Return to METRIA 12

Breeding Efforts in Cercis at North Carolina State University

Dennis J. Werner

Mailing Address:
North Carolina State University
Department of Horticultural Science
Campus Box 7609
Raleigh, NC 27695–7609

A paper from the Proceedings of the 12th Metropolitan Tree Improvement Alliance Conference (METRIA 12), Landscape Plant Symposium: Plant Development And Utilization, held in Asheville, NC, May 23-25, 2002, co-sponsored by the North Carolina State University, North Carolina Division of Forest Resources, USDA Forest Service Southern Region, North Carolina Landscape Association, North Carolina Association of Nurserymen, The Landscape Plant Development Center, The North American Branch of The Maple Society, and The International Ornamental Crabapple Society.

The redbud, Cercis canadensis L. offers numerous possibilities for genetic improvement and the development of new ornamental types. Variation in the species is remarkable, including purple leaves, variegated leaves, glabrous shiny leaves, pubescent leaves, male and female sterile forms, weeping growth habit, double flowered forms, purple fruits, dwarf forms, pink flowers, white flowers, dark–purple flowers, and variation in propagation potential via stem cuttings.

As many of you are aware, the late J.C. Raulston was particularly fond of redbuds, and acquired a remarkable collection of redbud taxa at the J.C. Raulston Arboretum (JCRA). The availability of this collection made it possible for us to 'jump right in' and begin hybridization efforts some years ago. This report will describe the major characteristics of the cultivars of redbud, our breeding experiences with these cultivars, and our general program objectives.

Descriptions of Cultivars

'Alba' – This name is given to perhaps any of the white–flowered forms present in the industry. The cultivar 'Dwarf Alba' found in the Arboretum demonstrates an attractive compact, spreading growth habit. 'Alba' types lack anthocyanin pigment in all plant parts, including newly emerging leaves and immature stems, typically pigmented in redbud, suggesting the lack of flower pigmentation is due to a mutation in the anthocyanin biosynthesis pathway, causing a disruption in anthocyanin synthesis. Some 'Alba' types available in the trade are seed propagated.

'Royal White' – A cultivar offered by Hidden Hollow Nursery in Belvidere, TN. It is described as having somewhat larger flowers than the usual 'Alba', and is more cold hardy.

'Forest Pansy' – An interesting purple–leaf cultivar with deep purple flowers. Although this cultivar is often described as being new to the nursery industry, according to Hubert Conlon, an extension specialist at the University of Tennessee, it originally was discovered in 1947 as a chance seedling at the former Forest Nursery in McMinnville, TN (communicated by Mr. Conlon in a letter to the editor in the January–February 2000 issue of Fine Gardening magazine). It is beautiful in bloom, and stunning as the intense purple foliage emerges in early spring. The intense purple color gradually fades as temperature increases during the growing season, such that in late June to early July the color has faded to a dull bronze in the Piedmont of North Carolina. Mr. Conlan also relates that 'Forest Pansy' is less cold hardy than other redbud cultivars. My observations of 'Forest Pansy' suggest that it is a shy producer of fruit, although this may simply be a function of the lack of adequate cross–pollination on trees I have observed. We have used 'Forest Pansy' in hybridizations with other cultivars, and find that it is a poor pollen producer, thus somewhat challenging to use as a male parent. In our studies, first generation progeny derived from hybridization of 'Forest Pansy' with green–leaf cultivars yield green–leaf offspring, but possessing greater than normal amounts of red pigmentation in the stems and leaf petioles. We will grow these first generation plants to maturity, obtain second–generation offspring from them, and attempt to recover purple–leaf progeny in that generation. Microscopic examination of leaves of 'Forest Pansy' in our lab reveals that the purple pigment is found only in the upper and lower epidermal cells, unlike purple leaf plums, where the purple pigment is distributed throughout all of the leaf cell layers.

'Silver Cloud' – An interesting variegated leaf form with unusual "splotched" leaves, at times almost appearing as if someone had flung green paint onto white leaves. The pattern and extent of variegation seems to be variable from year to year, probably influenced by environment and developmental stage of the tree and any particular leaf. Young leaves as they first emerge often appear to be completely white, acquiring patches of green chlorophyll as the leaf develops and ages. In his original description of 'Silver Cloud', Raulston characterized it as producing few flowers, probably a function of the young age of the tree at the time, but in maturity (20 years old?) the tree at the Arboretum, now gone, flowered abundantly. Flowers are a light purple, much less intensely pigmented than common redbud. According to Raulston, it was introduced in 1964 by Theodore Klein, Yellow–Dell Nursery, Crestwood, KY. The variegated character appears to be simply inherited. In our hybridizations of 'Silver Cloud' with three other cultivars, using

'Silver Cloud' both as a male and a female parent, all first generation offspring demonstrate normal green foliage. Dirr, in his 'Manual of Woody Landscape Plants', refers to this cultivar as a chimera (a genetic mosaic), based on the observation of occasional branches producing only non–variegated leaves. By definition, if 'Silver Cloud' is a chimera as proposed, no sexually derived offspring of 'Silver Cloud' should show variegation. We have grown out second–generation seed from our controlled crosses, and have demonstrated sexual transmission of the variegated condition, confirming that 'Silver Cloud' is not a chimera.

'Silver Cloud' is slower growing than standard redbud, and seems to show more extensive variegation when grown in partial shade. During periods of cool weather, the variegation is accompanied by the development of red pigment in newly emerged growth, providing a beautiful combination of colors.

'Flame' – 'Flame' is an interesting "double–flowered" cultivar in which the flowers contain about 20–25 petals, additional stamens, and typically a single, malformed pistil. Often described as a "new" cultivar, but actually discovered in the wild in 1902 in Illinois, and according to Dirr, named and introduced by Louis Geraldi Nursery, O'Fallon, IL In 1965. This cultivar was described in an article entitled "A double–flowered redbud (Cercis canadensis var. plena)", published in volume 23 of the Missouri Botanical Garden Bulletin in 1935. The extra production of petals in the flower distorts the typical legume flower morphology. Generally flowering about a week later than other redbuds, this cultivar appears to be almost totally female sterile. I have not observed a pod on the mature tree at the Arboretum in three years, although I have observed minimal fruit set on trees at Hidden Hollow Nursery in Belvidere, TN. However, pollen production is abundant, and we have successfully used 'Flame' as a male parent in controlled crosses. First generation hybrid plants derived from a cross of 'Flame' and 'Alba' were double flowered, confirming dominance of the double–flowered trait. Interestingly, propagation of 'Flame' via semi–hardwood cuttings has proven to be very easy, unlike most Cercis canadensis cultivars. F1 hybrid plants derived from the 'Alba' x 'Flame' cross also have rooted easily, demonstrating genetic transmission of rooting ability from 'Flame' to its offspring. Flower color is rose pink, and the tree is beautiful in bloom. 'Flame' is reputed to be less cold hardy than other redbuds. It grows very vigorously, and tends to demonstrate narrow branch angles and upright growth habit, and can break under ice or snow load.

'Wither's Pink Charm' – A bright pink–flowered form found in the mountains of Virginia around 1930 by D. Withers. Like many of the unusual redbuds, it is difficult to find in commerce. A very attractive tree, I had the chance to see the tree at the JCRA flower in 1998 and 1999 before it perished from canker disease in the summer of 1999. 'Appalachia' – A relatively new cultivar, 'Appalachia', sometimes referred to as

'Appalachian Red' – Although promoted as red flower, the flowers are a rich purple, close to red, particularly when just emerging from the bud, but not a true red. 'Appalachia' flowers later than most cultivars, typically at the same time as 'Flame'.

'Tennessee Pink' – Another pink flowered introduction, in my mind a truer pink than that shown on 'Wither's Pink Charm'. Discovered, introduced and marketed by Harald Neubauer, Hidden Hollow Nursery, Belvidere, TN.

'Pauline Lily' – Another pink flowered plant being propagated and marketed by Harald Neubauer, this cultivar was discovered in the mountains of West Virginia, and named after the wife of the discoverer. Not yet included in the Arboretum collection, Mr. Neubauer tells me this tree produces almost white flowers that have only a slight pink blush. Mr. Neubauer also states that 'Pauline Lily' flowers late, along with 'Appalachia' and 'Flame'.

'Covey' (Trademarked as 'Lavender Twist') – One of the two interesting arching/weeping cultivars available in the trade. This cultivar, a true C. canadensis type, has stems that show an interesting contorted, zigzag growth pattern, adding to the unique architecture. Leaf size is normal, to perhaps slightly larger than the species. The original plant was grown in the garden of Cornelia Covey of Westfield, NY since the mid–1960's, and propagated and patented by Tim Brotzman, Madison, OH. The plant patent number is 010328, awarded in 1998. A small tree, only about 3 years old, exists in the JCRA, but it flowers prolifically and already shows the typical weeping architecture. 'Covey' produces typical purple flowers, produces prolific pollen, and sets abundant fruit. Crosses of 'Covey' with standard non–weeping redbuds gives rise to normal progeny in the first generation; however, plants appear a bit weaker and slower growing than normal, and show a tendency to exhibit a hint of pendant architecture.

'Traveller' – The other weeping form available in the trade, this patented cultivar (plant patent 008640, issued 1994) is botanically a selection of C. canadensis var. texensis, found in the western to southern portion of the range, primarily Texas, Oklahoma, and Mexico. In my opinion, a more attractive cultivar than 'Covey' as the weeping growth and the glossy, leathery leaves typical of botanical variety 'texensis' combine to provide a beautiful landscape tree. Interestingly, 'Traveller' may represent a genetic dead–end from a breeding standpoint. Fruit set has never been observed on the JCRA trees, and in our experience, no functional pollen is produced. Flowers are typical purple. A testament to the fact that it "pays to keep your eyes open", 'Traveller' was discovered as a chance seedling in a row of trees by Dan Hosage, Jr. of Madrone Nursery, San Marcos, TX.

'Texas White' – A white–flowered cultivar of C. canadensis var. texensis. Like 'Traveller', 'Texas White' possesses attractive, glossy, leathery leaves and typical of variety 'texensis', showing a more compact growth habit. It is white flowered, and slightly later blooming than typical C. canadensis. Botanical variety 'texensis' appears to be sexually compatible with typical C. canadensis, as we have successfully hybridized 'Texas White' with 'Silver Cloud', with the long–term objective of developing variegated foliage forms with purple or white flower color possessing the compact growth, and glossy, leathery foliage typical of botanical variety 'texensis'.

'NC–3' – This accession is not a named cultivar, but a selection of C. canadensis var. mexicana obtained by J.C. Raulston. Not being familiar with the extent of variation in this botanical variety, I am uncertain if the tree in the Arboretum is typical. In my opinion, 'NC–3' is noteworthy for the very attractive purple pigmentation in its fruit that persists throughout the growing season. This selection also possesses pubescent stems and leaves, which appears to be transmitted to first generation progeny when used in crosses with normal non–pubescent C. canadensis. 'NC–3' is probably not worthy of cultivar status, but is a useful plant for breeding.

General Comments on Breeding of Cercis canadensis

The potential for breeding new and novel forms of redbud is great due to the high number of interesting genetic variants present in the species and the related botanical varieties 'texensis' and 'mexicana'. Having attempted controlled hybridizations between specific cultivars for 3 years, I agree with other breeders of this group that controlled hybridization is very difficult, fruit set is quite low, often less than 5%, and success varies greatly from year to year based on unknown factors. In our three years of hybridization, we have had reasonable success in one year, and very low set in the other two. Evidence we have obtained in the 3 years of experience working with redbud, and insights obtained by other breeders, suggests that redbud may be self–incompatible, meaning that 2 different cultivars growing in reasonable proximity to each other are needed for fruit set, similar to the situation in many plants such as apple. An isolated tree of a single cultivar would be predicted to set very few or no fruit. If true, instead of resorting to the laborious mechanics and typical low success rate of hand hybridization, it should be possible to grow trees in pots, place the two different potted cultivars one wishes to hybridize together in a screen cage to exclude foreign pollen, introduce some bumblebees (the primary pollinator) into the cage, and allow the bumblebees to make the crosses. Because of self–incompatibility, all seed produced on each parent tree should be a hybrid between the two cultivars. We have had limited success with this protocol, and it certainly warrants further trial.

To date, we have focused our efforts on combining the variegated leaf character from 'Silver Cloud' with the other interesting characters found in Cercis. We are also growing out a large population of open pollinated offspring derived from collecting seed from the 'Silver Cloud' tree at the JCRA. If we are correct in assuming that redbud is largely self incompatible, such seed (referred to as open pollinated seed) should represent crosses of 'Silver Cloud' with the entire range of the other cultivars growing in the same proximity. These trees are being grown at the Sandhills Research Station in Montgomery County, NC, along with open pollinated progeny from various other cultivars. These trees were established as 3–month–old transplants in May,1999. Some flowering occurred in spring, 2000 after only one growing season, particularly those populations derived from C. canadensis var. mexicana. Out of about 700 progeny derived from seed collected off of 'Silver Cloud', we have recovered 2 variegated offspring. These will be evaluated for flower color and other traits to ascertain if they vary significantly from 'Silver Cloud'. Other controlled hybridizations that have yielded progeny include:

All F1 progeny from each family have been established in isolation blocks at the Sandhills Research Station. F2 seed will be collected when trees flower and fruit, and segregating populations grown out to recover desired combinations of traits.

The financial assistance of the North Carolina Association of Nurseryman and the North Carolina Landscape Association is sincerely appreciated.

Return to METRIA 12

Web Crafter: Anne S. Napier ~ Email:

Format updated November 3, 2009