1922] REHDER, NEW SPECIES, VARIETIES AND COMBINATIONS 209 



partly referable to this variety, is based, as its name indicates on P. 

 pyrenaica Lapeyrouse which belongs to P. halepensis Mill, or more es- 

 pecially to its var. brutia A. Henry (P. brutia Ten.). This is clearly 

 stated by H. de Vilmorin (in Bull. Soc. Bot. France, xl. lxxvii. [1893]; 

 see also Ascherson and Graebner, Syn. Mitteleur. Fl. L 219), who shows 

 that Lapeyrouse in 1813 under P. Laricio described andunder stood a 

 variety of P. Laricio found in the Pyrenees, while in 1818 in the Supple- 

 ment he published under the name P. pyrenaica a new description based 

 on a tree growing in his park near Toulouse and supposed by him to be 

 the same as his P. Laricio from the Pyrenees. The tree in his park, 

 however, was not the Pine from the Pyrenees, but P. halepensis var 

 brutia A. Henry, probably raised from seed received from the Orient 



Potentilla fruticosa var. Purdomii, var. nov. 



A typo recedit praccique foliolis minoribus, 7-10 mm. longis, subtus 

 glaucrscentibus ad costam sparse longe pilosis ceterum glabris vel fere 

 glabris floribus pallide luteis. — Frutex erectus ramis tenuibus elongatis 

 laxe pilosis; folia 5-foliolata; foliola elliptico-oblonga vel obovato-oblonga, 

 acuta, margine leviter revoluta, supra obscura viridia laxe longe adpresse 

 pilosa: fiores 1-1.8 cm. diam. in corymbis plurifloris; bracteae calycinae 

 oblongo-oblanceolatae, sepalis subaequilongae; stamina circiter 25, stylis 



paullo longiora. 



Cultivated at the Arnold Arboretum, raised from seed collected by W. Purdom 

 in southern China and sent under Seed No. 848 in 1911; specimen collected Sept- 

 ember 6, 1922. 



Though this form of Potentilla fruticosa L. differing chiefly in the smaller 

 leaflets glaucescent and nearly glabrous beneath and in the pale yellow 

 flowers, is not strikingly different from the type and some of its variations, 

 it cannot be referred to any of the forms described, and is therefore pro- 

 posed here, though reluctantly in this polymorphous species, as a new 



variety. 



Rosa Maximowicziana var. Jackii, comb. nov. — R. coreana Keller in Bot. 

 Jahrb. xliv, 47 (1909), not R. koreana Komarov. — R. Kcllcri Baker 

 in Willmott, Gen. Rosa, I. 75 (1910), not Dalla Torre & Sarntheim. — 

 R. Jackii Rehder in Mitteil. Deutsch. Dendr. Ges. xix, 259 (1910) ;in 

 Bailey, Stand. Cycl. Hort. v. 2998 (1916). — R. M aximowiziana Nakai, 

 Fl. Sylv. Kor. VII. 2(5, t. 3 (1918), only as to tab. 3. 



This Rose differs from typical R. M axi moivicziana only in the absence 

 of the prickles on the stems and branches which are numerous at least 

 on the more vigorous shoots in the type. When I described R. Jackii 

 I knew R. Maximowicziana Regel only from the description, and as the 

 species was described as a dense upright shrub, the branches and branchlets 

 armed with prickles and bristles, I concluded that it must be an entirely 

 different plant. Since then, however, we have received material of typical 

 R. M aximowicziana from Manchuria as well as from Korea and find that 

 the habit is not upright, but sarmentose as correctly described by Nakai, 



