page d'accueil
commune de chaumont-gistoux
roses autres plantes horticulture environnementale liens, bibliographies, etc. contact
Botarosa > Roses > La Rosa (multiflora?) carnea of Redouté
Comment est faite une rose?
Pourquoi la botanique? 
L'herbier Crépin
Classer les roses
La Rosa multiflora var. carnea de Redouté et Thory
La Rosa multiflora var. carnea of Redouté (english)
L'énigme des roses en grappes
Les origines botaniques et géographiques de Rosa moschata Herrm.
Les aventures est-méditerranéennes de la rose gallique
La rose d'Abyssinie
Les roses des Quatre Saisons
Les Rosomènes
Rosa : descriptions types
Rosa : Typus descriptions (english)
La rose de Chine sauvage, un jeu de cache-cache botanique
Le rosier multiflore de Chine
Ward Sebba'auy, ancêtre des rosiers Bourbon
L'évolution du genre Rosa
Rosa mosqueta n'est pas Rosa moschata
Les roses créées à la roseraie communale
L'amateur crée ses roses
Enregistrer la diversité
Commer : réflexions en marge d'un "Concours Lépine" de la rose
Le jardin de roses, ou comment cultiver les roses en Belgique et ailleurs
2003 : année du bois mort
Liste des roses cultivées à Chaumont-Gistoux
List of the roses grown at Chaumont-Gistoux
Idées sur la conservation des variétés de roses
Ideas on Conservation of Roses
Die Bedeutung von Erhaltung
De l'utilité des grandes collections de roses
Les belles invasives
Questions et recommandations au Comité de conservation de la W.F.R.S.
Questions and recommandations to the Specialised Conservation Commitee of the World Federation of Roses Societies
Lord Penzance et Léonard Lille dans l'herbier Crépin
Humeur : Wichura, ou la rose à côté de ses pompes
Roseraie naturelle sans parfum ajouté

Investigation into a hairy rose bush

(First published in Roses Anciennes en France)

Between literature and nature, ...
... there are herbariums !
Journey to the islands and colonies...
... return to China ...
... detour by Indochina !
Stop press !!!

Between literature and nature, ...

At the beginning of the 19th century, in "Les Roses", Redouté and Thory described two cultivated rose bushes, one from China Rosa multiflora var. carnea , and the other perhaps from Japan Rosa multiflora var. platyphylla . They treated them as varieties of R. multiflora Thunb. But the latter, described by Thunberg in his Flora Japonica of 1784 from plants gathered in the Nagasaki region, was not officially introduced into Europe until around 1860.

The aim of this study was at the outset to try and define the botanical identity and geographical origin of the var. carnea, introduced hastily and thus one feels that it has played an important role in the genesis of our cultivated roses. This was to lead us to discover a very diversified group, which "Les Roses" of Redouté and Thory, leading work of its time, does not really take into account the scope and earliness of the introduction, at least in the colonies!
As Redouté and Thory have not left herbariums which could serve as reference, this is therefore the first obstacle to overcome. In actual fact, what is there to prove that the varieties sold in nurseries under the names of var. carnea and var. platyphylla are really those described and illustrated in "Les Roses"? I myself took a number of years before finding one (or several) variety which appeared to be like the var. carnea (one is often given 'Paul's Himalayan Musk' in its place), and I am still looking for something which satisfies me regarding the var. platyphylla ('de la Grifferaie' is the only thing I have received under this name up till now).

This is what Thory says about the geographical origin of the var. carnea : "Our rose grows spontaneously in China, this delectable region, where Flora reigns unrivalled; it was brought back by the honourable squire T. Evans, around 1804, and flowered for the first time in England in the nursery of Mr Colville. M. Boursault then had it brought from London to Paris in 1808 and it was not until four years later, in the month of August 1812 that it flowered in the garden of Doctor Cartier."

... there are herbariums !

China is large, almost a continent with its contrasting climatic areas which correspond to very diversified flora. The "packets" of the Crépin herbarium at the National Botanic Garden of Belgium (BR) contain considerable quantities of specimens under the appellation Rosa multiflora Thunb. Just for China alone, there is very varied material from shapes practically identical to the Japanese (mainly from the north-east to central China) to such different things (from the south and south-west) that it is difficult to understand why Crépin classified them with this species.
A more general examination however makes it simpler to characterize a fairly homogeneous regional group of flowers with very round buds with stipules and bracts deeply cut in strips in the heart of the Synstylae section. This group comprises R. multiflora, R. luciae, R. wichuraiana, etc. and the roses from the south and south-west of China classified by Crépin as R. multiflora.


Journey to the islands and colonies...

Before discussing the multiflora, I had found in another herbarium at BR, the herbarium of Africa, strange roses which made me think of "Hybrid Musk", created by the English clergyman, Reverend Pemberton (1) in the first half of the 20th century. These roses from the gardens of colonists, or naturalized in this corner of Africa (Tanzania) also had round buds, but they were very hairy (tomentum) on all parts of the inflorescence. In my mind, a similar hairiness, although much less dense, on the Hybrid Musk was one of the traces left by Rosa moschata in their morphology. Or else, these hairs came from elsewhere. On the roses from Tanzania the thickness of the hair was such that I could not help thinking of R. bracteata or R. clinophylla from the Bracteatae section (2).
During a first perusal in the herbariums of the Natural History Museum in Paris, I found here "hairy" roses very close to the Tanzanian roses, gathered in the colonies and French protectorates - Madagascar, the Île Bourbon and practically identical, the type of Rosa beauvaisii that Cardot noted "Countryside in the neighbourhood of Longtchéou, hedges and bushes, grows abundantly, April 1893". The flowers of these samples are semi-double to double and described as pink or "red". However, careful and scrupulous collectors are content to note "Rosa" on the herbarium without giving the name of the species.
I therefore come back to Thory's description for additional information on the morphology of the var. carnea, and I note that he describes a hairiness similar to what I had noticed in these roses from Tanzania and the French colonies. "The leaves [...] green and glabrous above, paler and pubescent underneath. [...] They are supported by a hairy petiole. [The] pedicels, as well as the common stem are covered with down similar to that on the petioles. The tubes of the calyx [...] are pubescent. The divisions of the leaf blade [...] are also pubescent ..."
At the same time (1820), Lindley in his "Rosarum Monographia" describes the same phenomenon for his Rosa multiflora : "Stipulae linear [...] downy beneath; petioles very villous; leaflets [...] hairy on both sides."
On returning to the multiflora of the Crépin herbarium, I discovered very similar specimens there which lead one to imagine the route followed to introduce this type of cultivars in Europe, but also perhaps independently in America - a specimen from San José de Costa Rica (3), in Central America (a twin which is also to be found in the general herbarium of Meise (Brabant)); - two specimens from Madeira under the local appellation of "roses of the Portuguese" (this name could lead one to think of a very much earlier introduction by Portuguese navigators, and the flora of the Madeira roses includes other things such as a white rose "with the fragrance of rice", closer to the "Noisette").
An interesting additional feature - the specimens from Madeira belong to two like shapes but different from one another:

  • The one possessing little leaflets (pointed like those of Rosa luciae) and very small flowers appears to be a vigorous climbing plant (4); the second plate of the same specimen shows strong sterile stems in full growth. This plant is similar to the one from San José de Costa Rica. This is the closest through all the above-mentioned characteristics of the var. carnea of Redouté. The colour of its flowers is not given.
  • The other possessing large, widely oval leaflets, is apparently identical to the roses of Tanzania, to certain from Madagascar and to the R. beauvaisii from Longtchéou. Here the flowers definitely have a pink or reddish, or even darker tendency (5).
    Is it not purely and simply from the var. platyphylla (6) described and illustrated in "Les Roses"?
    An important point. Thory only mentions the differences in proportion, colour of the flowers and roughness of the foliage between the var. carnea and platyphylla, but he does not indicate that the latter would be less hairy.


... return to China ...

But let us come back to the var. carnea. A number of authors attach it to a rose first of all considered as a botanical variety of Rosa multiflora, although it is probably genetically far removed: Rosa cathayensis (Rehder & E.H. Wilson) L.H. Bailey (synonym of R. multiflora var. cathayensis Rehder & E. H. Wilson), native of the Chinese provinces of Yunnan and Setchouan (7). These opinions are based on the work "Plantae Wilsonianae", which lists and describes the numerous plant species gathered by missionaries or during expeditions in China at the beginning of the 20th century. One has attributed to this rose the shape described by Plukenet as early as 1704 (1696 according to G. S. Thomas) as "Rosa sylvestris cheusanica, foliis subtus incanis, floribus purpureis parvis" (wild rose from China, underneath leaves covered with a whitish down, and small purplish flowers), but this short description could also very well fit the var. carnea, the var. platyphylla or others which are going to follow.
For the moment, herbariums of George Forrest and E. E. Maire borrowed from the Edinburgh Royal Botanical Gardens are being examined at Meise. At the time when these lines were written, the type specimen of the var.cathayensis of Rehder & Wilson and of other Chinese collections are the subject of a loan request from the Arnold Arboretum, in Boston.
If in the Edinburgh herbarium there are indeed specimens (uniquely from Yunnan) labelled var. carnea and var. platyphylla, practically all the specimens (23 in all) have hairy undersides to the leaflets, but only a few have sparse hairs in the inflorescence. On the other hand, the presence of glands
whether inflorescent or not, is frequent. Great differences from one specimen to another in the distribution of these, the size of the leaves and flowers as well as their colours and the duplicature of the corolla, give the impression that one is dealing with 23 different shapes.
Considerable consistency does however exist in the pale and dull colour of the two sides of the leaflets and the paleness of the thorns; the pedicels or stalks are thin and not very stiff, rather long in proportion to the size of the buds, 2 to 3 times longer than the R. multiflora of Japan, and forming a tighter angle with the main stems; articulation is little marked, unlike the R. multiflora, which often shows a notable thickening at this place; the inflorescence form corymbes or clusters which are not always very thick. This leaves the impression of a "semi-cultivated" population of regional origin (George Forrest often uses the expression "semi-cultivated" for his specimens of Rosa chinensis gathered in the same region), but very varied and where hybridization with other species is not to be excluded (8). Cultivating for a good number of years botanical or horticultural roses from the Synstylae section, I have been able to observe (for my greatest pleasure) their spontaneity to re-sow themselves and to hybridize amongst themselves.
Rosa cathayensis in any case does not appear to have been described from statistical studies of a wild population on a large scale.


... detour by Indochina !

There must be a botanical origin to the abundant hair on the rose bushes from the colonies and the var. carnea (I always look for an origin in nature before thinking about a mutation). One can of course hit a snag that the same plant is either more or less hairy depending on the climatic conditions under which it is grown. But if only the heat encourages the hairiness in the case we are studying, then why should the var. carnea of Redouté, grown in the temperate climate of the Paris region, have been so hairy if it comes from a species (Rosa cathayensis) that remains smooth and glabrous in the far warmer regions of the Yunnan or Setchouan? After some short detours through the Bracteatae section, with such similar hairs, I became immersed again in the group of the Musks, which comprise very ill-assorted things in the Crépin herbarium, often pubescent but less dense and with shorter hairs. Without success. Without thinking about it, I then started on a packet containing among others R. soulieana and R. wichuraiana, ... and there discovered Rosa tunquinensis Crépin !
Boulenger, in his "Révision des roses d'Asie de l'herbier Crépin" (Review of the roses of Asia in the Crépin herbarium) places on the borderline between musked and multiflora this species described by Crépin on the basis of material from the extreme south of China and from the north of the Indochinese penisula. Here the shapes are varied but particularly hairy (9); there are pink or red semi-double flowers. The likeness with the roses from the Colonies as well as R. beauvaisii is glaringly obvious, even though here the buds are more elongated and in fact make one think of the R. moschata group. The lateral appendages noticeably present and elliptical - whilst they are closely elliptical to linear in R. cathayensis- strengthens even more the impression of resemblance with R. moschata, but a little wider crown is added in this case, nearly of the same shape (10).  Boulenger also describes their R. cardoti and R. lecomtei from these regions, also hairy and which differ by the
glandulous nature of certain parts of the inflorescence, a detail which can vary within the same population. But in this material one also finds shapes with small single white flowers, from hedges near Hanoï and the Lan Mat mountain, and in western Tonkin, which give the impression of being more primitive, and perhaps closer to local wild populations.  Does this mean that one is now getting closer to the geographical and botanical origin of the roses which have contributed this hairiness?

One must not forget that on the biodiversity maps on a worldwide scale, these regions and even more so the mountain ranges which mark the frontier between Laos today and Thailand are shown in dark red!!! This means that diversity here is at a maximum and that at the present time one can still expect to discover unknown vegetal and animal species. Therefore why not new roses and even key species for the understanding of the genus and its earlier cultivars!
Herbariums collected more recently in the Seventies in Java during a Japanese expedition show a hairy form with single pink flowers morphologically intermediate between R. tunquinensis and R. beauvaisii. Perhaps it concerns naturalized plants in a region of plains at around 2,000m, but if this is not the case, the Rosa genus would therefore considerably overshoot the equator. Seeing the method of dispersion of its grains at long distance, and the distribution of various other genus of temperate origin in these regions (Rhododendron, Hydrangea, etc.), that would not be at all surprising.

Stop press !!!

Recently on perusing the multiflora of the Crépin herbarium to make an inventory of something completely different, by chance I came across the folders given over to the cultivated multiflora and here discovered specimens sent to Crépin from the Villa Thuret, at Antibes by Maurice de Vilmorin. They had been sent to him from China by missionaries, are hairy and .....very remontant (11) !!! Their growth is not described, but they look very like the rose with little leaflets from Costa Rica and Madeira, with just the buds a little more pointed and more pronounced appendages (influence of R. tunquinensis?). However Thory wrote this about the var. carnea : "... it is only four years afterwards in the month of August 1812, that it gave its flowers in the garden of Doctor Cartier ..." Further on, he also mentions: "... One must guarantee that they are protected from the cold using good cover; for M. Delaunay remarked, as a good gardener that the frosts of November and December 1812 destroyed the grafts and ungrafted seedlings which had been left in open ground... everywhere, in Paris and in the neighbourhood." Would not this modest hardiness be precisely the reason of the very southern geographical provenance of the var. carnea ?
 It would appear that here and there in south Africa and elsewhere nowadays one still comes across very old hairy, perpetual flowering multiflora rose bushes. I am hoping to be able to talk about them in the next newsletter, for I already know that there will be additional information on R. cathayensis; the type and other herbarium from the Arnold Arboretum are actually on the way to Meise - I have just received confirmation.



1. Roses such as 'Buff Beauty'  look very much like R. beauvaisii. In fact, morphologically, among the "Pemberton" one above all detects the influence of R. gigantea and these bushes of the cathayensis / tunquinensis which must have been involved several times in their genesis at the same time as other things, including Rosa moschata and perhaps genuine multiflora of Japanese origin. This complex question is far from new and it was not initiated in the West; see also note 8 on this subject.

2. For the Rosa, dense villous or woolly hairiness, spread to the inflorescence, at the hypanthium, even to the sepals seem to be characteristic of south Asia. One finds it again in the Synstylae for R. longicuspis among others (although this makes one think of an introgression of Chinenses in the Synstylae), but also with certain shapes of R. cymosa from the Banksianae section, with R. bracteata and clinophylla of the Bracteatae section, with R. sericea (individual variation within the populations). Rosa rugosa also has this feature on its stems, between the thorns ( phylogenetic link with the Bracteatae?)

3. Regarding these roses from the colonies and sea routes, central America and the West Indies were current stopovers when returning from the Indies so as to avoid the doldrums of the Gulf of Guinea; on the other hand, Father Labat mentions in his "Another journey to the islands of America"( "Nouveau voyage aux isles de l'Amérique"), already frequent exchanges of cultivated plants at the beginning of the 18th century between colonies on the Atlantic and the Indian Ocean.

4. A living bush of bygone origin found climbing up very high in a walnut tree in Belgium chez Jelena de Belder would seem to correspond to this variety, but its flowers are white... Jelena de Belder, at present vice-chairwoman of the R. H. S., a horticulturalist specialising in ligneous plants (world specialist on Hamamelis) has travelled widely and made many exchanges with Asia since the Sixties.

5. Living bushes morphologically close to Rosa beauvaisii, one found in Corsica recently by my friend Jan Balis, organiser in the Sixties of various exhibitions at the Royal Library Albert 1st in Brussels on the history of roses and cultivated plants; another received by Odile Masquelier from India under the name of R. multiflora var. carnea are at the moment undergoing genetic analyses in Lyon and seem identical. These shapes have fairly large leaflets and consequently do not correspond to the var. carnea such as is described in "Les Roses".


6. The Japanese origin described by Thory for the var. platyphylla is only based on this ill-defined evidence: "This magnificent variety, remarkable for its foliage and the colours of its petals, was introduced in France by M. Noisette, where it flowered in September of last year (1819). He himself discovered it in 1817, in the garden of a market gardener outside London who had obtained it from seeds received from Japan, and who gave him the whole plant." In Bean (1980), Desmond Clarke who compares it to var. 'Grevillei' (described in the Gardener's Magazine by Loudon in 1828 and perhaps introduced from Japan before 1809 by Charles Greville, one of the founders of the Horticultural Society) says of the latter, which could still probably be grown, that it has smaller and more numerous flowers, but nevertheless he suggests the collective appellation Rosa multiflora var. platyphylla for both, ... which is perhaps tenuous. Thomas mentions the appellation 'Scarlet Grevillea', which would be synonymous of 'Russelliana', for a bush which he says could be of the same origin as 'Grevillei'. Descriptions of morphology and colour of flowers does not allow one to differentiate 'Grevillei' from 'Russelliana'; however the former produces vigorous, not very frost resistant young shoots, which is not the case for the latter. The illustration of the inflorescence of 'Grevillei' in the same work undeniably reminds one of 'Russelliana', and this is not a copy of the illustration of the var. platyphylla by Redouté, contrary to the illustration further up on the same page, which is a copy of his var. carnea. 'Russelliana' has not the hairy inflorescence, and its stems do correspond to the saying "bramble-like texture", described by Loudon in Arboretum & Fruticetum Brittanicum for 'Grevillei'. They are covered with a mixture of thorns and needles of various sizes; should one perhaps see traces of a previous old hybridization with a species from another section (the Cinnamomeae, for example), like one feels that this is the case for the populations of Rosa maximowicziana, found in the wilds in Korea and Manchuria.
'Crimson Rambler', an old Japanese cultivar, probable hybrid between R. multiflora and R. wichuraiana has also been mistaken for var. platyphylla.

7. It is perhaps from this region that a non-hairy plant described by Lindley as producing fruit more easily than the var. carnea originated: "Its fruit has never before been described. For an opportunity of examining it I am obliged to Mr. Lambert, in whose possession is a specimen brought from China by Sir George Staunton, of what is certainly this plant, without the pubescence of peduncles and calyx; which is therefore deciduous." One must also mention the shape with single flowers indicated under cultivation by Thory, although he does not mention the question of hairiness : "This bush which we have not seen in flower, is noted here following the evidence of M. Noisette, who assures us that he has observed it in the Apothecaries Garden in London. Mr Anderson gave him a plant which he multiplied and which one can obtain from his nursery garden. [...] Vulg. Multiflore à fleurs simples."

8. Yunnan and Setchouan are considered as the cradle of rose growing in China. Only examining one species at a time gives a short-sighted view of reality; often I refer to the musk group or to the Chinenses section and between others to slides of the R. chinensis herbarium collected by Forrest in the region. The untypical stripped stipules of the Chinenses shown in 'Old Blush', as well as its slender and elongated pedicels could come from R. cathayensis. In other respects, R. godefroyae Carrière described and illustrated under the mistaken appellation of R. pissartii or R. nasturana in the The Genus Rosa of Willmott must be a hybrid of R. cathayensis.
Remontant and dwarfing mutations have been observed on the botanical Synstylae; it is not impossible that it is these that have brought this feature to Rosa chinensis Jacq. But if the latter is a hybrid, which genetics seems to confirm, one must expect not to find in its descendants a like image of itself, but more likely various rearrangements of features originating from its parents.

9. Boulenger says of hairiness: "...branches covered with very abundant pubescence, even spreading itself onto the thorns."  He likens it to the R.moschata var. dasyacantha Cardot, from Yunnan.

10. Thorough examination shows appendages and crowns intermediate in shape between R. cathayensis and R. tunquinensis among R. beauvaisii , roses from the colonies and Java mentioned at the end of the article.

11. Here is a translation of the letter in full attached to the herbarium:

"... Paris, 28th January 1898

Dear Sir, I am afraid I lack the time and talent to decide, from the two small samples of Rosa microcarpa Lindl. which by the way you will receive by registered post, whether the plant that has produced them has the legitimate titles to be named as such or whether it usurps its rights!
The aristate appearance, with pectinate stipules makes me fear that the good missionaries of Zi-ka-way have sent me a variety of multiflora
Th. I hope I am wrong! I only have today these two small samples
collected a month ago (out of season) but in May and during the summer
the two stock planted at Antibes should be able to supply them indefinitely.

Yours sincerely,

Maurice de Vilmorin ..."

ivan louette 2002


© ivan louette et Odile Masquelier, 2002.
Reproduction électronique autorisée moyennant mention de la source.
Republication papier autorisée, aux conditions suivantes.
Ni la Commune de Chaumont-Gistoux, ni les personnes agissant en son nom ne peuvent être tenues responsables de l'usage qui pourrait être fait des informations diffusées sur ce site Web.