Abstract
Salix babylonica, the classical “weeping willow” from China, is now regarded as including S. matsudana. The species is represented by several valid cultivars, including ‘Babylon’, which is proposed here for the first time for the female clone upon which the original species description may have been based. Since ‘Babylon’ is not cold hardy in many areas of the United States, the most commonly planted “weeping” willows are actually hybrids between ‘Babylon’ and S. alba or S. fragilis. Unfortunately, many of these hybrids suffer from cold-related twig dieback, their nomenclature is hopelessly confused, and several different clones may be growing under the same cultivar name. It is proposed that most of the older names used to denote cultivars and selections be abandoned and new names, based on plants propagated from living, documented specimens, be used in the future.
By the waters of Babylon, there we sat down and wept, when we remembered Zion. On the willows there we hung up our lyres.
Psalms 137
This particular translation of Psalm 137, cited by Zohary (6), is ample evidence that confusion with regard to weeping willows is not a recent phenomenon. The “willows” of the Psalm were really poplars (Populus euphratica Oliv.) and the “weeping” willow (Salix babylonica L.) is probably native to China.
The fact is that most of the “weeping” willows presently cultivated in the United States are not S. babylonica. Furthermore, if these weeping trees are known by any other name, that name is also probably incorrect. It has been well documented in the literature (1) and confirmed by our study that several different clones, even of different sexes, may be growing under the same name. Since there are more individual genotypes (clones) of weeping willows than there are names (cultivar names or otherwise), it may be best to discard most of the earlier names and start over. There is no reason to allow the present nomenclatural confusion to continue, or increase, when we have the means at our disposal to bring some order out of chaos.
The present study is a combination of literature research on plant nomenclature and experimental and observational research on plant biology. Its purposes are to justify a realistic appraisal of the cultivated weeping willows and offer solutions that will al low a more efficient and intelligent use of the germplasm available.
Nomenclature
Much of the historical and anecdotal information given below has been taken from Bean (1). Although this work contains an abundance of observations, no attempt was made to integrate disparate views. All of the plants presently cultivated in Europe and the United States under the epithet “Salix babylonica L. ” are female, and probably are genetically identical members of the same clone. In the listing that follows, we have proposed ‘Babylon’ as a cultivar name for this clone. According to Bean (1), Linnaeus first saw this weeping willow in the Clifford garden at Hartecamp in the Netherlands and described it in Hortus Cliffortinaus in 1738. Although the plant had not flowered, Bean (1) stated “there seems to be little doubt that Clifford’s willow was the common female clone”. However, when Linnaeus published the name “Salix babylonica” in Species Plantarum in 1753, his herbarium contained four specimens, presumably from different sources, purported to be this species. All of these specimens were of vegetative shoots, with no male or female catkins. The word “China” is written on one of these herbarium sheets. Thus, it cannot be proved that Linnaeus’ description was based on only one clone or that the choice of the specific name “babylonica” was made without knowledge of the Asiatic origin of the species.
It is, however, likely that the S. babylonica cultivated in the West (the female clone ‘Babylon’) was a highly atypical selection (from an Asiatic species) that had been introduced at various points along the ancient trade route through southwest Asia, to the Near East, and in ca. 1730 to Europe. The non-weeping taxon from China described by Koidzumi in 1915 as S. matsudana is probably more typical of the majority of wild S. babylonica. Rehder (2) and others placed certain non-weeping (or less-weeping) selections from China as formae under S. matsudana. Before 1915, it was probably thought reasonable to include both weeping and non-weeping trees under S. babylonica. Following the lead of Skvortsov (5), it is again fashionable, and probably biologically sensible, to consider S. babylonica as including S. matsudana.
In the listing that follows, we have attempted to include, and explain, most of the names that have been applied to S. babylonica or its “weeping” derivatives. As in previous checklists, VALID CULTIVARS are given in boldface capitals and INVALID names in lightface capitals.
ANNULARIS—based on S. babylonica f. annularis by Ascherson in 1864. Should be considered as synonymous with S. babylonica CRISPA.
BABYLON—This cultivar name is here proposed for the female clone that is widely grown in Europe and eastern United States as S. babylonica. Tree to 30 ft, with long pendulous branches; branchlets brown (not yellowish), glabrous; leaves lanceolate to linear lanceolate, 8-16 cm long, long acuminate, with cuneate base and serrulate margin; female catkins curved, to 2 cm long. We consider our trees (NA 44011) at the U.S. National Arboretum to be typical for this cultivar.
BLANDA—as S. x blanda by Andersson (1867) to denote a hybrid between S. babylonica and S. fragilis growing in Germany. Bean (1) considered this somewhat pendulous female clone as a cultivar, but we have seen several clones grown under this name. Even the putative parentage of the plant, judged only on herbarium specimens, is in doubt. Bean noted that the male parent may have been S. x rubens Shrank (S. alba x S. fragilis) or S. pentandra L. The name “blanda” should probably be used only as a hybrid specific epithet (S. x blanda) to denote putative hybrids between S. babylonica and S. fragilis.
CHRYSOCOMA—considered a cultivar by Bean, (1) listed under S. x sepulcralis Simonkai. Based on S. x chrysocoma Dode (1908), with S. vitellina pendula nova Spath, S. alba var. vitellina (Spath) Rehder, S. babylonica ramulis aureis Hort., and S. alba ‘Tristis’ Hort. in synonomy. Put into commerce by Spath (Berlin) in 1888. Origin unrecorded, and Bean expressed doubts as to its hybridity, but we believe it to be a hybrid between S. babylonica and S. alba. It is probable that many, if not most, of the “golden weeping willows” grown in the United States under the name ‘Tristis’ are this clone. However, until a type specimen is established and nursery propagation is standardized, we consider the use of this name to be of doubtful utility.
CORKSCREW—a common name for S. babylonica TORTUOSA.
CONTORTED (H.P. Kelsey and W.A. Dayton, Standardized Plant Names, 1942, p. 574)—as a clone (tortuosa) of S. matsudana. = S. babylonica TORTUOSA.
CRISPA—based on S. babylonica f. crispa (Loudon) Rehder (1949), originally considered a variety of S. babylonica by Loudon (1838). Branches pendulous and leaves twisted or spirally curved. According to H.J. Elwes and A. Henry, The trees of Great Britain and Ireland, 1913, Vol. VII, this is a female clone. A valid cultivar of S. babylonica.
DOLOROSA—a name used variously as S. dolorosa Hort., in synonomy with S. babylonica or as a variety of S. babylonica, as S. babylonica dolorosa Hort. A. Rehder, Bibliography of cultivated trees and shrubs, 1949, p. 76, placed S. babylonica var. dolorosa Rowen ex Rowlee in the synonomy of S. x blanda. Name probably not valid at any level.
DRAGON’S CLAW—mentioned by Bean (1) as a common name for S. matsudana ‘Tortuosa’ = S. babylonica TORTUOSA.
ELEGANTISSIMA—as S. x elegantisslma K. Koch (1871) to denote a hybrid between S. babylonica x S. fragilis. Excellent discussion of this name and its problems in Bean (1). Not a cultivar, best considered as a synonym for S. x blanda Andersson.
GOLDEN (H.P. Kelsey and W.A. Dayton, Standardized Plant Names, 1942, p. 573)—as a horticultural variety (ramulis aureis) of S. babylonica. Not a valid cultivar name.
GOLDEN CURLS (Girard Nurs., Geneva, Ohio, Cat. 1976, p. 29)—as S. tortuosa aurea pendula, a new goldenleaved willow “introduced by Charles Beardsley for first time this year". Had only been grown for four years prior to this date.
LAVALLEI—considered a valid cultivar of S. babylonica by Bean (1). A male clone, less pendulous than the female S. babylonica (BABYLON). May be the same as the male clone known in Japan as S. babylonica var. lavallei f. seiko Kimura. No plants listed under this name in American arboreta.
MATSUDANA—as S. matsudana Koidz. in 1915. Considered synonymous with S. babylonica.
NAPOLEON—according to Bean (1) plants of the true S. babylonica (BABYLON) were planted at the gravesite of Napoleon on St. Helena after his death in 1921. Whether the “Salix Napoleonis” sold by several nurserymen were of this clone or others is unknown. Not a valid cultivar name.
NIOBE—a common name of unknown origin and no particular botanical status, similar in its application to WISCONSIN (A. Rehder, Manual of cultivated trees and shrubs, 1940) as refering to a hybrid between S. babylonica and S. fragilis (S. x blanda Andersson). Name derived from tragic heroine in Greek mythology who wept for her murdered children. Not a valid cultivar name.
PENDULA—the name “pendula” has been applied to so many weeping or semi-weeping willows that it has lost any significance. Bean (1) considered PENDULA as a valid cultivar of S. matsudana, based on the female clone grown in Great Britain. However, a pendulous male clone sent to the Arnold Arboretum (U.S.) in 1908 is also mentioned. Might be considered as a valid name for the female clone if its identity can be verified.
PETZOLDI—as S. petzoldl Hort. and S. petzoldi pendula Hort. in synonomy of S. blanda (A. Rehder, Bibliography of cultivated trees and shrubs, 1949). Not a valid name at any level.
RAMULIS AUREIS—as S. babylonica ramulis aureis Hort., placed in synonomy of CHRYSOCOMA by Bean (1).
RINGLEAF (H.P. Kelsey and W.A. Dayton, Standardized Plant Names, 1942, p. 573)—as a common name for S. babylonica annularis and S. babylonica crispa. = S. babylonica CRISPA.
SACRAMENTO—as a cultivar of S. babylonica in G. Krussmann, Manual of cultivated broad-leaved trees and shrubs (Vol. Ill, p. 280), English translation published by Timber Press, Portland, Oregon in 1986. Tree less weeping, crown open, large leaved. Strangely, not found in G. Krussmann, Hanbuch der Laubegeholze (1978), on which the English translation was based. The herbarium of the U.S. National Arboretum contains a specimen collected in 1955 in Argentina from a plant (female) labeled S. babylonica var. sacramenta which had been received under that name from Switzerland. It is therefore not possible to validate either sacramenta or ‘Sacramento’, but any connection of the name with Sacramento, California can be ruled out.
SALAMONII—considered a cultivar by Bean (1) as synonymous with “Sepulcralis”, under S. x sepulcralis Simonkai (S. x salamonii Carr, ex Henry), a group of hybrids between S. alba and S. babylonica. First appeared on the property of Baron de Salamon before 1864 and was put into commerce by Simon-Louis Freres, Metz, France, in 1869. According to H.J. Elwes and A. Henry, the Trees of Great Britain and Ireland, 1913, Vol. VII, the young plants develop a handsome pyramidal crown, and only the tips of the branches are pendulous; only female trees known. Several trees with this growth habit have been observed in the Washington, DC area. Probably a valid cultivar name, but verified material should be introduced from Europe.
SEPULCRALIS—as S. x sepulcralis Simonkai (Oesterr. Bot. Zeitschr. 40:424.1890), S. babylonica x S. alba, a valid botanical epithet under which all hybrids between these two species must be placed. Not a cultivar.
SEPULCRALIS—S. x sepulcralls Simonkai (Oesterr. Bot. Zeitschr. 40:424.1890), S. babylonica x S. alba, a valid botanical epithet under which all hybrids between these two species must be placed. Not a cultivar.
SOLOMON (H.P. Kelsey and W.A. Dayton, Standardized Plant Names, 1942, p. 575)—as a common name for S. x sepulcralls, with S. salamoni in synonomy, and noted as S. alba x S. babylonica. Not a valid cultivar name.
THURLOW—a common name of unknown origin often applied to plants known as S. x elegantissima K. Koch (A. Rehder, Manual of cultivated trees and shrubs, 1940). Apparently a female clone with longer branches than the plant known as “blanda”. The plant should probably be considered as S. x blanda (S. alba x S. fragilis) but the name should not be used to denote a valid cultivar.
TORTUOSA (A. Rehder, J. Arnold Arb. 6:201-208, 1925)—as S. matsudana f. tortuosa. Based on S. matsudana var. tortuosa of Vilmorin (1924) and then in cultivation at Vilmorin’s nursery near Paris. Upright growing, with tortuous and twisted branches. A female clone.
TRISTIS—as S. alba ‘Tristis’ Hort., placed in the synonomy of CHRYSOCOMA by Bean (1). This is not to be confused with S. alba var. tristis (Seringe) K. Koch, which is a forma of S. alba with weeping branches. Plants grown in the United States under the name ‘Tristis’ are hybrids between S. babylonica and S. alba (S. x sepulcralis Simonkai) but the name cannot be unequivocally assigned to any particular clone.
UMBRACULIFERA (A. Rehder, J. Arnold Arb. 6:201-208, 1925)—as S. matsudana f. umbraculifera. Based on a 1906 USDA introduction by F.N. Meyer as Salix sp., PI. 17737. Plants growing at Chico, California had developed dense subglobose crowns (20 ft tall, 30 ft crown spread) in 6 years. Called “bread willow” by Chinese because crown resembled a loaf of bread. Sex not noted in 1925, but a specimen in the herbarium of the U.S. National Arboretum, grown from cuttings of the original introduction, is female.
UMBRELLA (H.P. Kelsey and W.A. Dayton, Standardized Plant Names, 1942, p. 574)—as a clone (umbraculifera) of S. matsudana. = S. babylonica UMBRACULIFERA.
VITELLINA—as S. alba var. vitellina (L.) Stokes in Bean (1), with S. vitellina L. in synonomy. The variety vitellina is a group of clones, some male and some female, having yellow twigs. Not a valid cultivar name under S. alba.
VITELLINA PENDULA—as S. alba var. vitellina pendula (Spath) Rehder, placed in synonomy under CHRYSOCOMA by Bean (1). This is the same clone described as S. x chrysocoma by Dode in 1908.
VITELLINA PENDULA NOVA—as S. vitellina pendula nova Spath, placed in the synonomy of CHRYSOCOMA by Bean (1). Plant put into commerce by Spath Nurs., Berlin, in 1888, but this clone was re-named S. x chrysocoma by Dode in 1908.
VITELLINA TRISTIS—considered a cultivar of S. alba by Bean (1). A semi-pendulous form of S. alba var. vitellina (L.) Stokes “received by the botanist Seringe from Baumann’s nursery in Alsace a few years before 1815. Probably a valid cultivar of S. alba; but not of S. babylonica or its hybrid derivatives.
WEEPING (H.P. Kelsey and W.A. Dayton, Standardized Plant Names, 1942, p. 574)—as a clone (pendula) of S. matsudana. Not a valid cultivar name.
WEEPINGGOLD (H.P. Kelsey and W.A. Dayton, Standardized Plant Names, 1942, p. 573)—as a clone of S. alba, with S. alba tristis and S. (alba) vitellina pendula as synonyms. Not a valid cultivar name.
WISCONSIN—a common name of unknown origin and no particular botanical status, usually used in the phrase “Wisconsin weeping willow” to refer to a taxon known botanically as S. x blanda Andersson (S. babylonica x S. fragilis), which may encompass several clones. The type for S. x blanda grew in Hessen, Germany and the only possible reason for the appelation “Wisconsin” might have been that a plant of this putative parentage was cold hardy somewhere in that state. Not a valid cultivar name.
Isozyme Studies
In January, 1983, we solicited dormant cuttings of willows from: the Arnold Arboretum, Jamaica Plain, Massachusetts; the Morton Arboretum, Lisle, Illinois; and the University of Minnesota Landscape Arboretum, Chanhassen, Minnesota. Our request was based on the plants listed in the 1979 microfiche edition of the Master Inventory of Botanical Taxa published by the Plant Sciences Data Center of the American Horticultural Society. We asked for cuttings of each and every entry, by accession number, of plants bearing the epithets “babylonica”, “blanda”, “elegantissima”, “matsudana”, “sepulcralis”, or “tristis”. These plants were propagated, along with similar plants from the collections of the U.S. National Arboretum, by sticking six to ten untreated dormant cuttings directly into a soil mix contained in large wire baskets.
In 1984, we analyzed the leaves and cambial tissue of each clone for isozyme variability (peroxidase, alcohol dehydrogenase), using established techniques of starch-gel electrophoresis (3, 4). First of all, it was apparent that all 10 taxa identified as “tristis” were hybrids between S. babylonica and S. alba. This relationship could be determined by comparing their banding patterns with our specimen of ‘Babylon’ (NA 44011) and a clone of S. alba (NA 44016). Secondly, the 10 “tristis” (or “chrysocoma”) taxa could actually be identified as four distinct clones! Interestingly, when we made the original requests for plant material, we had not realized that the four “tristis” accessions obtained from the Morton Arboretum represented only two clones. Our isozyme results did show this to be true, but also showed that these two “tristis” taxa were not identical.
Isozyme data also showed a remarkable similarity among an “elegantissima” from the Arnold Arboretum, a “sepulcralis” from the Morton Arboretum, and a “babylonica” from the University of Minnesota. None of the “babylonicas” from the Morton Arboretum or the University of Minnesota was identical to the supposedly correct specimen of ‘Babylon’ (NA 44011) at the National Arboretum. Taxa designated as “blanda” could likewise be identified as four distinct clones.
Using only two isozyme systems, we could not “fingerprint” the various clones as accurately as one might wish, but this was not our intention. We only wished to examine genetic variability. Which specimen was, indeed, the true “tristis”, or the true “Niobe”? Who can say? What we have shown is that many of the various names (species, hybrids, cultivars) under which weeping willows are presently cultivated in our major arboreta are virtually meaningless.
Disease Observations
The wire baskets containing the rooted willow clones were mulched in and overwintered in a coldframe in the National Arboretum lath house. During the spring of 1985, we noted a major “disease” outbreak on many of these plants. Black, sunken areas appeared on many of the stems and the portion of the stem distal to the “canker” died. The causal organism was cultured and identified as Botryosphaeria dothidea by Dr. R. Jay Stipes (VPI & SU, Blacksburg, VA). This condition was observed on National Arboretum specimen trees of ‘Babylon’ (NA 44011), ‘Tortuosa (NA 44014) and “tristis” (NA 8929, NA 43995) as well as plants of ‘Crispa’ and ‘Pendula’ from the Arnold Arboretum. At least one plant of every name was infected but at least one plant of every name seemed to be resistant. It was noteworthy that only one of eight clones from the Arnold Arboretum was resistant while eight of nine from the Morton Arboretum had no infection. These apparently resistant Morton trees included both clones of “tristis”, neither of which agreed isozymically with the National Arboretum trees under the same name. As a matter of interest, we attempted artificial inoculation of B. dothidea into stems of 1-year-old rooted cuttings in the greenhouse, but no symptoms developed. Apparently, cold weather or other stresses are necessary for symptom development on plants with a certain proportion of germplasm from S. babylonica. In 1986, we established a permanent field test plot for all of the clones included in this study.
Conclusions and Recommendations
It may be that a weeping willow, by any name, should be suspect. However, we think that agreement can be reached on establishing and maintaining the identities for the the various valid cultivars of S. babylonica and others recognized above. The various hybrid “weepers” are another matter. It may not, or, more likely, will never be possible to find, identify, or verify a clone to fit each name. There may be more clones than names, or more names than clones. Whatever the case, we recommend that the use of most, if not all cultivar names considered invalid in this paper be abandoned by the nursery trade. There is no need to continue the confusion already rampant.
There may, indeed, be some weeping willows in our arboreta and nurseries that are superior to most of the others. If so, we should be propagating them, naming, and using them. If nothing else, we should know what to expect when we purchase and plant a tree of a particular cultivar.
- © 1988, International Society of Arboriculture. All rights reserved.