A spreading shrub 6 to 8 ft high, its branches green, with a pruinose bloom, armed with scattered slender or stout prickles of unequal size. Leaves with five or more rarely seven leaflets; rachis downy, prickly beneath. Leaflets ovate, broadly oblong-elliptic or roundish, up to 2 in. long, obtuse or often shortly acuminate at the apex, dull green and glabrous above, hairy beneath on the veins and slightly so on the blade, deeply and sharply toothed. Stipules broad, with narrow, spreading auricles. Flowers solitary or in threes, terminal or from the axils of the upper leaves or reduced leaves, semi-double or double, white, of medium size. Pedicels 1 to 11⁄2 in. long, weakly glandular-bristly, the bristles more or less extending on to the ellipsoid receptacle. Sepals pinnated, with narrow, leafy tips, glandular-bristly on the back, reflexed and soon falling. Fruits said to ripen rarely.
An ancient hybrid of European gardens, long considered by botanists to be the double form of a wild species. However, in 1873 the Swiss rhodologist Christ described a wild plant which he identified as a hybrid between the dog rose R. dumetorum (corymbifera) and R. gallica, and remarked on its great similarity to R. alba. After this, with Crépin’s endorsement, R. alba came to be accepted as a hybrid of the parentage suggested by Christ, but recently Klastersky has suggested that it may be a complex hybrid deriving from R. gallica, R. arvensis, and some white-flowered member of the Canina group (Fl. Europaea, Vol. 2 (1968), p. 26). Like R. canina and its allies, R. × alba is highly polyploid, all forms examined being hexaploid. The wild, single-flowered rose known as R. alba by early writers is of uncertain identity. Probably white-flowered dog roses were meant, or hybrids between the Canina complex and R. arvensis. The R alba var. humilis Thory in Redouté is probably R. × polliniana.
Typical R. × alba is little known today, and indeed the above description is drawn from old accounts, illustrations, and herbarium specimens. The rose ‘Alba Maxima’ (see page 170) is probably the R. alba of Gerard and Parkinson, though it does not seem to be quite the same as the R. alba of old continental works.
R. incarnata Mill. R. alba var. incarnata (Mill.) Pers.; R. carnea Dum.-Cours.; R. provincialis var. incarnata (Mill.) Martyn – There has been much confusion over the name R. incarnata. As used by some pre-Linnaean botanists, it, or the plural Rosae incarnatae, meant what is now known as R. damascena, while some French botanists of the last century took Miller’s R. incarnata to be a form of R. gallica with sparsely armed branches and glandular-compound leaflets. A comparison of authentic herbarium specimens shows, however, that Miller’s R. incarnata is identical to the ‘Cuisse de Nymphe’ of French gardens. The English name for this – ‘Maiden’s Blush’ – is sometimes attributed to William Hanbury, who has a charming passage about it in his Compleat Body of Gardening (1770-1). In fact, Miller himself used it in the 1752 edition of his Dictionary.
The Maiden’s Blush differs from R. alba in the colour of the flowers, the almost unarmed stems, the more numerous leaflets (mostly seven), and the presence of numerous needle-like eglandular prickles on the flowering branchlets below the bracts. It is almost certainly the same as the R. incarnata of Parkinson and a very old rose.
Miller was apparently unacquainted with the ‘Great Maiden’s Blush’ (R. alba var. regalis Thory), which is now commoner in gardens. But it is listed in Weston’s Flora Anglicana (1775) as ‘Great Maiden’s Blush Rose’, with R. incarnata major as the Latin name.
The R. incarnata of Bot. Mag., t. 7035, is not Miller’s but a form of R. gallica.
R. canina aggregate × R. gallica. – As mentioned above, many authorities have considered that R. alba derives from a cross between R. canina or one of its near allies and R. gallica. Spontaneous hybrids between the latter and R. canina sens. strict. have been recorded from a few localities on the continent of Europe. They show the influence of both parents in their armature, taking large, hooked prickles from R. canina and bristles or needles from R. gallica, which also contributes a suckering habit. The leaflets are usually five in number against seven in R. canina. The flowers are larger than in R. canina, bright rose or purplish rose, borne on pedicels which are shorter than in R. gallica, smooth or glandular-bristly. Putative hybrids between R. gallica and R. coriifolia or R. afzeliana, both members of the R. canina aggregate, have also been recorded, but are less well authenticated. The wild rose found by Christ in Switzerland and said to resemble R. alba (see above) may have been R. gallica × R. coriifolia (Boulenger, Bull. Jard. Bot. Brux., Vol. 12 (1932), pp. 462-76 and 459-61).
Altogether some twenty names of specific form have been given to forms or putative forms of these hybrids. The oldest is R. collina Jacq., but Boulenger remarks that there is nothing in Jacquin’s description or figure to indicate the influence of R. gallica. The name used by Rehder for R. canina × R. gallica is R. × waitziana, but the rose so named by Trattinick is of uncertain identity. For R. × waitziana var. macrantha (Desp.) Rehd., see R. ‘Macrantha’, in alphabetical order.