in Vancouver, British Columbia, Canada - Davidsonia

Volume 18, Number 2April 2007DavidsoniaA Journal of Botanical Garden Science

Davidsonia 18:2 41EditorialThe paper by Wharton and Lancaster published in this issue servesto remind us of problems that arise when we bring rare or unusualplants into cultivation. Wharton’s account of re-introduction, in thiscase of Carrierea calycina, reports earlier at least partially successfulintroductions. Lancaster provides the background of successful cultivation.The satisfying part of their story is that their re-introductionsare from seed.Botanical gardens provide expertise for in situ conservation as wellas being repositories for ex situ activities. The ex situ function is not anissue for common forms, but movement of rarities into cultivation maylead to difficulties that arise from a collector’s urge to ‘have one of thosein our garden’. It is not clear from the Wharton and Lancaster paper ifthe distribution of seed was accompanied by a request for the recipientsto keep a concise diary of the handling and progress of the materialsto ensure the collection of (standardized) information to increase therecorded experience that came from growing these materials. The opportunityto have access to this and other seemingly rare and uncultivatedmaterial would be well exploited by asking each recipient to shareexperience gained during early cultivation of new material. This simplecontribution would ensure that the knowledge is retained and not left tochance. While there may be a few seeking to control technical informationfor profit, the Convention for Biological Diversity (CBD) providesa framework for information to be gathered at one or more botanicalgardens and shared with the country of origin.The ideal that botanical garden collections contain accessions of anadequate sound sample size (some say 5 seed source specimens and notmerely 5 copies of a single clone) presents a dilemma for collectors whofind a single specimen with perhaps only vegetative prospects. The col-Iain E.P. Taylor, Professor of Botany and Research Director,UBC Botanical Garden and Centre for Plant Research,6804 SW Marine Drive, Vancouver, BC, Canada, V6T 1Z4.iain.taylor@ubc.ca

42lection and successful propagation of vegetative material is an importantfirst step, it is important to train local people to collect seed, and toreturn to the site when seed is available. Both of these are important,though costly in both time and money; but they do contribute to thesponsoring botanical garden’s essential obligations to the Conventionfor Biological Diversity.The paper by Saarela is another example of how researchers canprovide reliable information when the biological conditions, in this caseflowering, are unknown. Individual ‘case studies’ such as this add tothe reliable body of horticultural knowledge. Eventually there will beenough natural history to launch a doctoral dissertation that will addressthe specific mechanisms of bamboo flowering.

Davidsonia 18:2 43The North American floweringof the cultivated fountainbamboo, Fargesia nitida(Poaceae: Bambusoideae), inVancouver, British Columbia, CanadaAbstractFargesia nitida, the fountain bamboo, is a hardy and attractivebamboo that is commonly cultivated in North America. Because itsreproductive structures were not known until recently, there have beenseveral taxonomic issues with this species. Fargesia nitida began floweringin 1993 in the United Kingdom for the first time since its originalcollection in its native China in 1886, and it has since been expected toflower in North America. Here I document its recent flowering in Vancouver,British Columbia, and I provide a morphological description offlowering individuals of this species.Introduction“With the commencement of flowering of some European plants of umbrellabamboo, we should soon observe flowering in this species in cultivationelsewhere. We may also predict that the fountain bamboo will soon comeinto flower. If so, we have one of nature’s most interesting phenomena unfoldingbefore us.”– Thomas R. Soderstrom, 1979The bamboosBamboos (subfamily Bambusoideae Luerss.) are a large clade ofgrasses (Poaceae) that includes 1,200–1,400 woody and herbaceousspecies (Grass Phylogeny Working Group, 2001; Bamboo PhylogenyJeffery M. Saarela, Research ScientistResearch Division, Canadian Museum of Nature, P. O. Box 3443, Station D,Ottawa, Ontario, K1P 6P4, Canada

46The genus FargesiaThe genus Fargesia includes clump-forming woody bamboos withshort-necked rhizomes and dense, spatheate, unilateral inflorescences(Stapleton, 1994a, b; Li et al. 2006). However, reproductive structuresare known only in about 25% of species (Gielis et al. 1999), and theoccasional acquisition of new flowering material for described Fargesiataxa has indicated that not all share the same reproductive morphology,which has frequently necessitated generic realignments (e.g., Stapleton,1994b; Li et al. 2006). Li et al. (2006) indicated that the genus includesabout 90 species (with 77 endemic to China), but they noted that itwould be more appropriate to include several of their Fargesia species ina segregate genus, Borinda. The generic circumscription of Fargesia, andtherefore its number of known species, remains uncertain.Two Fargesia taxa, F. murielae (Gamble) Yi, and F. nitida (Mitford) P. C.Keng ex T. P. Yi, are closely related species (Guo et al. 2001, 2002; Guoand Li, 2004) that are widely cultivated in Europe and North Americabecause of their extreme hardiness and attractiveness. Fargesia murielae(the umbrella bamboo or Muriel’s bamboo) (Figure 1, see rear cover)is characterized by its gracefully arching culms, whereas F. nitida (theFountain Bamboo) is more upright in stature. Additional characteristicsof the culm leaves, culms, and foliage leaves further distinguish thesetaxa (e.g., Stapleton 1995b; Li et al. 2006). Both species are native toChina, where they grow at high elevations (Li et al. 2006). The historyof the introduction of both species into the West has been documentedthoroughly (Stapleton. 1995a, b; Del Tredici, 1998;). In brief, F. nitidawas initially grown in the United Kingdom from seed collected in Chinain 1886 (Stapleton, 1995b), and F. murielae was collected live in China in1907 and subsequently propagated (vegetatively) at the Arnold Arboretumat Harvard University (Stapleton, 1995a; Del Tredici, 1998). Allindividuals of these species now grown in the West are believed to originatefrom these original Chinese collections. Despite their ecologicaland economic importance in their native habitats, and their importanceas ornamentals in Europe and North America, the generic affinities,morphological characteristics, species circumscriptions, and typificationsof these taxa have been extremely controversial, largely because

48Photo: J.M. SaarelaFigure 2. A flowering branch of Fargesia nitida.unilateral inflorescences (Stapleton, 1994a, b; Li et al. 2006). AlthoughI did not study the type specimens of the commonly cultivated Fargesiataxa, I determined these individuals to be F. nitida on the basis of theirrelatively upright culms. F. murielae, the other commonly cultivated Fargesiaspecies in North America, has culms that arch gracefully. Additionally,these plants growing on the UBC campus are known locally asF. nitida (Douglas Justice, personal communication, December 2006).Photo: J.M. SaarelaFigure 3. Inflorescence of Fargesia nitida [Saarela 798 (CAN)]. Scale bar = 5 mm.

Davidsonia 18:2 49Photo: E.P La FountaineFigure 4. Two individuals of Fargesia nitida outside the MacMillan building on theUniversity of British Columbia campus. In December 2006 the leftmost clump wasflowering vigorously, whereas the rightmost clump was not flowering. When thisphoto was taken (August 2007), the leftmost individual was nearly devoid of foliage,likely a response to its earlier flowering.There are numerous reports on plant enthusiast, nursery, and gardencentre websites indicating that F. nitida began flowering in North Americain the early to mid 2000s. This species was also observed floweringin Vancouver at the VanDusen Botanical Garden in 2005 (D. Justice,personal communication). At the University of British Columbia, I observedsix separate plantings of F. nitida; in December, three of thesewere flowering vigorously, and three were not. Because F. nitida reportedlydies after flowering (many garden centre websites indicate that theyare no longer selling “old generation” fountain bamboos because oftheir imminent flowering and death). I revisited each of the six plants inearly June, 2007, to determine if this was the case. At that time, each ofthe three individuals that were observed flowering vigorously in Decem-

50ber had some reproductive structures at the anthesis stage, suggestingthat these plants had probably flowered continuously over the previoussix month period, and that they had not finished flowering in early June.However, the density of reproductive structures on each of these plantswas substantially reduced (I did not formally quantify flowering density).In early June, the flowering clump outside of the UBC StudentUnion Building showed no signs of ill health – the plant was dense withfoliage and looked remarkably similar to its non-flowering neighbouringplant. Large portions of the two clumps that had flowered vigorouslybehind the MacMillan building, however, were nearly devoid offoliage, suggesting that flowering had affected their survival (Figure 4).The two individuals at this location that were not flowering in Decemberappeared vigorous and healthy in August 2007. It is not clear whythe three flowering plantings were not all affected in the same manner,but since the flowering cycle in these individuals is presumably not yetcomplete, it seems premature to determine conclusively their ultimatepost-flowering fate.Despite casual observations that F. nitida has flowered recently inNorth America (see above), I am not aware of any previous reportsin the scientific literature documenting this event — its first floweringin North America, approximately 120 years after its seed was firstcollected in China, and approximately 13 years after its first recordedflowering in the United Kingdom (Renvoize, 1993). Collections madefrom these flowering individuals of F. nitida (as well as some individualsnot in flower) in British Columbia therefore provided an opportunity tocharacterize the morphology of its reproductive structures, and fill thisgap in the taxonomic literature. The following detailed morphologicaldescription of leaf foliage and reproductive morphology of F. nitidais based on collections made from these cultivated specimens. I hopethat these details will help to clarify species circumscriptions of thisand other wild and cultivated Fargesia species. A detailed description ofculm foliage (i.e., culm sheaths and culm blades) is not provided herebecause the available collections did not have sufficient material fromthis portion of the plants. For a detailed description of culm morphology,see Li et al. (2006).

Davidsonia 18:2 51Description of leaf foliage and reproductivemorphology of Fargesia nitidaFargesia nitida (Mitford) P. C. Keng ex T. P. Yi, J. Bamboo Res.4(2): 30. 1985.Note—The taxonomy and nomenclature of Fargesia nitida is complex.For detailed information on the type specimen, correct applicationof the name F. nitida, and synonymy, see Stapleton (1995a,b, 1999), Li(1996), and Brummitt (1998).Foliage leaf sheaths glabrous, occasionally pubescent towards collar;margins not fused, usually smooth, sometimes pubescent near apex,hairs 0.2–0.5 mm long; auricles absent; oral setae 0.3–2.0 mm long,0.02–0.04 mm wide, not fused, straight or slightly sinuous, smoothwith occasional, minute, crater-like glands observable at 50x. Foliageleaf blades narrowly linear-lanceolate, 3–6.6 cm long, 4–7 mm wide,constricted at base, abaxial surface minutely pubescent, apex pubescent-scabrous,adaxial surface glabrous, margins smooth, serrulate, orwith antrorse hairs to 0.35 mm. Outer ligules a ciliolate rim 0.1–0.2mm. Inner ligules 0.4–0.9 mm, truncate or rounded, glabrous or pubescent,margins glabrous or ciliate, cilia to 0.1 mm. Synflorescences(2–)3–3.6 cm long, 0.7–1 cm wide, strongly compressed, subtended by3–4 spathes; spathes purple to brown, strongly enclosing synflorescencebase, strongly nerved, glabrous, margins usually glabrous but occasionallydensely ciliose, hairs to 0.7 mm, inner most spathe exceeding spikelet,often terminated by blade; rachis and pedicels terete, 0.2–0.3 mmin diameter, usually glabrous, occasionally minutely serrulate towardsnodes, longer hairs occasional particularly at nodes, nodes occasionallydensely hairy, hairs to 1 mm; branches inserted unilaterally, 1.4–1.8 mmlong. Spikelets 11.2–13.4 mm long, laterally compressed; florets 2 or3, the third floret usually sterile and smaller, disarticulating above theglumes. Glumes 2, lanceolate, purple, usually strongly keeled, sometimesflattened, particularly towards apex, surface glabrous, occasionallypubescent towards apex, keels glabrous, occasionally minutely serrulatetowards apex; first glume 4.7–6.5 mm long, 1-3–nerved; second glume

526.8–8 mm, 5-nerved. Lemmas 9.8–11 mm long, green at base becomingpurple towards apex, strongly keeled, strongly 7-nerved, keels smoothbecoming pubescent towards apex, margins smooth, backs glabrous(sometimes with occasional hairs) becoming scabrous towards apex;awns minute and mucro-like, 0–0.7 mm. Paleas 8.9–10.7 mm long, 1-3–nerved, bifid, divided apically for 0.2–0.7(–0.9) mm, sulcate, glabrousbut becoming scabrous towards apex. Stamens 3; anthers 3.6–4.8 mmlong, 0.5-0.9 mm wide, bright yellow; filaments 0.1 mm wide, translucent.Styles 2, feathery. Lodicules 2, 2.1–2.8 mm long, triangularshaped, glabrous, apex pubescent, hairs 0.2–0.4 mm long. Fruit notobserved.Specimens examined: British ColumbiaUniversity of British Columbia (UBC), N side of Student UnionBuilding (SUB), lower level, planted on either side of door, westmostclump, in flower, 13 Dec 2006, Saarela 798 (CAN, UBC); UBC, N sideof SUB, lower level, planted on either side of door, eastmost clump,not in flower, 13 Dec 2006, Saarela 799 (CAN); UBC, behind MacMillanbuilding, directly along wall, westmost clump, 13 Dec 2006, Saarela 801(CAN); UBC, behind MacMillan building, directly along wall, eastmostclump, 13 Dec 2006, Saarela 802 (CAN).AcknowledgementsI am grateful to Douglas Justice (UBC Botanical Garden) for information,Paul Hamilton (Canadian Museum of Nature) for help withmicroscope photography, and Lynn Gillespie (Canadian Museum ofNature), Christopher Sears (University of British Columbia), and ananonymous reviewer for constructive comments on earlier versions ofthe manuscript.ReferencesBamboo Phylogeny Group. 2005-2006. Bamboo Biodiversity. http://www.eeob.iastate.edu/research/bamboo/index.html. Accessed 30 July 2007.Bhattacharya, S., Das, M., Bar, R., and Pal, A. 2006. Morphological and molecularcharacterization of Bambusa tulda with a note on flowering. Annalsof Botany 98: 529-535.

Davidsonia 18:2 53Brummitt, R. K. 1998. Report of the Committee for Spermatophyte. Taxon47: 863-872.Clayton, W.D., Harman, K. T., and Williamson, H. 2006 onwards. GrassBase- The Online World Grass Flora. http://www.kew.org/data/grassesdb.html.Accessed 26 February 2007.Del Tredici, P. 1998. The first and final flowering of Muriel’s bamboo. Arnoldia58: 11-17Filgueiras, T. S., and Pereira, B. A. S. 1988. On the flowering of Actinocladumverticillatum (Gramineae: Bambusoideae). Biotropica 20: 164-166.Franklin, D. C. 2004. Synchrony and asynchrony: observations and hypothesesfor the flowering wave in a long-lived semelparous bamboo. Journal ofBiogeography 31: 773-786.Gielis, J., Goetghebeur, P., and Debergh, P. 1999. Physiological aspects ofdevelopment and experimental reversion of flowering in Fargesia murielae(Poaceae: Bambusoideae). Systematics and Geography of Plants 68: 147-158.Grass Phylogeny Working Group. 2001. Phylogeny and subfamilial classificationof grasses (Poaceae). Annals of the Missouri Botanical Garden 88:373-457.Guo, Z.-H., Chen, Y.-Y., Li, D.-Z., and Yang, J.-B. 2001. Genetic variation andevolution of the alpine bamboos (Poaceae: Bambusoideae) using DNAsequence data. Journal of Plant Research 114: 315-322.Guo, Z.-H., Chen, Y.-Y., and Li, D.-Z. 2002. Phylogenetic studies on theThamnocalamus group and its allies (Gramineae: Bambusoideae) basedon ITS sequence data. Molecular Phylogenetics and Evolution 22: 20-30.Guo, Z.-H., and Li, D.-Z. 2004. Phylogenetics of the Thamnocalamus groupand its allies (Gramineae: Bambusoideae): inference from the sequencesof GBSSI gene and ITS spacer. Molecular Phylogenetics and Evolution30: 1-12.Janzen, D. H. 1976. Why bamboos wait so long to flower. Annual Review ofEcology and Systematics 7: 347-391.John, C. K., and Nadgauda, R. S. 2002. Bamboo flowering and famine. CurrentScience 82: 261-262.Judziewicz, E. J., Clark, L. G., Londoño, X., and Stern, M. 1999. AmericanBamboos. Smithsonian Institution Press, Washington, D.C.

54Keeley, J. E., and Bond, W. J. 1999. Mast flowering and semelparity in bamboos:the bamboo fire cycle hypothesis. American Naturalist 154: 383-391.Li, D. Z. 1996. Proposal to conserve Sinarundinaria Nakai (Gramineae) with aconserved type. Taxon 45: 321-322.Li, D. Z. 1997. The flora of China Bambusoideae project – problems andcurrent understanding of bamboo taxonomy in China. In The Bamboos.Edited by G. P. Chapman. Academic Press, London.Li, Z., and Denich, M. 2004. Is Shennongjia a suitable site for reintroducinggiant panda: an appraisal on food supply. The Environmentalist 24: 165-170.Li, D., Guo, Z., and Stapleton, C. 2006. Fargesia Franchet. In Flora of China:Poaceae, Volume 22. Edited by Z. Y. Wu, P.H. Raven, and D.Y. Hong.Science Press and Missouri Botanical Garden Press, Beijing and St. Louis.pp. 74-96.Ramanayake, S. M. S. D., and Yakandawala, K. 1998. Incidence of flowering,death and phenology of development in the giant bamboo (Dendrocalamusgiganteus Wall. ex Munro). Annals of Botany 82: 779-785.Ramanayake, S. M. S. D, and Weerawardene, T. E. 2003. Flowering in a bamboo,Melocanna baccifera (Bambusoideae: Poaceae). Botanical Journal of the LinneanSociety 143: 287-291.Renvoize, S. A. 1993. In closing: Sinarundinaria nitida – in flower! BambooSociety (Great Britain) Newsletter 17: 24.Seifriz, W. 1920. The length of the life cycle of a climbing bamboo: a strikingcase of sexual periodicity in Chusquea abeitifolia Griseb. American Journalof Botany 7: 83-94.Soderstrom, T. R. 1979a. The bamboozling Thamnocalamus. Garden 3 (4): 22–27.Soderstrom, T. R. 1979b. Another name for the Umbrella Bamboo. Brittonia31: 495.Soderstrom, T. R., and Ellis, R. P. 1987. The position of bamboo genera andallies in a system of grass classification. In Grass Systematics and Evolution.Edited by T. R. Soderstrom, K. W. Hilu, C. S. Campbell, and M. E.Barkworth. Smithsonian Institution Press, Washington, D. C., U. S. A. pp.225-238.Stapleton, C. M. A. 1994a. The bamboos of Nepal and Bhutan Part I: Bambusa,Dendrocalamus, Melocanna, Cephalostachyum, Teinostachyum, and Pseudostachyum

Davidsonia 18:2 55(Gramineae: Poaceae, Bambusoideae). Edinburgh Journal of Botany 51:1-32.Stapleton, C. M. A. 1994b. The bamboos of Nepal and Bhutan Part II: Arundinaria,Thamocalamus, Borinda, and Yushania (Gramineae: Poaceae, Bambusoideae).Edinburgh Journal of Botany 51: 275-295.Stapleton, C. M. A. 1995a. Muriel Wilson’s bamboo. Bamboo Society (GreatBritain) Newsletter 21: 10-20.Stapleton, C. M. A. 1995b. Flowering of Fargesia nitida in the UK. BambooSociety (Great Britain) Newsletter 22: 17-22.Stapleton, C. M. A. 1996. The Fountain Bamboo – Fargesia nitida. BambooSociety (Great Britain) Newsletter 24: 19-20.Stapleton, C. M. A. 1997. Just when you thought it was safe to get back in thewater – comments on a proposal by D–Z Li to conserve Sinarundinaria.Bamboo Society (Great Britain) Newsletter 26: 25-26.Stapleton, C. M. A. 1999. Yushania vs. Sinarundinaria: good news and bad.Bamboo Society (Great Britain) Newsletter 31: 42-43.Stapleton, C. M. A. 2006. Proposal to conserve the name Arundinaria murielaeagainst A. sparsiflora (Poaceae: Bambusoideae). Taxon 55: 227-228.Stapleton, C. M. A. 2006–2007. Bamboo Identification. http://bamboo-identification.co.uk/index.html.Accessed 30 July 2007.Stapleton, C. M. A. 2007. Bambuseae Nees. In Flora of North America.Volume 24. Magnoliophyta: Commelinidae (in part): Poaceae, part 1.Edited by M. E. Barkworth, K. M. Capels, S. Long, L. K. Anderton, and M.B. Piep. Oxford University Press. pp. 15-17.Stern, M. J., Goodell, K., and Kennard, D. K. 1999. Local distribution ofChusquea tomentosa (Poaceae: Bambusoideae) before and after a floweringevent. Biotropica 31: 365-368.Tian, B., Chen, Y., Yan, Y., and Li, D. 2005. Isolation and ectopic expression ofa bamboo MADS-box gene. Chinese Science Bulletin 50: 217-224.Wu, Z. Y, Raven, P. H., and Hong, D. H. (Editors). 2006. Flora of China:Poaceae, Volume 22. Science Press and Missouri Botanical Garden Press,Beijing and St. Louis.Zhang, W., and Clark, L. G. 2000. Phylogeny and classification of the Bambusoideae(Poaceae). In Grass Systematics and Evolution. Edited by S. W. L.Jacobs and J. Everett. CSIRO, Melbourne, Australia. pp. 35-42.

56Carrierea calycina, the goat horntree – a two part account of itshistory in western cultivation andrecent reintroductionPart 1 P. WhartonCarrierea calycina Franchet – Salicaceae (Flacourtiaceae)In 1971 as a forestry undergraduate at the University College ofNorth Wales, Bangor, I often referred to ‘Trees and Shrubs Hardy in theBritish Isles’ by W. J. Bean (Bean, 1976). The entry for Carrierea calycina, atree in the Flacourtiaceae, contained information that the plant explorer,E. H. Wilson, had collected the species in 1908 and introduced it tothe United Kingdom from seed obtained in western Sichuan (Mupin).The plants were raised in the U.K. by Messers. Veitch and Sons, anddistributed in 1912 as part of their closing sale. Bean reported, “Itfirst flowered in this country in the garden of Capt. and Mrs. Desborough,Tulgey House, Broadstone, Dorset, in June 1929 and again thefollowing year, but this specimen, which provided the material for theplate in the Botanical Magazine (Turrill, 1949), died in 1931”. This storytypifies the generally ephemeral nature of the early Wilson introductions,to which Roy Lancaster alludes later in this article. Bean tells usit grew to flowering size at Bodnant, Gwynedd and Borde Hill, Sussex,but unfortunately, died out during or just after World War II. A plantat Kew, though seemingly healthy, died before flowering. It seemedthat the only surviving specimen in the British Isles was at Birr Castlein Eire, but a second was later found at Rowallane in Northern Ireland.I reintroduced this tree to cultivation some 20 years after I wasemployed as curator of the David C. Lam Asian Garden at the Uni-Peter WhartonUBC Botanical Garden and Centre for Plant Research,6804 SW Marine Drive, Vancouver, BC, Canada, V6T 1Z4Roy Lancaster58 Brownhills Road, Chandlers Ford, Eastleigh,Hampshire, S05 2EG, UK

Davidsonia 18:2 57Photo: Peter WhartonFigure 1. Dashahe Cathaya Reserve, Dalou Shan, northern Guizhou. October 1994.

58versity of British Columbia (UBC) Botanical Garden. A cooperativescientific agreement between UBC and the Nanjing Botanical GardenMemorial Sun Yat-sen, in Jiangsu, China and my friendship with Dr. HeShan-an, then Director at Nanjing, led in 1994 to an invitation to visitthe Dashahe Cathaya Reserve, a remote conservation area in the DalouShan in northern Guizhou, close to the Chongqing municipal regionborder in Sichuan (Figure 1). This reserve contained scattered stands ofCathaya argyrophylla (Cathay silver fir) (Figure 2). These trees are oftenperched precariously on overhanging limestone buttresses and are a distinctivefeature of the table-topped karst mountains in the Dalou Shan.We spent 15 memorable days exploring this little known reserve, whichis rich in both rhododendrons and a range of wildlife, including landcrabs. The entire Dalou Shan region was thickly forested and a majorrefuge for the Chinese tiger before 1959. Sadly, this majestic animal isjust one of many larger mammals that have been hunted almost to extinctionduring a time of calamitous forest destruction.My first sight of Carrierea calycina (the goat horn tree; ‘yang jiao shu’in Mandarin) was along the banks of the Dasha He, a small river thatflows through the centre of the reserve, when I noticed a tree withconspicuous claw-like capsules. I remembered a colour slide taken byTed Dudley (U.S. National Arboretum, Washington, D.C.) who photographedthis species (under collection # SABE 245) during the firstSino-American Botanical Expedition in 1980 to the Shennongjia regionof western Hubei. Dudley was particularly pleased to find the speciesin the Shennongjia, where it is uncommon, and this unexpected discoverygave me my own ‘botanical golden fleece’. As well, it was the firstrecord of the species in the Dashahe Cathaya Reserve. We collectedthe distinctive spindle-shaped capsules, which are similar in profile toa goat’s horn and often form pendulous clusters of three to five at thebranch tips. At maturity these long, bright green, lightly felted capsulessplit along three sutures, the valves arching backwards to form a structurereminiscent of the flowers of Clematis texensis. The silky, ribbedcapsule interiors are packed with small, flattened, winged seeds. A goat’shorn cornucopia, symbolizing abundance, is an accurate metaphor ofthis species’ fecundity. The distinctive fruits and the flat-topped crown

Davidsonia 18:2 59profile were the two characters needed to spot other specimens in thesurrounding moist woods. Wilson alludes to these very features in ANaturalist in Western China (Wilson, 1913)."Carrierea calycina, a wide-spreading flat-topped tree, is very common in rockyplaces by the stream-side, and was laden with torpedo-shaped, velvety-grey fruit whichwas not ripe."My collection PW 68, wasacquired at 320m (1050ft) from anopen crowned riverside specimenof just over 11m (36ft) tall. ThePhotos: Peter WhartonFigure 2. Cathaya argyrophylla (Cathay silver fir) in the Dashahe region perched at theedges of limestone crags.

60associated ligneous vegetation included a similar sized Magnolia insignisimmediately next to the goat horn tree. Nearby were fine specimens ofAcer davidii (PW 74) with massive leaves (similar to A. davidii ‘GeorgeForrest’), Liquidambar formosana, and overhanging specimens of Cyclocaryapaliurus and Cladrastis sinensis. The headman of a local village guided us toseveral other Carrierea locations the following day. My second collection(PW 84) was from a 15m (49ft) tree growing within vigorous secondarymesic forest. The trees here had obviously been ‘drawn up’ by thesurrounding forest, yet still displayed full, healthy crowns. They grewin small groups or as single trees at the base of moist, north-east facingrocky slopes. Despite the thin soils, they appeared to receive year roundmoisture. The soils at these sites ranged from moderately to weaklycalcareous. Leachate from the limestone cliffs above percolated downtowards the Dasha He, where the most of the goat horn tree populationsoccurred. Surrounding trees included, Betula luminifera, Cornusmacrophylla, Corylus chinensis and Idesia polycarpa (another member of theFlacourtiaceae [Salicaceae]). These grew amongst a rich herbaceousflora that included colonies of a curious, viviparous Senecio bulbiferusand a gargantuan Ligularia wilsoniana (giant groundsel) that had floweringstems up to 3m (9.8ft) tall. Wild pigs evidently excavate and relishthe fleshy roots of the groundsel. Small groups and individuals of theCarrierea were mixed within the forest along streams or where moisturewas flowing through the soil profile. It is a pioneer species that prefersmoist forest margins and riparian sites, which, if left undisturbed,will eventually give way to broadleaved evergreen forest dominated byhusky oak relatives, such as Castanopsis chunii, C. platyacantha and Lithocarpushancei. Much of the lowland forest of this reserve appeared tobe regrowth after clearing for subsistence rice cultivation several hundredyears ago. Vestiges of old terracing and abandoned walnut (Juglanscathayensis) groves were clearly visible. Today, the main threat to the reserveis illegal tree cutting and collection of traditional herbal plants byitinerant peasants from the Chongqing municipal region (Sichuan) whoenter from the north. The reserve is also home to a very rare walnutrelative, Annamocarya sinensis (Chinese hickory), which is at the northernlimit of its distribution.

Davidsonia 18:2 61Carrierea calycina has a wide distribution throughout Hubei, Hunan,Guangxi, Guizhou, Sichuan and Yunnan between 1300m–1600m(4265–5250ft) elevation. Deforestation at these low elevations hasgreatly fragmented its formerly extensive range. During his travels inwestern China in early part of the 20 th century, E. H. Wilson describedit in most complimentary terms:“This handsome tree is very rare in western Hupeh, but common in westernSzech’uan, especially by the side of woodland streams up to 1200m (3937ft) altitude.Whilst not forming a tall or even large tree, the much–branched and flat head iswide-spreading; the bark is grey and usually smooth, but in very old trees it becomesfurrowed and corrugated. The flowers are ivory-white and of much substance.”(Wilson, 1917).It is clear to me that Dashahe should be protected in the longtermand that more botanical fieldwork is warranted in this remarkableregion. After my return to Canada I distributed seed to botanical institutionsand individuals in the United Kingdom, USA, New Zealandand Ireland.Royal Botanic Gardens, Kew, U.K. (PW 68, 84)Wakehurst Place, RBG Kew, U.K. (PW 84)Ness Botanic Garden, University of Liverpool, U.K. (PW 84)Howick Gardens, Northumberland, U.K. (PW 84)Trewithen Gardens, Cornwall, U.K. (PW 84)Mr. Roy Lancaster, Hampshire, U.K. (PW 68, 84)National Botanical Gardens, Glasnevin, Ireland (PW 84)University of Washington Botanic Gardens, (PW 68, 84)Seattle, WA., U.S.A.Pukeiti Trust Garden, New Plymouth, N.Z. (PW 68, 84)Roy and I would certainly appreciate obtaining further informationon the location and condition of trees grown from the seeds that weredistributed.

62Botanical descriptionCarrierea is named after the French botanist and horticulturist Elie-Abel Carriere (1818-1896). The specific name calycina means ‘belongingto the calyx’, or, ‘with a well-developed calyx’.Growth habitIn the wild state, trees generally range from 6-9m (19-29ft) in opensituations, while in moist, forested locations trees can approach 15m(49ft). Wilson reported older trees with girths up to 1-1.6m (3.2-5.2ft),and occasional veteran trees approaching 2m. No tree I observed inthe wild had a diameter over 60cm (23”). The bark of young trees islight brown, with a slight silver cast, which contrasts beautifully withthe prominent chestnut coloured lenticels. These are liberally scatteredalong the trunk and main branches. Wilson describes the bark of veterantrees as ‘furrowed and corrugated’. Young trees, especially growingin open, waterside situations, tend to develop wide-spreading openbranched crowns and eventually assume almost flat-topped profiles.Most trees derived from PW collections growing in the U.K. and NewZealand are showing typical, open crowned characteristics. Some youngindividuals growing in moist, forested conditions at the UBC David C.Lam Asian Garden are vigorously upright, show persistence of apicaldominance and seem likely to achieve maximal mature height. Theyoung shoots are initially downy, becoming glabrous by late summer andhave a reddish-green colouration that fades at maturity to light brownwith grayish lenticels. The main lateral branches close to the trunk arestrongly horizontal, reminiscent of the related Idesia polycarpa. At theouter margins of the crown they assume a distinctively vertical pose,even with a full complement of summer foliage.Buds and leavesThe 3-8mm (0.11-0.31in) buds are alternate, rounded, reddish andfinely downy, and spilt to unfurl leaves in mid to late March (Figure 3,see front cover). The alternate, slightly pendulous leaves are glabrous,dark glossy green above and paler beneath. Leaf shape ranges fromovate or broadly ovate-lanceolate to oblanceolate with a gradual taperto a slender apex. Leaf bases are quite variable, ranging from rounded

Davidsonia 18:2 63or cordate to cuneate. Leaf size is 10-15 x 5-8 cm (4-6 x 1.9-3.1in) onmature flowering specimens, while on vigorous trees on moist sites theleaves often resemble those of Populus, often reaching 24-30 x 6-8cm(9.4-11.8 x 2.3-3.1in). The leaf margins are set with regular, 1mm pegliketeeth and have conspicuous, in-rolled flanges. Petiole length rangesfrom 2-5cm (0.8-2in) and the petiole curves downwards where it attacheswith the leaf-blade. There is a pair of small glands at the junctionof the leaf blade and petiole. These are pinkish in color (similar to thevegetative buds) and resemble those found in Populus. Leaves fall greenin the late autumn (November) if weather conditions remain frost-free.We have never recorded any autumn colour.FlowersPlants are remarkably variable with respect to flowering. Dr. QuentinCronk and his students have been studying the phylogeny of Carriereaand the genomics of the Salicaceae, and they have shed some lighton this sexual plasticity. Several genera in Flacourtiaceae (a considerablenumber of flacourts are now placed within the Salicaceae, (Cronk, 2005)including Carrierea, are functionally dioecious, yet sporadic occurrencesof monecious and hermaphrodite individuals occur. The fertile seedpodsof hermaphrodite trees are generally located at the terminal positionof the panicle. The same morphology occurs in the closely relatedgenus Idesia. It is interesting to speculate about the adaptive advantageof this wide range of sexual options. Both monoecy and especiallydioecy inhibit inbreeding (Quentin Cronk, pers. comm., 2005), whichcan have negative effects on a species’ long-term viability, but hermaphroditicbehavior may give the plant survival benefits, when geographicisolation, climatic stress or physical disturbances pose powerfulnatural dangers to their continuance.Individual flowers are borne in erect or spreading inflorescenceson lax terminal shoots (Figure 4). The flower masses resemble a candelabra.Wilson described them as “…curiously-shaped, waxy-whiteflowers borne in erect panicles” (Wilson, 1913. page 173). The rose-redrachis of the panicles is a most attractive foil for the pale flowers. Upto 10 flowers are produced, each with five heart-shaped, rugose sepals,which functionally replace normal petals. Flower colour varies from

64Photo: Peter WhartonFigure 4. A flowering specimen at the Darts Hill Garden in Surrey, BC. July 2006.ivory white through yellowish to greenish white, and Wilson remarkedthat he came across ‘blush’ flower forms. I suspect that these pinkishflower forms may be due to normal aging discolouration. The sepalsform beaker-shaped flowers, with the sepal tips slightly twisted, up to3.1 x 2.5cm (1.2 x 0.98in) in size. The upper sepal margin is lobed andraised-high in relation to the mid-point from the tip to its point of attachment.The centres of pistillate flowers are dominated by a large,vase-shaped, downy ovary, which has three radiating yellow stigmas atits apex. Forming a boss in staminate flowers or clustered at the baseof the ovary in hermaphrodite flowers are numerous short, purplish-redstamens.Lord and Lady Rosse, the owners of Birr Castle sent me a cardsizeprint of a beautiful painting of the goat horn tree in full flower in 2004by the noted Irish botanical artist Susan Sex. The Birr tree has floweredheavily in the last six years providing a wealth of subject matter for thisfine artist.

Davidsonia 18:2 65FruitFruits are distinctive spindle or claw shaped capsules, covered indense down (Figure 5). They range in size from 7-11 x 2cm (2.7-4.3 x0.78in) and are often born in clusters of three to five at the branch extremities.The capsules split into three lance shaped valves that re-curveat the tips revealing masses of papery winged seeds.Carrieria calycina in British Columbia, CanadaPlants grown from my seed collections were planted out in the DavidC. Lam Asian Garden, in UBC Botanical Garden in the autumn of 1997(Figure 6). They consisted of seven individuals of PW 68 and six ofPW 84, but we did not fully understand the importance of abundant, allseason, soil moisture to the vitality of this species. Recent phylogeneticresearch by Cronk and co-warkers at UBC has suggested some ecologicalsimilarities between Carrierea and poplars and willows.We planted our specimens on a range of sites from well-drainedto streamside locations, from deep shade to full sun. The plants thatthrived were those growing at waterside or on well watered, oxygenatedsites that were rich in organic matter. Drought and vole damage re-Photo: Dr. T.R. DudleyFigure 5. The spindle-shaped fruit of Carrierea calycina (SABE 245). Wild specimenphotographed at the Shennengjin region, western Hubei. August 1980.

66Photo: Peter WhartonFigure 6. Carrierea calycina growing at the UBC Botanical Garden in Vancouver,British Columbia.

Davidsonia 18:2 67duced our collection to 10 individuals. The largest trees are now (winter2006-07) just under 7m (23ft) high and growing with sustained vigor.In the winter of 2004-05 four trees were transplanted from drier sitesto sunlit streamside locations and all showed healthy extension growthin the summers of 2005 and 2006. So far, we have noticed no pest ordisease problems with our plants (Figure 7 - Front Cover).A number of young trees collected under PW 68 were distributedwithin southwestern BC in the late 1990’s, including the VanDusenBotanical Garden in Vancouver, the gardens of Margaret Charltonand Charlie Sale in North Vancouver and Francisca Darts (Darts HillGarden) in south Surrey, and the more exposed garden of Don Martynin Yarrow, BC. The specimen at VanDusen Garden is healthy but small.The Darts Hill Garden tree, a small 2m (6.5ft) specimen, stress-floweredheavily in the summer of 2005 after a severe bark injury, but subsequentlydied. This was the first individual to have flowered in NorthAmerica. Charlie Sale reports that his specimen is 7.5m (24.6ft) tall witha tapered, open crown to 4.5m (14.7ft). The vigor of this plant on asteep, south-west facing slope beneath a canopy of mature Pseudotsugamenziesii (Douglas fir) would appear to be out of character; althoughthere is year-round subsurface water percolating down the sloped siteand rainfall is close to 2635mm (103in) a year. This specimen has neverflowered.Don Martyn’s garden at Yarrow, in the lower Fraser Valley, 80 km(50miles) east of Vancouver has provided a case study. The tree isplanted in the open on flat terrain in heavy clay soils over sand. Hedescribes the tree as multi-stemmed, with a congested central crown 4m(13ft) high by 3m (9.8ft) wide. The Yarrow district gets 60–80kmh (40-50mph) winds every year and below 0 o C weather often for more than aweek at a time. Recorded temperature range since the tree was plantedwas -14° to 37 o C. Average rainfall is about 1785mm (70in).Carrierea calycina in New ZealandCarrierea calycina has been growing in the central Hawkes Bay areaof the North Island since the late 1970’s in the garden of Mike andCarola Hudson. They received their original plant from Hillier’s Nurs-

68ery around 1980/1981. It appears this plant was derived from from theoriginal Birr Castle plant (private communication from Peter Dummerto Roy Lancaster). Dummer, who was Hillier’s master propagator untilhis retirement in the 1980s, remembers receiving the cuttings from BirrCastle and rooting them under mist using a hormone powder and gettingabout 45% rooting strike. The Hudson plant initially grew well to4.5m (15 ft), but collapsed in a very dry summer despite heavy watering.A new basal sprout arose emerged and formed a mature 12m (39 ft)tree that flowers every year. Fruit and seed production depend on seasonalweather. Viable seed capsules only develop at the terminal flowerof each inflorescence, which suggests an origin from hermaphroditeflowers. Cuttings root well but never seem to thrive, and all have so fardied. The two oldest plants of seven grown from self-seed are now 9m(29.5ft) high and flower and fruit regularly. The Hudsons also grow onePW plant, but it has not yet flowered or fruited. They report that thisGuizhou tree has a more spreading habit than the more upright Wilsonintroduction, although Wilson maintained that his introductions hadflat-headed wide spreading crowns.Part 2 R. LancasterCarrierea calycina in British cultivationThe only tree of note I have seen in British or Irish cultivation isfrom the famous Wilson collection, presumably W. 1212, of 1908 collectedfor the Arnold Arboretum and growing at Birr Castle in Co.Offaly, Ireland. This tree, planted in 1916 was 16m (52ft) x 39cm (15in)when measured in 2000 for the Tree Register of Ireland. The recordshows that this tree was purchased from Veitch Nursery during the closingdown sale in 1914, which suggests that seed was sent by Wilson directlyto Veitch, although it is equally possible that the nursery receivedit as seed or as plants from the Arnold Arboretum. Curiously, plants ofCarrierea were being offered as early as 1912 by the Daisy Hill Nurseryof Newry, Co. Down who described it in their catalogue as an evergreenshrub.

Davidsonia 18:2 69Although, Wilson encountered this species on several earlier occasionsfor the Veitch Nursery, e.g. July 1903 (No. 3227) no precise locality;June 1904 (No. 4752) on Mt. Omei, Sichuan; and in western Hupeh(No. 1104) no precise locality or date; it is not recorded that he collectedother than herbarium specimens in flower. Plants presumably from thesame Wilson 1908 collection were subsequently grown by several othergardens in Britain and Ireland although few appear to have flourished.In A Garden Flora, describing trees and flowers grown in the gardenat Nymans, Sussex, between 1890 and 1915, a note by Muriel Messell(1918) records, “so far the hardiness of this tree, which Mr. Wilsonintroduced from China, has not been proved. Our plants were raisedfrom Mr. Wilson’s seeds, and one small tree has been planted out ofdoors for three years on a north-east slope, where it seems at home.”Sadly, there is no record of this tree or any others grown from this seedat Nymans that have survived the test of time.A tree, that is thought to have been planted around 1919, still growsat Rowallane in Northern Ireland. According to the former Head Gardener,Michael Snowden, now retired, this tree which may have comefrom the Donard Nursery of Co. Down in 1919 or, alternatively, theDaisy Hill Nursery in the same county, was planted in an area known asHolly Rock outside the main garden and was neglected if not forgottenfor many years. It presently stands at 5.5m (28ft), its stem forkingat 50cm (19in) above ground. It seems to have never flowered.Another tree recorded as growing in the Walled Garden at Rowallanein the 1920’s is no longer extant. The first mention of Carrierea calycinain the Gardener’s Chronicle appeared in the March 22 nd 1930 issuewhen F.E.G. Southampton, contributed a note in New or NoteworthyPlants. “It appears quite hardy in this country (Britain), plants havingcome through the severe weather experienced during the spring of 1928without suffering any apparent injury.” He further added “Unfloweredtrees here, planted in a sheltered position on soil of a heavy, clayeynature have reached a height of 6ft (1.8m).” It was in the same year thatflowering material from a tree growing at Tulgey Wood, Broadstone,Dorset was sent by a Mrs. Desborough to Kew to provide the illustra-

70tion for the Curtis Botanical Magazine entry which was not publisheduntil 1949.In autumn 1994 I received seed of Carrierea calycina, PW 68, fromPeter Wharton. I gave the seed to the Surrey plantsman, Harry Haywho, apart from having a noted garden and arboretum, has an uncannyskill at raising plants from seed or cuttings. Two years later, Harry returnedto me a tray of sturdy seedlings, which I potted on. One ofthese I planted in my garden whilst the rest were distributed to amongothers, Hergest Croft in Herefordshire, Abbotsbury Subtropical Gardensin Dorset, Dunloe Castle Hotel Gardens in Co. Kerry, Ireland andthe National Arboretum, Westonbirt, Gloucestershire.My plant settled in and began to put on growth. In May 2004, almost10 years after I received the seed, it produced its first flower buds fromthe tips of several shoots. A month later, these buds began to open andthe characteristic bell-shaped, greenish-cream, five-lobed downy calyx,2.5cm (0.98in) long, appeared. On examining these flowers it was clearthat they were male as there was no discernable ovary present, just aconspicuous brush of cream-coloured stamens of varying length to 2cm(0.78in). The flowers lasted for less than 10 days before falling intact.The same week, I took some cut specimens of the flowering stems to ameeting of The Garden Society held at the Royal Botanic Gardens Kewwhere, to my surprise, Maurice Foster produced flowering specimens ofa tree from the same Peter Wharton collection growing in his garden inKent. It was clear that the flowers of Maurice’s tree were female as theywere virtually no stamens present, only a downy ovary with a conspicuousflattened three-lobed stigma. The calyx in both sets of specimens,were a pale greenish-cream.On 6 th July 2004, I attended a gathering at Nymans in Sussex to celebratethe 50 th anniversary of the National Trust’s stewardship of thoseGardens. I was once more surprised to find the lunch tables decoratedwith cut flowering specimens of Carrierea calycina provided by the Earlof Rosse from the Wilson tree at Birr Castle. It was clear that they weretoo, were female.In 2005 my tree did not flower but more than made up for this in2006 when it flowered abundantly (flowers all male) from late June into

Davidsonia 18:2 71early July. Even when in full flower, this tree is not an eye-catcher as thepale green flowers do not stand out enough against the leaves to createthe necessary contrast. Fortunately, the foliage more than makes up forthis deficiency. The leaves are slightly pendulous so that their glossy darkgreen surface is well displayed on the wide-spread branches. Leaves onmy tree are from 20-33cm (9-13in) long, a third of which belongs tothe petiole. The blade varies from narrow elliptic to oblanceolate, up to7cm (2.75in) at the broadest, acuminate tip, broad cuneate to rounded,sometimes obliquely so at the base, regularly toothed with blunt or pegliketeeth, each with a circular glandular pit beneath. They are glabrousexcept for the shortly downy petiole and tufts of hair in the vein axilsbeneath, especially those of the lowermost veins. There is a small glandat the base of the blade either side of the petiole. When measured on13 th July 2006, this tree was 8m (20ft) tall with a girth at 1.5m (5ft) of27cm (10.5in). The lower branches are spreading, the upper ascendingthen spreading, creating a loosely layered appearance.I have another specimen grown from the Wharton 1994 introduction(PW 68) at Abbotsbury Subtropical Garden in Dorset. This hasgrown well and in June 2006 was approximately the same size as mine.It has yet to flower. Another of the same introduction planted in theWalled Garden at Dunloe Castle Hotel Garden was around 3m (10ft)in 2005. This has yet to flower. A tree seen in July 2006 growing in thewalled garden of Pan Global Plants at Frampton-on-Severn near Stroudin Gloucestershire was 3 x 3m (10ft x 10ft) having previously had its topblown out in a storm. It was received from the National Arboretum atWestonbirt as a seedling from the Peter Wharton 1994 collections (PW84). According to the nursery owner Nick Macer, this tree flowered(all female) in 2004,. In July 2006 the flowers appeared to be developingfruits which suggests that they had been pollinated. Short stamensor shriveled remains were present which points to these flowers beinghermaphroditic!Maurice Foster informed me that his tree flowered again in July 2006though not as prolifically as in 2004. Seed capsules were up to 3cm(1.75in) long, surrounded at their base by what look like the remains ofmany short stamens. One or two of the capsules contained developing

Davidsonia 18:2 73ReferencesBean, W.J. 1976. Trees and Shrubs Hardy in the British Isles, 8th edition.Volume 1. London: John Murray.Cronk Q.C.B. 2005. Plant eco-devo: the potential of poplar as a model organism.New Phytologist 166: 39-48.Messel, M. 1918. In, A garden flora trees and flowers grown in the gardens atNymans 1890-1915. By L. Messel. With illustrations by Alfred Parsons.London: Country Life.Turrill, W.B. 1949. Carrierea calycina Franch. Botanical Magazine N.S. 53: 1.Wilson, E.H. 1913. A Naturalist in Western China. London: Methuen & Co. p.239. New edition 1987. Everyman Publishers.Wilson, E.H. 1917. Plantae Wilsonianae 1:284 Cambridge: Cambridge UniversityPress.

Table of contentsEditorialTaylorThe North American flowering of thecultivated fountain bamboo, Fargesia nitida(Poaceae: Bambusoideae), inVancouver, British Columbia, CanadaSaarelaCarrierea calycina, goat horn tree - a two part accountof its history in western cultivationand recent reintroductionWharton and LancasterBook review.van WijkGleanings4143567476