Rosa centifolia L.

TSO logo

Sponsor this page

For information about how you could sponsor this page, see How You Can Help


Article from Bean's Trees and Shrubs Hardy in the British Isles

Recommended citation
'Rosa centifolia' from the website Trees and Shrubs Online ( Accessed 2023-10-03.


Common Names

  • Holland or Provence Rose


  • R. provincialis Mill., in part (1768), not J. Herrm. (1762)
  • R. gallica var. centifolia (L.) Reg.


An elliptic solid.
Lacking hairs smooth. glabrescent Becoming hairless.
Bearing glands.
(botanical) Contained within another part or organ.
Loose or open.
Egg-shaped; broadest towards the stem.
Egg-shaped solid.
Central axis of an inflorescence cone or pinnate leaf.
Enlarged end of a flower stalk that bears floral parts; (in some Podocarpaceae) fleshy structure bearing a seed formed by fusion of lowermost seed scales and peduncle.
Folded backwards.


There are no active references in this article.


Article from Bean's Trees and Shrubs Hardy in the British Isles

Recommended citation
'Rosa centifolia' from the website Trees and Shrubs Online ( Accessed 2023-10-03.

A lax shrub to about 5 ft high, its stems armed with numerous prickles of various sizes, the larger ones hooked, the others almost straight and narrowly based. Leaves drooping, with five or seven leaflets; rachis rough with stalked glands but not prickly. Leaflets broadly ovate, dull green and glabrous above, downy beneath, edged with large glandular teeth. Flower-buds broadly ovoid. Flowers nodding, solitary or few in a cluster, borne in late June or July, clear pink, very double, goblet-shaped from the incurving of the petals, becoming more lax when fully blown, exposing the tightly packed petaloids. Sepals spreading, longer than the flower-buds and covered, like the pedicels and receptacles, with sticky aromatic glands. Fruits roundish or ellipsoid, with a pulpy flesh. Redouté, Les Roses, Vol. I, p. 25, t.

This rose belongs to a small group of garden hybrids whose history can be traced back to the late 16th century. What is believed to be the first of the Holland roses was described by Clusius (Charles d’Escluse) in his Rariorum Plantarum Historia (1601). In 1589 he received two plants from John van Hogheland of a rose cultivated in Holland as, reputedly, the R. centifolia of Pliny, of which one survived and flowered with him (probably at Leyden) in 1591. The flowers were very double, some with about one hundred petals, the outer ones much larger than the inner, which took the place of the stamens. The fragrance was like that of ‘R. praenestina’ (which, sensu Clusius, is R. damascena) but with a suggestion of the scent of R. alba; the colour of the flowers was not unlike that of ‘R. praenestina’. He named it R. centifolia batavica (of Holland), though he himself doubted whether it really was the Rosa centifolia of Pliny, described as having scentless flowers. Clusius also mentions R. centifolia batavica II (altera). This he apparently never saw in flower, but he quotes van Hogheland, who sent him a plant in 1592, to the effect that it was the same as ‘Number I’ except in having smaller flowers. Linnaeus took ‘Number II’ as the type of R. centifolia, for no obvious reason but perhaps in the belief that ‘Number I’ was the very double ‘cabbage’ form and ‘Number II’ the normal form from which sprang the original Moss rose, which Linnaeus included in R. centifolia in the second edition of Species Plantarum.

The Holland rose had reached England by 1596, the date of Gerard’s Catalogus, and in the following year he described it in his Herball under the name Rosa Hollandica sive Batava, adding that it ‘is generally called the great Province Rose, which the Dutchmen cannot endure; for they say it came first out of Holland, and therefore to be called the Holland Rose; but by all likelihood it came from the Damask Rose, as a kind thereof, made better and fairer by art, which seemeth to agree with the truth.’ Parkinson, in his Paradisus (1629), gives a fuller description agreeing with that of Clusius, and uses a name similar to Gerard’s – R. provincialis sive Hollandica Damascena, ‘The Great double Damask Province or Holland Rose’.

It is not known, and perhaps never will be, whether the Holland rose was raised in the Low Countries or imported from somewhere in southern Europe or the eastern Mediterranean. Jean Bauhin, who saw it in flower at Pforzheim in Germany in July 1595, was told that it had been brought from ‘the city of Delphi’ and to have been purchased at a great price. But this sounds like a nurseryman’s tale and should be regarded with scepticism, if only because Delphi had started to crumble into ruins well over a thousand years before the Holland rose emerged. R. centifolia is obviously quite near to R. damascena, differing in its glandular-toothed leaflets, broader receptacle and fruits, and the spreading, not reflexed, sepals. From R. gallica, with which botanists have compared or even united it, it differs in its much taller growth, more prickly stems and nodding flowers. The gland-edged leaflets are not a point of distinction from R. gallica, since they are a common feature of some wild forms of that species. It is perhaps significant that there seems to be nothing in the old literature to suggest a Dutch origin for the dwarf forms of R. centifolia (var. pomponia); all have French vernacular names and some might well be older than the Holland rose. Dr Hurst’s view, however, was that it was bred from a cross between R. damascena and R. alba.

R. centifolia, now grown only by lovers of old-fashioned roses, has achieved immortality on the canvases of the Dutch and Flemish flower-painters and was first depicted by Jacques de Gheyn in a work dated 1603, not long after its introduction. Because of the complete doubleness of its flowers the Holland rose must have been sterile, and the few variants that existed up to the end of the 18th century must have descended from the original stock (of which there may have been more forms than those mentioned by Clusius) or have arisen later by sporting. By the middle of the 19th century ‘Cabbage Rose’ had become synonymous with ‘Provence Rose’, but earlier it had been recognised as a distinct variant, differing in its more fully double flowers, which had longer central petaloids and therefore lacked the hollow form of the classic R. centifolia. This form, which Andrews distinguished as R. provincialis multiplex, is beautifully portrayed in the plate reproduced in Blanche Henrey’s British Botanical and Horticultural Literature, Vol. II, facing p. 49, which orginally appeared in Edwards’s A Collection of Flowers (1783-95).

Early in the 19th century R. centifolia underwent a burst of evolution, following the appearance around 1800 of a single-flowered and fertile form, portrayed in Redouté, Les Roses, Vol. I, p. 77, t.


See Note on the name R. provincialis on p. 70.


See ‘Unique Blanche’, p. 202, the name by which this rose is usually known at the present time; Andrews (Roses, t. 20) called it R. provincialis alba. Although discovered in Suffolk, it grew in a hedge bordering the garden of a Dutch merchant and in this connection it is interesting to note that Clusius mentions a rose which he called R. centifolia batavica alba.

'Bullata' Rose Feuilles de Laitue

An exact counterpart of R. centifolia, apart from the luxuriant bullate (crinkled) foliage, of a brownish tint when young. It originated on the continent, before 1801. 5 ft. Very fragrant. (R. centifolia var. bullata Thory, in Redouté, Les Roses, Vol. I, p. 37, t.)

'Cristata' Crested Moss, Chapeau de Napoléon

A sport which originated about 1820, identical to R. centifolia in every respect except the flowers. These are of the same good pink, but slightly less globular, and the sepals are extended into many divided wings or appendages; they frame the bud with a parsley-like frill of green. The French name recalls the cockade-like effect of these wings. 5 ft. Sweetly fragrant. (R. centifolia cristata Prévost; R. centifolia muscosa cristata Hook., Bot. Mag., t. 3475; R. centifolia f. cristata (Prévost) Rehd.)

f. albo-muscosa (Bak.) Rehd

This name is based on the rose portrayed in Willmott, The Genus Rosa, Vol. II, p. 349, t., as R. centifolia albo-muscosa. This rose is almost certainly ‘Blanche Moreau’ (see p. 173). R. muscosa var. alba Thory, which Rehder gives as a synonym, is a white-flowered sport of the Common Moss rose (see ‘Muscosa’ below), either ‘Shailer’s White’ or the Bath Moss (Clifton Moss).

'Muscosa' Common Moss Rose

A sport from R. centifolia, characterised by the development of much-branched, moss-like aromatic glands on the calyx and pedicels, and the excessive glandularity of the leaf-rachis and branchlets. In size, foliage, and flower it is slightly less than R. centifolia itself, but both the rich pink colour and the fragrance are the same. This mutation first occurred on the continent before 1720, and was a great favourite in Victorian times (R. muscosa Ait.; R. centifolia f. muscosa (Ait.) Schneid.).The Common Moss sported on at least two occasions to white blooms – Shailer’s White Moss (1788) and the Bath or Clifton Moss shortly before 1818. What is grown today as ‘Shailer’s White’ may not be the original clone but is a charming rose, its flowers faintly blush-pink at the centre on opening (G. S. Thomas, The Old Shrub Roses, pl. 19). 4 ft. Sweetly fragrant. It has reverted to the Common Pink Moss in recent years.After the occurrence of a single-flowered sport of the Common Moss (R. centifolia f. andrewsii Rehd.; R. muscosa simplex Andr.) it became possible to raise seedling moss roses. The later hybrid moss roses derive from this, or from the moss form of the Autumn Damask (see R. damascena) crossed with Hybrid Chinas.

R parvifolia Ehrh. (1791), not Pall. (1788)

R. burgundiensis West., in part only (1770)
?R. burgundica Durande (1782)
R. burgundiaca Roess. (1802)
R. remensis DC. (1805)
R. pomponia var. remensis (DC.) Thory
R. gallica var. parvifolia (Ehrh.) Ser.
R. centifolia var. parvifolia (Ehrh.) Rehd

This rose is described on p. 194 as ‘Parvifolia’ but, having botanical status, it is mentioned here. Strictly it should be called ‘Burgundiaca’ and has generally been known as the Burgundy rose in Britain, but these or similar names have been used for the Centifolia pompons (see above) and would be more ambiguous than ‘Parvifolia’, which is founded on a well-established albeit illegitimate specific name.This rose, an old garden variety perhaps deriving from a plant found originally in the wild, was considered by Lindley to deserve the specific rank given to it by Ehrhart. It is certainly out of place under R. centifolia, in which Rehder includes it as a variety, and the only question is whether it is a mutant of R. gallica or a hybrid of it.Early accounts of this rose in French works are confusing. De Candolle, in Flore Française (1805), identified it with a rose grown in the Jardin des Plantes as R. remensis [of Reims], with the vernacular name Rose de Champagne; but the Rose de Rheims imported into Britain in the 1770s was different; as another vernacular name De Candolle adds ‘Rose de Meaux’, but neither of the two roses known in Britain under this name, and described earlier, were ‘Parvifolia’ (i.e., R. remensis as described by De Candolle). Other vernacular names for this rose given by Thory in Redouté, Les Roses, are also at variance with earlier accounts.Note on the name R. provincialis: In the 1768 edition of his Dictionary Miller gave as the nomenclatural type of R. provincialis the R. provincialis flore pleno ruberrimo of the Leyden Catalogue of 1720 (the so-called Index Alter, attributed to Boerhaave). This is of uncertain identity, but a reading of Miller’s account of R. provincialis and ‘Provence’ rose in the 1768 and earlier editions of the Dictionary makes it perfectly clear that his R. provincialis is R. centifolia as now understood, and not R. gallica var. officinalis – the name under which it appears in Rehder’s Bibliography. The author of Miller’s name was really Parkinson, who called the Holland rose R. provincialis seu [or] Hollandica Damascena. What Parkinson meant in this context by provincialis is not clear – perhaps simply ‘provincial’ or ‘of the United Provinces’, but, if so, he was departing from the nomenclature of some continental botanists of the 16th century, to whom R. provincialis meant the Damask rose.The name R. provincialis in Miller’s sense was taken up by Aiton in Hortus Kewensis (1789) and other works, but it dropped out of use after the publication of Lindley’s Monograph (1820), in which R. centifolia is used in the modern sense. Lindley also cleared up the confusion over the name R. centifolia, which Miller had applied to the double form of R. gallica known as the ‘Dutch Hundred-leaved’.It has been widely believed that R. centifolia is called ‘Provence rose’ because it originated in Provence. In fact we owe the name to Miller, who was simply translating provincialis as ‘of Provence’, its normal meaning. His nomenclature was not universally followed, however. Andrews, for example, used the more non-committal name ‘Province rose’. Lindley, too, evidently considered that Miller was wrong in associating R. centifolia with Provence and instead called it ‘Provins rose’ – an extraordinary blunder since the rose for which Provins was famous was a form of R. gallica (see p. 96). The similarity between the words ‘Provins’, ‘province’, and ‘Provence’ obviously invited confusion, especially as ‘Provins’ would have been pronounced ‘province’ in English. But the confusion seems to have existed only at the vernacular level. There is no evidence that the botanical name R. provincialis was ever used in the sense of ‘Rose of Provins’ until the student Herrmann did so in his doctoral thesis (1762).The belief that R. centifolia originated in Provence has led some writers on the history of the Rose to ransack The Romaunt of the Rose (believed to be partly by Chaucer) for evidence of its presence there in the Middle Ages. The work is of course a translation of parts of Le Roman de la Rose, and the evidence should have been sought in the original. There is none to be found there – or indeed in the English translation. The ‘Provincial roses’ (i.e., rosettes) of Hamlet Act III, Scene ii, were in all probability fashioned after the Rose of Provins or rather after the common Apothecaries’ or Officinal rose, for which the name ‘Rose of Provins’ was used by Shakespeare’s contemporary Gerard (see further under R. gallica var. officinalis).Although the name R. provincialis, without qualification, originally meant R. damascena, it was also used by some writers of the 16th century for other roses that were intermediate in flower colour between R. gallica and R. alba. Thus R. provincialis praecox was R. majalis and R. provincialis minor some dwarf rose with very double pink flowers.The account of R. provincialis in Willmott, The Genus Rosa (Vol. II, p. 359) is confused. The rose actually portrayed is one of the forms or hybrids of R. gallica var. holosericea; it has nothing to do with the R. provincialis of Miller.

var. pomponia [Roess.] Lindl.

R. pomponia Roess.
R. dijoniensis Roess.
R. burgundiaca Pers., not Roess.
?R. pulchella Willd

As usually defined, this variety differs from R. centifolia only in being smaller in all its parts. It seems to be really no more than an assemblage of miniature garden roses more or less agreeing with R. centifolia botanically but mostly disagreeing with it in floral style and probably of independent origin. The group, never a large one, is scarcely represented in gardens today. All the original forms originated in France, and several had been introduced to Britain by the 1770s.The nearest to a true Centifolia was the Rose de Bordeaux, also known as the Gros Pompon de Bourgogne (R. centifolia minor Roess., Die Rosen, t. 20 and of Thory in Redouté, Vol. III, p. 33, t.). According to Thory this form was fertile and many sub-varieties had been raised from it. The ‘Petite de Hollande’ (p. 195) belongs to this group.The Bordeaux rose was scarcely a dwarf. But two miniatures were introduced to Britain from France late in the 18th century as the ‘Rose de Meaux’ – Meaux being a town to the east of Paris. Of these the one remaining in cultivation (see p. 178) appears to be the Lesser de Meaux (Lawrance, Roses (1799), t. 50, as R. Pomponia; Curtis, Bot. Mag., t. 404 (1798), as R. provincialis var., and as Pompone rose or smaller de Meaux in text; Willmott, The Genus Rosa, Vol. II, p. 353, t., as R. pomponia). The Greater de Meaux is portrayed by Mary Lawrance (as the Rose de Meaux) in her plate 31 and seems very similar to the R. Pomponia of Roessig (Die Rosen, t. 37), which is the type of var. pomponia, and also to the R. pomponia of Redouté, Vol. I, p. 65, t. It is difficult to believe that this is the pompon of Bot. Mag., t. 407, as Thory suggested. The Greater de Meaux, which may no longer be in cultivation, differed obviously from the Lesser in its longer more slender pedicels, less pinnated sepals and more Centifolia-like flower; also presumably in its larger stature.Two other roses are placed under R. provincialis (i.e., R. centifolia) in the second edition of Hortus Kewensis (1810). One is the St Francis rose (Lawrance, t. 88), which was introduced to Britain in the 1770s. This had large flowers, but according to Trattinick it was dwarf in habit; he identified it with R. gallica regalis Thory in Redouté (Vol. II, p. 19, t.) and called it R. pumila. The other is the Rose de Rheims (Lawrance, t. 71), also with large flowers and not unlike the St Francis.