J Royal Hort. Soc. 66: 73-82, 242-250, 282-289 (1941)
Notes on the Origin and Evolution of our Garden Roses
C. C. Hurst, Sc.D., PH.D., F.L.S.

I. ANCIENT GARDEN ROSES (2000 B.C. TO A.D. 1800).

The introduction of the China Rose to England towards the end of the eighteenth century caused a complete revolution in the garden Roses of Europe, America and the Near East. The effects of this introduction from the Far East first became evident in the Noisette and Bourbon hybrids which appeared in the early years of the nineteenth century, so that we may take A.D. 1800 as a natural line of division between ancient and modern garden Roses.

The ancient Roses, for the most part, flowered only once a year, in the early summer, while the modern Roses bloom continuously from early summer to late autumn. In a favourable climate like the Riviera they may flower all the year round, since they are potentially perpetual. Recent research shows that this habit of continuous flowering is due to the action of a Mendelian recessive gene introduced into our modern Roses by the China and Tea Roses, already cultivated in China for a thousand years or more. A similar mutation was observed by the writer in the Botanic Gardens at Cambridge a few years ago in a batch of wild seedlings of one of the Chinese Musk Roses, R. Helenae Rehd. and Wils., from seeds received by Mr. Reginald Cory.

There was one ancient Rose which under favourable conditions of high culture and special pruning often produced a second crop of flowers in the autumn. This Rose was known in England as the Monthly Rose, in France as Quatre-Saisons, and in ancient Rome as Rosa bifera. It was, however, uncertain and sporadic in its second flowering and was in no sense a continuous bloomer like the China Rose. In the days of summer-flowering Roses a few scattered blooms in the autumn were highly prized.

A. THE RED ROSE OF LANCASTER (Rosa rubra [BLACKW.] 1757).*

*Syn. Rosa gallica L. 1759 (non 1753). R. gallica L. 1753 was a species of the Caninine, near to R. canina tomentella (Léman) and it was not until 1759 that LINNAEUS described the R. gallica of authors. In the meantime Mrs. ELIZABETH BLACKWELL had described and figured the species under the name of Rosa rubra (The Red Rose) in 1757.

The Red Rose is the foundation species from which most of our garden Roses have been evolved. The species, in its dwarf wild state, extends from France to Persia and has produced many natural hybrids with other species. The history of the garden Red Rose is lost in the night of time. According to GRAVEREAUX it was the Rose of the Persian Magi and the Median Fire Worshippers of the twelfth century B.C., who cultivated it for their religious ceremonies. According to their sacred writings of the ninth century B.C. (the Zend Avesta) this Rose was dedicated to an Archangel. In the fourth and third centuries B.C. the Red Rose was apparently cultivated by the ancient [74] Greeks and also at Miletus in Asia Minor, whence it was imported by the ancient Romans in the first century of our era. PLINY describes it as vivid in colour with not more than twelve petals. In the sixth century it appear as the Rose of the Codex Vindobonensis at Constantinople, which has been identified by GRAVEREAUX as the Dwarf Gallic Rose known as R. pumila. In the twelfth century the dark red Gallic Rose was cultivated by the Arabs in Spain with the tradition that it was brought from Persia in the seventh century.

THE APOTHECARY'S ROSE OF PROVINS (R. gallica officinalis THORY).

In the thirteenth century the town of Provins, south-east of Paris, became the centre of a great industry which persisted for more than six centuries, although its greatest production was in the seventeenth century. This industry arose through the discovery that a certain variety of Red Rose had the peculiar chemical property of preserving the delicate perfume of its petals even when dried and reduced to powder. The apothecaries of the period developed this discovery by making conserves and confections of Roses, in various forms, which gave rise to a great industry in the course of centuries. Documents show that in 1310 the town was able to offer presents of conserves and dried Roses to the Archbishop of SENS on his solemn entry, to CHARLES VII and JOAN OF ARC in 1429, to HENRY II in 1556, to LOUIS XIV in 1650, 1668, 1678 and 1681, to the Queen of LOUIS XV in 1725 and to the Emperor NAPOLEON I in 1814. In 1600 the historian of HENRY IV remarked that the main street of the town of Provins was peopled with apothecaries who make the famous conserve of Provins Roses which is sent all over France. The Provins Roses were also much appreciated in India and England, and in 1860, 36,000 kilos of Provins Rose petals were sent to America. This peculiar scented variety of the Red Rose grown at Provins for more than 600 years became known as the Apothecary's Rose and was beautifully figured by REDOUTÉ in 1817. There is a particularly good photograph of this Rose in EDWARD BUNYARD'S book on Old Garden Roses, 1936, Plate 25, under the name of 'Red Gallica.' For more than 300 years this Rose of Provins has been known in England as the Red Damask Rose although, as BUNYARD points out, it is botanically a gallica (rubra) and not a Damask. It was called a Damask because it was believed to have been brought originally from Damascus by a Crusader. The original plant, from which many thousands of bushes have been clonally propagated and grown at Provins and elsewhere, was said to have been brought to Provins from the Valley of Damascus by THIBAUT LE CHANSONNIER on his return from one of the Crusades. THIBAUT IV was King of Navarre and also Count of Champagne and Brie, which explains his interest in the town of Provins. Born in 1201, he was the author of 66 poems and evidently fond of Roses since he sings of them in its verses. In Le Roman de la Rose, written about 1260 (later translated by CHAUCER as Romaunt of the Rose), there is a reference to the Roses from the "land of the Saracens" and it is probable that we are [75] indebted to the Crusaders for the introduction from Eastern gardens of superior varieties of Red, Musk and Damask Roses.

ROSA MUNDI.

At some time in its history, the date of which is at present unknown, the ancient Rose of Provins gave rise by a bud-sport (somatic mutation) to the old-fashioned favourite striped Rose, Rosa Mundi, which occasionally reverts to its bud-parent. Rosa Mundi is often mentioned and figured by the old herbalists (CLUSIUS in 1583, BESLER in 1613, CASPAR BAUHIN in 1623) under various names, and there is a drawing of it by ROBERT in Paris at the Jardin des Plantes dated 1640. There is a popular tradition that this Rose is intimately associated with the Fair Rosamond of HENRY II, and in the older writers the name of the Rose is often written as one word, Rosamonde. The Fair Rosamond seems to have died in 1176, so that perhaps an earlier Crusader found the striped form in a Syrian garden and on his return presented it to her after giving it her name.

Both the ancient Rose of Provins and its bud-sport, Rosa Mundi, are tetraploid with 28 chromosomes in their body-cells and 14 in their male and famale germ-cells.

Up to the seventeenth century only a few varieties of the Red Rose were cultivated in Euope; in 1629 PARKINSON notes about a dozen; in the eighteenth century Dutch horticulturists, thanks to Van Eden and others, took up the breeding of Red Roses on a large scale, and the dozen varieties of Parkinson soon became a thousand.

Early in the nineteenth century, owing to the ardour and enthusiasm of the Empress Josephine, many of these were cultivated in France, England and Italy, whereas hitherto their cultivation had been confined to Holland, Belgium and Germany. Very few of these varieties of the Red Rose are now in cultivation, but owing to the courtesy and kindness of M. GRAVEREAUX of La Belle Roseraie de L'Hay I have been able to examine the chromosomes of a large number of the old Roses which are preserved in his collection. It may be said that all the garden varieties of the Red Rose examined are tetraploid with 28 chromosomes in the body-cells and 14 in both the male and female germ-cells, while many of the hybrids are triploids.

B. THE DAMASK ROSE (Rosa damascena [BLACKW. 1757]).

To the gardener the old Damask Roses are a very natural and charming group with their damask colouring, damasked pattern of the flowers and, above all, the damask fragrance of their perfume; he also likes to believe the old tradition of 1551 that they came originally from Damascus with the returning Crusaders.

The botanist is not so happy about the old Damask Roses although he has always been willing to admit that they are a natural group and has usually given them the rank of a species; he realizes, however, that their characters are very near to those of R. rubra and that these are sufficient to prove the origin of the Damask from the Red Rose. [76] At the same time he is aware that there are certain characters in the Damask Roses which are quite foreign to the Red Rose.

Analyses show that some of the Damask characters foreign to R. rubra are those of R. phoenicia Boiss. while the others are those of R. moschata Miller. It is evident therefore that the Damask Roses are all hybrids of rubra but that some are hybrids of phoenicia while others are hybrids of moschata.

Consequently the Damask Roses can be divided into two natural and distinct groups:

(1) x R. damascena (rubra x phoenicia). The Summer Damask Rose.
(2) x R. bifera (rubra x moschata). The Autumn Damask Rose.

(1) THE SUMMER DAMASK ROSE (x R. damascena [BLACKW.]).
THE YORK AND LANCASTER ROSE, ORD. TYPE.

To this group belongs the old striped Damask, the York and Lancaster Rose, which has often been confused in our gardens with the striped Red Rose, Rosa Mundi. There should be no difficulty, however, in distinguishing these two striped Roses, since (apart altogether from the striking differences between them in habit, armature, leaves and flowers) the pattern and colours of the striping are altogether different in the two Roses. In the Red Rosa Mundi the ground colour of the petals is pale rose-pink, irregularly but heavily striped and blotched all over with rosy-red. In the York and Lancaster the ground colour of ground colour of the petals is white, irregularly but lightly marked or blotched with blush-pink or rose, the striping and blotching being only partial. This famous Rose was faithfully described by JOHN PARKINSON in 1629, and there is little doubt that it is the Rose mentioned by SHAKESPEARE in Henry VI in the scene of the plucking of red and white Roses in the Temple Garden. In SHAKESPEARE'S Sonnet XCIX we get an intimate picture of the York and Lancaster Rose which will appeal to the practical gardener.

"The Roses fearfully on thorns did stand,
One blushing shame, another white despair;
A third, nor red nor white, had stolen of both
And to his robbery had annexed thy breath;
But, for his theft, in pride of all his growth."

These lines are of peculiar interest because they show that SHAKESPEARE had more than a casual acquaintance with the York and Lancaster Rose, he must have had personal experience in the growing the plant. My own experience is that this variety, in spite of its vigorous growth, has a poor constitution and does not last long without propagation and renewal.

The York and Lancaster Rose is a tetraploid with 28 chromosomes in its body-cells and 14 in its male and female germ-cells. The [77] chromosomes show definite signs of hybridity, with irregular pairing and weakened affinity. In two cases whole sets of seven chromosomes acted independently of the others in somatic divisions. Such disturbances are likely to affect metabolism and weaken the plant's constitution. The young pollen-grains were not healthy, about one-half being degenerate. The embryo-sac formation was normal and healthy.

THE HOLY ROSE.

* In view of the earlier and quite different R. sancta Andrews (1827), Rehder in 1922 amended R. sancta Richard (1848) to R. Richardii, and now both Richard's and Rehder's specific names become synonyms of R. damascena [Blackw.].

Another interesting but rather mysterious Rose, belonging to the first group of Summer Damask Roses, is the Holy Rose. Our knowledge of this Rose is somewhat scanty, yet it seems to go much farther back in history than any other garden Rose, and there are indications that it may have had a distant part to play in the evolution of our garden Roses. The little we know of its history is both curious and chequered. It was first found in Abyssinia, in the Christian Province of Tigre, whee it had been planted in the courtyards of religious sanctuaries. It was described by RICHARD in 1847 in his Flora of Abyssinia under the name of Rosa sancta.* The Holy Rose forms a low erect bush and is intermediate in its characters between the Red and the Phoenician Rose. Indeed, except for its single flowers with 5 to 6 petals, it would pass well for a dwarf Damask Rose. Since its phoenicia parent is not a cultivated plant, the Holy Rose is presumably a natural hybrid with its home in Asia Minor and Syria, where its two parents overlap, and the question arises how it came to be transported to Abyssinia. The story of ST. FRUMENTIUS may help to solve that problem. ST. FRUMENTIUS was born in Phoenicia about 300 A.D., and it is said that while on a voyage he was captured by Ethiopians, taken to Axum, the Abyssinian capital, and became the King's Secretary. He is said to have converted the Abyssinians to Christianity and secured its introduction to that country. In 326 he was consecrated Bishop of AXUM by ATHANASIUS at Alexander and died in 360.

In view of the above facts it seems likely that the Holy Rose may have been introduced to Abyssinia from Phoenicia (Syria) by the Phoenician Apostle of Ethiopia, ST. FRUMENTIUS, in the fourth century A.D. This also may explain why the Rose was planted within the precincts of the Christian churches in his diocese and thus preserved through the centuries.

THE ROSE OF THE TOMBS.

We next hear of the mysterious Holy Rose in the Tombs of Egypt. The Tomb in which it was found was in the cemetery of the town of Arsinoe of Fayoum in Upper Egypt, near to the Labyrinth Pyramid, and judging by other remains found alongside the Holy Rose it belonged to a period between the second and the fifth centuries A.D. The remains of this Rose were twined into garlands when found in [78] the Egyptian tomb in 1888 by the eminent English archaeologist, Sir FLINDERS PETRIE, who sent them to Kew Gardens for identification. Dr. OLIVER, having ascertained that they were Roses, forwarded them to Professor CRÉPIN, Director of the Brussels Botanic Garden. CRÉPIN reported that the nine Roses composing the garland were all of the same variety and were identical with Rosa sancta RICHARD cultivated at the present day in the courtyards of religious edifices in Abyssinia. CRÉPIN considered them to be a new form of R. gallica (R. rubra), but he did not think that they had ever been indigenous in Abyssinia or Egypt but that they had been introduced from Italy, Greece or Asia Minor. There may be some connexion between the garlands of Abyssinian Holy Roses in the Egyptian tomb and ST. FRUMENTIUS of Abyssinia, but until further evidence is obtained it remains a mystery.

THE MINOAN ROSE IN CRETE.

Finally we arrive at the Island of Crete, where Sir ARTHUR EVANS, in the course of his remarkable excavations, unearthed a fresco-painting near the ancient Palace of Cnossos that included a life-like representation of a Rose, which I saw in 1926 and which seemed to me to bear a striking resemblance to the Holy Rose of Abyssinia, Egypt and Asia Minor. It is hoped that further excavations will bring to light more of these Minoan Roses and add to our present scanty knowledge of them.

The central date of the great historic Minoan civilization of Crete may be put at about 3000 to 2000 B.C. It apparently maintained close contact with Egypt, Phoenicia and Greece, and there is no doubt that the beautiful Phoenician Roses would appeal to the high artistic sense of the Minoans. The selection of a single-flowered Rose by the Minoan artist for his fresco-painting does not necessarily mean that garden Roses were not cultivated by the Minoans; the double garden Roses of the time may have been too heavy for his artistic purpose.

So far as I have been able to trace, this Minoan fresco-painting is the earliest representation of a Rose in historical times.

(2) THE AUTUMN DAMASK ROSE (x R. bifera Poiret).

The Autumn Damask Roses are also very ancient and, like their parents rubra and moschata, have always been renowned for their fragrance and free-flowering qualities. We first hear of them in the Island of Samos towards the tenth century B.C., where they are said to have flowered twice a year and were freely used in the cult of Aphrodite (or Venus) which was later introduced to Greece and Rome, together with the Roses, which played a most important part in the ceremonies.

p. [79]

There is little doubt that the Roses which HERODOTUS mentions as growing in Macedonia in the gardens of Midas, each with 60 petals and surpassing all other Roses in fragrance, were the Autumn Damask Roses x R. bifera, and not R. centifolia as we used to believe. As we shall see later, R. centifolia as we know it, and as the eighteenth and nineteenth centuries knew it, first arose in Holland in the eighteenth century of our era and was therefore unknown to the ancient Greeks and Romans. Their Roses with a hundred petals, which naturally they called Rosa centifolia, were not our centifolia but evidently our bifera Damask Rose, which we know was cultivated by them at a later period at Paestum, Pompeii and other places. It is possible that Midas, who was King of Phrygia, may have brought the Roses of Herodotus to Macedonia from his gardens in Asia Minor and that afterwards they were distributed in Greece and Italy. Many Roses of the same kind were grown in Egypt for the Roman market, especially in the winter, as MARTIAL tells us "roses in winter bear the hightest price." In the first century B.C. VIRGIL, in his Georgics, sings of the Roses of Paestum and of their flowering twice a year, "biferique Rosaria Paesti," which can only refer to our Autumn Damask Rose. Frescoes of this Rose were found in the ruins of Pompeii, which was destroyed in 79 A.D., and PLINY himself perished in the catastrophe through his eagerness to get a new view of the great eruption of Vesuvius. In the tenth century, according to the Arab physician AVICENNA, the same Rose was cultivated on a large scale in Syria for making rose-water and for medicines. In the twelfth century it appears with the Arabs in Spain, and IBN-EL-AWAM describes our Damask Rose with much detail, suggesting that it came from the East.

THE ROSE OF ALEXANDRIA.

In the sixteenth century NICOLAS MONARDES, a Spanish physician, wrote a medical treatise on the Roses of Persia or Alexandria, which he said the Italians, Gauls, Germans and others call Damascenae because they believe them to have come from Damascus. His description agrees with the Autumn Damask Rose although he says nothing of its flowering twice a year. There is, however, indirect evidence that these Spanish Damask Roses did flower twice a year, for M. Lachaume, writing to the Journal des Roses in 1879 from Havannah, Cuba, states that in almost every Spanish home in Cuba there is the antique Quatre-Saisons Rose, which in Cuba is universally known as the Alexandria Rose.

All the varieties of Damask Roses so far examined, except one, prove to be tetraploid with 28 chromosomes in the body-cells and 14 in [80] the male and female germ-cells. The exception is a seedling Damask Rose in Kew Gardens which proves to be a pentaploid with 35 chromosomes in the body-cells, 21 in the female germ-cells and 14 in the male. This interesting Rose was raised at Kew from seeds of a hip gathered from the Rose growing on Omar's grave at Naishapur.

"When Omar died, the Rose did weep
Its petals on his tomb;
He would be laid, where North winds keep
The Rose in freshest bloom."
ANNA HILLS, 1884.

C. THE WHITE ROSE OF YORK (R. alba L.).

This familiar old garden Rose has persisted through the centuries and has been in general cultivation from the times of the Greeks and Romans. In the thirteenth century it was rather neatly described by the Universal Doctor of the Schoolmen, ALBERTUS MAGNUS, as "the white garden rose which has often 50 or 60 petals, it is very bushy and the branches are long and thin on a tree of which the trunk often attains the thickness of one's arm." In 1307 CRESCENTIUS, the Italian agriculturist, recommends it for planting a hedge. The Italian painters of the fifteenth century were fond of this Rose and portrayed the large 'Double White' and the 'Maiden's Blush' varieties (alba maxima and alba regalis), the flowers of which appear to be identical with those grown to-day. In the Wars of the Roses the 'Double White' English Rose was traditionally adopted as a badge by the Yorkists.

The old herbalists note and figure this Rose in the sixteenth century, and it usually appears first on their lists of Roses as "most ancient and knowne Rose ... King of all others " (PARKINSON). In 1753 LINNAEUS made it a species under the name of R. alba, and although it is obviously a garden plant most botanists have regarded it as a species until 1873, when CHRIST suggested that it was a hybrid between gallica (R. rubra) and canina, and CRÉPIN, with others, have agreed with this interpretation.

An analysis of the characters of x R. alba (L.) (1873) shows the definite influence of R. phoenicia Boiss. in this hybrid, and I feel bound to conclude, therefore, that x R. damascena, rather than R. rubra, was one parent, and that a white-flowered, almost prickleless form of R. canina was the other. Further, from the chromosome number we can say definitely that canina was the female parent and not the male. x R. alba is a hexaploid with 42 chromosomes; R. canina is a pentaploid with 28 chromosomes in the female germ-cells and 7 in the male; x R. damascena is a tetraploid with 14 chromosomes in both male and female germ-cells. Consequently, if R. canina had been the male parent, X R. alba would have been a triploid with 21 chromosomes instead of a hexaploid with 42.

The only white-flowered canina with few prickles known to me is R. C. Froebelii Christ, which grows in Kurdistan and possibly extends [81] to the Crimea and the Caucasus. This is the form of Briar known in gardens as R. laxa, and formerly used as a stock for budding garden Roses. It has no connexion whatever with the R. laxa of RETZIUS or of LINDLEY, nor, as REHDER suggests, with R. coriifolia Fries.

x R. alba has been reported as growing wild in the Crimea, and it may be that it originated there as a natural hybrid; it has also been reported from several places in Central Europe, but these may be garden escapes or otherwise of garden origin.

In the single and semi-double forms of x R. alba which are fertile, the female germ-cells carry 28 chromosomes while the male germ-cells carry only 14. All the seedlings examined have 42 chromosomes.

D. THE OLD CABBAGE ROSE (R. centifolia L.).

For many years most of us have believed that the Old Cabbage Rose of our great-grandmothers was the most ancient Rose in the world. We all followed the old herbalists who agreed that Rosa centifolia, the Queen of Roses and the Queen of Flowers, was cultivated by the Greeks and Romans, noted by HERODOTUS and named by THEOPHRASTUS and PLINY. We took it all in, and it seemed to us the most natural thing in the world for the classic Roses with a hundred petals and the most fragrant of perfumes to be our Rosa centifolia, especially as THEOPHRASTUS and PLINY called them by that name. Modern research has changed all that and will not be denied. It appears now that our R. centifolia did not actually arrive until the eighteenth century and is therefore the youngest of all our old Roses.

Thanks to the initial spade-work of the late EDWARD BUNYARD we are now in a position to trace the origin, or rather the evolution of the R. centifolia of Linnaeus, commonly known in England as the 'Provence' or 'Cabbage Rose.' Space will not allow full details here, and a brief summary must suffice.

Analysis shows that R. centifolia L. is not a wild species as many have supposed, nor is it a simple primary hybrid like the Damask Roses. It is a complex hybrid of four distinct wild species, R. rubra, R. Phoenicia, R. moschata and R. canina, and may therefore be presumed to have had a garden origin. Such a combination can, of course, be made in many different ways and we have no knowledge of the actual steps taken. The quickest way to arrive at the combination is no doubt the direct cross between x R. bifera (rubra x moschata) and X R. alba (canina x damascena), which would give the whole combination. Experience of plant breeding, however, leads one to suppose that, working under the usual system of trial and error, many steps were taken, both backwards and forwards, before the object was attained.

Evidence from many varied sources in literature and art show that R. centifolia L. was gradually and slowly evolved from the end of the sixteenth century to the beginning of the eighteenth century, when it apparently reached the perfection of a florist's flower. Definitely we [82] owe this superb Rose to the genius and industry of the Dutch breeders who first took the matter in hand about 1580, and persevered until they attained perfection about 1710.

Soon after that date the old Moss Rose appeared in Holland as a bud-sport of the perfected type of x R. centifolia, which was, of course, completely sterile sexually, owing to the complete doubleness of the flowers. Afterwards at least 62 other bud-sports appeared, either from the original x R. centifolia or from its sports. These were much cultivated in the first half of the nineteenth century, and a few are still grown to this day. At Burbage I grew and studied about twenty of these bud mutations for many years, and during that time several of them reverted to the bud-parent x R. centifolia without any variation, so far as I could see, which suggests that these sports came fro the stock of one seedling plant of x R. centifolia, existent about 1710.

The true type of x R. centifolia, i.e. the perfected form which produced the Moss Rose, is tetraploid with 28 chromosomes in the body-cells and 14 in the male and female germ-cells. Owing to doubleness the germ-cells do not mature and fail to reach the final stage. Only once did I succeed in getting a seed, and that failed to germinate.

Most of the bud-sports have the same number of chromosomes as their parent, but a few, like the Rose of the Painters, are triploid with 21 chromosomes in the body-cells and irregular numbers in the germ-cells which fail to mature.

* Syn. R. gallica L., 1759 (non 1753).

From the above notes it will be seen that the most important of our ancient garden Roses have originated from four wild species of Rosa, namely the Red Rose [R. rubra Blackw. 1757],* the Phoenician Rose (R. phoenicia Boiss.), the Musk Rose (R. moschata Miller) and the Dog Rose (R. canina L.). This completes a brief survey of the ancient foundation Roses from which our garden Roses, ancient and modern, have sprung. There are a few other ancient garden forms still in cultivation which have not been noted, such as the Double Cinnamon Rose, the Double Sweet Briar, the Double Sulphur Rose, and the Miniature forms, such as the Burgundy Rose, the Rose de Meaux and others of that kind; these have not been noted because up to now they have not been directly concerned in the evolution of our modern Roses. They have had their day but so far have not succeeded in establishing a dynasty. Included in this category are the less ancient forms of the eighteenth century, such as the Ayrshire Roses from R. arvensis Huds., the Evergreen Roses from R. sempervirens L., and the Burnet or Scots Roses from R. spinosissima L. 1762 (non 1753). Finally, there are the important Yellow and Copper Austrian Briars from R. lutea [Blackw.] which, though very ancient, did not come into the direct line of modern Roses until the dawn of the twentieth century. These will be noted in the next article on the origin and evolution of our modern Roses since A.D. 1800.

[242]

II. MODERN GARDEN ROSES (1800-1940)

GREAT events oft from little causes spring and it is a remarkable fact that the most important improvement of our garden Roses in the last century was due to the introduction from China of a minute Mendelian gene, so small that it is quite invisible to the naked eye. The chromosome thread on which this tiny Rose gene is borne is plainly seen under a high-powered lens of an ordinary microscope, but an electron microscope would be required to make out the outline of the gene itself.

A pair of these genes, one from the father and one from the mother, is present in every growing cell of our best modern Roses whether they be exquisite Teas, shapely Hybrid Teas, gay-coloured Pernets, dazzling Poulsens, or lowly Poly-Poms. In normal conditions of growth these genes cause a continuous pushing of shoots, with every shoot a flowering shoot, so that our modern Roses tend to flower all the year round, unlike the ancient roses which for the most part flowered naturally only once a year.

In Rose-breeding, except for certain complications met with in annual autumn-flowering species, this gene for perennial-flowering behaves as a simple Mendelian recessive to the dominant wild gene for annual flowering. It is evidently a mutation which still appears from time to time in a state of nature among the wild Roses of China just as it did a thousand or more years ago when the Chinese gardeners first discovered it. This mutation seems to be peculiar to China and to the genus Rosa although it is not limited to one species since I have found it in several Chinese species of Rosa.

China is a country rich in Roses, both wild and cultivated, indeed with its wealth of species it is considered by some botanists to be the original home and cradle of the genus. It was not, however, from these magnificent wild species that our modern garden Roses have sprung. That is reserved for future centuries. It was the humble dwarf China Rose that had been cultivated in Chinese gardens for more than a thousand years and bore the precious gene for continuous flowering which, transferred by hybridization to our Western garden Roses early in the nineteenth century, changed them as by a magic wand.

E. CHINA ROSES
(Sect. Indicae Thory).

Since the introduction of the Poly-Poms in 1875 China Roses have become rather old-fashioned but they are still grown and cultivated by those who appreciate the loveliness of translucent colours.

The earliest records of the China Rose are found in the Chinese screen paintings of the tenth century on which are portrayed Blush [243] China Roses which appear to be identical with the Blush Tea-scented China introduced to England in 1809. The earliest trace of the introduction of the China Rose to Europe that I have been able to find is in the National Gallery, London, where there is a painting by the Florentine artist ANGELO BRONZINO, dating from about 1529, which shows a smiling Cupid with his hands full of Pink Chinas Roses in the act of throwing them over Folly who is embracing Venus (BRONZINO No. 651). The small rose-pink flowers with translucent petals, incurved stamens, reflexed sepals and small firm ovate shining leaflet are precisely those of the Pink China and we may safely conclude that this Rose was cultivated in Italy early in the sixteenth century. It may have been this China Rose which Montaigne saw in flower at the Jesuit Monastery at Ferrara when he visited Italy in November 1578 and was told that the Rose flowered all the year round.

 * By some mischance the date of this specimen is given as 1704 by both WILLMOTT (1911) and BUNYARD (1936), and the latter refers to it as the Blush China although it is definitely a Crimson China.

Among the early botanical specimens in the British Museum there is a quaint little remnant of a Crimson China Rose neatly arranged in a paper vase in the Herbarium of GRONOVIUS (1690-1760). It is labelled "Chineesche Eglantier Roosen" and dated 1733.* It was on this specimen of GRONOVIUS that JACQUIN founded his figure of the Crimson China in 1768 under the name of R. chinensis. In 1750 PETER OSBECK, a pupil of LINNAEUS and a collector of specimens for the Linnean Herbarium, set sail for China and the East Indies. On his return to Sweden in 1752 he published a brief account of his travels in which he describes his discovery of Rosa indica in the gardens of the Custom House at Canton on October 29, 1751.

THE CHINA ROSES OF LINNAEUS.

The China Roses in the Herbarium of Linnaeus at the Linnean Society of London, include three Crimson Chinas, one Pink China, one Blush Tea China and one hybrid China which is apparently a cross with R. multiflora Thunb. The most interesting and important specimen is the Blush Tea China (Sheet 38) which LINNAEUS, in his own handwriting, indicates as his type specimen of R. indica. It is probable that this is the R. indica found by Peter Osbeck in 1751 in the Custom House garden at Canton. The Pink China specimen (Sheet 37), according to Linnaeus, came from the Botanic Garden at Upsala whither it had been brought from China, probably by Osbeck in 1752. It is the R. sinica of Murray in Linn. Syst. Veg. 1774 which according to the elder AITON (1789) was cultivated by PHILIP MILLER in 1759. This is an early date for the cultivation of the China Rose in England, but it seems hardly likely that AITON was mistaken because in the early part of that year he was a pupil of MILLER at Chelsea and thus had an opportunity of knowing the plant at first hand. MILLER was in close touch with LINNAEUS at that time and may well have received the Pink China direct from [244] Upsala. In 1769 Sir John Hill, in his Hortus Kewensis, records the cultivation of R. indica L. in the gardens of the Princess AUGUSTA at Kew House where AITON was in charge.

In the younger MARTYN'S French edition of MILLER'S Dictionary of 1785, it is stated that a Deep Red China Rose was cultivated in England at that time but I have been unable to trace its origin. French authors agree that the China Roses came to France through England but they give a bewildering array of dates of their introduction from China and India to England, ranging between 1710 and 1780, and so far I have been unable to confirm these from English sources. In 1781 the Pink China was introduced to Holland by the Dutch East India Company and in her Memoirs the Baroness D'OBERKIRCH relates that she saw this Chinese Rose in the gardens at Haarlem in 1782 and at once recognized the flower which was so often delineated on Chinese screens and fans. None of these early introductions of the China Rose to Italy, Sweden, Holland and England, however, seems to have played any part in the development of our modern Roses. That rôle was apparently reserved for four special English introductions of 1792, 1793, 1809 and 1824, each of which had a definite and permanent influence on the evolution of our garden Roses.

THE FOUR STUD CHINAS.

(1) SLATER'S CRIMSON CHINA, 1792.
(R. chinensis Jacq.)

*CR: Slater never had this rose, according to his gardener, James Main (1835).

The first of the four stud Chinas was SLATER'S Crimson China which was imported from China by GILBERT SLATER* of Knot's Green, Leytonstone, about 1792. The original plant was figured by CURTIS in Bot. Mag. in 1794 under the name of R. semperflorens and there is a specimen of it in the British Museum from Kew Gardens. SLATER'S Crimson China must have been introduced to France soon afterwards, for in 1798 CELS, the Paris nurseryman, THORY, the French botanist, and REDOUTÉ, the famous artist, commenced their breeding experiments with it in or about that year.

SLATER'S Crimson China reached Austria, Germany and Italy before the close of the eighteenth century and early in the nineteenth century in Italy it became the parent of the Portland Rose and thus grandparent of the first Hybrid Perpetual 'Rose du Roi.' SLATER'S Crimson China is still in cultivation in old gardens and there is an admirable modern figure of it in Willmott (1911) under the name of R. chinensis semperflorens. In its characters it is very near to HENRY'S Crimson China, collected in 1885 in the San-yu-tung Glen near Ichang in the Province of Hupeh, Central China, which is generally considered to be the wild species and original ancestor of the China Roses.

SLATER'S Crimson China differs from the wild species in its dwarf habit, semi-double flowers and perennial flowering, all of which are Mendelian characters, the genes for the first and last characters being closely linked in the same chromosome. [245]

Like most of the cultivated Crimson Chinas, SLATER'S form is a triploid with 21 chromosomes in the body-cells, 14 in the female germ-cells and 7 in the male. Consequently the pollen is very defective and on the average only one grain in seven is fertile. This 14 percent. fertility would no doubt be fatal to survival in a state of nature, but it is sufficient for the gardener and hybridist to raise new kinds of Roses. Some of the cultivated Crimson Chinas are, however, diploid with 14 chromosomes in the body-cells and 7 in the male and female germ-cells, having retained their wild simplicity. I found one of these diploid Crimson Chinas in the GRAVEREAUX collection at La Roseraie de l'Hay near Paris. In spite of its air of culture it was a typical wild Crimson China with single cherry-red flowers, while in habit it was a graceful short climber. Its diploid chromosome behaviour was entirely regular and normal as in a species, with no signs of hybridity, consequently the pollen and embryo-sacs were regular and the plant was as fully fertile as a wild species.

(2) PARSONS' PINK CHINA, 1793.
(R. chinensis Jacq. x R. gigantea Collett.)

*CR: I think he meant Dr. CARTIER.

PARSONS' Pink China was first seen in his garden at Rickmansworth in 1793, according to ANDREWS (1805), and was said by the younger AITON (1811) to have been introduced from China about 1789 by Sir JOSEPH BANKS. In the Banksian Herbarium there are three specimens of the Pink China, two of which closely resemble PARSONS' Pink China, one of these is marked "China prope Canton, Lord Macartney" and the other "Hort. Kew 1795 China." LINDLEY, in 1820, based his R. indica on the MACARTNEY specimen, crediting the collection of it to Sir GEORGE STAUNTON who accompanied Lord Macartney's embassy to China in 1792. It is therefore possible that PARSONS' Pink China was sent home by him to Sir JOSEPH BANKS, who was at that time Director of Kew. Soon after 1793 COLVILLE secured a stock of PARSONS' Pink China and sent it out as the Pale China Rose, presumably to distinguish it from SLATER'S Crimson China of 1792. PARSONS' Pink China soon arrived in France, for it was seen in the greenhouses of Dr. BARBIER* in Paris in 1798, and THORY tells us that he and REDOUTÉ started to raise seedlings from it in that year.

*CR: It seems just as likely that John Fraser was responsible. He introduced at least two pink Musk roses to England that were suspiciously similar to the Noisettes sent to France. Fraser also spent time in South Carolina, and probably would have known Champneys.

About 1800 PARSONS' Pink China appeared in North America, at Charleston in South Carolina, no doubt* through the agency of the two brothers LOUIS and PHILIPPE NOISETTE who were nurserymen in Paris and Charleston respectively, and about 1802, in the hands of JOHN CHAMPNEYS of Charleston, PARSONS' Pink China became a grandparent of the French Noisette Rose from which later on came our best Climbing Yellow Teas such as 'Maréchal Niel' and 'Gloire de Dijon.'

In 1805, at COLVILLE'S nursery in England, PARSONS' Pink China gave rise to the Dwarf Pink China, a miniature Rose known in England as the Fairy Rose or R. Lawranceana. LOUIS NOISETTE imported this Rose to France and called it Bengale Pompon. It was largely planted [246] at Lyon and in 1868 became a grandparent of the first Poly-Poms and ancestor of the Poulsen Roses.

About 1810 PARSONS' Pink China appeared in the French Island of Bourbon (Reunion) where the colonists used it to form hedges, and towards 1815, by natural crossing with the Autumn Damask Rose, also used as a hedge plant, it became a grandparent of the French Bourbon Rose and thus an ancestor of the Teas, Hybrid Perpetuals and Hybrid Teas.

* Genetical confirmation of this is found in the appearance of true-breeding gigantea characters in the original crosses of Noisette and Bourbon with the Blush and Yellow Chinas, giving rise to the race of Tea Roses (R. gigantea Collett).

PARSONS' Pink China is still in cultivation in country gardens and there is an excellent figure of it in REDOUTÉ (1817) under the name of R. indica vulgaris. Analyses of its characters show the influence of the Wild Crimson China (R. chinensis) in 16 of these, while the remaining 12 characters show the influence of another species, the Wild Tea Rose (R. gigantea). PARSONS' Pink China may therefore be regarded as a hybrid between chinensis and gigantea. It is not, of course, an ordinary primary hybrid produced directly* between the two species but rather a derivative hybrid derived after generations of crossings in Chinese gardens. Parsons' Pink China is a diploid with 14 chromosomes in the body-cells and 7 in both the male and female germ-cells. Although a diploid, its chromosomes are not regular in their behaviour and weak pairings in the germ-cell divisions lead to defective pollen and embryo-sacs and consequent sterility. In this respect the Pink Chinas behave as hybrids rather than pure species. Among the newer varieties of the Pink China I have found several triploid forms with 21 chromosomes which have no doubt arisen by duplication of the chromosomes in a pollen or egg-cell, as in the case of the triploid Crimson Chinas.

(3) HUME'S BLUSH TEA-SCENTED CHINA, 1809.

It was on this assumption that REHDER in 1915 reduced R. gigantea Collett to a variety of R. odorata Sweet, a name given by SWEET in 1818 to HUME'S Blush China.

HUME'S Blush China is said to have been imported in 1809 by Sir A. HUME, Bart., from the East Indies, at that time a comprehensive term which included China. It was figured by Andrews in 1810 under the name of R. indica odorata from a plant flowering in COLVILLE'S nursery. In the same year special arrangements were made by both the British and French Admiralties for the safe transit of plants of this new Tea-scented China to the Empress JOSEPHINE at Malmaison in spite of the fierce war that was raging between England and France at that time. In 1817 REDOUTÉ published a beautiful and accurate figure of the original plant under the name of R. indica fragrans which shows it to be a China considerably modified by the influence of the Wild Tea Rose (R. gigantea), the bias in favour of that species being about 2:1. It was not, however, a true Tea Rose as some have supposed† since it shows the influence of the Wild Crimson China (R. [247] chinensis) in 11 of its characters out of the 31 examined, the remaining 20 characters showing the influence of the Wild Tea Rose (R. gigantea). HUME'S Blush China must therefore be regarded as a derivative hybrid from gigantea and chinensis. Crossed with the Bourbon, Noisette and Yellow China towards 1830, HUME'S Blush China gave rise to the typical Tea Roses. Living material of this Rose is no longer available, but like its ancestral species it was most likely a diploid with 14 chromosomes. Material of its ancestral species the Wild Tea Rose (R. gigantea) from both Cambridge and Burbage proved to be diploid.

(4) PARKS' YELLOW TEA-SCENTED CHINA, 1824.
(R. chinensis Jacq. x R. gigantea Collett.)

*CR: By the 1830s, Parks' Yellow China had become confused with the 'Yellow Tea-scented', which was a self-seedling of the 'Blueh Tea-scented'.

The fourth and last of the Stud Chinas is PARKS' Yellow China* which was brought from China by PARKS for the R.H.S. in 1824. The following year it was imported to France by the enthusiastic Rose breeder HARDY of the Royal Luxembourg Gardens, where its novel colour made it a general favourite. There is a good figure of the original in the 1835 edition of REDOUTÉ under the name of R. indica sulphurea with a description by PIROLLE. At first sight, with its large yellow flowers, thick tea-scented petals and bright green leaves, PARKS' Yellow China looks more like a Tea than a China and reminds one rather of the yellow variety of R. gigantea discovered in Manipur by Sir George Watt in 1882. An analysis of its characters, however, shows the influence of 10 China characters to 20 of the Wild Tea Rose so that the plant must be regarded as a hybrid. PARKS' Yellow China was the ancestor of many remarkable Roses, crossed with the Noisettes it produced the typical Yellow Teas which, crossed with the Pink Teas derived from HUME'S Blush China and the Bourbon, gave rise to those exquisite shades of refined colouring peculiar to Teas in which the pink and yellow are indescribably mixed and blended. Through the Teas, PARKS' Yellow China was the ancestor of many of the Hybrid Teas, Pernets, Poly-Poms. and Poulsen Roses. No living material of PARKS' Yellow China has been available since1882, but there is little doubt that it was a diploid with 14 chromosomes since both ancestral species were diploid and crossed with the diploid Noisettes its descendants proved to be diploid.

In concluding the account of the four Stud Chinas which have had such a powerful influence in moulding our modern garden Roses, it may be interesting to note that China Roses similar to the four Stud Chinas are still cultivated in Chinese gardens. In South-Western China, in the Province of Yunnan some of them are largely used for hedges. The Wild Tea Rose (R. gigantea) also grows there naturally in ravines and on grassy hills, and in the same locality at the same elevation garden forms of the Crimson China (R. chinensis) are still cultivated there. The Port of Yunnan is Canton, from which the four Stud Chinas are presumed to have come. It seems therefore fairly safe to conclude that the four Stud Chinas which revolutionized our old garden Roses [248] originated in South-Western China in the Province of Yunnan and probably not far from Mengtsze.

F. THE NOISETTE ROSE.
MILLER'S WHITE MUSK x PARSONS' PINK CHINA.
R. moschata Miller X (R. chinensis Jacq. x R. gigantea Collett).

The honour of the first introduction of the Chinese gene for perennial flowering to our western Roses belongs to an American citizen, JOHN CHAMPNEYS, a wealthy rice planter of Charleston in South Carolina, who was a great gardener. About 1802 he fertilized the old White Musk Rose of MILLER with pollen of the new Pink China of PARSONS which he had recently received from France through the brothers Noisette of Charleston and Paris. The result was a handsome hybrid which combined the climbing habit, large open clusters of flowers and odour of the Musk with semi-double pink flowers of China and the handsome foliage of both. CHAMPNEYS called it R. moschata hybrida, but the Rose became such a great favourite in America that it came to be known as Champneys' Pink Cluster or the Champneys Rose. A few years afterwards, Philippe Noisette, the French nurseryman at Charleston, sowed seeds of CHAMPNEYS' hybrid and thus in the second generation raised the original French Noisette Rose and thus in the second generation raised the original French Noisette Rose which he sent to his brother, LOUIS NOISETTE of Paris, in 1814 who distributed it in Europe in 1819. This Rose was figured by REDOUTÉ in 1821 under the name of R. Noisettiana and quickly became a popular garden Rose. In the second generation Mendelian segregation took place and the Noisette was less tall and more compact in growth than CHAMPNEYS' Hybrid, the flowers were blush-white borne in large dense clusters equally fragrant, but above all they were produced continuously from June to the winter frosts.

Up to 1830 there was little variation in the Noisette race bred from seed except those with violet-crimson flowers which were hybridized with multiflora and chinensis, and those with lax flower clusters which were hybridized with R. sempervirens L. Among the latter was the old favourite 'Aimée Vibert' raised by VIBERT in 1828 and still prominent in old gardens.

Towards 1830 attempts were made to create a Yellow Noisette by crossing the Blush Noisette with PARKS' new Yellow China. This cross had far-reaching results since it gave rise not only to the Yellow Noisette but also to the Yellow Tea Roses. In 1830 'Lamarque' and 'Jaune Desprez' appeared and in 1833 SMITH'S Yellow, the last being a true dwarf Yellow Tea although classed as a Noisette on account of its origin. The first two were yellowish Noisettes which, selfed, gave Yellow Climbing Teas, 'Celine Forestier' in 1842 and 'Cloth of Gold' (Chromatella) and 'Solfatare' in 1843. 'Cloth of Gold' was the parent of the famous 'Maréchal Niel.' Further breeding swamped the old Noisette characters and transformed the Noisette into a Climbing Tea as seen in 'Gloire de Dijon' (1853), 'Marechal Niel' (1864), [249] 'Rêve d'Or' (1869), 'Bouquet d'Or' (1872), 'Caroline Kuester' (1872), 'William Allen Richardson' (1878) and 'Mme. Alfred Carrièr' (1879)

The original Blush Noisette proves to be a diploid with 14-chromosomes and all its early descendants including the Yellow Teas prove to be diploid except 'Gloire de Dijon' which is a tetraploid with 28 chromosomes. CHAMPNEYS' hybrid must therefore have been a diploid as its parents were. MILLER'S Musk Rose is now very rare, but it is still in the Cambridge Botanic Garden where it was examined.

G. THE BOURBON ROSE.
PINK AUTUMN DAMASK x PARSONS' PINK CHINA.
(R. rubra Blackw. x R. moschata Miller) x (R. chinensis Jacq. x R. gigantea Collett).

The Bourbon Rose, like the Noisette and all other new races of Roses, appeared in the second generation of a cross in accordance with Mendel's Laws of Segregation and Fixity. The first generation of the Bourbon was a natural hybrid between the Pink Autumn Damask and PARSONS' Pink China, found growing in a garden with its parents in the French Island of Bourbon in 1817 by the Parisian botanist BRÉON who was in charge of the Botanic Garden at the time. It was a vigorous and decorative hybrid intermediate in character between the Damask and China parents. Locally and in the adjacent island of Mauritius it was known as the 'Rose Edward.' In 1819 BRÉON sent seeds of the new hybrid to his friend JACQUES seeds Jacques raised the first French Bourbon Rose and called it 'Rosier de l'Ile Bourbon.' The new Rose was distributed in France about 1823 and in England about 1825. It was figured by REDOUTÉ in 1824 under the curious name of R. canina Burboniana. This Rose of the second generation was a better garden Rose than its parent the 'Rose Edward' partly because of its more compact habit but mainly by its abundant autumn flowering. Mendelian segregation had given it a double dose of the China gene for continuous flowering and it was a beautiful semi-double Rose with brilliant rose-coloured flowers and nearly evergreen foliage. From its Damask grandparent it inherited a delicious fragrance which was particularly marked in the late autumn months. During its reign of about half a century from 1820 to 1870 the Bourbon Rose contributed much to the development of our garden Roses. Crossed with HUME'S Blush China it helped to create our Pink Tea Roses and it was the main source of the typical Hybrid Perpetuals which, after all, were only the Bourbon 'writ large.' In 1825 the remarkable Bourbon 'Gloire des Rosomanes' appeared with its perennial scarlet-crimson flowers and vigorous habit of growth, and became the chief ancestor of the scarlet and crimson Hybrid [250] Perpetuals with their rich damask fragrance. Although from the first the Bourbon was a distinct type of Rose with its stout prickly stems, vivid rose-coloured flowers with rounded imbricated petals and broad leathery leaves, various breaks occurred from time to time through segregation as well as through hybridization. Between 1834 and 1841 the China reversion 'Hermosa' appeared independently with four different breeders and it is unlikely that all these were due to a China back-cross. In 1831 the Bourbon-Noisette 'Mme. Desprez' appeared adding purple to the rose colour which, in 1839, became a vivid crimson in 'Dr. Rocques' (Crimson Globe). 'Mme. Desprez' added to her laurels in 1842 by producing that wonderful Bourbon Rose 'Souvenir de la Malmaison' which in its turn produced the popular 'Gloire de Dijon.' In both cases the other parent was a Tea Rose and as in the case of the Noisette, repeated crossings with the Tea Roses towards the middle of the nineteenth century completely changed the character of the original Bourbon and improved it out of existence. Both the original Bourbon and 'Souvenir de la Malmaison' are triploids with 21 chromosomes, while 'Gloire de Dijon' is a tetraploid with 28. [282]

H. THE TEA ROSE (R. gigantea Collett).
HUME'S BLUSH AND PARKS' YELLOW CHINAS X BOURBONS AND NOISETTES.

The first typical Pink Tea Rose, appropriately enough, was called 'Adam' and was raised by a florist of that name at Rheims in 1833. Its characters are intermediate between those of HUME'S Blush China and the original Bourbon, and it is no doubt a hybrid between them.

The first typical Yellow Tea Rose was raised in England in the same year and was called SMITH'S Yellow; this was a cross made between the Blush Noisette and PARKS' Yellow China in an attempt to raise a Yellow Noisette and was duly sent out as such. Later, FOSTER of Devonport back-crossed the Yellow China with SMITH'S Yellow and raised in 1838 the beautiful 'Devoniensis' with its thick creamy-white petals, pale straw-pink centre and sweet fragrance. In 1839 appeared the remarkable Tea Rose 'Safrano,' with the outside petals bright rose and the inner petals butter-yellow; it appears to have been raised from PARKS' Yellow China crossed with Noisette 'Desprèz.' 'Safrano' became the head of a special line of yellow and copper Teas including 'Mme. Falcot' (1858), 'Perle des Jardins' (1874) and many others. Various intercrossings between the yellow Noisette-Teas and the pink Bourbon-Teas gave rise between 1840 and 1890 to the most exquisite Roses ever produced, but after that date they gradually lost their typical qualities by crossings with the new Hybrid Teas which they had helped to make, and to-day the typical Tea Rose has disappeared from general cultivation. All the typical Teas examined are diploid with 14 chromosomes, except 'Gloire de Dijon' which is tetraploid. The later Tea 'Lady Hillingdon' (1910) is a triploid with 21 chromosomes and is not a true Tea. It is to be hoped that the old Tea will be revived if only for crossing with the new Pernet and Poulsen Roses. We cannot afford to lose its valuable breeding qualities of colour, form and fragrance.

J. HYBRID CHINAS.
CHINAS, NOISETTES AND BOURBONS X FRENCH, PROVENCE, DAMASK AND OTHER SUMMER ROSES.

Early in the nineteenth century the term Hybrid China was used in a special sense to cover all hybrids between the perennial and the summer-flowering groups. At that time nearly all the Roses grown in gardens were summer-flowering, and of these the so-called French Rose (R. rubra Blackw.*) held pride of place, with more [283] than a thousand varieties in all the colours of the period. When the ever-blooming China came and was considered sufficiently hardy to plant outside about 1810, it was only natural that hybrids began to appear, and in 1815 the first two Hybrid Chinas came to light. The first was raised in England by Brown of Slough from Hume's Blush China fertilized by a French Rose, and was known as Brown's 'Superb Blush'. The second was one of Descemet's 10,000 seedlings rescued by Vibert, when the allied armies marched on Paris, and was known as 'Zulmé' or 'Bengale Descemet'. Up to 1830 about forty varieties of Hybrid Chinas appeared in France and two in England. Towards 1830 four superior varieties were produced which are particularly interesting to us, since they became the actual parents of the large Hybrid Perpetuals. One of these was 'Malton', a China-French hybrid, while the other three were 'Brennus', 'Athalin', and 'Général Allard', which were Bourbon-French hybrids. In flowers and foliage these varieties combined almost all that was beautiful in Roses but, like all the Hybrid Chinas without exception, they failed in one important character. They flowered only once a year owing to the Mendelian dominance of the gene for summer-flowering over the recessive gene for continuous flowering. These four Hybrid Chinas were all tetraploid with twenty-eight chromosomes, while the original Hybrid China was triploid with twenty-one.

K. HYBRID PERPETUALS
Hybrid Chinas X Portlands, Bourbons, and Noisettes

The first Hybrid Perpetual was the famous 'Rose du Roi', raised in 1816 by Souchet in the garden of the king at Sèvres, St Cloud, Paris, from the original Portland Rose whose history is chequered and somewhat obscured. It is known to have been in Dupont's nursery in Paris in 1809, and that Dupont obtained it from England and named it after the Duchess of Portland, who probably found it or obtained it from Italy early in the century. In England it was known as the Rosa Paestana or 'Scarlet Four Seasons', and was said to have been brought from Italy from the neighbourhood of the classic Paestum. The Portland Rose was a bright red verging on scarlet, and if treated well and pruned in a certain way it flowered twice a year, in summer and autumn. It was generally regarded as a cross between the French Rose and the Autumn Damask. Judging from Redouté's accurate figure of 1817 it is evidently a China-Damask-French hybrid which may well have originated in Italy, where, owing to the favourable climate, the China Roses had been largely cultivated in the open ever since their introduction about 1798. From the colour and dwarf habit of the Portland it may be presumed that the China parent concerned was Slater's Crimson China.

The 'Rose du Roi', with its charming crimson flowers, very double and very fragrant, and above all its continuous flowering throughout the season without special treatment and pruning, was certainly a great advance on its Portland parent, and it reigned supreme until the coming of the larger Hybrid Perpetuals in 1837, which displaced the smaller Portland and Bourbon Hybrid Perpetuals.

The fortunate raiser of these large-flowered Hybrid Perpetuals was Laffay, the well-known Rose breeder of Auteuil, who introduced the first typical Hybrid Perpetual 'Princesse Hélène' in 1837; he followed this up with 'Mme Laffay' and 'Duchess of Sutherland' in 1839, and finally crowned his successes in 1842 with that famous 'Rose de la Reine', with its large, strangely cupped flowers of a beautiful lilac-rose and fragrant as a Cabbage Rose. From 1837 to 1843 Laffay produced eighteen Hybrid Perpetuals of merit, and it is understood that all these were raised from seeds of the Hybrid Chinas (mostly Bourbon-French) selfed, and crossed with Portlands and Bourbons. In this way pairs of the recessive genes for continuous flowering were brought together and the Hybrid Perpetuals created. This also explains why many of the earlier and some of the later Hybrid Perpetuals sent out were not truly perpetual. From 1840 other breeders joined in, and in 1851 a new type appeared in 'Victor Verdier' fro a Bourbon-Tea cross. In 1852 'Jules Margottin' and the famous 'Général Jacqueminot'' arrived, the former a Brennus perpetual (Hybrid China selfed) and the latter a descendant of the Bourbon 'Gloire des Rosomanes'. 'The General', with its brilliant scarlet-crimson flowers and damask fragrance was a fertile tetraploid, and left a host of descendants of a similar type. From 1840 to 1890 the Hybrid Perpetual completely dominated the outside Rose garden until it was gradually replaced by the Hybrid Tea, which it helped to make. All the Hybrid Perpetuals examined are regular tetraploids with twenty-eight chromosomes, show traces of hybridity in their weak pairings, and are sterile to some degree.

L. HYBRID TEAS
Hybrid Perpetuals X Teas

The first Hybrid Tea was the favourite Rose 'La France', which was raised by Guillot in 1867 out of the Hybrid Perpetual 'Mme Victor Verdier' by the Tea 'Mme Bravy'. It was a worthy representative of its class with its silvery lilac-pink flowers of excellent shape and delicious fragrance. It combined the free-flowering habit, fine shape, and delicate colouring of the Tea with the hardiness and vigorous growth of the Hybrid Perpetual. In 1873 'Cheshunt Hybrid' and 'Captain Christy' appeared, but it was not until 1884 that the Hybrid Tea was recognized as a new group distinct from the Hybrid Perpetuals. Up to 1895, when 'Mme Abel Chatenay' and 'Mrs W. J. Grant' appeared, Hybrid Teas were mostly bred on the original lines, but afterwards they were not only bred inter se, but also back-crossed to the Hybrid Perpetual, giving 'Liberty', ...

[284]

...

In 1905 the Hybrid Teas began to change their original characters [285] owing to their hybridization with the new Pernet Rose which appeared in 1900. At first the influence of the Pernet Roses on the Hybrid Teas was slight, but gradually year by year the Pernet characters permeated the Hybrid Teas in geometrical progression, so that at the present time it is rare to see an original Hybrid Tea among the new Roses without a trace of the Pernet influence. These Roses are still called Hybrid Teas but they have lost their original characters. The cause of this revolution is that for the first time in a century a new species, R. lutea Miller, has been incorporated with our garden Roses.

All the Hybrid Teas examined prove to be tetraploid with 28 chromosomes.

M. THE PERNET ROSE.*
HYBRID PERPETUAL x AUSTRIAN BRIAR (R. lutea Miller).†

* The rules of nomenclature prevent the use of 'Pernetiana.'
Since there is a doubt about HERRMANN'S older name R. foetida of 1762
I am using the next oldest name, which is MILLER'S R. lutea of 1768.

The twentieth century, like the nineteenth, opened with the introduction of a new species to the garden Roses of the period. Last century it was R. chinensis from China with its habit of continuous flowering, this century it is R. lutea from Persia with its brilliant yellow flowers. Both species have revolutionized our garden Roses as no other species have done during the two centuries. R. lutea Miller is an ancient cultivated species grown by the Saracens in Syria, Tripoli, Tunis and Egypt in the twelfth century and in the same period cultivated by the Arabs and Moors in Spain, who had a tradition that it was brought from Persia in the seventh century. It is no doubt an oriental species and has been reported in awild state from the Crimea through Asia Minor and Persia to Turkestan, and even in the Punjab, Afghanistan and Tibet. Some of these, however, may be garden escapes, since it has been found in several places in Europe growing in Europe growing apparently wild. It was figured by GESNER in 1542, LOBEL in 1581, DODONAEUS in 1583, and exactly described by DALECHAMPS in 1587. CLUSIUS found it in Austria towards 1583 and introduced it to the Netherlands and England, where it has been known ever since as the Austrian Briar. There were two forms of this, the Austrian Yellow and the Austrian Copper, both well-known garden plants. The Yellow is the ordinary R. lutea, while the Copper is a bicolor form with the petals of two distinct colours, brilliant nasturtium-red on the upper inner surface and shaded yellow on the lower outer surface, the general effect being a coppery-red. CORNUTI in 1635 and MILLER in 1768 gave this form specific rank under the name of R. punicea, and JACQUIN in 1770 as R. bicolor. We now know that the Copper is simply a bud-sport of the Yellow and that often the two colours may be found on the same bush. This explains why so many of our dazzling Pernet Roses, bred from the yellow, present this fantastic bicolor effect. The Pernet Roses originally came from the Persian Yellow Rose, which appears to be only a double form of R. lutea. It was brought from Persia by Sir HENRY WILLOCK in 1838. [286]

After a thousand years of cultivation in gardens R. lutea had become notoriously sterile in both seeds and pollen and it was no easy matter to incorporate the species with the partially sterile garden Roses. All the more credit is therefore due to the persistent and untiring efforts of that great Rose breeder PERNET-DUCHER of Lyon which were ultimately crowned with success. From 1883 to 1888 he patiently and persistently crossed thousands of Hybrid Perpetuals with pollen of Persian Yellow, following a fixed idea of his own. All these crosses failed except one with 'Antoine Ducher' which gave him a few seeds in 1888. 'Antoine Ducher,' the grandmother of the Pernet Rose, was just an ordinary Hybrid Perpetual raised by DUCHER in 1866 with large, full, shining rose-red flowers and a strong old rose fragrance.

Two hybrids were reared and the first flowered in 1891. This was interesting in its preponderance of the paternal lutea characters, but being completely sterile it took no part in the creation of the Pernet race and was sent out as a Pillar Rose under the name of 'Rhodophile Gravereaux.' The other plant first flowered in 1893, but it was not until 1894 that it showed itself in its true colours. On the whole it was more like the mother Hybrid Perpetual in its habit of growth, large globular full flowers shaded with rose and in its delicious old rose fragrance. On the other hand, it resembled the father lutea in its red-brown wood, prickles, rounded leaflets, solitary flowers with bicolor petals which were orange-red within and golden-yellow without, and in its summer-flowering. This unnamed seedling was retained by PERNET as a breeder, and when mated with Hybrid Teas produced in the second generation the first Pernet Rose 'Soleil d'Or' and many which followed it. As in the cases of the Noisette and the Bourbon the second generation was necessary to fix the continuous flowering in the race, in accordance with Mendel's Laws. 'Soleil d'Or' was first exhibited at Lyon in 1898 and was sent out in 1900 as a forerunner of the Pernet Roses of the twentieth century. After 40 years' breeding the Pernet Roses have become thoroughly incorporated with the Hybrid Teas and have revolutionized their colours and colour patterns. One of the chief contributions made by the Pernet Rose is the production of deep yellow Roses, much deeper and richer but perhaps not so refined as the paler yellow of the Tea Rose. That was PERNET-DUCHER'S original aim in bringing in lutea and his object has been achieved. No one, however, could have foreseen the myriads of new shades, tones and patterns of colours which have been created in our garden Roses by the simple introduction of a yellow species. The peculiar glistening foliage and the numerous exaggerated prickles of the Pernet Roses are also rather unexpected in the product, constituting as they do new types of foliage and prickles which in a wild species we should probably regard as specific. Early in the century there was another side to the picture. Side by side with the good qualities of lutea introduced by the Pernets there were naturally some bad ones, such as bad constitution and winter die-back, liability to Black Spot and consequent early loss of leaves, flat-topped and quartered flowers, lack of fragrance or presence of disagreeable odours, and complete sterility of seeds. [287]

N. THE POLY-POMPON ROSE*
Japanese Multiflora x Dwarf Pink China
R. multiflora Thunb. x (R. chinensis Jacq. x R. gigantea Collett)

*CR: Sisley (1886) wrote that he had received his plant from M. Bonnet.

About 1860 Jean Sisley* of Lyon received from his son in Japan seeds of the wild R. multiflora of Thunberg, a strong climbing Rose with single white flowers, which was quite different from the Chinese Multiflora of gardens, which had pink or crimson double flowers. Guillot, the famous Rose breeder of Lyon, planted some of these Japanese Multifloras in his nursery, and in 1868 saved seeds from them. The result was a medley of forms in the flowers, large and small; single, semi-double, and full; white, rose, and cream; but all were climbers and summer-flowering like the mother parent, and all closely resembled the Noisette Rose in their wood and foliage. One of these had large, tinged white flower with two rows of petals and produced good seeds, which were sown in 1872, and thus produced in the second generation the first two Poly-Poms, 'Paquerette' and 'Mignonette'. Both varieties were dwarf Pompons a few inches high, while their fellows in the same batch of seedlings were tall climbers several feet high like their mother parent. Both were continuous in their flowering right through the season from May to December from their first year, while their tall fellows were summer-flowering in their second year. 'Paquerette' was a pure white, while 'Mignonette' was rosy-pink and white; both combined the characters of multiflora and the Dwarf Pink China, and it is evident that in 1868 the original Japanese Multifloras were naturally fertilized with Dwarf Pink Chinas, which at that time were commonly grown in the Lyon nurseries as specimens.

This Dwarf Pink China was known in the Lyon nurseries as Bengale Pompon, and was raised in England in 1805, where it was known as R. Lawranceana or the Fairy Rose. The first Poly-Pom 'Paquerette' was exhibited at Lyon in 1873 and distributed in 1875. 'Mignonette' was not sent out until 1881, but it became the chief ancestor of most of [288] our modern varieties through its remarkable offspring 'Gloire de Polyantha' (1887), and its line of 'Mme. Norbert Levavasseur' (1903), 'Orleans Rose' (1909), 'Edith Cavell' (1917), and 'Coral Cluster' (1921).

* These varieties are now classed with the Poulsen
Roses owing to their taller habit of growth.

In 1880 'Cecile Brunner' and 1883 'Perle d'Or' introduced the characters of the Tea Rose.* In 1903 the well-known Crimson Rambler became an ancestor of the Poly-Poms., and in 1911 'Dorothy Perkins' entered the list. In 1920 the Pernet Rose was introduced by 'Rayon d'Or' giving the Poly-Pom. 'Evaline.' One of the most popular Poly-Poms. of today is 'Eblouissant' (1918), which is a pure China descended from a dwarf Crimson China and is about the smallest of the Poly-Poms. 'Yvonne Rabier' (1910),* derived from the Japanese species R. Wichuraiana with its pure white flowers and handsome glossy foliage, is still a popular variety. Most of the modern varieties, however, while retaining their dwarf habit and continuous flowering, produce their flowers in tight bunches of rosettes of the Crimson Rambler type.

All the Poly-Poms. examined prove to be diploid with 14 chromosomes. The original 'Paquerette' and its ancestral species R. chinensis, R. gigantea and R. multiflora are all diploid.

The Pernet and lutea Poly-Poms. have not been examined, but they may be expected to be triploids with 21 chromosomes.

O. THE POULSEN ROSE.†
POLY POM. X HYBRID TEA.

The rules of nomenclature prevent the
use of 'Polyantha' or 'Floribunda.'

In 1924 a new and distinct group of garden Roses appeared from Denmark and caused quite a sensation in British and American Rose circles. They were originated by the Danish Rose breeder POULSEN by crossing the Poly-Pom. with a Hybrid Tea. He crossed the Poly-Pom. 'Orleans Rose' with pollen of the Hybrid Tea 'Red Star' and raised therefrom two seedlings of similar type and habit of growth but with different flowers. Both were vigorous growers, producing large open clusters of flowers on long stems which in 'Elsa Poulsen' were amaranth-pink with carmine reverse and semi-double, while those of 'Kirsten Poulsen' were bright cherry-scarlet and single.

by the appearance of the remarkable 'Karen Poulsen' with dazzling scarlet single flowers which defy description. This Rose was raised by POULSEN out of 'Kirsten Poulsen' by pollen of 'Vesuvius.' In 1935 in 'Anne Poulsen' there came a development towards the Hybrid Tea which seems to be a move away from the typical Poulsen rose. Its flowers are a lovely velvety crimson with a delightful fragrance, but the blooms are large and full, produced in small clusters, and the [289] general habit is not so vigorous. A similar type appeared in 1939 in Poulsen's Copper with flowers rosy-cerise on a copper base, large and double in small clusters, but the growth is vigorous. The latest Poulsen Roses to appear are Poulsen's Pink and Poulsen's Yellow, which seem to be a similar type. The future will show if these are to displace the original type, and other breeders are now taking a hand on somewhat different lines. The chromosomes of the Poulsens have not been examined, but most should be triploids with 21.

GENERAL.

Space will not allow me to deal with the evolution and development of various minor groups of garden Roses in which so many different species have been concerned in the nineteenth and twentieth centuries. Ayrshire Roses from R. arvensis Huds., Evergreen roses from R. sempervirens L., Sweet Briars from R. Eglanteria Mill., Scotch Roses from R. spinosissima L., Banksian Roses from R. Banksia Ait., Prairie roses from R. setigera Michx., Musk Roses from R. moschata Miller, Rugosa Roses from R. rugosa Thunb., Macartney Roses from R. bracteata Wendl., Cherokee Roses from R. laevigata Michx., Microphylla Roses from R. microphylla Roxb., Multiflora Climbers from R. multiflora Thunb., Wichuriana Climbers from R. Wichuriana Crep., and the old cultivated Roses of American species which, crossed with the China Rose, gave us the Old Boursault Roses which for more than a century have passed as hybrids of R. alpina L. All these interesting garden Roses must be left for future articles.

In conclusion, we find that the major groups of our garden Roses are descended from seven main species, namely, R. rubra Blackw., R. phoenicea Boiss., R. moschata Miller, R. canina L., R. chinensis Jacq., R. gigantea Collett, and R. lutea Miller. The specific characters of these species can still be traced in our garden Roses, and it is the happy combinations of these various characters that have given us the manifold beauties of the Rose.