bull. Sac, Bot. Belg. XXV. Compt. Rend. 9 (non Lind- 

 ley) (1S86), quoad specimen No. 861. — M. de Vihnorin in Jour. Hort. 

 Soc. Lond. XXVII. 490, fig. 140 (1902-03). 



Rosa sericea, f. pteracantha Franchet, PI, Delavay. 220 (1890). — Focke in 

 Not. BoL Gard. Edinburgh, V. 70 (PI. Chin. Forrest.) (1911). 



Rosa sericea frudu rubro aculeis decurrentihus Vilmorin & Bois, Frut. Vilmarin. 

 99, fig. (1904). 



Rosa sencea, var. pteracantha in Gard. Chron. ser. 3, XXXVIII. 260, figs. 98, 

 99 (1905). — Bean in GarJen, LXIX. 294, t. (1906); Trees (S: Shrubs BriL 

 Isl. 11. 442 (1914). — Hutchinson in Bot, Maa. CXXXIV. t. 8218 (1908). 



